Download full issue - PanamJAS

PAN-AMERICAN JOURNAL OF AQUATIC SCIENCES - PANAMJASEditor in Chief: María Cristina Oddone.Scientific Editors: Gonzalo Velasco, Pablo Muniz, Danilo Calliari, Douglas F. M. Gherardi.Invited Scientific Editors for this Volume: Margareth S. Copertino and Alexandre M. Garcia.Honorary members: Jorge P. Castello, Omar Defeo, and Kirk Winemiller.Graphic editors: Samantha E. G. Martins and Patricio Rivera Donoso.Special Graphic Editor for this Volume: Daniel Loebmann.PanamJAS is a non-profit Journal supported by researchers from several scientific institutions.PanamJAS is currently indexed inWeb of KnowledgeAquatic Sciences and Fisheries Abstracts (ASFA - FAO)Directory of Open Access Journals (DOAJ)Online Access to Research in the Environment (OARE)Thomson BiologyBrowser databaseIndexCopernicus InternationalSCOPUSQualis CAPES (Brazil)Portal de Periódicos CAPES (Brazil)Sistema de Bibliotecas SISBI-UFU (Brazil)Electronic Resources from Smithsonian Institution Libraries (USA)NewJour - Lauinger Library, Georgetown University (USA)Divulgador Científico Ensenadense (Mexico)SOCOL@R - Open Access Platform CEPIEC - Ministry of Education (China)PAN-AMERICAN JOURNAL OF AQUATIC SCIENCES2006, 1-22010, 5(2)Quarterly JournalISSN 1809-9009 (On Line Version) CDU 570Cover photo of this issue: Aerial photograph from Cassino Beach, Rio Grande municipality, state of Rio Grande doSul, Brazil. Picture taken by Daniel Loebmann.

Pan-American Journal of Aquatic SciencesSpecial Issue: Climate Change and Brazilian Coastal ZoneIntroduction to the Special Issue on Climate Change and Brazilian CoastalZone……........................................................................................................................IForeword.....................................................................................................................IXResearch articlesBrazilian coastal vulnerability to climate change.MUEHE, D......................................................................................................................173Potential vulnerability of the Brazilian coastal zone in its environmental, social, and technologicalaspects.NICOLODI, J. L. & PETERMANN, R. M..................................................................................184A methodology for assessing the vulnerability of mangroves and fisherfolk to climate change.FARACO, L. F. D., ANDRIGUETTO-FILHO, J. M. & LANA, P. C. A..............................................205Status of Eastern Brazilian coral reefs in time of climate changes.LEÃO, Z. M. A. N., KIKUCHI, R. K. P., OLIVEIRA, M. D. M. & VASCONCELLOS, V.......................224Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration overthe Brazilian continental shelf.CIOTTI, Á. M., GARCIA, C. A. E. & JORGE, D. S. F.................................................................236Regime shifts, trends and interannual variations of water level in Mirim Lagoon, southern Brazil.HIRATA, F. E., MÖLLER JR., O. O. & MATA, M. M.................................................................254Potential vulnerability to climate change of the beach-dune system of the Peró coastal plain - CaboFrio, Rio de Janeiro state, Brazil.MUEHE, D., BELLIGOTTI, F. M., LINS-DE-BARROS, F. M., OLIVEIRA, J. F. & MAIA,L. F. P. G......................................................................................................................267Historical assessment of extreme coastal sea state conditions in southern Brazil and their relation toerosion episodes.MACHADO, A. A., CALLIARI, L. J., MELO, E. & KLEIN, A. H. F................................................277Climatic changes in the coastal plain of the Rio Grande do Sul state in the Holocene: palynomorphevidences.MEDEANIC, S. & CORRÊA, I. C. S........................................................................................287

Representations in the Brazilian media of the impacts of climate change in the coastal zone.HELLEBRANDT, D. & HELLEBRANDT, L...............................................................................298Differences between spatial patterns of climate variability and large marine ecosystems in thewestern South Atlantic.GHERARDI, D. F. M., PAES, E. T., SOARES, H. C., PEZZI, L. P. & KAYANO, M. T.........................310An essay on the potential effects of climate change on fisheries in Patos Lagoon, Brazil.SCHROEDER, F. A. & CASTELLO, J. P……………………………………...........................320Long-term mean sea level measurements along the Brazilian coast: a preliminary assessment.LEMOS, A. T. & GHISOLFI, R. D..........................................................................................331Vulnerability and impacts related to the rising sea level in the Metropolitan Center of Recife,Northeast Brazil.COSTA, M. B. S. F., MALLMANN, D. L. B., PONTES, P. M. & ARAUJO, M...................................341Temporal changes in the seaweed flora in Southern Brazil and its potential causes.FAVERI, D., SCHERNER, F., FARIAS, J., OLIVEIRA, E. C. & HORTA, P. A.....................................350AnnexRio Grande Declaration. 1 st Brazilian Workshop on Climate Changes in Coastal Zones: CurrentKnowledge and recommendations. Universidade Federal do Rio Grande, Rio Grande, Brazil, 16-17September 2009……………………...............................................................................XII

Introduction to the Special Issue on Climate Change andBrazilian Coastal ZoneMARGARETH S. COPERTINO 1 , ALEXANDRE M. GARCIA 2 , JOSÉ H. MUELBERT 3& CARLOS A. E. GARCIA 41 Laboratório de Ecologia Vegetal Costeira, 2 Laboratório de Ictiologia, 3 Laboratório de Ecologia do Ictioplâncton,4 Laboratório de Estudos dos Oceanos e Clima.Instituto de Oceanografia, Universidade Federal do Rio Grande , Av Itália km 8, Rio Grande (RS), 96201-900, Brazil.doccoper@furg.brAbstract. The multidisciplinary Coastal Zone (CZ) network from the National Institute of Sciences andTechnology (INCT) for Climate Change, aims to evaluate the state of knowledge and coordinate projectsdealing with the effects of global climate changes in the country’s coastal zone. During its first year, CZfocused on literature review, historic data analysis and vulnerability studies. The studies were presentedduring the I Brazilian Workshop on Climate Change and Coastal Zones, and resulted in the fifteenscientific articles that comprise this Special Issue. Based on regional case studies and a few broad nationalassessments, results showed how the large Brazilian coast and their ecosystems are highly vulnerable toclimate variability, and which parameters and regions may be more impacted by global climate change.With only a few studies aiming to evaluate the vulnerability of biological, ecological and socio-economicparameters, along with the many deficiencies on basic knowledge, the country’s coastal and marine zonesare still neglected by climate change policies. Although a series of scientific goals still need to beachieved to better evaluate the effects of climate change on the Brazilian coastal zone, the present volumeconstitute a first step towards the provision of guidance to managers and policymakers.Key-words: Global Climate Change, impacts and vulnerability, oceanography, coastal geology, marineecology.Resumo. Introdução ao Volume Especial sobre “Mudança Climática e Zona Costeira Brasileira”. Arede Zona Costeira do INCT para Mudanças Climáticas, de caráter multidisciplinar e interinstitucional,objetiva de avaliar o estado do conhecimento e coordenar projetos que investiguem os efeitos dasmudanças climáticas em zonas costeiras brasileiras. Em seu primeiro ano, os projetos focaram emrevisões da literatura, análises de banco de dados e sobre vulnerabilidades dos ecossistemas. Osresultados foram apresentados durante o I Workshop Brasileiro de Mudanças Climáticas em ZonasCosteiras, resultando também nos 15 artigos deste Volume Especial. Tendo como base estudos de casoregionalizados e algumas avaliações nacionais, os resultados mostram como a extensa costa Brasileira eseus ecossistemas são altamente vulneráveis a variabilidade climática, e quais parâmetros e regiõespoderão ser mais impactados pelas mudanças climáticas globais. Com poucos estudos específicos queavaliem a vulnerabilidade de parâmetros biológicos, ecológicos e socio-econômicos, associado àsinúmeras deficiências de conhecimento básico, as zonas costeira e marinha brasileiras são aindanegligencias pelas políticas de mudanças climáticas. Ainda que uma série de metas científicas necessiteser alcançadas para melhor avaliar os efeitos das mudanças climáticas em zonas costeiras brasileiras, osartigos apresentados neste volume constituem passos preliminares para guiar a gestão e a elaboração depolíticas públicas sobre o tema.Palavras-chave: Mudança Climática Global, impactos, vulnerabilidade, geologia costeira, ecologiamarinha.Pan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

IIM. S. COPERTINO ET ALLIIntroductionGlobal Climate Change (GCC) has been atthe centre of scientific debate during the 21 st centuryand has received substantial attention from society,governments and the private sector worldwide.Despite active debate and recognized climateuncertainties, the scientific evidences for globalwarming remains robust and the conclusion thathuman activities are affecting the climate cannot beundermined (e.g. Schiermeier 2010). Indeed,evidences from hundreds of studies are suggestingthat the pace and scale of changes and its impactsare surpassing the predictions outlined by theIntergovernmental Panel on Climate Change FourthAssessment Report (IPCC 2007) (McMullen &Jabbour 2009, Hoegh-Guldberg & Bruno 2010,Nicholls et al. 2010, Stroeve et al. 2011). GlobalClimate Change (GCC) is recognized now as aplanetary crisis, which impacts represent anunprecedented risk for natural ecosystems andhuman civilization.Because coastal areas are directly affectedby sea-level rise, increases in air and seatemperature, exposure to extreme events and oceanacidification, these areas will be impacted mostseverely by the predicted GCC (Trenberth et al.2007). However, despite the significance of theclimate-related risks affecting the low-lying habitats,the paucity of scientific publications dealing withspecific and regional problems is remarkable(Nicholls 2007) and coastal and marine systems arevastly underrepresented in comparison withterrestrial systems (Richardson & Poloczanska 2008,Hoegh-Guldberg & Bruno 2010). Studies andpredictions relating to the impacts of climate changeon coastal and marine ecosystems are limited by thelack of long-term data, by the lower observationalcapacity and by a lack of integrated approaches tolink diverse aspects of science and society. The lackof information at the regional and local scalesproduces uncertainties about climate, and theseuncertainties hamper efforts to plan for the future(Malone et al. 2010). Furthermore, no coastal andmarine region on earth is exempt fromanthropogenic modifications and impacts that affectthe ecosystem equilibrium and capacity to bufferclimate change (Halpern et al. 2008).Brazil is a globally important country in thecontext of climate change. Owing largely to itshistorical deforestation rate, the country is theworld’s fourth-largest greenhouse-gas emitter(Nepstad et al. 2011). Moreover, Brazil dependsstrongly on its very abundant natural resources.Because of that, the country has therefore greatpotential to contribute to the reduction of GCC risks.Most of the discussion about Brazil’s contribution tomitigate global climate (Santili et al. 2005), as wellas studies about impacts and vulnerability of naturalecosystems to climate change, has been focussed onterrestrial ecosystems (e.g., Salazar et al. 2007,Lapola et al. 2009). However, the country’s largecoastal and marine zones as well as the country’srole in buffering climate change impacts andmitigating emissions are still neglected by manynational and international climate change forumsand policies (Copertino 2011). The Brazilian coast,as many other coasts, is highly vulnerable to presentdayclimate variability and may be profoundlyimpacted by the projected climate change (Muehe2006, Neves & Muehe 2008). These impacts wouldpose serious threats to coastal biodiversity,ecological functions and services to society,including the coastal zone’s carbon sequestrationcapacity. However, studies on Brazilian coastalvulnerability to GCC were few and isolated, owingin part to the fact that the country’s coastal zone isvery extensive.In an attempt to fill this gap in currentinformation, a Coastal Zone (CZ) project wasestablished within the framework and aims of theNational Institute of Sciences and Technology(INCT) for Climate Change and the BrazilianNetwork for Global Climate Change (RedeCLIMA). The full historic context of these largeresearch programs and their first results are found intheir activity report (INCT for Climate Change,2010). A brief overview is given by Garcia & Nobrein the foreword of this volume. Formed by nearly 50scientists, the multidisciplinary and interinstitucionalCZ network aims to evaluate the current state ofknowledge, identify deficiencies, establish protocols,integrate/coordinated projects and design scientificquestions in several GCC-related research topics.The goals of the project are to investigate theimpacts of climate change on the Brazilian coast andto determine the coast’s vulnerability to thesechanges. The CZ project also seeks to proposemitigation and adaptation measures to compensatefor and to buffer the impacts of climate change.To achieve these goals, the CZ projectorganised the “First Brazilian Workshop on ClimateChange and Coastal Zones”, which was hosted bythe Federal University of Rio Grande (FURG)(Figure 1). The workshop brought together expertsfrom fourteen national institutions and tookimportant steps that facilitated the attainment of thefirst project goals. In this three-day workshop,experts evaluated the status of the current knowledgeof Brazilian coastal systems and processes andPan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

Climate Change and Brazilian Coastal ZoneIIIpresented their preliminary results. In addition,participants offered recommendations for futurestudies. The information gathered in the workshopresulted in fifteen scientific articles that comprisethis Special Issue of the Pan-American Journal ofAquatic Sciences. In this editorial paper, we providean overview and we also highlight the main findingsof these articles, placing their importance andrelevance in the perspective of Brazilian coastalzones facing climate change.Overview of the volume contentThe Brazilian coastline is almost 9,000 kmlong and includes a variety of coastal features suchas sedimentary cliffs, large and deeply incisedestuaries, crystalline headlands and low-lyingcoastal plains. By conducting a broad introductoryreview on the morphology and vulnerability of theBrazilian coastal zone, Muehe explores how eachcoastal region will respond in different ways to theexpected climate changes and associated sea-levelrise. The author concludes that erosion, althoughirregularly distributed, is affecting the entireshoreline. The risk of erosion can be magnified bysea-level rise and by the increase in the frequencyand intensity of storms. The southern Braziliancoastline, for example, is exposed very often toextreme events, i.e., storm surges and storm waves,mostly associated with extra-tropical cyclones. Thefrequency and intensity of these extreme events andtheir effects on erosion over the past 30 years isanalysed by Machado et al., who report a total of 40extreme events associated with maximum erosionand surge elevation in Rio Grande do Sul State. Ifthe predicted sea-level rise occurs, the rise willcertainly increase the importance of the alreadyexistingstorm surge hazards by intensifyinginundation and the resulting erosion. In anothercontribution, Muehe et al. show that erosive trendsare not evident for some coastal-plain areas in Riode Janeiro State. However, the lack of sedimentsources can make the system highly vulnerable tosea-level rise. The sandy dune system is highlyvulnerable to changes in water balance. Thesechanges can affect the sparse foredune vegetationcover and have drastic consequences for sedimenttransport and coastal morphology. The study hasprofound implications for predicting the effects oftemperature increases and rainfall on the fragilemorpho-sedimentary balance in these coastal plains.The vulnerability of the Brazilian coastalzone to climate change and the hazards and risksinvolved are evaluate by Nicolodi & Petermann,who take into account the natural, social andtechnological characteristics of each coastalgeographic region. Based on broad-scale anddetailed information for each region, the authorsindicate the main economic sectors likely to beaffected and identify different levels of vulnerability.The areas identified as high and very highvulnerable should be at the top of the priority list forclimate change policies and plans. One of the mostvulnerable areas affected by sea-level rise is theRecife (Pernambuco State) metropolitan centre. Thisintensely populated area is highly exposed to coastalerosion and inundation and is the site of several landuse conflicts. Based on optimistic and pessimisticIPCC emission scenarios, Costa et al. conduct avulnerability and impact assessment for the region.They define coastal and estuarine flooding zones anddiscuss the natural, historical-cultural and economicresources at risk. Both studies provide guidance forpreventive strategies to adapt to the effects ofclimate change.The overall consensus seems to be thatdifferent Brazilian coastal regions will certainly beaffected and will respond in different ways toclimate changes. However, further predictions,particularly concerning sea-level rise, are verylimited. Lemos & Ghisolfi show that most of thetide-gauge measurements performed along the coastare not accurate. This widespread inaccuracy thusplaces a serious limitation on attempts to clearlydefine the effects of global warming on the mean sealevel along the Brazilian coast. These authorsinvestigate the causes for this lack of accuracy andreport that methodological, technological andinstitutional problems need to be addressed toprovide better estimates of the sea level.Brazil is home to the third-largest mangrovearea in the world (Giri et al. 2011). Brazilianartisanal fishers are highly dependent on thesemangrove systems, which occupy 80% of thecoastline and have been threatened by severalanthropogenic impacts. Faraco et al. propose aninnovative methodology to assess the vulnerabilityof mangrove system to climate change, adapted toBrazilian reality. The proposal includes not only sealevelrise estimates but also resilience and adaptationcapacity of the communities and the impacts ofBrazilian conservation policies. The integratedsocio-ecological diagnosis and approach may lead toa challenge in the development of managementpractices and more flexible policies, which are madewith the stakeholder´s participation, includingmitigation and adaptation strategies.Also important for conservation, fishing andtourism, Brazilian coral reefs are exposed to severalkinds of natural and anthropogenic impacts. Coralshave been degraded worldwide by eutrophication,Pan-American Pan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

IVM. S. COPERTINO ET ALLIpollution and overfishing (Hughes et al. 2007). Theyare also seriously threaten by climate changesimpacts such as ocean acidification and warmingthat are leading to decreased growth rates andincreasing bleaching and diseases (Anthony et al.2008, De´ath et al. 2009, Wild 2011). The status ofthe coral reefs of Brazil is reviewed by Leão et al.,who identify the reefs that are more stressed byanthropogenic impacts and more vulnerable toclimate change effects. The review provides animportant basis for establishing priorities for futureresearch and conservation plans for these uniqueBrazilian marine systems, which are marked by aresistant relict coral fauna and a high level ofendemism.The study of modes of large-scale climatevariability in relation to regional climate andoceanographic parameters can reveal trends that arefundamental to GCC studies. Focusing on theresponses of large marine ecosystems (LMEs),Gherardi et al. explore spatial patterns of correlationbetween several climate indices (Pacific DecadalOscillation, ENSO, North and South Atlantic andAntarctic Oscillation Index) and sea surfacetemperature anomalies (SSTA) for the SouthwestAtlantic. Their findings reveal distinct correlationpatterns for the Brazilian LMEs. The authors pointout that the response of LMEs to climatic variabilitymay not be controlled by the ecological criteria usedto define the LMEs. The characteristics of thedependence on the SSTA of productivity and trophicrelations in each of the Brazilian LMEs are such thatmixed responses are likely to be produced at theecosystem level. Therefore, despite the greatimportance of this framework for marine resourceassessment and management, the use of LMEresponses for monitoring GCC impacts requiresgreat caution.Changes in global primary production havebeen showed by several studies over the pastdecades, in association with multi-decadal climateand ocean variability (Chavez et al. 2011). Climatedriven changes in ocean temperature, nutrients andlight lead to changes in phytoplankton biomass andstructure, but the relative importance of the impactsand the biological responses are regionally varying(Marinov et al. 2010). The lack of long-termbiological observations makes it difficult to detectrobust changes with time at regional scales. Ciotti etal. used satellite data to conduct a first comparativeevaluation of surface chlorophyll over the entireBrazilian continental shelf (BCS), characterisingeach region. The authors conclude that the detectionof long-term trends is currently still not feasible forthe BCS. Despite the paucity of in situmeasurements and the optical complexity associatedwith Brazilian coastal waters, remote sensing ofocean color is undoubtedly the best available tool fora global description of chlorophyll and to predict theeffects of GCC on marine primary production. Theuse of these techniques in Brazilian science shouldbe enhanced.Rainfall distribution and hydrologicalbalance are likely to be affected by GCC (Meehl etal. 2007). Lake and lagoon levels are among themost apparent signals of change in water quantity.The water catchment surface area is much greaterthan the lake itself, and the water level can thereforevisibly reflect the influence of climate on a relativelylarge area (Williamson et al. 2009). By analysing 90years of water-level records for Mirim Lagoon(Brazil-Uruguay frontier), Hirata et al. find longtermchanges strongly associated with ENSO andidentified two regime shifts, apparently related to thecold and warm phases of the Pacific DecadalOscillation during the previous century. The lastwarm phase, between 1977 and 1998, significantlyaffected the water balance of a wide region inSouthern South America (Haylook et al. 2006,Agosta and Compagnucci 2008). Hirata and coauthorswork provides key insights for predicting thepossible effects of trends in regime shifts and ofclimate change on the water balance of the coastallagoons and wetlands of southern Brazil.Fisheries may be impacted by climatechanges in several ways. They may be affected bythe increasing temperature, changes in salinity andocean currents, and also through the changes innursery habitats and ecosystem primary production.Combining IPCC emission scenarios with regionalclimate models, climate and biological time seriesdata, and accounting for local anthropogenicalterations, Shroeder & Castello evaluate the futureimpacts of global warming on the fisheries of PatosLagoon. They predict that the main factor affectingfisheries in this region will be the acceleration of theoutflow current at the mouth of the lagoon. Thishydrological change will have important conesquencesfor the population dynamics of valuable fishresources and may produce changes to the fishingcalendar. Exploring a series of cause-effectrelationship, the authors call attention for the factthat climate changes impacts on the local fisheriescan be both negative and positive, depending on thefactor and on the considered species.Global warming is expect to affect thedistribution of marine communities and manyevidences of species shifting and changingabundance are raising from literature (Hoegh-Guldberg 2010). Previous biological knowledge andPan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

Climate Change and Brazilian Coastal ZoneVa comparison of present and past data allow DeFaveri et al. to find important changes in anintertidal macroalgal community in southern Brazil(Santa Catarina State). The results include theappearance of tropical and opportunist species notpreviously described in this region, and thedisappearance of local common taxa that used tooccur in the 70´s decade. Many marine populationresponses to temperature increases are expected tobe pronounced or to be initially detected intransitional warm-temperate regions, such asSouthern Brazilian region. On rock shores, however,climate change may not lead to a simple polewardshift in the distribution of intertidal organisms butmay cause localized extinctions, due to the inabilityof species to move into suitable habitats (Hawkins etal. 2008).Therefore, monitoring marine populations,with a focus on stenothermic species, can provideuseful indicators of climate change effects,particularly where historical temperature data are notavailable.The contributions appearing in the presentvolume deal mainly with climate variability and itspotential impacts during the previous decades or thepast century. In contrast, Medeanic & Correa gofurther back through recent geological time. Basedon radiocarbon and palynomorphic data from cores,these authors make a preliminary paleoreconstructionof climate, sea-level oscillation andenvironmental changes in the coastal plain of thesouthernmost state of Brazil during the Holocene(~10,000 B.P.). Their work highlights the potentialof palynomorph proxies as a tool to predict futurescenarios based on climatic change periodicity.The crucial role of scientific communicationand media coverage on the GCC issue is alsoconsidered in this volume. Hellebrandt &Hellebrandt review the coverage of GCC by theBrazilian media and identified critical points.Among these points is the predominance of issuesset by an international scientific and politicalagenda. The Brazilian media reproduces informationprovided by international agencies and news sourcesconnected with this agenda, which overlooks bothlocal reality and scientific expertise. Failure tocommunicate key messages about climate change,particularly if the regional context is not considered,can have negative implications for public awarenessand for establishing climate change policies.Final remarksIf the physics of climate change stillinvolves many uncertainties, the impacts of climatechange on coastal and marine ecosystems are stillvery much controversial, particularly in the case ofanalysis at the regional level. It seems relatively lesscomplicated to predict the effects of a changingparameter (e.g., rising temperature, acidification anddecreasing light) on isolated species from previousknowledge about their physiology, ecology andreproductive biology. However, many challengesarise when predicting the responses at thecommunity and ecosystem level, in addition to thecomplexity added by synergistic effects (Walther etal. 2002, Williams et al. 2008, Russel et al. 2009,Hoegh-Guldberg & Bruno 2010). Therefore, thechallenge posed to the worldwide scientificcommunity working with GCC and its impacts isparamount and will demand the best of our talentand creativity.The papers presented in this Special Issueconstitute a first step towards the provision ofguidance to managers and policymakers who mustaddress the impacts of GCC in the Brazilian coastalzone, particularly in relation to the Climate ChangeNational Plan. However, a series of scientific goalsstill need to be achieved to evaluate the effects ofclimate change on the Brazilian coastal zone. Thesegoals include the analysis and integration ofhistorical data, the application of standardisedprotocols, hypothesis testing, and continuous dataacquisition programmes specifically designed toobserve the coastal environments.For most Brazilian coastal ecosystems andregions, temporal and large spatial data is scarce forboth biotic and abiotic parameters. Where available,information is isolated, punctual in space, with shortand incomplete time series (if any). Most studiesalso lack integrated or comparative protocolconsistency. Scientific evaluations and futureplanning on the impacts of climate change alongBrazilian coastal zone will achieve development ifobservational systems are implemented andimproved to allow systematic monitoring programsof physical, chemical, biological and socialparameters. Experimental approaches can helpclimate change science and management, but onlyafter a better determination of questions andhypothesis, that are specific to each regionalcondition.By researching GCC, science and scientistscannot be apart from society. Public awareness andeducation may affect, at middle and long term,society opinion, behaviour and political decisions,contributing for the advance of climate changepolicies and actions. The Rio Grande Declaration,an open letter released soon after the Workshop andnow officially published in this volume, constitutes apreliminary attempt of CZ members and meetingparticipants to contribute to awareness and influencePan-American Pan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

VIM. S. COPERTINO ET ALLIpublic policies. The letter warns about climatechange problems, particularly the ones affecting thecoast, and claims for society and political action.By indicating deficiencies and revealingscientific hurdles, and pointing out the consequencesfor ecosystems and society, the present volume isexpected to offer new insights and stimulate futureAcknowledgmentsWe would like to acknowledge to theFederal University of Rio Grande and the Instituteof Oceanography, for all support given to theCoastal Zone project and by hosting the Workshop;to G. Velasco and M. C. Oddone, for all supportand contributions given towards realization ofthe present volume and also for critically reviewingthis editorial; to the thirty referees, whose reviewsand comments significantly improved themanuscripts; to D. Muehe and M. Mata forstudies of impacts, mitigation measures andadaptation strategies for Brazilian coastal zonesfacing Global Climate Change. It is also envisagedthat this initial step will foster new knowledge thatwill represent a significant Brazilian contribution tointernational scientific forums addressing thisimportant issue.reviewing this editorial; to U. Seeliger, E. Arraut andA. P. Soares, for language reviewing and editing; toD. Loebmann for formatting and publishing the finalmanuscripts; to L. Dalmas, for all support anddedication to the CZ project and the Workshop. Theproduction of this volume is a contribution of theNational Institute of Science and Technology(INCT) from Conselho Nacional de DesenvolvimentoCientífico e Tecnológico (CNPq - Proc. n°573797/2008 0).Figure 1. Participants of the I Brazilian Workshop on Climate Changes in Coastal Zones, hosted by the FederalUniversity of Rio Grande, between September 14 th and 16 th 2009. The event was promoted by Coastal Zone Networkfrom National Institute of Science and Technology (INCT) for Climate Change, and supported by CNPq.ReferencesAgosta, E. A. & Compagnucci, R. H. 2008. The1976/77 Austral Summer ClimateTransition Effects on the AtmosphericCirculation and Climate in Southern SouthAmerica. Journal of Climate, 21: 4365-4383.Anthony, K. R. N., Kline, D. I., Diaz-Pullido, G.,Dove, S. & Hoegh-Guldberg, O. 2008.Ocean acidification causes leaching andproductivity loss in coral reef builders.Proceedings of the National Academy ofSciences of the United States of America,105: 17442–17446.Chavez, F. P., Messie, M. & Pennington, J. T.2011. Marine Primary Production inRelation to Climate Variability and Change.Annual Review of Marine Science, 3:227-260.Pan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

Climate Change and Brazilian Coastal ZoneVIICopertino, M. S. 2010. Add Coastal Vegetation tothe Climate Critical List. Nature, 473: 255.De’ath, G., Lough, J. M. & Fabricius, K. E. 2009.Declining coral calcification on the GreatBarrier Reef. Science, 323: 116–119.Giri, C., Ochieng, E. , Tieszen, L. L., Zhu, Z.,Singh, A., Loveland, T. & Duke, N. 2011Status and distribution of mangrove forestsof the world using earth observationsatellite data. Global Ecology andBiogeography, 20 (1): 154-159.Halpern, B. S., Walbridge, S., Kappel, C. V.,Micheli, F., D’Agrosa, C., Bruno, J. F.,Casey, K. S., Ebert, C., Fox, H. E., Fujita,R., Heinemann, D., Lenihan, H. S., Madin,E. M. P., Perry, M. T., Selig, E. R.,Spalding, M., Steneck, R. & Watson, R.2008. A global map of human impact onmarine ecosystems. Science, 319: 948-952Haylock, M. R., Peterson, T. C., Alves, L. M.,Ambrizzi, T., Anunciação, Y. M. T., Baez,J., Barros, V. R., Berlato, M. A., Bidegain,M., Coronel, G., Corradi, V., Garcia, V. J.,Grimm, A. M., Karoly, D., Marengo, J. A.,Marino, M. B., Moncunill, D. F., Nechet,D., Quintana, J., Rebello, E., Rusticucci,M., Santos, J. L., Trebejo, I. & Vincent, L.A. 2006. Trends in Total and ExtremeSouth American Rainfall in 1960–2000 andLinks with Sea Surface Temperature.Journal of Climate, 19 (8): 1490-1512.Hawkins S. J., Moore P. J., Burrows M. T.,Poloczanska E., Mieszkowska N., HerbertR. J. H., Jenkins S. R., Thompson R. C.,Genner M. J. & Southward A. J.2010.Complex interactions in a rapidlychanging world: responses of rocky shorecommunities to recent climate change.Climate Research, 37: 123-133.Hoegh-Guldberg, O. & Bruno, J. F. 2010. TheImpact of Climate Change on the World’sMarine Ecosystems. Science, 328: 1523-1528.Hughes, T., Rodrigues, M., Bellwood, D. R.,Ceccarelli, D., Hoegh-Guldberg, O.,McCook, L., Moltschaniwskyj, N.,Pratchett, M. S., Steneck, R. S. & Willis, B.2007. Phase shifts, herbivory, and theresilience of coral reefs to climate change.Current Biology, 17: 360–365.INCT for Climate Change, 2009. 2010. ActivityReport. S. José dos Campos, 78pp.IPCC 2007: Summary for Policymakers. In:Climate Change 2007: The PhysicalScience Basis. Contribution of WorkingGroup I to the Fourth Assessment Report ofthe Intergovernmental Panel on ClimateChange [Solomon, S., D. Qin, M. Manning,Z. Chen, M. Marquis, K.B. Averyt,M.Tignor & H.L. Miller (eds.)]. CambridgeUniversity Press, Cambridge, UnitedKingdom and New York, NY, USA.Malone, T., Davidson, M., DiGiacomo, P.,Gonçalves, E., Knap, T., Muelbert, J. H.,Parslow, J., Sweijd, N., Yanagai, T. & Yap,H. 2010. Climate Change, SustainableDevelopment and Coastal OceanInformation Needs. Procedia EnvironmentalSciences, 1: 324–341.Marinov, I., Doney, S. C. & Lima, I. D. 2010.Response of ocean phytoplanktoncommunity structure to climate change overthe 21st century: partitioning the effects ofnutrients, temperature and light.Biogeosciences 7: 3941-3959.McMullen, C. P. & Jabbour, J. 2009. ClimateChange Science Compendium 2009. UnitedNations Environment Programme, Nairobi,EarthPrint. 68pp.Meehl, G.A., Stocker, T. F., Collins, W. D.,Friedlingstein, P., Gaye, A. T., Gregory, J.M., Kitoh, A., Knutti, R., Murphy, J. M.,Noda, A., Raper, S. C. B., Watterson, I. G.,Waver, A. J. & Zhao, Z.-C. 2007. GlobalClimate Projections. In: Climate Change2007: The Physical Science Basis.Contribution of Working Group I to theFourth Assessment Report of theIntergovernmental Panel on ClimateChange [Solomon, S., D. Qin, M. Manning,Z. Chen, M. Marquis, K.B. Averyt,M.Tignor and H.L. Miller (eds.)].Cambridge University Press, Cambridge,United Kingdom and New York, NY, USA.Muehe, D. 2006. Erosão e Progradação doLitoral Brasileiro. Ministério do MeioAmbiente e Programa de Geologia eGeofísica Marina, Brasília, 476 pp.Nepstad, D., Soares, B. S, Merry, F., Lima, A.,Moutinho, P., Carter, J., Bowman, M.,Cattaneo, A., Rodrigues, H., Schwartzman,S., McGrath, D. G., Stickler, C. M.,Lubowski, R., Piris-Cabezas, P., Rivero, S.,Alencar, A., Almeida, O. & Stella, O. 2011.The End of Deforestation in the BrazilianAmazon. Science, 326: 1350-1351Neves, C. F. & Muehe, D. 2008. Vulnerabilidade,impactos e adaptação a mudanças do clima:a zona costeira. Parcerias Estratégicas(Brasília), 27: 217-295.Pan-American Pan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

VIIIM. S. COPERTINO ET ALLINicholls, R. J., Wong, P. P., Burkett, V. R.,Codignotto, J. O., Hay, J. E., McLean, R.F., Ragoonaden S. & Woodroffe, C. D.2007. Coastal systems and low-lying areas.Climate Change 2007: Impacts, Adaptationand Vulnerability. Contribution of WorkingGroup II to the Fourth Assessment Reportof the Intergovernmental Panel on ClimateChange [M.L. Parry, O.F. Canziani, J.P.Palutikof, P.J. van der Linden & C.E.Hanson (Eds.)], Cambridge UniversityPress, Cambridge, UK, 315-356.Nicholls, R. J. & Cazenave, A. 2010. Sea-LevelRise and Its Impact on Coastal Zones.Science, 328: 1517-1520.Noone, K. J.; Nobre, C. A. & Seitzinger, S. 2010.The International Geosphere-BiosphereProgramme s (IGBP) Scientific ResearchAgenda for Coping with GlobalEnvironmental Change. In: H. G. Brauch,Ú. O. Spring, C. Mesjasz, J. Grin, P.Kameri-Mbote, B. Chourou, P. Dunay, J.Birkmann. (Org.). Coping with GlobalEnvironmental Change, Disasters andSecurity: Threats, Challenges,Vulnerabilities and Risks. 1249-1256.Richardson, A. J. & Poloczanska, E. S. 2008.Under-resourced, under Threat. Science,320: 1294-1295.Russell, B. D., Thompson, J. A. I., Falkenberg, L.J. & Connell, S. D. 2009. Synergisticeffects of climate change and localstressors: CO 2 and nutrient-driven changein subtidal rocky habitats. Global ChangeBiology, 15: 2153-2162.Salazar, L. F., Nobre, C. A. & Oyama, M. D.(2007). Climate change consequences onthe biome distribution in tropical SouthAmerica. Geophys. Res. Lett., 34: L09708.Santilli, M., Moutinho, P., Schwartzman, S.,Nepstad, D., Curran, L. & Nobre, C. A.2005. Tropical Deforestation and the KyotoProtocol. Climatic Change, 71: 267-276.Schiermeier, Q. 2010. The real holes in climatechange. Nature, 463: 284-287.Stroeve, J. C., Serreze, M. C., Holland, M. M.,Kay, J. E., Malanik, J. & Barrett, A. P.2011 The Arctic’s rapidly shrinking sea icecover: a research synthesis. ClimaticChange.Trenberth, K. E., Jones, P. D., Ambenje, P. G.,Bojariu, R., Easterling, D. R., Klein Tank,A. M. G., Parker, D. E., Renwick, J. A.,Rusticucci, M., Sond, B. & Zai, P. 2007.Surface and atmospheric climate change.In: Climate Change 2007: The PhysicalScience Basis. Contribution of WorkingGroup I to the Fourth Assessment Report ofthe Intergovermmental Panel on ClimateChange [Solomon, S., D. Qin, M. Manning,Z. Chen, M. Marquis, K.B. Averyt,M.Tignor and H.L. Miller (eds.)].Cambridge University Press, Cambridge.235-336.Walther, G.-R., Post, E., Convey, P., Menzel, A.,Parmesan, C., Beebee, T. J. C., Fromentin,J.-M., Hoegh-Gudberg, O. & Bairlein, F.2002. Ecological responses to recentclimate change. Nature, 416: 389-395.Wild, C., Hoegh-Guldberg, O., Naumann, M. S.,Colombo-Pallotta, M. F., Ateweberhan, M.,Fitt, W. K., Iglesias-Prieto, R., Palmer, C.,Bythell, J. C., Ortiz, J.-C., Loya, Y. & vanWoesik, R. 2011. Climate change impedesscleractinian corals as primary reefecosystem engineers. Marine andFreshwater Research, 62: 205–215Williams, S. E., Shoo L. P., Isaac J., Hoffmann,A. A. & Langham, G. 2008. Toward anIntegrated Framework for Assessing theVulnerability of Species to ClimateChange. PLOS Biology, 6: 2621- 2626Williamson, C. E., Saros, J. E., Jasmine, E.,Vincent, W. F. & Smol, J. P. 2009. Lakesand reservoirs as sentinels, integrators, andregulators of climate change. Limnologyand Oceanography, 54: 2273-2282.Pan-American Journal of Aquatic Sciences (2010), 5(2): I-VIII

F O R E W O R DCARLOS A. E. GARCIAScientific Coordinator of Coastal Zone NetworkUniversidade Federal do Rio Grande (FURG)Rio Grande, RS, BrazilCARLOS A. NOBREScientific Coordinator of Rede CLIMA and INCT for Climate ChangeInstituto Nacional de Pesquisas Espaciais (INPE)São José dos Campos, SP, BrasilThe Intergovernmental Panel on Climate Change (IPCC) projections for this century wouldcertainly result in a suite of biophysical and socio-economic impacts on the Brazilian coastal zone thatwould ultimately affect several sectors, including natural ecosystems, fisheries, transportation, tourism andrecreation, infrastructure and local communities.Global climate change will affect coastal zones where approximately 40% of world’s populationlives within 100 km of the coastline. The anticipated climate-related changes include an accelerated sealevel rise, a further increase in sea surface temperature, an intensification of tropical and extra-tropicalcyclones, large extreme waves and storms, alterations in the precipitation and run-off rates, oceanacidification, among others. These phenomena will vary from place to place at different time scales, buttheir impacts are most certainly to be negative on coastal societies.Coastal zones will face socio-economic and environmental problems, especially in low-lyingareas, which can be severely affected by seawater intrusion and local wave climate changes. For instance,the IPCC expects that, globally, mean sea level may rise as much as 88 cm by the end of the 21st century.One clear message that emerged from the last IPCC report is the urgent need to understand the impactsand assess the vulnerability of coastal zones to climate changes around the world.Although there has been a considerable increase in the understanding of the impacts of climatechanging on coastal zones and their ecosystems, there are still substantial knowledge gaps. Thevulnerability to climate change is highly variable, depending on the region and ecosystems underinvestigation. The capacity of adaptation of marine and coastal ecosystem will vary among species,communities, geographical regions and levels of system health and degradation. Rising in sea level andincrease in both frequency and intensity of storms would amplify the impacts on the detected areas underrisk. Coastlines under stress from human activities are particularly susceptible to global warming impactsand their vulnerabilities have to be urgently assessed. In highly urbanized areas floods rather than coastalerosion can cause strong impacts. Even worse, the historic drainage problem in certain coastal cities canbe magnified due to intense flooding causing a serious public health problem with groundwatercontamination, mosquito proliferation, and associated spread of diseases.The Brazilian Climate Change Research Programs: Rede CLIMA and INCT for Climate ChangeIn late 2007, the Brazilian Ministry of Science and Technology (MCT) created the BrazilianResearch Network on Climate Change (Rede CLIMA) with the following objectives: (a) generate anddisseminate knowledge and technology for Brazil to meet the challenges represented by the causes andeffects of global climate change; (b) gather data and information necessary to support Brazilian diplomacyin negotiations on the international regime for climate change; (c) develop studies on the impacts of globaland regional climate change in Brazil, with emphasis on the country's vulnerability to climate change; (d)consider alternatives for the adaptation of Brazil’s social, economic and natural systems; (e) investigatethe effects of changes in land use, and in Brazil’s social, economic and natural systems to the country’semissions that contribute to global climate change, and (f) contribute to the formulation and monitoring ofpublic policies on global climate change within the Brazilian territory, g) contribute to the conception andPan-American Journal of Aquatic Sciences (2010), 5(2): IX-XI

XC.A.E. GARCIA & C.A. NOBREimplementation of a Brazilian climate-related disaster monitoring and alert system, h) carry out studiesregarding Brazilian greenhouse gas emissions in support to the periodic national greenhouse gasinventories stipulated by the Presidential Decree n. o 7.390 of September 9 th 2010. One of the first productsof Rede CLIMA will be to produce regular analysis of Brazil’s climate and predicted climate change, withbetter precision for South America than the models developed in other countries that are availablenowadays, and with more specific analysis in certain strategic areas, among them Coastal Zones, for theformulation of national policies.From the scientific point of view, Rede CLIMA interacts closely with the National Institute ofScience and Technology for Climate Change (INCT for Climate Change), which began in 2008, and withthe São Paulo State Research Funding Agency (FAPESP) Research Program on Global Climate Changealso established in 2008. The INCT for Climate Change is funded by Brazil’s National Council forScientific and Technological Development (CNPq) and by FAPESP. It brings together the largest andmost far-reaching interdisciplinary network of environmental research institutions in Brazil, involvingover 90 research groups from 65 institutions and universities from Brazil and abroad, with over 400participants. The main goal of the INCT for Climate Change is to provide high quality and relevantscientific information needed to (a) understand climate functioning, variability and change, and (b) informadaptation and mitigation at local, national and international levels. The INCT for Climate Change isstructured in three scientific and one technological axes: scientific basis for global environmental change;research on impacts, adaptation and vulnerability; mitigation; and technological developments andproducts. FAPESP Research Program on Global Climate Change has objectives that are similar to those ofthe INCT for Climate Change and of Rede CLIMA, with a particular emphasis in the development of newtechnologies to mitigate and adapt to climate change.The Coastal Zone NetworkWith about 8.500 km, the large Brazilian coastal zone presents a variety of climates and coastalmorphologies, including several ecosystems such as sandy beaches and dunes, rock shores, coral reefs,estuaries, mangroves, salt marshes and seagrass meadows. The Brazilian’s coastal zone is a significantnational environmental benefit that is also fundamentally important to our lifestyle and economy. About20-30% of the Brazilian population lives by the coast, where several pressures already exist such as seareclamation, flooding, erosion and extreme weather events.Climate change will certainly affect in different ways the various coastal cities and coastalecosystems in Brazil. Due to its complexity, the Brazilian coastal zone cannot be investigated under theperspective of small projects addressing only few scientific questions. Dense population occupies theregion at sea-land interface with complex infrastructure in certain areas, added to a mosaic of differentecosystems, which are exposed to anthropogenic and natural impacts.Therefore, a coordinated research team was needed to investigate vulnerability, impacts andadaptation of the Brazilian coastal zone to climate change. To address this need, the Coastal Zone networkwas created within the context of Rede CLIMA and INCT for Climate Change. The first and major effortwas made towards establishing a multidisciplinary research team, comprising both regional andinstitutional representativeness, aiming at achieving national and international scientific impact.Coordinated by the Institute of Oceanography at the Federal University of Rio Grande (IO-FURG),Coastal Zone is formed by more than fifty researchers from 23 institutes, covering about 11 Braziliancoastal states. Taking into account that the analysis of the impacts of climate change on the Braziliancoastal zones is limited by a number of deficiencies, especially by basic knowledge about the physical,geological and ecological dynamics of these environments, the main CZ goals are: 1) to evaluate the stateof knowledge and identify the gaps; 2) to recommend future studies; 3) to establish protocols and 4) tocoordinate/integrate projects that investigate the vulnerability and the effects of climate change in coastalareas of Brazil in order to propose adaptive actions with the organized sectors of society. The CZresearchers focused first on making a preliminary assessment of the studies, including reviews, analysis ofpast data and vulnerability assessment of ecosystems and regions to climate.I Brazilian Workshop on Climate Change and Coastal ZonesThe Coastal Zone netwport has planned workshops to run on biennial basis, to better integrate thefinds. The “First Brazilian Workshop on Climate Change and Coastal Zones”, one of the first CZachievements, was held from September 14 th to 16 th 2009 at the IO/FURG. The Workshop aimed thePan-American Journal of Aquatic Sciences (2010), 5(2): IX-XI

ForewordXIdivulgation of Coastal Zone preliminary results, consolidating the research group, stimulating theintegration of its members and discussing methodological protocols and future research. The Workshopcovered topics within the areas of geology, geography, physical, geological and biological oceanography,ecology, fisheries, socio-economics, scientific divulgation and education. The event was attended by 200people, including scientists and students, from several national institutions. The Workshop successfullyachieved its goals, and its results were highlighted by the national scientific community and by the mediaand society in general. By bringing together different research areas and institutions, new collaborationswere established. Moreover, the Workshop established scientific basis, new and promising collaborationsand leaded the future of climate change research in Brazil.The Workshop also resulted in this Special Issue of the Pan-American Journal of Aquatic Sciences(PanamJAS), which includes 15 peer-reviewed scientific articles covering several physical, biological andsocial aspects of the Brazilian coast. The Issue is the first of several steps toward a detailed assessment ofthe impacts and vulnerability of the Brazilian coastal zone to climate change.Based on Workshop discussions, we released the Rio Grande Declaration, an open letter signedby Coastal Zone members and meeting participants. Forwarded to the main national media groups andpublic sectors, and also published in this volume, Rio Grande Declaration is a manifest from the scientificcommunity present or represented during the Workshop, warning about climate change impacts,particularly those affecting the coast. The letter highlights a series of scientific goals that still needs to bereached to adequately assess and monitor the effects of climate change on coastal ecosystems in Brazil.The successful implementation of the recommendations depends on political will and decisions, whichmust be long-term committed to the theme of Climate Change.We hope that this Special Issue will heighten the interest of the scientific community, managersand policy-makers, because a wide range of climate changes impacts are expected to influence theBrazilian coastal zones. Human response and our scientific capacity will play a major role in determiningthe success of the adaptation of the Brazilian coastal zone to climate change.Pan-American Journal of Aquatic Sciences (2010), 5(2): IX-XI

Brazilian coastal vulnerability to climate changeDIETER MUEHEUniversidade Federal do Rio de Janeiro (UFRJ), Programa de Pós-Graduação em Geografia, Instituto deGeociências, Cidade Universitária, Ilha do Fundão, 2194-611, Rio de Janeiro, Brazil. E-mail:dieter.muehe@gmail.comAbstract. The coastal zone of Brazil includes a variety of coastal forms, such as wide occurrence ofsedimentary cliffs, large and deeply incised estuaries, crystalline headlands and low-lying coastal plains.These forms will respond in different ways to the expected climate changes and associated sea level rise.The potential vulnerability of the distinct coastal types along the Brazilian coast to climate change isevaluated. Due to the low occupation of large sectors of the coastline, the risks are concentrated in theurbanized areas. The largest impacts are expected to be caused by floods. The absence of long-termobservations of oceanographic data or detailed altimetric maps presents a major difficulty for theevaluation of different risk scenarios at the local level and consequently for the application of measures tominimize these impacts over the population.Key words: Coastal vulnerability, climate change, coastal classification, BrazilResumo: Vulnerabilidade da costa brasileira a mudanças climáticas. A zona costeira brasileira secaracteriza por amplos domínios de falésias sedimentares e de rochas do embasamento com alternânciade baixadas costeiras e margens ao longo de estuários inseridos nos relevos elevados. Considerando osdistintos compartimentos geomorfológicos é apresentada uma avaliação da vulnerabilidade potencial emface de uma mudança climática e associada elevação do nível do mar. Os resultados indicam que a baixaocupação de grande parte da zona costeira faz com que as áreas de risco se concentram nas cidadescosteiras, sendo os riscos de inundação os de maior impacto sobre a população. A ausência deobservações contínuas de longa duração, assim como a falta de mapeamentos altimétricos de detalhe,representa a maior dificuldade na construção de cenários de risco a nível local e consequentemente para odesenvolvimento e aplicação de medidas de minimização dos impactos sobre a população.Palavras-chave: Vulnerabilidade costeira, mudança climática, classificação costeira, BrasilIntroductionThe climate changes forecasted by theIntergovernmental Panel on Climate Change (IPCC)will necessarily depend on the time and place beingconsidered.One of the forecasts for the evolution of theaverage surface temperature of the planet underdifferent greenhouse gas emission scenarios (IPCC2007) is that the temperature increase along theBrazilian coast will be less than 1 °C until 2030.But, this increase may be as high as 2 °C or 3 °C forthe last decade of this century (Fig. 1). The predictedincrease in temperature at higher latitudes is moresignificant and suggests that ice will melt inGreenland and Antarctica, resulting in rising sealevel rates. Sea level rates will already be increasingbecause of thermal expansion. The most direct resultof coastal geomorphological processes will be theadjustment, usually by erosion, of the coastline andthe increased vulnerability of low-lying areas toflooding.This first approximation regarding theclimate evolution already allows us to affirm that,from a geomorphic viewpoint, the changes broughtabout by sea level rise will be small in the next 20years, and the current trends will be more or lessmaintained. However, because of the significant risein temperature forecasted for the end of the century,geomorphic processes are expected to intensify,starting in the middle of the century, as a result ofboth the sea level rise and the intensification andfrequency of subtropical cyclogenesis. This cyclogenesisis associated with an increase in the seawater temperature, changes in sea level, increasedwind transport and destabilization of dune fields.Pan-American Journal of Aquatic Sciences (2010) 5(2): 173-183

174D. MUEHEFigure 1. Evolution of the average surface temperature projected by the IPCC in 2007.Nobre et al. (2007) produced climateprojections for the next 30 years of the century inSouth America for various emission scenarios. Interms of precipitation these predictions showdiscrepancies in the forecasted anomalies in terms ofeither increase or decrease. Although the discrepanciesbetween the results are small they will affectnegatively or positively areas that are alreadyexperiencing water deficit as in the NortheastRegion of Brazil, where the projections for thecoastal zone indicate anomalies varying from verysmall to an increase and even decrease in the amountof precipitation. However, even with no significativechange in rainfall, the augmented evaporationcaused by the temperature increase will result inaugmented water shortages in the semi-arid regions.The temperature forecasts by Nobre et al.(2007) for the coastal zone are similar to thescenario forecasted by the IPCC (2007) and indicatea temperature increase in the range of 2 ºC to 3 ºC.The perception of risk from sea level risewas incorporated into the ORLA Project of theMinistry of Environment (Ministério do MeioAmbiente 2004) through the establishment of setbacklines of 50 m in urban areas and 200 m in ruralareas. This distance would be measured landwardeither from the backshore limit or from the dune toeof the reverse side of foredunes when present. Theestablishment of the width of these set-back linesresulted from a conciliation between the results ofthe application of the Bruun rule (Bruun 1962,1988), for an increase of an 1 m sea level rise(Muehe 2001, 2004), and the impossibility of itspractical application in areas of very low gradient ofthe shoreface and inner shelf as in the north andnortheast regions where the results of the modelindicate very large recession of the coastline (Fig.2).The adopted limits represent a firstminimum restriction to be considered in theurbanization of coastal areas, and can be expandedaccording to local geomorphic characteristics suchas the known erosion rate and landscapeconservation.Coastlineretreat(m)1000010001001013 o 2 o 1 o 0.6 o 0.2 o 0.06 oSteepgradientIntermediategradientLowgradient NNESSSESENEd l,1 00d l,110 100 1000 10000Shorefacegradient (1:X)Figure 2. Coastline retreat in response to an elevation of1 m of relative sea level and a closure depth of the beachprofile of 5 m (dl, 1) and 10 m (dl, 100)), (Muehe 2001).NPan-American Journal of Aquatic Sciences (2010) 5(2): 173-183

Brazilian coastal vulnerability to climate change175The coastal vulnerability investigationsconducted by research groups associated to theProgram of Marine Geology and Geophysics(PGGM) (Muehe 2006) indicate that, in general,erosion is more prevalent than accretion, which isrestricted to sites with a significant input of fluvialsediments.Coastal morphology and its response to sea leveloscillationsHorizontal adjustments of the coastline arethe result of an intricate relationship between theoscillations in the sea level, changes in waveincidence, sediment availability, coastal morphologyand rock strength. Accretion of the shoreline mayoccur locally even during a period of sea level rise,provided there is sufficient sediment supply. Suchareas include coastal plains or terraces associated toa fluvial outlet. Nevertheless, large-scale marinetransgressions cannot be accompanied by a positivesediment balance and consequently result incoastline erosion. For example, during the lastmarine transgression, the coastline receded tens ofkilometers over the course of about 10,000 years.The relative sea level elevations of theBrazilian coast 120,000 years BP and 5,600 yearsBP were on the order of 8 m and 5 m higher than thecurrent level, respectively. During these transgresssions,marine sands were deposited in the form ofbeach barriers and beach ridge coastal plains. Thehigher sea level, that the present one, of the lasttransgression resulted, according to Pirazzoli (1996,cited in Bird 2008), from the postglacial isostaticcompensation, which was limited, in Brazil, to thecoast between Cape Calcanhar and the southernextremity of South America and excluded theAmazonian Coastal Region (Fig. 3).Curves of relative sea level variation wereestablished by Martin et al. (1979, 2003 amongothers) for Salvador (Fig. 4) and several locations insouthern and southeastern Brazil. These curvesconfirm a sea level elevation of up to 5 m, despitedisagreements about the oscillations after themaximum transgression (Angulo & Lessa 1997).The development of coastal barriers duringeach of the marine transgressions led to theformation of lagoons at the lee of the barriers andthe subsequent formation of coastal plains throughsediment infilling of the lagoons. The low altitudeand topography of these plains results in difficulty ofwater discharge, amplified during storms due to theblocking of the channel outlets by waves and tides.Therefore, these plains represent an area withpotential risk of flooding due to sea level rise (Fig.5). The response of a barrier to a sea level rise is toadjust by landward translation, a process thatinvolves wave overwash, the formation of overwashfans and even localized barrier rupture. Overwashdoes not occur when the barrier was formed during aperiod of higher sea level or had their heightincreased by the development of dunes. In any casethere will be erosion on the ocean side of the barrieras also at its backside in contact of a lagoon, ifpresent.Figure 3. Relative sea level variations during the Holocene.Adapted from Pirazzoli 1996, cited in Bird 2008.Sediment accumulation may occur in theform of coastal terraces or plains developed throughaccretion of a succession of beach ridges. This typeof accumulation occurred in the delta-shaped plainsof the São Francisco, Jequitinhonha, Doce andParaíba do Sul Rivers (Fig. 6), which were studied indetail by Dominguez et al. (1983, 1987, 1989), andMartin et al. (1984). Reversals of the longshoresediment transport direction is revealed by the changingalignment of beach ridges as a consequence ofFigure 4. Relative sea level curve for Salvador (adaptedfrom Martin, 2003).Pan-American Journal of Aquatic Sciences (2010), 5(2): 173-183

176D. MUEHEFigure 5. Highly developed double coastal barriers andenclosed lagoons in Barra da Tijuca, Rio de Janeiro.Photo D. Muehe.changes in the wave incidence in response to climatechanges, and those changes are reflected inadjustments of the shoreline through the process ofbeach rotation.In areas of high relief, as represented by thesedimentary deposits of the Barreiras Group andother morphologically similar areas which occurdiscontinuously from the coast of Amapá up to Riode Janeiro, the result of transgressions were theformation of coastal cliffs, in some places stillactive, preceded by narrow beaches and a nearshoreand inner shelf of low gradient (Fig. 7). A risingsea level will increase the erosion of the cliffs, aslow process whose speed depends on the cohesionof the material and the amount of sedimentreleased and maintained in the profile. Many of thecliffs that are still active probably represent anincomplete adjustment related to the post-glacialtransgression rather than the result of a recent rise insea level.The erosion of sedimentary cliffs leftvestiges of their past position in the form of lateriticconcretions deposited on the continental shelf, andthe released sediments represented an importantsource of sand for the formation of beaches, dunesand coastal terraces.Figure 7. Active cliffs in sedimentary deposits of theBarreiras Group. Buzios, Rio Grande do Norte. Photo D.Muehe.Figure 6. Beach ridges of the Doce River coastal plain.Image from Google Earth.Coastal Compartments and AssociatedVulnerabilitiesPhysiographic coastline classifications, forthe Brazilian coastline, have been proposed byseveral authors. The most largely known wasproposed by Silveira (1964) who consideredgeological, geomorphological, climatic and oceanographicaspects. More recent studies (Muehe 1998,2005, 2006, Muehe & Neves 1995, Dominguez2004, 2007) are in many aspects coincident with thePan-American Journal of Aquatic Sciences (2010) 5(2): 173-183

Brazilian coastal vulnerability to climate change177classification scheme of Silveira but aredifferentiated by their objectives and level of detail.The following classification (Fig. 8) is basedon the literature mentioned above, with emphasisplaced on the prevailing geomorphological aspectsand potential vulnerabilities associated with climatechange and its consequences (e.g., sea level rise andflooding):Figure 8. Coastal compartments.The tide and mangrove dominated coast of theNorth regionWith a wide continental shelf, highlyinfluenced by the water discharge and mud depositionof the Amazon, the coastal zone is submitted toa macro tidal regimen with tidal ranges of locally upto 10 m associated to strong tidal currents.Mangroves occur widely and correspondabout 76% of this formation of the whole Braziliancoastline (Muehe, 1998). They represent an importantecosystem for the maintenance of fishery as alsofor the protection of the coast. A rise in sea levelmay be compensated by a backward and lateralexpansion of the mangroves, but this expansion isfrequently limited by the presence of cliffs along theopen coast and estuaries. Depending of the rate ofsea level rise the adaption may not be sustained underrates higher that 7 mm a -1 as reported by Reed etal. (2008) for the Mid-Atlantic coast of the USA.Coastal erosion has been reported by ElRobrini et al. (2006) in Mosqueiro in the estuarinecoast to the north of Belém and along the AtlanticCoast in the region of Salinópolis, one of theprincipal resorts of the coast of Pará, and nearAjuruteua, whose beaches are in high demand onweekends and holidays.According to data from the BrazilianInstitute of Geography and Statistics (IBGE)interpreted by Strohaecker (2008), the population onthe coast is relatively high in relation to thepopulation of the state (89% in Amapá, 45% in Paráand 27% in Maranhão). However, this population ismainly concentrated in metropolitan areas, while therest of the coast is almost empty. Over 80% of thecoastal municipalities of Amapá and Pará havepopulation densities of less than 50 inhabitants/km 2 .Therefore, many erosion or flood events would onlyhave localized socio-economic impacts.According to a study by Tessler (2008), theflood risk varies from very high to high in the regionaround Macapá and in the northeast area of MarajóIsland and from high to medium in the city of Belémand the coast of Bragança in Pará. The risk is highalong the estuary on the island of São Luiz in Maranhãoand medium to high in the Parnaíba Delta.The Sediment Starved Coast of the NortheastThe coast is characterized by the dominanceof sedimentary cliffs of the flat topped Barreirasgroup and can be sub-divided according the degreeof water deficit into a semi-arid compartment in thenorth, including the States of Piauí, Ceará and theWest Coast of Rio Grande do Norte and a morehumid compartment in the south extending from theSouth Coast of Rio Grande do Norte to Salvador inBahia.Large dune fields occur widely and representa sink of the sands which otherwise would leadto a progradation of the coastline. An increase intemperature or decrease in precipitation will amplifythe eolian sediment transport and increase the vulnerabilityof the coastline due to the augmented transferof sediments from the shore to the continent.Beach rocks occur widely at some distancefrom the beach and acts like breakwaters, providingsome protection against the waves but also avoidingsediments to be carried to the beach, increasing thevulnerability to erosion. The overtopping of thesebarriers under a higher sea level will increase thewave activity at the beach which will have to adjustmorphodynamically to the new level of energy.In the semi-arid sector the most impactedsegments by erosion are in Ceará, to the north of theport of Pecém, and in Fortaleza. In Pecém becauseof the deposition of sediments at the lee of the portstructure, and in Fortaleza because of the retentionand diversion of the flow of sediment to the beachesof the city after the construction of a breakwater inorder to protect the Port of Mucuripe (Morais et al.2006). In Macau and Guamaré in Rio Grande doNorte, the recession of the coastline is endangeringPan-American Journal of Aquatic Sciences (2010), 5(2): 173-183

178D. MUEHEoil pumping stations (Vital et al. 2006). Accordingto these authors the construction of structuresperpendicular to the beach in Macau, Caiçara doNorte and Touros accelerates the erosion. Segmentsunder erosion of the sedimentary cliffs of the BarreirasFormation are associated to a slow retreat of thecliffs and represent a risk for constructions whenlocated too close to the edge of the cliff. Occasionalswell from the Northern Hemisphere reach the coastof Ceará in the summer months triggering erosion(Melo Fo & Alves, 1993, Melo Fo et al. 1995). Thefrequency of these events may increase as a result ofclimate change, and will contribute to the generaltrend of coastline destabilization.The population in the states of Ceará andRio Grande do Norte is concentrated along the coastand represents almost 50% of the population of thesestates, with densities ranging from 50 to 200inhabitants/km 2 (Strohaecker 2008). However,almost 40% of this population is concentrated incoastal metropolitan areas, therefore only 10% isdistributed in the rest of the coastal municipalities.According to Tessler (2008), the areas withthe greatest risk of flooding are located in theParnaiba metropolitan area in the State of Piaui, inthe metropolitan region of Fortaleza in Ceará, andalong the margins of the Apodi and do CarmoRivers, between the localities of Areia Branca andMossoró in Rio Grande do Norte.On the coast of the sedimentary cliffs, theerosion is widespread and occurs almost along thewhole coastline from the south of Rio Grande doNorte along the coast of Paraíba, Pernambuco andAlagoas. The opposite is true on the coast ofSergipe, where the abundant amount of sedimentdelivered by the rivers causes about 57% of theshoreline to be in equilibrium, while 21% is undererosion (Bittencourt et al. 2006).In Paraíba State, the segments under erosionrepresents about 42% of the 140 km of beaches(Neves et al. 2006).In Pernambuco, about 30% of the beachesare under erosion. The majority are experiencingsevere erosion due to natural factors, such as coastalcirculation and sediment deficit, while man-madeinterventions often intensifies this trend (Neves &Muehe, 1995; Manso et al. 2006). The occurrence ofbeach rocks and algae in the shoreface andinnershelf increase the sediment deficit bypreventing their transport in direction to the shore.Because of the proximity of buildings, the mostcritical areas are the beaches of Boa Viagem,Piedade, Candeias and Barra das Jangadas in Recife.In Alagoas, the vulnerability to erosionis caused by the reduced delivery of fluvialsediments. Erosion is concentrated primarily on thenorthern coast of the state, where tourist activity isconcen-trated (Araújo et al. 2006). According toDominguez (1995), the susceptibility of the coast toerosion is demonstrated by active cliffs of theBarreiras group, the almost absence of coastal plainsand Pleistocene terraces as also the occurrence ofsubmerged beach rocks attesting the retreat of thecoastline.In Sergipe segments under erosion arelocated in Atalaia Nova, north of Aracaju, and to thesouth of the São Francisco River outlet, where theVila do Cabeço was completely eroded. Areas ofhigh shoreline variability are located near the outletsof the Real, Vaza Barris and Sergipe Rivers, whereerosive episodes have caused significant materialdamage (Bittencourt et al. 2006).In Bahia, the coast between Mangue Seco, atthe outlet of the São Francisco River and Salvador,is in equilibrium (Dominguez et al. 2006).According to Strohaecker (2008) based onIBGE data, the general trend of the State populationis to concentrate in coastal municipalities. Around40% to 50% in Rio Grande do Norte, Pernambucoand Alagoas, 30% to 40% in Sergipe and Bahia and20 to 30% in Paraíba. The majority lives inmetropolitan areas with only 10 % distributed alongthe remaining coastal municipalities.The areas under greatest risk to flooding(Tessler 2008) are located in low-laying areas ofJoão Pessoa and Recife, in the urban area ofAracaju, in the coastal plain north of the Vaza BarrisRiver and in the low-lying areas of Salvador.The Mixed Coast of Sedimentary Cliffs and WaveDominated DeltasThe presence of the sedimentary cliffs of theBarreiras Group is still dominant but less continuousin the south. Beach ridge plains developed in frontof the Jequitinhonha and Caravela Rivers, in Bahia,Doce River in Espirito Santo and Paraíba do SulRiver in Rio de Janeiro State (Fig. 6). The changesin the alignment of the beach ridges associated tomodifications in the longshore sediment transportindicate the occurrence of alternations in thedominium between waves generated by the tradewinds and swell waves generated by cold frontsfrom the south. Therefore the compartment islocated in a region highly susceptible tomodifications in the dominium between tropical andsubtropical climatic-oceanographic processes.In Bahia, about 60% of the coast is inequilibrium, and 26% is under erosion, with intenseerosion occurring in fluvial outlets. Sedimentretention occur in Ilhéus and in unconsolidated capesPan-American Journal of Aquatic Sciences (2010) 5(2): 173-183

Brazilian coastal vulnerability to climate change179such as the coastal plain of Caravelas. Longstretches of the cliffs in south Bahia, fromCumuruxatiba to the border of Espírito Santo State,are experiencing long-term negative sedimentbudget (Dominguez et al. 2006). Strong coastalerosion with destruction of houses occurs in Mucuri,in south Bahia near the border with Espírito Santo.In Espírito Santo the coastline alternatesbetween long stretches under erosion or inequilibrium with few segments of accretion.Accretion is occurring in the coastal plains of DoceRiver, in the north, and Itabapoana River, at thesouthern border between the States of Espírito Santoand Rio de Janeiro (Albino et al. 2006).On the coast of Rio de Janeiro near theborder between Espírito Santo and Cape Frio,significant erosion is occurring on the southmargin at the outlet of the Paraiba do Sul Riverin Atafona where the sands are being trapped onthe inner continental shelf by mud brought by theriver and by the predominance of longitudinaltransport of sediments towards the south, out ofthe affected area (Muehe et al. 2006). Other areas atrisk of erosion in highly urbanized areas includethe coast of Macaé and Rio das Ostras (Muehe et al.2006).The population density of this compartmentis low except along the northern coast of Rio deJaneiro, where oil exploration on the continentalshelf led to migration to the cities of Macaé andCampos. Other cities with strong growth due totourism are Armação dos Búzios, Rio das Ostras andCabo Frio (Strohaecker 2008).According to Tessler (2008), the criticalareas at risk of flooding are the cities of Valença andIlheus in Bahia, São Mateus and Vitória in EspíritoSanto and Campos dos Goytacazes, Macaé and CaboFrio in Rio de Janeiro.The Double Barrier-Lagoon CoastThis compartment due to its almost eastwestalignment of the coastline is highly exposed tostorm waves from the south. The longshore sedimenttransport tends to be in equilibrium along a year,with the less frequent high energy waves (swell)from south and southwest being compensated by themore frequent waves from the southeast. Seasonallythis equilibrium is frequently disrupted by thepredominance of one of these wave incidencesresulting in short term beach rotation with erosion atone of the beach extremities and accumulation at theother as in the Ipanema-Leblon Beach in theMetropolitan area of Rio de Janeiro.From Cape Frio to the Marambaia Island,the coastline of this compartment shows signs ofinstability, with wave overwash and backshore scarpretreat (Muehe et al. 2006). Backshore retreat on theorder of 10 to 15 m were recorded in several placesand resulted largely from an exceptional storm thatoccurred in May 2001. This storm destroyed houses,kiosks and streets, mainly in the municipalities ofMaricá and Saquarema. Erosion also occurs at thelagoon side of the more landward locatedPleistocene barrier in contact with the Araruamalagoon, a long lasting geomorphological process thathas gradually reduced the width of this barrierassociated to an increase in the width of the lagoon.The extreme west of the compartment ischaracterized be a long and narrow barrierseparating the large Sepetiba Bay from the ocean.Localized overwash and gradual erosion of thelagoon shore of the barrier may result in temporarydisruption of the barrier during exceptional stormsand sea level rise. This disruption, in turn, will leadto the propagation of waves into the bay, withpossible effects on the port of Sepetiba.The occupation of the coastal barriers areconcentrated in the towns of Maricá, Saquarema,Figueira and Monte Alto and in the metropolitanarea of Rio de Janeiro. The expansion of theurbanization in these areas is moving very close tothe beach, increasing their vulnerability to erosion.In the metropolitan area of Rio de Janeiro,which includes the coast of Niterói, the increasedpopulation density makes the oceanic and estuarinecoast more vulnerable against erosion, floodingand landslides. The expansion of urbanization overlow lying areas of previously existing lagoons (e.g.,Barra da Tijuca) with limited drainage capacityrepresents risks that will significantly increase undera raised sea level and increased storm activity(Muehe & Neves 2008). Other critical areasidentified in the municipality of Rio de Janeiro arelocated next to the Meriti and Pavuna Rivers and inthe Sepetiba Bay.The Rocky Coast of the SoutheastThis compartment, which extends from IlhaGrande Bay in Rio de Janeiro to the Cape SantaMarta in Santa Catarina is characterized by theproximity of the Serra do Mar mountain range whichextends up to the coastline between Ilha Grande Bayin Rio de Janeiro and São Vicente in São Pauloresulting in a drowned landscape with a sequence ofhigh cliffs of Precambrian rocks and small coves.Coastal plains are small and sometimes absent. FromSão Vicente to the North of Santa Catarina, thereforeincluding the coast of Paraná, the coastline is formedby long beaches and wide coastal plains withimportant estuaries as in Santos and Cananéia in SãoPan-American Journal of Aquatic Sciences (2010), 5(2): 173-183

180D. MUEHEPaulo, Paranaguá and Guaratuba in Paraná and SãoFrancisco do Sul in Santa Catarina. From the northof Santa Catarina to the south of Santa CatarinaIsland the coastline becomes irregular with outcropsof the crystalline basement and small coastal plains.Southward of Santa Catarina Island and cape SantaMarta, in Santa Catarina, the coastline is formed bya sequence of beaches limited by rocky promontorieswith wide coastal plains and lagoons.Longshore sediment transport tends to bedirected to the north. The occurrence of extratropicalcyclones with strong winds, heavy rain andhigh waves has been a main threat and seems toincrease in frequency. Associated to an increase intemperature of the ocean water their occurrence willincrease and may also affect the coastline up to Riode Janeiro.Modifications of the coastline due to erosionare, in São Paulo, usually isolated and associatedwith natural or artificial obstacles that interrupt theflow of sediments (Tessler et al. 2006).In Paraná, the most significant modifycationsof the coastline occur on the estuarine outlets(e.g., the Superagui channel, Peças island, Mel island,Pontal do Sul, Ponta de Caiobá and Guaratuba).These modifications include both retreat andadvance of the shoreline and occurred on the orderof hundreds of meters in less than a decade (Anguloet al. 2006). The ocean coastline is presently stable.Areas most impacted by erosion are the beach resortsof Flamengo and Riviera and the central beachof Matinhos, restored through beach nourishment.In Santa Catarina, investigations were concentratedon the central north coast (Klein et al.2006) and on the island of Santa Catarina (Horn2006). On the continental coast, the risks associatedwith coastal erosion result from inappropriate landuse and frequent storms. The most critical points arelocated in Barra Velha, Piçarras and Penha. These a-reas are experiencing medium-intensity erosion, andBombinhas is experiencing low-intensity erosion.On the island of Santa Catarina, erosion is occurringthroughout the ocean coast. The greatest risk is tothe urban areas on the north coast of the island (e.g.,beaches of Canasvieiras, Cachoeira and the Ingleses)and on the northwest coast in the Barra da Lagoa.Urbanized areas on the east and south coastwith medium to high risk of erosion include Campeche,Armação and Pântano do Sul (Horn 2006).Areas with the greatest risk of flooding areidentified by Tessler (2008) and include theestuarine region of São Vicente and Santos in SãoPaulo, Paranaguá at the Bay of Paranaguá in Paraná,and in Santa Catarina at the southern shore of theBabitonga bay, at the estuary of the Itajaí-Açu Riverand at the localities of Palhoça and São José on thewest shore of the South Bay and Florianópolis at themargins of the North and South Bay.Strohaecker (2008) call attention to the highrate of population growth in the urban areas extendinghundreds of kilometers along the coastline.The Sandy Coast of Multiple Barriers of RioGrande do SulFrom cape Santa Marta to Chui, at theborder between Brazil and Uruguay, the coastline isformed by a long, wide, fine grained and monotonousbeach in front of a multiple barrier-lagoonsystem, with the widest lagoons represented by thePatos and Mirim Lagoons. Active dune fields developon top of the coastal barriers with dominant sandtransport to southwest. Storm surges are frequentlysubmitting the shoreline to a harsh wave climate.The beach shows a high morphodynamicmobility alternating between long stretches of retreatand advance (Toldo et al. 2006) and reversal of thistrend over time (Esteves 2008). This mobility hasbeen in most cases limited to the beach without adefinite retreat of the backshore. Very localizedsegments of coastal erosion were described byCalliari et al. (1998) and Speranski & Calliari (2006)and were related to wave convergence in Mostardas,to the south of the Mostardas lighthouse, betweenBojurú and Estreito and at a small segment nearCassino and in the far south near Chuí.The distribution of population along thecoast is low and mostly concentrated in urbancenters of second homes that attract nearly 100,000visitors during the summer. The main urban center islocated at the estuary near the mouth of the Lagoados Patos in Rio Grande. With about 200,000inhabitants the city is located in low laying areas ofthe coastal plain and present the highest risk offlooding of the entire Brazilian coast (Tessler 2008).The port of Rio Grande is one of the most importantin the country because its depth, its favorablelocation in relation to MERCOSUL countries andthe presence of important industrial and petrochemicalcomplexes.Final ConsiderationsThe Brazilian shoreline is experiencingerosion along the entire coast, but the erosion isirregularly distributed and often associated withriver outlets. Large segments of the coast are formedby sedimentary cliffs in areas of low occupationwhere erosion is slow. On beaches the erosionbecomes a risk when buildings are constructed toclose to the shore. Numerous low-lying coastalplains formed by the sedimentary fill of old lagoonsPan-American Journal of Aquatic Sciences (2010) 5(2): 173-183

Brazilian coastal vulnerability to climate change181and estuaries are very susceptible to the effects offlooding and represent a risk to urban areas.The already detected areas under risk willbe magnified by rising sea levels and the increase infrequency and intensity of storms associated withan elevation in ocean temperature. These risks willbe most significant in urban areas and especially inlarge coastal cities. In general flooding presents agreater risk than coastal erosion. Areas susceptibleto flooding already have drainage problems, whichwill become more critical with a rise in sea levelwhich in turn leads to groundwater contaminationand the spread of diseases both through water asalso due to the proliferation of mosquitoes andother transmittting agents. The implementation ofappropriate actions by the various levels of governmentare difficult to make because of the uncertaintyof the timing and magnitude of climate changes, thelack of observations of long-term temporaloceanographic variables and the absence of a wellestablished observational network, the absence ofdetailed altimetric surveys required to model andidentify the areas at greatest risk as also the smalltime window of the administrative mandate of eachgovernment in relation to the time required for theclimate change to become effective. Nevertheless,the potential risk of erosion and flooding that willresult from the expected climate change issignificant. However, the situation in Brazil is lesscritical than in many countries in terms of themagnitude of potential impacts to the population dueto the reduced occupation of large parts of thecoastal area. Nevertheless, the establishment ofintegrated networks of continuous monitoring ofoceanographic and climatic variables as also of thegeomorphologic changes in response to coastalprocesses is crucial in order to build up convincingevidence of the direction and intensity of climaticoceanographicchanges which will give thejustifycation to formulate appropriate actions forcoastal management at the different levels ofgovernment.ReferencesAlbino, J., Girardi, G. & Nascimento, K. A. do.(2006) Espírito Santo. 2006. Pp. 227-264. In:Muehe, D. (Ed.). Erosão e progradação dolitoral brasileiro. Ministério do MeioAmbiente, Brasília, 475 p.Angulo, R. J. & Lessa, G. C. 1997. The Braziliansea-level curves: a critical review withemphasis on the curve from Paranaguá andCananéia region. Marine Geology, 140: 141-166.Angulo, R. J., Soares, C. R., Marone, E., Souza, M.C. de, Odreski, L. L. R. & Noernberg, M. A.2006. Paraná. Pp. 347-400. In: Muehe, D.(Ed.). Erosão e progradação do litoralbrasileiro. Ministério do Meio Ambiente,Brasília, 475 p.Araújo, T. C. M. de, Santos, R. C. de A. L., Seoane,J. C. S. & Manso, V. do A. V. 2006. Alagoas.Pp. 197-212. In: Muehe, D. (Ed.). Erosão eprogradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília, 475 p.Bird, E. 2008. Coastal Geomorphology, anintroduction. Wiley, Chichester.Bittencourt, A. C. S. P., Oliveira, M. B. de &Dominguez, J. M. L. 2006. Sergipe. Pp. 131-154. In: Muehe, D. (Ed.) Erosão e progradaçãodo litoral brasileiro. Ministério doMeio Ambiente, Brasília, 475 p.Bruun, P. 1962. Sea level rise as a cause of shore e-rosion. Journal of Waterway, Port, Coastaland Ocean Enginering, ASCE, 88, 117-130.Bruun, P. 1988. The Bruun rule of erosion by sealevelrise: a discussion on large-scale two- andthree - dimensional usages. Journal ofCoastal Research, 4(4): 627-648.Calliari, L., Speranski, N. & Boukareva, I., 1998.Stable focus of wave rays as a reason oflocal erosion at the southern Braziliancoast. Journal of Coastal Research, SI 26:19-23.Dominguez, J. M. L. 1989. Ontogeny of astrandplain - evolving concepts on theevolution on the Doce River beach ridge plain(east coast of Brazil). InternationalSymposium on Global Changes in SouthAmerica during the Quaternary: Past-Present-Future, Extended abstracts. Spec.Publ. 1: 235-240.Dominguez, J. M. L. 1995. Regional assessment ofshort and long term trends of coastal erosionin northeastern Brazil. 1995 LOICZ - Land-Ocean Interactions in the Coastal Zone, SãoPaulo, 8-10.Dominguez, J. M. L. 2004. The coastal zone ofBrazil: an overview. Journal of CoastalResearch, SI 39, 16-20.Dominguez, J. M. L. 2007. As Costas do Brasil -proposição de uma nova tipologia. XICongresso da ABEQUA, Belém, 4-11 denovember 2007.Dominguez, J. M. L., Bittencourt, A. C. S. P. &Martin, L. 1983. O papel da deriva litorâneade sedimentos arenosos na construção dasplanícies costeiras associadas às desem-Pan-American Journal of Aquatic Sciences (2010), 5(2): 173-183

182D. MUEHEbocaduras dos rios São Francisco, Jequitinhonha,Doce e Paraíba do Sul. RevistaBrasileira de Geociências, 13(2): 98-105.Dominguez, J. M. L., Bitencourt, A. C. da S. P.,Santos, A. N., Andrade, A. C. da S.,Lavenere-Wanderley, A. A. de O., Silva, I. R.,Queiroz, I. G. de, Freitas, L. M. B. de,nascimento, L. & Silva, R. P. da. 2006. Bahia.Pp. 219-2006. In: Muehe, D. (Ed.). Erosão eprogradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília, 475 p.Dominguez, J. M. L., Martin, L. & Bittencourt, A.C. S. P. 1987. Sea level history andQuaternary evolution of river mouthassociated beach ridge plains along the eastsoutheastBrazilian coast: A summary. TheSociety of Economic Paleontologists andMineralogists, Sp. Publ., 41: 115-127.El-Robrini, M., Silva, M. A. M. A. da, Souza Filho,P. W. M. E, El-Robrini, M. H. S., SilvaJúnior, O. G. da & França, C. F. de. 2006.Pará. Pp. 41-86. In: Muehe, D. (Ed.). Erosãoe progradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília, 475 p.Esteves, L. S. 2006. Rio Grande do Sul -Variabilidade espaço temporal. Pp. 460-467.In: Muehe, D. (Ed.). Erosão e progradaçãodo litoral brasileiro. Ministério do MeioAmbiente, Brasília, 475 p.Horn, N. 2006. Santa Cataria-Ilha de Santa Catarina.Pp. 413-436. In: Muehe, D. (Ed.). Erosão eprogradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília, 475 p.IPCC, 2007. Intergovernmental Panel on ClimateChange. Working Group 1, ARA 4.Klein, A. H. da F., Menezes, J. T. de, Diehl, F. L.,Abreu, J. G. N. de, Polette, M., Sperb, R. M.& Sperb, R.C. 2006. Santa Catarina - LitoralCentro-Norte. Pp. 401-412. In: Muehe, D.(Ed.). Erosão e progradação do litoralbrasileiro. Ministério do Meio Ambiente,Brasília, 475 p.Manso, V. do A. V., Coutinho, P. da N., Guerra,N.C. & Soares Junior 2006. Pernambuco. Pp.179-196. In: Muehe, D. (Ed.). Erosão eprogradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília, 475 p.Martin, L. 2003. Holocene sea-level history alongeastern-southeastern Brazil. Anuário doInstituto de Geociências – UFRJ, 1: 13-24.Martin, L., Dominguez, J. M. L. & Bittencourt, A.C. S. P. 2003. Fluctuating Holocene sea levelin Eastern and Southeastern Brazil: evidencefrom multiple fossil and geometric indicators.Journal of Coastal Research, 19: 101-124.Martin, L., Flexor, J. M., Bittencourt, A. C. S. P. &Dominguez, J. M. L. 1984. Registro dobloqueio de circulação atomosféricameridiana na geometria dos cordões litorâneosda costa brasileira. 33Congresso Brasileirode Geologia, Sociedade Brasileira deGeologia, Rio de Janeiro, 1: 133-144.Martin, L., Flexor, J. M., Vilas Boas, G. S. &Bittencourt, A. C. S. P. 1978. Curve ofvariation of the sea-level during the last 7000years in a homogeneous section of theBrazilian coast (N of Salvador). InternationalSymposium on Coastal Evolution in theQuaternary. Special Publication No. 3 SãoPaulo, INQUA-SBG, 39-40.Melo Fo, E. & Alves, J. H. G. M. 1993. Nota sobre achegada de ondulações longínquas à costabrasileira. X Simpósio Brasileiro deRecursos Hídricos, Associação Brasileira deRecursos Hídricos, Gramado, RS, Brazil, 5:362-369.Melo Fo, E., Alves, J. H. G. M., Jorden, V., Zago, F.& Dias, W. S. S. 1995. Instrumentalconfirmation of the arrival of North Allanticswell to the Ceará coast. Proceeeding of the4th International Conference on Coastal andPort Engineering in Developing Countries –COPPEDEC IV, Rio de Janeiro, Brazil,1984-1996.Ministério do Meio Ambiente. 2004. Projeto Orla -subsídios para um projeto de gestão.Brasília, 104 p.Morais, J., Freire, G. S. S., Pinheiro, L. de S., Souza.M. J. N. de, Carvalho, A. M. de, Pessoa, P. R.S. & Oliveira, S. H. M. 2006. Ceará. Pp. 131-154. In: Muehe, D. (Ed.). Erosão eprogradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília, 475 p.Muehe, D. 1998. O litoral brasileiro e suacompartimentação. Pp. 273-349. In: Cunha, S.B. & Guerra, A. J. T. (Eds.). Geomorfologiado Brasil. Editora Bertrand Brasil S.A., Riode Janeiro, 388 p.Muehe, D. 2001. Critérios morfodinâmicos para oestabelecimento de limites da orla costeirapara fins de gerenciamento. RevistaBrasileira de Geomorfologia, 2(1): 35-44.Muehe, D. 2004. Definição de limites e tipologias daorla sob os aspectos morfodinâmico eevolutivo. In: Ministério do Meio Ambiente.2004. Projeto Orla - subsídios para umprojeto de gestão. Brasília, 13-32.Muehe, D. 2005. Aspecctos gerais da erosão costeiraPan-American Journal of Aquatic Sciences (2010) 5(2): 173-183

Brazilian coastal vulnerability to climate change183no Brasil. Mercator, 4(7): 97-110.Muehe, D. 2006. Erosion in the Brazilian coastalzone: an overview. Journal of CoastalResearch, SI 39, 43-48.Muehe, D. 2006. (Ed.). Erosão e Progradação doLitoral Brasileiro. Ministério do MeioAmbiente, Brasília, 476 p.Muehe, D., Lima, C. F. & Lins-de-Barros, F. M.2006. Rio de Janeiro. Pp. 265-296. In: Muehe,D. (Ed.). Erosão e progradação do litoralbrasileiro. Ministério do Meio Ambiente,Brasília, 475 p.Muehe, D. & Neves, C. F. 1995. The implication ofsea level rise on the Brazilian coast: apreliminary assessment. Journal of CoastalResearch, SI: 14, 54-78.Muehe D. & Neves, C. F. 2008. Vulnerabilidadesfísicas da orla. Pp. 59-79. In: Gusmão, P.P.,Carmo, P. S. do & Vianna, S. B. (Eds.). Riopróximos 100 anos - o aquecimento global ea cidade. Instituto Municipal Pereira Passos(IPP), Rio de Janeiro, 229 p.Neves, S. M., Dominguez, J. M. L. & Bittencourt, A.C. da S. P. 2006. Paraíba. Pp. 173-178. In:Muehe, D. (Ed.). Erosão e progradação dolitoral brasileiro. Ministério do MeioAmbiente, Brasília, 475 p.Neves, C. F. & Muehe, D. 1995. Potential impacts ofsea-level rise on the metropolitan region ofRecife, Brazil. Journal of Coastal Research,SI, 14: 116-131.Nittrouer, C. A., Kuehl, S. A., Figueiredo, A. G.,Allison, M. A., Sommerfield, C. K., Rine, J.M., Faria, L. E. C. & Silveira, O. M. 1996.The geological record preserved in Amazonshelf sedimentation. Continental Shelf,16(5/6): 817-841.Nobre, C. A., Salazar, L. F., Oyama, M., Cardoso,M., Sampaio, G. & Lapola, D. 2007.Mudanças climáticas e posíveis alteraçõesnos biomas da América do Sul. Ministériodo Meio Ambiente - MMA, Secretaria deBiodiversidade e Florestas - SBF, Diretoria deConservação da Biodiversidade - DCBio.Relatório No 6. 25 p.Pirazzoli, P. A. 1996. Sea-Level changes: the last20,000 years. Wiley, Chichester.Reed, D. J., Bishara, D. A., Cahoon, D.R.,Donnelly, J., Kearney, M., Kolker, A. S.,Leonard, L. L., Orson, R. A. & Stevenson, J.C. 2008. Site-Specific Scenarios for WetlandsAccretion as Sea Level Rises in the Mid-Atlantic Region. Section 2.1 In: Titus, J. G. &Strange, E. M. (Eds.). BackgroundDocuments Supporting Climate ChangeScience Program Synthesis and AssessmentProduct 4.1, EPA 430R07004, U.S. EPA,Washington, DC.Silveira, J. D. 1964. Morfologia do litoral. In:Azevedo, A. de, (Ed.), Brasil a Terra e oHomem. Companhia Editora Nacional, SãoPaulo, Brasil, pp. 253-305.Speranski, N. S. & Calliari, L. J. 2006. Rio Grandedo Sul - Padrões de refração de ondas para acosta do Rio Grande do Sul e sua relaçãocoma erosão costeira. Pp. 446-454. In:Muehe, D. (Ed.). Erosão e progradação dolitoral brasileiro. Ministério do MeioAmbiente, Brasília. 475 p.Strohaecker, T. M. 2008. Dinâmica populacional.Pp. 59-92 In: Muehe, D. (Ed.). Erosão eprogradação do litoral brasileiro. Ministériodo Meio Ambiente, Brasília. 242 p.Suguio, K., Martin, L. Bittencourt, A. C. S. P.,Dominguez, J. M. L., Flexor, J. M. &Azevedo, A. E. G. 1985. Flutuações do nívelrelativo do mar durante o QuaternárioSuperior ao longo do litoral brasileiro e suasimplicações na sedimentação costeira. RevistaBrasileira de Geociências, 15(4): 273-286.Tessler, M. 2008. Potencial de risco natural. Pp. 93-120. In: Macrodiagnóstico da zona costeirae marinha do Brasil. Ministério do MeioAmbiente, Brasília, 242 p.Toldo Jr., E. E., Almeida, L. E. S. B., Nicolodi, J. L.& Martins, L. R. 2006. Rio Grande do Sul -Erosão e acresção da zona costeira. Pp. 468-475. In: Muehe, D. (Ed.). Erosão e progradaçãodo litoral brasileiro. Ministério doMeio Ambiente, Brasília, 475 p.Vital, H. 2006. Rio Grande do Norte. Pp. 159-176.In: Muehe, D. (Ed.). Erosão e progradaçãodo litoral brasileiro. Ministério do MeioAmbiente, Brasília, 475 p.Received April 2010Accepted September 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 173-183

Potential vulnerability of the Brazilian coastal zone in itsenvironmental, social, and technological aspectsJOÃO LUIZ NICOLODI 1 & RAFAEL MUELLER PETERMANN 21 Instituto de Oceanografia, Universidade Federal do Rio Grande (FURG), Av Itália km 8, Rio Grande, RS, Brazil, CEP96201-900. E-mail: jlnicolodi@yahoo.com.br2 Datageo, Rua Valdevino V. Cordeiro 302, apto. 102, Ressaca, Itajaí, SC, Brazil, CEP 88307-370.Abstract. Climate change caused by human action can be considered a major challenge facinghuman kind in the 21 st century. Its potential to cause economic and social impacts is considerable,as it directly affects standards of living of coastal populations. This challenge can only beovercome through integrated actions taken by various sectors of society and supported by a deepknowledge of current and expected scenarios. This paper is a contribution to this knowledge, as itdefines the vulnerability level of the Brazilian coastal zone based on a combination ofenvironmental, social, and technological standards set forth in Macrodiagnóstico da ZonaCosteira e Marinha (Macrodiagnosis of the Coastal and Marine Zone) by the Ministry of theEnvironment in 2008. Low-lying, densely populated, socially underprivileged regions withintricate technological networks are the most vulnerable and require a prioritized integrated actionfrom policymakers. Throughout the entire country, several areas were rated as vulnerable orhighly vulnerable, particularly the metropolitan regions of Belém, capitals of the Northeast, Rio deJaneiro, and Santos. Its potential to cause economic and social impacts is considerable, as itdirectly affects standards of living of coastal populations. This challenge can only be overcomethrough integrated actions taken by various sectors of society and supported by a deep knowledgeof current and expected scenarios.Key words: climate changes, natural risk, social risk, technological riskResumo. Vulnerabilidade potencial das zonas costeiras brasileiras em seus aspectosambientais, sociais e tecnológicos. Considerado um dos grandes desafios a serem enfrentadospela humanidade no Século XXI, a resposta das sociedades aos efeitos de alterações nos padrõesclimáticos é fundamental no planejamento territorial, principalmente no que diz respeito às zonascosteiras. Mesmo alterações de pequena intensidade possuem potencial para causar impactoseconômicos e sociais consideráveis, com efeito direto na qualidade de vida das populaçõescosteiras. Este desafio somente poderá ser enfrentado a partir de ações integradas entre os diversossetores da sociedade e fundamentado no conhecimento profundo dos cenários atuais e previstos. Opresente artigo apresenta uma contribuição a este conhecimento, definindo o grau devulnerabilidade da zona costeira brasileira (em escala da União), com base em uma combinação decritérios ambientais, sociais e tecnológicos, definidos quando da publicação do Macrodiagnósticoda Zona Costeira e Marinha por parte do Ministério do Meio Ambiente em 2008. Regiões de baixaaltitude, densamente povoadas, socialmente carentes e com intrincadas redes tecnológicas são asmais vulneráveis e que demandam prioridade de ação integrada por parte dos tomadores dedecisão. Ao longo de todo o país diversas áreas foram classificadas com grau alto ou muito alto devulnerabilidade, com destaque para as regiões metropolitanas de Belém, capitais dos estados daregião nordeste, Rio de Janeiro e Santos.Palavras chave: mudanças climáticas, risco natural, risco social, risco tecnológicoIntroductionCoastal zones, in their seemingly simplelandscapes and usual dynamics, require at least thesame - or possibly a more complex - level ofconsideration as inland spaces, for they involveissues related to changing sea level, paleoclimate,and vegetation history. In other words, the coast, likemany other areas with ecological landscapes, mayalways be considered a heritage from earlierPan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone185processes, redesigned by the now-prevailing coastaldynamics. Therefore, one can say coastal areas arethree-way contact zones: land, sea, and climate, inaddition to the remarkable showcases of individualecosystems found in the land/sea mosaic comprisingthe total coastal space (Ab’Saber 2000).The Brazilian coast, 8,698 kilometers long,and covering some 514,000 square kilometers, is aperennial challenge to management, due to themultiplicity of situations existing in such territory 1 .There are approximately 300 coastal municipalitiesfacing the ocean, with privileged beaches for thedevelopment of tourist activities including leisure,fishing and many others. This dynamic landscape offast physical and socio-economic changes is home toapproximately 18% of the country’s population,inasmuch as 16 out of 28 metropolitan regions liealong the coast. These densely populated areascoexist alongside large, sparsely populated areas.These are the areas of small-scale commercial orsubsistence fishing communities, descended fromquilombolas [dwellers of communities of descendantsof fugitive African slaves], indigenous tribes,and other groups living in their traditional lifestyles.Considering the high level of preservation of theirecosystems, these areas will be the most relevant forpreventive environmental planning.In addition to the familiar environmentalissues affecting this part of Brazil’s territory,particularly with regard to causes and effects, therearises a new potential development in the shape ofclimate change. The need to adapt to this newdevelopment and mitigate the problems it has causedshould figure prominently on the agenda oflegislators and decision makers.Within this context, it is important tounderstand the interactions between oceans andcoastal zones and the climate change-relatedvariables. Moreover, it is essential to build astrategic vision of this part of the territory, so thatsteps may be taken in response to new scenarios ofglobal warming, rising sea levels and coastalerosion.UNESCO, through its IntergovernmentalOceanographic Commission (IOC) has beenconcentrating efforts to establish methodologies tohelp Member States in the difficult task ofidentifying hazards brought about by climate change1An extension value which takes into account theirregular coastline forming bays and recesses, amongother landforms. Of its 514,000 km 2 area, some 450,000km 2 cover 17 coastal states, and 395 municipalities,including inner water surfaces – while the rest consists ofBrazil’s Territorial Sea (MDZCM 2008).in coastal zones and making adaptations and actingto mitigate its undesirable effects.That has been a priority among IOC’sinitiatives, after the disaster of December 2004,when a tsunami hit several countries along theIndian Ocean. Together with the WorldMeteorological Organization (WMO), IOC isstarting to develop an initial multi-hazard warningsystem to guide governments in their decisions,especially with regard to integrated coastalmanagement (IOC 2009).In Brazil, efforts to build a technical andinstitutional structure able to withstand the effects ofclimate change are just getting underway. A recentstudy by TCU (the Federal Audit Office, similar tothe GAO in the USA) titled Auditorias de naturezaoperacional sobre políticas públicas e mudançasclimáticas (Operating audits on public policy andclimate change) has concluded that the country lacksa national-scale study on the vulnerability of itscoastline to climate change impacts (TCU 2009).TCU emphasizes that, among the fewexisting Brazilian coastal vulnerability studies, ahighlight is Macrodiagnóstico da Zona Costeira eMarinha (MDZCM), an instrument set forth by Law7661/88, which established the National CoastalManagement Plan.The MDZCM diagnosed the main aspects onthe Coastal and Marine Zone, mostly the changes inthe energy policy, which led to a considerableincrease in oil drilling, development, and extractionin this part of the territory, particularly after the statemonopoly was broken up. The current and potentialdimensions of the urban-manufacturing facilities andtheir interaction with other activities also went intothis diagnosis, which included information oninfrastructure, household and industrial wastewater,and toxic elements present in coastal municipalities,among others. The sources are identified bygeographic type of receiving bodies (estuaries, bays,beaches, etc).By combining a broad array of information,environmental hazard figures were generated which,in turn, measure threats to the living standards ofCoastal and Marine Zone populations. Locationswith a flooding potential, social risk potential, andtechnological risk potential could thus be identified(Nicolodi & Zamboni 2008).This paper attempts to identify, based ondata generated by MDZCM, the regions in theBrazilian coastal zone most vulnerable to the effectsof climate change, and thereby provide support for athorough assessment of the country’s vulnerability,and help fill the gaps identified by the Federal AuditOffice.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

186J. L. NICOLODI & R. M. PETERMANNVulnerability Analysis and the EnvironmentalRisk ConceptThe concept of risk is usually associatedwith an event which may or may not happen.However, the actual risk only occurs when assets arevaluable, whether materially or not, since there is norisk if the perception of losing something does notexist. Therefore, one cannot envision risk if there isno danger of losing something. In this case, societyfaces a risk.The notion of “possible loss”, which isintrinsic to risk, can be broken down into severalcomponents. When we examine spatial location, oreven spatial distribution of hazards, the connectionwith cities – or more precisely, urban centers –becomes more evident. This is because they are thespecific site of production and reproduction ofmanufacturing processes and a lifestyle which favorspopulation concentration, encourages manufacturingoutput, business relationships, and service provision(Castro et al. 2005).In this sense, risk assessment is based on therelationship between reliability and criticality ofcomplex systems, where the dynamic behavior ofnumerous variables must be captured in a select setof indicators capable of monitoring the interactionsthat actually occur along different time scales, i.e., inthe near, medium, and long term (Egler 2005).Environmental risk analysis must be seen asa dynamic indicator of relationships between naturalsystems, the productive structure, and the socialconditions of human reproduction at a given placeand time. It is therefore important to consider theassessment of environmental hazards as theconsequence of three basic categories:a) Natural Risk: related to processes andevents of a natural origin, or resulting from humanactivities. The nature of these processes is quitediverse on time and spatial scales, so the natural riskmay present differing levels of loss, as a result ofintensity (magnitude), spatial extent, and time ofactivity of the processes under consideration.b) Technological Risk: The technologicalrisk is inherent in productive processes and manufacturingactivities. The idea of technological dangerderives chiefly from manufacturing technology, as aresult of inherent flaws, as opposed to naturaldangers, perceived as an external threat (Castro et al.2005). Technological risk may be defined as apotential event that can be life-threatening in thenear, medium, and long term, as a result of investmentdecisions in the manufacturing structure.c) Social Risk: This category can beanalyzed and developed from different standpoints.It is often considered as the damage society (or partof it) can bring about. Another approach stresses therelationship between deprivation and vulnerability tonatural disasters. For the purposes of this study, wehave adopted the bias proposed by Egler (1996),where Social Risk is seen as the result of deprivationof social requirements for full human development,a fact that contributes to deterioration in standards ofliving. Its most obvious consequences are the lack ofadequate living conditions, expressed in terms ofaccess to basic services such as treated water,wastewater, and trash collection services. In the longterm, however, these can affect employability,income, and technical development of the localpopulation, as key elements to a full, sustainable,human development.Taking these three basic dimensions as astarting point for a broader concept of environmentalrisk, a methodology for its evaluation must build onthree basic criteria (Egler 1996):a) Vulnerability of natural systems, seen asthe level between the stability of biophysicalprocesses and unstable situations where there aresubstantial losses of primary productivity;b) Density and potential expansion of theproductive structure, which attempts to express fixedand flowing economic aspects in a certain area of thecountry in a dynamic concept;c) Criticality of housing conditions, in termsof the gap between current standards of living andthe minimum required for full human development.These definitions are in agreement withUNESCO’s IOC, which defines coastal vulnerabilityas the state of coastal communities (including theirsocial structure, physical assets, economy, andenvironmental support) that determine which areaffected to a greater or lesser extent by extremeevents (IOC 2009).The same Commission further establishesthat vulnerability analyses be conducted accordingto different – macro to micro – scales, depending onthe approach to be given by the national integratedcoastal management programs.In this study, the macroscale will be used todefine Brazilian coastal vulnerability by region, thusproviding inputs for planning responses for theirmitigation and adaptation.MethodologyAccording to IOC’s proposed methodology,five stages are necessary to make nationaland regional climate change adaptation plans:1) Identifying and quantifying the hazards;2) Measuring vulnerability; 3) Assessing the risk;4) Enhancing awareness and preparedness;5) Mitigating the risk. This study addresses stagesPan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone1871 and 2, which are the basis of the necessaryknowledge to define the other stages.Information generated by MDZCM(Nicolodi & Zamboni 2008) was used to prepare theoverview map on the vulnerability of the Braziliancoastal zone with relation to natural risk, social risk,and technological risk. To the crossing of suchresults were added spatial information on populationdynamics, geomorphology, use and occupation ofthe Exclusive Economic Zone (EEZ) andbiodiversity 2 . In all cases, specific geoprocessingroutines were resorted to, along with IDRISI andARCGIS9 3 software.The analysis scale of the issues addressed inMDZCM and the vulnerability analyses of thecoastal zone proposed by this paper is 1:1,000,000.This scale corresponds to the scope of the area ofstudy and enables practically all existing map basesto be included in the analysis context.Natural risk charts are a direct product of acombination of altimetry aspects 4 with populationdata, added to assessment of vulnerability levels toinundation caused by extreme weather events, heavyrains, and prospects of a higher sea level.The altimetry information was modeled intogeographic information systems 5 , and became adigital model of the Coastal Zone, to which data onthe local population were added by subdistricts,provided by IBGE, the Brazilian Institute ofGeography and Statistics, according to the censusupdate provided by IBGE in 2006.Upon refining the five levels of potential naturalrisk, 6 coastal process information was considered,through the use of statistical techniques (weightedaverages). Eroded coastal areas added value byshowing the regions more prone to flooding, sinceerosion tends to destroy natural barriers such asrestingas (beach ridges with scrubby vegetation),dunes, sea cliffs, mangroves, etc. On the other hand,coastal areas with a sediment accretion and, conesquently,shoreline progradation, subtracted valuewhen determining risk ranges (Muehe 2006).2 The inclusion of these variables did not necessarilyoccur during the analyses of Geographic InformationSystems, but rather, during the descriptive analysis of theresults.3 Such routines include Boolean operations with maps,attribute analysis of georeferenced databases, and multistandardevaluations.4 The altimetry data came from SRTM-NASA, availableon the U.S. Geological Service.5 Such modeling took place at the Geography Departmentof the Federal University of Rio de Janeiro (UFRJ) whenthe data base’s MDZCM was prepared.6 These are: very high, high, moderate, low, and very low.When weighting the factors, thecombination of land elevations below 10 m abovesea level and marine erosion was considered themost critical indicator of coastal environmentvulnerability to floods. The risk potential could thenbe assessed by cross-referencing this informationagainst the population data by subdistrict. An exampleof a natural risk map can be seen in figure 1.Regarding social risk definition, the level ofincome of that part of the population earning up tothree (3) times the minimum wage was used asbackground data, based on IBGE 2000 censusresults by district. Area ranking according to socialrisk potential 7 was obtained by crossing income datawith the number of homes lacking garbagecollection and wastewater services. The rankingsystem thus considered dwellings where wastes aredisposed of in rudimentary cesspools, ditches, rivers,lakes, or into the ocean to be “lacking basicsanitation”. Regarding the destination of solid waste,the ranking system considered those dwellingswhere garbage is burned or buried, thrown intobackyards or streams, the ocean, or ponds as “homeslacking garbage collection”. An example of a socialrisk map can be seen in figure 2.As to technological risk, the data came fromsources referred as technological, such as, for example,power generating units or manufacturing facilities.Their construction methodology resulted fromthe number of industry employees per city in relationto the industry’s polluting potential. The definitionof polluting potential followed the methodologyproposed by RAIS, the Annual List of Social Informationissued by the Labor Ministry (2002) 8 .The data resulting from the crossing of thisinformation were grouped into four categoriesrepresenting technological risk potential levels (low,medium, high, and very high). Moreover, the mapsinclude the location of thermal plants according tothe fuel used, natural gas and oil production andextraction activities, and oil industry-related7 These can be: very low, low, medium, high, and veryhigh.8 Types of industry according to polluting potential: (a)very high: Rubber, Tobacco, Leather, Chemicals, Mining,Non-Metal Ores; (b) high: Metallurgy, Textiles,Foodstuffs and Beverages, Paper and Printing; (c)medium: Mechanics, Rolling Stock, Footwear, Wood andFurniture; (d) low: Electronics and Communications,Civil Construction, Public Utilities. It should be pointedout that IBAMA measures the polluting potential ofmanufacturing activities, especially with regard to theCadastro Técnico Federal (a mandatory registration listof companies in polluting, or potentially polluting,industries) (http://www.ibama.gov.br/cadastro).Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

188J. L. NICOLODI & R. M. PETERMANNFigure 1. Example of a Natural Risk Map.Figure 2. Example of a Social Risk Map. The risk potential is shown in yellow at subdistrict level.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone189Figure 3. Example of a Technological Risk Map. The pie chart indicates the risk potential at each municipality. Thetown population appears as a gray background.facilities (pipelines, refineries, etc). This mappingactivity is based on population estimates bymunicipality, made in 2006, which give an idea ofthe number of people potentially affected by anaccident involving technological risk. This is relatedto the various stages of the productive activity, fromthe extraction of raw materials to the marketing ofgoods. An example of a technological risk map canbe seen in figure 3.The definition of Brazilian coastvulnerability was made in five hierarchical levels,following the same categories as for Social Risk,Technological Risk, and Inundation Risk. For datacrossingoperations the following values wereestablished per category: Very Low ≤ 1, Low >1 and≤ 2, Medium >2 and ≤ 3, High >3 and ≤ 4, VeryHigh >4 and ≤ 5.The first stage consisted in establishing asole Technological Risk rating for themunicipalities. This was calculated by the weightedaverage between risk potential categories. A processexample may be seen in Table I.The results were ranked in five intervalsusing the Geometrical Interval algorithm, availableas an ArcGIS function. In this system, the classintervals are based on a geometrical series. Thegeometric coefficient in this classifier can changeonce (to its inverse) to optimize the intervals. Thealgorithm creates these geometrical intervals byminimizing the square sum of elements per class.The reclassified technological risk potentialcan then be spatially crossed with the social risk andthe natural risk data. To this end, phrases andsentences similar to those used in mathematics wereused to describe Boolean operations. The modelused to describe the GIS sentences, involved thelogical combination of the vector maps throughconditional operators which supported theassumption to which the analysis was directed. TheVulnerability Index obtained from this crossing wasdeveloped through the simple average between thethree types of risk involved (Fig. 4).Results and DiscussionFrom the onset of European colonization,the establishment of populations and socio-economicutilization of coastal areas have been increasinglyintense. This territory occupation by about onequarter of the Brazilian population, has begun withthe appropriation of common spaces in the CoastalPan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

190J. L. NICOLODI & R. M. PETERMANNTable I. Example of weighted average calculation for the new categories of Technological Risk.Final RiskMunicipality V. High Risk High Risk Medium Risk Low Risk(Weighted Average)Araruama 3605 1536 4608 2614 3087.2Armação de Búzios 60 192 576 310 244Arraial do Cabo 3565 220 660 266 1515.5Cabo Frio 4615 3324 9972 3234 5196.7Casimiro de Abreu 230 828 2484 1744 1100.1Iguaba Grande 90 4 12 156 58.1Macaé 63535 13608 40824 32586 39982.2Maricá 1335 1040 3120 934 1575.9Rio das Ostras 275 452 1356 3432 1008.2Saquarema 565 440 1320 296 652.6Final Risk = [(V.High Risk*5)+(High Risk*4)+(Medium Risk*3)+Low Risk*2)]/(5+4+3+2)Figure 4. The spatial crossing made between the three risk types: (a) Natural Risk; (b) Social Risk; and (c)Technological Risk. The result is shown by (d) Coast vulnerability, obtained by simple average.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

192J. L. NICOLODI & R. M. PETERMANNestuaries and sediment accretions which jointly looklike rias (drowned river valleys), and the reason whythey are called re-entrâncias (recesses) in Maranhãostate. Many of them resulted from prograded muddydeposits, forming long landforms more or lessperpendicular to the coast. The North coast, hereshown from Oiapoque (in Amapá state) to the southof Maranhão, is a high energy segment, with astrong sediment mobility, very much influenced bythe intense water and sediment discharge from theAmazon River and the hydrodynamic factors of theocean, particularly tides (Muehe & Nicolodi 2008).Such characteristics provide lowvulnerability levels to this area, which change whenanalyzing the existing metropolitan regions. TheNorth coast of Brazil is characterized by scantyhuman presence, consisting mainly of traditionalextractive and foraging communities, large emptyareas, dozens of municipalities with a smallpopulation density, but with important regionalcomplexes such as Macapá and metropolitanconcentrations in Belém and São Luís (Strohaecker2008).The geomorphologic characteristics of thenorth coast of Pará form a physical barrier to anintensive process of population settlement on thecoast. But a few parts of this segment have had adisorderly population growth. Population density inthis area is approximately 27 inhabitants per km 2 , incontrast to other sections with a density of 3.5inhabitants per km 2 . Significant values can benoticed just in the Belém area and its surroundings(of about 220 inhabitants/km 2 ).São Luís is located in the GolfãoMaranhense (Maranhão’s Big Gulf) region. São Luíshas the state’s only significant populationconcentration (> 170 inhabitants per km 2 ) on thiscoastal lowland. Therefore, only the area around theMaranhão state capital is highly vulnerable.In addition to this analysis of populationdynamics, it should be noted that the coverage ofwaste collection services seen in the North Region ismuch lower than in other regions of Brazil, and it isalso the one with the worst provision of this basicservice: 6,790 tons/day.This situation, further to the data on basicsanitation, leads to a coefficient ratio between thetotal population and the population exposed to socialrisk of 33.7% for the North Region, which, inabsolute figures can be translated into 2.206,138inhabitants, most of them residing in the capitals andtheir outskirts (Astolpho & Gusmão 2008). This databecomes even more relevant when considered a fewresults of global assessments made by IPCC, whichconfirm the fact that disadvantaged populations, whoare less able to adapt, are the most vulnerable(Marengo 2006).Adding to high levels of vulnerability of themetropolitan areas in the North, is the associationbetween the metal-mechanical complexes and thepaper and pulp industry on the coast of the Pará andMaranhão states, with massive investments in theproduction of metallic minerals such as iron andaluminum, and extensive planted land used toproduce pulp. This is a determining factor thatincreases the technological risk and vulnerability ofthe Coastal Zone at critical points, as is the case ofBarcarena, in Pará state, and São Luís, in Maranhão(Egler 2008).Northeast RegionThe Coastal Zone of the Northeast Region,marked here by the coast between the north of Piauístate and the south of Bahia state, features a greatdiversity of ecosystems, with distinct physical andgeomorphologic characteristics affected by a broadrange of pressure vectors, which ultimately definethe region’s vulnerability.Unlike the North, where only metropolitanareas were found to be highly vulnerable, theNortheast alternates between the five vulnerabilitylevels which do not necessarily have a directrelationship with population dynamics.In geomorphological terms, the upper part ofthe region is dominated by sedimentary deposits ofthe Barreiras group, in front of which numerousdune fields have developed, fed by sedimentscoming from the inner continental shelf, as, forexample, the Parnaíba River Delta and Jericoacoara,in Ceará.In Rio Grande do Norte state one can see theBarreiras Group sea cliffs and a wide developmentof active dune fields along the entire coast. One cannotice the natural barrier formed by the dunes on theriver estuaries, which leads to insufficient drainageand forms swampy valleys, in addition to anincreased number of estuaries and mangrovesstarting in Paraíba state, as a result of a higherprecipitation volume.To the east of this dune field, the ParnaíbaRiver estuary displays a coastal stretch consideredmedium to highly vulnerable, particularly due to thestrong erosion caused by the cyclic floods that hitthe Parnaíba River downstream during the highwaterseason (Fig. 6).The coast of Ceará, marked here and thereby higher land, has a large number of eroded coastalsections linked to moving barchan dune fields, theBarreiras terrace deposits, and outcrops of thecrystalline basement. In addition, the Ceará coast hasPan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone193a low population density, except near Fortaleza,where this density is higher, and the vulnerabilitylevel is also high. In the Aracati area, wherevulnerability is medium to high, the factors affectingthis ranking are related to the significant shortage ofbasic sanitation, the accelerated manufacturingdevelopment, and the increase of shrimp farmingactivities and tourism.Another especially vulnerable area is thevicinity of Mossoró, in the innermost portions of thecoastal region. This situation occurs due to a numberof factors, including the existing low-lying areas thattend to flood owing to the drainage of Apodi andMossoró rivers, an acute shortage in the provision ofbasic services, and an intricate logistic oil and gasnetwork, which extends all the way to the area nearMacau, where the Guamaré Natural Gas Plant islocated. In this section, the highlight is the coastalerosion, which is so strong, that it is alreadyaffecting oil industry equipment installed in the area(Muehe 2006).When analyzing this region’s social risk, onecan see that the situation is critical in large centers,particularly in Natal, João Pessoa, and Fortaleza.The lack of sanitation in these areas issignificantly greater than the lack of garbagecollection services. In Fortaleza, the data on theshortage of garbage collection services show atendency toward solving the problem, while thesewerage situation is of extreme concern in almostall municipalities and districts. This same situation,although less severe, can be noticed in Maceió,Aracaju, and nearby areas (Astolpho & Gusmão2008).In the Northeast, the population exposed tosocial risk is 25.71% of the total population, which,in absolute numbers, may be translated into12.286,455 inhabitants potentially more vulnerableto the effects of climate changes.In the central part of the Northeast, themain areas of higher vulnerability includethe metropolitan areas of Natal, João Pessoa, andRecife (Fig. 7). According to Neves et al. (2006)about 42% the coast of Paraíba is exposed to theeffects of erosion. Similar geomorphologiccharacteristics extend to the coast of Pernambuco,which has a higher population density, comparedwith the coast of Paraíba and Alagoas. Along thisentire segment, low to medium natural riskspredominate, with the exception of the areas with thehighest urban concentration (João Pessoa andRecife) and deeply eroded segments (Paulista,Itapojuca, Suape, Cabo de Santo Agostinho, andRecife).Another factor that adds to the region’s highvulnerability is the displacement of the chemicalscomplex to the Northeast coast along Salvador,Aracaju, and Maceió, due to the expansion of theenergy boundary on the coast. This fact has broughta massive concentration of pipelines, terminals, andplants. The surroundings of the Recôncavo Baianoarea and the cities of Aracaju (SE), Maceió (AL)Recife-Cabo (PE), and Macau-Guamaré (RN) arehighlights in this process, where the energyproduction equipment increases the exposure toenvironmental hazards (Egler 2008).In the southern portion of the NortheastRegion, the most outstanding morphological featureis the São Francisco River delta, site of the country’sworst coastal erosion. Bittencourt et al. (2006)indicate as likely causes the embankmentinterventions to contain the river flow upstream fromits mouth, mainly those related to the construction ofhydroelectric plants, which implies great potentialinundation of inner drainage areas, which rates thissection as of high risk.From the São Francisco River to theCaravelas River plain, there is a general tendency forthe coastline to prograde and stretches with sea cliffsof the Barreiras Group to erode. Dune fields appearnear the mouth of the São Francisco and on thenorthern coast of Bahia. Near Salvador, the BarreirasRiver is replaced by outcrops of Pre-Cambrian andCretaceous crystalline basement.Higher-than-average coastal sections,coupled with a low density population, have amedium to low vulnerability. At some places, thislevel is high only where population density is higherand basic sanitation is deficient, i.e., at Valença,Ilhéus, and Porto Seguro - urban centers combinedwith river mouths (Fig. 8).In the Salvador metropolitan area, the highvulnerability levels are not only related to thesefactors, but also to high technological risk posed bythe Camaçari manufacturing complex, namely theoil industry, and particularly the Landulpho AlvesRefinery, the Candeias Natural Gas Production Unit,and the Termobahia, Rômulo Almeida, andCamaçari thermal plants (Fig. 9).Southeast RegionThe coast of Espírito Santo state and thenorth coast of Rio de Janeiro state are geomorphologicalboundaries of the Northeast coast. Thisstretch is dominated by Tertiary terraces and the seacliffs (Barreiras), Pre-Cambrian crystalline promontoriesand Quaternary plains of river and sea origin.Between the mouths of São Mateus and Itabapoanarivers, the Barreiras terraces and sea cliffs extendalong the entire coast, displaying live and deadPan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

194J. L. NICOLODI & R. M. PETERMANNFigure 6. Vulnerability Map of the North Region, showing the states of Piauí, Ceará, and Rio Grande do Norte.Figure 7. Coastal vulnerability of Paraíba, Pernambuco, Alagoas, and Sergipe states.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone195cliffs, and marine abrasion terraces. The sedimentarycoastal plains are little developed, and the plain atthe Doce River mouth is the most relevant.This stretch consists of sections of mediumto low coastal vulnerability. Only three sites wereranked as vulnerable (medium to very high levels):the Doce River, Vitória, and the inner drainage areasof the Paraíba do Sul River (Fig. 10).In the case of the Doce River, one can seethat the combination of the above-mentionedconditions, coupled with high levels of coastalerosion, make the region of São Mateus andConceição da Barra more vulnerable.What adds to this situation is the fact that thestretch of the Coastal Zone between Mucuri, on thesouthern coast of Bahia, to the center-north ofEspírito Santo, especially near Linhares and Aracruzin Espírito Santo, is specializing in the production ofpulp for the foreign market, as can be seen from theconcentration of equipment used by the paper andpulp industry, particularly the continuousdimensions of the area involved (Egler 2008).The Doce River drainage at the end of itsflow at Linhares occurs on low ground showingmarginally to its main flow a number of tributariesconnected to ponds. Attributing higher risk levelsrelates to inundation potential for low land with arate of human occupation slightly above the region’saverage. Vitória, Vila Velha, and Guarapari have thehighest population densities in low coastal areas,with population densities above the Espírito Santostate coast average.The drainage of the Paraíba do Sul River, inthe Campos dos Goytacazes area, occurs at near sealevel land, through densely populated areas boundby the Pre-Cambrian crystalline complex. Thisgeomorphologic setting, associated with the populationdensity of northern Rio de Janeiro state, aretypical of the vectors leading to high vulnerabilitylevels in the area. The town of Atafona, on the southbank of the Paraíba do Sul River mouth, has one ofthe most intense erosive phenomena of the entireSoutheast coast of Brazil (Muehe et al. 2006).The stretch between Cabo Frio andGuanabara Bay has a rim formed by narrow ridgesseparated by rocky headlands, with the developmentof lagoons behind the ridges. This section is alsoknown as the Lake Region. The general direction ofthe coastline, which directly exposes this coastalstretch to the south (with waves from the southquadrant) and, from time to time, to the action ofheavy storms, which explains the strong erosion.The highest levels of vulnerability identifiedon the eastern coast of Rio de Janeiro state are in theareas of São João da Barra and Macaé, which, in thelast two decades, have experienced a sharp urbandevelopment linked to oil prospecting activities onthe contiguous continental shelf (Fig. 11). In CaboFrio, the increase in population in the urban areas, ina land that displays higher landforms (promontoriesand hills) and low-lying coastal plains, lead to anincreased potential hazard to which the area isexposed (Tessler 2008).The Guanabara Bay region is one of themost emblematic cases in Brazil, with regard tovulnerability. Its low topography lies along ageological fault that extends toward the ocean fromthe crystalline complex. To this depression convergeall drainage networks from Serra do Mar mountainrange at the back of the bay, which were blocked attheir low flows by high sea levels during theHolocene.In contrast with the ocean beaches located atits outer edges, constantly exposed to storm cyclesoriginating from south quadrants, the inner baycoastline is affected only occasionally by morepowerful events. Its vicinity, however, concentratesone of the highest population densities in the country9 , sometimes along the lower river streams thatflow into the system. In extreme tidal situations followedby heavy rain on the mountain range (associatedwith the passage of frontal systems whichdrown the drainages in their lower flows) the innerbay coastline, which is lower, is exposed to inundationevents (which increases the volume of rivers).In addition to this context, Rio de Janeirohas the highest ratio in Brazil, between the exposedpopulation (78%) and its total population, equivalentto 11.194,150 people – some 5 million of which inthe capital alone. Data on the social risk of thisportion of the Brazilian territory are alarming, asshown in (Fig. 12).In addition to these factors which lead tohigh vulnerability, the Rio de Janeiro metropolitanarea holds a petrochemical complex, with anintricate network of refineries 10 , natural gas plants 11 ,gas pipelines, and offshore oil fields.The location of a coastal mountain rangenear the existing shoreline, west of Guanabara Bay,with its promontories marking small individual beachesand conspicuous inlets and sedimentary plains9 Rio de Janeiro is the state with the greatest totalpopulation residing in metropolitan areas (75.2%).Additionally, the state includes most coastalmunicipalities with population densities over 1,000inhabitants/km 2 , as is the case of Rio de Janeiro City andNiteroi, the towns of Baixada Fluminense (in the state’slow-lying area) and the outskirts of the metropolitan area.10 Duque de Caxias and Manguinhos Refineries.11 REDUC I and II and Cabiunas I, II, and III.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

196J. L. NICOLODI & R. M. PETERMANNFigure 8. Urban centers in Bahia state, where vulnerability is high due to a high population density and an inadequatebasic sanitation service.Figure 9. The metropolitan area of Salvador. High vulnerability levels linked to high technological risk.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone197Figure 10. Medium to very high vulnerability level locations: the Doce River, Vitória and the inner drainage areas ofthe Paraíba do Sul River.Figure 11. Higher vulnerability levels identified on the eastern Rio de Janeiro state coast associated with the São Joãoda Barra and Macaé areas.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

198J. L. NICOLODI & R. M. PETERMANNFigure 12. Social risk map of Guanabara Bay, in Rio de Janeiro. Graphic forms in purple represent the lack of sewers insubdistrict households. The shortage of garbage collection services is shown in blue. (Adapted from MDZCM, 2008).Figure 13. High vulnerability level in the metropolitan Rio de Janeiro area. The coastal area south of Guanabara Bayhas a low vulnerability.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone199formed in the mountain range recesses, shape a geomorphologicregion of many different ground levelsoccupied by permanent, low density populations.As is common during summer on most ofthe Brazilian coast, beaches that are far frombig cities get a large inflow of temporary population.Therefore, most of this coast does not present ahigh vulnerability level (Fig. 13).The group ofcities near Santos known as Baixada Santista,which includes the Santos bay and estuary, as wellas the surrounding urban areas, contain Brazil’slargest sea port and manufacturing complexes onthe small estuary and coastal plains, which havedeveloped around channels, on the foothills ofSerra do Mar. The region’s high population density,its typical socio-economic features, and itsgeomorphologic configuration of a pronouncedretreat in the crystalline complex, determined, foralmost the entire area, a high vulnerability level(Fig. 14).Figure 14. Baixada Santista and the Santos estuary. A combination of socio-economic, technological andgeomorphologic features resulted in high vulnerability.Another factor that makes the entire regionmore vulnerable is a visible concentration ofmanufacturing facilities between Santos and Macaéwhere there are oil and gas extraction fields,terminals, and pipelines, thermal and nuclear powerplants, and a host of chemicals, metal andmechanical complexes. Furthermore, energyboundaries are being expanded toward the Southcoast, with an increase in oil prospection in theSantos Basin, plus the construction of pipelines, andan expanded chemical industry in Paranaguá.South RegionA scenery of broad amphitheaters is thepredominant geomorphologic feature in theParanaguá Bay region, which includes the coastalarea south of Baixada Santista to Itajaí, on the coastof Santa Catarina state. This segment contains threemajor seaports (Paranaguá, São Francisco do Sul,and Itajaí). These municipalities and/or theirsurroundings have significantly higher populationdensities than the average population per km 2 of theSoutheast coast of Brazil. This mixture oftopographic and population factors, the socioeconomicimportance of these urban centers, plusthe instability factors affecting the shoreline, producemedium to high vulnerability levels (Fig. 15).On the coast of Santa Catarina, the Joinvillearea, the Itajaí Valley, and Greater Florianópolishave very high vulnerability levels, because of theirhigh urban concentrations in areas below 10 metersabove sea level. Floods as those that occurred in1983, 1984, and recently in the November 2008disaster, in which 135 people perished and over 1.5Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

200J. L. NICOLODI & R. M. PETERMANNFigure 15. Vulnerability of the northern part of the South Region. The topography, population density, and socioeconomicfactors of urban centers generate medium to high vulnerability levels.Figure 16. The high and very high vulnerability region corresponds to the distal portion of the Itajaí-Açu River basin,which has undergone frequent inundation in recent years, particularly during recent events in 1983, 1984, and 2008.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone201million people were injured, fully confirmed thisvulnerability (Fig. 16).The southern part of Santa Catarina state upto the border of Uruguay, is characterized by sandybarriers highly exposed to a strong wave and stormregimen with a tidal range of less than 1 m.Numerous lagoons developed behind the barrierswith only few outlets.Although the geomorphological contextalone cannot explain the high vulnerability levels, itis important to emphasize that this region is criticalfor the occurrence of uncommon, extreme events ofgreat magnitude, as was the case of CatarinaHurricane, which struck in 2004 and all but wipedout the bordering area between the two southernmoststates of Brazil.The only place defined as highly vulnerablein Rio Grande do Sul state is the Rio Grande areanear the Patos Lagoon outlet, which is kept openthrough two 4.5 km long jetties (Rio Grande barpiers). This scenario includes the main urban centerin the inner estuary, with a population of around200.000 inhabitants living on low, flat ground andover areas expanded by the water surface landfill.The land occupied by housing coexists with spacesdominated by the activities of one of the country’smost important ports, combined with an expandingmanufacturing and petrochemicals complex of greatrelevance to the state (Fig. 17).The role of the port of Rio Grande in thispart of the area of high vulnerability, should be consideredin conjunction with the Metropolitan AreaFigure 17. From the south of Santa Catarina state to the border of Uruguay, vulnerability is relatively low, with theexception of the urban area of Rio Grande. This region is subject to weather events of great magnitude, such asHurricane Catarina in 2004.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

202J. L. NICOLODI & R. M. PETERMANNof Porto Alegre, as regards the lagoon area wherethey are located. The likelihood of an increasedtrading of energy, goods and services and theimplementation of new plants in the area due to itsMercosur standing, are specific elements that willprobably increase the threat of technological hazardson the South coast in the coming decades (Egler2008).ConclusionsThe Coastal Zone is the most dynamicgeographical area in the country, since the timewhen the country was a colony of Portugal, andconnections from structural centers directed internalflows directly to seaports, next to which the firsturban centers were established (Moraes 1999).The analysis of the combination between alikely unchanging tendency of this scenario inthe near future and the context of global climatechanges, points inevitably to the importance ofundertaking a realistic coastal management,with priority actions, and human and financialresources.Knowing about the mesoregions more orless vulnerable to the impacts of the direct effects ofclimate change is essential for the public authoritiesto make their decisions. These effects are directlylinked to three major types of causes, defined inthis paper as natural risk, social risk, andtechnological risk. The combination of theseconcepts, when applied to the national territory haveenabled the definition of the five levels ofvulnerability used, illustrating the scene presented asa challenge to be faced by integrated coastalmanagement in Brazil, especially in the currentcontext of climate change.From this standpoint, the IntergovernmentalOceanographic Commission (IOC) has definedthe climate change-related risks as shown in TableII. With the exception of tsunamis, Brazil isexposed to varying levels of the other risks definedby IOC.In addition to the risks to which theBrazilian coast is directly exposed, other factorsare expected to indirectly influence the dynamicsof this part of the territory. According to Marengo(2006), modeling carried out by IPCC indicatepossible significant changes to the outflow ofthe largest Brazilian rivers: an increased volumein the Plata and Paraná River basins and a decreasein the Amazon and the Pantanal basins. Thevariation in those water volumes will lead to aTable II. Definition of climate change-related risks to Coastal Zones, according to the IntergovernmentalOceanographic Commission (IOC, 2009). and the relations with the observed in Brazil.Risk Definition Vulnerable areasRapid onsethazardsCumulative orprogressivehazardsTsunamiStorm surgeExtremewind-forcedwavesLong-term sealevel riseCoastalerosionA series of ocean wavesgenerated by displacement of theocean floor from an earthquakes,volcanic events, or large asteroidimpacts.Temporary rise in sea levelcaused by intense stormassociated with low pressure andstrong onshore winds.Extreme instances of wavesgenerated by winds somewhere inthe ocean, be it locally orthousands of kilometers away.Global sea level rise due to athermal expansion of oceans andincreased melting of land-basedice.Loss of coastal land caused bywaves, tides currents or drainagethat can be enhanced by each ofother hazards.Because the Brazilian coast is a"passive" coast, such events aren’texpected, although they can’t bediscarded.The entire coast, especially the southeastand south, due to greater energyinvolved in the dynamics of this coastalregion.The entire coast, especially the states ofSanta Catarina and Rio Grande do Sul.There’s no availability long data seriesin the country, with the exception of Riode Janeiro and São Paulo. In these cases,there is evidence of elevation about 3mm/year.There is evidence of coastal erosion inseveral areas of the Brazilian coast. Thephenomenon is more complex when itcomes to urban coasts.Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

Potential vulnerability of the Brazilian coastal zone203new sediment transport regime and its consequenceson the shoreline.These effects were identified by Neves &Muehe (2008). They include: a) coastal erosion andprogradation; b) damaged coastal protection works;c) structural or operating losses at ports andterminals; d) damaged urban construction work incoastal cities; e) damaged structural or operatingsanitation work; f) exposure of undergroundpipelines or structural damage to exposed pipelines;g) saline intrusion in estuaries; h) saline intrusion inaquifers; i) mangrove evolution; j) damaged coralreefs.The scene has been set and there is nodoubt the challenge of adapting and mitigating theconsequences of such events is enormous, andcannot be faced without a detailed technicalreference study consisting of micro- and macroscalevulnerability assessments.The results obtained are included in thisreference study based on a georeferenced data base,and can potentially assist in dealing with two of thevarious issues raised by the Federal Audit Officeduring its audit of public policies and climate change(TCU 2009):1 – Brazil has no vulnerability study of itscoast against the impacts of changing climate on anational scale.2 – The country’s available data areinsufficient to build climate change impact scenariosin coastal areas.The main sectors likely to be affected in aclimate change scenario include ports and tourism.Brazil has a port sector that moves an annual 700million tons of various goods and accounts for over90% of all exports. One example was the destructionof the Itajaí seaport by heavy rains that hit SantaCatarina state in November 2008. Port reconstructionwork will require over R$ 320 million, inaddition to downtime losses estimated at US$ 35million per day.In the case of tourism, it is worth noting thatthe largest investments have been made ininfrastructure work in coastal zones. For example, ofthe 14 tourist centers covered by the PRODETUR /NE-II program, with a US$ 400 M funding, 12 arelocated in the Coastal Zone 12 .These are examples of situations which theBrazilian society should be prepared to handle. Theanalysis of coastal zone vulnerability should guidethe priority given to government actions. The areasdefined as of high or very high vulnerability shouldbe on the top priority list when decisions and plans12 Source: http://migre.me/3H24g access on 11/27/2008.are made.In terms of institutional planning, Nicolodi& Zamboni (2008) analyzed the main actionsundertaken by the Federal Government in the coastalzone and found that, although the management toolsdeveloped between 1996 and 2006 have broughtsome advance 13 , integrated strategic planning is stillincipient.An integrated strategic planning mustinclude the variables related to climate changevulnerability, especially when analyzing geographicaction priorities.Neves & Muehe (2008) reported the followingactions that should make up the mentionedintegrated strategic planning:• permanent (long term) environmental monitoring;• proposing effective municipal legislationgoverning urban occupation;• effective state policies on coastalmanagement;• directing federal action efforts: legislationand education;• action monitoring and coordination;• identification sources of funds, theirapplication, and forms of control;• planning and prioritization of studies toundertake traditional actions (retreat, accommodation,and protection).Key initiatives to address the “climatechange in coastal areas” theme, such as the GlobalOcean Observation System (GOOS) 14 , linked to theIntergovernmental Oceanographic Commission(IOC), or the surveys on coastal erosion made by theMarine Geology and Geophysics Program(PGGM) 15 must be encouraged as a way ofmaintaining a structural base for decision-making byinstitutions responsible for the country’s coastal andmarine management.13 The authors identified as main instruments: ProjetoOrla (Rim Project) Agenda 21, Planos DiretoresMunicipais (Municipal Master Plans), ConselhosMunicipais de Meio Ambiente – CMMA (MunicipalEnvironmental Councils), Zoneamento EcológicoEconômico Costeiro – ZEEC (Coastal Ecological andEconomic Zoning), Áreas de Exclusão Temporária deÓleo e Gás (Areas of Temporary Oil and Gas Exclusion),Sistema Nacional de Unidades de Conservação (NationalSystem of Conservation Units), Mapeamento daSensibilidade do Litoral ao Óleo (Mapping of the CoastalSensitivity to Oil).14 The Brazilian component of this program may beaccessed from: www.goosbrasil.org15 The results are organized in the book Erosão eProgradação do Litoral Brasileiro (Muehe 2006).Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

204J. L. NICOLODI & R. M. PETERMANNReferencesAb’Saber, A. N. 2000. Fundamentos daGeomorfologia Costeira do Brasil Inter eSubtropical. Revista Brasileira deGeomorfologia, 1(1): 27-43.Astolpho, S. M. & Gusmão, P. P. 2008. PotentialSocial Risk. Pp. 121-148. In:Macrodiagnóstico da Zona Costeira e Marinhado Brasil. In: Zamboni, A. & Nicolodi J. L.(Eds.). Macrodiagnóstico da Zona Costeirae Marinha do Brasil. Ministério do MeioAmbiente, Brasília, 242 p.Bittencourt, A. C. P. Oliveira, M. B. & Dominguez,J. M. L. 2006 - Sergipe. Pp. 212-218. In:Muehe, D. (Ed.). Erosão e progradação dolitoral brasileiro. Ministério do MeioAmbiente e Programa de Geologia e GeofísicaMarinha (PGGM), Brasília, 476 p.Castro, C. M., Peixoto, M. N. O. & Rio, G. A. P.2005. Riscos Ambientais e Geografia:Conceituações, Abordagens e Escalas.Anuário do Instituto de Geociências –UFRJ, 28(2): 11-30.Egler, C. A. G. 1996. Risco Ambiental como critériode gestão do território. Territory, 1: 31-41.Egler, C. A. G. 2005. As Cartas de Risco Ambiental,Social e Tecnológico do NovoMacrodiagnóstico da Zona Costeira. IEncontro Temático: Gestão Integrada deBacias Hidrográficas e da Zona Costeira,Ministério do Meio Ambiente, Itajaí, SC, CD-ROM.Egler, C. 2008. Potencial de Risco Tecnológico. Pp.149-172. In: Zamboni, A. & Nicolodi, J. L.(Eds.). Macrodiagnóstico da Zona Costeirae Marinha do Brasil. Ministério do MeioAmbiente, Brasília, 242 p.IOC – Intergovernmental OceanographicCommission. 2009. Hazard awareness andrisk mitigation in integrated coastal areamanagement. United Nations Educational,Scientific and Cultural Organization(UNESCO), Paris, 143 p.Marengo, J. A. 2006. Mudanças Climáticas Globaise seus Efeitos sobre a Biodiversidade.Caracterização do clima atual e definição dasalterações climáticas para o territóriobrasileiro ao longo do Século XXI. Ministériodo Meio Ambiente, Brasília, 212 p.MDZCM – Macrodiagnóstico da Zona Costeira eMarinha do Brasil. 2008. In: Zamboni, A. &Nicolodi, J. L. (Eds.). Ministério do MeioAmbiente, Brasília, 242 p.Moraes, A. C. R. 1999. Contribuições para a gestãoda zona costeira do Brasil: elementos parauma geografia do litoral brasileiro. São Paulo:Hucitec /Edusp. 229 p.Muehe, D., Lima, C. F. & Lins-de-Barros, F. M.2006. Pp. 265-296. In: Muehe, D. (Ed.).Erosão e progradação do litoral brasileiro.Ministério do Meio Ambiente e Programa deGeologia e Geofísica Marinha, Brasília. 476p.Muehe, D. 2006. Erosão e progradação do litoralbrasileiro. Ministério do Meio Ambiente ePrograma de Geologia e Geofísica Marinha(PGGM), Brasília, 476 p.Muehe, D. & Nicolodi, J. L. 2008. Geomorfologia.Pp. 23-58. In: Zamboni, A. & Nicolodi J. L.(Eds.). Macrodiagnóstico da Zona Costeirae Marinha do Brasil. Ministério do MeioAmbiente, Brasília, 242 p.Neves, C. F. & Muehe, D. 2008. Vulnerabilidade,impactos e adaptação a mudanças doclima: a zona costeira. CGEE Strategicpartnerships. Brasília, 27: 217-296.Nicolodi, J. L. & Zamboni, A. 2008. GestãoCosteira. Pp. 213-241. In: Zamboni, A. &Nicolodi J. L. (Eds.). Macrodiagnóstico daZona Costeira e Marinha do Brasil.Ministério do Meio Ambiente, Brasília, 242 p.Strohaecker, T. M. 2008. Dinâmica Populacional.Pp. 59-92. In: Zamboni, A. & Nicolodi J. L.(Eds.). Macrodiagnóstico da Zona Costeirae Marinha do Brasil. Ministério do MeioAmbiente, Brasília, 242 p.TCU – Federal Audit Office, 2009. Auditorias denatureza operacional sobre políticaspúblicas e mudanças climáticas – Adaptaçãoem Zonas Costeiras. Reporting Justice AroldoCedraz, Brasília, 62 p.Tessler, M. 2008. Potencial de Risco Natural. Pp.93-120. In: Zamboni, A. & Nicolodi J. L.(Eds.). Macrodiagnóstico da Zona Costeirae Marinha do Brasil. Ministério do MeioAmbiente, Brasília, 242 p.Received January 2010Accepted May 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 184-204

A methodology for assessing the vulnerability of mangrovesand fisherfolk to climate changeLUIZ F. D. FARACO 1,4 , JOSÉ M. ANDRIGUETTO-FILHO 2,4 & PAULO C. LANA 31 Parque Nacional Saint-Hilaire/Lange, Instituto Chico Mendes de Conservação da Biodiversidade, Av. Paranaguá,729, sl. 02, Matinhos, Paraná, Brasil. E-mail: xicofaraco@yahoo.com2 Departamento de Zootecnia, Universidade Federal do Paraná, Rua dos Funcionários, 1540, CEP 80050-035,Curitiba, Paraná, Brasil.3 Centro de Estudos do Mar, Universidade Federal do Paraná, Av. Beira Mar, s/n, CP 50002, CEP 83255-000, Pontaldo Sul, Pontal do Paraná, Paraná, Brasil.4 Programa de Pós-graduação em Meio Ambiente e Desenvolvimento, Universidade Federal do Paraná, Rua dosFuncionários, 1540, CEP 80050-035, Curitiba, Paraná, Brasil.Abstract. Mangroves are putatively vulnerable to climate change, especially sea level rise,depending on factors such as coastal topography and the presence of barriers to landwardmigration. Usage patterns of mangrove resources can also affect their ability to respond to change.Brazilian artisanal fisherfolk are highly dependent on mangrove resources and services, whichmakes them also vulnerable to climate change. These populations have to cope with high levels ofuncertainty related to the availability of natural resources, and to social and political contexts,such as biodiversity conservation policies. Besides being protected by many different laws,mangroves are also included in no-take protected areas. This may contribute to their resilience asnatural systems, but can enhance the vulnerability of human populations. We propose herein aresearch methodology for assessing the vulnerability to climate change of the social-ecologicalsystem mangroves - fisherfolk, by analyzing exposure to sea-level rise, sensitivity and adaptivecapacity, and the impacts of conservation policies on these elements, particularly the effects ofcoastal protected areas in southern Brazil. An integrated social-ecological diagnosis may lead tomore flexible policies, elaborated with stakeholders’ participation, more adequate to local realitiesand more inclusive of strategies for mitigation and adaptation to climate change.Keywords: adaptive capacity, sea-level rise, Brazil, protected areas, social-ecological systemsResumo. Metodologia para análise da vulnerabilidade de pescadores e manguezais àsmudanças climáticas. Os manguezais são vulneráveis às mudanças climáticas, especialmente àelevação do nível do mar. Sua capacidade de resposta depende da topografia costeira, da presençade barreiras à migração e de padrões de uso dos recursos naturais. Pescadores artesanais no Brasilsão dependentes de recursos e serviços ambientais dos manguezais, sendo também vulneráveis àsmudanças climáticas. Eles lidam com incertezas relacionadas à disponibilidade destes recursos, e acontextos sociais e políticos. Mesmo protegidos por diversas normas, os manguezais também sãoincluídos em unidades de conservação de proteção integral. Isso pode contribuir para suaresiliência, mas pode, por outro lado, aumentar a vulnerabilidade das populações humanas.Propomos nesse trabalho uma metodologia para avaliar a vulnerabilidade de manguezais epopulações costeiras às mudanças climáticas, usando como estudo de caso uma área no litoral suldo Brasil. A metodologia baseia-se na análise da exposição à elevação do nível do mar, dasensibilidade e da capacidade adaptativa, e dos impactos das atuais políticas de conservação,especialmente as unidades de conservação, sobre esses elementos. Um diagnóstico sócioecológicointegrado pode contribuir para políticas mais flexíveis, elaboradas com a participação de todos osinteressados, mais adequadas às realidades locais e que incluam estratégias de adaptação àsmudanças climáticas.Palavras-chave: capacidade adaptativa, elevação do nível do mar, Brasil, áreas protegidas,sistemas socioecológicosPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

206L. F. D. FARACO ET ALLIIntroductionVulnerability of coastal populations andecosystems is a multi-concept which includeshazard exposure, sensitivity (the magnitude of lossesthat potentially result from exposure to the hazard)and adaptive capacity, or the capacity to respond toimpacts and prepare ahead of them, through copingstrategies and long-term adaptation to a certainthreat (Kelly & Adger 2000, Brooks 2003, Turner etal. 2003, Füssel 2007).In projected climate change scenarios, themain threats to coastal populations and ecosystemsare sea-level rise, the intensification of extremeweather events and ecosystem changes (Nicholls etal. 1999, Nicholls et al. 2007, Gilman et al. 2008).Other expected impacts are a rise of up to 3 ºC onsea surface temperature, changes in precipitation andfresh water input, salt water intrusion into soils andcoastal aquifers, and ocean acidification (Sterr et al.2000, Adger et al. 2005, Nicholls et al. 2007). Theseclimate alterations will have varied effects on coastalecosystems and human populations, with a likelyincrease on flooding and loss of wetlands (Nicholls2004), flooding of populated areas andinfrastructure, resulting in severe economic impacts(Zhang et al. 2004, Wu et al. 2008), and changes inthe availability of natural resources, withconsequences for the livelihoods of those that relydirectly on them for survival, such as traditional orneo-traditional coastal populations, includingfisherfolk (Badjeck et al. 2010).Exposure to these threats is directly linked tothe position of human settlements and ecosystems onwhich they depend in relation to the sea and toregions prone to the occurrence of sea-level rise andextreme weather events (Smit & Wandel 2006).Sensitivity, often treated as equivalent to exposure,depends on the number of people, the infrastructureand the extension of ecosystems exposed to thehazard, and on the level of dependency on naturalresources of the considered population (Tuler et al.2008). Adaptive capacity depends, in the case ofhuman populations, on a series of factors linked toaccess to assets. In the case of ecosystems, adaptivecapacity can be treated as analogous to ecologicalresilience, which is the capacity of a system torecover after a disturbance while maintaining itsfunctionalities (Walker et al. 2004). It will beaffected, among other factors, by the degree ofecosystem degradation and the exploration levels ofits natural resources. In any case, adaptive capacityis a result of the system’s ability to self-organize,learn and adapt (Walker et al. 2004, Adger 2006).In this article we aim to present amethodology for assessing the vulnerability toclimate change of both mangroves and fisherfolk,jointly conceived as a Social-Ecological System -SES (Folke et al. 2002, Folke et al. 2003).We dothis by first presenting a general review of the mainelements that determine the vulnerability of theseecological and social systems to the major expectedeffects of climate change on coastal areas. We thenpresent a case study of the coastal region of the Stateof Paraná, southern Brazil, where artisanal fishingvillages coexist with extensive mangrove forests.This scenario is rendered more complex by theexistence of several protected areas on the region.We suggest that the strict preservation of these areas,mainly for biodiversity conservation, can have bothnegative and positive effects on the vulnerability ofmangroves and artisanal fisherfolk to climatechange. Furthermore, these varying effects can workin opposite directions, for example, enhancing thevulnerability of fishing villages, by diminishing theiroptions for livelihood diversification, while loweringthe vulnerability of mangroves, by protectingadjacent land and allowing them to adapt to sealevelrise.Based on this case study we propose amethodology to assess the joint vulnerability toclimate change of mangroves and fisherfolk, whichwe believe can be used in other settings and is basedon the so-called vulnerability or contextual approach(Ford et al. 2010). In this perspective, whichconsiders vulnerability of social-ecological systemsas a starting point, studies must focus on definingthe current processes related to the socialconstruction of vulnerability. That is, how differentsocial, economic and political characteristics,processes and trends determine, in the present,distinct levels of vulnerability. The goal is todevelop policies that are able to improve futureperspectives considering social and environmentalchanges (Kelly & Adger 2000, Van Aalst et al.2008). This approach becomes more relevant if weconsider that, in general, lack of adaptive capacity isthe main factor contributing to increase thevulnerability of human societies, although physicalexposure to hazards is also an important componentof vulnerability of social and natural systems(Nicholls et al. 2007).On a second level of analysis we proposespecific indicators to analyze the impact of theprotected areas on the elements that composevulnerability. Most protected areas in the world wereestablished based on availability of space or politicalviability, without considering climate, or based on astatic view of climate issues. Even though climatechange will probably affect the distribution ofPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk207species, so that many of them may move out ofprotected areas, these will continue to be animportant tool for biodiversity conservation.However, in this new context, it becomes moreurgent to integrate conservation policies with othergeneral strategies for management of landscapes andnatural resources. And it also becomes essential thatthe selection of areas to be protected and themanagement of them and of the landscapes in whichthey are inserted are done with climate change as anexplicit parameter (Hannah et al. 2002, Hannah etal. 2007).Conservation actions should not be limitedto protecting large tracts of ecosystems, but shouldalso consider dynamics of change and adaptation ofsocial-ecological systems and the building ofadaptive capacity in human communities, includingactions aimed at diversifying income sources so thatthese populations become less dependent on naturalresources and are better able to cope with and adaptto the expected and unexpected impacts of climatechange (McClanahan et al. 2008). For this to beachieved, there is a need to identify which elementsof climate change will bring the most importanteffects on each ecosystem and human population,and what will these effects be (Sterr et al. 2000,Hulme 2005). Thus, it is of great importance tounderstand how social and natural systems currentlyrespond to climate-related disturbances, in order toprovide a basis for the development of these newadaptive strategies capable of achieving biodiversityconservation, both inside and outside of protectedareas, together with the social and economicsustainability of human populations.Mangroves and their vulnerability to climatechangeAmong coastal ecosystems, mangroves areof great ecological, economic and social importance.Occupying most of the protected and semi-protectedcoasts in tropical and subtropical regions, theystabilize coastlines, prevent erosion and function asa barrier to storms. They provide refuge, feeding andreproduction sites for a great variety of animals,including commercially important species, andultimately help to sustain and restore fishing stocks.Mangroves are also a source of organic matter forother coastal ecosystems; they provide adequatesites for aquaculture; their sediments are sinks forpollutants and terrigenous sediments; and they haveaesthetic and spiritual value for many human groups.Besides these functions, many mangrove productsare directly explored by coastal populations,especially wood, used as fuel and building material,but also tannins and other plant extracts (Lacerda2002, Agrawala et al. 2003, Walters et al. 2008,Valiela et al. 2009).Mangroves are amongst the most threatenedcoastal ecosystems. In the Americas, an estimated38% of mangrove areas have already been lost, at anannual rate of 3.62% (Valiela et al. 2009). However,South America had the lowest rate of mangrove lossamong all world regions, only 0.18%, or 4,000hectares, in the 2000 - 2005 period (FAO 2007).Threats to mangroves have two main origins: on theone hand human occupation and unsustainablepatterns of resource usage threaten their existenceand limit available space for migration, besidesaffecting factors such as sediment supply, thevolume of groundwater and the discharge ofnutrients and pollutants. On the other hand, theeffects of global climate change, especially sea-levelrise, pressure the frontward margin of mangroves,causing erosion, tree mortality and loss of forestarea. As active contributors to the degradation ofmangroves, coastal populations may end upeliminating the very ecosystem that provides themresources and protection against the impacts ofclimate change (Taylor & Sanderson 2002).The ability of mangroves to respond to sealevelrise depends on many factors, including coastaltopography and the presence of barriers to landwardmigration. This response depends on their ability toaccumulate sediments and promote accretion, aprocess which is regulated by a series ofgeomorphological, climatic and hydrologicalcontrols over sediment supply, primary production,decomposition, subsidence and autocompaction, allof which are extremely variable from one site toanother (Cahoon & Hensel 2006). Vulnerability alsodepends on their ability to migrate, following sealevelvariations. Though significant increases in totalmangrove area have been recently reported for thenortheastern Brazilian coast (Maia et al. 2006,Lacerda et al. 2007, Lacerda 2009), migration andcolonization of new areas may be limited by humanoccupation of adjacent areas, which restricts thisecosystem’s capacity to adapt to new conditions(Scavia et al. 2002). Mangroves are unable to followsea-level rise when the surface elevation rate islower than the relative sea-level elevation rate. Thishas been observed in some recent studies (Gilman etal. 2008), although some other studies have shownthe opposite (Alongi 2008), indicating there is aneed for more long-term observations in a largernumber of sites.Mangroves will probably suffer acombination of positive (rise in atmospherictemperature and CO 2 concentration) and negative(rise in saline intrusion and erosion) effects ofPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

208L. F. D. FARACO ET ALLIclimate change, and the balance between the twowill largely depend on site-specific factors (Saenger2002). Because of the multiplicity of expectedresponses of mangroves to these changes, a morerealistic approach would be to categorize anddelimitate forests according to their level ofvulnerability.Less vulnerable mangroves would be thoselocated in areas with high tidal range (> 5 meters), inhumid tropical coasts and/or in areas close to themouths of large rivers or on their margins, in remoteareas with little human occupation, in areas withlarge nutrient supply, those growing on deep soils,with available space for landward migration and inregions with large extensions of well developedmangrove stands, which are a source for propagulesand seeds. These conditions are found, for example,on the northern coast of Brazil. Highly vulnerablemangroves would be those situated on small islands,growing on calcareous soils, in regions without largerivers, in arid regions, in places subject to groundsubsidence, in areas with low tidal range and with nosediment supply, and mangroves whose expansion isblocked by human occupation or a steep slope(McLeod & Salm 2006, Alongi 2008, Lovelock &Ellison 2007). Therefore, the main variables thatshould be considered for the analyses of mangroveresponses to relative sea-level rise are: topography,sediment sources, rate of sediment supply, area ofthe drainage basin, tidal range, coastal dynamics andthe mean rate of sea-level rise (Soares 2009).Even though the predicted impacts ofclimate change on mangroves will vary betweendifferent ecosystems and regions, it is important toconsider that they will combine with, and evenintensify, other stress factors, potentially aggravatingoverall conditions. Thus, the survival of theseecosystems in a climate change scenario depends ontheir adaptive capacity, but also on the intensity withwhich human activities are undermining thiscapacity (Scavia et al. 2002). For example, incomparison with prior sea-level rise events inEarth´s history, nowadays most coastal regions areaffected by human activities, including cities andinfrastructure, which limit the possibility ofmangroves migrating towards the continent inresponse to climate change. In addition, if weconsider the large number of people living close tomangroves and depending directly on them forsurvival, and at the same time functioning as asource of impact and contributing to lower theirresilience, it becomes of surmount importance toanalyze these systems together, hence our SESapproach. The development of management policiesand strategies for land occupation and resource usein coastal environments must consider these multipleelements, or otherwise, there is a risk that bothmangroves and human populations will lose(Walters et al. 2008).Fisherfolk and their vulnerability to climatechangeAmong coastal populations, those thatdepend on the direct use of natural resources, suchas fisherfolk, are especially vulnerable to climatechange. Worldwide, an estimated 120 million peopledepend directly on fishing for their survival, 95% ofwhich live in developing countries, where the greatmajority is engaged in artisanal fisheries (Allison &Ellis 2001). Artisanal fisherfolk in tropical andsubtropical regions are usually highly dependent,directly or indirectly, on resources and servicesprovided by mangroves, which makes them jointlyvulnerable to climate change, since those thatdepend on marine resources as a source of food arehighly vulnerable to its impacts, both in terms ofhealth and food security, and in economic terms(Nicholls et al. 2007).It is important to discriminate the differentdimensions that compose the vulnerability offisheries systems. This vulnerability is dynamic overtime due to changes in the characteristics of threats,the exposure to them, the sensitivity of the systemand the adaptation actions. A “fisherfolkmangroves”SES may be exposed to different kindsof threats (environmental, economic and political),may have distinct sensitivities to these threats andmay also be more or less resilient (Tuler et al. 2008).Small-scale fisheries face a permanent stateof uncertainty, due to the natural variability of fishstocks and because these stocks are declining as aresult of overfishing, bad management practices, andother factors (Jackson et al. 2001, Mullon et al.2005, Pauly et al. 2005). For these populations,which usually have lower adaptive capacity, socialand biophysical resilience are closely connected andclimate change can increase the uncertaintiesregarding the availability of natural resources (Dolan& Walker 2004), and, as a consequence, threatentheir biological survival and social reproduction.Climate change will bring direct impacts on marinebiodiversity, such as changes in reproduction andmigration periods of several species, an increase indiseases, changes in latitudinal and longitudinaldistribution patterns, changes in population size andcommunity composition and changes in thehydrological cycle, with effects on biodiversity andenvironmental services (Gitay et al. 2002). Much ofthis is already evident in different ecosystems andbiomes all over the planet (e.g. Walther et al. 2002,Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk209Parmesan & Yohe 2003, Parmesan 2006). Thesechanges can alter seasonal and distributional patternsof fish species that are explored by artisanalfisheries, potentially impacting local livelihoods.Faced with the restrictions imposed on them by thenatural environment, these fisherfolk are forced toadapt to the seasonal distribution and the ecology offish, both strongly influenced by climate (Iwasaki etal. 2009).Climate change will also bring greatervariability and uncertainty regarding weatherconditions, which impact directly on artisanalfisherfolk, whose fishing gear limit their mobilityand ability to operate in adverse conditions. In faceof these changes in the environment, traditionalknowledge accumulated by these populations, whichused to guide them during their fishing activities inan efficient and safe way, may become useless (Ford& Smit 2004).Uncertainties also derive from political,economic and social contexts. Market variationsand changes and inadequacies of the rules thatregulate fishing activities are sources of variabilityand stress that constantly threaten the livelihoods offisherfolk (Marschke & Berkes 2006). Other factorssuch as the lack of external institutional support anderosion of traditional resource use systems can leadto a rise in the vulnerability of these livelihoods(Kalikoski et al. 2010). In addition to fisheriesmanagement rules, small-scale fishermen are alsosubject to the effects of other types of rules, such asthose concerned with biodiversity conservation. Forthe sake of biodiversity, many restrictions areimposed on the access to and usage of coastalenvironments and resources. These actions mayfunction as factors that increase the vulnerability ofthe system by making it more sensitive or bylimiting its capacity to respond.These factors (environmental laws, environmentaldegradation, increase in variability anduncertainty related to climate and fish stocks)may result in a reduction of the adaptive capacity ofcoastal populations, potentially aggravating theeffects of climate change. As an example, fishingvillages that depend on a small number of species,and that have few options for diversifyingtheir income sources, tend to be very vulnerableto changes in fish stocks. On the other hand,the consequences of climate change can makethe implementation of biodiversity conservationpolicies ever more difficult, if they fail to takeinto account this new scenario, and if they failto include in their elaboration and implementtationprocesses those that are directly affected bythem.A case study: the coast of Paraná State, southernBrazilWith environmental problems and landoccupation patterns partially similar to otherdeveloping countries, the coast of Brazil is alsoexposed to extreme climatic events, such as stormsurges and flooding, with risks for natural systems,infrastructure and human settlements. The rise ofmean sea-level in Brazil, which is already occurringin most measuring sites, although still small, tends toadd to the effects of these other phenomena,bringing such consequences as an acceleration ofcoastal erosion, a magnification of flooding events,the rise of water tables and increased salinization ofrivers, estuaries and aquifers (Szlafsztein 2005).Most human occupation on the Braziliancoast derives from urbanization and the expansion ofactivities such as tourism, ports, commerce andindustry, which are concentrated on the roughly 55%of the coast which are more densely populated(Neves & Muehe 2008). In these areas, the impactsof extreme events and climate change tend to beeconomically and socially important, as they affectgreat concentrations of infrastructure and humanpopulations. On the remaining parts of the coast,there is a predominance of SES in which humanpopulations, such as artisanal fisherfolk, dependdirectly on the exploration of natural resources, withmany of them still using traditional practices.A fundamental interface between the socialand the ecological components of a fisheries systemis the relation between fishing populations and thecoastal environments from which they extract theresources that sustain their livelihood. In Brazil,mangroves play an important role on supportingboth coastal ecosystems and human populations.Mangroves occur along most of the Brazilian coast,from the extreme north (Cabo Orange, Amapá,04º30’ N) to the city of Laguna, in the southern stateof Santa Catarina (28º56’ S), covering mostintertidal areas (Schaeffer-Novelli et al. 1990).Coastal populations in Brazil who rely on the directexploration of marine resources are highlydependent on mangrove ecosystems. As an example,in the northern state of Pará, at the estuary of theCaeté River, in the city of Bragança, over 80% ofthe population base their livelihoods on mangroves,and around 68% obtain income directly frommangrove products (Glaser 2003).In spite of their importance for humanpopulations, there are few studies dealing with thevulnerability to climate change of coastalecosystems in Brazil. Even less common are studiesthat analyze both biophysical and socioeconomicaspects in an integrated manner. Some studiesPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

210L. F. D. FARACO ET ALLIfocused on observing changes in the distribution ofmangroves as a response to relative sea-level rise(Almeida et al. 2008). Gathering results from thisand other studies, Soares (2009) has proposed aconceptual model for the study of the response ofmangroves to climate change, but it focuses,fundamentally, on biophysical aspects.In the extensive mangrove forests of thenorthern coast of Brazil, many long-term studieshave been developed, especially as part of theMADAM project (Berger et al. 1999). Some ofthese focused on biophysical dynamics, such as thetemporal analysis of mangrove distribution byCohen & Lara (2003), who concluded that mangrovestands are losing area in the seaward margin andmigrating landward, possibly as a response torelative sea level rise, but that this migration islimited by local topography. Other studies analyzedthe dependency of local populations on mangroveresources (Glaser 2003), while there were alsostudies that related the response of mangroves tosea-level rise to socioeconomic matters such as landuse and occupation (Lara et al. 2002). In otherregions of the Brazilian coast, studies have measuredextension, retraction and migration of mangroveforests, but without relating them directly to climatechange (e.g. Lacerda et al. 2007).The coast of the Brazilian southern state ofParaná is dominated by the Paranaguá EstuarineComplex (PEC), whose physical, chemical andbiological properties were described by Lana et al.(2001). It has extensive intertidal flats, whichtotalize around 295 km 2 , mostly covered bymangroves (Fig. 1). The whole region is part of aBiosphere Reserve and of the Atlantic RainforestBiome. Around 70% of the region’s surface area isstill covered by this type of forest and associatedecosystems, in stark contrast with most of theBrazilian coast, where this ecosystem has beenlargely destroyed (SOS Mata Atlântica/INPE 2009).The study case site (Fig. 1) is centeredaround the northern part of the PEC, in themunicipality of Guaraqueçaba, where there are twocoastal protected areas (PAs), managed by theBrazilian Federal Government: GuaraqueçabaEcological Station (created in 1982, it encompassesFigure 1. The Paranaguá Estuarine Complex, its extensive mangroves, numerous fishing villages and the no-takeprotected areas that dominate the northern part of the estuary. Source: Adapted from an original map designed by Prof.Mauricio A. Noernberg, CEM/UFPR.Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk211around 11,500 hectares of mangroves around theBays of Laranjeiras and Pinheiros) and SuperagüiNational Park (created in 1989, it protects around34,000 hectares in the islands of Superagüi andPeças, which include a variety of coastalecosystems).Guaraqueçaba is a sparsely occupied areawith a population of 7,890 inhabitants distributedover 2,315 km 2 . It is one of the poorestmunicipalities in the state of Paraná, with a HumanDevelopment Index of 0.659 (IPARDES 2010).Fishing is the main activity in around 40 villageslocated on the margins of the estuary where there arean estimated 2,100 families of artisanal fisherfolk,most of them living close to mangroves and theaforementioned PAs (Martin & Zanoni 1994,IPARDES 2010). Although of little regionaleconomic importance, fishing is locally of highsocial and economic importance (Borges et al.2006).Artisanal fisheries in this area have beenchanging over time due mainly to factors such asmarket changes, demographic dynamics(immigration) and technical innovation, which ledpart of the local fisheries to more intensified andmarket-oriented practices, while some of the otherpractices have disappeared (Andriguetto-Filho 2003,Andriguetto-Filho et al. 2009).Environmental problems in the coasts ofBrazil and Paraná affect fisherfolk in diverse ways,causing a series of conflicts, such as thedisplacement of these populations to inappropriateareas, disputes over fishing grounds among artisanaland industrial fisheries and aquaculture,contamination and depletion of fishing stocks,among others. Cleavages are also observed amongartisanal fisherfolk, between those that aretraditionally linked to fishing and the opportunists,and between those that use predatory techniques andthose that avoid them (Andriguetto-Filho 1999).These problems threaten the survival and socioeconomicreproduction of these populations.Besides environmental degradation and thedisputes with industrial fisheries, there are evidencesof conflicts between environmental regulations andthe economic activities of local populations,especially those engaged in small-scale fishing andagriculture. Many studies point to environmentalconflicts associated with the creation of PAs in thecoast of Paraná, especially in the municipality ofGuaraqueçaba (Andriguetto-Filho 1993, Martin &Zanoni 1994, Zanoni & Miguel 1995, Pedroso Jr.2002, Cunha et al. 2004, Miranda 2004, Teixeira2004).Brazilian mangroves are also included in notakePAs, besides being protected by many differentlaws (Martin & Lana 1994). This may contribute tothe resilience of natural systems, but can also affectthe vulnerability of human populations. Large tractsof Atlantic Rainforest and mangroves in Paranácompose an area of high priority for biodiversityconservation, being classified as of extremely highbiological importance by the Brazilian Ministry ofEnvironment (MMA 2007). A great variety ofecosystems and species of interest for conservationis reflected in the existence of many PAs in thisregion. These areas have been created here since the1980s and include no-take reserves as well as“sustainable use” ones. They cover a large portion ofthe region: around 76% of the northern coast ofParaná is included inside PAs, of which 13%(59,440 hectares) in no-take reserves such asNational Parks and Ecological Stations. There arealso specific rules to protect the Atlantic Rainforest,which also limit the possibilities for occupation ofthe land and usage of natural resources. This standsin stark contrast with the process of rapid destructionof natural ecosystems that characterized the firstcenturies of human occupation of the Brazilian coast(SOS Mata Atlântica/INPE 2009).In no-take PAs, and especially on those thatinclude mangroves, there is a prevalence of strictconservation rules, which aim at completely banningthe direct use of natural resources. Mangroves werethe first ecosystem in the region to be included in ano-take PA (the above mentioned GuaraqueçabaEcological Station is composed mainly ofmangroves). For this ecosystem, strict no-take rulesapplied equally to all mangroves represent asimplistic view of the ecosystems structural andfunctional characteristics, because they are based onthe misconception that all mangrove stands areequally productive, that resources are equallydistributed and even that human groups access andexplore all areas in the same way and with the sameintensity. This last assumption also ignoresterritoriality relations between human groups andeven the community conservation mechanisms thatmay be in use by them. All of this makes these rulesnot only inadequate but also inefficient, and evenunfair, paradoxically resulting in open-accesssituations and in the environmental degradation ofthese ecosystems. Extensive discussions on thenovel conflicts created by environmental legislation,mainly in the northern sector of the ParanaguáEstuarine Complex, were provided by Martin &Zanoni (1994), Lana (2003) and Raynaut et al.(2007).Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

212L. F. D. FARACO ET ALLIIn spite of the fact that most mangroves inthis region are located inside no-take PAs, localpopulations use mangrove fishing resources on aregular basis. Although mangrove products, such ascrabs (Ucides cordatus), oysters (Crassostrearhizophorae) and shrimp (Penaeus schmitti),account for a relatively small percentage of totalartisanal fish production, there are somecommunities that rely on crab extraction and oystersemicultivation as their main economic activities. Itis also evident that a practical situation of openaccess to these resources has lead tooverexploitation. Additionally, a general crisis infisheries in the region has led to an increase in theseactivities in local mangroves. In some areas there isa direct relation between the fishing village and thenearby mangrove, with examples of locallyestablished rules of access. But in many casesmangrove resources are explored by people comingfrom distant places, which is a source of conflictsbetween different groups of fisherfolk in this region(Miranda 2004).Even when they are not included in no-takePAs, the direct exploration of mangrove resources,except fishing resources, is considered illegal inBrazil, which is also a source of conflicts with thosepopulations that traditionally explore them (Glaser& Oliveira 2004). In the end, this situation ofpermanent illegality experienced by those thatexplore mangrove resources results in a normativeinsecurity and in a reduction of acceptance ofenvironmental rules, with growing hostility betweenlocal populations and environmental authorities. Itresults also in the adoption of economicallyinefficient, ecologically inappropriate and sociallyunequal practices by these populations (Glaser et al.2003).Therefore, although they might havecontributed for the conservation of natural resourcesin this region, land management and biodiversityconservation policies also generate conflicts whenthey restrict occupation of certain areas and prohibitexploration of resources. The impacts of these rulesare unevenly distributed among different socialgroups. This situation results in negative impacts onthe livelihoods of the populations that inhabit areasconsidered important for conservation. This is one ofthe reasons why official management actions are,more often than not, inefficient in protecting naturalresources, especially common pool resources such asfishing and mangrove ones.In a context such as the one observed in thecoast of Paraná, one of the poorest regions in thestate (Pierri et al. 2006), this scenario becomes evenmore problematic, as local small-scale fishingpopulations have limited access to political, financialand social assets, which aggravates the impacts thateven small fluctuations in natural resourcesavailability, or restrictions on the access to theseresources, can have on their livelihood and survival.Therefore, environmental regulations become one ofthe sources of variability and disturbance for theselivelihoods, acting together with otherenvironmental (variations in fish stocks, environmentaldegradation, extreme weather events) andeconomic (market fluctuations, low income, lack orinadequacy of support mechanisms) factors, andcontributing to increase the vulnerability ofpopulations and ecosystems to climate change. Thevulnerability of the SES of small scale fisheries inthis context puts at risk the survival conditions ofthousands of people, compromising their socialreproduction.As we recognize the connections betweencoastal ecosystems and human populations, there isa growing need for interdisciplinary research on theeffects of climate change and to translate researchresults into better policies. It is important tocomparatively understand these dynamics, analyzinghow this multitude of factors affects the livelihoodsand adaptive capacity of coastal populations andecosystems, in order to better adapt conservationstrategies, with an eye to both biodiversityconservation and social reproduction of humanpopulations. Research in this domain must find waysto influence the making of environmental policiesand rules.A methodology for assessing vulnerability toclimate change and the impacts of protectedareasThe development of conservation policiesfor coastal zones must consider the need tounderstand the different systems - socioeconomic,geomorphologic and ecologic - in an integratedmanner, so that vulnerability can be analyzed for thecoastal area as a whole. Although climate is one ofthe main sources of hazards for the coastal zone, it isnot the only source of change and vulnerability. Itneeds to be considered together with other factors sothat management instruments are useful forintegrated coastal management. Thus, whiledeveloping methodologies for analyzingvulnerability, the components of this analysis mustprovide information about all the processes thatdefine the behavior of the whole system (McFadden& Green 2007).A research framework aiming to analyze therelations between social and ecological systems hasto face the challenge of understanding cross-scalePan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk213interactions between phenomena and processes.Nevertheless, considering that responses to theimpacts of climate change will consist primarily ofindividual responses to day-to-day changes on alocal scale, there is a need for this type of study tohave a multiscale perspective which can be appliedto the analysis of adaptive capacity on the level ofthe communities (Dolan & Walker 2004).Therefore, the proposal for working with alocal case study as the starting point is based on theidea that the information to be collected can form thebasis for a bottom-up analysis aimed at elucidatingsome of these interactions, contributing with studiesof global change, which have usually focused onglobal scale models as the starting point, thenextrapolating to the regional and local levels(Wilbanks & Kates 1999). With a focus on the localcontext, extrapolations could work on the oppositedirection, emerging from comparisons betweendifferent communities – for example, usingproximity to no-take protected areas as theindependent factor – or even between differentsocieties, comparing the results of the small-scalestudy with similar realities in other countries – forexample, where artisanal fisherfolk and mangrovesmay coexist under different conservation policies.These comparisons would aim on identifyingthose characteristics of communities and theirenvironments that contribute to enhancing orlowering vulnerabilities, and the elements of theadaptation strategies that turn out to be moreefficient (Smit & Wandel 2006). This would resultin scientific explanations of the specific realities, butnot necessarily on guidelines that could beuniversally applied to the formulation of policies,because the great variety of social and ecologicalcontexts makes it difficult to develop homogeneousmanagement solutions.Supported by this logical background, wepropose herein a research framework to assess thevulnerability to climate change of mangroves andfisherfolk. Using the case study of the ParanaguáEstuarine Complex in southern Brazil as an examplewe further include in the proposed methodologyindicators to analyze the impacts of biodiversityconservation actions, especially no-take PAs, on thisvulnerability. Following this perspective, theframework aims to understand how fisherfolkrespond to changes in the status of the assets(biophysical, cultural, political and institutional) onwhich they base their livelihoods, and if this status isaffected by environmental changes or changes inaccess and entitlements to these assets, specificallyas a result of climate change and the implementationof no-take PAs.Specific steps in the methodology include:(a) Evaluation of the vulnerability of mangroves toclimate change, especially regarding their exposureand sensitivity to sea-level rise, and their adaptivecapacity; (b) Evaluation of the vulnerability offisherfolk populations to climate change,considering the exposure of villages to sea-level riseand extreme climatic events, their position inrelation to no-take PAs, and the elements thatenhance or diminish their adaptive capacity; and (c)Analysis of the effects of no-take PAs on thesevulnerabilities, through impacts on sensitivity andadaptive capacity of both fisherfolk and mangroves.Such an approach can also be useful to analyzeregional biodiversity conservation policies regardingtheir adequacy to deal with climate change relateddynamics, aiming to identify how they can beadapted to contribute to building adaptive capacity,both of mangroves and fisherfolk, to respond tothese changes.To achieve these objectives, we propose ananalysis of vulnerability in two sections,corresponding to two different scales (regional andlocal) and considering the main components ofvulnerability: exposure, sensitivity and adaptivecapacity. By analyzing the components ofvulnerability in different situations we aim tounderstand which elements of the system are directlyaffected by change, being it regulatory,environmental, social or economic (Tuler et al.2008), but with an emphasis on the impacts ofexisting PAs on the factors that conditionvulnerability.Preliminary steps in the research projectshould include the definition of the spatial andtemporal scales for the study, consideringbiophysical and socioeconomic criteria, and the timescale in which management decisions are taken;collection of information on the biophysicalenvironment and on socioeconomic and culturalcharacteristics of the area; identification of specificrules and policies that affect the area; and, choosingthe specific sites for detailed data collection,according to the population, ecosystems and policiesof interest (Harvey et al. 1999).We divide the analysis according to the threecomponents of vulnerability and the threesubsystems being considered: social (fisherfolk),natural (mangroves) and social-ecological (representingthe interaction between the other twosubsystems; in our case study, it concerns mainly thefisheries production system, or the patterns ofutilization of natural resources by coastalpopulations). It should also be considered thatfactors influencing vulnerability can be both socialPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

214L. F. D. FARACO ET ALLIand ecological, and internal or external to the system(Füssel 2007).Vulnerability should always be measured inrelation to specific environmental hazards, whichcan vary according to the specific characteristics ofthe setting under consideration. For the coastalpopulations and environments of our case study, wedefined three expected effects of climate change asthe main threats: relative sea-level rise, an increasein the frequency and intensity of extreme climaticevents, and an increase in uncertainty and variabilityrelated to the availability of fishing resources.Exposure to these threats is mainly relatedwith the characteristics of the sites where villagesand mangroves are located, especially theirproximity to the sea and the topography and slope ofthe terrain. Thus, for the analysis of this element adigital elevation model (DEM) of the coastal areaneeds to be constructed. This can be accomplishedthrough the use of remote sensing, with a highresolutionsatellite image of the region, or, ideally,with LIDAR (Light Detection and Ranging) data(Gesch 2009).To improve the quality of the model, theupper boundary of mangrove forests should bedelimitated and considered as equivalent to the meanhigh tide in the region. The DEM can then beconstructed through interpolation of existingelevation data, considering this high tide line and thefirst topographic contour available in local maps (10meters in the case of the Brazilian coast). To furtherimprove the accuracy of the model, the interpolationcan be fed with a series of elevation data pointsestablished along the margins of the estuary, in thearea between the sea and the first topographiccontour.This digital elevation model, even if notaccurate enough to produce detailed scenarios ofareas prone to future flooding, especially if LIDARdata is not available, will allow a classification ofthe coastline in categories representing differentlevels of exposure to sea-level rise and flooding. Themain goal is to identify the location and topographyof mangroves and villages and the land-use patternsin the low-lying areas, up to 10 meters above sealevel.This focus on topography and on the existenceof barriers to mangrove migration is justified by thelack of consistent data on local and regional sealevelrise. In this case, the analyzed factors work assurrogates or indirect indicators of this dynamics,producing information about the response of theseecosystems to climate change (Gilman et al. 2007).The last component of exposure to be measured isthe magnitude and frequency of occurrence ofextreme climatic events. Information on these eventscan be obtained both from meteorological andhistorical records and from interviews with localinhabitants. We suggest as a proxy for this themeasurement of the number of fishing days lost dueto bad weather conditions.For the analysis of exposure, we can alreadyidentify opposite effects on the social and naturalsubsystems. A gentle slope means a higher exposureto sea-level rise and storm surges, increasing thevulnerability of villages, considering that the flatterthe land the larger the area that would be flooded.But, for mangroves, a gentle slope in adjacentlandward areas means that they have available spaceto migrate towards the continent, although land-usepatterns may create barriers to this migration. In ourspecific setting, this spatial analysis of land adjacentto mangroves will also evaluate the situation of localfishing villages regarding their sensitivity to meansea-level rise and the impacts of protected areaswhich exclude human occupation. In this case, thefactor that will be analyzed is “coastal squeeze”, thatis, if a rise in sea-level will put pressure on humanoccupations and if these will be able to respond.Many villages in this region are placed between theocean and the PAs, being susceptible to coastalsqueeze if sea level rises.Still regarding sensitivity, which relates toeffects that hazards can have on the systems, thegeneral housing, health and food conditions ofhuman populations should be measured, togetherwith their dependency on climate-influencedresources (such as fish). For this, we propose, basedon the Livelihood Vulnerability Index methodology(Hahn et al. 2009), to focus on water and foodsources and storage capacity, the percentage ofhousehold members that lost work or school daysdue to health problems, and the distance to theclosest health service (Table I). For mangroves,mapping of the total area occupied by the ecosystemand the forest types potentially affected areindicators of this component.Following this analysis of exposure andsensitivity, some villages, and the mangrovesassociated to them, identified as potentially highlyvulnerable, can be chosen for a more detailedevaluation of factors affecting their adaptivecapacity. That is, besides exposure and sensitivity tobiophysical risk, the other main component ofvulnerability will be analyzed: the capacity of thesepopulations and ecosystems to cope and adapt tochange, the factors that contribute to it, and, in ourspecific case, how the existence of no-take PAsaffects these factors.This analysis of adaptive capacity can bedivided in two steps. The first one focuses on currentPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk215vulnerability, considering exposure to risks andcoping strategies observed in the present, and basedin knowledge of the environment, availableresources and existing strategies. A second step aimsto create future scenarios by estimating changes thatmight occur and possible adaptation actions basedon behaviors already demonstrated in the past, andtheir adequacy to these scenarios. What isconsidered is how communities have dealt withextreme events and disturbances, what conditionsmay change and what opportunities exist for futureadaptation (Ford & Smit 2004).This type of analysis considers currentconditions in these communities, with factors andprocesses that contribute to enhance or diminishtheir capacity to respond to changes, to variationsand to the unexpected. This perspective recognizesthe importance of factors that are not directlyconnected to climate, such as sources of subsistence,assets, access to resources, institutionalarrangements, etc., that condition the vulnerability ofthese populations. This means that vulnerability toclimate change is analyzed together with othersources of stress, with emphasis on the ability ofpeople to respond to risks, changes and threats,potentially generating adaptation proposals thatdiminish vulnerability to climate change through thereduction of exposure or minimization of otheradverse factors. Therefore, besides identifying themost relevant matters for the adaptive capacity ofcommunities, the aim is also to understand theimportance of climatic stresses in comparison toother sources of disturbance to these livelihoods(Tschakert 2007).McClanahan et al. (2008) proposed an indexof adaptive capacity composed of the followingvariables: recognition by the population of thecausalities; anticipation of changes; mobility andoccupational multiplicity of the population; socialcapital; material assets; and available infrastructureand technology. In a similar way, Yohe & Tol(2002) developed a method to estimate adaptivecapacity using eight factors linked to technologicaloptions, availability of resources and theirdistribution in the population, structure ofinstitutions that are important in decision making,the stock of human and social capital, the access torisk spreading processes, characteristics and abilitiesof the decision makers and the perception of thepublic in relation to the causes of stress and themeaning of being exposed to it.Other authors applied and discussed thepertinence of more specific and detailedmethodologies such as “Community RiskAssessment” and “Participatory Rapid Appraisal”(Van Aalst et al. 2008), or the “SustainableLivelihood Approach” (Iwasaki et al. 2009). Whatthese methods have in common is a bottom-upapproach, the direct involvement of communitiesand a focus in analyzing vulnerability to currentevents, as well as strategies and policies based oncurrent and real experiences, in different scales.The “Sustainable Livelihood Framework”(Adato & Meinzen-Dick 2002, Baumann 2000,Divakarannair 2007) is a widely used approach forthe study of the livelihoods of these communitiesthat depend directly on natural resources, as well asthe study of the importance of biophysical, social,cultural, economic, political and institutional factorsthat determine the options held by these populations.This method considers that livelihoods are linked toassets composed of human capital (education,knowledge, health, nutrition, workforce), naturalcapital (the natural resources explored by thecommunity), physical capital (the availableinfrastructure, such as fishing gear and housing),financial capital (savings, credit, income), socialcapital (networks, cooperation, access toopportunities, organization) and political capital(policies, institutions and processes that link theindividual or group to external power structures).It is also important to include in this analysiselements of Environmental History, through, forexample, interviews with elders and analysis ofaerial photographs of the region, that may indicatehistorical patterns of land usage and occupation. Theaim is to understand how the measured factorsbehaved prior to the existence of PAs, and in thisway, relativize the impacts of them on these factors.Considering the studies already mentioned,and the objectives of this work, and also consideringother sources that discuss indicators and criteria forthe evaluation of vulnerability and resilience in SESs(Marschke & Berkes 2006, Tschakert 2007, Tuler etal. 2008, Hahn et al. 2009, Kalikoski et al. 2010),we selected a number of indicators, linked mainly toincome sources of fishermen and their relation tomangroves. Among the indicators usually utilized invulnerability assessments, we considered thefollowing to be more informative and usefulregarding the objectives of this work:- Income: total income, income distributionin communities, diversification of income sources,existence of stable income sources (retirementpayments, etc.), proportion of economically activepopulation.- dependency of communities on fishing andmangrove resources: importance of these resourcesfor their subsistence, access conditions andtendencies of variation on their availability,Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

216L. F. D. FARACO ET ALLIconsidering environmental degradation,overexploitation, climate change, and the restrictionsimposed by environmental rules.- fishing strategies: types of boats andfishing gear available; number of days they areprevented of going out to sea by bad weather; safetyregarding availability of assets related to theirlivelihood (property of means of production, naturalresources and housing, including conditions ofaccess to them); if they have experienced changes infishing activities (diversification, dislocation offishing grounds) in response to economic, politicaland climatic changes.- Market relations: whether they trade theproducts they capture or produce only in localmarkets, only in external ones, or in both; level ofdependency on middlemen.- Organization capacity of the community:means of organization and participation indiscussion forums, associations, etc.; perceptionsregarding the efficacy of these forums; responses ofthe community to previous catastrophic events(storm surges, oil spills).- Adaptation and learning strategies,including social cohesion mechanisms: relations ofhelp and exchange of information inside thecommunity and between communities, regardinglivelihood activities; participation of youngermembers on livelihood activities (knowledgetransmission).- Environmental and fisheries managementpolicies and institutional factors: impacts of thesepolicies on livelihoods; existence of financialsupport programs (loans, unemployment insurance,fishing ban period insurance) and the level offishermen participation in these programs.To assess the linked elements that composethe adaptive capacity of the SES, we propose ameasurable index of the stress level oranthropization level of those mangroves that areused by these populations. Estimates on the presentstate of mangrove forests and fishing resources, suchas oyster, mussels and crabs, including theavailability of resources and how, where and in whatintensity they are explored, can be used as a proxy ofthe impacts of human activities on the ecosystem'sresilience.This information will help to characterizethe main component of the relation betweenmangroves and fishermen: the usage patterns ofmangrove resources by these human populations.Additionally, the level of human usage of aparticular mangrove shows its importance for thatpopulation. If a highly explored and usefulmangrove has a low resilience to currentdisturbances and projected climate change, it mustbe the object of adequate management, one thatcontributes to increase the resilience of the SES as awhole.The effect of protected areas, or other factorof interest, on the vulnerability of mangroves andfisherfolk should be dealt with at a second level ofanalysis. For our case study, we propose fourspecific indicators of these effects: (a) the distanceof villages to the closest no-take protected area,indicating the probability of them suffering with“coastal squeeze” and of potential conflicts withbiodiversity conservation norms; (b) the proportionof the income of local fisherfolk that comes frommangroves located inside no-take PAs, indicatingthe impact that these areas can have on theirlivelihood if these strict norms are fully enforced; (c)the proportion of local inhabitants that used to havea more diversified livelihood, practicing agricultureand extractivism, and who abandoned theseactivities because of the prohibitions brought onthem by PAs; and, (d) the proportion of inhabitantsthat have suffered other type of restrictions on theirlivelihoods, such as limitations on improvement ofhousing conditions, due to PAs.These data can be obtained from a numberof different sources and utilizing a variety oftechniques. Part of the social and economic data isavailable in government agencies, from projectsdeveloped in the region by non-governmentalorganizations, and as published scientific literature.Primary data shall be obtained directly on fishingvillages using semi-structured interviews, contactswith key informants (such as community leaders andprotected area managers) and direct observation ofspecific forums. Biological data from mangroves canbe collected directly on site, using scientificsampling techniques, or with the help of localfishermen.Table I summarizes the steps that areproposed in this methodology. They are divided intwo scales (regional and local) and categorizedaccording to the component of vulnerability theyrefer to, the type of capital (human, social, political,financial, natural and physical), the type ofinformation to be collected and analyzed, themethod and the source and type (quantitative orqualitative) of data.After gathering the data it must be decidedwhether they will be summarized into an index ofvulnerability. Most of the indicators proposed in thismethodology are quantitative and can beparameterized and used to compose such an index.Although a focus on quantitative indicators andbuilding of a composite index can oversimplify aPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk217Table I. Summary of the indicators that compose the proposed methodology, considering the case study described above.VulnerabilitycomponentExposureSubsystemMangrovesType of CapitalNaturalIndicatorTopography/SlopeInformation collected/analyzedSurface elevation data; contour curves;upper limit of mangroves; location of PAs;relative sea-level rise scenariosMethodRemote sensing;spatial analysis;digital elevationmodelExposureFisherfolkNaturalPhysicalSlope; distancefrom village tothe seaSurface elevation data; contour curves,upper limit of mangroves; location of Pas;location of fishing villages; relative sealevelrise scenariosRemote sensing;spatial analysis;digital elevationmodelExposureSensitivitySensitivitySESFisherfolkFisherfolkNaturalPhysicalSocialSocialSocialHistoricoccurrence ofextremeclimatic eventsFood and WaterHealthFrequency of climatic events thatsignificantly affect the villages; number offishing days lost due to bad weatherType of water source; water storagecapacity; food storage capacity, percentageof food coming from own production;number of months a year when there isshortage of foodPercentage of members of household wholost work or school days due to healthproblems; distance to closest health serviceCollection ofsecondary data;interviews with localinhabitantsSecondary data;field observations;household surveysHousehold surveysSensitivitySensitivityAdaptivecapacity/Ecological resilienceMangrovesSESMangrovesNaturalNaturalPhysicalSocialPoliticalHumanNaturalMangrove typeDependency onfishingresources/diversity ofincome sourcesEcosystemhealthTotal area and mangrove types potentiallyaffected by sea-level riseProportion of income derived from fishingand mangroves resources; location ofexploited resourcesAbundance of fishing resources inmangroves and usage patterns; signs ofstress in the ecosystemRemote sensing;spatial analysisSecondary data;household surveysSampling of primarydataAdaptivecapacity/Ecological resilienceMangrovesNaturalCapacity tocope with sealevelrisePresence of barriers/patterns of occupationof landward areas; level of protection ofmangroves and landward areas; proximity torivers and size of drainage basinRemote sensing;spatial analysis;field surveysData sourceSatellite images;LIDAR dataSatellite images;LIDAR dataGovernmentagencies;scientificpublications;fisherfolkGovernmentagencies;scientificpublications;fisherfolkFisherfolkSatellite imagesGovernmentagencies;scientificpublications;fisherfolkDirectmeasurement inmangrovesSatellite images;directobservationsData typeQuantitativeQuantitativeQuantitativeandqualitativeQuantitativeandqualitativeQuantitativeQuantitativeandqualitativeQuantitativeandqualitativeQuantitativeQuantitativeandqualitativeScaleRegionalRegionalLocalLocalLocalRegionalLocalLocalLocalPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

218L. F. D. FARACO ET ALLITable I. Continued.VulnerabilitycomponentSubsystemAdaptivecapacityFisherfolkAdaptivecapacityFisherfolkSESAdaptivecapacityFisherfolkAdaptivecapacityFisherfolkAdaptivecapacityFisherfolkSensitivity/AdaptivecapacityFisherfolkType ofCapitalFinancialPhysicalSocialFinancialSocialSocialPoliticalHumanSocialPoliticalFinancialSocialIndicatorIncome level,distributionand diversityFishingstrategiesdiversityMarketconnectionsCommunityorganizationAdaptationand learningstrategiesImpacts ofbiodiversityconservationpoliciesInformation collected/analyzedDiversity of income sources; percentage of incomecoming from external/stable sources; dependency rateNumber of fisheries practiced in the village; percentageof fisherfolk that own boats and fishing gears; frequencyof usage of mangrove resourcesNumber of places where fisherfolk trade their products;percentage of those that depend on middlemenExistence and number of community organizations;perception of effectiveness of external help duringcatastrophic events; responses of community duringprevious catastrophic events.Frequency and type of help and exchange of informationinside the community and between communities,regarding livelihood activities; percentage of inhabitantswith knowledge of the threats related to climate changeDistance of villages to no-take protected area; proportionof income of local fisherfolk that comes from mangroveslocated inside no-take PAs; proportion of localinhabitants who abandoned livelihood activities andsuffered restrictions because of PAs.MethodSampling ofsecondary data;household surveysSecondary dataHousehold surveysInterviews with keyinformantsSecondary dataHousehold surveysInterviews with keyinformantsSecondary dataInterviewsDirect observationof forumsFocal groupdiscussionsInterviews; focalgroup discussionsInterviews; focalgroup discussionsData sourceGovernment agencies;scientific publications;fisherfolkGovernment agencies;scientific publications;FisherfolkGovernment agencies;scientific publications;FisherfolkGovernment agencies;scientific publications;FisherfolkFisherfolkFisherfolk; keyinformantsData typeQuantitativeQuantitativeQuantitativeQuantitativeandqualitativeQualitativeandquantitativeQuantitativeandqualitativeScaleLocalLocalLocalLocalLocalLocalPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk219complex reality, this can be useful for comparisonsbetween villages regarding their vulnerability toclimate change, the effects of protected areas and therelative importance of each vulnerability component,which can help in defining management priorities.Furthermore, qualitative information can also beused to guide interpretations of the observedsituation. An important decision for composing avulnerability index is whether each sub-componentshould have a different weight. This demandscareful judgment of the reality being studied. Wepropose to follow the “balanced weighted averageapproach” used in composing the LivelihoodVulnerability Index (Hahn et al. 2009), where eachsub-component contributes equally to the overallindex. This allows for a clear identification of thecontribution of each indicator for the composition ofthe overall vulnerability of a population orecosystem, facilitating comparisons and pointingmanagement actions towards the most relevantsituations.ConclusionsSuch a complex situation as the onedescribed requires adequate management measures,ReferencesAdato, M. & Meinzen-Dick, R. 2002. Assessing theimpact of agricultural research on povertyusing the sustainable livelihood framework.Environment and Production TechnologyDivision Discussion Paper N o . 89, IFPRI(March), Washington, DC, 57 p.Adger, W. N. 2006. Vulnerability. Global EnvironmentalChange, 16: 268-281.Adger, W. N., Hughes, T. P., Folke, C., Carpenter,S. R. & Rockström, J. 2005. Social-EcologicalResilience to Coastal Disasters. Science, 309:1036-1039.Agrawala, S., Ota, T., Risbey, J., Hagenstad, M.,Smith, J., Van Aalst, M., Koshy, K & Prasad,B. 2003. Development and Climate Changein Fiji: Focus on Coastal Mangroves. COM/ENV/ EPOC/ DCD/ DAC (2003)4/ FINAL,Organization for Economic Co-operation andDevelopment, Paris, France, 56 p.Allison, E. H. & Ellis, F. 2001. The livelihoodsapproach and management of small-scalefisheries. Marine Policy, 25(5): 377-388.Almeida, P. M. M., Campos, N. S., Chaves, F. O.,Estrada, G. C. D., Rosado, B. B., Silva, J. E.S. & Soares, M. L. G. 2008. Análise dopadrão de colonização de uma planíciehipersalina por espécies de mangue na regiãodo rio Piraquê, baía de Sepetiba (Rio defocused not only on understanding and managing ofimmediate sources of impact on social andecosystem process, but also on increasing theresilience of SESs so that they can support theseimpacts, especially those derived from globalclimate change. The challenge is to develop newpractices and management policies which allow forthe adaptation of productive systems to change, and,therefore, their viability and sustainability. For thisto be achieved we must acquire sound knowledgeabout the elements that compose the vulnerability toclimate change of the SES formed by mangrovesand fisherfolk and the effects of biodiversityconservation policies on these elements. In thiscontext, interdisciplinary research becomes essentialin the characterization and analysis of the differenttypes of impacts, the social and economic practicesof human populations, and the vulnerability ofecosystems, of the environmental services theyprovide and of coastal populations. A sounddiagnosis may lead to more flexible policies,elaborated with stakeholders’ participation, moreadequate to local realities and more inclusive ofstrategies for mitigation and adaptation to climatechange.Janeiro). III Congresso Brasileiro de Oceanografia,Fortaleza, Ceará, Brasil. p. 1-4.Alongi, D. M. 2008. Mangrove forests: resilience,protection from tsunamis, and responses toglobal climate change. Estuarine Coastaland Shelf Science, 76: 1-13.Andriguetto Filho, J. M. 1993. Institutional Prospectsin Managing Coastal EnvironmentalConservation Units in Paraná State, Brazil.Eighth Symposium on Coastal and OceanManagement, New Orleans. Coastal zone '93proceedings - ASCE, p. 2354-2368.Andriguetto-Filho, J. M. 1999. Sistemas técnicos depesca e suas dinâmicas de transformação nolitoral do Paraná, Brasil. PhD. Thesis.Universidade Federal do Paraná, Curitiba,Brasil, 242 p.Andriguetto-Filho, J. M. 2003. A mudança técnica eo processo de diferenciação dos sistemas deprodução pesqueira no Litoral do Paraná,Brasil. Desenvolvimento e Meio Ambiente,8: 43-58.Andriguetto-Filho, J. M., Krul, R. & Feitosa, S.2009. Analysis of natural and social dynamicsof fishery production systems in Parana,Brazil: implications for management andsustainability. Journal of Applied Ichthyology,25: 277-286.Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

220L. F. D. FARACO ET ALLIBadjeck, M-C., Allison, E. H., Halls, A. S. & Dulvy,N. K. 2010. Impacts of climate variability andchange on fishery-based livelihoods. MarinePolicy, 34: 375-383.Baumann, P. 2000. Sustainable livelihoods andpolitical capital: Arguments and evidencefrom decentralization and natural resourcemanagement in India. OverseasDevelopment Institute Working Paper 136.ODI, London, UK, 44 p.Berger, U., Glaser, M., Koch, B., Krause, G., Lara,R., Saint-Paul, U., Schories, D. & Wolff, M.1999. An integrated approach to mangrovedynamics and management. Journal ofCoastal Conservation, 5: 125-134.Borges, L. M. M., Maulin, G. C. & Andriguetto-Filho, J. M. 2006. Analysis of income sourcesof fishers’ families on the coast of the State ofParaná, Brazil. Journal of Coastal Research,39: 1268-1272.Brooks, N. 2003. Vulnerability, risk and adaptation:A conceptual framework. WorkingPaper 38, Tyndall Center for Climate ChangeResearch, University of East Anglia, Norwich,UK.Cahoon, D. R. & Hensel, P. 2006. High-resolutionglobal assessment of mangrove responses tosea-level rise: a review. Symposium onMangrove Responses to Relative Sea LevelRise and Other Climate Change Effects,Society of Wetland Scientists 27 th InternationalConference, 9–14 July 2006, CairnsConvention Centre, Cairns, Australia, 9-17.Cohen, M. C. L. & Lara, R. J. 2003. Temporalchanges of mangrove vegetation boundaries inAmazonia: Application of GIS and remotesensing techniques. Wetlands Ecology andManagement, 11: 223-231.Cunha, L. H. O., Rougelle, M. D., Von Behr, M. F.& Diegues, A. C. 2004. Comunidadeslitorâneas e Unidades de ProteçãoAmbiental: convivência e conflitos - O casode Guaraqueçaba, Paraná. Série dedocumento e relatórios de pesquisa.NUPAUB-USP, São Paulo, Brasil, 55 p.Divakarannair, N. 2007. Livelihood Assets andSurvival Strategies in Coastal Communities inKerala, India. PhD. Thesis. University ofVictoria, Victoria, Canadá, 254 p.Dolan, A. H. & Walker, I. J. 2004. Understandingvulnerability of coastal communities toclimate change related risks. Journal ofCoastal Research, SI 39: 1-8.FAO. 2007. The World’s Mangroves 1980-2005.Food and Agriculture Organization of theUnited Nations, Rome, FAO Forestry Paper153, 89 p.Folke, C., Carpenter, S., Elmqvist, T., Gunderson,L., Holling, C. S. & Walker, B. 2002.Resilience and Sustainable Development:Building Adaptive Capacity in a World ofTransformations. Ambio, 31(5): 437-440.Folke, C., Colding, J. & Berkes F. 2003. Synthesis:Building resilience for adaptive capacity insocial-ecological systems. Pp. 352-386. In:Berkes, F., Colding, J. & Folke, C. (Eds.).Navigating Social-Ecological Systems:Building Resilience for Complexity andChange. Cambridge University Press,Cambridge, UK, 400 p.Ford, J. D. & Smit, B. 2004. A framework forassessing the vulnerability of communities inthe Canadian Arctic to risks associated withclimate change. ARCTIC, 57(4): 389-400.Ford, J. D., Keskitalo, E. C. H., Smith, T., Pearce,T., Berrang-Ford, L., Duerden, F. & Smit, B.2010. Case study and analogue methodologiesin climate change vulnerability research.WIREs Climate Change, 1: 374-392.Füssel, H-M. 2007. Vulnerability: A generallyapplicable conceptual framework for climatechange research. Global EnvironmentalChange, 17: 155-167.Gesch, D. B. 2009. Analysis of Lidar Elevation Datafor Improved Identification and Delineation ofLands Vulnerable to Sea-Level Rise. Journalof Coastal Research, SI 53: 49-58.Gilman, E. L., Ellison, J. & Coleman, R. 2007.Assessment of mangrove response toprojected relative sea-level rise and recenthistorical reconstruction of shoreline position.Environmental Monitoring and Assessment,124: 112-134.Gilman, E. L., Ellison, J., Duke, N. C. & Field, C.2008. Threats to mangroves from climatechange and adaptation options: A review.Aquatic Botany, 89: 237-250.Gitay, H., Suárez, A., Watson, R.T. & Dokken, D. J.(Eds). 2002. Cambio climático ybiodiversidad. Grupo Intergubernamental deexpertos sobre el cambio climático,PNUMA/UNEP, 85 p.Glaser, M. 2003. Interrelations between mangroveecosystems, local economy and socialsustainability in the Caeté estuary, NorthBrazil. Wetlands Ecology and Management,11: 265-272.Glaser, M., Berger, U. & Macedo, R. 2003. Localvulnerability as an advantage: Mangroveforest management in Pará state, North BrazilPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk221under conditions of illegality. RegionalEnvironmental Change, 3-4: 162-172.Glaser, M. & Oliveira, R. S. 2004. Prospects for theco-management of mangrove ecosystems onthe North Brazilian coast: Whose rights,whose duties and whose priorities? NaturalResources Forum, 28: 224-233.Hannah, L., Midgley, G. F., Lovejoy, T., Bond, W.J., Bush, M., Lovett, J. C., Scott, D. &Woodward, F. I. 2002. Conservation ofbiodiversity in a changing climate.Conservation Biology, 16(1): 264-268.Hannah, L., Midgley, G., Andelman, S., Araujo, M.,Hughes, G., Martinez-Meyer, E., Pearson, R.& Williams, P. 2007. Protected area needs ina changing climate. Frontiers in Ecology andthe Environment, 5(3): 131-138.Hahn, M. B., Riederer, A. M. & Foster, S. O. 2009.The Livelihood Vulnerability Index: Apragmatic approach to assessing risks fromclimate variability and change - A case studyin Mozambique. Global EnvironmentalChange, 19(1): 74-88.Harvey, N., Clouston, B. & Carvalho, P. 1999.Improving coastal vulnerability assessmentmethodologies for integrated coastal zonemanagement: An approach from SouthAustralia. Australian Geographical Studies,37(1): 50-69.Hulme, P. E. 2005. Adapting to climate change: isthere scope for ecological management in theface of a global threat? Journal of AppliedEcology, 42: 784-794.IPARDES. 2010. Caderno Estatístico do Municípiode Guaraqueçaba. World Wide Webelectronic publication accessible at http://migre.me/3aH7z. (Accessed 05/03/2010).Iwasaki, S., Razafindrabe, B. H. N. & Shaw, R.2009. Fishery livelihoods and adaptation toclimate change: a case study of Chilikalagoon, India. Mitig Adapt Strateg GlobChange, 14: 339-355.Jackson, J. B. C., Kirb, M. X., Berher, W. H.,Bjorndal, K. A. & Botsford, L. W. 2001.Historical overfishing and the recent collapseof coastal ecosystems. Science, 293: 629-38.Kalikoski, D. C., Quevedo Neto, P. & Almudi, T.2010. Building adaptive capacity to climatevariability: The case of artisanal fisheries inthe estuary of the Patos Lagoon, Brazil.Marine Policy, 34(4): 742-751.Kelly, P. M. & Adger, W. N. 2000. Theory andpractice in assessing vulnerability to climatechange and facilitating adaptation. ClimateChange, 47: 325–52.Lacerda, L. D. 2002. Mangrove Ecosystems:Function and Management. 1. ed.Heidelberg: Springer Verlag, 292 p.Lacerda, L. D. 2009. Manguezais, ecossistemaschavesob ameaça. Oceanos: origens,transformações, e o futuro: Mudançasclimáticas e desafios ambientais. 1 ed.Duetto, São Paulo, 1: 76-82.Lacerda, L. D., Menezes, M. O. T. & Molisani, M.M. 2007. Changes in mangrove extension atthe Pacoti River estuary, CE, NE Brazil due toregional environmental changes between 1958and 2004. Biota Neotropica, 7(3): 67-72.Lana, P. C. 2003. Manguezais, legislação e gestão deáreas costeiras: o caso da Baía de Paranaguá.Pp. 313-331. In: Vieira, P. F. (Ed.).Conservação da diversidade biológica ecultural em zonas costeiras. Editora Aped,Florianópolis, Brasil, 527 p.Lana, P. C., Marone, E., Lopes, R. M. & Machado,E. C. 2001. The subtropical estuarine complexof Paranaguá Bay, Brazil. Ecological Studies,144: 131-145.Lara, R., Szlafsztein, C., Cohen, M., Berger, U. &Glaser, M. 2002. Implications of mangrovedynamics for private land use in Bragança,North Brazil: a case study. Journal ofCoastal Conservation, 8: 97-102.Lovelock, C. E. & Ellison, J. C. 2007. Vulnerabilityof mangroves and tidal wetlands of the GreatBarrier Reef to climate change. Pp. 237-269.In: Johnson, J. E. & Marshall, P. A. (Eds.).Climate Change and the Great BarrierReef: A Vulnerability Assessment. GreatBarrier Reef Marine Park Authority andAustralian Greenhouse Office, Australia.Maia, L. P., Lacerda, L. D., Monteiro, L. H. U. &Souza, G. M. E. 2006. Atlas dos Manguezaisdo Nordeste; Avaliação das Áreas deManguezal dos Estados do Piauí, Ceará,Rio Grande do Norte, Paraíba e Pernambuco.Fortaleza. Superintendência Estadualdo Meio Ambiente, 125 p.Marschke, M. & Berkes, F. 2006. Exploring strategiesthat build livelihood resilience: a casefrom Cambodia. Ecology and Society, 11(1),42. World Wide Web electronic publicationaccessible at http://migre.me/3aEPD.(Accessed 21/01/2010).Martin, F. & Lana, P. C. 1994. Aspectos jurídicosrelativos à proteção dos manguezais da Baíade Paranaguá. III Simpósio de Ecossistemasda Costa Brasileira, 1: 107-112.Martin, F. & Zanoni, M. 1994. Conflits d’usage surles mangroves de la Baie de Paranaguá,Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

222L. F. D. FARACO ET ALLIParaná, Brèsil. Urbanisation et preservationou utilisation rationnelle des resources?Journal d' Agriculture Traditional etBotanique Applique, 36 (2): 237-260.McClanahan, T. R., Cinner, J. E. & Maina, J. 2008.Conservation action in a changing climate.Conservation Letters, 1: 53-59.McFadden, L. & Green, C. 2007. Defining‘vulnerability’: conflicts, complexities andimplications for Coastal Zone Management.Journal of Coastal Research, SI 50(Proceedings of the 9th International CoastalSymposium): 120–124. Gold Coast, Australia.McLeod, E. & Salm, R. V. 2006. ManagingMangroves for Resilience to ClimateChange. IUCN, Gland, Switzerland. 64 p.Miranda, R. B. 2004. Dinâmicas de apropriação esaberes comunais dos manguezais e de seusrecursos bênticos de interesse econômico nocomplexo estuarino da baía de Paranaguá,Paraná. PhD. Thesis. Universidade Federal doParaná, Curitiba, Brasil, 349 p.MMA. 2007. Áreas Prioritárias para aConservação, Uso Sustentável e Repartiçãode Benefícios da Biodiversidade Brasileira:World Wide Web electronic publication,accessible at http://migre.me/3aAcN.(Accessed 13/09/2009).Mullon, C., Fréon, P. & Cury, P. 2005. Thedynamics of collapse in world fisheries. Fishand Fisheries, 6: 111–120.Neves, C. F. & Muehe, D. 2008. Vulnerabilidade,impactos e adaptação a mudanças do clima: azona costeira. Parcerias Estratégicas, 27:217-295.Nicholls, R. J. 2004. Coastal flooding and wetlandloss in the 21 st century: changes under theSRES climate and socio-economic scenarios.Global Environmental Change 14: 69–86.Nicholls, R. J., Hoozemans, F. M. J. & Marchand,M. 1999. Increasing flood risk and wetlandlosses due to global sea-level rise: regionaland global analyses. Global EnvironmentalChange, 9: S69-S87.Nicholls, R. J., Wong, P. P., Burkett, V. R.,Codignotto, J., Hay, J. E., McLean, R. F.,Ragoonaden, S. & Woodroffe, C. D. 2007.Coastal systems and low-lying areas. Pp.315-356. In: Parry M. L., Canziani, O. F.,Palutikof, J. P., Van der Linden, P. J. &Hanson, C. E. (Eds.). 2007. Climate Change2007: Impacts, Adaptation andVulnerability. Contribution of WorkingGroup II to the Fourth Assessment Reportof the Intergovernmental Panel on ClimateChange. Cambridge University Press,Cambridge, UK, 982 p.Parmesan, C. 2006. Ecological and EvolutionaryResponses to Recent Climate Change. AnnualReview of Ecology, Evolution, andSystematics, 37: 637-69.Parmesan, C. & Yohe, G. 2003. A globally coherentfingerprint of climate change impacts acrossnatural systems. Nature, 421: 37-42.Pauly, D., Watson, R. & Alder, J. 2005. Globaltrends in world fisheries: impacts on marineecosystems and food security. PhilosophicalTransactions of the Royal Society B, 360: 5-12.Pedroso Jr, N. N. 2002. Etnoecologia e Conservaçãoem Áreas Naturais Protegidas: Incorporando osaber local no Parque Nacional de Superagui.MSc. Thesis. Universidade Federal de SãoCarlos, São Carlos, Brasil, 91 p.Pierri, N., Angulo, R. J., Souza, M. C. & Kim, M. K.2006. A ocupação e o uso do solo no litoralparanaense: condicionantes, conflitos etendências. Desenvolvimento e MeioAmbiente, 13: 1737-1767.Raynaut, C., Zanoni, M., Ferreira, A. D. & Lana, P.C. 2007. Sustainability: Where, When, forWhom? Past, present and future of a localrural population in a protected natural area(Guaraqueçaba, Brazil). In: Maida, C. (Ed.).Sustainability and Communities of Place(vol 5. Studies in environmental anthropologyand ethnobiology). Berghan Books,Oxford, 1: 21-40.Saenger, P. 2002. Mangrove Ecology, Silvicultureand Conservation. Kluwer AcademicPublishers/The Society for MangroveEcosystems/International Tropical TimberOrganization, Dordrecht, The Netherlands/Okinawa, Japan, 360 p.Scavia, D., Field, J. C., Boesch, D. F., Buddemeier,R. W., Burkett, V., Cayan, D. R., Fogarty, M.,Harwell, M. A., Howarth, R. W., Mason, C.,Reed, D. J., Royer, T. C., Sallenger, A. H. &Titus, J. G. 2002. Climate Change Impacts onU.S. Coastal and Marine Ecosystems.Estuaries, 25(2): 149–164.Schaeffer-Novelli, Y., Cintrón-Molero, G., Adaime,R. R. & Camargo, T. M. 1990. Variability ofmangrove ecosystems along the Braziliancoast. Estuaries, 13(2): 204-219.Smit, B. & Wandel, J. 2006. Adaptation, adaptivecapacity and vulnerability. GlobalEnvironmental Change, 16: 282–292.Soares, M. L. G. 2009. A conceptual model for theresponses of mangrove forests to sea levelPan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Vulnerability of Mangroves and Fisherfolk223rise. Journal of Coastal Research, SI 56(Proceedings of the 10 th International CoastalSyposium): 267-271, Lisboa, Portugal.SOS Mata Atlântica/INPE. 2009. Atlas dosRemanescentes Florestais da MataAtlântica, Período 2005-2008, RelatórioParcial. Fundação SOS MataAtlântica/Instituto Nacional de PesquisasEspaciais, São Paulo, SP, Brasil, 156 p.Sterr, H., Klein, R. & Reese, S. 2000. Climatechange zones: an overview on the state-of-theartof regional and local vulnerability assessments.FEEM working paper series 38.Szlafsztein, C. F. 2005. Climate change, sea-levelrise and coastal natural hazards: a GIS basedvulnerability assessment, state of Pará, Brazil.Human Security and Climate Change; anInternational Workshop, Holmen FjordHotel, Asker, Oslo, 1-31.Taylor, D. & Sanderson, P. G. 2002. GlobalChanges, Mangrove Forests and Implicationsfor Hazards along Continental Shorelines. Pp.203-222. In: Sidle, R. C. (Ed.).Environmental Change and GeomorphicHazards in Forests. IUFRO Research Series,v. 9. CABI, New York, USA, 241 p.Teixeira, C. F. 2004. A Proteção Ambiental emGuaraqueçaba: Uma Construção Social. PhD.Thesis. Universidade Federal do Paraná,Curitiba, Brasil, 314 p.Tschakert, P. 2007. Views from the vulnerable:understanding climatic and other stressors inthe Sahel. Global Environmental Change,17: 381–396.Tuler, S., Agyeman, J., Da Silva, P. P., Lorusso, K.R. & Kay, R. 2008. Assessing vulnerabilities:Integrating information about driving forcesthat affect risks and resilience in fishingcommunities. Human Ecology Review, 15:171–184.Turner II, B. L., Kasperson, R. E., Matson, P. A.,McCarthy, J. J., Corell, R. W., Christensen,L., Eckley, N., Kasperson, J. X., Luers, A.,Martello, M. L., Polsky, C., Pulsipher, A. &Schiller, A. 2003. A framework forvulnerability analysis in sustainability science.Proceedings of the National Academy ofSciences, 100(14): 8074-8079.Valiela, I., Kinney, E., Culbertson, J., Peacock, E. &Smith, S. 2009. Global losses of mangrovesand salt marshes. Pp. 108-142. In: Duarte, C.M. (Ed.). Global loss of Coastal Habitats:Rates, Causes and Consequences.Fundación BBVA, Bilbao, Espanha, 184 p.Van Aalst, M. K., Cannon, T. & Burton, I. 2008.Community level adaptation to climatechange: the potential role of participatorycommunity risk assessment. GlobalEnvironmental Change, 18: 165–179.Walker, B. H., Holling, C. S., Carpenter, S. C. &Kinzig, A. P. 2004. Resilience, adaptabilityand transformability. Ecology and Society,9(2): art. 5. World Wide Web electronic publicationaccessible at http://migre.me/3aF0o(Accessed 15/12/2009).Walters, B. B., Rönnbäck, P., Kovacs, J. M., Crona,B., Hussain, S. A., Badola, R., Primavera, J.H., Barbier, E. & Dahdouh-Guebas, F. 2008.Ethnobiology, socio-economics andmanagement of mangrove forests: A review.Aquatic Botany, 89: 220-236.Walther, G., Post, E., Convey, P., Menzel, A.,Parmesan, C., Beebee, T. J. C., Fromentin, J.M., Hoegh-Guldberg, O. & Bairlein, F. 2002.Ecological responses to recent climate change.Nature, 416: 389-395.Wilbanks, T. J., Kates, R. W. 1999. Global changein local places: how scale matters. ClimaticChange, 43(3): 601-628.Wu, S., Najjar, R. & Siewert, J. 2008. Potentialimpacts of sea-level rise on the MidandUpper-Atlantic Region of the United States.Climatic Change, 95(1-2): 121-138.Yohe, G. & Tol, R. 2002. Indicators for social andeconomic coping capacity—moving toward aworking definition of adaptive capacity.Global Environmental Change, 12: 25–40.Zanoni, M. M. & Miguel, L. A. 1995. Impacts despolitiques de protection de l'environnementsur les pratiques paysannes (Guaraqueçaba-Brésil). In: Fertilité du Milieu Et StratégiesPaysannes Sous Les Tropiques Humides.Actes du séminaire, Montpellier, France.Zhang, K., Douglas, B. C. & Leatherman, S. P.2004. Global Warming and Coastal Erosion.Climatic Change, 64: 41– 58.Received April 2010Accepted October 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 205-223

Status of Eastern Brazilian coral reefs in time of climate changesZELINDA M. A. N. LEÃO, RUY K. P. KIKUCHI, MARÍLIA D. M. OLIVEIRA& VIVIAN VASCONCELLOSUniversidade Federal da Bahia, Centro de Pesquisa em Geofísica e Geologia, Instituto de Geociências. RuaBarão de Jeremoabo s/n, Campus Universitário de Ondina, Salvador 40170-115, Bahia, Brasil.E-mails: zelinda@ufba.br; kikuchi@ufba.br; mariliad@ufba.br; vivianvasconcellos@hotmail.comAbstract. Brazilian reefs comprise the largest and the richest reefs of the Southwestern AtlanticOcean. Indicators of reef vitality reveal that reefs located less than 5 km from the coastline, theinshore reefs, are in poorer conditions than those located more than 5 km off the coast, theoffshore reefs. The inshore reefs are the most impacted by the effects of eutrophic watersassociated with sewage pollution, high sedimentation rates and water turbidity, and the mostexposed to the effects of bleaching and infectious diseases. From 1998 to 2005, long-term seawater thermal anomaly events, equal or higher than 1ºC, were responsible for more than 30% ofbleached corals in the inshore reefs. In the offshore reefs of the Abrolhos area, bleaching wasmilder, but the reefs are strongly threatened by the incidence of diseases that have escalated inprevalence from negligible to alarmingly high levels in recent years. Although bleaching and coraldisease have not yet caused mass mortality in the Brazilian reefs, these natural disturbancesassociated with the effects of global climate changes and human-induced activities, could lead thereefs to higher levels of degradation.Keywords: Coral bleaching, Sea surface temperature anomaly, AbrolhosResumo. Estado dos recifes de coral da costa leste Brasileira em tempo de mudançasclimáticas. Os recifes de coral do Brasil são os maiores e mais ricos recifes do oceano AtlânticoSul-Ocidental. Indicadores da vitalidade dos corais revelaram que os recifes localizados menos de5 km da linha de costa, os recifes costeiros, apresentam condições inferiores aos recifes de altomar, que estão afastados mais de 5 km do continente. Os recifes costeiros além de estaremseveramente impactados pelos efeitos da eutrofização das águas, associada à poluição de esgotosdomésticos, altas taxas de sedimentação e turbidez, estão ameaçados pela ocorrência de eventos debranqueamento e doenças dos corais. De 1998 a 2005 eventos de longa duração das anomaliastérmicas da água do mar de 1ºC ou mais, foram responsáveis por mais de 30% do branqueamentonos recifes costeiros. Nos recifes de alto mar da área de Abrolhos, apesar dos eventos debranqueamento terem sido mais suaves, os recifes estão ameaçados pela incidência de doençascuja prevalência alcançou níveis alarmantes nos últimos anos. Embora os eventos debranqueamento e a ocorrência de doenças não tenham provocado, ainda, mortalidade em massados corais, estes distúrbios associados aos efeitos dos fenômenos climáticos globais e de açõesinduzidas pela atividade humana constituirão uma grave ameaça que poderá levar os recifes aníveis elevados de degradação.Palavras-chave: Branqueamento de coral, anomalia térmica da água superficial do mar, AbrolhosIntroductionIn the coastal zone of Eastern Brazil, coralreefs are one of the most prominent marineecosystems, comprising the largest and richest areaof reefs in all of the southwestern Atlantic Ocean.Studies during the last two decades have shown thatthese coral reefs are experiencing increasingdegradation due to a combination of large-scalenatural threats (e.g. sea level oscillations and ENSOevents), and of more local scale anthropogenicstressors, such as accelerated coastal development,Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Brazilian coral reefs in time of climate changes225reef eutrophication, marine pollution, tourismpressure, over-exploitation of reef resources,overfishing and destructive fisheries and, morerecently, the introduction of non-indigenous invasivespecies (De Paula & Creed 2004, Leão & Kikuchi2005, Leão et al. 2008). At the global scale, thereare many cases in which these threats have alreadycaused a reef phase shift away from corals (Riegl etal. 2009).In Eastern Brazil, several parameters of reefvitality, among them the living stone coral cover, thedensity of reef building coral species and of coralrecruits, and the abundance of macroalgae, indicatethat, overall, the reefs located closer to the mainlandare in poorer condition than those that are more than5 km from the coastline (Kikuchi et al. 2010). Alowering of sea-level that occurred after 5,000 yearsBP, along that part of the Brazilian coast, placed theinshore reefs closer to the coastline and mobilizedthe western continent-derived siliciclastic sedimenttoward the eastern reef systems. This event exposesthem to increased runoff and sedimentation andintense solar radiation, as well as threats induced byhuman activities (Leão & Kikuchi 2005).Bleached corals have been seen in EasternBrazil since the summer of 1982/1983 (Z.M.A.N.L.pers. observation), but the first published records ofcoral bleaching date from the summer of 1993/1994,after the occurrence of a worldwide El-Niño event(Castro & Pires 1999, 2001). Since then, there arerecords of bleached corals coincident with everyoccurrence of sea water temperature anomaly alongthis part of the Brazilian coast (Leão et al. 2008).There seems to be a strong linkage between coralbleaching and periods of elevated sea surfacetemperatures along the coast of Brazil. On the otherhand, coral diseases have flourished worldwide sincethe 1980s (Harvell et al. 1999, 2002, Rosenberg &Loya 2004), but only recently has the incidence ofcoral diseases in the Brazilian reefs increased(Francini-Filho et al. 2008). Both coral bleachingand diseases seem to be intensified by warming ofthe sea surface temperature, and they are affectingmostly the inshore reefs.This work presents a synthesis of the statusof the coral reefs from Eastern Brazil based on datacollected during the last decade. It was examined avariety of environmental factors in an effort todistinguish the dominant attributes on the intensityof bleaching. The combination of these studiesincrease the chances of making predictions concerningthe effects of the expected temperature increaseon reef organisms, and to set regional priorities forresearch and conservation of the reefs in the face ofglobal climate changes.Materials and MethodsStudy area: The Eastern Brazilian reefs arespread along about 800 km of the coastline of thestate of Bahia, between 12º and 18º S (Fig. 1). Thispart of the Brazilian coast has a tropical climate,with rainfall ranging from 1300 mm y -1 in itsnorthernmost part to a maximum of 2000 mm y -1in the southern region. Average air temperaturesrange from 23 ºC in the winter to 28 ºC in thesummer. The most significant wave front directionsare from the NE, E, SE and SSE. NE and E windinducedwaves have periods of 5 sec and heights of1.0 m, and SE and SSE wind-induced waves haveperiods of 6.5 sec and 1.5 m heights (Bittencourt etal. 2005). Spring tides vary from 1.7 m in thesouthernmost region to 3.0 m at the extreme north.The temperature of the surface water varies fromaround 24 ºC (winter) to 28 ºC (summer).Study reefs: The studied reefs comprise two groups:inshore and offshore. The inshore reefs are adjacentto the coast or a few kilometers from the shoreline(< 5 km). They include fringing reefs and shallowbank reefs, both in depths from 5 m to 10 m in thefore-reef zone and are not longer than 1 km. Theoffshore reefs consist of reef structures of variabledimensions, from a few meters up to 20 km, and arelocated more than 5 km off the coastline, at variousdepths. They include coral knolls, patch and bankreefs, and isolated coral pinnacles. Besides these,there are also oceanic shelf edge reefs that occur atthe border of the continental shelf, with widths up to3 km and a relief of about 30 m at depths of 50 m.These reefs must have started to grow earlier in theHolocene, at lower sea level stands, and are nowveneered with a deeper water community (Kikuchi& Leão 1998).Reef-building fauna: Brazilian reefs were builtby a low diversity coral fauna rich in endemicspecies (Laborel 1970). Twenty three species ofstony corals and five species of hydrocorals areregistered along the Brazilian coast, and eighteencorals and four hydrocorals occur on the EasternBrazilian reefs (Table 1). From these, six species areendemic of the Western South Atlantic waters, somehave affinities with Caribbean coral forms and someare remnants of a more resistant relict fauna datingback the Tertiary time, which was probablypreserved during Pleistocene sea level low stands ina refuge provided by the sea-mountains off theAbrolhos Bank (Leão et al. 2003). These archaicspecies are the most common forms in all studiedreefs. They are the three species of the genusMussismilia: M. braziliensis Verrill 1868, M. hispidaPan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

226Z. M. A. N. LEAO ET ALLIVerrill 1868 and M. hartti Verrill 1868 and thespecies Favia leptophylla Verrill 1868. The othertwo endemic species are F. gravida Verrill 1868 andSiderastrea stellata Verrill 1868, both related to theCaribbean coral fauna. The species M. braziliensisand F. leptophylla are the Brazilian corals that showthe greatest geographical confinement, because theyare, so far registered, only found along the coast ofthe state of Bahia. The species S. stellata and F.gravida have a broader distribution along the coastof Brazil, and are the most common corals in theshallow intertidal pools of reef tops. Along thewhole coast of Brazil, the reefs of Abrolhos in itseastern region, have the largest number of coralspecies (eighteen), but this number reducesnorthward the state of Bahia (see Table 1)(Laborel 1969, 1970, Belém et al. 1982, Leão 1982,Araujo 1984, Nolasco 1987, Castro 1994, Leão et al.2003, Neves et al. 2006, 2008, Amaral et al. 2008,Kikuchi et al. 2010).Field procedures: the status of coral reefs wasassessed through visual census using band transects.In most reefs it was applied the methodology of theAGRRA protocol (Atlantic and Gulf Rapid ReefAssessment http://www.agrra.org, Ginsburg et al.1998, last updated in 2007), except for the reefsfrom Todos os Santos Bay, where the video transectmethod was applied. Reef assessments began in1998, when three shallow reef sites from the NorthCoast of the state of Bahia were surveyed using belttransects positioned parallel to the coastline. Startingin 1999, the reefs were assessed using the AGRRAmethod that was designed for assessing andcomparing reef status in the Atlantic Ocean,including reefs in the Gulf of Mexico. In theAGRRA protocol, the assessment of reef status isperformed along six 10 m long transects, where theline intercept method is applied for coral cover.Colony diameter, dead surfaces, bleaching anddiseases are analyzed in a belt 1 m wide along theline transects, and five quadrats (25 cm X 25 cm) pertransects are used to estimate the relative abundanceof algal types: macro, turf (< 1 cm high), andcrustose corallines, the average canopy height ofmacroalgae, and the density of coral recruits(colonies ≤ 2 cm).The video-transect technique consists of twodistinct phases: a) field data acquisition, whenimages taken along a belt-transect are recorded by avideo-camera, and b) the identification of organismson the screen of a computer, from the images acquiredin the field. Studies using this technique werecarried out in several reef areas around the world,and are described in Aronson et al. (1994), Carleton& Done (1995), Aronson & Swanson (1997).Figure 1. Location of reefs along the coast of Eastern Brazil. A - North Coast; B - C- Todos os Santos Bay and Tinharéand Boipeba islands; D - E- Cabrália and Itacolomis reefs; F – Abrolhos. Black arrows indicate studied reef sites.Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Brazilian coral reefs in time of climate changes227In Brazil, the videography technique wasintroduced in 2003 (Dutra et al. 2006) with thepurpose of assessing the status of coral reefs withinTodos os Santos Bay, which was re-evaluated late in2007 (Cruz 2008). The purpose of using this techniquein the reefs of Todos os Santos Bay was itsprevious successful performance in water with lowvisibility, due to the short distance between the videocamera and the surface of the reefs. This conditionof low visibility is common in the bay waters.The assessed reef areas: Six areas along the coastof the state of Bahia were assessed: North Coast,Todos os Santos Bay, Tinharé and Boipeba islands,Cabrália, Itacolomis reefs and Abrolhos (see Fig. 1).Information of location, depth of reefs, date ofsurvey, the values of sea surface temperatureanomalies during survey, the measured parametersand the applied methods are shown in Table II. Thereefs of the North Coast were assessed before thedevelopment of the AGRRA protocol. They weresurveyed using 3 x 1 m wide belt transects 20 mlong completing a total area of 60 m 2 in each reefsite (Leão et al. 1997, Kikuchi & Leão 1998,Kikuchi 2000, Leão & Kikuchi 2005). Themethodology of the AGRRA protocol was applied inthe reefs of Tinharé and Boipeba islands, Cabrália,Itacolomis and Abrolhos. The Tinharé/Boipeba,Cabrália and Itacolomis reefs were surveyed onlyonce or two times, but in the Abrolhos area reefswere surveyed four times. In all these reefs thesurveyed area summed 60 m 2 in each site (Kikuchiet al. 2003 a, b, Kikuchi et al. 2010). Most surveyswere performed between the months of March andApril after the occurrence of sea water thermalanomalies (Leão et al. 2008).The reefs of Todos os Santos Bay wereassessed twice (2003 and 2007), applying themethodology of the video transects. In 2003, eightgroups of reefs in the inner region of the bay wereinvestigated (Dutra et al. 2006), and in 2007 (Cruz2008, Cruz et al. 2009) twenty-three reef sites wereassessed, being eight located at the entrance of thebay and fifteen in the interior of the bay; eight ofthem were the same reefs investigated by Dutra etal. (2006).Two in situ investigations were carried outin order to evaluate the behavior of coral bleachingin fixed colonies of specific coral species. The firstwas performed in 2006, in corals located at theentrance of Todos os Santos Bay, where thirtycolonies of the coral Montastrea cavernosa and ofSiderastrea spp were surveyed for a year todetermine their rate of growth and investigate theoccurrence of bleaching (Chaves 2007). The otherone was performed in the Abrolhos area during thesummer and winter seasons of the years 2006, 2007and 2008, in fixed colonies of the coral Montastreacavernosa, located on the tops and lateral walls of14 reef sites, to evaluate the percent of bleachingand its progression rate during the period ofinvestigation (Meirelles 2009).Surveys for coral diseases were carried outyearly from 2001 to 2007 on twenty-eight sitesdistributed along the Abrolhos area, during summerperiods (January to April) when northeasternwinds prevail, water visibility is from 5 to 10 m,and sea surface temperatures are relatively higher(~25-28 o C). The disease progression rates wasestimated for one site (Portinho Norte, within theAbrolhos Archipelago) between April 14 and July 4,2006, and the disease prevalence was determined attwo sites (Pedra de Leste, within Parcel das Paredesreefs, and one site in Timbebas reef) betweenJanuary and March 2007 (Francini-Filho et al. 2008,2010). These surveys were conducted by theConservation International-Brazil (CI Brazil), as partof the Marine Management Areas Science Program,Brazil Node.ResultsReef condition: the following parameters wereconsidered to define the status of the EasternBrazilian reefs: percent living coral cover, density oflarger corals (> 20 cm diameter), density of coralrecruits (≤ 2 cm diameter) and percent of macroalgae.The reefs surveyed on the North Coast in1998 revealed very low values for living coral cover(

228Z. M. A. N. LEAO ET ALLITable I. Occurrence of coral and hydrocoral species in reefs from Eastern Brazil.NorthCoastTodosSantos BayTinhare/BoipebaCabráliaItacolomis AbrolhoscoastalAbrolhosislandsAbrolhosoffshoreAgaricia agaricites X X X X X X X XAgaricia fragilis X X X X X XAstrangia braziliensis X XFavia gravida X X X X X X X XFavia leptophylla X X X X X XMadracis decactis X X X X X X XMeandrina X X X X X XMontastrea cavernosa X X X X X X X XMussismilia X X X X X X X XMussismilia harttii X X X X X X X XMussismilia hispida X X X X X X X XPhyllangia americana X X XPorites astreoides X X X X X X X XPorites branneri X X X X X X X XScolymia wellsi X X X X X X XScolymia cubensisXSiderastrea stellata X X X X X X X XSiderastrea radians XStephanocoeniaXMillepora alcicornis X X X X X X X XMillepora nitida X X X X X XMillepora braziliensis X X XStylaster roseus X X# Species 14 14 16 16 15 17 17 22In the Abrolhos area, average values forthe reefs from the coastal area were as follows:living coral cover ranged from 5.6 to 11%, thedensity of large corals varied from 11 to 144colonies per reef site (60 m 2 ), the number of coralrecruits per square meter was 13.6 to 35, andvalues for macroalgae abundance varied from 0.6 to11.9%. In the fringing reefs bordering the Abrolhosislands, coral cover values ranged from 6.8 to17.3%, the density of larger corals varied between57.7 and 155.3 colonies per reef site, coral recruitsranged from 13.6 to 35 m 2 , and the abundanceof macroalgae ranged from 0 to 22.5%. In thechapeirões of the Parcel dos Abrolhos, the valuesof the reef indicators were: 14.9% for living coralcover, 107.5 for number of large coral coloniesper reef site; 32.1 m 2 for density of coral recruits and6.3% for macroalgae abundance (Kikuchi et al.2010). Comparing these data averaged for theinshore and offshore reefs, which were acquiredalong belt transects that covered an area of 60 m 2in each reef site (Tab. III), one sees that thevalues of the measured coral parameters (livingcover, density of larger colonies and of recruits) forthe inshore reefs (North Coast, Tinharé / Boipebaand Cabrália) are much lower than those forthe offshore reefs (Itacolomis and Abrolhos), whichare opposite the average values of the abundance ofmacroalgae.In Todos os Santos Bay the data about coraldiversity originated from the survey performed in2003 by Dutra and collaborators (2006) comparedwith the information given in Laborel’s descriptionfrom the 1960’s (Laborel 1969, 1970), show that themajor differences between these two surveys are theidentified coral species present in the reefs from theinterior of the bay. Some species cited in Laborel’sdescription were not found in the 2003 survey, suchPan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Brazilian coral reefs in time of climate changes229Table II. Information of coral reefs surveyed along the coast of Eastern Brazil. n = number of reef sites. SST = Sea SurfaceTemperatureReefs Location Depthrange(m)Date ofsurveySSTAnomalyMeasuredparametersMethodNorth Coastn=6Tinharé /Boipeban=4Cabralian=6Itacolomisn=3Abrolhoscoastal reefsn=20Abrolhosislandsn=12Abrolhosoffshore reefsn=6Todos SantosBayn=32Salvador CityYatch ClubAbrolhosseveral reefsN=1412.3805º S38.0200º W13.4910º S38.9024º W16.2397º S38.9526º W16.8976º S39.0909º W17.4670º S -18.0203º S39.0004º W -38.9883º W17.9673º S -17.9794º S38.7018º W -38.7080º W17.9977º S38.6713º W12.4720º S-13.1037º S38.3134º W-38.4532º W12.5900º S38.3100º W17.4701º S-17.5732º S39.0141º W-38.3020º W4.5-8.5 1998 1.0 ºC > 1 week Coral cover, colonydiameter, # coral species,coral recruits3.5-4.0 200320041.0 ºC > 1 week0.25 ºC < 2 weeksCoral cover, colonydiameter, # coral species,coral recruits, coral deadsurface, bleaching,disease, algal abundance4.8-6.8 2004 No SST anomaly Coral cover, colonydiameter, # coral species,coral recruits, coral deadsurface, bleaching,disease, algal abundance2.0-3.5 2005 0.75 ºC > 2 weeks Coral cover, colonydiameter, # coral species,coral recruits, coral deadsurface, bleaching,disease, algal abundance2.0-8.0 20012002200320053.5-8.0 20002001200220056.0-9.0 20002001200220033.0-8.0 2003n=92007n=230.50-0.75 ºC< 1 week1.0 ºC > 2 weeks0.75 ºC > 2 weeks0.25 ºC < 2 weeks0.50-0.75 ºC< 1 week0.75 ºC > 2 weeks0.25 ºC < 2 weeks0.50-0.75 ºC< 1 week1.0 ºC > 2 weeks1.0 ºC > 2 weeksNo SST anomalyCoral cover, colonydiameter, # coral species,coral recruits, coral deadsurface, bleaching,disease, algal abundanceCoral cover, colonydiameter, # coral species,coral recruits, coral deadsurface, bleaching,disease, algal abundanceCoral cover, colonydiameter, # coral species,coral recruits, coral deadsurface, bleaching,disease, algal abundanceCoral cover, # coralspecies, algal abundance,other benthics (sponge,zoanthus, soft coral)2.0-4.5 2006 No SST anomaly Rate of coral growth andbleaching4.3-15.0 200620072008No SST anomalyNo SST anomalyNo SST anomalyCoral bleachingprogressionBelt transect(3x1m wide x 20m long)60m 2 /siteBelt transect (AGRRA)(6x1m wide x 10m long)60m 2 /siteBelt transect (AGRRA)(6x1m wide x 10m long)60m 2 /siteBelt transect (AGRRA)(6x1m wide x 10m long)60m 2 /siteBelt transect (AGRRA)(6x1m wide x 10m long)60m 2 /siteBelt transect (AGRRA)(6x1m wide x 10m long)60m 2 /siteBelt transect (AGRRA)(6x1m wide x 10m long)60m 2 /siteBelt transect (Video)(10x0.20m wide x 20mlong)40m 2 /siteIn situ measurement in 30coral colonies ofSiderastrea sp andMontastrea cavernosaIn situ measurement incolonies of Montastreacavernosaas Mussismilia braziliensis, Meandrina braziliensis,Porites branneri, Stephanocoenina michelini andMillepora nitida. The most common species thatoccurred in both the 1960 and the 2003 surveys areMontastrea cavernosa, Siderastrea stellata,Mussismilia hartti, M. hispida and Favia gravida.There are also some species that Laborel did notidentify but were found in 2003: Madracis decactis,Agaricia agaricites, Porites astreoides and Scolymiawellsi. Comparing data from the living coral coverof the same reefs investigated in 2003 by Dutra et al.(2006) with a survey performed by Cruz in 2007(Cruz 2008, Cruz et al. 2009), it seems that thevalues found for five of these reefs in 2007 arelower than the ones registered in 2003 (see TableIV). Regarding the presence of the coral species,Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

230Z. M. A. N. LEAO ET ALLICruz (2008) and Cruz et al. (2009) report thesame species that Dutra et al. (2006) registered intheir earlier survey of the reefs located in theinterior of the bay.Coral bleaching events: the first record of coralbleaching in Bahia occurred in 1994, in the area ofAbrolhos, affecting an average of 70% of coralcolonies (Castro & Pires 1999). In 1998, a seasurface temperature anomaly started in mid January(summer in the southern hemisphere), attainedits climax in mid March and early April, andfaded away at the end of May. During this event,two hotspots were registered, one on the NorthCoast of the State (Dutra 2000) and the other inAbrolhos (Leão et al. 2008). In both areas, theestimated sea surface temperature anomaly, at about1 °C, matched measurements of sea temperature inthe field, which ranged between 29.5 °C and30.5 °C. An evaluation during 13 months in a reeffrom the North Coast registered an average of 60%corals bleached (Fig. 2).During surveys performed from 2000 to2005, the average of corals bleached ranged from0.2% during years with short-term sea surfaceanomalies, equal or less than 0.75 o C, to up to 50%in years with anomalies equal or greater than 1 o C formore than one week.Table III. Average values of coral parameters and macroalgae abundance of inshore and offhore reefs from EasternBrazil, measured along belt transects covering an area of 60 m 2 in each reef site. Inshore reefs = North Coast, Tinharé /Boipeba and Cabralia. Offshore reefs = Itacolomis and Abrolhos.Living coralcover %Density colonies> 20 cm#colonies . 60 m 2Density coral recruits#colonies . m 2Macroalgae%Inshore reefs 3.6±2.4 22.0±22.4 1.6±1.8 43.2±18.4Offshore reefs 11.6±3.5 102.0±42.7 27.1±8.5 8.2±6.7Table IV. Average (± SD) living coral cover measured during the surveys of 2003 (Dutra et al. 2006) and 2007 (Cruz2008) in reefs located in the interior of Todos os Santos Bay.Reefs Mangueira Pedras Alvas Dentão Cardinal Poste 1 Poste 4 Poste 5 Poste 62003Survey13.5±4.7 27.9±1.8 7.7±6.5 23.1±2.3 8.9±0.2 18.7±1.5 8.1±1.5 10.1±0.72007Survey8.4±5.7 13.1±5.8 0.7±0.4 27.0±5.4 2.3±1.2 21.0±8.7 4.6±2.3 19.2±4.3Sea surface anomalies of 1 °C or a littlehigher occurred in 2003 in Tinharé/Boipeba andAbrolhos. In Tinharé/Boipeba, the sea surfacetemperature rose at the end of February anddissipated at the end of April. Coral bleachingwas registered at the beginning of May, reachingan average of 27%. In Abrolhos, a sea surfacetemperature anomaly started in mid February,reached its climax during the whole month ofMarch and dissipated at the end of April. Coralbleaching was observed in mid March, with anaverage of 13% colonies bleached (Fig. 2).During the hotspot event that occurred in SouthernBahia in 2005, two reef areas were affected,Itacolomis and Abrolhos, with an average of 25% ofcoral colonies bleached. A less extensive coralbleaching event was observed in the Itacolomisreefs, where an average of 11% of corals was foundto be affected (Fig. 2). In those areas, the sea surfacetemperature started to rise in mid March, reachinganomalies of 0.75 to 1 o C at the beginning of April,and being completely dissipated at the end of themonth.During years with short term sea wateranomalies with maximum values of 0.75 ºC, thesurveys performed in the Abrolhos area (2000, 2001,2002), and in reefs from Tinharé and Boipeba(2004), revealed values of coral bleaching of lessthan 10% (Fig. 3).During the investigation performed infixed coral colonies at the entrance of Todos osSantos Bay in 2006, when sea water temperatureanomalies did not occur, the averagenumber of colonies of Montastrea cavernosa withsignals of bleaching were from 3.5 (out of agroup of 30 colonies – 11.6%) per month duringthe summer, and three colonies per month (10%)during the winter. Siderastrea spp did not reachmore than two bleached colonies (within a groupof 30 monitored colonies – 6.6%) per month duringthe summer, and less than that during the wintermonths (Chaves 2007). The investigation of theAbrolhos reefs with fixed colonies of Montastreacavernosa, from 2006 to 2008, a period without seawater temperature anomalies along the coast ofBrazil, revealed that bleaching events affectedless than 10% of the investigated coral colonies(Meirelles 2009).Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Brazilian coral reefs in time of climate changes231Figure 2. Average percentages of bleached coral coloniesmeasured during longer term events of sea watertemperature anomalies ≥ 1 ºC, along the coast of EasternBrazil. TSB = Todos os Santos Bay, T/B = Tinharé andBoipeba islands.Figure 3. Average percentages of bleached coral coloniesmeasured during short term events of sea watertemperature anomalies ≤ 0.75 ºC. St. B. S = Santa BarbaraSouth; St. B. W. = Santa Barbara West; St. B. N. = SantaBarbara North; T/B = Tinharé and Boipeba islands.Occurrence of coral diseases: according to thedescription given by Francini-Filho et al. (2008), thefirst coral disease detected in the Abrolhos reefs wasin January 2005, comprising only a few affectedcolonies of Siderastrea spp and Mussismiliabraziliensis. Since then, the number of sites showingdiseased corals increased sharply. Four types ofdiseases were detected affecting the reef-buildingcorals: black-band, red-band, white plague and darkspots. The most affected coral species were: Faviagravida, F. leptophylla, Mussismilia braziliensis,M. hartti, M. hispida, Porites astreoides andSiderastrea spp. Because Mussismilia braziliensis isan important coral constructor of reefs in Abrolhos,the white-plague like disease prevalence andprogression rate was investigated. At two sites(Leste and Timbebas reefs), this disease prevailedduring the sampling period, and its linearprogression rate was estimated at 0.18±0.06 (SE)cm 2 day -1 , while the area progression rate wasestimated at 0.21±0.07 (SE) cm 2 day -1 . According tothe above authors, a model estimating the loss of M.braziliensis based on these numbers predicts that ifthe current rate of mortality of M. braziliensis ismaintained, about 40% of this coral cover will belost in the next 50 years, and if there is an increase indisease severity in successive decades, M.braziliensis will be nearly extinct in less than acentury. Francini-Filho et al. (2010) identified aseasonal prevalence of the white plague-like diseasein the summer compared to the winter, which meansthat the disease is temperature dependent. Accordingto these authors, this result supports the hypothesisthat warmer oceans are facilitating the proliferationof coral diseases worldwide.Discussion and RecommendationsThe reef condition: overall, inshore reefs of EasternBrazil (North Coast, Tinhare and Boipeba andCabralia), which are located less than 5 km from thecoast, are in poorer condition than the offshore reefs,which are located more than 5 km off the coast(Itacolomis and Abrolhos) (see Table III). Previousworks have shown that the inshore reefs areexperiencing stress resulting, chiefly, from highersedimentation rates and water turbidity (Dutra et al.2006), an abnormal increase in nutrients associatedwith sewage pollution in the coastal waters (Costa Jret al. 2000, 2006), and an elevated rate of bioerosion(Santa-Isabel 2001, Reis & Leão 2003), besidesbeing highly used by fishing and tourism. Theseinhospitable conditions must have been deleteriousto the reef-building corals of these reefs. Althoughthe vital conditions of the offshore reefs seem to bequite stable, both in the Archipelago and in theChapeirões, there are two sites (Leste and southernSt. Barbara), where the values of the selected coralindicators are rather low. Leste is the closest reef inthe Abrolhos area to the coastline (~10 km) and,therefore, the most exposed to processes acting inthe continent coastal zone, besides being one of thereefs most affected by coral disease (Francini-Filhoet al. 2008). The southern part of St. Barbara Islandis located inside the limits of the Abrolhos NationalMarine Park, but it is the preferred site of tourists fordiving and snorkeling, because of the protection itoffers for anchoring (Spanó et al. 2008).Thus, the setof parameters assessed show that there is a clearPan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

232Z. M. A. N. LEAO ET ALLIdistinction between the inshore and offshore reefs,with some sites from the latter group alreadydecayed to conditions quite similar to those of theinshore reefs. These sites must be closely monitoredand need to be target of management actions.Occurrence of coral bleaching and disease: Thestrongest coral bleaching events registered along thecoast of Eastern Brazil have been associated withwarming of the sea surface temperature. When seasurface anomalies reached values of 1 ºC during oneto two weeks, as occurred in 1998, 2003 and 2005along the whole coast of the state of Bahia, morethan 30% of bleached corals were registered in theinshore reefs along an extent of about 500 km. Inthese periods, the reefs located in the northernmostpart of the state (e.g., the North Coast and Todos osSantos Bay) were the most affected. In this area, theaverage bleaching of colonies reached values above50% (see figure 2). In the offshore reefs ofAbrolhos, the percentage of bleaching was milder,ranging from 8 to 18% in 2003 and reaching anaverage of 25% in 2005.When thermal anomalies were between 0.50to 0.75 ºC for less than one week, as occurred in2001 and 2002, or around 0.25 ºC, as in 2000 and2004, the average coral bleaching was lower than10%, even in the inshore reefs of Tinhare andBoipeba (see figure 3). During 2006, 2007 and 2008,no thermal anomalies occurred along the coast ofBrazil, and signals of bleaching in coral coloniesduring experiments on the northern (Todos os SantosBay) and southern (Abrolhos) coasts of EasternBrazil were very mild, remaining lower than 12% ofthe surveyed coral colonies.Overall, based on these occurrences, we seea strong linkage between strong coral bleaching andperiods of elevated sea surface temperatures inEastern Brazil. Additionally, reefs already impactedby processes operating in the coastal zone, theinshore reefs, are most susceptible to bleaching.Other examples include the reefs of Jamaica andSouth Florida, which are threatened by anthropogenicimpacts and have been strongly affected bysuccessive bleaching events (Hughes et al. 2003,Goldberg & Wilkson 2004).Adding to those pieces of evidence, aBayesian model developed by Krug (2008), basedon the maximum sea surface temperature accumulatedin five years, light attenuation in water, rainfallmagnitude, zonal and meridional wind fields derivedfrom remote sensing and analysis and reanalysisdata, shows the complex nature of bleaching patternsin Eastern Brazil. This model points to a scenariowhere factors such as the period, the intensity and/orthe geographical position of the bleached corals,determine the bleaching pattern. The model showsevidence that high sea surface temperaturecontrolled bleaching events on the North Coast andTodos os Santos Bay. However, in the Abrolhosarea, where bleaching events were weaker than thenorthern reefs, there is a large amount of variabilityof surface temperature, precipitation and winds,suggesting that, for example, the 2003 bleachingevent was mostly influenced by low watertransparency and increased rainfall. Thus, the modelconcludes that the bleaching pattern in EasternBrazil is a highly complex system that might beresponding to both global and regional forcingfactors.Although Brazilian reefs have beencoexisting for a long time with natural occurrence ofextreme environmental conditions, such as highsedimentation rates, low light penetration levels andperiodic thermal stress, coral bleaching has not yetcaused coral mass mortality. Such threat could occurin the future, considering that a worldwideintensification of warm sea surface temperatureanomalies is expected (Hoegh-Guldberg 1999,Maynard et al. 2009). In addition, it is known thatthere is also a positive link between coral diseaseincidences and stresses arising from global warming(Selig et al. 2006; Bruno et al. 2007; Ainsworth &Hoegh-Guldberg 2009). Anomalously higher watertemperatures may enhance the probability of coraldisease outbreaks by increasing the abundance orvirulence of pathogens or by increasing the hostsusceptibility, as bleached corals are moresusceptible to diseases than healthy ones (Muller etal. 2008). Moreover, coral bleaching and diseasemust be also intensified by the rapid deterioration ofcoastal waters, a factor that will enforce thedegradation of the Brazilian inshore reefs.As stated by Francini-Filho et al. (2008), theconsequences of a coral decline due the intensificationof diseases, for the maintenance of reefintegrity, in a region with high endemism levels, arenot yet totally predictable. Nevertheless, in thespecific case of the relatively low diversity Brazilianreef ecosystem, it may be more catastrophic thanpreviously anticipated. It is therefore important toincrease our knowledge of the reef processes and todevelop a better capacity in Brazil for implementingstrategies that will enhance the chances for the reefsto resist future sea surface temperature warming.Thus, urgent actions regarding the conservation ofthis reef region is needed. Many initiativesconcerned with coral reef protection, managementand recovery have been developed in Brazil duringthe last few years (Rodriguez-Ramirez et al. 2008),Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Brazilian coral reefs in time of climate changes233but much effort is still needed for the effectiveconservation of the reefs. The survival of these reefswill depend upon an appropriate understanding of allprocesses involved in the reef ecosystem functioningand maintenance, and on the effective managementand sustainable use of their resources. The complexitiesinvolving important processes like euthrophication,spread of coral diseases and coral bleachingdemand deliberate actions toward a long-term reefmonitoring and increase in society’s awareness.ReferencesAinsworth, T. D. & Hoegh-Guldberg, O. 2009.Bacterial communities closely associated withcoral tissues vary under experimental andnatural reef conditions and thermal stress.Aquatic Biology, 4(3): 289-296.Amaral, F. M. D., Steiner, A. Q., Broadhurst, M. K.& Cairns, S. D. 2008. An overview of theshallow-water calcified hydroids from Brazil(Hydrozoa: Cnidaria), including the descriptionof a new species. Zootaxa, 1930: 56-68.Araujo, T. M. F. 1984. Morfologia, composição,sedimentologia e história evolutiva do recifede coral da Ilha de Itaparica, Bahia. MSc.Dissertation. Universidade Federal da Bahia,Salvador, Brasil. 92 p.Aronson, R. B., Edmunds, P. J., Precht, W. F.,Swanson, D. W. & Levitan, D. R. 1994.Large-scale, long-term monitoring of Caribbeancoral reefs: simple, quick, inexpensive techniques.Atoll Research Bulletin, 421: 1-19.Aronson, R. B. & Swanson, D. W. 1997. Videosurveys: uni and multivariate applications.Proceeding 8 th International Coral ReefSymposium, Panamá, 2: 1441-1446.Belém, M. J. C., Rohlfs, C., Pires, D. O. & Castro,C. B. 1982. S.O.S. Corais. Revista CiênciaHoje, 5: 34-42.Bittencourt, A. C. S. P., Dominguez, J. M. L.,Martin, L. & Silva, I. R. 2005. Longshoretransport on the northeastern Brazilian coastand implications to the location of large scaleaccumulative and erosive zones: An overview.Marine Geology, 219: 219–234.Bruno, J. F., Selig, E. R., Casey, K. S., Page, C. A.,Willis, B. L., Harvell, C. D., Sweatman, H. &Melendy, A. M. 2007. Thermal stress andcoral cover as drives of coral diseaseoutbreaks. Plos Biology, 5(6): 1220-1227.Carleton, J. H. & Done, T. J. 1995. Quantitativevideo sampling of coral reef benthos: largescaleapplication. Coral Reefs, 14: 35-46.Castro, C. B. 1994. Corals of Southern Bahia. Pp.161-176. In: Hetzel, B. & Castro, C. B. (Eds.).AcknowledgementsThe data presented in this article originatedfrom several projects with financial supportfrom various sources: MCT, CNPq, FAPESB,FINEP, CAPES and CI-Brazil through the MarineManage-ment Areas Program – Brazil Node. Theauthors are grateful to the administrators of theAbrolhos National Marine Park for logisticalsupport, as well as to all those that participated infield works.Corals of Southern Bahia, Rio de Janeiro,Editora Nova Fronteira. 185 p.Castro, C. B. & Pires, D. O. 1999. A bleaching eventon a Brazilian coral reef. Brazilian Journalof Oceanography, 47: 87-90.Castro, C. B. & Pires, D. O. 2001. Brazilian coralreefs: what we already know and what is stillmissing. Bulletin of Marine Science, 69(2):357-371.Chaves, E. M. 2007. Crescimento linear ebranqueamento dos corais Siderastrea spp eMontastrea cavernosa na Baía de Todos osSantos. BSc. Monography, Faculdade deTecnologia e Ciência. 49 p.Costa Jr, O. S., Leão, Z. M. A. N., Nimmo, M. &Atrill, M. 2000. Nutrification impacts on coralreefs from Northern Bahia, Brazil.Hydrobiology, 440: 307-316.Costa Jr, O. S., Attrill, M., Nimmo, M. 2006.Seasonal and spatial controls on the deliveryof excess nutrients to nearshore and offshorecoral reefs of Brazil. Journal of MarineSystems, 60: 63-74.Cruz, I. C. 2008. Recifes de corais da Baía de Todosos Santos, caracterização, avaliação e identificaçãode áreas prioritárias para conservação.MSc. Dissertation. Universidade Federal daBahia, Salvador, Brasil, 102 p.Cruz, I. C., Kikuchi, R. K. P. & Leão, Z. M. A. N.2009. Caracterização dos recifes de corais daárea de preservação ambiental da Baía deTodos os Santos para fins de manejo, Bahia,Brasil. Revista da Gestão CosteiraIntegrada, 9: 16-36.De Paula, A. F. & Creed, J. C. 2004. Two species ofthe coral Tubastraea (Cnidaria, Scleractinia)in Brazil: a case of accidental introduction.Bulletin of Marine Science, 74(1): 175-183.Dutra, L. X. C. 2000. O branqueamento de coraishermatípicos no litoral norte da Bahiaassociado ao evento El-Niño 1998. BSc.Monography, Universidade Federal da Bahia,Salvador, Brasil. 78 p.Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

234Z. M. A. N. LEAO ET ALLIDutra, L. X. C., Kikuchi, R. K. P. & Leão, Z. M. A.N. 2006. Todos os Santos Bay Coral Reefs,Eastern Brazil, revisited after 40 years.Proceeding 10 th International Coral ReefSymposium Okinawa, 1: 1090-1095.Francini-Filho, R., Moura, R. L., Thompson, F. L.,Reis, R. M., Kaufman, L., Kikuchi, R. K. P. &Leão, Z. M. A. N. 2008. Diseases leading toaccelerated decline of reef corals in the largestSouth Atlantic reef complex (Abrolhos Bank,eastern Brazil). Marine Polution Bulletin,56: 1008-1014.Francini-Filho, R., Reis, R. M., Meirelles, P. M.,Moura, R. L., Thompson, F. L., Kikuchi, R.K. P. & Kaufman, L. 2010. Seasonalprevalence of white-plague like disease on theendemic Brazilian reef coral Mussismiliabraziliensis. Latin American Journal ofAquatic Research, 38: 292-296.Ginsburg, R. N., Kramer, P. A., Lang J. C. & Sale,P. 1998. Atlantic and Gulf Rapid ReefAssessment (AGRRA) revised rapidassessment protocol (RAP) http://agrra.org(Assessed: 1998 to 2008).Goldberg, J. & Wilkson, C. R. 2004. Global Threatsto Coral Reefs: Coral Bleaching, GlobalClimate Change, Disease, Predator Plagues,and Invasive Species.Pp 67-92. In: Wilkson,C. (Ed.) Status of Coral Reefs of the World2004. GCRMN, Australian Institute of MarineScience, Townsville, Australia, 301 p.Harvell, C. D., Kim, K., Burkholder, J. M., Colwell,R. R., Epstein, P. R., Grimes, D. J., Hofmann,E. E., Lipp, E. K., Osterhaus, A. D. M. E.,Overstreet, R. M., Porter, J. W., Smith, G. W.& Vasta, G. R., 1999. Emerging marinediseases: climate links and anthropogenicfactors. Science, 285: 1505-1510.Harvell, C. D., Mitchell, C. E., Ward, J. R., Altizer,S., Dobson, A. P., Ostfeld, R. S. & Samuel,M. D. 2002. Climate warming and disease riskfor terrestrial and marine biota. Science, 296:2158-2162.Hoegh-Guldberg, O. 1999. Climate change, coralbleaching and the future of the world’s coralreefs. Marine and Fresh Water Research,50(8): 839-866.Hughes, T. P., Baird, A. H., Bellwood, D. R., Card,M., Connolly, S. R., Folke, C., Grosberg, R.,Hoegh-Guldberg, O., Jackson, J. B. C.,Kleypas, J., Lough J. M., Marshall, P.,Nystron, M., Palumbi, S. R., Pandolfi, J. M.,Rosen, B. & Roughgarden, J. 2003. Climatechange, human impacts and the resilience ofcoral reefs. Science, 301: 929-933.Kikuchi, R. K. P. 2000. Modificações nacomunidade dos corais construtores dosrecifes da costa norte da Bahia durante aregressão holocênica. PhD. Thesis, Institutode Geociências, Universidade Federal daBahia, Salvador, Brasil. 148 p.Kikuchi, R. K. P. & Leão, Z. M. A. N. 1998. Theeffects of Holocene sea level fluctuations onreef development and coral community structure,North Bahia, Brazil. Anais da AcademiaBrasileira de Ciências, 70: 159-171.Kikuchi, R. K. P., Leão, Z. M. A. N., Testa, V.,Dutra, L. X. C. & Spanó S. 2003a. Rapidassessment of the Abrolhos Reefs, EasternBrazil (Part 1: stony corals and algae). AtollResearch Bulletin, 496: 172-187.Kikuchi, R. K. P., Leão, Z. M. A. N., Sampaio, C. L.S., & Telles. M. D. 2003b. Rapid assessmentof the Abrolhos Reefs, Eastern Brazil (Part 2:fish communities). Atoll Research Bulletin,496: 189-204.Kikuchi, R. K. P., Leão, Z. M. A. N. & Oliveira, M.D. M. 2010. Spatial patterns of AGRRAvitality indexes in Southwestern AtlanticReefs. International Journal of TropicalBiology, 58(1): 1-31.Krug, L. A. 2008. Um estudo diagnóstico eprognóstico do branqueamento de corais naplataforma leste do Brasil. MSc. Dissertation,Instituto Nacional de Pesquisas Espaciais,Brasil. 87 p.Laborel, J. L. 1969. Les peuplements des madreporairesdes cotes tropicales du Brésil. AnnalesUniversitê d’Abidjan, Serie E, II(3): 1-260.Laborel, J. L. 1970. Madreporaires et hydrocoralliairesrecifaux des cotes bresiliennes.Systematique, ecologie, repartition verticale etgeographie. Annales de l’Institut Oceanographiquede Paris, 47:171-229.Leão, Z. M. A. N. 1982. Morphology, geology anddevelopmental history of the southernmostcoral reefs of Western Atlantic, AbrolhosBank, Brazil. PhD Thesis, RSMAS,University of Miami, Florida, USA, 218 p.Leão, Z. M. A. N & Ginsburg, R. N. 1997. Livingreefs surrounded by siliciclastic sediments:the Abrolhos coastal reefs, Bahia, Brazil.Proceeding 8 th International Coral ReefSymposium, Panamá, 2: 1767-1772.Leão Z. M. A. N. & Kikuchi, R. K. P. 2005. A reliccoral fauna threatened by global changes andhuman activities, Eastern Brazil. MarinePolution Bulletin, 51: 599-611.Leão, Z. M. A. N., Kikuchi, R. K. P. & Testa, V.2003. Corals and Coral Reefs of Brazil, Pp. 9-Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Brazilian coral reefs in time of climate changes23552. In: Cortés, J. (Ed.). Latin America CoralReefs, Elsevier Science, 508 p.Leão Z. M. A. N., Kikuchi, R. K. P. & Oliveira, M.D. M. 2008. Branqueamento de corais nosrecifes da Bahia e sua relação com eventos deanomalias térmicas nas águas superficiais dooceano. Biota Neotropica, 8(3): 69-82.Maynard, J. A., Jonhson, J. E., Marshall, P. A.,Eakin, C. M., Goby, G., Shuttenberg, H. &Spillman, C .M. 2009. A strategic frameworkfor responding to coral bleaching events in achanging climate. Environmental Management,44(1): 1-11.Meirelles, P. M. 2009. Branqueamento no coralMontastrea cavernosa (Linneaus 1766) nosrecifes de Abrolhos, Bahia – Brasil. BSc.Monography, Universidade Federal da Bahia,Instituto de Biologia, 107 p.Muller, E. M., Rogers, C. S., Apitzack, A. S. &Woesik, R. V. 2008. Bleaching increaseslikelihood of disease on Acropora palmata(Lamarck) in Hawksnest Bay, St. John, USVirgin Islands. Coral Reefs, 27(1): 191-195.Neves, E. G. 2004. Complexo Siderastrea: espéciesdistintas? Significado da variabilidade dogênero Siderastrea de Blainville, 1830(Anthozoa, Scleractinia) no Brasil. PhD.Thesis, Universidade de São Paulo. 458 p.Neves, E. G., Jonhsson, R., Sampaio, C. L. &Pichon, M. 2006. The occurrence of Scolymiacubensis in Brazil: revising the problem of theCaribbean solitary mussids. Zootaxa, 1366:45-54.Neves, E. G., Andrade, S. C., Silveira, F. L. &Solferini, V. N. 2008. Genetic variation andpopulation structuring in two brooding coralspecies (Siderastrea stellata and Siderastrearadians) from Brazil. Genetica (The Hague),132(1): 379-384.Nolasco, M. C. 1987. Construções carbonáticas dacosta norte do Estado da Bahia (Salvador aSubauna). MSc. Dissertation, UniversidadeFederal da Bahia, Salvador, Brasil, 143 p.Poggio, C. A. Leão, Z. M. A. N. & Mafalda-Junior,P. 2009. Registro de branqueamento sazonalem Siderastrea spp. em poças intermareais dorecife de Guarajuba, Bahia, Brasil. Interciência,34(7): 502-506.Reis, M. A. C. & Z. M. A. N. Leão. 2003.Bioerosion rate of the sponge Cliona celata(Grant 1826) from reefs in turbid waters,North Bahia, Brazil. Proceeding 8 thInternational Coral Reef Symposium,Panamá, 1: 273-278.Riegl, B., Bruckner, A., Coles, S. L., Renaud, P. &Dodge, R. E. 2009. Coral reefs threats andconservation in an era of global change.Ecology and Conservation Biology, 1162:136-186.Rodriguez-Ramirez, A., Bastidas, C., Cortes, J.,Guzman, H., Leão Z., Garzon-Ferreira, J.,Kikuchi, R., Ferreira, B. P., Alvarado, J. J.,Jimenez, C., Fonseca, A. C., Salas, E., Nivia,J., Fernandez, C., Rodriguez, S., Debrot, D.,Croquer, A., Gil, D., Gomez, D. I., Navas-Camacho, R., Reys-Nivia, M. C., Acosta, A.,Alva-rado, E., Pizarro, V., Sanjuan, A.,Herron, P., Zapata, F. A., Zea, S., Lopez-Victoria, M. & Sanchez, J. A. 2008. Status ofcoral reefs and associated ecosystems inSouthern Tropical America: Brazil, Colombia,Costa Rica, Panama and Venezuela. Pp. 281-294. In: Wilkson, C., Status of Coral Reefsof the World: 2008. Global Coral ReefsMonitoring Network (GCRMN). Townsville,Australia, 298 p.Rosenberg, E. & Loya, Y., 2004. Coral Health andDisease. Springer-Verlag, Berlin, 509p.Santa-Izabel, L. M. 2001. Caracterização da bioerosãointerna dos recifes de Guarajuba comênfase nos macroperfuradores bivalvos, sipunculídeose poliquetas, Litoral Norte do estadoda Bahia. PhD. Thesis, Univeridade Federalda Bahia, Salvador, Brasil. 141 p.Selig, E. R., Harvell, C. D., Bruno, J. F., Willis, B.L., Page, C. A., Casey, K. S. & Sweatman, H.,2006. Analyzing the relationship betweenocean temperature anomalies and coraldisease outbreaks at broad spatial scales.In: Phinney, J. T., Strong, A., Skrving, W.,Kleypas, J. & Hoegh-Guldberg, O. (Eds.).Coral Reefs and Climate Change: Scienceand Management. AGU Monograph Series,Coastal and Estuarine Studies, 61: 111–128.Spanó, S., Leão, Z. M. A. N. & Kikuchi, R. K. P.2008. Diagnóstico do estado de conservaçãodos recifes em franja do Parque NacionalMarinho dos Abrolhos. OLAM Ciência &Tecnologia, 8(2): 245-277.Veron, J. E. N. 2000. Corals of the World. Vol 3.Australian Institute of Marine Sciences andCRR Qld Pty Ltd, Australia, 469 p.Received October 2009Accepted June 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 224-235

Temporal and meridional variability of Satellite-estimates of surfacechlorophyll concentration over the Brazilian continental shelfÁUREA M. CIOTTI 1 , CARLOS A. E. GARCIA 2 & DANIEL S. F. JORGE 11 Laboratório Aquarela, UNESP- Campus Experimental do Litoral Paulista, Praça Infante Dom Henrique S/N ParqueBitaru, São Vicente (SP), 11330-900, Brazil.2 Instituto de Oceanografia, Universidade Federal do Rio Grande , Av Itália km 8, Rio Grande (RS), 96201-900, Brazil.Abstract. Forecast of biological consequences of climate changes depend on both long-termobservations and the establishment of carbon budgets within pelagic ecosystems, including theassessment of biomasses and activities of all players in the global carbon cycle. Approximately 25% ofoceanic primary happen over continental shelves, so these are important sites for studies of global carbondynamics. The Brazilian Continental Shelf (BCS) has sparse and non-systematic in situ information onphytoplankton biomass, making products derived from ocean color remote sensing extremely valuable.This work analyzes chlorophyll concentration (Chl) estimated from four ocean color sensors (CZCS,OCTS, SeaWiFS and MODIS) over the BCS, to compare Chl and annual cycles meridionally. Also,useful complementary ocean color variables are presented. Chl gradients increased from the centralregion towards north and south, limited by estuarine plumes of Amazon and La Plata rivers, and clearannual Chl cycles appear in most areas. In southern and central areas, annual cycles show strong seasonalvariability while interannual and long-term variability are equally important in the remaining areas. Thisis the first comparative evaluation of the Chl over the BCS and will aid the understanding of its long-termvariability; essential initial step for discussions of climate changes.Keywords: South West Atlantic, ocean color remote sensing, chlorophyll concentration, AnnualVariability, CDOM index, Fluorescence Line HeightResumo. Variabilidade temporal e meridional de estimativas de Satélite da concentração declorofila superficial na plataforma continental brasileira. Previsões de conseqüências biológicasnos oceanos às alterações do clima dependem de bases de dados longas e da quantificação das trocas decarbono nos ambientes pelagiais, incluindo a caracterização das biomassas e atividades de organismoschavedo ciclo global de carbono. Cerca de 25% da produtividade primária global acontece nasplataformas continentais, assim essas são regiões de estudo essenciais na dinâmica ciclo de carbono. NaPlataforma Continental Brasileira (PCB), as informações sobre a biomassa do fitoplâncton são esparsas,tornando dados de satélite da cor do oceano valiosos. Nesse trabalho, analisamos concentrações declorofila (Chl) estimadas por quatro sensores (CZCS, OCTS, SeaWiFS e MODIS) sobre a PCB paracompará-la meridionalmente e sazonalmente. Apresentamos ainda duas variáveis (linha de fluorescênciae o índice de Matéria Orgânica Dissolvida) que podem ser usadas para a interpretação da Chl. Osgradientes de Chl enfatizam duas áreas extremas, influenciadas pelas plumas dos rios Amazonas e LaPlata, e forçantes interanuais nas regiões centrais. Os resultados são a primeira comparação da Chl naPCB, que poderá guiar estudos futuros para o entendimento de suas variabilidades em longa escala, etapainicial fundamental para estudos sobre mudanças climáticas.Palavras-chave: Atlântico Sul Ocidental, Cor do Oceano por sensoriamento remoto,Concentração de Clorofila, Variações Anuais, Altura da Linha de Fluorescência, Índice de MODCIntroductionThe discussion of effects of climate changein oceanic biological processes remainscontroversial. While reports show fast increases inboth CO 2 concentrations in the atmosphere and inocean’s temperatures (IPCC 2007), the biologicalconsequences directly related to these changes areoften questionable. Detailed budgets for the globalcarbon cycle are oversimplified (Houghton 2007),Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration237and current predictive models do not accommodatethe role of biological processes. More importantly,the lack of long-term biological observations makesit difficult to detect robust changes with time in theoceanic biota. Therefore, efforts are needed toquantify abundances and distributions of keybiological marine players in the carbon cycle, suchas phytoplankton (Anderson 2005), and to describequantitatively the processes in which they participate(Le Quere et al. 2009).In the open ocean, the influence of climatechanges in processes mediated by phytoplankton aregenerally based on trends with time, derived fromvariables such as chlorophyll concentration orprimary productivity (PP) rates (Sarmiento et al.2004). Decreases in PP rates have been related toenhancement of vertical stratification in warmwaters (Polovina et al. 2008), while increases areassociated with increases in sea water temperature incold regions (Rysgaard et al. 1999). Biologicaleffects are also link to episodic events such asincreases on chlorophyll concentration afterpassages of hurricanes (Goldenberg et al. 2001). Incontinental shelves, expected changes includealtering freshwater outflows (Belkin 2009) and thestrength and temporal scales of costal upwellingevents (Bakun 1990, Mote & Mantua 2002), andhence nutrient inputs and circulation patterns. Allthese biological processes will vary regionally,making it imperative to characterize them withinthese scales.Long-existent observational sites thatinclude systematic acquisition of biological dataare infrequent in the global ocean, but recentinternational programs and funding efforts (Dickeyet al. 2009, Johnson et al. 2009) will certainlychange this panorama the future (e.g.complementary sensors in Argo floats, seehttp://www.argo.ucsd.edu/). Nonetheless, a keysource for observing and understanding the upperocean layer remain satellite data, which providesinoptic views of biological oceanographic processesthrough variables retrieved from ocean colorsensors. The spectral reflectance emerging from theoceans are used in empirical algorithms andmodeling techniques to estimate phytoplanktonbiomass (i.e, chlorophyll concentration; see O'reillyet al. 1998) and other dissolved and particulatecomponents (Carder et al. 1986, Ciotti et al. 1999,Lee et al. 1998, Maritorena et al. 2002, Roesler &Perry 1995). Ocean color data combined withinformation on sea surface temperature,downwelling irradiance and mixed layer depth allowfor estimates of primary production rates (Campbellet al. 2002, Carr et al. 2006, Friedrichs et al. 2009).More recently, a number of bio-optical models weredesigned to discriminate among phytoplankton typesor communities (Alvain et al. 2005, Ciotti &Bricaud 2006, Sathyendranath et al. 2004).Freely available ocean color data existssince 1978 (McClain 2009) being global for the past12 years (http://oceancolor.gsfc.nasa.gov). Logically,decade-long time series are inadequate tosubsidize climate change studies, but these data hasimproved our knowledge on annual cycles ofphytoplankton biomass (Kahru et al. 2004,Longhurst 1995), and can also be helpful to identifyinter-annual variability (Henson and Thomas 2007).It is essential to understand how phytoplanktonbiomass change in time in order to developconceptual models on phytoplankton dynamics, andocean color is still the only source of information inmany regions where observational studies have beensparse.The Brazilian Continental Shelf (BCS)occupies over 40º degrees of latitude (Fig. 1) andcontains significant regional differences concerningmainly its extension, the influence of offshorecirculation, the overall area of the continental shelfand continent runoff inputs (Castro & Miranda 1998,Castro et al. 2006). Thus, the physical processescontrolling nutrients and light availability forphytoplankton growth or accumulation varymeridionally. Surveys performed on the BCSdescribing temporal and spatial distributions ofphytoplankton biomass and primary production havebeen fairly unsystematic, but latitudinal differenceswere reported and temporal patterns have alreadybeen showed in some areas (Brandini 1997, Gaeta &Brandini 2006, Ciotti et al. 2006). It is importantalso to mention that these sparse observations havebeen derived from a variety of methods, whichdespite of being standard oceanographic procedureshave never being inter-compared to date.It is acknowledged that ocean color productshave been developed for open ocean and theiruse over continental shelves can be sometimesproblematic, especially in areas receivingconsiderable amounts of continental outflows. Theglobal algorithms that retrieve chlorophyllconcentration (Chl) presume a covariance betweenin situ Chl and the relative contributions of all otheroptically active components in the light field, thus,so when them effectively absorb blue-light, which isthe case for detritus or colored dissolved matter(CDOM), Chl is overestimated. The MODIS/Aquaocean color sensor has additional spectral red bandsthat are sensitive to the natural fluorescence forchlorophyll present in living cells (Esaias et al.1998, Gower et al. 2004). NASA is currentlyPan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

238Á. M. CIOTTI ET ALLIdistributing these data as “evaluation products”named fluorescence line height (FLH). Theinterpretation of FLH is not trivial (Huot et al. 2005,2007), and many issues regarding the role ofphytoplankton physiology and taxonomy over theFLH signal and on the actual efficiency offluorescence by phytoplankton living at the surfaceof the ocean remain unresolved (Schallenberg et al.2008). Nonetheless, in a first approximation, FLHregistered over continental shelf areas that receivesignificant continental outflow can be an alternativeor a complement for blue-green ratios algorithms’for Chl estimates, as the contribution of detritus andCDOM to FLH tend to be minimal (Gower et al.2004). Even in the open ocean, CDOM is animportant optical component for light absorption(Siegel et al. 2002), and recently, an additional bioopticalproduct has been developed (Morel & Gentili2009) and distributed by NASA - the CDOM indexthatintends to offer a metric option to observeCDOM versus phytoplankton influences in a givenarea.Figure 1. The Brazilian Continental Shelf (BCS) and thesubdivision used in this work (Areas 1 to 7 from North toSouth). Image color gradients show the annual standarddeviation of the annual mean chlorophyll concentration(mg.m -3 ) in log scale, for the combined 2000 to 2005 dataderived from SeaWIFs. Line represents the 200 m isobaththat defines the continental shelf boundary.The studies conducted over the BCS usingocean color data were mostly regional, except bythat from Gonzalez-Silvera et al. (2004), which alsoincluded observations in the open ocean. Publishedwork comprise studies on the behavior of NorthBrazil Current (Johns et al. 1990, Richardson et al.1994, Fratantoni & Glickson 2002), the confluenceof Brazil and Malvinas Currents (Garcia et al. 2004,Saraceno et al. 2005, Barre et al. 2006, Gonzalez-Silvera et al. 2006), Amazon & La Plata Plumessignatures (Froidefond et al. 2002, Hu et al. 2004,Del Vecchio & Subramaniam 2004, Cherubin &Richardson 2007, Garcia & Garcia 2008, Piola et al.2008, Molleri et al. 2010), and the tracking ofmesoscale features (Bentz et al. 2004). Validationand verification of ocean color models have alsobeen conducted (Garcia et al. 2005, 2006, Ciotti &Bricaud 2006).Descriptions and comparative analyses areneeded to better understand seasonal and interannualchanges of the phytoplankton standing stock overthe BCS, as space and time characterizations arecrucial initial steps for studies of eventual globalchange effects. In this work, we gathered allavailable remote sensing ocean color data for theBCS, derived from 4 ocean-color sensors, toquantify and compare observed Chl, FLH and theCDOM-index over time. We divided the BCS into 7large regions following Castro et al. (2006). Ourmain objectives were: i) to examine meridionalvariability in chlorophyll concentrations derived bysatellite; ii) to assess the quality of the dataavailable, concerning mainly data coverage; iii)inter-compare chlorophyll products from twosensors (SeaWiFS and MODIS/Aqua); and iv) toestablish annual cycles and interannual variability ofchlorophyll concentration for all regions. We willalso describe spatial and temporal trends of FLH andthe CDOM-index over the subareas, and comparethese indices to Chl.Data and MethodsSatellite DataData from mapped, 8-Day, global compositeimages (L3) were downloaded from Ocean ColorHome Page (http://oceancolor.gsfc.nasa.gov) andconsist of the entire data sets available for CoastalZone Color Scanner (CZCS), Ocean ColorTemperature Sensor (OCTS), Sea-viewing WideField-of-view Sensor (SeaWiFS) and ModerateResolution Imaging Spectroradiometer on the Aquasatellite (MODIS/Aqua) up to 31 December 2009(see Table I). These data sets are reprocessed fromtime to time, which is necessary for sensorcalibration and algorithm improvements. In the pre-Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration239sent work, data from SeaWiFS refer to reprocessing2009.1 (December 2009), data from MODIS/Aquarefer to reprocessing 1.1 (August 2005), and datafrom both CZCS and OCTS were last reprocessed inOctober 2006. MODIS/Aqua has been fullyoperational since its launch while a number of gapsfor SeaWiFs data, due to problems with theinstrument, occurred in 2008 (January 3 to April 3;July 2 to August 18) and 2009 (April 24 to June 15;July 3 to July 17; August 31 to November 5; andNovember 14 to 30). Further details on each sensorand respective data sets can be found in the officialdistribution site (http://oceancolor.gsfc.nasa.gov).Products distributed as Level 3 used in thiswork included chlorophyll concentration (allsensors), sea surface temperature (MODIS/Aqua),fluorescence line height (MODIS/Aqua) and theCDOM index (SeaWiFS), computed by therespective standard global algorithms and masks.Note that spatial resolutions for L3 images are 4 kmfor CZCS and MODIS/Aqua, and 9 km for OCTSand SeaWiFS, and were preserved as such.Images were processed using SeaDAS(v5.4) - a multi-platform software freely distributedby NASA (http://oceancolor.gsfc.nasa.gov/seadas).The bathymetry dataset from SEADAS was used toset apart image pixels occurring over the continentalshelf, assumed here as those where local depthsranged from 20 to 200 m. The lower limit of 20 mwas set to exclude the effects of both localgeography and possible bottom effects. For eachsatellite product and sensor, the proportion of datacoverage was computed as the number of pixels withvalid data (i.e., those not masked by land, clouds oralgorithm’s failures) over the total number of pixelsexpected between 20 and 200 meters. We used 8-Day spatial resolution of all products and sensorsthat yielded 46 observations per year. We alsogrouped SeaWiFS and MODIS/Aqua chlorophyllconcentration and data coverage by seasons, so thatFall refers to images starting on days of the year 80to 162, Winter images from days 168 to 258, Springimages from days 264 to 346 and Summer imagesfrom days 352 to 361 and days 1 to 74.Meridional Division and Basic StatisticsThe BCS was divided into 7 (seven) largeareas, which are a slight modification of the divisioninto six compartments proposed by Castro andMiranda (1998). The original 6 (six) areas weredivided according with physical processes andpresence of distinct water masses. Our modificationscheme was basically shifting latitudinal boundariesnecessary to i) keep the areas more comparable insize; and ii) to accommodate, within each area,similar values for mean and standard deviations ofsatellite surface chlorophyll (Fig. 1). For each 8-Daysatellite product and area, basic statistics (mean,median, standard deviation) were computed, butrespective time series represent the median values,rather the mean, in order to focus our discussion oncentral values and also to avoid extreme or toolocalized values. Nonetheless, we present the entirestatistical distributions by area and ocean colorsensor (see Fig. 2).Annual Cycle and Long-Term Variability ofChlorophyllChlorophyll data for SeaWifs were averagedover the 46 8-Day composites to produce a pseudoclimatology,or general mean, for the 12 years (up to31 December 2009). The 8-Day general means werein turn used to compute 8-Day “anomaly” values forall the composites. The “chlorophyll anomalies”were then log-transformed and used to model theamplitude and phase of the mean annual cycleobserved over each area throughout the 12-yearseries. The annual mean model is based on anonlinear least square procedure (see Garcia et al.2004 for details), that assumes a sinusoidal formwith a single amplitude over the 46 compositesregistered in a year. The ratio of the varianceexplained by the annual model to the total variance -correlation of determination (r 2 ) - for each timeseries per area was also calculated to verify thegoodness of the model, and here describe theconsistence of a seasonal variability in thechlorophyll concentration. It is important to note thatlow correlation of determination observed per areareflects both the inadequacy of the single amplitude(e.g., seasonal cycles can have more than anoscillation per year) and also interannual variability,but in this case we expect the latter to be moreimportant as the majority of BCS is located intropical and subtropical areas.To observe long-term variability inSeaWiFS Chl anomaly per area, we first applied arunning mean filter (equivalent to 46, or a year of 8-Day mean observations), to create smoothed series,each being normalized by their respective standarddeviation. The filtered data were then subjected tospectral analysis, which used the variancepreservingform of the energy spectrum (Emery &Thomson 2004) so to make the total energy of thesignals observed among areas comparable. These aresimple techniques, that were chosen to accommodateour present goals, but future work should include anumber of more sophisticated time-series techniques(e.g., EOFs; wavelets) to better understand the longtermvariability of chlorophyll fields in the differentPan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

240Á. M. CIOTTI ET ALLIFigure 2. Box plots showing the range and statistics of chlorophyll values observed by the different sensors over theseven subareas of the Brazilian Continental Shelf. Red lines represent the medians and each box represents the 50%percentile (i.e., mean is the center of the box). Note that the distributions are derived from different temporal resolutionsfrom each sensor (see Table I). Red crosses are outliers.zones of the BCS and their relation to otherenvironmental variables (ex. river discharge, windstress, surface current, etc.).Results and DiscussionDespite of the smaller area compared to theopen ocean, continental shelves account for about25% of oceanic global primary production(Longhurst et al. 1995). However, it is difficult toquantify time and spatial variability of phytoplanktonbiomass over these areas. A number ofcomplex, and not yet fully quantifiable processes,interfere on observed phytoplankton standing stocks,usually estimated with chlorophyll concentration(Chl). General spatial patterns for Chl do exist in thesurface ocean, resulting from differences in nutrientand light availability for phytoplankton growth (orChl accumulation) governed by regional and globalphysical processes (see Holt et al. 2009 andreferences therein). However, other important termsto be considered when Chl distributions arecompared in time and space include less understoodprocesses as grazing, sedimentation and advectionrates for phytoplankton cells (Banse 1994) becausethey depend on ecological and physiologicalcomponents that are difficult to observe. Overcontinental shelves, all chemical and biologicalprocesses are associated with larger Chl inPan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration241Table I. Characteristics of the four ocean color sensors that can provide information on the Brazilian Continental Shelfon chlorophyll concentration. Note operating periods.Spectral Bands (only visible and near infrared only)Sensor Central wavelength (nm) Spectral Spatial resolution Operating datesresolution (nm) (m)CZCS 1 443; 520; 550; 67020825 24/10/1978 -750OCTS 2 412; 443; 490; 520565; 670765; 865SeaWiFS 3 412; 443; 490; 510555; 690765; 865MODIS 4 /Aqua531443; 488; 551667; 678; 936412; 748; 869469; 555; 1240858905; 137594064510020204020204051010152025304050700110010001000100010005002501000100025022/6/198603/9/1996 -29/6/199729/8/1997 -present04/07/2002 -presentObs: 1 Coastal Zone Color Scanner; 2 Ocean Color Temperature Sensor, 3 Sea-viewing Wide Field-of-view Sensor,4 Moderate Resolution Imaging Spectroradiometercomparison with the open ocean, within also a morecomplex physical dynamics (Longhurst 2006). Thus,we stress that the patterns presented here formeridional and seasonal Chl are a result from anumber of processes yet to be quantified, so theyneither translate directly into primary production norexportation rates for phytoplankton carbon.Remote Sensing Chlorophyll Observed in theBCSThe meridional and seasonal distributions ofChl over the large and heterogeneous BCS (Fig. 1)can be related to the equally diverse dominantphysical influences (Castro & Miranda 1998). Whendata from the four ocean color sensors are compared,similar general statistical distributions for Chl (Fig.2) emerged, especially for SeaWiFS andMODIS/Aqua sensors, which result from almost adecade of their simultaneous data acquisition (TableII). The OCTS sensor operated for less than a year,and there are only about 30 8-Day images availablefor the BSC. Lacks in data acquisition (Table II)represent besides instrument problems, cloudcoverage and algorithm failures.The highest degree of Chl variability wasfound in the latitudinal extremes of BCS (Areas 1and 7) while Chl was remarkable constant in Area 2.The highest and lowest mean Chl values were foundin Areas 7 and 3, respectively (Fig. 2) differingabout an order of magnitude from each other(e.g., ~1.0 versus ~0.1 mg.m -3 ). Meridional 8-Daymedian Chl variability as a function of time and theChl statistical distribution (Figs. 2 and 3) suggest theimportance of the distinct degrees of seasonality inthe different areas, but as indicated by some recentstudies on physical processes (e.g., Dotori & Castro2009) we also expect a significant contribution ofinterannual variability may also be expected.Data coverage by the distinct sensors alsoshowed important seasonal patterns (Fig. 4) that willbe assessed below, but we found no significantstatistical relationships between data coverage andChl for a given Area or sensor. For instance, bothSeaWiFS and MODIS/Aqua coverage wereremarkable poor in Area 1 during the summermonths, period when Chl was also low, while Areas3 and 7 showed good data coverage year round, anda clear seasonal Chl pattern. Figures 3 and 4 pointout the discontinuity for CZCS data coverage in theBSC (see also Table II), which is a result of NASA’sstrategy of turning off the sensor when it was outsidethe main interest areas due to limited on-boardstorage data and to preserve the sensor (Mcclain,2009). Indeed, CZCS had its life expectancyexpanded for several years, but unfortunately, thelow data coverage over the BSC prevent studiesto integrate its data with those acquired by thecurrent operational sensors (e.g., SeaWiFs andMODIS/Aqua - but see also www.ioccg.org/sensors/current.html) in order to produce longertime-series. Recent literature show, for areas whereCZCS data is available, detectable trends in Chlwith time that suggested alterations in the marinebiota (e.g., Antoine et al. 2005). Studies like thesePan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

242Á. M. CIOTTI ET ALLIat the BSC are, unfortunately, limited today to12 years of continuous ocean color data data(i.e., SeaWiFs and MODIS/Aqua), and none ourchosen Areas has showed strong positive ornegative Chl trends with time. It is worthmentioning that understanding the impacts of globalchange demands long-term biological observationsand that will be jeopardized for the BCS if thecurrent NASA’s programs are discontinued. Thus,it is imperative to create and maintain new oceancolor programs and to involve the Brazilianscientific community on using and interpreting itspotential products.Seasonal and Meridional Chl variabilityThe seasonal mean Chl per area derived bySeaWiFS and MODIS/Aqua (Table III), place Area3 as a boundary of low Chl that separates gradientstowards north and south. Maximum Chl per areaincreases steeply towards north (3-fold in Area 2 andan order of magnitude in Area 1) occurring duringfall while towards south, Chl gradients were gradualand occurred during winter. In Areas 3 and 4, Chlwas high and similar in fall and winter, and low andsimilar in spring and summer. Area 2 showed asingle and modest Chl peak in the fall. Periods withminimum Chl values also varied meridionaly, andwere observed during spring in Area 1 and duringsummer in Areas 5, 6 and 7.The seasonal and meridional distribution ofmean data coverage per Area and instrument (Tab.III), showed that only half of the 8-Day compositeimages were available most of the year in Area 1and that data coverage was about 13 to 18% duringsummer and 23 to 27% during fall. Area 2 has alsoshowed poor data coverage during summer (29 to33%) and fall (44 to 53%).It was possible to model significant annualcycles in all Areas, except for Area 2 (Table IV).The best fits were found in Areas 7 and 4,suggesting that the main processes for accumulationand loss of phytoplankton biomass operate ratherseasonally, especially in Area 7 (r 2 =0.78). Allremaining Areas showed correlations of 0.52 to 0.58to the annual model, probably a result of importantinter-annual variability as in Area 1 in 2001 andArea 6 during 2007 (Fig. 5), for instance.The meridional patterns of Chl were indeedrelated to the main reported seasonal changes inhydrography. In Area 1, the most prominent featureis the Amazon River discharge, which providesseasonally dissolved colored material, nutrients,sediments and detritus to larger extensions of thecontinental shelf. During the rainy seasons (summerand spring) larger river discharge is combined withchanges in both wind direction (from SE to NE) andin transport rates of the North Brazil Current(Molleri et al. 2010). Both the size and theorientation of the Amazon plume are responsible forthe magnitude and seasonal Chl variability and alsoreflect the amount of particles and dissolved loadthat can interfere in the Chl algorithm performance.Chl changes in Areas 2 and 3 are likely to beassociated with variability in the biologicalprocesses, probably driven by changes in watertemperature and irradiance, since neither major riverflow nor relevant oceanographic features (e.g.,upwelling) are reported for those two areas, whichare the least studied ones of our BSC subdivision. InArea 2, no significant parameters could be found forthe annual variability in Chl (Table IV), whichshowed a fall Chl peak only about 25% higher thanthat for the remaining of the year. Area 3 showed thelowest Chl in comparison to all other Areas, with astrong seasonal cycle. The annual amplitudesdetected by the SeaWiFs time series were 0.04 and0.12 mg.m -3 for Areas 2 and 3, respectively (TableIV). It is important to remember that Area 3 is thesmallest despite occupying over 10 degrees oflatitude.Areas 2 to 4 are also under direct influenceof the seasonal and meridional migration of theInter-Tropical Convergence Zone (ITCZ) in thesouth Atlantic and thus they will likely respond toany mode of change in the wind fields. Northern andweak winds occur in the summer, enhancing verticalstratification that in turn is expected to disfavor Chlaccumulation. In the winter, southern and strongerwinds have been already linked to increases in Chlin Area 4 due to the erosion of the picnocline andconsequent fertilization of the surface layers (Ciottiet al. 2006). In addition, Area 4 comprises theAbrolhos region with a shallow continental shelf,allowing addition of nutrients by ressuspension ofbottom water masses. The seasonal influence ofwind regimes on vertical mixing is suggested by thegood adjustment of the Chl series to the annualmodel (Table IV). Abrolhos, and a number of banksadjacent to Area 4 affect the flow and direction ofthe Brazil Current (BC). As a consequence, the mainflow of the BC shows mesoscale instabilities andmeanders (Gaeta et al. 1999) that have a significantimpact on the circulation of the Areas towards thesouth (Calado et al. 2008, Silveira et al. 2008).Areas 5 and the northern third of Area 6encompass the most studied regions of the BSC (seereview in Castro et al. 2006). Areas 5 and 6 wereseparated at São Sebastião Island latitude because oftheir important distinct oceanographic features. Area5 includes Cabo São Tomé and Cabo Frio, wherePan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration243Table II. Division for the BSC in subareas based on Castro et al. (2006). For each sensor and area, columns representthe number of 8-Day images with no data; n is the length of each sensor data sets (up to the end 2009 for thoseoperational) for 8-Day images. Also, the total number of Pixels over the delimited continental shelf (20 to 200 m): 9 kmfor OCTS and SeaWiFS and 4 km for CZCS and MODIS/Aqua.Area and latitudinalrangeSeaWIFS(n=568)Modis/Aqua(n=345)OCTS(n=30)CZCS(n=353)Number of 9 KmpixelsNumber of 4 Kmpixels1 - 04N-01S 37 3 3 282 1742 69182 - 01S-05S 35 2 2 240 561 22373- 05S-15S 36 2 0 259 258 10174- 15S-21S 39 2 1 285 656 25875- 21S-24S 39 2 3 298 689 27276- 24S-28S 37 2 0 297 1229 49377- 28S-34S 37 2 0 270 1288 5168Figure 3. Time series of 8-Day median chlorophyll concentration over the seven Areas of the BCS (see Table I forlimits), provided by the four ocean color sensors (see legend) during their respective operation periods.Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

244Á. M. CIOTTI ET ALLIcostal upwelling occurs in the summer (see Guentheret al. 2008 and references). The upwelling plumestend to move offshore in São Tomé, and towardssouth through the inner and median continentalshelves in Cabo Frio. In the revision by Guenther etal. (2008), it is presented that the upwelling of theSouth Atlantic Central Water (ACAS) to the surface,shifs the system from oligotrophic to eutrophicconditions, and thus, Chl accumulation would beexpected during the summer (period where theupwelling events are more intense), which was notshown in the present analyses (Table III). Indeed,maximum Chl occur in winter in Area 5. Numericalmodels and observations, however, show that theshelf currents respond to mesoscale wind fields,which are most intense during winter in response topassages of atmospheric cold front systems (Dotori& Castro 2009). Also, it is important to note that astrong deep Chl maximum associated with thepermanence of ACAS intrusions at the subsurfacehave been observed in Area 5 (Sumida et al. 2005)which perhaps the ocean color sensors do not notefficiently detect (André 1992). Other possible causeof the seasonal Chl pattern observed in Area 5 (and6) is related to the flow of BC that in the summer iscloser to the coast (Silva Jr. et al. 1996) inducingFigure 4. Time series of 8-Day median data coverage (i.e., 1 would represent that 100% of the subarea had valid data)concentration over alternated subareas of the BCS (see Table I), provided by the four ocean color sensors (see legend).Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration245coastal downwelling of oligotrophic warm waters atAreas 5 and 6. Upwelling events in Cabo Frio areusually short, and our approach compared largeAreas and 8-Day image composites, therefore, therelative importance of the coastal upwelling processmay have been masked. Nonetheless, we cannotexclude the role of biological factors such as grazingand sedimentation explaining these patterns. Notealso that the upwelling plumes tend to stay in Area5, but the modeled annual amplitudes for both Areas4 and 5 are exactly the same (Tab. IV). Mean Chl isonly slightly higher in Area 5 during winter, whilethe goodness for the annual fit is higher in Area 4,suggesting that interannual influences on Chlaccumulation are more important in Area 5 than inArea 4.In winter, Areas 6 and 7 (but mainly Area 7,as we discussed below) receive a coastal water massoriginated from the south portion of Brazil, withnutrients from the outflows of both Rio de la Plataand Lagoa dos Patos (Piola & Romero 2004,Pimenta et al. 2005). Indeed, the relativecontribution of Colored Dissolved Organic Mater(CDOM) has been showed to be higher in winter inArea 6 (Ciotti unpublished data) which may implythat the Chl values could be overestimated due topresence of CDOM. Note that the goodness for thefit to the annual model in Area 6 is about the sameof that observed in Area 5, suggesting importantinterannual forcing in Area 6 as well.Area 7 comprises the southernmost portionof Brazilian continental shelf waters. The influenceof both La Plata River and Pato’s Lagoon dischargeson ocean color imagery has been recently reported(Garcia & Garcia 2008, Piola et al. 2008). Theextent of the La Plata plume over the Brazilian shelfis associated with both surface winds and riverdischarge (Piola et al. 2008, Garcia & Garcia 2008).During winter, the La Plata plume extension isassociated with more intense and persistentFigure 5. Annual cycles of chlorophyll concentration (Chl, mg.m -3 ) adjusted for each subarea (red solid line) versusobserved Chl values in each 8-Day period (blue doted line).Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

246Á. M. CIOTTI ET ALLIfrom Area 7 to Area 3 during winter (Table III) thatis probably related to the seasonal migration of LaPlata River plume reflected in the amplitudes for theannual fits up to Area 3 (Table IV).Comparison Between SeaWiFS and MODIS/Aqua Chl and Data CoverageBecause of the inconsistencies for both CSZC andOCTS data acquisition over the BCS, our seasonalcomparisons over the seven Areas were performedonly for SeaWiFS and for MODIS/Aqua (Table III),during periods when the 8-Day composite imageswere available for both sensors. The non-parametricstatistical tests to access significant differencesbetween sensors for the same sample size (Table V)show statistically comparable Chl for the twosensors in most, but not in all Areas. Significantdifferences were mainly related to data coverage,especially during winter and fall, when SeaWiFSdata coverage was higher in Area 1 butMODIS/Aqua data coverage was higher in theremaining Areas. Differences in Chl were detectedin Areas 1, 2 and 3, with no apparent seasonalpatterns. Note that we used data from MODIS/Aquareprocessing 1.1, performed in August 2005.Figure 6. A variance-preserving plot of power energy (inmg2 m-6) for the normalized deseasoned chlorophyll timeseries from Areas 1 and 3 (above), Areas 4 and 5(middle), and Areas 6 and 7 (below). The vertical dashedlines stand for ½ year, 1 year, 2 year and 4 years cycles,from right to left, respectively.northeast winds rather than with increases indischarge by the La Plata River. Garcia and Garcia(2008) demonstrated the annual cycle as the mostdominant mode of variability over the southernBrazilian continental shelf, and the associated highamplitude in the annual cycle (see Table IV) ismainly controlled by the seasonal variation in theincursion of La Plata plume. The northward La PlataRiver plume extension varies from year to year(Piola et al. 2008) leading to a strong interannualvariability in Chl is also expected in the region. Thefew observations also show larger in situ chlorophyllconcentration during spring and winter in Area 7(Ciotti et al. 1995). The meridional gradient of ChlLong-term variability of chlorophyll based onSeaWiFS 12-years periodThe analyses of the interannual variability inChl fields given by SeaWiFS dataset (Sept. 1997 toDec. 2009) show clear seasonal cycles all regions,except in Area 2 (Table IV). In addition, the amountof energy contained in Chl anomalies derivedspectra was an order of magnitude higher in Areas 1and 7 (Fig. 6) due to the presence of AmazonasRiver (Area 1) and La Plata River (Area 7). Exceptfor Area 2, the results suggest long term variation (>2 years) in most of BCS. Note that the vertical axisin Figure 6 represent the total energy, given in[chla]2, where the signal variance is preservedwithin the frequency spectrum. A 4-years signal wasobserved in Areas 4 to 7 a pattern already detectedin southern BCS (Garcia & Garcia 2008), whoassociated this signal with both cycles of La Platariver discharge and alongshore winds. In theremaining areas, the cycles may be associated withlong-term changes on the Trade winds regime andmesoscale features.Complementary informationIt is outside the scope of this work to detailother variables than Chl and data coverage by thefour ocean color sensors. However, the presence of asouth to north gradient of Chl values in the winterthat may be related to the seasonal meridionalPan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration247excursion of the La Plata river plume, lead us toinvestigate the general patterns observed in twoadditional ocean color products that maycomplement and aid the interpretation of Chlpatterns (Fig. 7). The hypothesis to test is that thesouth-north Chl gradient observed in winter reflectspartially the influence of continental outflows onover-estimatives for Chl due to the presence ofCDOM.As mentioned before, MODIS/Aquaprovides data on sea surface temperature (SST) andchlorophyll fluorescence line height (FLH) that cancomplement Chl, and a new ocean color product(available for SeaWiFS and MODIS/Aqua) - theCDOM-index – can illustrate the relative importanceof this component over phytoplankton. In a firstapproximation, FLH appears to confirm the trendsshown by Chl:i) Area 3 showed the lowest FLH andCDOM-index, while the maximum were observed inboth Areas 1 and 7. However, Area 2 is no longerthe least variable, being replaced by Area 3 for bothFLH and CDOM-index, which suggest that theseasonal variability on Area 3 may be a seasonalcontribution of CDOM by changes in wind field.There is a contrast between Area 1 and 7 regardingthe CDOM-index, much more evident in the south.Note also that Areas 5 and 6 have different meanFLH but similar Chl, with an apparent largercontribution of CDOM in Area 6 that may explainthis divergence. A discussion in depth of thesedifferences among products would require validationprograms and the development of semi-analyticalregional models. Our goal here is to point out thatdata are already available for scientific interests inphysical and biological processes in the BCS.Although these products are not free of somefundamental errors, they may be used to askscientific questions and design better surveys.Figure 7. Statistical distributions for all available data per Area for Sea Surface Temperature (SST, MODIS/Aqua),Chlorophyll concentration (log scale; Chla SeaWiFS), Fluorescence Line Height (FLH, MODIS/Aqua) and the CDOMindex (SeaWiFS).Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

248Á. M. CIOTTI ET ALLITable III. Seasonal chl cycles and proportion of data coverage (i.e., 0 would represent no data while 1 would representthat all pixels had valid data) per region (Seawifs MODIS). Values represent the overall means for all 8-Day medianvalues of Chl observed in each area. These means were computed for periods with concurrent data acquired by bothsensors, up to 31 December 2009.SeaWiFSSeason Fall Winter Spring SummerArea andlatitudinalrangeCHLMean(Std)1 - 4N-1S 2.17(2.05)2- 1S-5S 0.32(0.23)3- 5S-15S 0.16(0.05)4- 15S-21S 0.23(0.07)5- 21S-24S 0.28(0.11)6- 24S-28S 0.27(0.11)7- 28S-34S 0.66(0.42)COVERMean(Std)0.27(0.20)0.44(0.23)0.62(0.18)0.74(0.21)0.86(0.20)0.79(0.21)0.72(0.24)CHLMean(Std)0.54(0.57)0.27(0.03)0.15(0.04)0.26(0.07)0.38(0.12)0.55(0.43)1.61(0.79)COVERMean(Std)0.53(0.17)0.71(0.15)0.70(0.15)0.73(0.20)0.86(0.21)0.75(0.24)0.77(0.20)CHLMean(Std)0.25(0.07)0.25(0.03)0.10(0.02)0.16(0.05)0.28(0.11)0.25(0.11)0.78(0.52)MODIS/AquaCOVERMean(Std)0.41(0.12)0.58(0.12)0.73(0.18)0.52(0.30)0.57(0.30)0.48(0.28)0.74(0.25)CHLMean(Std)0.91(1.72)0.25(0.07)0.11(0.02)0.15(0.03)0.23(0.21)0.18(0.04)0.30(0.13)Season Fall Winter Spring SummerArea andlatitudinalrangeCHLMean(Std)1 - 4N-1S 1.72(1.83)2- 1S-5S 0.36(0.24)3- 5S-15S 0.18(0.06)4- 15S-21S 0.24(0.09)5- 21S-24S 0.27(0.09)6- 24S-28S 0.25(0.10)7- 28S-34S 0.59(0.32)COVERMean(Std)0.23(0.17)0.53(0.23)0.67(0.17)0.83(0.16)0.93(0.10)0.87(0.16)0.82(0.19)CHLMean(Std)0.50(0.690.26(0.030.17(0.060.28(0.070.36(0.12)0.46(0.32)1.47(0.82)COVERMean(Std)0.46(0.14)0.78(0.10)0.73(0.14)0.83(0.15)0.91(0.16)0.83(0.22)0.82(0.17)CHLMean(Std)0.29(0.56)0.26(0.03)0.11(0.02)0.17(0.05)0.26(0.11)0.24(0.08)0.76(0.54)COVERMean(Std)0.28(0.11)0.57(0.11)0.71(0.15)0.61(0.27)0.62(0.30)0.54(0.25)0.77(0.18)CHLMean(Std)0.57(0.85)0.28(0.07)0.13(0.06)0.15(0.02)0.20(0.05)0.19(0.06)0.31(0.11)COVERMean(Std)0.18(0.13)0.29(0.17)0.62(0.24)0.58(0.27)0.69(0.29)0.65(0.28)0.78(0.18)COVERMean(Std)0.13(0.10)0.34(0.16)0.65(0.21)0.69(0.24)0.77(0.23)0.70(0.25)0.80(0.15)Recommendations and ConclusionsWe have shown that the ocean color data areadequate for time series studies but they arecurrently only a decade-long and quantification oflong-term trends is not feasible at present time forthe BCS. Our simple analyses have shown however,the presence of long-term and interannual variabilitythat must be taken into account in future studies onBCS. Systematic ocean color measurements frommulti-instrument, multi-platform and multi-yearobservations are needed to understand how annualand decadal-scale climate variability affects thegrowth of phytoplankton on the continental shelves.It is important to have satellite ocean color dataavailable for the scientific community to accesschanges in phytoplankton biomass, dissolvedmaterial and other derived products. However,SeaWiFS, MODIS, and MERIS sensors are eitherwell beyond or nearing the end of their design lives(Mc Clain 2009). The continuity for these productswill only be achieved over the next decades if aneffort is made to launch new ocean color sensors.The data from new sensors have also to beopen to all ocean color researchers, including thepre-launch characterization and on-orbit calibrationPan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration249Table IV. Correlation values of the annual model for eachregion and their respective amplitudes. All values arestatistically significant (>99% level) except for region 2.Area andlatitudinal rangeR 2 to theAnnual fitAmplitudemg.m -31 - 4N-1S 0.524 0.442- 1S-5S 0.083 0.043- 5S-15S 0.567 0.124- 15S-21S 0.640 0.165- 21S-24S 0.542 0.166- 24S-28S 0.576 0.237- 28S-34S 0.783 0.40data, so a critical and constructive discussion is setin place to guarantee the quality of the products andthe preservation of the experience from pastprograms. Among the new planned sensors are theOcean Color Monitor (OCM-2, India) and theVisible-Infrared Imaging Radiometer Suite (VIIRS,United States). Brazilian and Argentineangovernments will also launch an ocean color sensor(SABIA-MAR) and at present the design of potentialinstruments is under discussion.For studying the BCS in the future, we mustbe aware of the necessity of both improving existingocean color algorithms and validate ocean colorproducts. Regional algorithms may also be neededin certain areas. Recent improvements in fieldobservation capabilities and the increase in numberReferencesAlvain, S., Moulin, C. Dandonneau, Y. & Breon, F.M. 2005. Remote sensing of phytoplanktongroups in case 1 waters from global SeaWiFSimagery. Deep-Sea Research Part I-OceanographicResearch Papers, 52: 1989-2004.Anderson, T. R. 2005. Plankton functional typemodelling: running before we can walk?Journal of Plankton Research, 27: 1073-1081.André, J. M. 1992. Oceanic color remote sensingand the subsurface vertical structure ofphytoplankton pigments. Deep-Sea Research,39(5): 763-779.Antoine, D., Morel, A., Gordon, H. R., Banzon, V.F. & Evans, R. H. 2005. Bridging ocean colorobservations of the 1980s and 2000s in searchof long-term trends. Journal of GeophysicalResearch, 110: 1-22.Bakun, A. 1990. Global Climate Change andIntensification of Coastal Ocean Upwelling.Science, 247: 198-201.Banse, K. 1994. Grazing and zooplanktonproduction as key controls of phytoplanktonproduction in the open ocean. Oceanography,7: 13-20.of experts in bio-optical algorithms (empirical andsemi-analytical) will definitely push this line ofresearch forward, but programs must be designedand conducted in long term. However, the Brazilianocean color community is concentrated in fewinstitutions in the South and Southwest Brazilian(e.g., Garcia et al. 2005, Kampel et al. 2009).Products and models will improve with time, if along-term program is established for all thenecessary steps and not only for data acquisition.The experience of countries like UK and USA hasshown that it is crucial to complement remotesensing programs with a net of observational keylocations over the continental shelves wheresystematic and repeatable surveys are executed formany years. Besides acquiring and validating remotesensing data, we have also to be concerned with theprocessing and distribution of these data andproducts to the scientific community. Results of oursimple comparison among Chl, FLH and CDOMindexfields are encouraging, but the interpretationof biological and physical mechanisms associatedwith spatial distributions of these products must beimproved. There is also a need to incorporateecological modeling as part of ocean color datainterpretation, and the ocean color groups mustinvest in this line, or better the different groupsshould work together.Barre, N., Provost, C. & Saraceno, M. 2006. Spatialand temporal scales of the Brazil-MalvinasCurrent confluence documented bysimultaneous MODIS Aqua 1.1-km resolutionSST and color images. Natural Hazards andOceanographic Processes from SatelliteData, 37: 770-786.Belkin, I. M. 2009. Rapid warming of Large MarineEcosystems. Progress in Oceanography, 81:207-213.Bentz, C. M., Lorenzzetti, J. A. & M. Kampel. 2004.Multi-sensor synergistic analysis of mesoscaleoceanic features: Campos Basin, south-easternBrazil. International Journal of RemoteSensing, 25: 4835-4841.Brandini, F. P., Lopes, R. M., Gutseit, K. S., Spach,H. L. & Sassi, R. 1997. Planctolologia naplataforma continental do Brasil. Diagnose erevisão bibliográfica. Rio de Janeiro, MMA-CIRM-FEMAR, 196 p.Calado, L., Gangopadhyay, A. & Silveira, I. C. A.2008. Feature-oriented regional modeling andsimulations (FORMS) for the western SouthAtlantic: Southeastern Brazil region. OceanModelling, 25(1-2): 48-64.Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

250Á. M. CIOTTI ET ALLICampbell, J., Antoine, D., Armstrong, R., Arrigo,K., Balch, W., Barber, R., Behrenfeld, M.,Bidigare, R., Carr, M. E., Esaias, W.,Falkowski, P., Hoepffner, N., Klefer, D.,Lohrenz, S., Marra, J., Morel, A., Ryan, J.,Vedernikov, V., Waters, K. Yentsch, C. &Yoder., J. 2002. Comparison of algorithms forestimating ocean primary production fromsurface chlorophyll, temperature, andirradiance, Global Biogeochemical Cycles,16(3): 1-15, doi:10.1029/2001GB001444.Carder, K. L., Steward, R. G., Paul, J. H. & Vargo,G. A. 1986. Relationships between Chlorophylland Ocean Color Constituents as TheyAffect Remote-Sensing Reflectance Models.Limnology and Oceanography, 31: 403-413.Carr, M. E., Friedrichs, M. A. M., Schmeltz, M.,Aita, M. N., Antoine, D., Arrigo, K. R.,Asanuma, I., Aumont, O., Barber, R.,Behrenfeld, M., Bidigare, R., Buitenhuis, E.R, Campbell, J., Ciotti, A. M., Dierssen, H.,Dowell, M., Dunne, J., Esaias, W., Gentilli,B., Gregg, W., Groom, S., Hoepffner, N.,Ishizaka, J., Kameda, T., Le Quére, C.,Lorenz, S., Marra, J., Mélin, F., Moore, K.,Morel, A., Reddy, T. E., Ryan, J., Scardi, M.,Smyth, T., Turpie, K., Tilstone, G., Waters, K.& Yama-naka, Y. 2006. A comparison ofglobal estima-tes of marine primaryproduction from ocean color. Deep-SeaResearch Part I Topical Studies inOceanography, 53: 741-770.Castro, B. M. & Miranda, L. B. 1998. Physicaloceanography of the western Atlantic continentalshelf located between 4º N and 34º S –coastal segment (4W). The Sea, 11: 209-251.Castro, B. M., Brandini, F. P., Pires-Vanin, A. M. S.& Miranda, L. B. 2006. MultidisciplinaryOceanographic Processes On the WesternAtlantic Continental Shelf Between 4N and34S, p. 259-293. In: A. R. Robinson & K.Brink (Eds.). The Global Coastal Ocean.Interdisciplinary Regional Studies andSyntheses. The Sea. Harvard UniversityPress.Cherubin, L. M. & Richardson, P. L. 2007.Caribbean current variability and theinfluence of the Amazon and Orinocofreshwater plumes. Deep-Sea Research PartI-Oceano-graphic Research Papers, 54:1451-1473.Ciotti, A. M. & Bricaud, A. 2006. Retrievals of asize parameter for phytoplankton and spectrallight absorption by colored detrital matterfrom water-leaving radiances at SeaWiFSchannels in a continental shelf region offBrazil. Limnology and Oceanography-Methods, 4: 237-253.Ciotti, A. M., Cullen, J. J. & Lewis, M. R. 1999. Asemi-analytical model of the influence ofphytoplankton community structure on therelationship between light attenuation andocean color. Journal of GeophysicalResearch-Oceans, 104: 1559-1578.Ciotti, A. M., Gonzalez-Rodriguez, E., Andrade, L.,Paranhos, R. & Carvalho, W. F. Clorofila a,Medidas Bio-ópticas e ProdutividadePrimária. 2006. In: Valentin, J. L. (Ed.).Características hidrológicas da regiãocentral da Zona Econômica Exclusivabrasileira (Salvador, BA, ao Cabo de SãoTomé, RJ). Brasília: Ideal, p. 6172 (SérieDocumentos REVIZEE/SCORE Central).Ciotti, A. M., Odebrecht, C., Fillmann, G. & MöllerJr, O. O. 1995. Freshwater outflow andSubtropical Convergence influence onphytoplankton biomass on the southernBrazilian continental shelf. Continental ShelfResearch, 15: 1737-1756.Del Vecchio, R. & Subramaniam, A. 2004. Influenceof the Amazon River on the surface opticalproperties of the western tropical North AtlanticOcean. Journal of Geophysical Research,109, C11001, doi: 10.1029/2004JC0 02503.Dickey, T., Bates, N., Byrne, R. H., Chang, G.,Chavez, F. P., Feely, R. A., Hanson, A. K.,Karl, D. M., Manov, D., Moore, C., Sabine, C.L. & Wanninkhof, R. 2009. The Nopp O-Scope and Mosean Projects AdvancedSensing for Ocean Observing Systems.Oceanography, 22: 168-181.Dotori & Castro. 2009. The response of the SaoPaulo Continental Shelf, Brazil, to synopticwind. Ocean Dynamics, 59: 603-614.Emery, W. J. & R. E. Thomson. 2004. DataAnalysis Methods in Physical Oceanography,Second ed., Elsevier, Amsterdam,Holan-da, 638 p.Esaias, W. E., Abbott, M. R., Barton, I., Brown, O.B., Campbell, J. W., Carder, K. L., Clark, D.K., Evans, R. H., Hoge, F. E., Gordon, H. R.,Balch, W. M., Letelier, R. & Minnett, P.J.1998. An overview of MODIS capabilities forocean science observations. IeeeTransactions on Geoscience and RemoteSensing, 36: 1250-1265.Fratantoni, D. M. & Glickson, D. A. 2002. NorthBrazil current ring generation and evolutionobserved with SeaWiFS. Journal of PhysicalOceanography, 32: 1058-1074.Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration251Friedrichs, M. A. M., Carr, M. E., Barber, R. T.Scardi, M., Antolne, D., Armstrong, R. A.,Asasuma, I. Behrenfeld, M. J., Bultenhuls, E.T., Chal, F., Christian, J. R., Ciotti, A. M.,Doney, S. C., Dowell, S.C., Dunne, M.,Gentilli, B., Gregg, W., Hoepffner, N.,Ishizaka, J., Kameda, T., Lima, I., Marra,J.,Mélin, F., Moore, J.K., Morel, A.,O'Malley, R.T., O'Relly, J., Saba, V. S.,Schmeitz, M., Smyth, T. J., Tjiputra, J.,Waters, K., West-berry, T. K. & Winguth, A.2009. Assessing the uncertainties of modelestimates of primary productivity in thetropical Pacific Ocean. Journal of MarineSystems, 76: 113-133.Froidefond, J. M., Gardel, L., Guiral, D., Parra, M.& Ternon, J. F. 2002. Spectral remote sensingreflectances of coastal waters in FrenchGuiana under the Amazon influence. RemoteSensing of Environment, 80: 225-232.Gaeta, S. A. & Brandini, F. P. 2006. Produçãoprimária do fitoplâncton na região entre oCabo de São Tomé (RJ) e o Chuí (RS). In:Rossi-Wongtschowski, C. L. & Madureira. M.S. P. (Eds.). O Ambiente Oceanográfico daPlataforma Continental e do Talude naRegião Sudeste-Sul do Brasil. EDUSP, SãoPaulo, 219-264.Gaeta, S. A., Lorenzzetti, J. A., Miranda, L. B.,Ribeiro, S. M. S., Pompeu, M. & Araujo, C.1999. The Vitória eddy and its relation to thephytoplankton biomass and primaryproductivity during the austral fall of 1995l.Archive of Fishery and Marine Research,47(2/3): 253-270.Garcia, C. A. E. & Garcia, V. M. T. 2008.Variability of chlorophyll-a from ocean colorimages in the La Plata continental shelfregion. Continental Shelf Research, 28:1568-1578.Garcia, C. A. E., Garcia, V. M. T. & Mcclain, C. R.2005. Evaluation of SeaWiFS chlorophyllalgorithms in the Southwestern Atlantic andSouthern Oceans. Remote Sensing ofEnvironment, 95: 125-137.Garcia, C. A. E., Sarma, Y. V. B., Mata, M. M. &Garcia, V. M. T. 2004. Chlorophyll variabilityand eddies in the Brazil-Malvinas Confluenceregion. Deep-Sea Research Part Ii-TopicalStudies in Oceanography, 51: 159-172.Garcia, V. M. T., Signorini, S., Garcia, C. A. E. &Mcclain, C. R. 2006. Empirical and semianalyticalchlorophyll algorithms in thesouthwestern Atlantic coastal region (25-40degrees S and 60-45 degrees W). InternationalJournal of Remote sensing, 27: 1539-1562.Goldenberg, S. B., Landsea, C. W., Mestas-Nunez,A. M. & Gray, W. M. 2001. The recentincrease in Atlantic hurricane activity:Causes and implications. Science, 293: 474-479.Gonzalez-Silvera, A., Santamaria-del-Angel, E. &Millán-Núñez, R. 2006. Spatial and temporalvariability of the Brazil-Malvinas Confluenceand the La Plata Plume as seen by SeaWiFSand AVHRR imagery, Journal ofGeophysical Research, 111, C06010,doi:10.1029/2004JC002745.Gonzalez-Silvera, A., Santamaria-Del-Angela, E.,Garcia, V. M. T., Garcia, C. A. E., Millan-Nunez, R., & Muller-Karger, F. 2004.Biogeographical regions of the tropical andsubtropical Atlantic Ocean off SouthAmerica: classification based on pigment(CZCS) and chlorophyll-a (SeaWiFS)variability. Continental Shelf Research, 24:983-1000.Gower, J. F. R., Brown, L., & Borstad, G. A. 2004.Observation of chlorophyll fluorescence inwest coast waters of Canada using theMODIS satellite sensor. Canadian Journalof Remote Sensing, 30: 17-25.Guenther, M., Gonzalez-Rodriguez, E., Carvalho,W. F., Rezende, C. E., Mugrabe, G. & Valentin,J. L. 2008 Plankton trophic structure andparticulate organic carbon production during acoastal downwelling-upwelling cycle. MarineEcology Progress Series, 363: 109-119.Henson, S. A. & Thomas, A. C. 2007.Phytoplankton scales of variability in theCalifornia Current System: 1. Interannual andcross-shelf variability. Journal of GeophysicalResearch-Oceans, 112, C07017, doi:10.1029/2006JC004039Holt, J., Harle, J., Proctor, R., Michel, S., Ashworth,M., Batstone, C., Allen, I., Holmes, R.,Smyth, T., Haines, K., Bretherton, D. &Smith, G. 2009. Modelling the global coastalocean. Philosophical Transactions of theRoyal Society a-Mathematical Physical andEngineering Sciences, 367: 939-951.Houghton, R. A. 2007. Balancing the global carbonbudget. Annual Review of Earth andPlanetary Sciences, 35: 313-347.Hu, C. M., Montgomery, E. T., Schmitt, R. W. &Muller-Karger, F. E. 2004. The dispersal ofthe Amazon and Orinoco River water in thetropical Atlantic and Caribbean Sea:Observation from space and S-PALACEPan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

252Á. M. CIOTTI ET ALLIfloats. Deep-Sea Research Part Ii-TopicalStudies in Oceanography, 51: 1151-1171.Huot, Y., Brown, C. A. & Cullen, J. J. 2005. Newalgorithms for MODIS sun-induced chlorophyllfluorescence and a comparison withpresent data products. Limnology andOceanography-Methods, 3: 108-130.Huot, Y., Brown, C. A. & Cullen, J. J. 2007.Retrieval of phytoplankton biomass fromsimultaneous inversion of reflectance, thediffuse attenuation coefficient, and Suninducedfluorescence in coastal waters,Journal of Geophysical Research, 112,C06013, doi: 10.1029/2006JC003794.IPCC, 2007: Climate Change 2007: The PhysicalScience Basis. Contribution of WorkingGroup I to the Fourth Assessment Report ofthe Intergovernmental Panel on ClimateChange. Solomon, S., Qin, D., Manning, M.,Chen, Z., Marquis, M., Averyt, K. B., Tignor,M. & Miller, H. L. (Eds.). CambridgeUniversity Press, Cambridge, UnitedKingdom and New York, NY, USA.Johns, W. E., Lee, T. N., Schott, F. A., Zantopp, R.J. & Evans, R. H. 1990. The North BrazilCurrent Retroflection - Seasonal Structure andEddy Variability. Journal of GeophysicalResearch-Oceans, 95: 22103-22120.Johnson, K. S., Berelson, W. M., Boss, E. S., Chase,Z., Claustre, H., Emerson, S. R., Gruber, N.,Kortzinger, A., Perry, M. J. & Riser, S.C.2009. Observing Biogeochemical Cycles atGlobal Scales with Profiling Floats andGliders Prospects for a Global Array.Oceanography, 22: 216-225.Kahru, M., Marinone, S. G., Lluch-Cota, S. E.,Pares-Sierra, A. & Mitchell, B. G. 2004.Ocean-color variability in the Gulf ofCalifornia: scales from days to ENSO. Deep-Sea Research Part Ii-Topical Studies inOceanography, 51: 139-146.Kampel, M., Lorenzzetti, J. A., Bentz, C. M., Nunes,R. A., Paranhos, R., Rudorff, F. M. &Politano, A. T. 2009. Simultaneous Measurementsof Chlorophyll Concentration by Lidar,Fluorometry, above-Water Radiometry, andOcean Color MODIS Images in the SouthwesternAtlantic. Sensors, 9(1): 528-541.Le Quere, C., Raupach, M. R., Canadell, J. G.,Marland, G., Bopp, L. T., Ciais, P., Conway,T. J., Doney, S.C., Feely, R. A., Foster, P.,Friedlingstein, P., Gurney, K., Houghton, R.A., House, J. I., Huntingford, C., Levy, P. E.,Lomas, M. R., Majkut, J., Metzl, N., Ometto,J. P., Peters, G.P., Prentice, I. C., Randerson,J. T., Running, S. W., Sarmiento, J. L., Schuster,U., Sitch, S., Takahashi, T., Viovy, N.,van der Werf, G. R. & Woodward, F. I. 2009.Trends in the sources and sinks of carbondioxide. Nature Geoscience, 2: 831-836.Lee, Z. P., Carder, K. L., Mobley, C. D., Steward, R.G. & Patch, J. S. 1998. Hyperspectral remotesensing for shallow waters. I. Asemianalytical model. Applied Optics, 37:6329-6338.Longhurst, A. 1995. Seasonal cycles of pelagicproduction and consumption. Progress inOceanography, 36: 77-167.Longhurst, A. 2006. Ecological Geography of theSea. Academic Press, 2 nd edition.Longhurst, A., Sathyendranath, S., Platt T. &Caverhill, C. 1995. An Estimate of GlobalPrimary Production in the Ocean fromSatellite Radiometer Data. Journal ofPlankton Research, 17: 1245-1271.Maritorena, S., Siegel, D. A. & Peterson, A. R.2002. Optimization of a semianalytical oceancolor model for global-scale applications.Applied Optics, 41: 2705-2714.Mcclain, C. R. 2009. A Decade of Satellite OceanColor Observations. Annual Review ofMarine Science, 1: 19-42.Molleri, G. S. F., Novo, E. M. L. D. & Kampel, M.2010. Space-time variability of the AmazonRiver plume based on satellite ocean color.Continental Shelf Research, 30: 342-352.Morel, A. & Gentili, B. 2009. A simple band ratiotechnique to quantify the colored dissolvedand detrital organic material from ocean colorremotely sensed data. Remote Sensing ofEnvironment, 113: 998-1011.Mote, P. W. & Mantua, N. J. 2002. Coastalupwelling in a warmer future. GeophysicalResearch Letters, 29: 2138-2142,doi:10.1029/2002GL016086.O’Reilly, J. E., Maritorema, S., Mitchell, B. G.,Siegel, D. A., Carder, K. L., Garver, S. A.,Kahru, M. & McClain, C. 1998. Ocean colorchlorophyll algorithms for SeaWiFS. Journalof Geophysical Research-Oceans, 103:24937-24953.Pimenta, F. M., Campos, E. J. D., Miller J. L. &Piola, A. R. 2005. Numerical study of theriver discharge and wind effects over the PlataRiver plume intrusion the southeastern SouthAmerican Shelf, Brazilian Journal ofOceanography, 53(3-4): 129-146.Piola, A. R. & Romero, S. I. 2004. Analysis ofspace-time variability of the Plata RiverPlume. Gayana, 68(2): 482-486.Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Temporal and meridional variability of Satellite-estimates of surface chlorophyll concentration253Piola, A. R., Romero, S. I. & Zajaczkovski, U. 2008.Space-time variability of the Plata plumeinferred from ocean color. Continental ShelfResearch, 28: 1556-1567.Polovina, J. J., Howell, E. A. & Abecassis, M. 2008.Ocean’s least productive waters areexpanding. Geophysical Research Letters,35, L03618, doi: 10.1029/2007GL031745.Richardson, P. L., Hufford, G. E., Limeburner, R. &Brown, W. S. 1994. North Brazil CurrentRetroflection Eddies. Journal of GeophysicalResearch-Oceans, 99: 5081-5093.Roesler, C. S. & Perry, M. J. 1995. In-SituPhytoplankton Absorption, FluorescenceEmission, and Particulate BackscatteringSpec-tra Determined from Reflectance.Journal of Geophysical Research-Oceans,100: 13279-13294.Rysgaard, S., Nielsen, T. G. & Hansen, B. W. 1999.Seasonal variation in nutrients, pelagicprimary production and grazing in a high-Arctic coastal marine ecosystem, YoungSound, Northeast Greenland. Marine EcologyProgress Series, 179: 13-25.Saraceno, M., Provost, C. & Piola, A. R. 2005. Onthe relationship between satellite-retrievedsurface temperature fronts and chlorophyll ain the western South Atlantic, Journal ofGeophysical Research, 110: C11016, doi:10.1029/2004JC002736.Sarmiento, J. L., Slater, R., barber, R., Bopp, L.,Doney, S. C., Hirst, A. C., Kleypas, J.,Matear, R., Mikolajewicz, U., Monfray, P.,Soldatov, V., Spall, S. A. & Stouffer, R. 2004.Response of ocean ecosystems to climatewarming, Global Biogeochemical Cycles, 18,GB3003, doi: 10.1029/2003GB002134.Sathyendranath, S., Watts, L., Devred, E., Platt, T.,Caverhill, C. & Maass, H. 2004. Discriminationof diatoms from other phytoplanktonusing ocean-colour data. Marine EcologyProgress Series, 272: 59-68.Schallenberg, C., Lewis, M. R., Kelley, D. E. &Cullen, J. J. 2008. Inferred influence ofnutrient availability on the relationshipbetween Sun-induced chlorophyll fluorescenceand incident irradiance in the BeringSea, Journal of Geophysical Research, 113,C07046, doi: 10.1029/2007JC004355.Siegel, D. A., Maritorena, S., Nelson, N. B., Hansell,D. A. & Lorenzi-Kayser, M. 2002, Globaldistribution and dynamics of coloreddissolved and detrital organic materials,Journal of Geophysical Research,107(C12): 3228, doi: 10.1029/2001JC000965.Silva Jr., C. L., Kampel, M., Araújo, C. E. S. &Stech, J. L. Observação da penetração doramo costeiro da corrente das Malvinas nacosta Sul-Sudeste do Brasil à partir deimagens AVHRR. In: Simpósio Brasileiro deSensoriamento Remoto, 8, 1996, Salvador,Brasil. Anais. São José dos Campos: INPE,1996. p. 787-793. 1 CD-ROM.Silveira, I. C. A. Da, Lima, J. A. M., Schimdt, A. C.K., Ceccopieri, W., Sartori, A., Fransisco, C.P. F. & Fontes, R. F. C. 2008. Is the meandergrowth in the Brazil Current System offSoutheast Brazil due to baroclinic instability?Dynamics of Atmosphere and Oceans, 45(3-4): 187-207.Sumida, P. Y. G., Yoshinaga, M. Y., Ciotti, A. M. &Gaeta, S.A. 2005. Benthic response to upwellingevents off the SE Brazilian Coast. MarineEcology Progress Series, 291: 35-42.Received May 2010Accepted August 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 236-253

Regime shifts, trends and interannual variations of water level inMirim Lagoon, southern BrazilFERNANDO E. HIRATA 1 , OSMAR O. MÖLLER JÚNIOR 2 & MAURICIO M. MATA 21 School of Earth and Atmospheric Sciences, Georgia Institute of Technology, 311 Ferst Drive, 30332. Atlanta, GA,USA. E-mail: fernando.hirata@gatech.edu2 Instituto de Oceanografia, Universidade Federal do Rio Grande (FURG) – Campus Carreiros, Av. Itália, Km 8, s/n,CEP 96201-900. Rio Grande, Rio Grande do Sul, Brazil.Abstract. Long-term changes of water level in Mirim Lagoon, southern Brazil, are stronglyassociated with the El Niño-Southern Oscillation (ENSO). However, the relationship between thewater lever record and the Southern Oscillation Index changed during the last century. Twodifferent regimes are identified. The first regime (1912-58) is drier than the second (1958-2002).Shifts in variance and wavelet analyses suggest that the extreme floods of 1941 and the subsequentdrought of 1943-45 were very unusual climatic events. The following years, from 1945 to 1963,appear to be a transition towards a new climatic state characterized by more frequent El Niños. Apositive trend for the entire period (> 1 m) was detected and is significant. The trend is apparentlyrelated to the higher frequency of ENSO warm events in the Pacific Ocean during the second halfof the last century and references in literature report that this trend affects a wide region insubtropical South America. The record from Mirim Lagoon, spanning over 90 years, is a rare andvery important indicator of long-term climatic variations and should be maintained and monitoredin order to assess potential environmental changes.Keywords: El Niño-Southern Oscillation, hydrological cycle, climate changesResumo. Mudanças de regime, tendências e variações interanuais de nível na Lagoa Mirim,sul do Brasil. Mudanças de longo período no nível da Lagoa Mirim, sul do Brasil, são fortementeassociadas ao El Niño-Oscilação Sul. Entretanto, a relação entre nível registrado e o Índice deOscilação Sul mudou ao longo do século passado. Dois regimes diferentes são identificados. Oprimeiro regime (1912-58) é mais seco que o segundo (1958-2002). Mudanças na variância eanálise por ondeletas sugerem que a extrema inundação de 1941 e a seca subsequente de 1943-45foram eventos climáticos singulares. Os anos seguintes, entre 1945-63, parecem ser uma transiçãopara um novo estado climático, caracterizado pela maior frequência de El Niños. Uma tendênciapositiva maior que 1 m para todo o período foi detectada e é significante. A tendência parecerelacionada à maior freqüência de El Niños no Oceano Pacífico durante a segunda metade doséculo passado e referências na literatura indicam que essa tendência afeta uma vasta região daAmérica do Sul subtropical. O registro da Lagoa Mirim, cobrindo mais de 90 anos, é um raro eimportante indicador de variaçõesclimáticas de longo período e deve ser monitorada com afinalidade de avaliar potenciais mudanças ambientais.Palavras-chave: El Niño-Oscilação Sul, ciclo hidrológico, mudanças climáticasIntroductionMirim Lagoon (ML) catchment basin is partof Patos-Mirim hydrographic system, whichcomprises a portion of Rio Grande do Sul State(southern Brazil) and eastern Uruguay (Fig. 1). ThePatos-Mirim system drains an approximate area of200,000 km 2 , with Patos Lagoon alone drainingnearly 145,000 km 2 . This hydrographic systemexerts strong impacts on the adjacent costal areathrough the input of freshwater and nutrients (Ciottiet al. 1995). Its average discharge is 2,400 m³/s.However, during El Niño (EN) years, discharge mayrise above 12,000 m³/s (Möller et al. 2001) inducingdrastic changes in the regional ecosystem andcontinental shelf circulation and composition. ThisPan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

Climatic variations in Southern Brazil255excessive discharge is associated with largeinterannual variations of rainfall over the basin. Theaverage precipitation rate over the Patos-Mirimdrainage basin is 1,200 mm/year. Above averagerainfall in excess of 2,000 mm/year was observedduring El Niño events and the lowest average valueon record (near 800 mm/year) was observed duringthe 1988 La Niña (LN) (Beltrame & Tucci 1998).The mean surface area of ML is approximately3,749 km 2 (185 km long and 20 km wide). Itscatchment basin includes almost 55,000 km 2 (47% inBrazil, 53% in Uruguay) and ML is linked to thePatos Lagoon estuary through a 76-km narrownatural channel called São Gonçalo.The term “coastal lagoon” usually refers towater bodies along the coast with one or moreconnections with the ocean (Bird, 2008). Therefore,although ML is traditionally known as a coastallagoon, it resembles more closely an overflowinglake. In the past, Patos Lagoon brackish waterscould reach ML through the São Gonçalo channel,damaging inundated rice crops in the region. Toavoid salt penetration upstream, a subsurface damwas built across the channel in 1977 to block denserbrackish waters but still allowing surface fluxes andnavigation. The dam is 3.2 m high and is placed in across section where the mean depth is around 5 m.After the dam was built, ML waters remained freshall the time, with an estimated mean overflow intoPatos Lagoon of 700 m 3 /s (Machado 2007). Changesof outlet conditions (as the subsurface dam) shouldonly marginally influence mean overflowing lakelevels (Bengtsson & Malm 1997).In addition to rainfall-runoff relationship,the residence time of Patos-Mirim system is alsoinfluenced by synoptic scale atmosphericphenomena. With a narrow connection restrictingfresh-seawater exchanges, circulation within thesystem is driven by the combined effect of windsand runoff.Dynamically, the passage of a cold front andthe associated strong south-southwesterly windsdrive a water level set up in the northern region ofboth Mirim and Patos Lagoon (e.g. Möller et al.2001). This could favor ML discharge, but alsodrives shelf waters against the coast, pushingseawater into the estuary and balancing the pressuregradient. Thus, seawater inflow and the wide floodplains along São Gonçalo channel slow down MLrunoff. With the weakening of the southerlycomponent of the winds and the establishment ofnortheasterlies, seawater retreats and the southwardpointingpressure gradient inside the lagoons acts toequalize the water level throughout the system,driving Patos Lagoon waters towards its estuarinearea. This effect, together with the southwardmovement of ML waters, partially dams ML again,increasing its residence time. Northeasterly windsare predominant during the whole year withincreasing importance of southwesterlies duringwintertime, associated with atmospheric cold frontspropagating over the region on time scales rangingfrom 3 to 11 days (Stech & Lorenzetti 1992).In general, Southeastern South America(SSA - Southern Brazil, Northeastern Argentina andUruguay) experiences positive rainfall anomaliesduring EN events and negative anomalies during LNevents (Grimm et al. 1998, 2000). This associationalso holds for streamflow anomalies, which presentinterannual cycles of 3.5 and 6.3 years, coherentwith El Niño – Southern Oscillation (ENSO) cycles(Robertson & Mechoso 1998).The relationship between ENSO and riverrunoff in Negro and Uruguay Rivers were exploredby Mechoso & Perez-Irribaren (1992). They found atendency for below average streamflow from Junethrough December during LN events and a slightlytendency for above average streamflow fromNovember through February during EN years.Figure 1. Map showing the La Plata River basin with(1 - Paraguay River, 2 - Paraná River, 3 - Uruguay Riverand 4 - Negro River) and Patos-Mirim basin. The detailshows the locations where the water level is observed (a -Santa Isabel and b - Santa Vitória do Palmar). The thickblack line represents the São Gonçalo channel.Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

256FERNANDO E. HIRATA ET ALLIThe study of Robertson & Mechoso (1998)also found a near-decadal cycle (approximately 9years) most marked in the streamflows of Paranáand Paraguay Rivers. The authors pointed out thatthis cycle is associated with sea surface temperature(SST) anomalies over the tropical North AtlanticOcean, mostly significant in austral summer. Anapparent co-variability of the 9-year cycle and SSTanomalies south of Greenland, also suggested arelationship of this cycle with the North AtlanticOscillation (NAO). Negative SST anomalies wouldbe associated with enhanced Paraná and ParaguayRivers streamflow. They also suggested a relation tothe decadal variability of the summer monsoonsystem and the southward moisture flux associatedwith the low-level jet along the eastern flank of theAndes Mountains. This impact would regulate arainfall see-saw between the region influenced bythe South Atlantic Convergence Zone and thesubtropical plains of SSA. The see-saw pattern isdescribed by Nogués-Paegle & Mo (1997).Numerical simulations presented by Robertson et al.(2000) showed that NAO interannual fluctuationsare highly correlated with SST anomalies in thetropical and subtropical South Atlantic Ocean,accompanied by variations in the summer monsoonover South America.Rainfall is the major source of long-termvariability in the hydrological cycle over SSA oncevariations in evaporation seem to be less important(Berbery & Barros 2002). Positive trends ofprecipitation were detected over the region andrelated to a change to more negative SouthernOscillation Index (SOI) conditions in the tropicalPacific Ocean since the 1960’s (Haylock et al.2006). Genta et al. (1998) focused attention on theexistence of long-term trends of streamflow in fourmajor rivers in the region (Uruguay, Negro, Paranáand Paraguay). They reported a general increase instreamflow beginning in the mid-1960s consistentwith a decrease in the amplitude of the seasonalcycle. In the case of the Negro River, the positivetrend began almost 15 years earlier, just after theextreme drought of 1943-45. Examining SSTanomalies in eastern equatorial Pacific Ocean, theysuggested that an important component of theobserved increase in streamflow is associated withlarge-scale low-frequency variability of the globalclimate system and the long-term trend is alsopossibly associated to changes in the Amazonregion.Long-term behavior of Pacific Ocean SSTconditions is usually associated with interdecadaloscillations (Mantua et al. 1997, Zhang et al. 1997).Abrupt changes from one phase of this interdecadalcycle to another are commonly referred to as regimeshifts. In 1977, the leading principal component ofSST anomalies of the North Pacific changed frommostly negative to mostly positive values (Mantua etal. 1997). Since then, there is uprising ambiguity inthe use of the term “regime shift” (Overland et al.2008). According to these authors, confusion arisesas a consequence of: (1) the shortness of climaticdatasets, (2) the lack of evidence on the existence ofdifferent stable modes of the climate system (eachstable mode would characterize one regime), and (3)the different usages of the term “regime” amidclimate scientists. Therefore, a clear definition of theterm is necessary.Here, we follow the “displacement”viewpoint of regime shifts and use the algorithmdesigned by Rodionov (2004) to identify them. Thedisplacement concept is solely based on time seriesanalysis of relatively short records (

Climatic variations in Southern Brazil257of the Federal University of Pelotas (Rio Grande doSul, Brazil). The monthly mean is calculated fromdaily records measured at two locations (SantaVitória do Palmar and Santa Isabel, see Fig. 1 forreference). The long-term mean was removed beforethe application of any statistical procedure and fromnow on we will refer to the ML water levelanomalies simply as water level unless otherwisestated.Statistical analyses described below werecarried out in 3 steps. First, the monthly time serieswas transformed into a mean annual water levelseries in order to reduce the effects of serialcorrelation prior to the regime shift analysis. Second,the monthly time series was used to test theexistence and significance of a long-term trend. Last,the monthly values were used again to explore thetemporal variability of the record through waveletanalysis.High serial correlation impacts the rejectionrate of the null hypothesis of no regime shift. Thus,monthly means were used to construct a series ofmean annual water level. This procedure reduced thelag-1 serial correlation from more than 0.9 to nearly0.30 (Fig. 3). Then, a prewhitening procedure wasperformed using a lag-1 autoregressive approach(von Storch 1995) corrected by the InverseProportionality with 4 corrections (IP4) method(Rodionov 2006). Rodionov (2006) demonstratedthat this method keeps the rejection rate of the nullhypothesis close to the target significance level (0.1)for series with autocorrelations as high as 0.6 priorto prewhitening. The author considers the analysiswith a prewhitening procedure a more conservativemethod because it increases the chance of missing atrue regime shift but, if a shift is detected, itssignificance can be accurately estimated.The existence of regime shifts of mean andvariance were tested using a sequential dataprocessing technique based on the Student’s t test(Rodionov 2004). From the number of independentobservations, the Student's test determines thedifference in mean necessary for a significant shiftof regimes to occur. Considering a time series, foreach new observation, the algorithm tests the nullhypothesis of the existence of a regime shift usingthe cumulative sum of normalized anomalies. Theparameters of the analysis were the same as thoseused by Rodionov (2006) in his analysis of theannual Pacific Decadal Oscillation index (cutofflength = 15, Huber weight parameter = 1 and targetsignificance level = 0.1). The test for a shift invariance was applied over the difference between theoriginal mean annual time series and theprewhitened time series using the same cutoff lengthand significance level.The algorithm is very sensitive to the choiceof these parameters. The cutoff length determinesthe minimum length of a possible regime. As we aredealing with annual averages, a cutoff length of 15indicates that a regime, if identified, will have 15 ormore years. The choice of 15 years was made inorder to avoid the selection of near-decadal cycles asregimes. The Huber weight parameter is a way ofreducing the impact of outliers, allowing an evenmore conservative analysis. The value of Huber’sparameter indicates the value above which anobservation is considered an outlier (number ofstandard deviations). Then, any outlier is weightedinversely proportional to their distance from themean of the regime.In overflowing lakes as ML, water levelobservations only allow the conclusion that theclimate has been excessively humid (if it is the case)and even drastic climatic changes have minorinfluences on lake levels and small amplitudeinterannual variations (Bengtsson & Malm 1997).Still, the circulation and dynamics of the Patos-Mirim system, as discussed in the introduction,induces a higher autocorrelation value with longerlags. This means that ML water level behavior maynot resemble that of a pure overflowing lake andmay show a clearer impact of climatic forcings.Therefore, our choice of cutoff length and Huber’sparameter intend to select regimes longer than 15years and avoid an excessive flattening of extremeevents due to the possible smallness of long perioddifferences in the record.The trend and its significance were testedusing monthly mean values. A linear regression in aleast-squares sense was used to estimate the trendand its significance was assessed by a Monte Carlotechnique to manage the impact of serial correlation(Livezey & Chen 1983).Periodicities were explored using waveletanalysis (Torrence & Compo 1998). Waveletanalysis was carried out using a Morlet wavelet(wavenumber 6) with an initial scale of 6 months.The Morlet wavelet was chosen because it isnonorthogonal (better suited for time series analysiswith expected continuous variations in waveletamplitude) and complex (better adapted to captureoscillatory behavior). Moreover, arbitrary choices ofdifferent nonorthogonal and complex wavelets donot qualitatively change the results (Torrence &Compo 1998). The variation rate of scales was set to0.25 corresponding to approximately 271 scales. Themaximum scale is 22 years and the series were zeropadded before convolutions.The ML water level time series was thenPan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

258FERNANDO E. HIRATA ET ALLIcompared with monthly times series of the SOI andthe NAO station based index through cross andcoherence wavelet analysis (Grinsted et al. 2003).Relative phase relationship in the plots is shown byarrows, where in-phase behavior (no lag covariation)is denoted by arrows pointing to the rightand anti-phase relationship, by arrows pointing tothe left. When ML water level leads by 90º, thearrows point straight down. The time lag betweenthe two signals can be estimated by the phaserelationship:time lag = [ (Φ × π/180) × λ] / 2;Where Φ is the angle of the arrow and λ isthe wavelength (in this case, the periodcorrespondent to a given frequency band).The SOI time series used here is thestandardized sea level pressure difference betweenDarwin (Australia) and Tahiti. It was obtainedonline from the Australian Bureau of Meteorology(BOM) Internet site (http://migre.me/3wXkX). Themonthly NAO station based index is provided byJim Hurrell’s webpage (http://migre.me/3wXtE) atthe Climate Analysis Section of the National Centerfor Atmospheric Research (CAS, NCAR). The indexis the difference of normalized sea level pressuresbetween Ponta Delgada (Azores, Portugal) andStykkisholmur/Reykjavik (Iceland). The probabilitydensity function of the NAO index time series ishighly bimodal (two discrete peaks) and, assuggested by Grinsted et al. (2003), it wastransformed to percentiles in order to enhance theresults of the analysis.ResultsBefore removing the mean of the series, it ispossible to observe that the maximum water level of4.8 meters occurred near the end of the record whilethe minimum value is reached by the beginning of1943 (anomalies in Fig. 2a). The 1943-45 period,recognized as the worst drought of the last century inSSA by Genta et al. (1998), is also evident in theML water level time series.The hydraulic behavior of ML induced by itsmorphology and dynamics is apparent in the timeseries. Figure 3 shows lagged autocorrelations forthe water level time series with and without theseasonal cycle. The high values and slow decay ofthe autocorrelation coefficient toward higher lagsindicate that the ML water level has strongdependence from month to month. This characterristicsupports the idea that the synoptic scaledynamics that was suppose to drive water exchangebetween ML and Patos Lagoon estuary is notcapable to overcome the long memory of the timeseries variability.Rodionov (2004)’s sequential method todetect regime shifts in the annual mean water levelseries revealed only one shift, in 1958. Before 1958,the annual mean water level was approximately -0.28 m (-0.34 m considering Huber's weightedmean). After the shift, the mean water level jumpedto 0.27 m (weighted mean of 0.20 m). Theconfidence level (CF) of the difference between themeans before and after the shift (tested by aStudent's two-tailed test with unequal variance) is6.54 x 10 -4 . This very small value indicates that it isvery unlikely that this shift had occurred by chanceor as a result of red noise. The existence of twodifferent regimes agrees with the hypothesis ofGenta et al. (1998) for their streamflow series, in thesense that the later period has a higher mean waterlevel than the first one. No signal of shift wasdetected around 1977, when the subsurface dam wasbuilt in the São Gonçalo channel.The test for a shift in variance resulted in 5different regimes. The first and last shifts in variance(1919 and 2001) are too close to the record ends,being smaller than the cutoff frequency of 15 years,and should be viewed with caution. From 1912 to1918 the estimated variance was 0.84. The shift of1919 has a CF of 5.55 x 10 -4 , suggesting that it hadnot occurred by chance. The following period, from1919 to 1946, had a smaller variance of 0.10 and, in1947, a second regime shift was detected (CF = 5.33x 10 -5 ). The variance jumped to 1.10 for 15 years. In1963, the variance decreased to 0.51 (CF = 6.29 x10 -4 ) and remained in this level for 38 years, until2001, when the last shift is detected (CF = 0.57).The CF in the last shift is poor and theregime is much shorter than the cutoff length. Thisshift should be taken as "on test" and may beconfirmed by new observations. Figure 2b highlightsthe regime shifts described above. The varianceregimes will be later compared with the waveletanalysis results.After the statistical detection of two regimeswith significantly different mean annual water level,the monthly time series is used to detect a differencein the annual cycle between the two periods.Estimating the annual cycle from the mean valuesfor each month of the year results in a shift to higherwater levels during the entire year, accompanied bya longer season with positive values (Fig. 4). In thiscase, the amplitude of the annual cycle is smaller forthe second period (1.30 m from 1912 to 1957 against1.21 m from 1958 to 2002), consistent with theresults of Genta et al. (1998), who used the samemethod to present differences in the cycle. If thePan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

Climatic variations in Southern Brazil259Figure 2. a) Mirim Lagoon monthly water level time series and the linear trend of anomalies (1.06 m from 1912 to2002, significant at 99% - thick dashed line). b) The average annual water level anomalies and regimes of mean andvariance identified by Rodionov's sequential method. c) The Southern Oscillation Index (SOI) time series. d) The NorthAtlantic Oscillation station index (NAO).annual cycle is estimated by calculating itsamplitude for all years and averaging the respectivevalues, the scenario changes. The amplitude ofthe annual cycle during the first period becomessmaller than that of the second regime (1.67 m,with a 95% confidence interval between 1.46 and1.88 m, against 1.88 m with a confidence intervalbetween 1.70 and 2.07 m). A Mann-Whitney U-testindicates that this change in amplitude is notsignificant at 0.05 significance level, although thesame test applied to the two regimes identifiedby Rodionov's algorithm reveals that theirprobability distribution function has changed signifycantlywith the shift in 1957-58. Thus, it is still notclear if the water level annual cycle has changedover the years.The estimation of a linear trend using thewater level anomalies resulted in a positive increaseof 11.9 mm/year (1.0669 m from 1912 to 2002),significant at 99% (Fig. 2a). This linear increase wasnot corrected to take into account vertical motions ofthe terrain and might be affected by isostaticadjustment or tectonic subsidence. The closest tidegauge station with a near-centennial record (83years) is in Buenos Aires (Argentina). Raicich(2008) estimated the sea level rise recorded inBuenos Aires from 1905 to 1987 to be around 1.57mm/year with peaks associated with high freshwaterdischarges of the La Plata River during EN events.The author also indicated that the United StatesGeological Survey consider the area betweenBuenos Aires and Southern Brazil as a region of lowseismicity. Jelgersma (1996) argued that subsidencein coastal sedimentary lowlands is slow (a fewcentimeters/century) and consists of basementsubsidence enhanced by subsidence due to isostaticloading. Therefore, even if the increase in ML waterlevel is influenced by both sea level rise and landsubsidence, their magnitude would be at least oneorder smaller than the observed trend.Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

260FERNANDO E. HIRATA ET ALLIUsing wavelet analysis, it is possible toverify that main periodicities are concentrated in twofrequency bands (Fig. 5). There is a broad bandcentered at 4 years, in agreement with the ENSOperiodic band ranging from 2-7 years and commonlyfound in environmental records worldwide (e.g.Allan, 2000). These periods are also related to largescale atmospheric circulation anomalies that reachSouth America during ENSO events as described byGrimm et al. (1998). This relationship is confirmedby cross wavelet between the water level record andthe SOI time series (Fig. 6a). Grinsted et al. (2003)pointed out that cross wavelet analysis showsregions in the time-frequency domain where twotime series present high spectrum power. If twoseries are physically related, it is expected aconsistent or a slowly varying phase lag. The arrowsFigure 3. Serial correlations coefficients for MirimLagoon water level anomalies (demeaned), for theanomalies with the annual cycle removed (deseasoned)and for the annual mean water level anomaly.Figure 4. Annual cycles for the two regimes withsignificant different annual mean water level. Vertical barindicate the standard deviation for each month.inside the significant area of Figure 6a suggest ananti-phase relationship in a frequency band rangingfrom 4 to 5 years (48 to 60 months) and localized intime from the late 1930s to the early 1980s. Theanti-phase relationship indicates that when MLwater level is positive (negative), SOI is negative(positive). This is expected since negative values ofSOI are indicative of El Niño events. The waveletcoherence presented in Figure 6b indicates that bothseries co-vary in a frequency band ranging fromnearly 3 to 6 years around 1940 and after 1990,predominantly showing the same anti-phasebehavior inside the 0.05 significance regions.A near-decadal cycle even more energeticthan the ENSO-associated cycles was detected in theanalysis. This periodicity was already identified byRobertson & Mechoso (1998) for streamflow seriesin SSA and was most marked in Paraná andParaguay rivers. Cross wavelet analysis betweenSOI and ML water level do not show significantpeaks near the 10-year frequency band, but thewavelet coherence presents a wide significant regionnear the 6-year band. As arrows inside the 0.05significance level region at this frequency bandindicate that SOI leads ML water level byapproximately 2 years.Robertson & Mechoso (1998) suggested anassociation between SSA streamflow variability andthe NAO. Hurrel et al. (2003) reported that the NAOhas a signal with a period around 8-10 years. WhenML water level time series is compared with thetransformed NAO index time series, there is a smallregion in the time-frequency domain, near the 6-yearband and localized in time around 1960, where crosswavelet indicates common energy concentration(Fig. 6c). Approximately the same significant regionis detected in wavelet coherence presented in Figure6d, with both time series varying in phase duringthis period.DiscussionStatistical analysis of the ML water leveltime series allowed the detection of two differentregimes of mean water level and a significantpositive trend. The first regime, with negative meanrelative to the record, ranges from 1912 to 1957 andthe second, positive relative to the record, spansfrom 1958 to 2002. The study of Genta et al. (1998)showed a significant probability of larger medianstreamflow for Negro and Uruguay Rivers between1970 and 1995 than the median before 1940. Theyalso detected a consistent decrease in the annualcycle in the later period for the Uruguay Riverstreamflow. The 30-year running mean for NegroRiver streamflow (a basin located just next to MLPan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

Climatic variations in Southern Brazil261basin) increased monotonically since the late 1940’s.The same increasing trend was detected by theauthor for another neighbor basin, the UruguayRiver, but only after the mid 1960’s.In ML, the first regime (1912-1957)embraces two major droughts (1917 and 1943-45).Both events were also observed by Mechoso &Perez-Irribaren (1992), while Genta et al. (1998)described the 1943-45 drought period as“extraordinarily strong anomalous climate events”.The one in 1917 coincides with a LN year and itis possible to observe that both the ML waterlevel wavelet transform (Fig. 5) and the crosswavelet between the water level series and SOI(Fig. 6a) show significant energy around the 4-yearband at that time. The later drought is consideredthe longest dry event of the last century and wasnot related to a LN event. However, Figure 5indicates that there is significant power concentratedin a wide frequency band (corresponding to periodsbetween 4 and 12 years). Cross wavelets fromFigures 6a and 6c suggest a significant relationshipbetween ML water level and both SOI and NAO,but on shorter timescales (3 to 8 years). Thereforewe conclude that neither ENSO nor NAO are ableto completely explain those extreme drier conditionsover SSA. However, global-scale tropospheric andstratospheric circulation anomalies during theearly 1940’s may be a result of a particular state ofthe climate system associated with the strong1941-42 EN (Brönnimann 2005) and the longlastingdrought may be related to pre-existentconditions set up by that unusually extreme warmENSO event.The existence of two regimes withsignificantly different means agrees with Genta et al.(1998) although a change in the water level annualcycle is not clear. The regime shift in mean isapparently associated to a shift of the ENSO-SSAteleconnection towards higher spectrum frequencies.From 1912 to 1957, when the mean regime wasnegative, both cross wavelet and wavelet coherencebetween ML water level and SOI exhibit a highcommon energy and co-variation in a frequencyband centered between 3 and 8 years. During thesecond regime, when the mean anomaly jumped to apositive value, this frequency band of significantassociation is concentrated between 1 and 5 years.This supports the idea of Haylock et al. (2006)that wetter conditions in SSA are associatedwith a higher frequency of ENSO warm events.Figure 6c also suggests that even the weakassociation between ML water level and NAOobserved during the first regime is not observed inthe second period.The shift in the ML mean water level is hardto be addressed, but large-scale conditions of theclimate system may help to assess its causes. Longtermchanges in ENSO are partly described as aninterdecadal oscillation of SST anomalies over thePacific Ocean, with a well expressed El Niño-likeregime prevailing from mid-1920s to 1942-43 andagain since 1976-77 (Zhang et al. 1997). Thisinterdecadal oscillation is also connected to the mostpowerful principal component analysis mode ofstreamflow variability calculated using North andFigure 5. Wavelet transform (left) and global wavelet spectrum (right) of ML water level record. The white line in thewavelet transform plot represents the cone-of-influence and black contours represent the 5% significance level againstred noise. The red dashed line in the global spectrum represents a 95% confidence level.Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

262FERNANDO E. HIRATA ET ALLIFigure 6. a) Cross wavelet analysis between ML water level and SOI time series (upper left). b) Wavelet coherencebetween ML water level and SOI time series (upper right). c) Cross wavelet analysis between ML water level and NAOindex time series (lower left). d) Wavelet coherence between ML water level and NAO index time series (lower right).The 95% significance against red noise is shown as a thick contour. Relative phase relationship is shown by arrows (inphasepointing right, anti-phase pointing left and ML water level leading by 90º pointing straight down).South American rivers together (Dettinger et al.2000). Zhang et al. (1997) also pointed out that theinterdecadal pattern of the Pacific Ocean presented achange around 1957-58. According to them, thischange was analogous to the one in 1976-77 but hasreceived less attention because its subsequent warmphase was shorter. The mid-1950's shift of ML waterlevel regime may be related to these basin-widechanges in the Pacific described by Zhang et al.(1997).Shifts in variance detected in the previoussection support the idea of a regime from theearly 1920’s until the late 1930’s. The waveletcoherence in Figure 6d suggests a co-variation ofthe water level and SOI from the 1930’s to thelate 1970’s. Allan (2000) reported more robustENSO periods in the 1910’s, 1950’s, 1970’s and1980’s and less energetic periods between 1920’sand 1940’s and during the 1960’s. Again, shifts invariance seem to agree with this argument untilthe 1960’s. Then, the variance decreases in the1963 shift and remains leveled off until the end ofthe record. This discrepancy may also be relatedto the positive shift in the mean that occurred in1958.Zhang et al. (1997) argued that theinterdecadal behavior of Pacific anomalies weredifferent in 1942-43 when compared to thevariations that took place in 1957-58 or 1976-77.The high-variance water level regime that began justafter the great drought of 1943-45 and lasted until1963 may reflect a transition period towards a newbasic climatic state (considering the observed lowvariance of the two major regimes of ML waterlevel). Although the nature of the early 1940’sclimate anomalies is not known, their strong signalin ML water level record and the Pacific anomaliesreported by Zhang et al. (1997) suggest a worldwideclimatic disturbance as proposed by Brönnimann(2005).Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

Climatic variations in Southern Brazil263The near-decadal cycle revealed by thewavelet analysis consistently lags SOI by 2 years.The study of Robertson & Mechoso (1998) cited theassociation of streamflow with SSTa south ofGreenland, which would be a possible link to theNAO. Cross wavelet and wavelet coherence betweenML water level and the transformed NAO indexindicate a sharp region of possible association nearthe 8-year band around 1960. Hurrel et al. (2003)suggested that the 1960’s were characterized byanomalously high surface pressures and severewinters from Greenland across northern Europe,during a negative NAO regime. Although the resultsdescribed here do not reject the hypothesis ofassociation between the NAO and SSA surfaceclimate, the overall signal on ML basin is punctualand barely significant. This may result from the factthat NAO impacts climate mostly during northernwinter (austral summer) and this is the dry season insouthern Brazil. Decadal variations of the basic stateduring this season may not be as important as theworld-wide ENSO impact on the 7 year timescales.Whether the positive trend identified here isa consequence of natural oscillations of the climatesystem or a result of anthropogenic forcing is stillunclear. Recently, Church et al. (2008) suggestedthat volcanic activity may be related to a globalcooling of the upper ocean and the offset of theincreasing rate of sea-level rise between 1963 and1991. A combination of a robust ENSO periodduring the 1950’s and the subsequent period of highvolcanic activity might led to the regimeconfiguration of positive mean and low variancefrom 1958/1963 to the end of the water level recordanalyzed here.Considering human impact around MLbasin, surrounding regions in Uruguay have beensubjected to intensive cattle and sheep grazing sincethe early 1940’s (García-Rodríguez et al. 2002).Overbeck et al. (2007) estimated a 25% decrease innatural grassland area in southern Brazil since thelate 1970’s due to the expansion of agriculturalactivities (mainly grazing, irrigated rice crops andmore recently Eucalyptus sp. plantations). Marqueset al. (2004) reported that nearly 89% of Uruguayanirrigated rice is cultivated on ML basin. On the otherhand, Baldi & Paruelo (2008) used satellite data todemonstrate that ML region presented a small rate ofchange from grass to cropland when two periods inthe last 30 years were compared (1985-89 and 2002-04), suggesting that major changes may haveoccurred before 1985. Nonetheless, Gautreau (2010)showed evidences that forest loss on the Uruguayanbanks of ML could be explained by rice cropextension but improved conditions for forest growthin Uruguay during the last century could be aconsequence of increased rainfall. Medeanic et al.(2010) used algal palynomorphs from anothercoastal lagoon in southern Brazil to demonstrate atendency of increasingly humid conditions duringthe last century, with a marked anthropogenicimpact detected after the 1970’s. All thesereferences lead to the conclusion that human impactis important. However, the increase in water level isremarkably high, especially considering that waterdiverted from ML to irrigated rice productionnegatively impacts the water balance of the lagoon.The positive trend in ML is consistent withthe trends found by Genta et al. (1998) for fourmajor rivers in SSA. A positive trend in rainfall overa large region in SSA is presented by Haylock et al.(2006) at least between 1960 and 2000. In theAmazon basin, Marengo et al. (1998) found noevidences of changes in the 20 th century, whereasslow increases in rainfall were found fornortheastern Brazil. Collinschon et al. (2001)analyzed river flow and rainfall from 1900 to 1995on the Paraguay River basin. They detectedincreased river flow since 1970 associated withchanges in rainfall patterns (increased frequency ofrainfall events) and suggested that at least part of therunoff increase should be due to deforestation.Moreover, the authors showed that, in Africa, theCongo River flow presented the exactly oppositebehavior of Paraguay River throughout the lastcentury, indicating a possible large-scale climaticconnection between the continents.A modeling study by Cook et al. (2004)pointed out that the West African Monsoon, inboreal summer, generates a Walker-type circulationwith low-level convergence and wet conditions overAfrica and divergence and drier conditions overnortheastern South America and tropical Atlantic.Thus, a weakening of the African Monsoon wouldlead to higher precipitation over northeastern Brazil.In fact, the African Monsoon presented a reductionin precipitation during the second half of the lastcentury and numerical modeling experimentsindicate Atlantic SST variability as the main driverof the observed rainfall decline (Paeth & Hense2004). Janicot et al. (1998) showed that divergentanomalies over the tropical Atlantic, associated withEl Niño events, may lead to a weaker AfricanMonsoon especially if there are positive SSTanomalies over the eastern tropical Atlantic as well.Moron et al. (1995) showed evidences of stronger ElNiño impact over the West African Monsoon after1970. On the other hand, Diaz et al. (1998)demonstrated that the Atlantic Ocean may impactrainfall over southern Brazil and Uruguay and thatPan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

264FERNANDO E. HIRATA ET ALLIthis impact may be independent of ENSO. In theirstudy, unusually high precipitation was observed in1959 (a neutral ENSO year) and associated withSST anomalies in the Atlantic basin. It is also worthto note that all trends described by these studies(Genta et al. 1998, Marengo et al. 1998, Collinschonet al. 2001, Paeth & Hense, 2004 and Haylock et al.2006) identify an increasing (decreasing) trend ofprecipitation or river flow in South America (Africa)starting around 1960. In summary, the tendency ofmore frequent warm ENSO events and a possibleinfluence of long-term variations of the Atlanticbasin may be responsible for these changes inrainfall. Last, the modeling experiments of Paeth &Hense (2004) suggested that increasingconcentration of greenhouse gases may lead towarmer SST conditions in the Atlantic and astronger African Monsoon. If the argumentspresented here are correct, the strengthening ofthe African Monsoon would possibly lead to anothershift, now towards a drier regime in South America.Climatic variations or changes, as describedhere, impact a wide and important region of theSouth American continent, with many differentecosystems. Patos Lagoon, for instance, may suffer astrong limnification process of its estuarine area,influencing species distribution and abundance.Ecological changes on interannual timescales suchas those observed in shallow-water fish assemblage(Garcia et al. 2001, 2004) and algal palynomorphsReferencesAllan, R. J. 2000. ENSO and climatic variability inthe past 150 years. Pp 3-55. In: Diaz, H. F. &Markgraf, V. (Eds). El Niño MultiscaleVariability and Global and RegionalImpacts. Cambridge University Press,Cambridge, United Kingdom, 496 p.Baldi, G. & Paruelo, J. M. 2008. Land-use and landcover dynamics in South American temperategrasslands. Ecology and Society, 13(2): 6.Beltrame, L. F. S. & Tucci, C. E. M. 1998. Estudopara gerenciamento e avaliação dadisponibilidade hídrica da bacia da LagoaMirim. Instituto de Pesquisas HidráulicasTechnical Report, UFRGS, Porto Alegre,Brazil, 128 p.Bengstsson, L. & Malm, J. 1997. Using rainfallrunoffmodeling to interpret lake level data.Journal of Paleolimnology, 18: 235-248.Berbery, E. H. & Barros, V. R. 2002. TheHydrologic Cycle of the La Plata Basin inSouth America. Journal of Hydrometeorology,3: 630-645.Bird, E. C. F. 2008. Coastal Geomorphology: an(Medeanic et al. 2010) may become permanent withthe abundance of freshwater. Higher discharge of thePatos-Mirim system and the La Plata River wouldintroduce considerable modifications of temperature,salinity and nutrient loads in this heavily fisheryexploredcontinental shelf also affecting theecological settings of coastal waters (Paes & Moraes2007). Because different time series spanning such along period of time are not common in SSA, the MLwater level record may be used as an indicator forchanges in the regional hydrological cycle andenvironmental conditions. Its maintenance andoperational monitoring is vital to track regime shiftsand trends in the region in order to developenvironmental policies and to manage the ecosystemand the anthropogenic impact on the neighboringareas.AcknowledgementsF. E. Hirata was sponsored by Coordenaçãode Aperfeiçoamento de Pessoal de Nível Superior(CAPES). O. O. Möller Júnior and M. M. Mataacknowledge CNPq research grants 302812/2007-5and 301623/2006-6, respectively. This paper is acontribution from the South Atlantic ClimateChange (SACC) consortium that has been funded bythe Inter-American Institute for Global ChangeResearch (IAI) through grant CRN2076. The IAI issupported by the US National Science Foundationgrant GEO-0452325.Introduction. John Wiley & Sons, Chichester,England, 340 p.Brönnimann, S. 2005. The global climate anomaly1940-1942. Weather, 60(12): 336-342.Church, J. A., White, N. J., Aarup, T., Wilson, W.S., Woodworth, P. L., Domingues, C. M.,Hunter, J. R. & Lambeck, K. 2008.Understanding global sea levels: past, presentand future. Sustainability Science, 3: 9-22.Ciotti, A. M., Odebrecht, C., Fillmann, G. & Möller,O. O. 1995. Freshwater outflow andSubtropical Convergence influence onphytoplankton biomass on the southernBrazilian continental shelf. Continental ShelfResearch, 15: 1737-1756.Collinschon, W., Tucci, C. E. M. & Clarke, R. T.2001. Further evidences of changes in thehydrological regime of the River Paraguay:part of a wider phenomenon of climate change?Journal of Hydrology, 245: 218-238.Cook, K. H., Hsieh, J.-S. & Hagos, S. M. 2004. TheAfrica-South America Intercontinental Teleconnection.Journal of Climate, 17: 2851-Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

Climatic variations in Southern Brazil2652868.Dettinger, M. D., Cayan, D. R., McAbe, G. J. &Marengo, J. A. 2000. Multiscale streamflowvariability associated with El Niño/SouthernOscillation. In: Diaz, H. F. & Markgraf, V.(Eds). El Niño Multiscale Variability andGlobal and Regional Impacts. CambridgeUniversity Press, Cambridge, UnitedKingdom, 496 p.Diaz, A. F., Studzinski, C. D. & Mechoso, C. R.1998. Relationships between precipitationanomalies in Uruguay and Southern Braziland Sea Surface Temperatures in the Pacificand Atlantic Oceans. Journal of Climate, 11:252-273.Garcia, A., Vieira, J. P. & Winemiller, K. 2001.Dynamics of the shallow-water fishassemblage of the Patos Lagoon estuary(Brazil) during cold and warm ENSOepisodes. Journal of Fish Biology, 59: 1218-1238.Garcia, A., Vieira, J. P., Winemiller, K. & Grimm,A. M. 2004. Comparison of 1982-1983 and1997-1998 El Niño effects on the shallowwaterfish assemblage of Patos Lagoonestuary (Brazil). Estuaries and Coasts, 27:905-914.García-Rodríguez, F., Mazzeo, N., Sprechmann, P.,Metzeltin, D., Sosa, F., Treutler, H. C.,Renom, M., Scharf, B. & Gaucher, C. 2002.Paleolimnological assessment of humanimpacts in Lake Blanca, SE Uruguay. Journalof Paleolimnology, 28: 457-468.Gautreau, P. 2010. Rethinking the dynamics ofwoody vegetation in Uruguayan campos,1800-2000. Journal of Historical Geography,36: 194-204.Genta, J. L., Perez-Irribaren, G. & Mechoso, C. R.1998. A recent increasing trend in streamflowof rivers in Southeastern South America.Journal of Climate, 11: 2858-2862.Grinsted, A., Moore, J. C. & Jevrejeva, S. 2003.Application of the cross wavelet transformand wavelet coherence to geophysical timeseries. Nonlinear Processes in Geophysics,11: 561-566.Grimm, A. M., Ferraz, S. E. T. & Gomes, J. 1998.Precipitation anomalies in Southern Brazilassociated with El Niño and La Niña events.Journal of Climate, 11: 2863-2880.Grimm, A. M., Barros, V. R. & Doyle, M. E. 2000.Climate variability in Southern SouthAmerica associated with El Niño and La Niñaevents. Journal of Climate, 13: 35-58.Haylock, M. R., Peterson, T. C., Alves, L. M.,Ambrizzi, T., Anunciação, Y. M. T., Baez, J.,Barros, V. R., Berlato, M. A., Bidegain, M.,Coronel, G., Corradi, V., Garcia, V. J.,Grimm, A. M., Karoly, D., Marengo, J. A.,Marino, M. B., Moncunill, D. F., Nechet, D.,Quintana, J., Rebello, E., Rusticucci, M.,Santos, J. L., Trebejo I. & Vincent, L.A..2006. Trends in total and extreme SouthAmerican rainfall in 1960-2000 and links withsea surface temperatures. Journal of Climate,19: 1490-1512.Hurrel, J. W., Kushnir, Y., Ottersen, G. & Visbeck,M. 2003. An overview of the North AtlanticOscillation. Geophysical Monograph, 134:1-35.Janicot, S., Harzallah, A., Fontaine, B. & Moron, V.1998. West African Monsoon Dynamics andEastern Equatorial Atlantic and Pacific SSTAnomalies (1970–88). Journal of Climate,11: 1874-1882.Jelgersma, S. 1996. Land subsidence in coastallowlands. Pp 47-62. In: Milliman, J.D. & Haq,B.U. (Eds). Sea level rise and coastalsubsidence: Causes, consequences, andstrategies. Kluwer Academic Publishers,Dordrecht, Netherlands, 369 p.Livezey, R. E. & Chen, W. Y. 1983. Statisticalsignificance and its determination by MonteCarlo Techniques. Monthly WeatherReview, 111: 46-59.Machado, G. 2007. Demanda e disponibilidadehídrica no sistema Lagoa Mirim-São Gonçalo– Rio Grande do Sul. Revista DiscenteExpresões Geográficas, 3: 61-82.Mantua, N. J., Hare, S. R., Zhang, Y., Wallace, J. M.& Francis, R. C. 1997. A Pacific interdecadalclimate oscillation with impacts on salmonproduction. Bulletin of the AmericanMeteorological Society, 78: 1069-1079.Marengo, J. A., Tomasella, J. & Uvo, C. R. 1998.Trends in streamflow and rainfall in tropicalSouth America: Amazonia, eastern Brazil, andnorthwestern Peru. Journal of GeophysicalResearch, 103: 1775-1783.Marques, M., Knoppers, B., Lanna, A. E., Abdallah,P. R. & Polette, M. 2004. Brazil Current -GIWA Regional assessment 39. UnitedNations Environment Programme, Nairobi,Kenya, 192 p.Mechoso, C. R. & Perez-Irribaren, G. 1992.Streamflow in Southeastern South Americaand the Southern Oscillation. Journal ofClimate, 5: 1535-1539.Medeanic, S., Hirata, F. & Dillenburg, S. 2010.Algal palynomorphs response to environmen-Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

266FERNANDO E. HIRATA ET ALLItal changes in the Tramandaí Lagoon,Southern Brazil, and climatic oscillationsduring the 20th century. Journal of CoastalResearch, 26(4): 736-742.Möller, O. O., Casting, P., Salomon, J. C. & Lazure,P. 2001. The Influence of Local and Non-Local Forcing Effects on the Subtidal Circulationof Patos Lagoon. Estuaries, 24: 297-311.Moron, V., Bigot, S. & Roucou, P. 1995. Rainfallvariability in subequatorial America andAfrica and relationship with the main SSTmodes (1951-1990). International Journalof Climatology, 15: 1297-1322.Nogués-Paegle, J. & Mo, K. 1997. Alternating wetand dry conditions over South America duringsummer. Monthly Weather Review, 125:279-291.Overbeck, G. E., Müller, S. C., Fidelis, A., Pfadenhauer,J., Pillar, V. D., Blanco, C. C., Boldrini,I. I., Both, R. & Forneck, E. D. 2007.Brazil's neglected biome: The South BrazilianCampos. Perspectives in Plant Ecology,Evolution and Systematics, 9: 101-116.Overland, J., Rodionov, S., Minobe, S. & Bond, N.2008. North Pacific regime shifts: Definitions,issues and recent transitions. Progress inOceanography, 77: 92-102.Paes, E. T. & Moraes, L. E. S. 2007. A new hypothesison the impact of the El Niño/La Niñaupon the biological productivity, ecology andfisheries of the Southern Brazilian Bight.Pan-American Journal of Aquatic Sciences,2: 94-102.Paeth, H. & Hense, A. 2004. SST versus climatechange signals in West African rainfall: 20thcentury variations and future projections.Climatic Change, 65:179-208.Raicich, F. 2008. A review of sea level observationsand low frequency sea-level variability inSouth Atlantic. Physics and Chemistry ofthe Earth, 33: 239-249.Robertson, A. W. & Mechoso, C. R. 1998.Interannual and decadal cycles in river flowsof Southeastern South America. Journal ofClimate, 11: 2570-2581.Robertson, A. W., Mechoso, C. R. & Kim, Y-.J.2000. The influence of Atlantic sea surfacetemperature anomalies on the North AtlanticOscillation. Journal of Climate, 13: 122-138.Rodionov, S. N. 2004. A sequential algorithm fortesting climate regime shifts. GeophysicalResearch Letters, 31, L09204, doi: 10.1029/2004GL019448.Rodionov, S. N. 2006. Use of prewhitening inclimate regime shifts detection. GeophysicalResearch Letters, 33, L12707, doi:10.1029/2006GL025904.Stech, J. L. & Lorenzetti, J. A. 1992. The responseof the South Brazil Bight to the passage ofwintertime cold fronts. Journal ofGeophysical Research, 97(C6): 9507-9520.Torrence, C. & Compo, G. P. 1998. A practicalguide to wavelet analysis. Bulletin of the A-merican Meteorological Society, 79: 61-78.von Storch, H. 2005. Misuses of statistical analysisin climate research. Pp 11-26. In: von Storch,H. & A. Navarra (Eds). Analysis of ClimateVariability Applications of Statistical Techniques.Springer, New York, USA, 352 p.Zhang, Y., Wallace, J. M. & Batistti, D. S. 1997.ENSO-like interdecadal variability: 1900-93.Journal of Climate, 10: 1004-1020.Received December 2009Accepted August 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 254-266

Potential vulnerability to climate change of the beach-dune system ofthe Peró coastal plain - Cabo Frio, Rio de Janeiro state, BrazilDIETER MUEHE 1 , FÁBIO M. BELLIGOTTI 1 , FLÁVIA M. LINS-DE-BARROS 2 ,JULIO F. DE OLIVEIRA 3 & LUIZ F. P. G. MAIA 41 Programa de Pós-Graduação em Geografia, Instituto de Geociências, Universidade Federal do Rio de Janeiro(IGEO/UFRJ), Cidade Universitária - Ilha do Fundão. Caixa Postal 68537, Rio de Janeiro. CEP 21941-972E-mail: dieter.muehe@gmail.com.2 Instituto Multidisciplinar– Universidade Federal Rural do Rio de Janeiro. Rua. Governador Roberto Silveira s/n,Moquetá, Nova Iguaçu, Rio de Janeiro, RJ. CEP 26020-740.3 Programa de Pós-Graduação em Geociências - Universidade Federal do Rio Grande do Sul. Av. Bento Gonçalves9500, Prédio 43.113 S-207B – Agronomia, Porto Alegre, RS. Cx. P. 15.001. CEP 91509-900.4 Departmento de Meteorologia, Instituto de Geociências, Universidade Federal do Rio de Janeiro (IGEO/UFRJ),Cidade Universitária - Ilha do Fundão. Caixa Postal 68537, Rio de Janeiro. CEP 21941-972.Abstract. The Peró coastal plain is one of the three morphological compartments located between Búziosand Arraial do Cabo, on the east coast of Rio de Janeiro State, characterized by the occurrence of dunefields, driven by northeasterly winds, obliquely oriented in relation to the coastline. In order to evaluate thestability of the coastline as well as of the dune system a topographical and sedimentologicalcharacterrization of the different morphodynamic units was made, consisting of topographic andbathymetric profiling, evaluation of eolian sediment transport through sediment traps, sediment samplingfor grain size analysis and comparison of aerial photographs. The results indicate that during the period ofinvestigation sediment transport from the beach-foredune to the main dune field was largely inhibited bythe presence of vegetation, and that sediment transport was limited to the reworking of its own sedimentstock. Climatic oscillation toward drier periods, whether the result of global climate change or localvariation in the duration of the dry period, will probably trigger the eolian sediment transport due to thereduction of the vegetation cover resulting in the migration of the main dune field towards urban areas. Nocoastal erosion was observed, whether through comparison of aerial photographs of 1970 and 2000 orduring one year of monthly topographic profiling. Nevertheless, a rise in sea level might result in anerosive adjustment of the shoreline as well as in an increase in the flooding of the low lying areas.Key words: Eolian transport, dune stability, beach profile, shoreface profileResumo: Vulnerabilidade potencial às mudanças climáticas do sistema duna-praia da planície costeirado Peró - Cabo Frio, Rio de Janeiro, Brasil. A planície costeira do Peró é um dos três compartimentoscosteiros localizados entre Armação dos Búzios e Arraial do Cabo, no litoral leste do Rio de Janeiro,que se caracterizam pela presença de campos de dunas dispostas obliquamente à linha de costa porefeito do transporte induzido pelo vento nordeste. Para a avaliação da estabilidade da linha de costa e dosistema de dunas foi efetuada uma caracterização topográfica e sedimentológica das diversas unidadesmorfológicas. Os levantamentos consistiram de perfis topográficos e batimétricos, avaliação do transporteeólico por meio de armadilhas de areia durante doze meses consecutivos, coleta de sedimentos para análisegranulométrica e análise de fotos aéreas. Os resultados mostram que no momento atual o sistema dedunas se alimenta do seu próprio estoque sedimentar, sem aporte significativo do sistema praia-dunafrontal. Oscilações do clima para o mais seco, independentemente de mudanças globais, podem reativar otransporte eólico em função da redução do recobrimento vegetal reiniciando a migração das dunas emdireção a áreas urbanizadas. A linha de costa não apresentou tendência de erosão, no entanto uma elevaçãodo nível do mar implicará num ajuste erosivo através de retrogradação da mesma, e aumento dadificuldade de escoamento com resultante inundação das áreas baixas à retaguarda das dunas frontais.Palavras-chave: Transporte eólico, duna, perfil de praia, antepraiaPan-American Journal of Aquatic Sciences (2010) 5(2): 267-276

268D. MUEHE ET ALLIIntroductionThe Peró coastal plain constitutes, togetherwith Tucuns and Cabo Frio coastal plains, one of thethree geomorphological compartments that stretchfrom Cape Búzios, herein called Búzios, to Arraialdo Cabo, characterized by the presence of sandydunes, which lie obliquely to the coastline, driven bythe prevailing northeasterly wind (Fig. 1). As this isan area of environmental protection, it is stillrelatively well preserved thus being suitable for arepresentative case study of the coastal processesoperating in the region as related to its present andpotential vulnerability to climate change.Figure 1. The dune fields of Tucuns, Peró and Cabo Frio,between Búzios and Arraial do Cabo, with location of thestudy area (white rectangle).Three distinct geomorphologic unitscharacterize the study area (Fig. 2):i. The beach-foredune ridge system in front of adeflation plain;ii. The deflation plain with isolated parabolic dunes;iii. The active dune field which intercepts andoverlaps a paleo-lagoon, currently filled withsediments.The foredune ridge extends behind thebeach from which it receives the sediments. Itrepresents, when depleted of vegetation, a source ofsediment for the dune field, which is developingbehind it and forms a considerable barrier, with its 7m high dunes, to the erosion of the coast whosestability depends on its vegetation cover andsediment budget.The deflation plain is located between theforedune ridge and the active dune field and itpresents a few semi-active parabolic dunes.The active dune field consists of NE-SWtrending transgressive dunes and is located in thedistal sector of the sandy terrace. Its starting point islocated in the northern extremity of the beach and itextends obliquely to the coastline, partiallyintercepting a paleo-lagoon forming the westernlimit of the deflation plain. As the dune field movesaway from the coastline it becomes wider andencroaches on the lower slope of the coastal rangeassuming the characteristics of a climbing dune(Rangel & Castro 2005), while its western slip faceprecipitates towards an urban zone, having alreadyled to the abandonment of a residential estate(Rangel & Castro 2005). The mobility of this dunefield has been studied by Dourado & Silva (2005)who estimated a displacement of 125 m between1965 and 2001 and an average displacement rate of3.5 m/year, with great variation in the differentsectors of the dune field. Because the sediment inputfrom the foredune field, has been almost completelyinterrupted by the vegetation cover of the foredunesand of the deflation plain, the mobility of the dunefield results essentially from the reworking of itsown sediment stock. Sediment stability thereforedepends essentially on the rainfall regime for themaintenance of the vegetation cover and theconsequent inhibition of the wind transport as doesthe stability of the coastline in terms of coastalerosion.Local climate, according to Barbiére (1975,1984), is warm and semi-arid (BSh), with mean temperaturesbetween 25.2 ºC in February and 20.5 ºCin August; mean monthly rainfall of is 60 mm,except in December when it stays above 100 mm;mean monthly evaporation rate is 60 mm. The dryperiod, when the monthly precipitation drops belowFigure 2. Geomorphologic units of the Peró coastal plainwith the location of the foredune ridge, the deflation plainand the active dune field. The dotted line represents theinferred contours of a former lagoon partially interceptedby the active dune field. Image Google Earth.Pan-American Journal of Aquatic Sciences (2010) 5(2): 267-276

Peró coastal plain vulnerability26960 mm, occurs between June and September whilethe rainy season starts in December, with decreasingamounts of rain up to March. Winds are predominantlyfrom northeast during the whole year alternatingwith short periods of winds from southeast andsouthwest associated to the arrival of cold fronts.Wind velocities usually stay between 4 and 6m/s;higher values are observed during the winter.The present study seeks to identify presentdaymorphodynamic processes and the possibleeffects of climate change on them.MethodologyMonthly topo-bathymetric profiles weremeasured over twelve months in three positionsalong the beach arc and along the longitudinal axisof three parabolic dunes; the main goal was tocharacterize the topography of the foredunes and toassess the mobility of the beach and shoreface aswell as that of the parabolic dunes of the deflationplain (Fig. 3).Topo-bathymetric profiles were measured atpositions P2, P5 and P7 (Fig. 3), and adjusted to theaverage sea-level on the basis of the predicted tidesfor the Porto do Forno (Arraial do Cabo), followingthe methodology described by Muehe et al. (2003)and Muehe (2004). The profiles extended from theback of the foredune field (profiles P5 and P7) andfrom the public walk (profile P2), up to the outerlimit of the surf zone through conventionaltopographic leveling. The profiles were extended togreater depths, for the determination of the depth ofclosure and sediment sampling, with a kayak, aportable echo sounder a hand-held GPS, inaccordance with the methodology developed byBelligotti & Muehe (2007).In accordance with the studies undertakenby Belligotti (2009), the depth of inflection of thebathymetric profile was considered for thedetermination of the depth of closure, the depth atwhich vertical variations of topography arenegligible (Hallermeier 1981), as a more practicalalternative in view of the enormous discrepancybetween the values obtained with the differentequations proposed in the literature (Hallermeier1981, Birkemeier 1985, Houston 1995, Wang &Davis Jr 2007) which are, in their turn, greatlyinfluenced by the wave parameters adopted.The determination of the shorefaceequilibrium profile, for purposes of comparison withthe measured profile, was obtained with the use ofDean’s equation (1991) expressed by the relation:z x= Ax2 / 3where the depth “z” at a distance “x” from thecoastline is defined as a function of the mediandiameter of the sediments or of the correspondingsettling velocity of (ω s ), expressed by the scaleparameter “A”, determined in its turn by the relation:A =0.440,0067⋅ωsFor a rough assessment of a potential retreatof the coastline corresponding to a given rise in sealevelthe equation of Bruun (Bruun, 1962, 1988) wasemployed:SLGR =hwhere:R = erosive retreat of the coastline due to rise in sealevel(m)S = rise in sea-level (m)L = length of the active profile (m)h = height of the active profile (m)G = proportion of eroded material which ismaintained on the active profileThe height of the active profile isdetermined by the sum of the land height (sandbarrier, terrace, foredune) and the depth of closure,and the active length is the horizontal distancebetween the land height and h. The proportion ofmaterial retained on the profile (G) is considered tobe 1, because of the difficulty of assessing thequantity of material made available by erosion andretained on the submerged profile. This is certainlyone of the sources of error in the assessment of theamplitude of the retreat.Sediment traps were installed at a fewselected locations (Fig. 3) for the assessment ofsediment transport from the foredunes to thedeflation plain and were monitored during the periodof the surveys. The traps were constructed inaccordance with Leatherman (1978) and consisted ofa 100 cm long PVC tube whose upper half wasprovided with two openings on each side, one ofthem being closed with a silk-screen in order toretain the sediments but still allowing the passage ofthe wind. The lower part was buried and served assupport for an acetate cylinder in which the grainscollected by the trap were retained. The traps wereplaced with the opening in the direction of theprevailing wind, of about 60º, well defined insatellite images by the tracks of the trajectory of thesedimentary transport.Sediment samples were collected for grainsize analysis at each of the traps and along thebeachface (Fig. 3), where the sampling was carriedout at 500 m intervals on the top of the foredune, ofthe berm and on the beachface. For an adequaterepresentation of the berm sediments 30 cm deepPan-American Journal of Aquatic Sciences (2010), 5(2): 267-276

270D. MUEHE ET ALLItrenches were dug, and one side of the trench wasscraped to obtain compound representative samples.Thirteen sediment samples were obtained from thedune field and its associated deflation plain and afurther 27 samples were collected from the dunebeachsystem. Grain size analysis was made throughdry sieving according the method described by Folk& Ward (1957).The stability of the coastline was assessedby means of a comparison between aerialphotographs from 1970 with orthophotocharts from2000/2002.Climatic data for Arraial do Cabo/Cabo Friowere obtained at following sources: Corpo deBombeiros Militar do Estado do Rio de Janeiro(CBMRJ) for rain, Centro de Previsão do Tempo eEstudos Climáticos (CPTEC) for cold fronts, andRede de Meteorologia do Comando da Aeronáutica(REDEMET) for hourly wind direction and speed.ResultsSediment characterizationSamples collected on the beach (beach faceand berm) and on the top of the foredunes, along thebeach arc, indicate an increase in median grain sizediameter from the southern end toward the center ofthe beach arc, and a decrease towards the northernend. The median grain size is coarsest in the vicinityof the center of the beach arc (P5), and decreasestowards both extremities of the beach. This suggeststhat the region near the center of the beach is adispersal center for the finer sedimentary fractionswhich are displaced preferentially towards the north,possibly being the result of an increase in waveenergy under storm conditions (waves from the SE)due to the more exposed position of this segment.The localized increase in the grain size of the beachsediments is not so clearly reflected in the sedimentsof the foredunes because of the inability of the windto transport sediments greater than 0.2 mm (2.3 phi)(dashed horizontal line in Fig. 4).The finer sands that are more easilytransported by wind are found in the final third ofthe northern extremity of the beach arc (Profile P7),precisely where the beginning of the active dunefield is located.In figure 5 it is presented a correlationbetween the median diameter and the respectivestandard deviation of the sediment samples collectedon the berm, foredune and on the deflation plainwith associated parabolic dunes. The area delineatedon the graph by a broken ellipse indicates the grainsize characteristics of the foredunes whosesediments present a median diameter equal to or lessthan 0.2 mm (2.3 phi) and a standard deviation ofless than 0.7 mm (0.5 phi). The occurrence ofsediments of larger median grain size and higherstandard deviation in the dunes of the deflation plainsuggests the occurrence of residual depositsresulting from the winnowing of the finer fractionsor even a mixture with the sediments of the deflationsurface.Figure 4. Median grain size distribution along the beachand foredune field. The dashed horizontal line representsthe local grain size threshold for eolian sediment transportas depicted in figure 5 indicating a convergence in grainsize between beach and foredune sediments toward thenorthern sector (profile P7) of the beach.Figure 3. Location of the sediment traps (ο), of the topobathymetricprofiles (Pn) and of the beach-foredunesediment samples (Δ). Image Google Earth.Sediment trapsThrough out the monitoring period onlythree traps, 7, 11 and 14, registered a significantcapture of sand. The first two were located on thewindward side of parabolic dunes of the deflationPan-American Journal of Aquatic Sciences (2010) 5(2): 267-276

Peró coastal plain vulnerability271plain and the last on the flank of the active dunefield. At the other traps, sediment capture was nil orvery small. Therefore the few areas of significantwind transport are associated to places withoutvegetation cover at the inner or lateral flank of thedunes resulting in remobilization of their ownsediment stock, suggesting that there is nocontinuous transport from the foredunes to thedeflation plain. This becomes clearer when the trapsare grouped by location. Trap 7 presented significantsediment capture while at traps 5 and 4 located,respectively, outside the trailing ridge of theparabolic dune and at the deflation plain upwindfrom the dune, sediment trapping varied from nil tovery small. Trap 11 was also located on thewindward side of a reactivated parabolic dune. Trap14 was located on the flank of the active dune fieldwhere sediment transport is active because of acomplete absence of vegetation. Trap 12 was placedin a similar position, though closer to the beach, inorder to assess the flow of sediment which fed thisdune field from the beach. Surprisingly, this trapcaptured no sediment, suggesting that the main dunefield is also largely being fed by its own sedimentreservoir without receiving any significant contributionfrom the beach.Our results strongly suggest that themaintenance of the vegetation cover is fundamentalto inhibit the transport of sand, which is extremelysensitive to the water balance as illustrated by thesediment mobility that occurred in the areas with novegetation cover (traps 7 and 14). The observedpattern reflects a strong seasonality with expansionof vegetation cover and decrease of eolian sedimenttransport (Fig. 6).During the period of investigation the rainyseason, with monthly rainfall above 60 mm,extended from November 2007 to April 2008 withthe highest levels of precipitation in January andMarch (Fig. 7) while, according to Barbiére (1975),there is normally a decrease in precipitation afterDecember. The expected relationship between highprecipitation and low eolian sediment transport wasnot found in November, January or August. Asshown in figure 7, the number of rainy days in eachmonth was fewer than five, mainly associated withthe passage of cold fronts. After the rain the sandsdeprived of vegetation became once againsusceptible to transport by winds. Thus, for theeolian sediment transport, frequency of rainfall ismore important than the amount, whereas for themaintenance of vegetation the amount of monthlyprecipitation is still important.A better explanation for the observed patternof eolian sediment transport was found when themonthly duration of the highest wind speeds wastaken into consideration. During the period ofinvestigation mean monthly wind speed variedbetween 7 m/s and 9 m/s, with a constant standarddeviation of ±2 m/s, while the highest values rangedbetween 9 m/s and 14 m/s. When considering onlywinds from 50º to 70º, which represent the maindirection of transport, and velocities ≥ than 9 m/s therelation with eolian sediment transport issignificantly higher (Fig. 8). The 9 m/s windvelocity threshold was chosen as being the highestmean monthly wind velocity.Figure 5. Graphic correlation between median grain sizeand standard deviation of the beach, foredune anddeflation plain sediments. The samples inside the ellipserepresent the grain size characteristics of the sourcesediments that feed the parabolic dune and the activedune field; these sediments are characterized by lowstandard deviation (well to very well sorted) and mediangrain size finer than 2.3 phi (0.2 mm).Figure 6. Sediment accumulated in the traps reflectsalternating humid and dry weather conditions.In the light of the preliminary results theimportance of the vegetation cover is clear, both ininhibiting sediment mobility and in producing theopposite effect when removed. During the monitoringperiod it became apparent that the activityof off-road vehicles (buggies and motorcycles) alsotriggered the remobilization of sediments by dama-Pan-American Journal of Aquatic Sciences (2010), 5(2): 267-276

272D. MUEHE ET ALLIging the vegetation cover, and resulting in blowoutsand modification of the topography (Fig. 9).Mobility of the Parabolic Dunes of the DeflationPlainThe comparison of the topographic profilesof three parabolic dunes did not indicate any netdisplacement, despite the intense mobilization ofsediments on the windward side of the depositionallobe. This mobilization is reflected in the greatermorphodynamic variability of this face as comparedto that on the slip face (Fig. 10).Considering the present climate conditions,the parabolic dunes are reasonably stable, withno input of sediments from the foredune-beachsystem, due to the, even sparse, vegetation coverof the foredune ridge as also of the deflation plain;in contrast, there is intense sediment remobilizationon the windward, vegetation depleted side of thedunes,due to the reworking of their own sediments. Themaintenance of the vegetation cover is, however,extremely dependent on the water balance, withlonger dry periods leading to the destabilization ofthe system due to the intensification of the windtransport. This is well exemplified by the 1959 aerialphotograph, which indicates sediment transport fromthe foredunes in the direction of the active dunefield, through transgression over the deflation plain(Fig. 11).When the annual rainfall rates measured inCabo Frio (from 1921 to 1968) and Arraial do Cabobetween (1969 and 1986) are compared with theaverage rainfall of the same period (Fig. 12) largeFigure 9. Destruction of the vegetation cover andsubsequent incision of the terrain due to the traffic of offroadvehicles. Location landward from position P6,figure 3.Figure 7. Monthly precipitation and indication of thenumber of rainy days and number of cold fronts.Figure 8. Duration of wind velocities over 9 m/s indicatingthe close relationship between sediment transport(Fig. 6) and duration of strongest winds from 50-70º.Figure10. Overlay of the 12 monthly topographic profilesalong the longitudinal axes of three parabolic duneslocated on the deflation plain (see Fig. 3). No netadvances of the base of the slip face of the dunes weredetected. The last profile of each dune is represented by ared line. Location of profiles in figure 3.Pan-American Journal of Aquatic Sciences (2010) 5(2): 267-276

Peró coastal plain vulnerability273oscillations are observed, with the occurrence ofalternating dry and rainy periods. In the 1950´s themonthly rainfall fell far below the average duringfive successive years what might explain thereactivation of the wind transport as indicated by the1959 photograph (Fig. 11). This phenomenon maybe recurrent during occasions of water deficit: anaerial photograph from 1976 shows some sedimenttransport in the direction of the active dune field,though less evident than that seen in the 1959photograph. It must be borne in mind that in 1975the annual rainfall was above the average value andso the effects of the water deficit were not soremarkable.Erosion and Coastal FloodingIn the study area, the georeferencing ofolder aerial photographs is extremely difficult due tothe sparse land occupation and the consequent lackof points of reference. Images from 1970 and 2000were compared by drawing the position of theinternal limit of the berm in each image; no evidenceof significant change in the position of the coastlinewas found. Dias et al. (2007) had found a differentresult by inferring erosion of the coastline based onsatellite images and a digital terrain model.However, in a later study, Dias et al. (2009)reappraised this interpretation as they had thenidentified a period of coastal advance of some 30mbetween 1959 and 1976, and of retreat, also of 30m,between 1976 and 2003. This result should beaccepted with caution because these analyses werebased on aerial photographs by taking the limits ofthe vegetation cover as points of reference, besides afew roads and buildings, what may lead tosignificant errors. Beyond that, it is possible thatchanges may have occurred in the vegetation coverat the foredune-beach interface which could lead toan equivocal interpretation of the position of thecoastline. Assuming however, that the amplitude ofthe erosive process was compensated byprogradation of equal extent, the final result does notdisagree with the findings of the present study.The analysis of the time series of thetopographic profiles of the foredune-beach system(Fig. 13) did not identify the occurrence of erosion,apart from the morphodynamic variability of thebeach with periods of gains and losses typical ofbeaches of the intermediate morphodynamic stage(Short 1999, Calliari et al. 2003). A similar resultwas obtained by Pereira et al. (2008).The bathymetric profiles present significantlandward increase in slope at 9 and 10 m depth.Equilibrium profiles (Dean 1991) despite indicatingequilibrium of the upper shoreface are not adjustedto the remaining measured profiles, with “excess” ofsand in profile P7 and lack in profiles P2 and P5(Fig. 14). This may indicate a lack of sedimentstending to an erosive adjustment of the profile or acompound profile in which the intermediateshoreface presents a profile with equilibrium of itsown, as proposed by Inman et al. (1993).Figure 11. Aerial photograph taken in 1959, by theCruzeiro do Sul company. The bright white areas at thecenter and at the northern edge of the beach showsediment transport from the foredunes to the active dunefield.Figure 12. Annual rainfall in Cabo Frio/Arraial do Caboshowing significant deviations from the average of the1921-1986 period, represented by the black lineparallel to the x-axis. No data were collected during theperiod of 1960 to 1961. In 1982 registered rainfall wasonly 91.8 mm.Pan-American Journal of Aquatic Sciences (2010), 5(2): 267-276

274D. MUEHE ET ALLIFigure 15. Estimated erosive adjustment of differentsectors of the coastline to distinct sea level rise scenariosaccording to the Bruun rule. Profile locations in figure 3.Figure 13. Overlay of the 12 monthly topographicprofiles of the foredune-beach system indicating theabsence of net erosion during the period of observation.The last profile is represented by a red line. Location ofprofiles in figure 3.Figure 14. Topo-bathymetric profiles of the foredunebeach-shorefacesystem compared to the equilibriumprofile (broken line) according to the model proposed byDean (1991).The non-identification of an erosive trend ofthe coastline (not taking into account the possiblelack of sediments in the middle shoreface) isauspicious in the sense that it indicates a certainrobustness of the coast in the face of the variabilityof the coastal processes.However, a rise in sea-level will tend toresult in an erosive adaptation of the coastline asthere are no sediment sources to compensate for thisrise. The application of Bruun´s model (1962, 1988),as described above, to provide a rough estimate ofthe amplitude of this response to different scenariosof sea level rise, is presented in figure 15; the modelpredicts an increase in the amplitude of the retreatfrom the center to the extremities of the beach arc aresult that stems essentially from the lower gradientof the shoreface, associated with a smaller grain sizein the northern and southern sectors of Peró beach.For the urban area, behind profile P2, thiserosive adaptation will mean the destruction of thepublic walk in front of the urban area, unless thebeach is preserved by means of artificial landfill.Another vulnerability factor arises from the shallowdepth of the water table and the poor drainage of thedeflation plain with the consequent renewedflooding of the low-lying areas that were extensivelyflooded during a period of intense rainfall thatoccurred in the period from January to May 2008.ConclusionThe Peró coastal plain, with its system ofsandy dunes, presents a highly sensitive environmentalequilibrium, as climatic oscillations tendingto drier weather, independently of global changes,can disturb its delicate morpho-sedimentary balance.Alterations in the water balance, whether by dimini-Pan-American Journal of Aquatic Sciences (2010) 5(2): 267-276

Peró coastal plain vulnerability275shed rainfall or increased evaporation, may lead to areduction of the vegetation cover of the foredunesand of the deflation plain, thus re-starting sedimenttransport from the foredunes to the active dune fieldthat, in turn, will resume its displacement towardsthe urban areas. Sea level rise and increasedstorminess will lead to both, an erosive adjustmentof the coastline by means of its retreat and to increasedflooding of the low-lying areas of the deflationplain and of the paleo-lagoon located behind theactive dune field. Other consequences include thelandward displacement of the foredunes and thetransference of sediments as blowouts and parabolicdunes towards the deflation plain and the activedune field, as described by Psuty & Silveira (2010).The observations made on the Peró coastalplain may also be applied to other coastal dune fieldsthat occur in Rio de Janeiro State in areas withhigher rainfall rates such as those observed to theReferencesBarbiére, E. B. 1975. Ritmo climático e extração desal em Cabo Frio. Revista Brasileira deGeografia, 37(4): 23-109.Barbiére, E. B. 1984. Cabo Frio e Iguaba Grande,dois microclimas distintos a um curtointervalo de tempo. In: Lacerda, L. D.,Cerqueira, R. & Turcq, B. (Eds.). Restingasorigem, estrutura, processos. Niterói,CEUFF. p. 3-14.Belligotti, F. M. 2009. Avaliação metodológica daprofundidade de convergência (profundidadede fechamento) de perfis de três praias deenergia moderada a alta no litoral do Rio deJaneiro. M.Sc. Thesis. Programa de Pós-Graduação em Geografia, UFRJ. 130 p.Belligotti, F. M. & Muehe, D. 2007. Levantamentodo perfil da antepraia (shoreface) com uso deecobatímetro portátil e mini-embarcação. XXSimpósio da ABEQUA.Birkemeier, W. A. 1985. Field data on the seawardlimit of profile change. Journal ofWaterway, Port, Coastal and OceanEngineering, 111: 598-602.Bruun, P. 1962. Sea level rise as a cause of shoreerosion. Journal of Waterway, Port, Coastaland Ocean Engineering, ASCE, 88: 117-130.Bruun, P. 1988. The Bruun rule of erosion by sealevelrise: a discussion on large-scale two- andthree - dimensional usages. Journal ofCoastal Research, 4(4): 627-648.Calliari, L. J., Muehe, D., Hoefel, F. G. & Toldo Jr.,E. 2003. Morfodinâmica praial: uma brevenorth, in the Paraíba do Sul river plain, and to thesouth, in the beach barriers found in front ofAraruama and Saquarema lagoons as well as ofSepetiba bay (Marambaia beach barrier) wherelocalized active blowouts already indicate occasionaleolian sediment transport. As some of these areas areexperiencing increased occupation adequate setbacklines should be established in order to avoid futureloss of propriety.AcknowledgementsThe authors wish to thank the Departamentode Recursos Minerais (DRM) and AMPLA for theirrespective authorizations of the use of the aerialphotographs and orthophotocharts. We would,further, thank CNPq and CAPES for the graduateand research scholarships as also for two anonymousrevisers whose suggestions substantially improvedthe present document.revisão. Revista Brasileira de Oceanografia,52: 63-78.CBMERJ – Corpo de Bombeiros Militar do Estadodo Rio de Janeiro. Dados pluviométricos daEstação Meteorológica de Cabo Frio, athttp://migre.me/3b5Lr.CPTEC. Centro de Previsão do Tempo e EstudosClimáticos/Instituto Nacional de PesquisasEspaciais. Boletim Climanálise. CachoeiraPaulista, at http://migre.me/3b5Kk.Dean, R. G. 1991. Equilibrium beach profiles.Characteristics and applications. Journal ofCoastal Research, 7: 53-84.Dias, F. F., Pereira, R.S., Seoane, J. C. S. & Castro,J. W. A. 2007. Utilização de imagens desatelite, fotografias aéreas, MDT'S e MDE noestudo de processos costeiros em Cabo Frio -RJ. XIII Simpósio Brasileiro deSensoriamento Remoto, Anais. Florianópolis,UFSC, 1: 56-58.Dias, F. F., Seoane, J. C. S. & Castro, J. W. A. 2007.Evolução da linha de praia do Peró, CaboFrio/RJ nos últimos 7.000 anos. Anuário doInstituto de Geociências - UFRJ, 32(1): 9-20. www.anuario.igeo.ufrj.br.Dourado, F. A. & Silva, A. S. 2005. Monitoramentodo avanço da frente de dunas na região doPeró, Cabo Frio, Rio de Janeiro. XIISimpósio Brasileiro de SensoriamentoRemoto. Goiânia, Brasil, 2957-2064.Folk, R. L & Ward, W. C. 1957. Brazos river bar: astudy in the significance of grain sizeparameters. Journal of SedimentaryPan-American Journal of Aquatic Sciences (2010), 5(2): 267-276

276D. MUEHE ET ALLIPetrology, 27: 3-26.Hallermeier, R. J. 1981. A profile zonation forseasonal sand beaches from wave climate.Coastal Engineering, 4: 253-277.Houston, J. R. 1995. Beach-fill volume required toproduce specific dry beach width. CoastalEngineering Technical Note CETN II-32. USArmy Corps of Engineers. 8 p.Inman, D. L., Elwany, M. H., & Jenkins, S. A. 1993.Shore rise and bar-berm on ocean beaches.Journal of Geophysical Research, 98(18):181-199.Leatherman, S. P. 1978. A new eolian sand trapdesign. Sedimentology, 25: 303-306.Muehe, D. 2004. Método de levantamento topo-batimétricodo sistema praia-antepraia. RevistaBrasileira de Geomorfologia, 5(1): 95-100.Muehe, D., Roso, R. H. & Savi, D. 2003. Avaliaçãode método expedito de determinação do níveldo mar como datum vertical para amarraçãode perfis de praia. Revista Brasileira deGeomorfologia, 4(1): 53-57.Pereira, T. G., Correa, W. B., Rocha, T. B. &Fernandez, G. B. 2008. Considerações sobre aMorfodinâmica Costeira e da MorfologiaSubmarina no Arco de Praia do Peró, litoraldo Rio de Janeiro. VII Simpósio Nacional deGeomorfologia & I Encontro Latino-Americano de Geomorfologia, BeloHorizonte, Brasil.Psuty, N. P. & Silveira, T. M. 2010. Global climatechange: an opportunity for coastal dunes?Journal of Coastal Conservation, 14(2):153-160.Rangel, F. E. & Castro, J. W. A. 2005. Soterramentoda estrada do Guriri na Praia do Peró - CaboFrio/RJ, associada a dunas escalonares. XCongresso da Associação Brasileira doQuaternário - ABEQUA, 1: 133-138.REDEMET. Rede de Meteorologia do Comando deAeronáutica. Dados meteorológicos daEstação Meteorológica de Superfície deArraial do Cabo, at http://migre.me/3b5PIShort, A. D. 1999. Wave-dominated beaches. In:Short, A. D. (Ed.). Handbook of beach andshoreface morphodynamics. John Wiley &Sons, Ltd., Chichester, England.Wang, P. & Davis Jr, R. A. 2007. Profundidade defechamento e perfil de equilíbrio de praia: umestudo de caso em Sand Key, Florida. RevistaMercator, 6(12): 51-68.Received May 2010Accepted September 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010) 5(2): 267-276

Historical assessment of extreme coastal sea state conditions insouthern Brazil and their relation to erosion episodesARTHUR A. MACHADO 1* , LAURO J. CALLIARI 1 , ELOI MELO 2 & ANTONIO H. F. KLEIN 31 Institute of Oceanography, Federal University of Rio Grande, Av Itália Km 8, CEP: 96201-900, Rio Grande, RS,Brazil.2 Engineering School, Federal University of Rio Grande, Av. Italia Km 8, CEP 96201-900, Rio Grande, RS, Brazil.3 Department of Geosciences, Federal University of Santa Catarina, Campus Universitário Trindade, CEP 88040-970,Florianópolis, SC, Brazil.* Corresponding author: oceaam@yahoo.com.brAbstract. Intense cyclonic weather systems in southern Brazil generate ocean storms which can,in a temporal scale varying from few hours to a day completely erode a beach profile from itsmaximum accretion state. Mid-latitude cyclogenesis with low pressure centers in the deep oceanand along the coast increases the intensity of the Mid-Atlantic storms causing storm surges andstorm waves. Preliminary results from a hindcast of wave energy at deep water (100 m),performed with a wave model using winds from reanalysis ( period 1979 - 2008), indicated a totalof 40 extreme events (wave height above 6 m). These events cause maximum erosion and surgeelevation on the order of 62.96m³/m and 1.827 m respectively. Four patterns of synoptic situationscapable of generating extreme events were identified. Among the 40 events, 53.66% had thetrajectory of Pattern II and 26.82% were associated to Pattern III, representing both 80% of thetotal. Coastal erosion episodes where mostly associated with Pattern II, while Pattern III causedthe highest surges. In a climate change scenario this study shows no important differences in theamount of the extreme events along the last thirty years.Key words: storm surge, extra-tropical cyclones, wave height, NCEP/NCAR ReanalysisResumo. Avaliação histórica das condições extremas de mar na costa do sul do Brasil e suarelação com episódios de erosão. Sistemas meteorológicos como ciclones extratropicais de altaintensidade que ocorrem no sul do Brasil geram ondas de alta energia, que podem levar um perfilde praia de um estágio máximo acrescivo ao máximo erodido em poucas horas. A ciclogênese emmédias latitudes, com centro de baixa pressão, contribui para a intensificação das tempestades doMeio do Atlântico, causando marés meteorológicas (storm surges) e ressacas (storm waves).Resultados preliminares para um estudo de energia das ondas em águas profundas (100 m),utilizando um modelo de ondas com dados de vento de reanálises (período 1979 - 2008),indicaram 40 eventos extremos (6 m de altura de onda). Alguns desses eventos geraram erosão de62,96 m³/m e 1,827 m de elevação do nível do mar. Foram identificados quatro padrões desituações sinóticas geradoras de alturas de ondas acima de 6m. Entre os 40 eventos, 53,66%tiveram a trajetória do Padrão II e 26,82% estavam associados ao Padrão III, ambos representando80% do total. Episódios de erosão costeira geralmente são associados ao Padrão II. Já o Padrão IIIé responsável pela maior elevação do nível do mar. Diferenças significativas na quantidade deeventos extremos ao longo dos últimos 30 anos não foram observadas.Palavras-chave: maré meteorológica, ciclone extra-tropical, altura de onda, NCEP/NCARReanálisesIntroductionAfter the accretion period which occursbetween December and March, storms beginning inApril start the erosion cycle of the southernBrazilian sandy beaches. Generally, erosion iscaused by extreme sea state events which combinehigh waves and high storm surges. Sinceastronomical tides have higher amplitude in thisregion during April, and the storms can last a fewPan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

278A. A. MACHADO ET ALLIdays, it is not uncommon that episodes of severeerosion occur during the high water spring tideperiod (Calliari et al. 1998). These storms are mostlyassociated with high intensity extra-tropical cyclonesthat generate wind waves which can change a beachprofile from its maximum accretion state tocomplete erosion during a period that can vary fromfew hours to a few days.Regarding the occurrence of extra-tropicalCyclones in South America, Gan (1992), analyzing10 years of data (from 1979 to 1988) has found thatthe majority of events happen in winter (8 events),followed by autumn (6), spring (4) and summer (3).Gan & Rao (1991) identified two cyclogenesisregions in South America: one in Argentina (42.5° Sand 62.5° W) related to the baroclinic instability ofthe westerly winds and another in Uruguay (31.5° Sand 55° W) associated with the baroclinic instabilitydue to the presence of the Andes. Recently, a thirdregion between 20° and in the 35°S located insouthern and south-eastern Brazil was identified(Reboita et al. 2010).Mid-latitude cyclogenesis with low pressurecenters in the deep ocean and along the coastincreases the intensity of Mid-Atlantic stormscausing extreme storm surges and storm waves(Calliari et al. 1998). The “surge” in a specificinstant is represented by the difference between theobserved and the astronomical tide and can be eitherpositive or negative causing rapid increase ordecrease in sea level, respectively (Pugh 1987).Storm surges are the major geological risk inlow coastal areas. They are often associated withsignificant losses of life and property. Climatechange, with rising sea level and changing stormtracks, will modify the regional distributions of thesehazards (von Storch & Woth 2008).The two main sources of storm surges are:changes in atmospheric pressure and the exchange ofmomentum between the wind and the sea surface. Ingeneral, the effects associated with atmosphericpressure is less than 10% of the total, being the windshear stress at the sea surface the main component(Marone & Camargo1994). Additionally, sea levelelevations at the shore can be further amplified bythe presence of shelf waves and by the pilling up ofwater due to wave breaking processes at the surfzone (known as “wave set up”) (Marone & Camargo1994).Observations of synoptic weather conditionsand sea level elevation done by Parise et al. (2009)showed that the highest sea level elevation eventsresulted from the action of SW winds which blowparallel to the main NE-SW coastline orientation inthe region, a result that can be explained by thepilling up of water at the coast due to the Corioliseffect (i.e. Ekman transport). The monitoring carriedout by Saraiva et al. (2003) from April 1997 to July1999 on Cassino Beach indicated the highestfrequency of the storm surge in autumn (65%),followed by similar values in summer and spring(15%) and lower values in winter (5%). All thestorm surges observed by Saraiva et al. (2003) wereassociated with extra-tropical cyclones.Coastal erosion has been causing substantialalterations along the coastline of the Rio Grande doSul (RS) state in southern Brazil for quite sometime. In the less occupied areas in the central littoral,coastal erosion caused habitat loss of foreduneridges and inflicted local damage to a lighthouse(Conceição lighthouse) and small beach resorts atLagamarzinho beach (Barletta & Calliari 2003). Inmore developed regions of the northern littoral, suchas Cidreira, Tramandaí and Imbé beaches, coastalerosion is aggravating, leading to severe loss ofpublic and private property (Esteves et al. 2000,Toldo Jr. et al. 1993). At the southern littoral,Hermenegildo beach, located near the Uruguayanborder, has had homes, roads and power linessystematically destroyed (Calliari et al. 1998,Esteves et al. 2000). Additionally, in severalstretches of the RS coastline, beach erosion causesexposure of peat and muddy lagoonal outcropsleading to a decrease in the quality of beachrecreation (Calliari et al. 1998) (Fig. 1).In a climate change scenario, the presentstudy aims at assessing in detail the synopticsituations that give rise to extreme sea state events inSouthern Brazil and determining trends in theatmospheric patterns and path lines ofmeteorological systems associated with them. Theerosional impact on the coastline and the stormsurges caused by these extreme events are alsoinvestigated. Case studies of selected extreme eventsthat generated strong beach erosion are alsodiscussed in detail.Materials and MethodsExtreme eventsIn the absence of sufficiently long data sets,we had to resort to numerical models to infer theoccurrence of extreme events. The results usedherein were extracted from a comprehensive studythat is being currently carried out by the third authorand are still preliminary (see Melo et al. 2010). Inthat on-going study, the wave generation modelWave Watch III (WW3) (Tolman 2002) forced withreanalysis winds from NCEP (National Centers forEnvironmental Prediction) was used to reconstructsea state conditions off the southern Brazilian coastPan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

Extreme coastal sea state conditions in southern Brazil279Figure 1. Study area and location of cited sites.from 1979 to 2008. “Extreme” events were thenselected based on the criteria that the reconstructedsignificant wave height (Hs) at a point in 100 mdepth off Rio Grande city exceeded the 6 m mark.Preliminary results indicated that a total of 40extreme events occurred in the studied period.Synoptic scenarios associated with extreme eventsThe reanalysis dataset was created throughthe cooperative efforts of the NCEP and NCAR(National Center for Atmospheric Research) (Kalnayet al. 1996) to produce relatively high-resolutionglobal analyses of atmospheric fields over a longtime period. The reanalysis dataset (R-1 of NCEP /NCAR) was used to characterize atmosphericconditions that originated these 40 extreme events.To do so, meridional and zonal components of thewind and atmospheric pressure at the 995 mbar levelwere used. For both a spatial resolution of 2.5° x2.5°, and a temporal resolution of 6 hours (0000,0600, 1200, 1800 UTC) restricted between 60°S -15°S and 90°W - 20°W was adopted. In order tobetter characterize the path of the systems thatgenerated extreme events, a threshold vorticity lesseror equal than (ζ 10 ) -5 x 10 -5 s -1 was adopted. Thereanalysis dataset is available at sitewww.cdc.noaa.govData analysisAnalysis of variance (ANOVA) was used toverify differences between the numbers of events at3 years interval along the 30 years. Data normalityPan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

280A. A. MACHADO ET ALLIwas tested through the Kolmogorov-Smirnov testand the homogeneity of the same ones through theLevene test (Zar 1999).Results and DiscussionIn the period between the years of 1979 and2008, 40 events of significant wave height (Hs)above 6 m occurred considering as reference theposition of (32°54’S, 50°48’W) at 100 m waterdepth. The yearly mean number of events was 1.33with a minimum of 0 events and a maximum of 4events in the year of 1999. The standard deviationwas 0.958/year.The ANOVA result of the 10 groups joinedat 3 years interval shows no significant difference inthe number of extreme wave height events along the30 years period (F(9, 20) = 1.4815, p = 0.22141(Fig. 2A). The correlation graphic between thenumbers of events and the thirty 30 years periodshowed a positive but weak correlation (r = 0.30636)(Fig. 2B).Anumber of eventsBnumber of events4,54,03,53,02,52,01,51,00,50,0-0,5-1,01979-1981 1985-1987 1991-1993 1997-1999 2003-20051982-1984 1988-1990 1994-1996 2000-2002 2006-20083 years4,54,03,53,02,52,01,51,00,50,0-0,51975 1980 1985 1990 1995 2000 2005 2010yearFigure 2. (A) Mean of the number of extreme events ofthe 10 groups, Vertical bars denote 0.95 confidenceintervals, (B) Correlation of the number of extreme eventsover the study period, the dashed line denote 0.95confidence intervals and the continuous line denote theregression line.Webster et al. (2005) found an increase inthe number of tropical cyclones and cyclone days aswell as tropical cyclone intensity over the past 35years, in an environment of increasing sea surfacetemperature. A large increase was seen in thenumber and proportion of hurricanes reachingcategories 4 and 5. The largest increase occurred inthe North Pacific, Indian, and Southwest PacificOceans, and the smallest occurred in the NorthAtlantic Ocean.From the analysis of the meteorologicalscenarios, four patterns of synoptic situationscapable of generating extreme events were identified(Fig. 3):• PATTERN I: Cyclogenesis in the southernArgentinean coast with a displacement to the eastand a trajectory between 47.5°S and 57.5°S;• PATTERN II: Cyclogenesis in the southernUruguayan coast with a displacement to the east anda trajectory between 28ºS and 43°S;• PATTERN III: Cyclogenesis in the southernUruguayan coast with a displacement to thesoutheast and a trajectory between 32°S and 57.5°S;• PATTERN IV: High-pressure center generatingan easterly wind.Was not observed at the study area asignificant difference in the frequency of thepatterns of cyclone trajectories along time. Thepattern with the greatest number of extreme eventswas Pattern II with 22 of the 40 +1 events. Eleven(11) events were associated with Pattern III and four(4) events were associated to both Pattern I and IV.The value of 41 related to the sum of all the patternsis due to an event that occurred on 07/21/1996, inwhich, two parallel extra-tropical cyclonesresembling Patterns I and II occurred simultaneously.Case StudiesIn this section, a selection of extreme seastate events was used to assess both, the specificmeteorological scenarios associated, and theresponse that was observed on the coast in terms ofbeach erosion.July 21 th , 1996An extreme wave height event occurred onJuly 21 th , 1996. The meteorological scenario showstwo extra-tropical cyclones parallel to each otherrepresenting Patterns I and II. Due to its eastern pathand the following of a high pressure system in therear, a long southwest wind fetch of more than 3000km was formed over the Atlantic off the SouthAmerican coast (Fig. 4).Pan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

Extreme coastal sea state conditions in southern Brazil281This event caused the maximum erosionprofiles recorded in 1996. At places located betweenSolidão and Estreito lighthouses the maximumeroded volume reached 62.96 m³/m (Barletta &Calliari 2003) (Fig. 5).April 18 th , 1999The meteorological scenario on this extremeevent was unusual since the path of the cyclone thatdeveloped off the RS coast formed a loop withoutmuch forward motion (Fig. 6).Severe erosion was observed at Hermene-gildo beach. Prior to the storm, this beach resort had110 beachfront houses. During the storm, 22 houseswere destroyed or highly damaged. This singlestorm was also able to destroy the majority ofcoastal protection structures including 20% of allbeachfront houses. However, as it was laterobserved, all the coastal protection structures werebuilt on top of the foredunes without any foundationunderneath them, being, in this way, susceptible toundermining. Esteves et al. (2000) indicated that thiswas the process that caused most of the structures tocollapse (Fig. 7).Figure 3. Path of the four synoptic situations: (A) Pattern I, (B) Pattern II, (C) Pattern III, (D) Pattern IV.Figure 4. (A) Trajectory, (B) Synoptic situation, wind field (knots) and pressure (mbar).Pan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

282A. A. MACHADO ET ALLIthe initial profile was already eroded by the winterstorms. Regarding the meteorological scenario, itcan be observed the development of a long windfetch from S to SW (Fig. 9). The associationbetween this wind pattern and the NE-SWorientation of the shoreline favored the extra highrise in sea level observed on the coast due to theCoriolis effect (Parise et al. 2009).Figure 5. Shoreline (m) and volume changes (m³/m)between Estreito and Solidão lighthouses, after the stormof July 1996. Modified from Barletta & Calliari (2003).May 25 th , 2003This event was, actually, monitored by anoceanographic buoy moored offshore Rio Grandecity (Minuano buoy from the PNBOIA, Programmoored at a depth of 70 m) since it caused muddeposition at Cassino beach (Calliari & Faria 2003).Buoy measurements, which did not includedirection, are displayed in Table I. Maximumsignificant wave height came very close to the 7 mmark. The path described by the extra-tropicalcyclone on this event resembled Pattern II (Figure8A). A large low pressure center can be observedmoving slowly towards the E (Fig. 8B).September 04 th , 2006This event coincided with one of the extratropicalcyclones studied by Parise et al. (2009),who shows that this particular storm caused a surgeof 1.827 m. Although the surge was very high, beacherosion was low (-8.14 m 3 /m) the reason being thatTable I. Wave data from the Minuano buoy (100 mdepth) (Calliari & Faria 2003).Date/Hours Wave height (m) Period (s)25/05/2003- 00:0025/05/2003- 02:0025/05/2003- 07:0025/05/2003- 13:0025/05/2003- 17:0025/05/2003- 20:006.96.76.95.65.66.911.610.711.611.114.216.0Video-images from an ARGUS system(Holman & Stanley 2007) analyzed before and afterthe onset of the extra-tropical cyclone allowed thequantification of changes in beach width at Cassinoduring the event (Parise et al. 2009) Timex imagesfrom the same system display the surge reaching thedunes and the differences of the surf zone width witha third bar appearing during the storm surge (Fig.10). Maximum values of storm surges of the order of1 m, 1.4 m and 1.9 m in the coast of the RS havebeen found by Calliari et al. (1998), Saraiva et al.(2003) and Parise et al. (2009), respectively. Duringthis event great part of Cassino beach was floodedwhen the water reached the first avenue close to thebeach.Studies done by Saraiva et al. (2003) andParise et al. (2009) pointed out that the maximumelevation of the surge occurs mainly 24 hours afterthe cyclone formation (Tab. II).Figure 6. (A) The path in loop, (B) Synoptic situation, wind field (knots) and pressure (mbar).Pan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

Extreme coastal sea state conditions in southern Brazil283Figure 7. Beach profiles done before and after the event of 18/4/1999. Modified from Esteves et al. (2000).Figure 8. (A) Trajectory, (B) Synoptic situation, wind field (knots) and pressure (mbar).Figure 9. (A) Trajectory, (B) Synoptic situation, wind field (knots) and pressure (mbar).Pan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

284A. A. MACHADO ET ALLIFigure 10. Timex images of the Argus system installed at Cassino beach, during a normal surf-zone (left) and duringthe storm surge (right). LOG-FURG/2007 - http://www.praia.log.furg.br/Figure 11. Synoptic situation of the of high pressure center (anticyclone) moving toward the east. Wind (knots) andpressure (mbar)Pan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

Extreme coastal sea state conditions in southern Brazil285Table II. Time interval between the formation of thecyclone and the maximum surge elevation.MONITORING 6h 24h 36h 48h1997 to 1999 (Saraiva et al. 2003) 10% 45% 10% 30%2006 to 2007 (Paris et al. 2009) 9% 39% 26% 26%Special CasesAmong the 40 extreme wave height events,four were generated by strong easterly windsassociated with large anticyclonic system, whichalso displays the path of the high-pressure centerbetween March 04 and 05 of 1996. This eventgenerated waves from the east quadrant as indicatedby the wind field shown in figure 11.ConclusionThis study shows no important differencesin the amount of extreme events along the last thirtyyears. The mean number of events obtained was1.33 per year. To these events data of wind velocityand vorticity, atmospheric pressure and sea levelelevation were added. Effects of extreme events onthe coast caused maximum erosion and surgeelevation on the order of 62.96m³/m and 1.827 m,respectively.Among the 40 events studied, 22 (53.66%)had the trajectory of Pattern II with Cyclogenesis tothe south of the Uruguayan coast with a path to theeast and a trajectory between 28°S and 43°S.Cyclones associated with Pattern III, represented26.82% (11 events). Those two types represent 80%of the total extreme events. The relationship betweenthe coastal erosion and these extreme events is clear,as observed from Parise et al. (2009), being thecyclones associated with Pattern II the most erosiveones, whereas those associated with Pattern III theones that cause highest surges.The reanalysis dataset of NCEP provedvery useful in this type of analysis, becausealthough the spacing of the data 2.5 degrees,cyclones and anticyclones studied have diametersabove 1000 km, thus presenting a good answerto the synoptic situation that caused each extremeevent studied.ReferencesBarletta, R. C. & Calliari, L. J. 2003. An assessmentof the atmospheric and wave aspectsdetermining beach morphodynamic characterristicsalong the central coast of RS state,Southern Brazil. Journal of CoastalResearch, 35(SI): 300-308.Calliari, L. J., Tozzi, H. A. M. & Klein, A. H. F.1998. Beach morphology and CoastlineErosion Associated with Storm Surge inSouthern Brazil- Rio Grande to Chuí, RS.Anais da Academia Brasileira de Ciências,70(2): 231-247.Calliari, L. J. & Faria, G. A. F. 2003. Bancos delama: na praia do cassino: formação,implicações geomorfológicas, ambientais eriscos costeiros. Estudo de caso: maio de2003. IX Congresso da AssociaçãoBrasileira de Estudos do Quaternário(ABEQUA), Recife, Pernambuco, Brazil,CD-ROM 5 p.Esteves, L. S., Pivel, M. A. G., Silva, A. R. P.,Barletta, R. C., Vranjac, M. P., Oliveira, U. R.& Vanz, A. 2000. Beachfront OwnersPerception of Beach Erosion along anArmored Shoreline in Southern Brazil.Pesquisas em Geociências, 27(2): 93-109.Gan, M. A. & Rao, B. V. 1991. Surface ciclogenesisover South America. Monthly WeatherReview, 119: 293-302.Gan, M. A. 1992. Ciclogêneses e ciclones sobre aAmérica do Sul. PhD Thesis, InstitutoNacional de Pesquisas Espaciais, INPE, SãoJosé dos Campos – São Paulo, Brasil 183p.Holman, R. A. & Stanley, J. 2007. The history andtechnical capabilities of Argus. CoastalEngineering, 54: 77–491.Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W.,Deaven, D., Gandin, L., Iredell, M., Saha, S.,White, G., Woollen, J., Zhu, Y., Leetmaa, A.,Reynolds, B., Chelliah, M., Ebisuzaki, W.,Higgins, W., Janowiak, J., Mo, K. C.,Ropelewski, C., Wang, J., Jenne, R. & Joseph,D. 1996. The NCEP/NCAR 40-yearreanalysis project. Bulletin of the AmericanMeteorological Society, 77: 437–472.Marone, E. & Camargo, R. 1994. MarésMeteorológicas no litoral de Estado doParaná: O Evento de 18 de agosto de 1993.Revista Nerítica, 8(1-2): 73-85.Melo, E., Romeu, M. A. R. & Hammes, G. R. 2010.Condições extremas de agitação marítima aolargo de Rio Grande a partir do ModeloWW3. IV Seminário e Workshop emEngenharia Oceânica - FURG(SEMENGO), Rio Grande Rio Grande doSul, Brasil, 20 p.Parise, C. K., Calliari, L. J. & Krusche, N. 2009.Extreme storm surges in the south of Brazil:atmospheric conditions and shore erosion.Brazilian Journal of Oceanography, 57(3):175-188.Pugh, D. T. 1987. Tides, surges and mean seaPan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

286A. A. MACHADO ET ALLIlevel. A handbook for Engineers andScientists, John Wiley & Sons Ltd, NewYork, 472 p.Reboita, M. S., Rocha, R. P. & Ambrizzi, T. 2010.South Atlantic Ocean cyclogenesisclimatology simulated by regional climatemodel (RegCM3). Climate Dynamics, 35:1331-1347.Saraiva, J. M. B., Bedran, C. & Carneiro, C. 2003:Monitoring of Storm Surges on CassinoBeach. Jounal of Coastal Research, 35: 323-331.Toldo JR., E. E., Dillenburg, S. R., Almeida, L. E. S.B., Tabajara, L. L., Martins, R. R. & Cunha,L. O. B. P. 1993. Parâmetros morfodinâmicosda Praia de Imbé, RS. Pesquisas, 20(1): 27-32.Tolman, H. L. 2002. User manual and systemdocumentation of WAVE-WATCH IIIversion 2.22. NOAA/NWS/NCEP/IOMBTech. Note 222. 133 p.von Storch, H. & K. Woth 2008. Storm surges,perspectives and options. Sustainability.Science, 3: 33-44.Webster, P. J., Holland, G. J., Curry, J. A. & Chang,H. R. 2005. Changes in Tropical CycloneNumber, Duration, and Intensity in a WarmingEnvironment. Science, 309: 1844-1846.Zar, J. H. 1999. Biostatistical analysis. EnglewoodCliffs, New Jersey: Prentice-Hall, 663 p.Received January 2010Accepted December 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 277-286

Climatic changes in the coastal plain of the Rio Grande do Sul state inthe Holocene: palynomorph evidencesSVETLANA MEDEANIC 1 & IRAN C. S. CORRÊA 21 Laboratório de Oceanografia Geológica, FURG/Instituto de Oceanografia, Av. Itália, km 08, Campus Carreiros, RioGrande, RS, Brazil. CEP 96201-900. E-mail: svetlanamedeanic@furg.br2 Centro de Estudos de Geologia Costeira e Oceânica, UFRGS/Instituto de Geociências, Av. Bento Gonçalves, 9500,Porto Alegre, RS, Brazil. CEP 91509-900. E-mail: iran.correa@ufrgs.brAbstract. Climatic changes during the Holocene were the principal factors influencing thecoastal plain evolution and sea level oscillations. Climatic fluctuations were responsible forlittoral sedimentation and palaeoenvironmental changes. Palynological data, obtained fromcore-drillings, performed at the bottom of the Patos and the Tramandaí lagoons and in adjacentareas, were used for palaeoclimatic reconstructions. Palynomorphs, represented by algal andfungal palynomorphs, pollen and spores of vascular terrestrial and aquatic plants,microforaminifers, and scolecodonts showed their indicative value for palaeoclimaticreconstructions. The samples, dated by 14 C method, allowed the comparison of the results whenmaking interpretations. Increase in marine algal palynomorphs indicated sea level rise as aresult of global temperature increase. The beginning of the marine transgression was about 10kyrs BP after the last glacial period when temperature raised. Transgressive maximum (4-6 kyrsBP) was characterized by sea-level rise at about 4-5 m. Posterior regression began as result oftemperature fall and drier climate, forcing dune formation in the coastal plain. Palynomorphrecords from the lagoon sediments in the coastal plains have a great potential and may serve fordetail paleoclimatic reconstructions of the past and predictions of climatic changes in thefuture.Keywords: Palynology, Palaeoecology, Lagoon sediments, Extreme Southern BrazilResumo. Mudanças climáticas na Planície Costeira do Rio Grande do Sul no Holoceno:evidências de palinomorfos. As mudanças climáticas ocorridas durante o Holoceno foram osprincipais fatores que influenciaram na evolução da zona costeira e nas oscilações do nível domar. Flutuações climáticas foram responsáveis pela formação dos sedimentos litorâneos e asmudanças paleoambientais. Dados palinológicos, obtidos a partir de furos de sondagens etestemunhos executados no fundo da laguna dos Patos e da laguna de Tramandaí e nasadjacências, foram usados para reconstruções paleoclimáticas. Os palinomorfos representadospelas algas, fungos, grãos de pólens e esporos de plantas vasculares terrestres e aquáticas,microforaminiferos e escolecodontes mostram seus valores indicativos para as reconstruçõespaleoclimáticas. As amostras datadas pelo método de 14 C nos permitiram comparar osresultados obtidos quando foram efetuadas as interpretações. Aumento de palinomorfos de algasmarinhas indicou a subida do nível do mar como resultado do aumento das temperaturasglobais. Início da transgressão marinha foi a cerca de 10 ka AP depois da última glaciaçãoquando as temperaturas subiram. O máximo transgressivo (4-6 ka AP) foi caracterizado pelasubida do nível do mar de 4-5 m. A regressão posterior começou como resultado da queda dastemperaturas e aridização do clima, que causou a formação de dunas na zona costeira. Osregistros de palinomorfos nos sedimentos lagunares das zonas costeiras possuem um grandepotencial nas reconstruções paleoclimáticas do passado e também como previsões de mudançasclimáticas do futuro.Palavras-chave: Palinologia, Paleoecologia, Sedimentos Lagunares, Extremo Sul do BrasilPan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

288S. MEDEANIC & I. C. S. CORREAIntroductionPaleoclimatic changes during the Quaternary(approximately 2.5 Ma BP) were relatively rapidin the past as compared with past climate changesin the last 60 Ma. Understanding of climateoscillations during the Quaternary is importantto evaluate variability of the natural environmentsin the past and probably use of these data for abetter understanding of actual climate situationand prediction of the future climate. Climaticchanges occurred during the Holocene (10 ka BP tillpresent) were the main factors influencing on thecoastal plain evolution of the Rio Grande do Sulstate, Brazil. The oscillations of global temperaturescaused relatively high frequency changes on sealevel. The climatic modifications were alsoresponsible for the littoral sediments depositionin the coastal plains and palaeoenvironmentalevolution.There are many different methods (proxydata) for the Quaternary paleoclimate reconstructionsin the world (Bradley 1999). Palynologicalanalysis is one of the most important methods,providing information from continents tocomplement other proxy data about paleoclimates(Webb 1991, Bradshaw 1994, Ledru et al. 1998,Bradley 1999). In addition to pollen and spores ofvascular plants, sediments usually include differentnon-pollen palynomorphs, which are organic-walledmicrofossils composed of sporopollenin-like orchitin (pseudochitin) polymers (Traverse 1988).These palynomorphs as a rule are more resistant tocorrosion and oxidation than pollen and sporescomposed of sporopollenin (van Geel 1976,Komárek & Jankovská 2001). These palynomorphsare predominantly represented by cysts of Acritarcha(marine phytoplankton) and Dynophyta, zygospores,coenobiums and colonies of Chlorophyta, differentfungal palynomorphs, palynomicroforaminifers, andscolecodonts. The forms, usually classified bypalynologists as cysts of Acritarcha arePrasinophycean phycomata (Colbath & Grenfell1995). In spite of uncertainties, regarding of theirprecise biological affinities, acritarchs areconsidered as remains of cysts of algal protists(Tappan 1980, Strother 1996). Prasinophycean algaeare a class of green algae, which are known todayfrom freshwater to marine environments, although inrecent forms only the marine taxa produce afossilized phycomata. The vast majorities ofPrasinophycean phycomata are recovered frommarine sediments and/or are associated with marineorganisms. Furthermore, based on their distribution,morphology, and composition, most Prasinophyceanphycomata are assumed to be phytoplankton, andtherefore were the primary producers of the ancientmarine ecosystem during the Proterozoic andPaleozoic (Martin 1993).The study of non-pollen palynomorphsstarted at the early 1970s in the Netherlands, andsince then, the interest in their use for palaeoclimaticreconstructions is growing (van Geel 1976, van Geel& van Hammen 1978, van Geel et al. 1980/81, vanGeel & Aptroot 2006).In the Quaternary lagoon sediments ofcoastal plains, the palynomorph variability is morediverse than in continental deposits. In addition toalgal and fungal palynomorphs, there are frequentpalynoforaminifers (Zamora et al. 2007, Medeanicet al. 2009), Prasinophycean phycomata, and cystsof Dynophyta (Grill & Quatroccio 1996, Medeanicet al. 2001, Weschenfelder et al. 2008) indicative ofsediments deposited under sea water influence.Scolecodonts are palynomorphs, encountered insediments formed in shallow water environmentsnear the coast (Lorscheitter 1983, Cordeiro &Lorscheitter 1994).The different scientific projects regardingHolocene paleoclimate and palaeoenvironmentalreconstructions in the coastal Plain of the RioGrande do Sul State, Brazil based on palynomorphstudy started in 1980s (Lorscheitter 1983, Cordeiro& Lorscheitter 1994, Lorscheitter & Dillenburg1998, Medeanic et al. 2003, Medeanic 2006,Medeanic & Corrêa 2007). All obtained informationso far allowed us to make proxy evaluation ofclimatic oscillations occurred since the last Pleistoceneglaciations. The lowering of temperature thattime caused the regression of the ocean withnegative amplitude of 120-150 m. The temperaturerise at about 10 kyrs BP (the Early Holocene) wasthe reason of the beginning of the marinetransgression. The maximum of marine transgressionoccurred at about 5-6 kyrs BP (the MiddleHolocene) with positive amplitude of sea level riseat about 4-5 m. Posterior marine regression at about3-3.5 kyrs (the Late Holocene) was characterized bysea level fall at about 2 m. In this paper, we presenta revision of our previous published results, basedon palynomorph study focusing on the Holoceneclimatic paleoreconstructions.Materials and MethodsStudy areaThe study area is situated in the southernpart of the coastal plain of the Rio Grande do Sulstate (Fig. 1). The climate of this region is warmtemperate,due to the joint influence of the warmBrazil and cold Falkland currents (Vieira & RangelPan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

Climatic changes in the coastal plain of the Rio Grande do Sul2891988). The average annual temperature is 18 °C,with monthly averages of 24.6 °C in January and13.1 °C in July. The mean annual precipitation isabout 1,200 mm. The coastal plain has a diversegeomorphology, with sandy beaches, dunes,freshwater, brackish-water, salt marshes, andwetlands (Costa et al. 1997).In the Quaternary, the coastal plain wassubjected by glacio-eustatic sea level oscillations(Villwock & Tomazelli 1995, Corrêa et al. 1996).During the Holocene, a vast part of the presentBrazilian coastal plain was flooded by seawaters(Angulo & Lessa 1997, Martin et al. 2003).The last glacio-eustatic sea level rise (transgressivephase) began at about 10 kyrs BP. The maximumof that transgression occurred around 5,600 yrs BPcharacterized by a sea level rise of approximately5 m above the present sea level. The followingregressive stage led to the present sea level.Sandy sediments along the coast weredeposited during regressive-transgressive eventsin the Quaternary. Lagoon formation was connectedto the evolution of the Holocene barrier-lagoonalsystem (Villwock & Tomazelli 1995). The lagoonalsediment deposits started about 8 kyrs ago(Toldo et al. 2000). They consist mostly of mudor muddy sands, having an average thickness ofabout 6 m.According to Ramos (1977), the presentday vegetation of the coastal plain of the Rio Grandedo Sul state consists of the plant communities ofdry fields, humid depressions, flooded depressions,peat soils, flooded soils, freshwater marshes andcoastal subtropical forests. A great specific diversityof species of Poaceae and Cyperaceae characterizesthis region. The modern state of vegetation is aresult of the effect of natural factors during theHolocene and the anthropogenic impact, especiallyduring the last century.Sample collectionsA total of more than 100 samples werecollected from cores T-64, TBJ-02, B-2, situated inthe estuarine part of the Patos Lagoon, and fromcore FS-20, performed at the Cassino Beach region.Other 35 samples were collected from the cores FS-10 and T-14 performed at the bottom of theTramandaí Lagoon and adjacent area (Fig. 1).Samples represented by mud and muddy sandswere used for the palynomorph study. Samples,enriched by organic matter, were radiocarbon datedat Beta Analytic Inc., Florida, USA. Based onthe correlation of the core T-64 (a sample 140 cmdeep) with an adjacent core identical insedimentological and seismical characteristics(Toldo et al. 2000), we concluded an approximateage of 5,500-6,000 yrs BP.The chemical treatment of the samplesfollowed Faegri and Iversen (1975) using HCl (10%)and NaOH (5%). The use of HF was avoided inorder to preserve siliceous remains, such assilicoflagellate skeletons and diatoms. Separation ofinorganic and organic substances was carried outusing ZnCl 2 solution with a density of 2.2 g/cm 3 .Results and DiscussionThe principal palynomorphs, identified fromthe Holocene lagoon sediments, were pollen andspores of vascular terrestrial and aquatic plants, algalpalynomorphs (zygospores, coenobiums andcolonies of Chlorophyta, Prasinophycean phycomata,and Dynophyta cysts), fungal palynomorphs(ascospores, hypnodia, fruit bodies), palynoforaminifers,and scolecodonts. Besides, silicoflagellateskeletons of Dictyocha were found in some samples,which were corresponded to transgressive maximum.The microphotographs of the most frequentpalynomorphs, identified from the Holocene lagoonsamples in the coastal plain of the Rio Grande doSul state are shown in the plate.Our reconstructions of paleoclimate werebased on the ecological characteristics of allregistered and identified palynomorphs taxa. Onlythe most representative palynomorphs and theirecological characteristics were mentioned forpalaeoclimate reconstructions in the Table I. Thesepalynomorphs were zygospores, coenobiums andcolonies of Chlorophyta, whose identifications weremade according to literature (van Geel 1976, vanGeel & van der Hammen 1978, Jankovská &Komárek 2000). They were from freshwaterenvironments, and may also be evidence offreshwater input into saline aquatic environmentsduring pluvial periods (Medeanic et al. 2003,Medeanic 2006). Besides, dinoflagellate cysts andPrasinophycean phycomata were found, which areindicators of marine environments or sea waterinfluence by tides and/or marine transgressions(Dale 1976, Hoek et al. 1995, Grill & Quattroccio,1996).The palynoforaminifers represent chitinousinner tests of different benthic and planktonicforaminifers which are widely spread in the oceansand in the seas (van Veen 1957, Pantic &Bajaktarevic 1988). The informal classification ofpalynoforaminifers was based on morphology,including number of chambers and the types ofchambers arrangement. Biological affinities of thedifferent morphological types of palynoforaminifershave not been established yet. Chitinous fungalPan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

290S. MEDEANIC & I. C. S. CORREATable I. Indicative values of palynomorphs encountered in the Holocene lagoon sediments in the coastal plain of RioGrande do Sul for paleoclimatic reconstructions.Palynomorphs Distribution Palaeoclimatic implicationsPollen and spores of terrestrialand aquatic vascular plantsMICROALGAEDinoflagellate cystsPrasinophycean phycomataChlorophytaBotryococcusSpirogyra, Pediastrum,Zygnema, MougeotiaPseudoschizaeaFUNGIGlomusTetraploaHyphaeSCOLECODONTSPALYNOFORAMINIFERSAll ecosystems of the coastalplainsThe oceans and the seasThe oceans and the seasFreshwater-brackish waterenvironmentsFreshwater fluvial, lacustrineenvironmentsGrow at the edge of streamsand in pondsPrincipal plant in the coastalplain that fix and supportdunesSporadically in salt andbrackish-water marshesNumerous hyphae in organicrich sedimentsSalt marshes and beachesOceans and seasIncrease in frequency, abundance and taxonomicvariety from the lagoon samples indicate onaugmentation of fresh-water input into lagoons (morehumid climate)Increasing of marine water influence(transgressions) occurred as a results of temperatureriseIncreasing of marine water influence(transgressions) occurred as a results of temperatureriseBrackish-water lagoon environments, increasingin the past was connected with drier climateSignificant freshwater input into lagoons mayserve as indicator of humidity increaseSignificant freshwater input into lagoons mayserve as indicator for humidity increaseIncrease in frequency in lagoon sediments maybe indicative for the shallow lagoons and an proximityof the coast indicative for drier climateSalt and brackish-water marshes and mangrovesIndicative of humid and warm climate,freshwater input during pluvial periods and transportby riversIndicators of shallow water basins, the beaches,showing the coast proximity to the lagoons occurredwhen climate changes became drierSea level rise (transgressions) caused bytemperature increasepalynomorphs, resistant to destruction and importantfor palaeoenvironmental reconstructions have beenrecently reported (Jarsen & Elsik 1986, van Geel &Aptroot 2006).Use of fungal palynomorphs for the elevatedsalinity environments of the coastal plains, spreadinthe Quaternary, has not been sufficientlyelaborated yet. The abundance and taxonomicvariability of mycorrhizic Glomus and itsimportance for dune stabilization was shown byCordazzo & Stümer (2007), who studied Glomus inthe roots of Panicum racemosum, a species that fixesand supports dunes. Limaye et al. (2007) registeredan increase in Glomus in the sediments, formedduring the Late Glacial Period of the Pleistocene inIndia, related to active erosion processes under a dryand continental climate. Fungal palynomorphs ofTetraploa are common in the salt and brackish-watermarshes of coastal plains (Medeanic et al. 2001,Limaye et al. 2007).Figure 1. Map of the coastal plain of the Rio Grande doSul State, Brazil, with the location of the core-drillingsites.Pan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

Climatic changes in the coastal plain of the Rio Grande do Sul291Scolecodonts are the jaws of polychaeteannelids. They are fossilized due to their chitinousteeth and dwelling tubes. Representative amounts ofscolecodonts are important indicators of sedimentdeposition near the shelf or near the beaches(Limaye et al. 2007). Silicoflagellate skeletons ofDictyocha are important marine indicator for thelagoon sediments subjected by the marine influencewhen temperatures were higher than in present(McCarthny & Loper 1989, Medeanic & Corrêa2007).For the reconstructions of palaeoclimatechanges occurred in the different periods of theHolocene, based on palynomorphs, the absolute agedata of some samples on 14 C were used (Table 2).Present day, using both absolute age data, andpalynomorphs records, we can detect some periodsof distinctive climatic changes.Early HoloceneSome samples whose ages correspond to9,620+/-160 and 9,400+/-140 yrs BP includedpalynomorphs, indicated on relatively humid andlower than present day temperatures. That time,freshwater marshes, sometimes subjected by seawaterentrance, were wide spread (Medeanic &Dillenburg 2001, Weschenfelder et al. 2008). Thinlayers (30-40 cm) of peat were formed under suchclimatic conditions. Then, 8,620+/-170 yrs BP, adrier climate caused a decrease in the areas offreshwater marshes and an increase of xerophylousand halophylous herbaceous plants. At the sametime, marine influence in the coast increased, whichwas revealed by marine palynomorphs (prasinophyceanphycomata, cysts of dinoflagellates, andpalynoforaminifers) appearance in the lagoonsediments.Middle HoloceneFurther results point to climatic and environmentalchanges in the coast during the MiddleHolocene. Palynomorphs from one sample, dated as7,840+/-140 yrs BP indicated a relatively humidclimate and sea level rise, influenced on notablespreading of the salt-and-brackish-water marshes.Ahead of marine influence, freshwater marshes wereinhabited by ferns, mesophylous and aquatic herbs.The obtained data delay with oscillate character ofclimate during the Middle Holocene. The period oftime since 7,570+/-150 and 7,370+/-150 yrs BP wascharacterized by dry and hot climate, resulted in lessdense vegetation cover, decrease in taxonomicvariety of plants. In the coast, dunes were morespread, and lagoon were subjected by sea waterentrance, ampli-tude of sea-level rise continuouslygrew (Medeanic et al. 2001, 2003, Weschenfelder etal. 2008). Maximum of sea water rise was detectedfrom the samples dated as 5,500-6,000 and 4,940+/-80 yrs BP. There was a notable increase in marinealgae palynomorphs and Dictyocha skeletons (Fig.2). Frequency of freshwater algal palynomorphs(Spirogyra, Pediastrum, Zygnema, Mougeotia,Pseudoschizaea) was very low, indicating salinityincreasing in paleolagoon in the maximum sea-levelrise (Fig. 3). Predominance of Botryococcuscolonies in the paleolagoon was evident. Hot and dryclimate was the reason of relatively pure vegetationcover in the coastal plain – small frequency of pollenof aquatic vascular plants and arboreal pollen (Fig.4). Increase in humidity caused more spreading offerns, arboreal and herbaceous terrestrial and aquaticplants (Figs. 4-5). The final of marine transgressionwas a result of temperature lowering and climatedrying. Sea level fall led to regressive stage.Late HoloceneThat time interval corresponds to regressivestage of the Holocene. We have not yet radiocarbondata on absolute age of sediments from the differentdepths of the core drillings. But palynomorphs datafrom the lagoon sediments indicate that oscillateclimatic changes occurred after transgressive stage.Based on the evaluation of algal palynomorph taxafrequency and relation (%) between BotryococcusTable II. The cores, absolute age of studied sediments and interpretations of climate, based on palynomorphs.Cores14 C dating, (yrs BP) References Paleoclimatic reconstructionsFS-10 9620+/-160 Medeanic & Dillenburg 2001 Relatively humidB-2 9400+/-140 Weschenfelder et al. 2008 Relatively humidFS-10 8620+/-120 Medeanic & Dillenburg 2001 DrierFS-10 7840+/-140 Medeanic & Dillenburg 2001 More humidT-14 7570+/-120 Medeanic et al. 2003 Relatively dryTBJ-02 7370+/-150 Medeanic et al. 2001 Relatively dryB-2 7370+/-150 Weschenfelder et al. 2008 Relatively dryT-64 ~5500-6000 Medeanic et al. 2001 Higher temperature and humidityFS-20 4940+/-80 Clerot 2004 Higher temperature and humidityPan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

292S. MEDEANIC & I. C. S. CORREAFigure 2. Percentage palynodiagram of algal palynomorphs from the samples of the core T-64 (adapted from Medeanic& Corrêa 2008).Figure 3. Percentage palynodiagram of algal palynomorphs of Chlorophyta from the samples of the core T-14 (adaptedfrom Medeanic 2006).Pan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

Climatic changes in the coastal plain of the Rio Grande do Sul293Figure 4. Generalized percentage palynodiagram from the samples of the core-drilling T-64: I – pollen and spores ofterrestrial vascular plants, II – pollen and spores of aquatic plants, III – pollen of Poaceae, IV – palynomorphs ofChlorophyta, V – marine algal palynomorphs, T – transgressive stage, R – regressive stage.Figure 5. Percentage palynodiagram of palynomorphs counted from the samples of the core T-14: AP – arboreal pollen,NAP – non-arboreal pollen, T – transgressive stage, R – regressive stage, 1 – mud, 2 – sand, 3 – more humid climate, 4– drier climate (adopted from Medeanic et al. 2003).Pan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

294S. MEDEANIC & I. C. S. CORREA(indicative for drier climate) and Pediastrum(pointing to an increase in humidity) from lagoonsediments of the core drilling T-64, we concludeddifferent sub/stages of climatic changes duringregressive stage: drier-more humid-drier-morehumid, drier and humid (Fig. 3). In the lagoonsediments formed in the periods of drier climate, agreat number of scolecodonts and various fungalpalynomorphs were recorded, indicating a lagoonshallowing and on the coast (beach) proximity.Plate captions: 1,2 – Prasinophycean phycomata, 3 – Spiniferites, 4 – Dictyocha, 5,6 – Operculodinium, 7 – dinoflagellatecyst undetermined., 8 – Botryococcus, 9 – Mougeotia, 10, 11 – Spirogyra, 12,13 – Pseudoschizaea, 14 – Zygnema,15 – Pediastrum, 16 – Glomus, 17, 18 – Tetraploa, 19 – microforaminifera, 20,21 – scolecodonts. Scale: 10 µm.Pan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

Climatic changes in the coastal plain of the Rio Grande do Sul295Samples, corresponding to more humid period hadmore freshwater algal palynomorphs (Zygnema,Mougeotia, Spirogyra, Pediastrum, Pseudoschizaea),which were carried to paleolagoons by fluvialfresh-water flows (input) when it was heavy raining.The same sequence of climatic changesbased on analysis of increase-decrease frequenciesof arboreal pollen (AP), pollen of aquatic herbs(NAP aquatic), and spores of ferns is shown inFigures 4 and 5. The end of the regressive stage(approximately last two thousand years) ischaracterized by notable increase in humidity,reflected in increasing in the lagoon sediments offreshwater algal palynomorphs and NAP aquatic.The more dense vegetation cover in dunes, saltbrackishwater marshes, and arboreal and shrubswere reconstructed.ConclusionsIn this paper, we tried to show some timeintervals of climatic changes occurred in the coastalplain of the Rio Grande do Sul state during the last10 kyrs BP based on dated by 14 C samples andpalynomorphs study from different core-drillings.ReferencesAngulo, R. J. & Lessa, G. C. 1997. The Brazilian sealevel curves: a critical review with emphasison the curves from the Paranaguá and Cananéiaregion. Marine Geology, 140: 141–161.Bradley, R. S. 1999. Paleoclimatology. ReconstructingClimates of the Quaternary. Secondedition. Institute. Geophysics series, 68.Elsevier Academic Press, Amsterdam, 631 p.Bradshaw, R. H. W. 1994. Quaternary terrestrialsediments and spatial scale: the limits to interpretation.Pp. 239-252. In: Traverse, A. (Ed.).Sedimentation of Organic particles. CambridgeUniversity Press, Cambridge, 544 p.Clerot, L. C. P. 2004. Estudo da Barreira IV naregião do Cassino, Rio Grande do Sul - RS.Evolução e caracterização como reservatório.Graduate course, Universidade Federal doRio Grande do Sul, Brasil, 83 p.Colbath, G. K. & Grenfell, H. R. 1995. Review ofbiological affinities of palaeozoic acidresistant,organic walled eukaryotic algalmicrofossils, including .acritarchs. Review ofPalaeobotany and Palynology, 86: 287-314.Cordazzo, C. V. & Stümer, S. L. 2007. Ocorrência defungos micorrízicos arbusculares em Panicumracemosum (P. Beav.) Spreng (Poaceae) emdunas costeiras do extremo sul do Brasil.Atlântica, Rio Grande, 29(1): 65–68.Cordeiro, S. H. & Lorscheitter, M. L. 1994.Indicative value of some palynomorphs allowedpaleoreconstructions of the relative climate changes,sea level oscillations, and coastal palaeoenvironmentalchanges.Currently, we have a limited number ofprofiles conducted on the coastal plain of the RioGrande do Sul and only a few data on radiocarbonabsolute age. Therefore, our conclusions regardingclimatic changes that occurred in this region duringthe last 10 kyrs BP are general and preliminary.Nevertheless, our obtained results clearly show thatpalynomorphs are important indicators of climatechange and could be use to predict future scenariosbased on the climatic change periodicity (morehumid – drier, lower temperature – higher temperature)revealed in the present study.AcknowledgementsThis study was supported by CNPq(Conselho Nacional de Desenvolvimento Científicoe Tecnológico), grants no 300001/2008-8 and303956/2006-2. We are grateful for anonymousreviewers for their useful corrections and suggestionsthat helped improving this manuscript.Palynology of Lagoa dos Patos sediments, RioGrande do Sul, Brasil. Journal of Paleolimnology,10: 35–42.Corrêa, I. C. S., Martins, L. R. S., Ketzer, J. M.M.,Elias, A. R. D. & Martins, R. 1996. Evoluçãosedimentológica e paleogeográfica da plataformacontinental sul e sudeste do Brasil.Notas Técnicas, 9: 51–61.Costa, C. S. B., Seeliger, U., Oliveira, C. P. L. &Mazo, A. M. M. 1997. Distribuição, funções evalores das marismas e pradarias submersasno estuário da Lagoa dos Patos (RS, Brasil).Atlântica, 19: 67–85.Dale, B. 1976. Cyst formation, sedimentation andpreservation: factors affecting Dinoflagellateassemblages in recent sediments fromTrondheims fjord, Norway. Review of Palaeobotanyand Palynology, 22(1): 39–60.Faegri, K. & Iversen, J. 1975. Text-book of pollenanalysis. Scientific Publications. Amsterdam,295p.Grill, S. C. & Quatroccio, M. E. 1996. Fluctuacioneseustáticas durante el Holoceno a partir deregistro de paleomicroplancton: arroyoNaposta Grande, sur de la provincia deBuenos Aires. Ameghiniana, 33(4): 435–442.Jankovská, V. & Komárek, J. 2000. Indicative valueof Pediastrum and other coccal green algae inPalaeoecology. Folia Geobotanica, 5: 59–82.Pan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

296S. MEDEANIC & I. C. S. CORREAJarsen, D. M. & Elsik, W. C. 1986. Fungalpalynomorphs recovered from recent riverdeposits, Lungwa Valley, Zambia. Palynology,10: 35–60.Hoek, C., Mann, D. G. & Jahns, H. M. 1995. Algae:An introduction to Phycology. CambridgeUniversity Press, Cambridge, XIV, 623 p.Komárek, J. & Jankovská, V. 2001. Review of theGreen algal Genus Pediastrum: Implicationfor Pollen-analytical research. BibliotecaPhycológica, Berlin-Stuttgart, 108, 127 p.Ledru, M. P., Salgado-Labouriau, M. L. &Lorscheitter, M. L. 1998. Vegetation dynamicsin Southern and Central Brasil duringthe last 10.000 yr BP. Review of Palaeobotanyand Palynology, 99(2): 131–142.Limaye, R. B., Kumaran, K. P. N., Nair, K. M. &Padmaiai, D. 2007. Non-pollen palynomorphsas potential palaeoenvironmental indicators inthe Late Quaternary sediments of the westcoast of India. Current Science, 92(10):1370–1382.Lorscheitter, M. L. 1983. Evidences of seaoscillations of Late Quaternary in Rio Grandedo Sul, Brazil, provided by palynologicalstudies. Quaternary of South America andAntarctic Peninsula, 1: 53–60.Lorscheitter, M. L. & Dillenburg, S. R. 1998.Holocene palaeoenvironments of the northerncoastal plain of Rio Grande do Sul, Brazil,reconstructed from palynology of Tramandailagoon sediments. Quaternary of SouthAmerica and Antarctic Peninsula, 11: 73–97.Martin, F. 1993. Acritarchs: A review. BiologicalReview, 68: 475–538.Martin, L., Dominguez, J. M. L. & Bittencourt, A. C.S. P. 2003. Fluctuating Holocene Sea Level inEastern and Southeastern Brazil: Evidencefrom Multiple Fossil and GeometricIndicators. Journal of Coastal Research,19(1): 101–124.McCartney, K. & Loper, D. E. 1989. Optimizedskeletal morphologies of silicoflagellategenera Dictyocha and Distephanus. Paleobiology,5: 283–298.Medeanic, S. 2006. Freshwater algal palynomorphrecords from the Holocene deposits in thecoastal plain of Rio Grande do Sul. Reviewof Palaeobotany and Palynology, 141(1-2):83–101.Medeanic, S. & Corrêa, I. C. S. 2007. SilicoflagellateDictyocha Ehrenberg from the MiddleHolocene sediments in the coastal Plain ofRio Grande do Sul, Brazil. Revista Espanõlade Micropaleontologia, 39(3): 73–86.Medeanic, S. & Dillenburg, S. R. 2001. The EarlyHolocene palaeoenvironment history of theTramandaí Lagoon (RS, Brazil). I Congressodo Quaternário de Paises de LínguasIbéricas, Lisboa, 402–405.Medeanic, S., Dillenburg, S. R. & Toldo Jr., E. E.2001. Novos dados palinológicos datransgressão marinha pós-glacial emsedimentos da Laguna dos Patos. RevistaUniversidade Guarulhos, 6(6): 64–76.Medeanic, S., Jankovská, V. & Dillenburg, S. R.2003. The implications of green algae(Chlorophyta) for palaeoecological reconstructionof the Holocene lagoon system in theTramandai Lagoon region, Rio Grande do Sul,Brazil. Acta Palaeobotanica, 43: 113–123.Medeanic, S., Torgan, L. C., Clerot, L. C. P. & Bahidos Santos, C. 2009. Holocene marinetransgression in the coastal plain of RioGrande do Sul, Brazil: Palynomorph andDiatom evidence. Journal of CoastalResearch, 25(1): 224–233.Pantic, N. & Bajaktarevic, Z. 1988. Nannoforaminiferain palynological preparations andslides from Mesozoic and Tertiary deposits inCentral and Southeast Europe. Revue dePaléobiologie, Benthos-86, (2): 953–959.Ramos, R. F. 1977. Composição florística e ecologicado delta do Rio Tramandaí, Rio Grandedo Sul. Master’s thesis, Universidade Federaldo Rio Grande do Sul, Brazil, 131 p.Strother, P. K. 1996. Acritarchs. In: Jansonius, J. &McGregor, D. C. (Eds.). Palynology: Principlesand Applications. American Associationof Stratigraphic Palynologists Foundation,Dallas, 216 p.Tappan, H. 1980. The Palaeobiology of PlantProtista. Freedman Company, San Francisco,1028 p.Toldo Jr., E. E., Dillenburg, S. R., Corrêa, I. C. S.&Almeida, L. E. S. B. 2000. Holocenesedimentation in Lagoa dos Patos, Rio Grandedo Sul, Brazil. Journal of Coastal Research,16: 816–822.Traverse, A. 1988. Paleopalynology. UnwinHyman, Boston, 600 p.van Geel, B. 1976. Fossil spores of Zygnemataceae inditches of a prehistoric settlement in Hoogkarspel(The Netherlands). Review of Palaeobotanyand Palynology, 22: 337–344.van Geel, B. & Aptroot, A. 2006. Fossil ascomycetesin Quaternary deposits. Nova Hedwigia,82(3, 4): 313–329.van Geel, B., Boncke, S. J. P. & Dee, H. 1980/81. Apalaeoecological study of an Upper LatePan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

Climatic changes in the coastal plain of the Rio Grande do Sul297Glacial and Holocene sequence from “deBorchert”, the Netherlands. Review of Palaeobotanyand Palynology, 31: 367–448.van Geel, B. & van der Hammen, T. 1978.Zygnemataceae in Quaternary Colombiansediments. Review of Palaeobotany andPalynology, 25: 377–392.van Veen, F. R. 1957. Microforaminifera. Micropaleontology,3(1): 74–74.Vieira, J. P. & Rangel, S. R. S. 1988. PlanícieCosteira do Rio Grande do Sul: Geografiafísica, vegetação e dinâmica sóciodemográfica.Saagra, Porto Alegre, 256 p.Villwock, J. A. & Tomazelli, L. J. 1995. Geologiacosteira do Rio Grande do Sul. NotasTécnicas, 8: 1–45.Webb III, T. 1991. The spectrum of temporalclimatic variability: current estimates and theneed for global and regional time series. In:Global Changes of the Past (R.S. Bradley,ed.). Boulder University Corporation forAtmospheric Research: 61–82.Weschenfelder, J., Medeanic, S., Corrêa, I. C. S. &Aliotta, S. 2008. Holocene paleoinlet of theBojuru region, Lagoa dos Patos, SouthernBrazil. Journal of Coastal Research, 24(1):99–109.Zamora, N., Medeanic S. & Corrêa, I. C. S. 2007.Microforaminíferos como indicadores paleoambientais:estudo palinológico na região sulda Costa Rica, América Central. Gravel, 5:75–87.Received December 2009Accepted June 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 287-297

Representations in the Brazilian media of the impacts of climatechange in the coastal zoneDENIS HELLEBRANDT 1 & LUCENI HELLEBRANDT 21 University of East Anglia, School of International Development, Norwich, United Kingdom. Email:d.hellebrandt@uea.ac.uk2 Universidade Federal do Rio Grande, Centro de Estudos de Economia e Meio Ambiente, Rio Grande-RS, Brasil.Abstract. This study aims to examine texts which deal with climate change in the coastal zone, andlooks specifically at the coverage of Brazilian daily newspapers with national and regional circulation.The paper begins by introducing the study with a selective review of recent literature on climate changeand media, looking both at international and Brazilian research. The literature review also focus on theimpacts of climate change in the coastal zone and its links with policy, particularly in Latin America. Thetheoretical basis of the research is presented - social construction of meaning - with the questions whichguide the work and a summary of the methodology. The final sections show results, analysis andconclusions with particular comments on the frequency of the coverage at national and regional levels anda detailed look at elements of the news discourse in a specific case. The predominance of issues set by aninternational scientific and political agenda in the Brazilian media and relative absence of references tothe coastal setting on the national coverage point to the need of an urgent review of priorities in the masscommunication of scientific and environmental themes in Brazil.Keywords: newspapers, sea-level rise, social constructionResumo. Representações na mídia brasileira dos impactos das mudanças climáticas na zonacosteira. Esta pesquisa busca examinar textos que tratam especificamente do tema mudançasclimáticas em zonas costeiras, com interesse na cobertura de jornais diários brasileiros decirculação nacional e regional. Inicia-se com uma introdução do estudo através da revisão seletivada literatura recente sobre mídia e mudanças climáticas, focando estudos no âmbito internacional ebrasileiro. Em seguida são focados os impactos da mudança climática em zonas costeiras e sualigação com políticas públicas, particularmente no caso da América Latina. A base teórica –construção social do significado – é apresentada a seguir, assim como as perguntas da pesquisa eum sumário da metodologia. As seções finais mostram resultados, análise e conclusões,especificamente comentários sobre a frequência da cobertura em níveis nacional e regional, bemcomo um tratamento detalhado de elementos do discurso empregado pela mídia em um casoparticular. A predominância de tópicos na mídia brasileira que são determinados por uma agendacientífica e política internacional, somada a relativa ausência de referências ao contexto costeirona cobertura de jornais nacionais sugerem a necessidade urgente de revisão das prioridades nacomunicação de massa de temas científicos e ambientais no Brasil.Palavras-chave: jornais, aumento do nível do mar, construção socialIntroductionClimate change and mediaClimate change began to receive widespreadattention in the United States and United Kingdommedia by the late 1980s, triggered by a series ofevents ranging from scientists making their views onthe rapid global warming explicit for the first time,politicians reactions to the growing concern thathumans were implicated in the global climatechange and the creation of the United NationsIntergovernmental Panel on Climate Change, orIPCC (Carvalho & Burgess 2005, Boykoff &Roberts 2007). Since then, the coverage of climatechange by the international English-speaking mediahas steadily increased (Boykoff 2008), although ithas recently shown signs of a slight decrease(Revkin 2009). The trend of general increase incoverage is largely associated with either specificevents (such as the release of Al Gore’s film ThePan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Representation of climate change in the media299Inconvenient Truth, as well as his shared Nobelprize with the IPCC) or large scale disasters, such asthe effects of Hurricane Katrina in the US (Boykoff& Roberts 2007, Boykoff & Roberts 2008). In spiteof a tendency for improvement in terms of richercontextualisation and broader coverage of differentaspects of the issue – for example, relatively fewerattention to “climate skeptics” and greater space forthe political dimension of climate change - theoverall quality of the coverage has been criticised(Carvalho 2007, Boykoff 2008).Research has drawn attention to the howideology and politics can determine the framing ofnews on climate change (Smith 2005, Carvalho2007), specifically highlighting how supposedlyobjective scientific views can be given differentmeanings according to political interests. Carvalho(2007: 240) suggests a “politicised reading ofscience reports in the press” (italics as in theoriginal), adding that “audiences could engage in amore active interpretation of representations ofknowledge in the media and in a criticalunderstanding of their implications”. Boykoff &Boykoff (2007) examine specifically how thenormative dimension of journalism has interferedwith the communication of human contribution toclimate change. The authors mention how political,economic and professional norms are interrelated,and show how journalistic (professional) norms arethe main factor causing the failure by the media tocommunicate “central messages” about theanthropogenic causes of climate change, andconclude that this “occurs not only because ofcomplex macro-political and economic reasonsrooted in power relations, but also, in part, becauseof the micro-processes that undergird journalism”(Boykoff & Boykoff 2007).As the Brazilian media is concerned, thecoverage of climate change issues has shown anincrease similar to that recorded in the Englishspeakingmedia, and the general pattern of peaks incoverage related to specific events was alsoidentified (ANDI, 2009). However, research on theBrazilian media representations of climate changeare scarce, as noted by the authors of theaforementioned study, which quotes the almost totallack of records in the Scielo database (ANDI, 2009).Likewise, in spite of looking at a broader searchbase in our own review of the Brazilian literature onthe theme, we were able to find only relatively fewstudies (Lavezzo Filho & Nunes 2004, Costa &Lages 2008, Silva Junior & Bortotti 2009). Only bywidening the scope are we able to find relevant,even if indirect, references on communication ofscientific knowledge to the general public. Forinstance, the survey on public perception of sciencecarried out by the Ministry of Science andTechnology in 2006 (MCT 2007) showed that asizeable portion of the Brazilian public is interestedin scientific issues (41% of 2004 respondents).Furthermore, the survey also showed how the publicconsidered journalists to be the most trustworthysources of information (27%), well above scientistsworking at state universities (17%) (MCT 2007).These findings suggest that there is clearly potentialpublic interest in a theme such as climate change inBrazil, and that coverage of the theme by the mediaseems to play an important role in how peopleconstruct the meaning of related issues - forexample, impacts of climate change on coastal areas.Climate change in the coastal zone – science andpolicyNicholls et al. (2007) draw attention to howthe understanding of the implications of climatechange on coasts has improved and summarise thefindings in their assessment as what they called“important policy relevant-messages” (Table I). Thereport clearly stresses the high cost of the impacts ofclimate change, both when short-term events (e.g.floods) and long-term process (e.g. sea-level rise)are considered. It also highlights the need forpolicies which are able to anticipate and respond toevents and processes at both time scales. The generalaim of such synthesis, or “executive summary” as itis called, is to communicate scientific findings to theactors involved in the policy-making processes,above all those making the decisions. Therefore, byfollowing how these “messages” (as they arereferred to by the authors themselves) circulateamongst the general public it is possible to assesseffectiveness of the communication of scientificknowledge and policy advice.There is a well-developed body of researchwhich investigates the vulnerability and adaptationof coastal communities to climate change (Allison etal. 2009). This literature highlights particularities ofclimate change impacts on human population incoastal settings. Firstly, poor people living in coastalzones show relatively higher exposure andsensitivity to impacts from climate change incomparison with well-off groups or those livinginland – for example, as reviewed by Daw et al.(2009) in the case of small-scale fisherfolk.Secondly, livelihoods based on coastalnatural resources are affected by composite impactsof climate change and social and economicprocesses (Glavovic & Boonzaier 2007). Forinstance, small-scale fisheries are affected not onlyby the ecological outcomes of rainfall variability orPan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

300D. HELLEBRANDT & L. HELLEBRANDTTable I. Policy issues related to the impacts of climate change in the coastal zone, according to the assessment by theIPCC Working Group II on “Coastal Systems and Low-lying Areas” (Nicholls et al. 2007: 317). Issues selected have“very high” confidence levels according to the IPCC: experts have 9 out of 10 chances of being correct in theirpredictions or judgements. Statements in the “policy issues” column are direct quotes from Nicholls et al. (2007),page 317.Policy issuesHighlighted aspects“Coasts are experiencing the adverseconsequences of hazards related to climate and sealevel.”“Coasts will be exposed to increasing risks,including coastal erosion, over coming decades due toclimate change and sea-level rise.”“The impact of climate change on coasts isexacerbated by increased human-induced pressures.”Costs and loss of lives due to extreme events such asstorms and floods.Increase in floods and cyclones, coral mortality andloss of wetlands. Impacts on fisheries and sources offreshwater with “serious implications for the well-being ofsocieties.”Growth in human population, particularly, theincrease of settlements in coastal areas.changes in sea-surface temperature, but also by theinteraction of these physical processes with so-called"adaptive strategies" - for example, changes towatershed management, land use and economicdevelopment brought about by a move from fisheriesto aquaculture (Badjeck et al. 2010). Studies in thefield highlight the need for the integration of climatechange in policies and management regimes appliedto coastal areas, particularly in the cases wherelivelihoods of poorest people and ecologicalsustainability are at stake (Badjeck et al. 2010,McIlgorn et al. 2010).Thus, impacts on the coastal zone are theresult of the interaction of different multiplestressors - sources of undesirable change - climatechange being only one of them. It is argued thatpolicies may become a stressor in their own right ifpolicy-making does not consider the complexity ofthe coastal zone (Bunce et al. 2010). According tothis view, a fuller representation by the media ofclimate change affecting coastal areas would requirea treatment of both climate-related impacts andsocial and economic development. The contextuallisationof climate change by the media becomescrucial to its understanding as its impacts arecompounded by processes such as diverse asecosystem conservation, urbanisation, tourism andfisheries development (O’Brien & Leychenco 2000,Bunce et al. 2010).Climate change and public policy in the coastalzone – the Latin American contextThe most up-to-date national level Brazilianpolicy aimed at tackling climate change, the“National Action Plan” (Rosa 2009), is the result ofan extensive consultative process carried out by the“Brazilian Forum of Climatic Change”. This was aninitiative led by the Brazilian government includinggovernment officials, scientists, state companiesfrom the energy sector and international NGOs (Rosa2009: 45). The term “coast” is directly mentionedonly once in the synthesis of the “action plan”provided by Rosa (2009), and there are no otherindirect references to coastal or marine areas - theclosest reference is a mention to “river banks”, yetspecifically in the context of reforestation. Coastalzone figures in the “action plan” in the context ofdata generation - “Instalment of systems to collectdata on the sea level on the Brazilian Coast” - underone of the three major components of the plan“Vulnerability and Transversal Actions” (Rosa2009: 47). This is in clear contrast with the attentionpaid by the “action plan” to forests, with repeatedquotes in several sections, focussing on the need tocontrol deforestation as a means to curb Brazilianemissions of greenhouse gases (Rosa 2009).The apparent lack of attention to coastalissues in a high level policy plan is not easilyexplained, as the relevance of climate change tocoastal populations is well established in policy andacademic circles (Nicholls et al. 2007, among othersaforementioned).In the case of Latin America, for example,governmental policy originally aimed at improvingliving conditions of coastal urban settlements hasbeen found to have the opposite effect, increasingthe vulnerability of human population. In a review ofpolicies applied in Buenos Aires, Murgida &Natenzon (2009: 149) highlight a program aimed atproviding housing “for [the] underprivilegedpopulation” which did not consider “scenariosrelated to climate change and variability, [neither]the flood [...] recurrence maps” for Buenos Aires.The authors conclude that this is an example of how“the lack of an integrated vision” betweengovernment, urban planners and other sectors maycause “maladaptation, as well as an [increase in] riskand new aspects of vulnerability”.Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Representation of climate change in the media301Similarly, Romero et al. (2009: 225) foundthat “[the] lack of urban planning and managementare allowing [for careless] watershed urbanization”,with particularly negative impact in the case ofcoastal cities with “complex topography”, such asValparaiso (Chile). The increased risk of floods andlandslides in coastal areas caused by extremeweather events is frequently highlighted in theliterature (Romero et al. 2009, Sant’Anna Neto &Roseghini 2009). For instance, findings from a studyby Sant’Anna Neto & Roseghini (2009: 244) suggestthat the seasonal pattern of rainfall in the northerncoast of São Paulo (Brazil) “presents a greatpotential to cause perturbations and reach a greaternumber of victims”, because rainfall tends to bemore intense and concentrated in short periods oftime when a large number of people to movetemporarily into the region for the tourism season.Summing up, the review of the literatureshows that, firstly, there is indisputable evidence ofhow climate change can have direct and specificeffects in the coastal zone. Secondly, researchcarried out in diverse contexts points to commonproblems related to the complex interaction betweenphysical, social and economic processes in coastalareas. Finally, another concern shared by severalstudies is the need for improvement in publicpolicies aimed at coastal development, whichcurrently do not seem to effectively take climatechange issues into account.These considerations indicate how highlydiverse factors may affect the ways in which themedia represents climate change impacts on thecoast. As a result, we argue that an attempt tounderstand this process of representation requiresboth an appraisal of inherent features of scientificjournalism (and the media sector as whole) and thecomplex nature of environmental and politicaldimensions of climate change.Material and methodsTheoretical framework - Media and representationThis research is based in a theoreticalframework which understands risk as being sociallyconstructed (Dake 1992, Beck 1992). By adoptingthis stance we assume that the perception of riskamong social actors will depend on theirbackground, the context in which they live, theaccess to sources of information and ability toarticulate responses to phenomena, among otherelements which directly or indirectly contribute tothe process of social construction of meaning. It isimportant to note that such perspective acknowledgesthe material existence of the natural andhuman phenomena, e.g. storms and floods, isindependent of what humans think about them. Onthe other hand, the meaning that these same stormsand floods have to us, for example, as causes ofdeath and material loss, is the result of the complexinterplay between individual perception and socialrelationships (Johnson 1986).Such considerations are relevant for thisresearch precisely because they imply that theunderstanding of perception of phenomena is onlypossible if we look at the cultural aspects underlyingit. In the specific case of this research it means thatif we wish to examine how the media represents acomplex set of phenomena such as climate change,we need to look at who are the actors involved -media, scientists, politicians – and what are therelationships between them and their connection tothe phenomena in question (Beck 1992). It alsoimplies that all actors in this process have the abilityto interpret phenomena and, consequently, tointerfere in the way they are given meaning.Therefore, the journalist is not seen as objective andimpartial messenger, neither the public as passivereceptor of news.More specifically, we adopt the concept of“cultural circuit”, which has been applied to studiesof media representation of environmental issues(Burgess 1990), and particularly of climate change(Carvalho & Burgess 2005). In this model (Fig. 1),the production and circulation of texts is madepossible through encoding and decoding processes,which are determined by specific contexts. Texts,understood as content in any form, are produced bythe media (encoded according to diverse norms andcriteria – visual, linguistic, professional, economic,institutional, etc.), and are then circulated in the“public sphere”. The readers will then consumethese texts, decoding them in the “private sphere”(Habermas 1989). In the following stage, readers notonly decode the texts, but also provide feedback andinput for another cycle of production.The construction of meaning during the productionstage is determined by a professional context inwhich technology and institutions play a major role.Meanwhile, in the consumption of texts meaning isconstructed in a personal context, in whichsubjectivity and everyday actions prevail (Carvalho& Burgess 2005). The assumption of a reader who isable to actively select which and how issues arerelevant, interpret these issues and transform theirmeaning is also central to our approach (Burgess etal. 1991). It has been shown how not only readers,but all actors constantly revise and contest themeaning of environmental issues and their portrayalin the media (Smith 2005).Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

302D. HELLEBRANDT & L. HELLEBRANDTFigure 1. Analytical framework: “Circuits of culture” model, adapted from Carvalho and Burgess (2005). The newsproduction cycle consists of three phases (numbered arrows in the diagram): news are produced (1) and distributed(2) in the public sphere. Then, news are consumed (3) in the private sphere and subsequent cycles are originated (lightgrey arrows).Research questions and methodologyThis research project is a component of aBrazilian network aimed at the multi andinterdisciplinary study of climate change in thecoastal zone (“Rede CLIMA – INCT para MudançasClimáticas, Zonas Costeiras”). This project seeks toanswer the following research questions: (1) howjournalistic coverage contextualises the climatechange theme, especially in terms of social andeconomic processes, (2) how different actorsdetermine the framing of theme in the media, withspecial focus on the role of institutions and (3) howthe media articulates the discourse on factorsaffecting vulnerability and adaptive measures whichare particularly relevant in coastal zones (forexample, urbanisation and integrated coastalmanagement).The research aims to examine the coverageof daily newspapers at three levels – local (definedby the towns in the estuarine area of the PatosLagoon (Southern Brazil), state/regional (State ofRio Grande do Sul) and national – for a 12 monthperiod (second semester of 2008 and first semesterof 2009). Texts which deal specifically with climatechange in the coastal zone have been selected andorganised in a database. The texts will be exploredwith qualitative techniques and critical discourseanalysis (Bazeley 2007, Wodak & Meyer 2006). Atthe time of writing the research is part of an ongoingproject and its scope will be gradually expanded infollowing stages, when we expect to include othermedia, as well as extend the analysis to otherperiods. The choice of geographical context is due tothe very nature of the research, which is integratedwith another existing project (SACC-HD) in whichwe look at the perception of climate change andvulnerability among fisherfolk in the small-scalefisheries of the Patos Lagoon.Results and discussionAs mentioned above, the findings obtainedso far are the result of ongoing analysis and limitedto part of the total database of articles. Nonetheless,it was possible to identify patterns in the generalcoverage which are in line with those shown inthe literature. Moreover, we were also able to selecta case of particular relevance to the understandingof how the media discourse is developed withrespect to our specific research focus. Findingsfrom both quantitative analysis and in-depth casePan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Representation of climate change in the media303study are presented and discussed in the followingsections.The analysis was developed in two stages,defined on the one hand by the geographic scope ofthe newspapers, and on the other hand by aspecificity of the news coverage. To begin with,we analysed the coverage of climate change inits broadest sense, as done by the newspapers aimedat audiences in all Brazilian territory. Then, wenarrow our focus to target the coverage of climatechange issues only in the context of the coastalzone. Also, a newspaper primarily aimed at thesouthern Brazilian audiences is included. Bypresenting the analysis in the structure outlinedabove we intend to gradually introduce the theme tothe reader: first, with a general view of how theBrazilian media has handled the climate changeissue, then with a more localised and specificanalysis. The inclusion of the newspaper withregional scope at the national level would lead theanalysis of results in a different direction – it couldbe equally valid, but would not support the gradualinterpretation aimed in this paper, and would notbe helpful in terms of the comparison with otherstudies which look at national media coverage ofclimate change (e.g. ANDI 2009).Frequency and level of the coverageFirstly, we looked at texts containing theexpression “mudança climática” (climate change) inthe online editions of three of the Braziliannewspapers with largest readership and nationwidedistribution – “A Folha de Sao Paulo”, “O Estado deSao Paulo” e “O Globo” - during the first semesterof 2009. The number of texts selected was asfollows: “Folha”=279, “Estadao”=528 and “OGlobo”=423. The frequency of texts throughout theperiod is shown in the figure 2. There is a relativelysimilar pattern in the coverage of differentnewspapers, with a few pronounced peaks. Theincrease in coverage seems to be related to specificevents: in the end of January the surge in news wasrelated to several events: the World EconomicForum in Davos, a meeting of the IPCC held in SãoJosé dos Campos and, above all, to the new climatechange policy being implemented by the US underBarack Obama. In contrast, at end of March, thecoverage peaked in response to a single event: theWWF campaign called “The Planet Hour” (“A Horado Planeta”) in which major cities around the worldswitched off their lights to show support for energysaving policies and control of CO 2 emissions.Finally, in mid April, the US policies lead the newscoverage again when a change in the status of CO 2(which became to be regarded as pollutant)reinforces the move in policy towards more controlof emissions.These findings are corroborated by the trendobserved in other studies, where the newspaperreinforces the move in policy towards moreFigure 2. Frequency of articles containing the expression “climate change” first semester 2009.http://migre.me/38jIWPan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

304D. HELLEBRANDT & L. HELLEBRANDTcontrol of emissions.These findings are corroborated by the trendobserved in other studies, where the newspapercoverage of climate change is guided by specificevents (Boykoff & Roberts 2007), and shows aparticularly tendency by Brazilian newspapers toincrease its coverage of the theme in response tointernational events. A very similar tendency of anational coverage largely dictated by aninternational agenda was observed in 2007 in thestudy organised by ANDI (2009) in Brazil.Accordingly, Billet (2010) mentions the “stronginternational focus” of the news addressing climatechange in India between 2002 and 2007.In the next stage of analysis we narrow ourfocus, adding the regional level of news coverage(represented by the newspaper “Zero Hora”)and restricting the selection of texts to thosecontaining simultaneously the terms “mudançaclimática” (climate change) and “costa” (coast). Theresulting number of texts was as follows:“Folha”=14, “Estadao”=42, “O Globo”=22 and“ZH”=2. The frequency distribution of texts isshown in figure 3. A smaller number of texts wasexpected, but the extremely low count in theregional newspaper is somewhat surprising as itsreadership encompasses the whole state of RioGrande do Sul and tends to provide national as wellas regional news coverage, thus could have moreclosely followed the coverage given by nationalnewspapers - in 2008 “Zero Hora” circulatedapproximately 180,000 copies/day, compared to246,000 by “Estadao”; 281,000 by “O Globo” and311,000 by “Folha, the largest circulation in Brazil(ANJ 2010).It is worth mentioning that several termswith equivalent or closely related meaning wereused in the search and selection of texts: forexample, “aquecimento global” (global warming),“mudança global” (global change),“praia” (beach),“litoral” (litoral), etc. Despite the low quantity, theresults from the combination “mudança climática”(climate change) and “costa” (coast) yielded thelargest number of texts between all possible relevantterm matches.The paucity of articles dealing with climatechange in coastal settings is remarkable when oneconsiders the significance of climate-related risks inthese areas (Nichols 2007). The texts in this subsetfollow the general pattern observed above and didnot refer to the specific Brazilian coastal context.These findings help to put into perspective therelative absence of consideration for coastalprocesses in the national policies aimed at tacklingclimate change (Rosa 2009). The tendency inquestion also suggest that the mismatch betweenexisting public policies and the needs of vulnerablecoastal populations in Brazil an Latin America(Romero et al. 2009, Sant’Anna Neto & Roseghini2009, Murgida & Natenzon 2009) is unlikely to bechallenged by a mass media which responds toissues set by an agenda removed from localFigure 3. Frequency of articles containing the expression “climate change” and “coast” first semester 2009.Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Representation of climate change in the media305concerns. Our study indicates that not only thenational newspaper coverage of climate-relatedimpacts on the coast was arguably insufficient, but itwas practically non-existent in regional newspapers.Where the debate outlined in the previous paragraphis concerned, one group of occurrences seemedparticularly interesting, as it was the only occasionin which all four newspapers reported almostsimultaneously on the theme of climate change incoastal zone. Indeed, on closer examination itproved to be a revealing case, which we present indetail below.How (and why) we are told about sea level riseBetween 15 and 16 March 2009 the onlineeditions of the three papers analysed reported onexactly the same story: a rise in sea level threatenedto flood New York by 2100. The news originated ina scientific paper published online in the journalNature Geoscience, on that exact same day (15March 2009) (Yin et al. 2009). All three Brazilianpapers only reproduced stories written originally byinternational news agencies (Reuters and EFE), withalmost no differences in the content – with notableexception for the version presented by “Zero Hora”in which it says that the sea level “might increasefrom 36 to 51 cm”, while the text in the “Folha”article reads “will increase”.The path taken by the news in this case isrepresented in figure 4. It also shows the mainfeatures determining the context in each step ofnews production and consumption (in italics). Theblue arrows represent the steps followed by the textuntil the “final” reader. The grey arrows representthe possible feedback from the reader towards thebeginning of a new cycle of news production, andpotentially changing human behaviour andinfluencing phenomena. The grey cross at the centreof the diagram represents the potential interactionbetween different contexts. This representation ofhow news were produced is based on both empiricalevidence and our intepretation, which is in turnrelated to the analytical framework followed in thisresearch, as articulate in detail below.Each step in figure 4 can be clearlyidentified based on direct analysis and/or inferencesfrom the research paper and the news articles. Actorsand the relationships depicted in the sequence ofevents throughout the production of the news followfrom the analysis of the paper and articles: forexample, the intermediary role of international newsagencies is not assumed or extrapolated, it arisesfrom the simple annotation of sources in the openingof each news article. On the other hand, the featuresof the context in each step are not presented in thetexts themselves and are infered - though in allinstances presenting objective aspects of eachcontext. For example, the requirement of instrumentsfor gathering of information about naturalphenomena or the existence of deadlines and editingin newspapers production are well-known, concreteaspects of these settings.Figure 4. News circuit: stages of production and consumption of news in the story about sea level rise and flood threatto New York in 2100.Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

306D. HELLEBRANDT & L. HELLEBRANDTThe contexts are obviously more complex than ahandful of concepts can describe, and those shownin the diagram seek to add to the interpretation of thenews cycle without reference to excessivelytechnical or subjective descriptions which woulddetract from the analysis and add bias to theinterpretation of results.The cycle itself is related to a particularinterpretation of news production, as indicated abovein the section “Theoretical framework – mediarepresentation” (Material and Methods). Suchinterpretation follows from an understanding ofsociety which assumes the existence of a private anda public sphere (Habermas 1989). As this study isconcerned, each sphere is related to different butinterconnected contexts of production andconsumption of news; the production of news takesplace initially in the public sphere, whereprofessional, technological and institutional factorsdetermine the context – shown in figure 4 asacademic institutions, journals and news outlets. Forinstance, the role of the market-related aspects of thepublic sphere in determining the social constructionof news has been focus of media studies for morethan a decade – for a comprehensive review, seeGamson et al. (1992).In the following stages, news are consumedin the private sphere, where personal and subjectivefactors are predominant in defining the context – infigure 4 represented by the individuals who make upthe different groups of news readers – teachers,researchers, politicians, etc. These steps are notinsulated from each other, quite the opposite, thereare feedbacks and interferences, as well asindividual actors which can stride both spheres - asnoted, for example, by Doulton and Brown (2009) inthe case of academic researchers and nongovernmentalactors which can be involved inseveral stages of the production and consumption ofnews. These connections cannot be fully appreciatedat the level of analysis aimed at in this paper, neithercan they be fully depicted in a single diagram.Nonetheless, we seek to ackowledge their existence,and draw attention to their potential role in the newscycle production by representing them as greyarrows in figure 4.The latter aspect is particularly relevant to acritical perspective on how the media representsenvironmental issues. It connects direclty to whatCarvalho and Burgess (2005) call “diachronicmodel” (see Fig. 1), a explanation of how news areproduced and consumed which expressly aims toinclude changes to society through time. In thatperspective, the so-called “circuits of culture” arenot closed circuits, but a succession of interrelatedcycles which allow us to understand howconstruction of news changes along with changes insociety and culture. By adopting this view we maybe able to unravel the mechanisms which lead tolong-established patterns of news production, suchas the focus on extreme climatic events (Boykoff &Roberts 2007), as well as explain sudden changes tothe news agenda, as in the case of attractive newsheadlines offered by celebrities statements (Boykoff,2009). The diachronic model of circuits of culturehelps to clarify the roles of different actors and theexistence of feedbacks along construction of news,as the interaction of research community and mediaand the consequent evolving focus on climatechange thorugh 19 years of news coverage demonstratedby Burgess and Carvalho (2005).Some aspects of this case are illustrative ofhow the social construction of meaning affects thecommunication of scientific findings, as well as theirrelevance for policy. Firstly, the focus on the impactof sea level rise in New York seems to render thenews more attractive. The coverage in all threepapers suggests that editors and/or journalists fromindependent news companies, thus irrespective ofspecific professional contexts, judged the same newsto be relevant. As to what criteria they used, we canonly speculate, but it might be related to the fact thatthe text stresses the risk of floods and extremeevents, with views on possible disaster scenarios.The tendency to report on episodic events has beenidentified in the literature (Boykoff & Roberts,2007), which also points out to the loss of contextthat the excessive focus on the “disaster” aspect cancause to the reporting of environmental news.Furthermore, these findings suggest that despite theexistence of well-defined policy issues relatingcoasts and climate change in the Brazilian and LatinAmerican context contex (see introduction), themedia tend to ignore this connections in therepresentation of the theme. In this case, this is doneby literally reproducing news related to a foreigncontext. A similar detachment between news andpolicy has been noted in other studies, whichhighlight how the media focus on narrow issues indetriment of broader discussion of social and policyrelatedtopics (Hayes et al. 2007).Secondly, the reproduction of content frominternational news agencies might be related tofinancial and institutional aspects of journalismwhich are reason for concern in the specific caseof reporting environmental and scientific issues:these are not regarded as priority and often do notreceive funds or editorial support, with fewcorporations interested to maintain specialised staff(Hannigan, 2006, Gamson et al. 1992). As a result,Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Representation of climate change in the media307environmental news end up being covered withready-made content which allows the newscompanies to save on resources and personnel. Forthese reasons, it is reasonable to argue that we aremore likely to be informed about the impact of sealevel rise in New York, as told by indirect sources,than to hear an account of how the Brazilian coastcould be affected by climate change, told by a localjournalists with the ability to provide rich context tothe story.ConclusionsFindings from this study show that Braziliannewspapers tend represent the climate change themealong similar lines used by the English-speakingpress, that is, largely following high-profile events.The predominance of issues set by an internationalscientific and political agenda in the Brazilian mediaand relative absence of references to the coastalsetting on the national coverage point to the need ofa urgent review of priorities in the communication ofscientific and environmental themes in Brazil.Moreover, as it has also been noted in other studies,the coverage of climate change in Braziliannewspapers is overwhelmingly concentrated in a fewmajor vehicles of large circulation. The study byANDI (2009) showed how in comprehensive sampleof 50 newspapers only six – four general dailypapers and two specialised in economic isssues, allwith national circulation – published 37% of allnews articles on climate change issues between 2005and 2008 in Brazil. The authors repeated the study in2008 and found that 48% of the coverage of climaterelatednews was done by those same newspapers(ANDI, 2009). These trends represent a considerableReferencesAllison, E. H. Perry, A. L. Badjeck, M.C. Adger, W.N. Brown, K. Conway D. Halls, A. S. Pilling,G. M. Reynolds, J. D. Andrew, N. L., Dulvy,N. K. 2009. Climate change and fisheries: acomparative analysis of the relative vulnerabilityof 132 countries. Fish and Fisheries.10(2): 173–96ANDI: Agência de Notícias dos Direitos da Infância.2009. Mudanças Climáticas na ImprensaBrasileira. Uma análise comparativa de 50jornais nos períodos: Julho de 2005 a junhode 2007, Julho de 2007 a dezembro de 2008.Acessible at: http://migre.me/3aK8k.(Accessed in 04/30/2010).ANJ. 2010. Major Brazilian newspapers by dailycirculation. Accessible at http://migre.me/3aKbv. (Accessed in 04/30/2010).Badjeck, M. C. Allison, E. H. Halls, A. S. & Dulvy,obstacle to the advancement of a coverage whichreflects the existing knowledge of coastal processesand their relation to public policy. Science has beenable to approach the impacts of climate change oncoastal zones in a nuanced manner, acknowledgingits complexity and generating knowledge potentiallyapplicable to improving peoples lives in vulnerableareas. However, the mass media vehicles analysedseemed to represent the issues in a detached way,focussed on issues removed from Brazilian reality,overlooking both local problems and scientificexpertise. The concetration on topics which mustappeal to a broad national audience leaves littleroom for localised accounts of impacts, while aspecialised agenda may limit the coverage, in thecase of Brazilian newspapers, to economic issues.Such findings and interpretations reinforce therelevance of a critical perspective on the study of therepresentation of climate change impacts in theBrazilian media. Further, the coastal zone emergedas a clearly valid focus of this research effort, giventhe urgent need of concerted policies andaccompanying communication aimed at increasingthe chances of preventing disasters and implementingsuccessful adaptive measures for coastalhuman populations.AcknowledgementsLuceni Hellebrandt is funded by the CNPq(grant no. 381476/2009-0). This research is linked tothe project SACC-HD Climate change, oceanographicvariability and the artisanal fisheries in theSW Atlantic: a human dimension approach(CRN2076), funded by the Inter-american Institutefor Global Change Research – IAI.N. K. (2010), ‘Impacts of climate variabilityand change on fishery-based livelihoods’,Marine Policy. 34(3): 375-383.Bunce, M. Brown, K. & Rosendo, S. 2010, Policymisfits, climate change and cross-scalevulnerability in coastal Africa: howdevelopment projects undermine resilience.Environmental Science & Policy. 13(6):485-497.Bazeley, P. 2007. Qualitative data analysis withNVivo. London: Sage.Beck, U. 1992. Risk Society: Towards a NewModernity. London: Sage.Billet, S. 2010. Dividing climate change: globalwarming in the Indian mass media. ClimaticChange. 99:1–16Boykoff, M. 2008. The real swindle. Nature Reports- Climate Change. Nature. 2: 31-32Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

308D. HELLEBRANDT & L. HELLEBRANDTBoykoff, M. T. & Roberts, J. T. 2007. MediaCoverage of Climate Change: CurrentTrends, Strengths, Weaknesses. UNDPHuman Development Report Office. 54p.Boykoff, M. T., Boykoff, J. M., 2007. Climatechange and journalistic norms: A case-studyof US mass-media coverage. Geoforum. 38(6), 1190–1204.Boykoff, M. & Goodman, M. 2009. "ConspicuousRedemption: Promises and Perils of CelebrityInvolvement in Climate Change" Geoforum40: 395–406Burgess, J. 1990. The Production and Consumptionof Environmental Meanings in the MassMedia: A Research Agenda for the 1990s.Transactions of the Institute for BritishGeography. 15: 139-161.Burgess, J., Harrison, C. M. & Maiteny, P. 1991.Contested meanings: The consumption ofnews about nature conservation. Media,Culture and Society. 13, 499–519.Carvalho, A. & Burgess, J. 2005. Cultural circuits ofclimate change in UK broadsheet newspapers,1985-2003. Risk Analysis. 25(6): 1457-1469.Costa, L. M. & Lages, L. T. S. 2008. MudançasClimáticas e Desmatamento: uma análisedo discurso dos jornais Folha de São Pauloe O Globo no contexto da divulgação dorelatório final do IPCC em 2007,apresentado no IV Encontro Nacional daAnppas 4, 5 e 6 de junho de 2008 Brasília -DF – Brasil. Accessible at: http://migre.me/3aKTD (Accessed in 30/09/2009).Dake, K. 1992. Myths of nature-culture and thesocial construction of risk. Journal of SocialIssues, 48: 21–37.Daw, T. Adger, N. Brown, K. Badjeck M. C. 2009.Climate change and capture fisheries. In:Cochrane, K. , De Young, C. , Soto, D. editors.Climate change implications for fisheries andaquaculture: overview of current scientificknowledge. Fisheries technical paper 530.Rome, Italy: FAO. p. 95–135.Doulton, H. & Brown, K. 2010. Ten years to preventcatastrophe? Discourses of climate changeand international development in the UKpress. Global Environmental Change 19:191–202Gamson, W. A., Croteau, D., Hoynes, W. & Sasson,T. 1992. Media Images and the SocialConstruction of Reality. Annual Review ofSociology, 18: 373-393Glavovic, B. C. & Boonzaier, S. 2007. Confrontingcoastal poverty: Building sustainable coastallivelihoods in South Africa. Ocean & CoastalManagement. 50(1-2), 1 - 23.Habermas, J. 1989. The Structural Transformationof the Public Sphere. Cambridge: PolityPress.Hayes, M. Ross, I. E. Gasher, M. Gutstein, D. Dunn,J. R. & Hackett, R. A. 2007. Telling stories:News media, health literacy and public policyin Canada. Social Science & Medicine 64:1842–1852Hannigan, J. 2006. Environmental sociology.London: Routledge.Johnson, R. 1986. The story so far and furthertransformations? In D. Punter (Ed.),Introduction to Contemporary CulturalStudies (pp. 277–313). London: Longman.Lavezzo Filho, L. F. & Nunes, L. H. 2004.“Informações Climáticas e a Sociedade: opapel da mídia, com ênfase na revista Veja”resumo do XII Congresso Interno de IniciaçãoCientífica da Unicamp, 2004. UniversidadeEstadual de Campinas. Accessible at:http://migre.me/3aKLu (Accessed in30/04/2010).MCT. 2007. Percepção Pública da Ciência eTecnologia. Departamento de Popularização eDifusão da C & T, Secretaria de Ciência eTecnologia para Inclusão Social, Ministérioda Ciência e Tecnologia. 69p. Accessible at:http://migre.me/3aKPI (Accessed in30/04/2010).McIlgorm, A. Hanna, S. Knapp, G. Le Floc, H.P.Millerd, F. & Pan, M. L. 2010. How willclimate change alter fishery governance?Insights from seven international case studies.Marine Policy. 34(1), 170-177.Murgida, A. N. & Natenzon, C. E. 2009. Social"downscaling". A few reflections on adaptationin urban environments. p. 139-156. In:Dias, P. L. S., Ribeiro, W. C., Sant’AnnaNeto, J. L. & Zullo Jr., J (Eds.). 2009. Publicpolicy, mitigation and adaptation to climatechange in South America. Universidade deSão Paulo, Instituto de Estudos Avançados.280 p.Nicholls, R. J., Wong, P. P. Burkett, V. R.Codignotto, J. O. Hay, J. E. McLean, R. FRagoonaden, S. & Woodroffe, C.D. 2007:Coastal systems and low-lying areas. ClimateChange 2007: Impacts, Adaptation andVulnerability. Contribution of WorkingGroup II to the Fourth Assessment Report ofthe Intergovernmental Panel on ClimateChange, Parry, M. L. Canziani, O. F.Palutikof, J. P. Van der Linden, P. J. &Hanson, C. E. (Eds.), Cambridge UniversityPan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Representation of climate change in the media309Press, Cambridge, UK, 315-356.O’Brien, K. L. & Leichencko, R. M. 2000. Doubleexposure: assessing the impacts of climatechange within the context of economic globalization.Global Environmental Change,10: 221-232.Revkin, A. 2009. News Coverage of Climate Entering‘Trance’? Accessible at http://migre.me/3aLmv (Accessed in 30/04/2010).Romero, H., Molina, M. Moscoso, C. Sarricolea, P.& Vásquez, A. 2009. Climate change andurban sustainability of Chilean metropolitancities. p. 211-227 In: Dias, P. L. S. Ribeiro,W. C. Sant’Anna Neto, J. L. & Zullo Jr., J(Eds.). 2009. Public policy, mitigation andadaptation to climate change in SouthAmerica. Universidade de São Paulo,Instituto de Estudos Avançados. 280 p.Rosa, L. P. 2009. Brazil and climate change:proposal of a governmental plan of action. p.39-48. In: Dias, P. L. S. Ribeiro, W. C.Sant’Anna Neto, J. L. & Zullo Jr., J (Eds.).2009. Public policy, mitigation and adaptationto climate change in South America.Universidade de São Paulo, Instituto de EstudosAvançados. 280 p.Sant’Anna Neto, J. L. & Roseghini, W. F. F. 2009.extreme pluvial impacts and their socioenvironmentalrepercussion on the northernSão Paulo coast. p. 231-247. In: Dias, P. L. S.Ribeiro, W. C. Sant’Anna Neto, J. L. & ZulloJr., J (Eds.). 2009. Public policy, mitigationand adaptation to climate change in SouthAmerica. Universidade de São Paulo,Instituto de Estudos Avançados. 280 p.Silva Junior, F. & Bortotti, B. 2009. Ageograficidade e alguns eventos climáticosocorridos no município de Londrinarelatados pelo jornal Folha de Londrina noperíodo de 1996 a 2005. Publicado em 12ºEncontro de Geógrafos da América Latina, 3 a7 de abril de 2009 – Montevidéu – UY.Accessible at: http://egal2009.easyplanners.info/area07/7392_Silva_Junior_Francisco.doc(Accessed in 29/12/2010).Smith, J. 2005. Dangerous news: Media decisionmaking about climate change risk. RiskAnalysis, 25: 1471.Wodak, R. & Meyer, M. 2006. Methods of criticaldiscourse analysis. London: Sage.Yin, J. Schlesinger, M. E. & Stouffer, R. J. 2009.Model projections of rapid sea-level rise onthe northeast coast of the United States.Nature Geoscience. 2: 262-266.Received May 2010Accepted October 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 298-309

Differences between spatial patterns of climate variability and largemarine ecosystems in the western South AtlanticDOUGLAS FRANCISCO MARCOLINO GHERARDI 1 *, EDUARDO TAVARES PAES 1 , HELENACACHANHUK SOARES 1 , LUCIANO PONZI PEZZI 1,2 & MARY TOSHIE KAYANO 21 Divisão de Sensoriamento Remoto – INPE, Av. dos Astronautas, 1758, São José dos Campos, SP, 12227-010, Brazil.2 Centro de Previsão de Tempo e Estudos Climáticos – INPE*Corresponding author: douglas@dsr.inpe.brAbstract. Despite their importance for environmental management, the response of Large MarineEcosystems (LMEs) to climate changes is unlikely to be controlled by the ecological criteria usedto define them. This is because productivity and trophic relations are endmembers of a chain effectthat starts with physical processes not necessarily bounded by LMEs. Correlation fields werecalculated for climate indices and sea surface temperature anomalies (SSTA) for the southwestAtlantic to identify interannual correlation patterns. Significant correlations indicate that theinfluence of El Niño/Southern Oscillation (ENSO) along the north and east coasts of Brazil is notcoincident with the boundaries of LMEs. The antisymmetric (opposed signs) correlation pattern ofthe Tropical South Atlantic (TSA) observed in the South Brazil (SB) LME, during the warm PDOphase, may be related with the northeast-southeast SST dipole. It is possible that both the TSA andthe Antarctic Oscillation Index (AAO) have distinct influences on the Brazilian LMEs dependingon the geographic location and time scale. The PDO multidecadal and the ENSO interannualinduced SSTA variability along the Brazilian coast exhibit a complex spatial dynamics againstwhich ecosystem functioning should be tested to provide clues as to how the LMEs might respondto these climate forcings.Keywords: Ecosystems, Tropical South Atlantic, climate indices, global changesResumo. Diferenças entre os padrões espaciais de variabilidade climática e dos grandesecossistemas marinhos no oeste do Atlântico sul. A despeito de sua importância para o gerenciamentoambiental, é improvável que a resposta dos grandes ecossistemas marinhos (GEMs) às mudançasclimáticas seja regida pelos critérios ecológicos usados em sua definição. Isto acontece porque a suaprodutividade e relações tróficas são o resultado de uma cadeia de eventos que começa nos processosfísicos que não seguem, necessariamente, os contornos dos GEMs. Campos de correlação foramcalculados para índices climáticos e temperatura da superfície do mar (TSM) para o oeste do Atlântico sulpara a identificação dos padrões interanuais. Correlações significativas indicam que a influência do ElNiño Oscilação Sul (ENOS) ao longo da costa brasileira não é coincidente com os limites espaciais dosGEMs. Os padrões antisimétricos de correlação do índice do Atlântico Tropical Sul (ATS) no sul doBrasil durante a fase quente da Oscilação Decenal do Pacífico (ODP) pode estar relacionada com o dipolode TSM. É possível que tanto o ATS quanto a Oscilação Antártica (OA) apresentem infuências distintasnos GEMs do leste e do sul do Brasil. A variabilidade interanual da TSM ao longo dos GEMs no Brasil,induzida tanto pela ODP quanto pelo ENOS, exibe uma dinâmica espacial complexa contra a qual ofuncionamento dos ecossistemas devem ser testados.Palavras-chave: Atlântico Tropical Sul, índices climáticos, mudanças globaisIntroductionLarge Marine Ecosystems (LMEs) havebeen used as a framework for the assessment andmanagement of marine resources and they aredefined according to four ecological criteria:bathymetry, hydrography, productivity, andtrophically related populations (Sherman 1991). It isassumed that ecosystem variability in LMEs can becaptured by indicators grouped into five modules:Pan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

Climate variability and large marine ecosystems in the western South Atlantic311productivity, fish and fisheries, pollution andecosystem health, socioeconomics and governance(Duda & Sherman 2002). Recent efforts to relatelarge scale climatic changes to ecological processessuch as fisheries biomass yields in LMEs haveindicated the presence of emergent trends inducedby global warming (Sherman et al. 2009). It hasbeen suggested that increasing sea surfacetemperature (SST) exceeding the levels expectedfrom the warm phase of the Atlantic MultidecadalOscillation affected the Icelandic Shelf, NorwegianSea and Faroe Plateau, a region responsible for 5%of the global fisheries yields. The perceived increasein fisheries yields in these regions is a result ofincreased availability of zooplankton leading toimproved feeding conditions of zooplanktivorousspecies in the Northeast Atlantic. Similarly, fisheriesbiomass yields are increasing in the proposed NorthBrazil (NB) and East Brazil (EB) LMEs (Fig. 1),where changes in production seem to be respondingpredominantly to overexploitation rather thanclimate warming (Sherman et al. 2009).Figure 1. Map of South America showing the location ofLarge Marine Ecosystems in Brazil. 1) North BrazilLME, 2) East Brazil LME, 3) South Brazil LME.Integrative research tackling the influenceof large and mesoscale processes on biologicalsystems of the western south Atlantic has gainedmomentum in the last years. Variations in meanSST, cloud cover and turbidity in Bahia State,eastern Brazil, induced by the 1997-98 ENSOcaused partial mortality of octocorals andactiniarians, but had limited impact on scleractiniancommunities in coral reefs (Kelmo et al. 2003).ENSO is also known to have caused increasedprecipitation and reduced salinity of the PatosLagoon estuary (southern Brazil), driving awayeuryhaline species (Garcia et al. 2001). Remotesensing data show that cold-core eddies form at theBrazil-Malvinas Confluence (BMC) region andrapidly separate from the mean flow withchlorophyll concentration in their cores higher thanthe surrounding waters (Garcia et al. 2004). Thewestern boundary Brazil Current (BC) also produceswarm core rings that can be expelled to the shelf(Souza & Robinson 2004). The interannualvariability in the number of warm core rings shed inthe BMC may be forced by the AntarcticCircumpolar Current (Lentini et al. 2002), whichhelps propagating Pacific ENSO signals to theAtlantic ocean (Peterson & White 1998).Despite increasing evidences that point tothe importance of large geographical areas as targetunits for ecosystem-based management (Belkin2009), the response of these areas to climate changesis unlikely to be controlled by the ecological criteriaused to define the LMEs. The reason for that is verysimple, climate changes can act directly onphysiology, behaviour, mortality and distribution,and indirectly on productivity, structure andcomposition of the ecosystem (Brander 2007).Productivity and trophic relations are endmembersof a chain effect that starts with physical processesnot necessarily bounded by LMEs as they aredefined today. The response of the tropical Atlanticto climate variability depends on atmosphericteleconnection mechanisms and on basin-scale SSTgradients acting on different time scales (Lanzante1996, Enfield & Mayer 1997, Alexander et al. 2002,Giannini et al. 2004, Hastenrath 2006). Mechanismsinclude the upper-tropospheric Rossby-wave trainthat extends from the equatorial eastern Pacific tothe northern tropical Atlantic and the east-westdisplacement of the Walker circulation during ElNiño years (Hastenrath 1976, Kayano et al. 1988). Itis clear that the interplay of local dynamics andremote forcing, including the ENSO, is responsiblefor the observed SST anomalies over the tropicalAtlantic (Nobre & Shukla 1996).It is plausible to assume that if the spatialextent of SST anomalies can only partially affect aLME, the resulting changes in fisheries biomassyields or geographical distribution of species withinit may not be statistically detectable. As a result,unbiased assessment of climate change impactscould be hampered and national governments wouldPan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

312D.F.M GHERARDI ET ALLIface too many uncertainties regarding theeffectiveness of response policy measures. Also,managers responsible for mitigation programmesand action plans aimed at dealing with climatechange impacts on marine and coastal ecosystemsmay find it difficult to envisage the necessary sitespecificmanagement strategies for multiple stressors(Higgason & Brown 2009). This is an importantissue because the reduction in carrying capacity canbe coupled with density-dependence effects onbiomass changes of small pelagic fish species, suchas observed with the Japanese sardine (Yatsu et al.2008). As large-scale climatic-induced regime shiftsare modulated by local physical conditions, this willmost likely impose time-lagged changes onbiological production at lower trophic levels (e. g.mesozooplankton). Recen-tly, Gigliotti et al. (2010)showed that the interannual variability of eggconcentration of the Brazilian sardine can be relatedto the expansion and contraction of the spawninghabitat. The Brazilian sardine is capable of exploringsuitable spawning sites provided by the entrainmentof the colder and less saline South Atlantic CentralWater (SACW) onto the shelf due to the combinedeffect of coastal wind-driven and meander inducedupwelling.The purpose of this paper is to present someexploratory results that point to importantdifferences between the spatial patterns ofcorrelation of climate indices and SST anomalies(SSTA), and the geographic arrangement of LMEsfor Brazil (North Brazil, East Brazil, and SouthBrazil Shelf), as recently discussed in Sherman et al.(2009). The consequences of such differences to thestudy of the impacts of climate variability in theseLMEs are discussed.Materials and MethodsClimate indices calculated as SST anomalyaverages were obtained for three different areas:Niño 3 limited at 5° S, 5° N and 150° W, 90° W;Tropical North Atlantic (TNA) bounded at 5.5° N,23.5° N and 15° W 57.5° W; and Tropical SouthAtlantic (TSA) at 0°, 20° S and 10° E 30° W(Fig. 2). These indices are the same used in otherstudies of tropical Atlantic climate variability(Enfield et al. 1999, Kayano et al. 2009) and areavailable at http://www.esrl.noaa.gov/psd/. The SSTdata used in this work are the monthly gridded seriesfrom 1948 to 2008, with a spatial resolution of 2° inlatitude and longitude, derived from the version 3 ofthe reconstructed SST data set, described by Smithet al. (2008). These data can be freely downloadedfrom http://migre.me/3Hy49. The AntarcticOscillation Index (AAO), also known as theSouthern Hemisphere Annular Mode (Kidson 1988,Thompson & Wallace 2000) is calculated byprojecting the monthly mean 700 hPa geopotentialheight (normalized) anomalies poleward of 20°Sonto the leading Empirical Orthogonal Function(EOF) mode of these anomalies from 1979 to 2000.The AAO describes a mass seesaw between thesouthern mid and high latitudes, with positive(negative) values representing above (below) normalgeopotential height in the midlatitudes and below(above) normal geopotential height in the highlatitudes. The monthly AAO dataset corresponds tothe period from 1979 to 2007, available athttp://www.cpc.noaa.gov/products/precip/CWlink/daily_ao_index/aao/aao.shtml.Figure 2. Location of areas from which the climate indices have been calculated.Pan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

Climate variability and large marine ecosystems in the western South Atlantic313In order to determine the spatial patterns ofinterannual SSTA variability along the Braziliancoast associated with global climate change,correlations between Niño 3, TNA, TSA and AAOindices and the SST anomaly field were calculatedfor the area between 10ºN to 40ºS and 62ºW to26ºW. The influence of the Pacific DecadalOscillation (PDO) shift on correlations wasinvestigated by dividing the complete time series(1948 to 2008) in the cold PDO phase from 1948 to1976, and the warm PDO phase from 1977 to 2008(Mantua et al. 1997). Correlations were carried outfor each grid point in the study area using linearlydetrended, standardized and filtered data. Filteringprocedure made use of a Morlet wavelet as abandpass filter (Torrence & Compo 1998) to retainonly the interannual variability between 2 and 7years. The cross correlation time lags used in theanalyses apply to all grid points and were selectedbased on the higher significance value obtained foreach climate index. The statistical significance of allcorrelations has been assessed by Student’s t-test ata 95% confidence level and only significantcorrelations are presented in the results section. Thenumber of degrees of freedom (DOF) wasdetermined by dividing the total time length of theseries by the time lag needed to achievedecorrelation time closest to zero (Servain et al.2000, Kayano et al. 2009). Only the lower values forthe number of DOFs were adopted, so that the test isthe most severe.ResultsFor the sake of simplicity, all correlationsbetween the climate indices and SSTAs along theBrazilian coast will be referred to only in terms ofthe indices used in each case. Maximum positivecorrelations of 0.7 with the Niño 3 are found alongthe eastern coast and offshore the northern coast ofBrazil and to the north of the equator after a time lagof eight months (Fig. 3). This time lag has been alsoreported by Lanzante (1996) and indicates that underan El Niño (La Niña) the surface waters in theseareas of the tropical Atlantic are anomalouslywarmed (cooled) eight months later. Possibly, themost striking aspect of the correlation fields is themarked spatial differences between the cold andwarm PDO phases, namely the lack of positivecorrelation in the South Brazil (SB) LME during thewarm PDO. Positive correlations with values up to0.6 appear at the SB LME and up to 0.7 offshore forthe cold phase only. It is also important to note thatduring the warm PDO, correlations in the EB LMEare mostly located in its southern half, characterizedby high (up to 0.7) values. This suggests a separationbetween the two halves of the EB LME, in terms ofthe decadal SSTA variability. It is worth noting thatto the north of the equator correlations aresignificantly lower for the warm phase of the PDOthan for the cold phase.The TNA achieves higher positivecorrelations after one month lag but has a limitedimpact on the SSTA along the Brazilian coast (Fig.3). During the cold phase of the PDO the northBrazil coast experienced the highest correlations, butthese are greatly reduced moving offshore in thefollowing warm phase. In fact, there is no significantcorrelation for the TNA along the north Braziliancoast during the warm phase. On the other hand,significant correlation of 0.5, restricted to a smallarea in the eastern coast in the cold phase, evolvesinto a wide northwest-southeast correlation band.This extends towards the subtropical portion of thecentral south Atlantic in the subsequent warm phase.So, for the warm PDO, an anomalous warmed(cooled) TNA relates to an anomalous warmed(cooled) subtropical South Atlantic.Not surprisingly, the TSA achieves thehighest correlation (with zero lag) in the northernand eastern portions of Brazilian coast, with thelatter being more developed during the warm PDOphase (Fig. 3). This appears to be the result of theproximity with the area of the tropical Atlanticwhere the TSA is calculated. Furthermore, thepositive correlation pattern also resembles the SSTequatorial mode previously detected by Zebiak(1993) and Wagner & da Silva (1994). Theseauthors showed that a significant part of theobserved SST interannual variability in the tropicalAtlantic is related to an internal Atlantic equatorialmode similar to the ENSO in the Pacific. A newfeature, however, emerges for the warm phasecharacterized by negative correlation values as highas -0.6 along the southern limit of the SB LME. Forthe warm PDO phase, anomalously warm (cold)surface waters in the TSA relate to cold (warm) thannormal surface waters in the South Atlantic to thesouth of 35°S. It is worth noting the lack ofsignificant correlations in the area under theinfluence of the South Atlantic ConvergenceZone(SACZ), similar to the observed pattern for theNiño 3 and TNA indices. The SACZ is an elongatedconvective band that originates in the Amazon basinextending to the southeastern Atlantic Ocean,responsible for extreme precipitation events andstrongly influenced by warm ENSO events thatfavors its persistence over de Atlantic (Carvalho etal. 2004).It takes six months for the AAO to developthe highest positive correlations along the SB LMEPan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

314D.F.M GHERARDI ET ALLIFigure 3. Significant correlation maps for Niño 3, TNA and TSA during the cold (left) and warm (right) PDO phaseand their respective time lags. Hatched rectangles indicate areas of spurious correlation. Black straight lines correspondto the limits between LMEs in Brazil and were added for reference. Color bar is in (nondimensional) units ofcorrelation.Pan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

Climate variability and large marine ecosystems in the western South Atlantic315and two years (24 months) to develop negativecorrelations along the NB and EB LMEs, as well aspositive correlations restricted to the southern coast(Fig. 4). The most relevant aspect of thesecorrelations is that this is the only index to presentsome relation with the interannual variability ofSSTA within the SACZ region. The fact thatsignificant correlations also developed with a 24months delay with inverted signals in tropicaland southern Atlantic deserves further attention.This aspect is analyzed through the sequentiallagged correlation maps from lag 0 to 24 months(Fig. 4). Significant positive correlations appearoffshore the southern coast between 25°S and30°S at lag 1 month. Gradually, these correlationsintensify and occupy a large area between 25°Sand 35°S from lag 1 to 9 months. Significantnegative correlations develop along the NB LME bylag 9 months. As these negative correlationsintensify, significant (negative) correlations start todevelop between 15° S and 20° S in the SouthAtlantic, and the positive correlation center between25° S and 40° S splits into two centers by lag17 months. With time, all correlations intensify, withthe strongest values being settled by lag 24 months.This pattern shows, indeed, three main centerswith the largest magnitude of correlation: twonegative centers, one along the NB LME andanother along the EB LME, and one positive centeralong the SB LME.DiscussionThere are striking differences between thecorrelation fields for Niño 3 and the proposedseparation of LMEs along the Brazilian coast. Thisis also the case when the correlation fields for thecold and warm PDO phases are projected onto theseLMEs. We have found that the time lag necessaryfor maximum correlations to be reached is the same(eight months) for both PDO phases, suggesting thatthe mechanisms connecting the Pacific and theAtlantic Ocean are comparable. During the coldPDO phase the boundary between the NB and theEB LMEs intersects a large region where SSTAs arehighly correlated with the tropical Pacific variability(Fig. 3). It is striking, however, the coincidencebetween the significant correlations and the southernlimit of the EB LME. High correlations betweenNiño 3 and SSTA cover the SB LME only partiallyand are mostly centered offshore (at 35°S, 45°W).This contrasts with our results for the NB and EBLMEs, where high correlations are found close tothe coastline. This gives a strong indication thatmarine organisms living on the continental shelf ofthe SB LME, which are sensitive to SSTAs, maytake longer to react the impact of ENSO events. Wedo know, however, that ENSO induced increase inprecipitation in the Patos Lagoon estuary has anegative impact on euryhaline species (Garcia et al.2001). Looking at the warm PDO phase, the EBLME includes areas with high and medium positivecorrelations and also a large area where nosignificant correlation was detected. This means thattrophically related populations in the EB LME maynot respond consistently to a remote climatic forcingsuch as the El Niño because environmentalconditions expressed as SSTAs co-vary differentlyinside this region. This spatial discontinuity ofcorrelations within the EB LME can pose somethreat on pelagic species that rely on the thermalstructure of the west tropical Atlantic, such as thealbacore Thunnus alalunga during their reproductivephase (Frédou et al. 2007). Two other emblematicexamples of the spatial and temporal effects ofclimate on marine pelagic ecosystems are providedby Stenseth et al. (2002), the Peruvian anchovycrash in 1972 and the zonal displacement of thePacific skipjack tuna following the eastwarddisplacement of the warm pool during ENSO events.A further complicating factor is thatecological processes sensitive to long term (e.g.decadal) environmental changes are likely to besubmitted to different regimes (see North et al.2009) within the EB LME. The same complicatingfactor is even more evident in the South Brazil (SB)LME, where high positive correlation with El Niñohas been detected during the PDO cold phase, but nosignify-cant relation was found in the subsequentwarm phase (Fig. 3). So, the PDO-related multidecadaland ENSO-related interannual SSTAvariability along the Brazilian coast exhibit acomplex dynamics against which ecosystemfunctioning should be tested to provide clues as tohow NB, EB and SB LMEs might respond to theseclimate forcings. Besides, if one considers thehypothesis that the PDO can be represented as a rednoise process, then extreme values or rapid shiftsmight occur when fortuitous random phasingcombine contributions of different frequencies(Overland et al. 2010).The TNA and TSA indices are long knownas indicators of the principal modes of TropicalAtlantic Variability (TAV), namely meridionalSSTA gradients, which are important for the climateof the tropical Atlantic and the surrounding landmasses (Enfield et al. 1999). The reason to includethese indices in our analyses is to portray a balancedview of the inter-basin and within-basin influence onthe SSTA of the Brazilian LMEs. The extent towhich the TNA and TSA interact with SSTA alongPan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

316D.F.M GHERARDI ET ALLIthe Brazilian LME can be explored in the correlationmaps of Fig. 3. These maps show a scenario wherecorrelations with TNA are limited to the southernhalf of the EB LME during the warm PDO phase,with a single correlation area between 15° and 25° S,and points to a possible indirect influence via ENSOteleconnection over the region. In fact, the TNAitself is forced by the ENSO and is likely to be ofmarginal importance to the NB and EB LME ifcompared to the meridional propagations of SSTAsin the tropical Atlantic (Andreoli & Kayano 2004). Itis worth mentioning the significant differencesbetween cold and warm PDO correlations for theTNA. This is a recurrent feature that highlights theimportance of decadal variability in shaping spatialpatterns of LME vulnerability to climate change.Interpretations regarding the TSA should bemade with caution due to the proximity of the EBLME with the region from which the index has beencalculated. However, the antisymmetric (opposedsigns) correlation pattern observed in the SB LME,during the warm PDO phase, may be related withthe northeast-southeast SST dipole suggested byGrodsky & Carton (2006). It is beyond the scope ofthe present work to discuss the applicability of theterm dipole but our results point to a basin scaleinterannual relation between the SB LME SSTA andthe so-called TAV. It is important to highlight thatnot only TAV may have an impact in SB LME but itstrengthened after 1977, since it was absent duringthe PDO cold phase. Again, the negative correlationsof TSA found only for the southern half of the SBFigure 4. Significant correlation maps for AAO (warm PDO phase only) indicating the time and space evolution ofcorrelation fields along four time lags (from zero to 22 months). Black lines correspond to the limits between LMEs inBrazil and were added for reference. Color bar is in (nondimensional) units of correlation.Pan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

Climate variability and large marine ecosystems in the western South Atlantic317LME suggest that possible random phasing withremote forcing of SSTA is likely to produce mixedeffects in this LME.The leading EOF of monthly sea surface heightanomalies (SSHA) shown by Grodsky and Carton(2006), indicates a region of shallow thermoclinein the southwest Atlantic between 25° and 35° Scoincident with negative SST EOF scores.They interpreted the westward 10 cm interannualSSHAs as Rossby waves produced by thermoclineanomalies due to local and equatorial air-seainteractions. These waves would propagate alongthe Agulhas Eddy Corridor (AEC). The regionof shallow thermocline is coincident with a fieldof negative correlations with the AAO withouttime lag that dominates the northern half of the SBLME (Figure 4). This means that positive AAOindices are correlated with cold of SSTAs andstronger westerly circumpolar flow (positive midlatitudepressure anomalies) as previously indicatedby Thompson and Wallace (2000). Coupled oceanatmospheremodel results suggest that much of thevariability of the south Atlantic poleward of 30° Shas a direct relation with the Southern HemisphereAnnular Mode (hence, the AAO; Hall & Visbeck,2002). Positive AAO is related with increasedpoleward ocean heat transport at 30° S and areduction at 50° S, with associated 0.05° C increasein SSTA in the subtropics. The importance ofAAO as a source of large-scale interannualvariability in the tropical and Southern AtlanticOcean can be also inferred from positive correlationcenters along the northern and eastern coastand positive correlations observed between 25° Sand 35° S, 15 and 22 months ahead of the AAOfor the warm PDO phase (Figure 4). Whether thisis a result of enhanced Ekman drift and convergenceof heat it is not possible to ascertain, but theobserved correlations indicate that the AAO exertsa strong influence on the environmental conditionsalong the LMEs.It is possible that both TSA and AAO havean influence on the EB and SB LME acting indifferent ways depending on the geographic locationand time scale. This influence seems to beparticularly conspicuous during the warm PDOphase. In this preliminary report, we can only bespeculative, but this is not to say that there are noevidences for the influence of the spatial scale.Indeed, it is quite the opposite, looking at theinterplays among population dynamics, climatechange and fisheries throughout the Atlantic, it isseen that at the basin scale patterns of variations arespatially structured (Rouyer et al. 2008).The current knowledge on coupled oceanatmospheredynamics tells us that LMEs are tiedtogether by wind stress forcing, Ekman drift andheat transport. All of these are important agents thatcontrol the pelagic food-web structure, includingprimary productivity, mesozooplankton biomass andthe position of spawning habitats of pelagic fishes(Kiorboe 2008). Changes in surface currents, windstress and heat flux can have an impact on the longtermdynamics of zooplankton functional groups,leading to regime shifts in the ecosystem functioningfrom bottom-up to top-down control (Molinero et al.2008). If monitoring and management of LMEs areto become an effective means to respond to climaticimpacts on marine biodiversity and productivity,then the physical linkages between oceanatmospheredynamics and the pelagic ecosystem ona regional and basin scale have to be explicitlyconsidered.ConclusionsThe above results are preliminary findingsthat aim at exploring the spatial patterns ofcorrelation between climate indices and the SSTAalong the Brazilian LMEs at the interannual timescale. Significant correlations indicate that there is aseparation between the north and east Brazilcoasts located halfway between the boundaries ofthe EB LME. The SSTAs in the SACZ regionshowed no significant correlation with Niño 3, TNAand TSA, but are correlated with the AAO duringthe warm PDO phase. Possibly, the most evidentpattern that surfaced from the results is the influenceof the PDO phase shift causing dramatic changesin the spatial distribution of correlations. Thecorrelation patterns for the TSA and AAO seemto have a better fit with Brazilian LME duringthe warm PDO phase (1977-2007). During thisphase of the PDO, while the largest magnitudecorrelations are found in the EB and SB LME forthe TSA, they are centered in the three LME areasfor the AAO. It is, strongly recommended thecombined use of coupled ocean-climate andecological models as a means to elaborate thepossible mechanisms linking climate change andthe functioning of LMEs in Brazil. The assumptionthat LMEs delimited along the Brazilian coastcoherently respond to global climate changes, andthat these can be used to monitor their impactsshould be taken with caution. It is clear that, asfar as their dependence on SSTA is concerned,productivity and trophic relations in each of theBrazilian LMEs are likely to generate mixedresponses at the ecosystem level. This would, inturn, induce policy makers to react to a confoundedscenario of environmental change.Pan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

318D.F.M GHERARDI ET ALLIAcknowledgementsThe authors thank the support provided bythe National Institutes of Science and Technologyfor Climate Changes - Coastal Zones. ComputationalReferencesAlexander, M., Blade, I., Newman, M., Lanzante, J.R., Lau, N-C. & Scott, J. D. 2002. TheAtmospheric Bridge: The Influence of ENSOTeleconnections on Air–Sea Interaction overthe Global Oceans. Journal of Climate, 15:2205-2231.Andreoli, R. V. & Kayano, M. T. 2004. Multi-scalevariability of the sea surface temperature inthe Tropical Atlantic. Journal of GeophysicalResearch, 109: C05009.1-C05009.12.DOI: 10.1029/2003JC002220.Belkin, I. M. 2009. Rapid warming of large marineEcosystems. Progress in Oceanography, 81:207-213.Brander, K. M. 2007. Global fish production andclimate change. Proceedings of the NationalAcademy of Sciences, 104: 19709-19714.Carvalho, L. M. V., Jones, C. & Liebmann, B. 2004.The South Atlantic Convergence Zone: intensity,form, persistence, and relationships withintraseasonal to interannual activity and extremerainfall. Journal of Climate, 17: 88-108.Duda, A. M. & Sherman, K. 2002. A new imperativefor improving management of large marineecosystems. Ocean and CoastalManagement, 45: 797-833.Enfield, D. B. & Mayer, D. A. 1997. TropicalAtlantic SST and its relation to El Niño-Southern Oscillation. Journal of GeophysicalResearch, 102: 929-945.Enfield, D. B., Mestas, A. M., Mayer, D. A. & Cid-Serrano, L. 1999. How ubiquitous is thedipole relationship in tropical Atlantic seasurface temperatures? Journal of GeophysicalResearch, 104: 7841-7848.Frédou, F. L., Frédou, T., Travassos, P., Lins, J.,Arfelli, C., Andrade, H. A. & Hazin, F. 2007.Distribuition, catch and length composition ofthe albacore tuna (Thunnus alalunga) caughtby the longline fishery in the South Atlanticocean. Colllective Volume of ScientificPapers ICCAT, 60(2): 518-526.Garcia, C. A. E., Sarma, Y. V. B., Mata, M. M. &Garcia, V. M. T. 2004. Chlorophyll variabilityand eddies in the Brazil-Malvinas ConfluenceRegion. Deep-Sea Research II, 51: 159-172.Garcia, C. A. E., Vieira, J. P. & Winemiller, K. O.2001. Dynamics of the shallow-water fishassemblage of the Patos Lagoon estuaryhelp by Mr. Cristiano P. Oliveira is alsoacknowledged. HCS is funded by a fellowship fromCNPq no. 381661/2009-1.(Brazil) during cold and warm ENSOepisodes. Journal of Fish Biology, 59: 1218-1238.Giannini, A., Saravanan, R. & Chang, P. 2004. Thepreconditioning role of tropical AtlanticVariability in the development of ENSOteleconnection: implications for the predictionof Northeast rainfall. Climate Dynamics, 22:839-855.Gigliotti E. S., Gherardi, D. F. M., Paes, E. T.,Souza, R. B. & Katsuragawa, M. 2010.Spatial analysis of egg distribution andgeographic changes in the spawning habitat ofthe Brazilian sardine Sardinella brasiliensis.Journal of Fish Biology, 77: 2248–2267.Grodsky, S. A. & Carton, J. A. 2006. Influence ofthe tropics on the climate of South Atlantic.Geophysical Research Letters, 33:L06719.1-L06719.4.DOI:10.1029/2005GL025153.Hall, A. & Visbeck, M. 2002. Synchronousvariability in the southern hemisphereatmosphere, sea ice, and ocean resulting fromthe annular mode. Journal of Climate, 15:3043-3057.Hastenrath, S. 1976. Variations in low-latitudecirculation and extreme climatic events in thetropical Americas. Journal of theAtmospheric Sciences, 33: 202–215.Hastenrath, S. 2006. Circulation and teleconnectionmechanisms of northeast Brazil droughts.Progress in Oceanography, 70: 407-415.Higgason, K. D. & Brown, M. 2009. Local solutionsto manage the effects of global climate changeon a marine ecosystem: a process guide formarine resource managers. ICES Journal ofMarine Science, 66: 1640-1646.Kayano, M. T., Rao, V. B. & Moura, A. D. 1988.Tropical circulations and the associatedrainfall anomalies during two contrastingyears. Journal of Climatology, 8: 477-488.Kayano, M. T., Oliveira, C. P. & Andreoli, R. V.2009. Interannual relations between SouthAmerican rainfall and tropical sea surfacetemperature anomalies before and after 1976.International Journal of Climatology, 29:1439-1448.Kelmo, F., Attrill, M. J. & Jones, M. B., 2003.Effects of the 1997-1998 El Niño on thePan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

Climate variability and large marine ecosystems in the western South Atlantic319cnidarian community of a high turbidity coralreef system (Northern Bahia, Brazil). CoralReefs, 22(4): 542-550.Kidson, J. W. 1988. Interannual variations in theSouthern Hemisphere circulation. Journal ofClimate, 1: 1177-1198.Kiorboe, T. 2008. A mechanistic approach toplankton ecology. Princeton University Press,209 p.Lanzante, J. R. 1996. Lag relationships involvingtropical sea surface temperatures. Journal ofClimate, 9: 2568-2578.Lentini, C. A. D., Olson, D. B. & Podesta, G. P.,2002. Statistics of Brazil current ringsobserved from AVHRR: 1993–1998.Geophysical Research Letters, 29, 1–4.Mantua, N. J., Hare, S. R., Zhang, Y. & Wallace, J.M., Francis, R.C. 1997. A Pacific interdecadalclimate oscillation with impacts onsalmon production. Bulletin of the AmericanMeteorological Society, 78: 1069-1079.Molinero, J. C., Ibanez, I., Souissi, S., Buecher, E.,Dallot, S. & Nival, P. 2008. Climate controlon the long-term anomalous changes ofzooplankton communities in the NorthwesternMediterranean. Global Change Biology, 14:11-26.Nobre, P. & Shukla, J. 1996. Variations of seasurface temperature, wind stress, and rainfallover the tropical Atlantic and South America.Journal of Climate, 9: 2464-2479.North, E. W., Gallego, A. & Petitgas, P. 2009.Manual of recommended practices formodelling physical - biological interactionsduring fish early life. ICES CooperativeResearch Report, 295, 111 p.Overland, J. E., Alheit, J., Bakun, A., Hurrell, J. W.,Mackas, D. L. & Miller, A. J. 2010. Climatecontrols on marine ecosystems and fishpopulations. Journal of Marine Systems, 79:305-315.Peterson, R. G. & White, W. B. 1998. Slow oceanicteleconnections linking the AntarcticCircumpolar Wave with the tropical El Nin˜o-Southern Oscillation. Journal of GeophysicalResearch, 103: 24,573– 24,583.Rouyer, T., Fromentin, J.-M., Ménard, F., Cazelles,B., Briand, K., Planet, R., Planque, B. &Stenseth, N. C. 2008. Complex interplaysamong population dynamics, environmentalforcing and exploitation fisheries.Proceedings of the National Academy ofScience of the U.S.A., 105: 5420-5425.Servain, J., Wainer, I., Ayina, H. L. & Roquet, H.2000. The relationship between the simulatedclimatic variability modes of the TropicalAtlantic. International Journal ofClimatology, 20: 939-953.Sherman, K. 1991. The large marine ecosystemconcept: research and management strategyfor living marine resources. EcologicalApplications, 4: 349-360.Sherman, K., Belkin, I. M., Friedland, K. D.,O'Reilly, J. & Hyde, K. 2009. Acceleratedwarming and emergent trends in fisheriesbiomass yields of the world's large marineecosystems. Ambio, 4: 215-224.Smith, T. M., Reynolds, R. W., Peterson, T. C. &Lawrimore, J. 2008. Improvements toNOAA’s Historical merged Land Oceansurface temperature analysis (1880-2006).Journal of Climate, 21: 2283-2296.Souza, R. B. & Robinson, I. S. 2004. Lagrangianand satellite observations of the BrazilianCoastal Current. Continental ShelfResearch, 24: 241-262.Stenseth, N. C., Mysterud, A., Ottersen, G., Hurrell,J. W., Kung-Sik, C. & Lima, M. 2002.Ecological effects of climate fluctuations.Science, 297: 1292-1296.Torrence, C. & Compo, G. P. 1998. A practicalguide to wavelet analysis. Bulletin of theAme-rican Meteorological Society, 79: 61-78.Thompson, D. W. J. & Wallace, J. M. 2000. Annularmodes in the extratropical circulation. Part 1:month-to-month variability. Journal ofClimate, 13: 1000-1016.Yatsu, A., Aydin, K. Y., King, J. R., McFarlane, G.A., Chiba, S., Tadokoro, K., Kaeriyama, M. &Watanabe, Y. 2008. Elucidating dynamicresponses of North Pacific fish populations toclimatic forcing: Influence of life-historystrategy. Progress in Oceanography, 77:252-268.Wagner, R. G. & Da Silva, A. M. 1994. Surfaceconditions associated with anomalous rainfallin the Guinea coastal region. InternationalJournal of Climatology, 14: 179-199.Zebiak, S. E. 1993. Air-sea interaction in theequatorial Atlantic region. Journal ofClimate, 6: 1567-1586.Received December 2009Accepted June 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 310-319

An essay on the potential effects of climate changeon fisheries in Patos Lagoon, BrazilFÁBIO DE ANDRADE SCHROEDER 1 & JORGE PABLO CASTELLO 21 Institute of Oceanography, University of Rio Grande, Av Itália Km 8, RS, Brazil. CEP 96201-900. Email 1 :tazzoceano@yahoo.com.br; Email 2 : docjpc@furg.brAbstract. Important artisanal fisheries depend on the Patos Lagoon estuarine resources (southernBrazil). About 3,500 fishermen work in this region. Some resources, notably pink shrimp, whitemouthcroaker and grey-mullet are estuarine-dependent species as their life cycles depend on abrackish water environment. The ENSO cycle and climate changes may impact the estuary. In thisessay, we sought to qualitatively analyze these environmental changes impact on estuarinesecondary production and fisheries. We ponder the currently available knowledge about thespecies and regional climate models which show a maximum 2ºC increase in the next 30 years,with rainfall rates showing little change. However, the lagoon outflow should increase due tochanges on the hydrographic basin, which would increase the limnic and decrease the salineinfluence in the estuary. This scenario may impact the biology and dynamics of estuarinedependentspecies and their fisheries, because temperature influences metabolism, which affectsthe growth of individuals. Larvae natural mortality may increase due to metabolic stress, althoughincreased growth rates may also reduce the vulnerability period of young to predation. A decreasein the species maximum size is also expected, as well as a shift in biomass peaks and changes inthe fishing calendar.Keywords: temperature and rainfall increase, estuarine fishery resources, climate impactsResumo. Um ensaio sobre os efeitos potenciais das alterações climáticas sobre a pesca naLagoa dos Patos, Brasil. Importantes pescarias artesanais dependem dos recursos estuarinos daLagoa dos Patos (sul do Brasil). Estima-se que 3.500 pescadores trabalhem na região. Algunsrecursos (camarão-rosa, corvina, tainha) são estuarino-dependentes, tendo seus ciclos de vidaligados ao ambiente salobro. Este ensaio analisa qualitativamente os impactos que as mudançasclimáticas terão sobre a produção secundária e a pesca estuarina. Considerou-se os conhecimentosdisponíveis sobre a biologia e dinâmica das espécies envolvidas e os modelos climáticos. Estesapontam um aumento máximo de 2ºC na região nos próximos 30 anos, com pouca mudança naprecipitação. Entretanto, a vazão da lagoa deve aumentar, devido às mudanças antrópicas na baciada Lagoa dos Patos, aumentando a influência límnica e reduzindo a influência salina no estuário.Nesse cenário, esperam-se impactos na biologia, dinâmica e pesca das espécies estuarinodependentes,pois a temperatura influencia o metabolismo, afetando o crescimento individual. Amortalidade natural larval aumentaria, devido ao estresse metabólico, mas o aumento das taxas decrescimento reduziria o período de vulnerabilidade juvenil à predação, diminuindo a mortalidadenatural. É esperada uma diminuição no tamanho máximo das espécies, assim como umdeslocamento dos picos de biomassa e mudanças no calendário de pesca.Palavras-chave: aumento de temperatura e chuva, recursos pesqueiros estuarinos, impactosclimáticosIntroductionUntil the Industrial Revolution (earlynineteenth century), anthropogenic actions (e.g.,deforestation, the regional extinction of species,destruction of the natural landscape) had onlyregional or perhaps even continental impacts. Afterthe Industrial Revolution, however, the impacts ofhuman activities became global (based on thepremise that climate changes are the result ofPan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

Potential effects of climate change on fisheries in Patos Lagoon321anthropogenic actions and that they are related to thecarbon concentration in the atmosphere) (IPCC2007).The Industrial Revolution marks thebeginning of the use of fossil fuels, resulting in theproduction of greenhouse gases such as carbondioxide (IPCC 2007). These gases increase theretention of heat in the atmosphere, causing globalwarming and climate change on the planet (Mitchell1989, Cline 1991). At present, there is a great deal ofemphasis within the academic community oncontinuing studies on the consequences of globalwarming, how it may be prevented and on theadoption of mitigation measures.The delay of many countries in adoptingmeasures to reduce greenhouse gas emissions haveresulted in steady and rapid increase in levels ofcarbon in the atmosphere, such that the currentconcentration is the highest in the last 450,000 years(King 2007).The carbon absorption capacity of theoceans, which are the largest absorbers of carbondioxide on the planet, seems to have reached its limitand has decreased over recent years (Khatiwala et al.2009). This means that more greenhouse gases willremain in the atmosphere and will continue to driveclimate change.According to Solomon et al. (2009), manyof the effects of climate change are alreadyirreversible (King 2007) because of the high degreeof inertia of the oceans and other climate controlsystems, which respond slowly to environmentalchanges (Caldeira et al. 2003). If emissions werereduced to zero today, the temperature of the planetwould still continue to rise for over a thousand years(Solomon et al. 2009).Thus, reversing the climate change processand returning the planet to a pre-Industrial Revolutionclimate appears to be an almost impossibletask. Because of this, it will be necessary to makeadjustments to the new climate of the planet byundertaking measures to minimize further impactson mankind.The analysis of likely future climate changescenarios and characterization of their biotic impactcan support this adaptation.The study presented in this essay is aqualitative analysis of the influences of climatechange in the Patos Lagoon estuary and thebiological response of some fish species there. It isbased on the best information that is currentlyavailable without making a distinction betweennatural and anthropogenic effects. The synergisticeffects of living beings and their adaptivecapabilities are also disregarded.Patos LagoonPatos Lagoon (Fig. 1), which is locatedin the coastal plain of Rio Grande do Sul, is250 km long and has an average width of40 km, covering a total area of approximately10,360 km 2 . This makes it the largest coastallagoon of the choked type in the world. It can beclassified as a shallow lagoon having an averagedepth of 5 m. The topography of the mainlagoon body is characterized by major natural andartificial channels (8 - 9 m), wide adjacent areas(

322F. A. SCHROEDER & J. P. CASTELLOWhy the Patos Lagoon Estuary?This estuary in southern Brazil was used forthis study because it is an area that has been studiedby the University of Rio Grande (FURG) in detailsince the 1970s (Seeliger et al. 1998). It is also aregion with a high degree of socio-environmentalimportance in the Southwest Atlantic (Seeliger et al.1998) as a fishing ground for artisanal fisheries, as anursery area for larvae and juveniles of manyspecies (Chao et al. 1985, Haimovici et al. 2006)and because of its agro-industrial, tourist and portactivities (Seeliger et al. 1998).Rice production in the region reached morethan eight million tons in 2008 (InstitutoRiograndense de Arroz 2009). The port of RioGrande (the only seaport in the state) processedmore than 21.5 million tons of cargo in 2003 andserved as a port for almost three thousand ships(Superintendência do Porto de Rio Grande 2009).The artisanal fishermen community in theregion, which depends on these resources, fluctuatesaround approximately 3,500 people who are directlyinvolved in fishing (Haimovici et al. 2006), highlightingthe socio-economic importance of theregion.Materials and MethodsClimate ModelsFor future climate projections, we used thefuture scenarios A2 (pessimistic) and B2 (optimistic)of the Intergovernmental Panel for Climate Change(IPCC 2007). The A2 scenario describes a veryheterogeneous world that is characterized by a highdegree of self-sustenance with the identities of localpopulations being preserved. Birth rates in thisscenario continue to be higher than mortality rates,resulting in continuous population growth.Economic development is regionally oriented, andeconomic growth and technological changes aremore fragmented and slower than in other scenarios.The B2 scenario describes a world with anemphasis on local solutions for economic, social andenvironmental sustainability. It is a world with acontinuously growing global population, though at arate lower than in the A2 scenario, and withintermediate levels of economic development.Technological change will be less rapid and morediverse than in scenarios B1 and A1. This scenario isalso more oriented towards environmental protectionand social equality focused on both local andregional processes.From these scenarios, the National Institutefor Space Research (INPE) created thirty-year futureclimate models (2010 to 2040), aimed atunderstanding expected changes in rainfall and localair temperature (Marengo 2007).Time SeriesIn this study, the following time series wereused: rainfall and river flow of the main riverscontributing to the Patos Lagoon basin (Costa et al.in press) (Fig. 2) rainfall (Steinmetz 2007, Costa etal. in press) (Fig. 3) and air temperature in the cityof Pelotas, which is located in the margin of theestuary region (Steinmetz et al. 2007) (Fig. 4).Fishing ResourcesAmong the 28 species that are exploitedthrough fishing in this region, five have historicallybeen the most important: pink shrimp (Farfantepenaeuspaulensis), corvina/whitemouth croaker(Micropogonias furnieri) miragaia/black drum(Pogonias cromis), tainha/grey-mullet (Mugil platanus)and bagre/white sea catfish (Genidens barbus).However, according to the present profile ofartisanal fishing, only three species (pink shrimp,whitemouth croaker and grey-mullet) were selectedfor the study because the black drum and catfishhave been heavily over-exploited and their catchesare no longer significant (Haimovici et al. 2006).Together, the three species represent more than30% of the landings and nearly 45% of therevenue from fishing in this region (Haimovici et al.2006, Castello et al. 2009). Thus, it is possible tocharacterize them as the most important socioeconomicresources of the estuary. As such, it can beassumed that characterizing the impacts on thesethree fisheries is equivalent to characterizingthe impacts on the entire estuarine fishing in theregion.The reproductive peak of the whitemouthcroaker takes place from September to October,with spawning occurring at the mouth of the estuary.The eggs and larvae of this species penetratethe estuary by being passively transported by thecurrent, particularly in late spring and early summer,and come to occupy the shallower marginalareas. As their development progresses, the croakerswill move to the deeper areas of the estuary,and they will return to the sea upon reachingadulthood (Castello 1985, Oliveira & Bemvenuti2006).The grey-mullet also spawns in the open seabetween the coasts of the neighbour states of RioGrande do Sul and Santa Catarina in the late fall andearly winter. Juveniles then migrate to shallowerwaters and enter the estuaries, particularly during thespring. Upon reaching maturity, they return to theocean in shoals, generating the so-called “mulletrace” (Oliveira & Bemvenuti 2006).Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

Potential effects of climate change on fisheries in Patos Lagoon323Figure 2. Rainfall time series (right) in the drainage basin and flow (left) of the main rivers contributing to PatosLagoon. The study period is 40 years (1960 to 2000) for the rainfall series and 90 years (1912 to 2002) for the riverflow series. (Costa et al. in press).Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

324F. A. SCHROEDER & J. P. CASTELLOPink shrimp breed in the spring on the shelfat the coastal front of Santa Catarina. The larvaemigrate in a passive manner into the Patos Lagoonestuary in the form of megalopae (D'Incao 1991).They will remain there for three to four months inthe shallower embayment areas of the estuary untilthey reach the juvenile or pre-adult stage, at whichpoint they will return to sea in autumn-winter(Ribeiro et al. 2004).Combining the INPE predictions and thetime series, we chose to create a future climatescenario for the next 30 years.A conceptual model was developed based onthe main stages of the life cycle of a cohort (Fig. 5).From this, we identified the stages and parameters ofthe life cycle that could be altered by climatechange.In Figure 5, the rectangles represent thebiomass (B) of a cohort at each time point (T). Ateach time transition, there is an effect of growth (G),natural mortality (M) and, after a certain age, fishingmortality (F). The sub-index 1 represents larvae andjuveniles, 2 represent the recruits, and 3 representsFigure 3. Rainfall time series for the city of Pelotas (citymarginal to the estuary) during the period from 1880 to2002. The line shows a variation of 184 mm in 111 years(Steinmetz et al. 2007b).Figure 4. Air temperature anomaly time series for the cityof Pelotas (from the portion of the city marginal to theestuary) during the period from 1951 to 2006 (Steinmetz2007).the sexually mature (breeders). Tr and Lr identifythe age and size at the time of recruitment,respectively. Analogously, Tc and Lc identify ageand size at the time of the first catch, respectively.Age and length at first maturation, which is the timeof first breeding, are designated by Tm and Lm.Closely connected to this whole process, there ismigration between the mating, breeding and feedingsites.To carry out this study, we used thefollowing premises:- All of the determinants that affect climate, andestuarine characteristics and processes and/or thebiology of fishing resources were consideredwithout distinguishing between natural andanthropogenic forces.- The adaptive capacities of species were notconsidered.- The 30-year scenario was chosen as the mostprudent because future knowledge shall come intoplay if predictions are made for longer periods.Results and DiscussionFuture ScenarioPrecipitation and salinityThe INPE climate models (Marengo, 2007)do not indicate rainfall anomalies for this region,i.e., climate change will not affect rainfall in thedrainage basin of Patos Lagoon in the next thirtyyears independent of the analyzed scenario (A2 orB2). However, the spatial and temporal resolution ofthese climate models is low and large-scale and doesnot take into consideration local characteristics ofeach region.During the discussions of the First BrazilianWorkshop on Climate Change in Coastal Zones(held in Rio Grande/RS; 13-16/09/2009), the need toincrease our knowledge on the specifics of howclimate change is occurring was remarked. Thiswould attend the purpose of producing models thatare better predictors of climate change in the future(Rede Clima 2009).In contrast with the INPE models, the flowtime series of the major rivers contributing to thePatos Lagoon (Fig. 2) shows a clear upward trend.The Taquari River nearly tripled its flow over thepast 65 years; the flow of the Jacuí almost doubled;the Camaquã had a smaller increase of about 20%;and the contribution of the Mirim Lagoon remainedalmost stable with a slight increase. This means thatthere is an increasing amount of fresh water reachingthe estuary.This increase in outflow is associated withrising rates of rainfall in the drainage areas of theserivers (Fig. 2). At the same time there is aPan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

Potential effects of climate change on fisheries in Patos Lagoon325superimposed decadal cycle. Costa et al. (in press),for a shorter time period, found a much higherincrease in rainfall rates (up to 40% in 40 years)when compared with the data from Pelotas.Toldo et al. (2006) points to an increasein sedimentation rate in Patos Lagoon in the last150 years, i.e., before and after the agriculturalcolonization of the banks of its hydrographicbasin. Soil impermeabilization (due to building citiesand roads), destruction or reduction of riparianvegetation and agriculture all decrease the rates ofwater penetration into the soil, causing morewater (and therefore more sediment) to be carried toPatos Lagoon. This process, along with increasedrainfall in the drainage areas of the rivers basins,may explain the increased outflow of contributingrivers.Figure 5. Conceptual model of the evolution of a cohort.This scenario may be relieved by the highrates of evaporation in the Patos Lagoon, estimatedto be approximately 600 m 3 /s (Hirata & Möller2006). As evaporation is directly proportional totemperature, increased heat in the region shouldincrease evaporation rates, reducing the amount ofwater available.Assuming that freshwater input will increase(in a linear fashion or in a decadal cycle) in thesystem despite the evaporation rates, more freshwater will be carried to the estuary over the next 30years resulting in ‘limnification’, therefore,decreasing the size of the estuary (Costa et al. inpress).The ‘limnification’ process increases thefresh water in the estuary, reducing the salinity in itsentire area of influence. Therefore, the position ofzero salinity which determines the upper limit of theestuary, should move towards the sea, reducingthe area of influence of the salt and the brackishwater that is present in the southern region ofPatos Lagoon (Costa et al. in press). Still, it isexpected that this process shall be intensified whenincreasing the length of the rocky man-made pair ofjetties located in the mouth of the estuary (Fernandeset al. 2005).An increase of the outflow of the estuaryis also expected because there is an increase inthe volume of displaced water into the lagoon.In parallel, the speed at which the water exitsthrough the rocky jetties should increase, resultingboth from the volume of the outflow andrenovation work that is done on them (Fernandes etal. 2005).TemperatureAir temperature is expected to rise in theregion. The INPE model points to an elevation ofbetween one and three degrees Celsius, dependingon the season and model that is analyzed. Althoughthere are no historical data on water temperature inthe lagoon and estuary, the low average depth andresidence time of the water there indicate that it willfollow the trends that have been verified for the air(i.e. increasing temperature).The time series (Fig. 4) shows an upwardtrend in temperatures in the region, reinforcing theconditions outlined in these models.Estuary of Patos Lagoon - 30 yearsIn light of the data presented above, it canbe predicted that, in 2040, the estuary would besmaller, less salty and warmer, with stronger outflowand weaker inflow currents at the mouth of theestuary.Response of fish stocksOver time, the evolution of a fish cohort thatis not fished behaves according to the conceptualmodel in figure 6. The number of individuals in thecohort is maximal at time zero, i.e., immediatelyafter larval eclosion. As these individuals grow, thecohort biomass increases, while mortality reducesthe number of individuals. The cohort biomass peakoccurs when the natural mortality removes a largeenough number of individuals that the growth of thesurvivors can no longer compensate the biomass ofthe cohort. This point occurs at the intersection ofthe curves for number and weight of individuals(Fig. 6).When fishing is added to the model (Fig. 7),the curve representing the number of individualsdeclines even more due to increased mortality fromdeath by fishing. The biomass and numbers of acohort decrease with the increase in fishingmortality.Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

326F. A. SCHROEDER & J. P. CASTELLOEstuary Reduction and Limnic InfluenceThe reduction in the estuary area shouldresult in an increase in inter-and intra-specificcompetition among brackish water species, for bothspace and food. If we start with the premise that thenumber of individuals should not decline (at least atfirst instance), the population density in shallowareas with conditions of suitable salinity andvegetation will be higher. This means thatembayment areas, with seagrass meadows, whichare conducive to the proliferation of fish andcrustaceans (Seeliger et al. 1998), should bear agreater density of consumers.It is thus likely that the available food, whichpreviously supported a lower density of individualsin areas with brackisk waters, will be competed formore severely. This situation can create stress in thepopulations, increasing mortality of the estuarinedependentspecies. On the other hand, grey-mullet,which lives and growths in the fresh water area, willprobably be affected during recruitment time.While studying the effects of ‘El Niño’ onthe estuary, Garcia & Vieira (2001) noticed that thephenomenon caused a local reduction in salinity dueto high rainfall in the lagoon drainage basin. He alsofound that freshwater species, particularly Parapimelodusnigribarbis, Astyanax eigenmaniorum andOligosarcus jenynsii, had subsequently becomemore frequent in the estuary.These species will also compete for spaceand food with the estuarine-dependent species, andmay, especially in the case of the carnivore O.jenynsii, represent one more predator of larvae andjuveniles in the region.The influence of ‘limnification’ can causechanges in the dynamic movements of estuarinedependentspecies. ‘El Niño’ events reduce therecruitment of juvenile grey-mullets, negativelyaffecting the next growing season (Vieira et al.2008) and the reproductive migration (“mulletrace”).However, the variation in salinity should notaffect the biology of the species, since thefluctuation of salinity in the estuary is common,ranging from marine to limnic conditions (Seeligeret al. 1998). These changes are, however,ephemeral, while the changes that will occur due toclimate change are expected to be of a more constantnature.Thus, with the reduction of the estuary sizeand the influence of ‘limnification’, increasedmortality of the estuarine-dependent species andchanges in migration and population dynamics areexpected, reducing the available population forfishing.Figure 6. Model of a "virgin" cohort. The curve Lrepresents the weight of individual species. Curve Brepresents the biomass curve, and N is the total number ofindividuals.Figure 7. Model of an exploited cohort. Curve Lrepresents the size of individual species, B the biomass,Be the biomass with exploitation, N the number ofindividuals and Ne the number of individuals withexploitation.Increased temperatureTemperature is determinant to metabolism infish, in a manner that the second is proportional tothe first and that affects the rate of growth at all lifestages. Because the Patos Lagoon is shallow(average depth of 5 m), variations in air temperatureare quickly transmitted to the water column.Krummenauer et al. (2006) pointed out tothe low water temperature of the estuary in theautumn-winter period as a limiting factor for the cultivationof pink shrimp throughout the year. Therefore,a higher temperature could extend the growingseason for shrimp and increase their growth rate.Okamoto et al. (2006) discussed theadvantages of higher temperatures on the developmentof juvenile grey-mullet. At a higher temperature,the feed-to-weight gain ratio increases (i.e., thesame amount of food generates a greater amount ofPan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

Potential effects of climate change on fisheries in Patos Lagoon327body mass). Higher growth and fattening rates withhigher temperatures are also observed.This rapid growth should affect the age offirst maturity and result in an earlier start forreproduction in the species. Thus, the maximumlength of the species (L ∞ ) should decrease (Beverton& Holt 1957, Weatherley 1972).With individuals growing faster, the biomasspeak of the species should shift, occurring earlier inthe year. However, as the L ∞ should decrease, thepeak of biomass should follow this trend, decreasingas well.With higher growth rates, larvae will quicklypass through the periods of greater vulnerability,being exposed to “windows” of predation for lesstime. This should cause a decrease of the naturalmortality rate, allowing more individuals to reachyouth.However, increased temperature can alsocause metabolic stress on larvae, causing an increasein their natural mortality. Okamoto et al. (2006)found 5% higher mortality rates at highertemperatures compared with lower temperatures.Yet, metabolic stress is a response totemperature stress, or a sudden change intemperature. Climate change results in small,incremental changes in warming rates of less thanone degree Celsius every ten years. Therefore, thereis the possibility that physiological adaptations mayarise to circumvent the effects of temperature changeon metabolism.With accelerated growth rates and likelychanges in the timing of migrations, the fishingcalendar for the main species in the estuary maychange. Haimovici et al. (2006) have, in fact,previously reported changes in this calendar (Fig. 8).It is worth noting that there was a decreasein the peak of production, and an increase in theoverall time of harvesting. There were changes ofsome harvested species, for example, catfish,between the periods from September-November(spring) to June-September (winter).These changes are not unambiguously aresult of climate change. They may be associatedwith specialization and adaptation of fishermen todifferent patterns of abundance caused by overfishing(Kalikoski & Vasconcellos 2007). However,because of changes in travel patterns and populationdynamics, we cannot dismiss the possibility ofchanges occurring as a result of climate change.Temperature and salinity are known factors fortriggering migrations into or outward the estuary;therefore, an increase in temperature and a decreasein salinity may affect the timing of migration ofestuarine dependent species (Pitcher 1993).Currents at the Mouth of the EstuaryThe larvae, post-larval individuals andjuveniles of the three main fishery resources enterpassively into the estuary, carried out by currents(Möller et al. 2009, Vieira et al. 2008, Garcia &Vieira 2001). Thus, the balance between the inflowand outflow currents of the estuary is of paramountimportance for successful migration to occur.Möller et al. (2009) have discussed theeffects of rainfall anomalies (and consequently, theacceleration of the output current of the estuary) onthe harvest of shrimp in the estuary and found aninverse relationship in which more rain resulted inless shrimp in the following harvest.Vieira et al. (2008) performed a similaranalysis for the grey-mullet, concluding that, withincreased rainfall intensity and outflow current,juvenile recruitment would be hampered, reducingamounts of both current and future catches.Thus, the increase in the speed of outflowcurrents constitutes the largest impact on thesefisheries, hindering the entry of larvae and postlarvaeinto the estuary and affecting the life cycles ofthese species.Figure 8. Calendar of artisanal fisheries in the 1960s(above) and 1990s (below). The lines represent theproportion of the total catch in each month (Haimovici etal. 2006).Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

328F. A. SCHROEDER & J. P. CASTELLOSedimentation ratesToldo et al. (2006) have demonstratedthat the sedimentation rate of the Patos Lagoonhad increased from 5-11 times over the past150 years. Suspended material that is transportedby water increases the turbidity of the watercolumn, reducing light penetration and,thus, impairing primary production, which isgenerally already at a low level in estuaries(Barnes 1974).Visual predation by fish, such as thewhitemouth croaker, is also affected by increasedturbidity (Figueiredo & Vieira 2005). Turbidityreduces the visual field and acuity, thus undermininga visual strategy of predation. This may altertheir food intake and consequently, their growthrates.Marshes and submerged grasslandsSubmerged marshes and seagrasses arethe nurseries of the main fish stocks in the estuary.They provide shelter from predators, which havedifficulty traveling in these densely vegetatedenvironments.These regions are also providers of foodbecause they trap organic matter that is carried outby currents and retain the debris that is generatedlocally. They also house a whole community ofbenthos and plankton and even bacterial biofilms onthe vegetation, which serve as food for larvae andjuveniles.Just as these environments trap organicmatter, they also cause the retention of sediments.With the increase in the amount of suspendedmatter, sedimentation rates at these sites should alsoincrease.Thus, silting of seagrass meadows andmarshes can occur, reducing the footprint ofthese highly productive areas. Copertino (2010)shows that climate change may impact theestuarine seagrass fields, reducing its area.Therefore, the “nursery” area would also bereduced, increasing mortality of the species thatdepend on it.In contrast, scavenging species such aspink shrimp and grey-mullets would benefit fromthis condition. The nutrition of these speciescomes from two sources, allochthonous andautochthonous. With more material beingtrapped, the allochthonous amount that isavailable increases, so there is more food forscavengers. This could result in increased growthrates and decreased natural mortality for thesespecies, representing a trend in the oppositedirection to the previous case.ConclusionClimate change will have both positive andnegative impacts on fish stocks in the Patos Lagoonestuary.For the next 30 years, the main factoraffecting fisheries in this region should be theacceleration of the outflow current at the mouth ofthe lagoon. This current will be intensified, both byclimate changes and by anthropogenic actions fromthe occupation of the banks of the Lagoon and fromwork on the pair of jetties located in the mouth ofthe estuary.With decreases in the age of recruitment,catch and first maturity, changes are also expected tooccur in population structure and dynamics. There isalso expected to be a reduction in the maximum sizeof specimens (L ∞ ) and their life expectancy. Anadvance in the timing and a reduction in the size ofthe biomass peak are also expected (Fig. 9), as arechanges in migration patterns.Importantly, no single factor can beconsidered totally positive or negative. Many of theeffects are beneficial when analyzed from oneperspective and detrimental when analyzed fromothers. The best example is the increase in turbiditythat should affect the predatory strategy of thewhitemouth croaker but will introduce more foodinto the system for the grey-mullet. Future, morecomplex models and new data may increase ourlevel of understanding of climate change effects.Figure 9. Model of an exploited cohort under conditionsof climate change. The curve L represents the size ofindividual species, La the individual size with climatechanges, B the biomass, Be the biomass with exploitation,Ba the biomass with climate changes, Bea the exploitedbiomass with climate change, N the number ofindividuals, Ne the number of individuals withexploitation, Na the number of individuals with climatechanges and Nae the number of individuals withexploitation and climate change.Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

Potential effects of climate change on fisheries in Patos Lagoon329AcknowledgmentsThis work is part of the major project“Climate change, oceanographic variability and theartisanal fisheries in the SW Atlantic: a humanReferencesBarnes, R. S. K. 1974. Estuarine Biology. Studies inBiology n. 49, Edward Arnold, London, 76 p.Beverton, R. J. H. & Holt, S. J. 1957. On theDynamics of Exploited Fish Populations. In:Fisheries Investigations, H. M. StationeryOffice, London, 19(2).Caldeira, K., Jain, A. K. & Hoffert, M. I. 2003.Climate sensitivity, uncertainty and need forenergy without CO 2 emission. Science, 299:2052-2054.Chao, L. N., Vieira, J. P. & Pereira, L. E. 1985.Estuarine fish community of the dos PatosLagoon, Brazil: A Baseline Study. Pp. 429-450. In: Yañes-Arancibia, A. (Ed.). Fishcommunity ecology in estuaries and coastallagoons: Towards an ecosystem integration.UNAM Press, Mexico DF, Mexico, 654 p.Castello, J. P. 1985. La ecología de losconsumidores del estuario de Lagoa dosPatos, Brasil. Pp. 383-406 In: Yanez-Arancibia, A. (Ed) Fish Community Ecologyin Estuaries and Coastal Lagoons: Towardsan Ecosystem Integration. UNAM Press,Mexico DF, Mexico, 654 p.Castello, J. P., Sunyé, P., Haimovici, M. &Hellebrandt, D. 2009. Fisheries in southernBrazil: a comparison of their management andsustainability. Journal of Applied Ichthyology,25(3): 243-364.Cline, W. R. 1991. Scientific basis for the greenhouseeffect. The Economic Journal, 101:904-919.Copertino, M.S., 2010. In: McKenzie, L. J., Yoshida,R. L. & Unsworth, R. (Eds.). Seagrass-watchnews. Issue 40, Seagrass-watch HQ,16 p.Costa, C. S. B., Seeliger, U. & Bemvenuti, C. E. Inpress. Diagnóstico de alterações hidrológicasdevido ao impacto das mudançasclimáticas sobre o ecossistema costeirotemperado brasileiro através da vegetaçãoe do macrozoobentos. Ministério do MeioAmbiente (Série Biodiversidade), Brasília.D’Incao F. 1991. Pesca e biologia de Penaeuspaulensis na Lagoa dos Patos, RS. Atlântica,13: 159-169.Fernandes, E. H., Cecílio, R. O. &Schiller. R. V.2005. Estudo da influência dos molhes da barrade Rio Grande sobre a circulação do estuárioda Lagoa dos Patos. Vetor, 15(2): 40-57.dimension approach”, support by the Inter-AmericanInstitute for Climate Change (IAI). We like to thankDr. Osmar Möller Jr and Dr. Patrizia Raggi for theadvices.Figueiredo, G. M., Vieira, J. P. 2005. Diel feeding,daily food consumption and the predationimpact of whitemouth croaker (Micropogoniasfurnieri) in an estuarine environment. MarineEcology, 26(2): 130-139.Garcia, A. M. & Vieira, J. P. 2001. Aumento dadiversidade de peixes no estuário da Lagoados Patos durante o episódio El Niño 1997-1998. Atlântica, 23: 85-96.Haimovici, M., Vasconcellos, M., Kalikoski, D. C.,Abdalah, P., Castello, J. P. & Hellebrandt, D.2006. Diagnóstico da pesca no litoral do RioGrande do Sul. In: Isaac, V. J., Martins, A. S.,Haimovici, M. & Andriguetto, J. M. (Eds.). Apesca marinha e estuarina do Brasil noinício do século XXI: recursos, tecnologias,aspectos socioeconômicos e institucionais.Universidade Federal do Pará, Belém, 188 p.Hirata, F. E. & Möller, O. O. 2006. Interannualvariability of rainfall and streamflow in PatosLagoon Basin: Preliminary Results. 8°ICSHMO. INPE, São José dos Campos, 1:775-780.Instituto Rio Grandense do Arroz 2009. WorldWide Web electronic publication,Accessible at http://www.irga.rs.gov.br/,(accessed 04/12/2009).IPCC 2007. Summary for policymakers. In:Solomon, S., Qin, D., Manning, M., Chen, Z.,Marquis, M., Averyt, K.B., Tignor, M. &Miller, H.L. (Eds.). Climate Change 2007:The Physical Science Basis. Contribution ofthe Working Group I to the Forth AssessmentReport of the Intergovernmental Panelon Climate Change. Cambridge UniversityPress, Cambridge and New York. 996 p.Kalikoski, D. C. & Vasconcellos, M. 2007. The roleof fishers knowledge in the co-management ofsmall-scale fisheries in the estuary of PatosLagoon, Southern Brazil. p. 289-312. In:Haggan, N., Neis, B. & Baird, I. G. (Eds.).Fisher’s Knowledge in Fisheries Scienceand Management. UNESCO, Paris, 437 p.Khatiwala, S., Primeau, F. & Hall, T. 2009.Reconstruction of the history of antropogenicCO 2 concentrations in the ocean. Nature, 462:346-349.King, M. 2007. Fisheries Biology, Assessment andManagement. 2 nd . Ed. Blackwell Pub. 382 p.Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

330F. A. SCHROEDER & J. P. CASTELLOKrummenauer, D., Wasielesky, W. W., Cavalli, R.O., Peixoto, S. & Zogbi, P. R. 2006.Viabilidade do Cultivo do camarão-rosaFarfantepenaeus paulensis (crustacea,decapoda) em gaiolas sob diferentesdensidades durante o outono no sul do Brasil.Ciência Rural, 36(1): 252-257.Marengo, J. A. 2007. Mudanças climáticas globaise seus efeitos sobre a biodiversidade:caracterização do clima atual e definiçãodas alterações climáticas para o territóriobrasileiro ao longo do século XXI (sérieBiodiversidade, v. 26). Ministério do MeioAmbiente, Brasília, 212 p.Mitchell, J. F. B. 1989. The greenhouse effect andclimate change. Reviews of Geophysics,27(1): 115-139.Möller, O. O., Castello, J. P. & Vaz, A. C. 2009. Theeffect of river discharge and winds on theinterannual variability of the pink shrimpFarfantepenaeus paulensis production inPatos Lagoon. Estuaries and Coasts, 32: 787-796.Okamoto, M. H., Sampaio, L. A. & Maçada, A. P.2006. Efeito da temperatura sobre ocrescimento e a sobrevivência de juvenis datainha Mugil platanus Günther, 1880.Atlântica, 26(1): 61-66.Oliveira, A. F. & Bemvenuti, M. A. 2006. Ciclo devida de alguns peixes do estuário da Lagoados Patos, RS, informações para o ensinofundamental e médio. Cadernos de EcologiaAquática, 1(2): 16-29.Pitcher, T. (Ed.) 1993. Behaviour of teleost fishes.2nd Ed. Fish and fisheries Series 7. Chapmanand Hall, 715 p.Rede Clima – INCT para Mudanças Climáticas.2009. Accessible at http://mudancasclimaticas.zonascosteiras.com.br/workshop (accessed08/12/2009).Ribeiro, P. R. C., Nunes, M. T. O. & Quadrado, R. P.2004. Pescador da Lagoa dos Patos:restabelecimento da capacidade produtivado sistema ambiental da pesca artesanal doextremo sul do Brasil – Camarão. FNMA(Fundo Nacional do Meio Ambiente).Accessible at http://www.ceamecim.furg.br/avatool/avas/ensino/bib-files/1699.doc.(Accessed 04/11/2009).Seeliger, U., Odebrecht, C. & Castello, J. P. (Eds)1998. Os ecossistemas costeiro e marinho doextremo sul do Brasil. Ecoscientia, RioGrande, 341 p.Solomon, S., Plattner, G. K., Knutti, R. &Friedlingstein, P. 2009. Irreversible climatechanges due to carbon dioxide emissions.PNAS 106(6): 1704-1709. Accessible athttp://www.pnas.org/cgi/doi/10.1073/pnas.0812721106 (Accessed 08/12/2009).Steinmetz, S. 2007. Impacto das mudançasclimáticas globais sobre o arroz irrigado no suldo Brasil. V congresso Brasileiro de ArrozIrrigado, Pelotas, RS.Steinmetz, S., Wrege, M. S., Herter, F. G. & Reisser,C. R. 2007. Influência do aquecimento globalsobre as temperaturas máximas, mínimas emedias anuais na região de Pelotas, RS. XVCongresso Brasileiro de Agrometeorologia,Aracajú, SE.Superintendência do Porto do Rio Grande 2009.Accessible at http://www.portoriogrande.com.br (accessed 04/12/2009)Toldo Jr., E. E., Dillenburg, S. R., Corrêa, I. C. S.,Almeida, L. E. S. B., Weschenfelder, J. &Gruber, N. L. S. 2006. Sedimentação de Longoe Curto Período na Lagoa dos Patos, Sul doBrasil. Pesquisas em Geociências, 33: 79-86.Vieira, J. P., Garcia, A. M. & Grimm, A. M. 2008.Evidences of El Niño effects on the mulletfishery of the Patos Lagoon estuary. BrazilianArchives of Biology and Tech, 51(2): 433-440.Weatherley, A.H. 1972. Growth and ecology of fishpopulations. Academic Press, London, 293 p.Received March 2010Accepted June 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 320-330

Long-term mean sea level measurements along theBrazilian coast: a preliminary assessmentANGELO TEIXEIRA LEMOS 1,2 & RENATO DAVID GHISOLFI 1,31 Universidade Federal do Espírito Santo, Departamento de Oceanografia, Laboratório de Pesquisa e Simulação Sobrea Dinâmica do Oceano, Av. Fernando Ferrari 514, Goiabeiras, Vitória, Espírito Santo, 29075-910, Fone: +55 27 40097787.2 E-mail: angelolemos@gmail.com3 E-mail: rdrghisolfi@dern.ufes.brAbstract. The main objective of this work is to present a brief historic review focused on sealevel measurements along the Brazilian coast. Furthermore, it aims to describe the protocolsas well as the state-of-the-art and future challenges regarding the mean sea level (MSL)estimates along the Brazilian coast. The Brazilian initiative of measuring the sea level can begrouped into two distinct periods. The first period basically involved the setup and maintenance oftide gauges, focusing on obtaining information for navigation and harbor applications, elaborationof nautical charts and altimetric surveys, which did not require accurate estimates. The secondphase, from the 1990s to date, is marked by an improvement in the establishment ofreference levels (either local or the Vertical Datum) and the creation of PTNG (permanent tidenetwork for geodesy) along with more precise and accurate estimates using continuous GPS(CGPS), gravimeters and altimetry. In conclusion, it is believed that continuation of currentefforts to improve MSL measurements, implementation and maintenance of a geocentricdatum, use of altimetric information and incorporation of geodesic measures as well ascrustal movements control, are technical approaches that allow for the development oflong time-series data appropriate for application in studies regarding effects of climate change onMSL.Keywords: Mean sea level, mean sea level measurement, tide, Brazilian coast, tide gaugeResumo. Medições a longo prazo do nível média do mar ao longo da costa Brasileira:uma avaliação preliminar. O objetivo deste trabalho é apresentar uma breve revisão históricafocada nas medições do nível do mar ao longo da costa brasileira, além de descrever osprotocolos, o estado-da-arte e os desafios futuros em relação a estimativa do nível médio domar (NMM) ao longo da costa brasileira. A iniciativa brasileira de medição do nível do marpode ser agrupada em dois períodos distintos. O primeiro período, envolvendo basicamente ainstalação e manutenção dos medidores de maré, com foco na obtenção de informaçõespara navegação e atividades portuárias, elaboração de cartas náuticas e levantamentos altimétricos,que não requerem estimativas precisas. A segunda fase, a partir da década de 1990 até à dataatual é marcado por uma melhoria no estabelecimento dos níveis de referência (local ou Datumvertical) e a criação da RMPG (rede maregráfica permanente para geodésia), juntamente comestimativas mais precisas e exatas usando medidas contínuas de GPS (CGPS), gravímetros ealtimetria. Conclui-se, então, que a continuidade dos esforços atuais voltados para a melhoriadas medições do NMM, isto é, a implementação e manutenção de um datum geocêntrico, o usode informações altimétricas e a incorporação de medidas geodésicas, bem como o controle demovimentos da crosta terrestre, se constituem em abordagens técnicas necessárias que permitirãoo desenvolvimento de longas séries temporais de dados adequadas para aplicação em estudossobre os efeitos das alterações climáticas no NMM.Palavras-chave: Nível médio do mar, medida de nível médio do mar, maré, costa brasileira,marégrafo.Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

332A. T. LEMOS & R. D. GHISOLFIIntroductionThe mean sea level (MSL) is defined as theaverage of the daily oscillating processes of rise andfall of tides and all the disturbing processesassociated with the meteorological effects andseasonal cycles (Pugh 2004). Understanding thefactors that influence MSL are important owing tothe impact an eventual rise or fall of sea level mayhave on human activities, especially on continentalborders. To evaluate MSL, a multidisciplinaryapproach comprising several oceanic andatmospheric processes of different spatial andtemporal scales like termohaline processes, currents,long waves, meteorology, atmospheric pressure,wind curl, evaporation, precipitation, riverdischarge, crustal movements, tides, glaciology andeustatic changes, is required (Lisitzin 1974). Anypossible change in one or more of these processeshas the ability to deform the marine surface andthus, change the sea level (Dalazoana 2005).According to the 4º Report of Assessment ofthe Intergovernmental Panel on Climate Change(IPCC 2007), the main processes affecting theoceans in a scenario of global warming are seawaterthermal expansion and melting of ice caps.According to the report, the MSL derived from tidegauge records on a global scale during the lastcentury points toward a gradual rise in sea level, andthe projections for the current century indicate anaverage elevation of approximately 1.7 mm year -1 .Despite being one of the main causes of MSLelevation, thermal expansion of oceans is toocomplicated to be measured. Different featuresrespond in different ways to thermal expansion in aneventual warming. For example, when tropical seasurface water is heated, it will expand more easilythan the deep cold waters (Pugh 2004). According toHoughton (2004), if the first 100 m of the ocean,with an average original temperature of 25ºC, showa temperature increase of 1°C, then the local depthwill increase by 3 cm. However, the first 100 m ofthe ocean include the mixing layer, a stratumsusceptible to atmospheric changes, which makes itdifficult to estimate thermal expansion. Thus, tocalculate the increase in MSL, it is necessary to useoceanic-atmospheric coupled models. Some resultsfrom these models show a gradual increase in oceanvolume as a consequence of the observed increase inatmospheric temperature since the last century (Pugh2004). The sea level rose at a rate of approximately0.3-0.7 mm year -1 in the last century (IPCC 2007),and has increased from 0.6 to 1.1 mm year -1 duringthe last decade.Accurate measurement of sea level to theorder of millimeters is a challenging task, mainlydue to the technological dependence of instrumentsand techniques used over the years. There are twomethods for measuring sea level: direct and indirect.Direct measurement is performed in situ usingmetric ruler, tide gauges, reference levels, etc. Incontrast, indirect measurement involves theestimation of sea level from altimetry usingsatellites.In Brazil, the first measurements of sea levelusing tide gauges started toward the end of the 19 thand the beginning of the 20 th century, under theresponsibility of the Navigation and HydrographicBureau (DHN). These estimates were reserved foruse on harbor applications to obtain tidalcomponents, and/or for the elaboration of nauticalcharts. Majority of the in situ measurements lastedfor no longer than a lunar month, covering onespring and neap tide.The main objective of this work is to presenta brief historic review focused on the sea level alongthe Brazilian coast. In addition, it is intended todescribe the protocols as well as the state-of-the-artand future challenges regarding MSL estimatesalong the Brazilian coast. It is important to point outthat this work is focused on the long-term absolutesea level measurements, and not on the relative sealevel within the scale of decades, which is used inengineering surveys with local application.Materials and MethodsEquipment and protocols for measuring MSLSea level mensuration can be carried outdirectly and indirectly. The method is considered tobe direct when the equipment is installed in situ, likethe tide stakes and tide gauges. As these methods arerelatively cheap, easy to handle and do not requiresophisticated technology, they were initially used toassess the MSL. Indirect method of sea levelmeasurement makes use of altimetry satelliteestimates and represents a new technology (in usesince the 1990s) as well as a more accuratetechnique than the direct estimates. Nevertheless,both the methods have their advantages anddisadvantages, and the readers are referred to thereports by UNESCO (1985, 1994, 2002, 2006) forfurther information.In Brazil, direct estimates are generally usedto measure sea level. Nowadays, the two mostcommon and recommended tide gauges are the floattide gauge and the radar tide gauge. The float tidegauge is basically a weight floating inside a tubeimmersed partially in water. The tube prevents thefloat from moving under the action of winds andwaves for short periods. The float keeps itself linkedPan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

Long-term mean sea level measurements along the Brazilian coast333to a cable and one pulley. When the cable isdisplaced on the pulley, an encoder transforms thismovement into a measurement of sea leveloscillation. Despite the equipment being simple toinstall and widely used, it is susceptible to manyerrors, such as sedimentation in the installation site,biological encrustations and crustal movements. Theradar tide gauge is a new technology recentlyapplied in Brazil. The main advantage of this systemis its installation (it is located outside the water) andthus, is not susceptible to temperature and densityoscillations. It measures the distance between theair–sea interface and the equipment through anacoustic signal. However, a major disadvantage ofthis system is the energy demand in case of use inlong-term research.Protocols for measuring MSL using tide gaugesSince tide gauges are being currently used inBrazil to measure MSL, we will further discuss theprotocols associated with this technique.The choice of system to measure sea leveldepends mostly on the purpose for which the data isto be used. Aspects such as costs, accuracy, locationand duration of measurements must be taken intoaccount to make the best decision possible. Forexample, harbor operations demand an accuracy ofabout 0.1 m (Pugh 2004). Hence, there is no need ofthe equipment to be sophisticated and a low cost tidegauge may be used accordingly.Scientific studies focusing on MSL, on theother hand, require a more accurate estimate ofabout 0.01 m. In this case, the installation must havean adequate number of reference levels (RRNNs) tomonitor possible changes in the ruler(s) position, tocarry out a topographic–geodesic monitoring of thereference levels and tide gauges, besides a digitaldata record (Fig. 1). These items will be discussedlater.Nevertheless, once the choice is made, it isessential to follow its basic recommendationsaccordingly to obtain a valid and usefulmeasurement.The use of metric ruler is the simplest wayto measure sea level. Despite being quite susceptibleto positioning mistakes, it is still used as acalibration reference to verify and/or correcteventual vertical displacements of already installedtide gauges.Tide gauges are relatively simple to install,widely known and do not require sophisticatedtechnology to operate. However, there are manyerrors and some disadvantages associated with them.For example, they are for local use (i.e., theinformation is restricted to the point being sampled),are subject to operational errors (e.g., biologicalincrustation), crustal movements, meteorologicalfactors, geographic positioning of the levels’references, etc.There are many types of tide gauges. Theycan be classified, for example, according to thephysics employed to obtain the information. Thereaders can refer to UNESCO (2002) for a completelist of tide gauges and their operating systems. Toensure accuracy of results, all of them commonlyfollow the same basic installation requirements. Anymeasurements of height should have a referencelevel relative to that plane. A reference plane is welldefined locally on a stable surface, free of anyinfluence of vertical and horizontal movements,Figure 1. Scheme of the system of measurement using a tide gauge. Adapted from UNESCO (2002).Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

334A. T. LEMOS & R. D. GHISOLFIerosion, sedimentation, weathering, etc. Usually, arock is used as a stable surface. When greateraccuracy is required, or even for safety reasons incase one or more of them are destroyed, it is stronglyrecommended to use more than one reference plane(three, at least; UNESCO 1985). These local RRNNsare called Local Fixed Datum. Having more thanone reference level guarantees the long term stabilityof a sea level time-series. In Brazil, the majority oftide gauges in operation are located in the harborzones that are currently expanding their installationareas, and hence, it is quite possible to have some ofthe RRNNs destroyed during this process. The maindisadvantage of this system (having multipleRRNNs) is that it is susceptible to crustalmovements that may cause errors to the order ofmillimeters. Such errors must be removed usingcontinuous GPS (CGPS) and gravimeter informationif the data is to be used in studies regarding MSLvariation due to climate change. CGPS is a GPS thatrecords information continuously so that one caninfer horizontal and vertical crustal movements.The local datum RRNN should also bereferenced to the Vertical Datum. The VerticalDatum can be defined as a standard mark to whichany height measurement over the national territory isreferenced.According to UNESCO (1985), the localdepthtide gauge installation site should be at least 2m below the surface in a low-tide regime. Estuaries,promontories, straits, and low-tide impoundmentzones should, if possible, be avoided. Besides theRRNNs, the tide gauges should also have their ownreference surface level, the so-called zero tide gauge(ZTG). The ZTG is a horizontal surface thatindicates the mark zero. This mark can be any pointlocated below the equipment. Table I summarizesthe basic procedures that must be taken into accountwhen a sea level measurement system based on tidegauges is set up. The readers are referred to thereports by UNESCO (1985, 1994, 2002) for moredetailed information.ResultsThe Brazilian initiative of measuring MSLThe first sea level measurements date backto the beginning of the last century, between 1910and 1920, under the responsibility of DHN andINPH (National Institute of Hydrologic Research).Initially, the focus was toward navigation and harborapplications, elaboration of nautical charts andaltimetric surveys (Neves 2005). After the creationof Portobras (Brazilian Harbor, or ‘PortosBrasileiros’), the INPH became responsible for theinstallation and maintenance of all the equipmentinstalled in harbors.Over the decades, a total of 281 sitesthroughout the Brazilian coast, and a few offshoreones, were sampled (Fig. 2). Most of the samplingdid not last for more than a month and were carriedout during the 1970s. As described earlier, the datawere used on specific applications, mainly harboractivities, and determination of tide components andamplitude.Almost simultaneously, another set ofinformation about the MSL was used to establish theBrazil Vertical Datum. Between 1919 and 1920, theextinct Brazilian General Chart Commissionoperated a tide gauge in the city of Torres in the RioGrande do Sul state.Despite the fact that such information nolonger exists, those observations were referenced toa geodetic mark of the Geographic Service Board (orTable I. Basic procedures to be considered when a tide-gauge sea level measurement scheme is set up. Adapted fromUNESCO (1985, 1994, 2002).CharacteristicDescriptionInstallationLocals of strong erosion, sedimentation, and hydrodynamics must be avoidedReference levelsAt least 3, arranged radiallyRulesShould be used for calibration and operational control purposeVertical DatumAll the RRNNs should be referenced to the DatumCGPSThe equipment should be used for continuous monitoring of tide gauges’ positionsTopographic–gravimetric control It should be used to monitor the gravitational field of the installation siteData recordsMust be digitalSamplingAt least hourly, although high-frequency sampling is recommendedPhysical protectionBuilt around the tide gauge to prevent damagesAccuracyMinimum 0.01 m (Scientific studies)Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

Long-term mean sea level measurements along the Brazilian coast335Figure 2. Geographic distribution of sites where tidegaugesmeasurements were carried out during the lastcentury (red dots). Adapted from FEMAR (2000).‘Diretoria do Serviço Geográfico – DSG’), whichwas included in a new research leveling linebetween the city of Criciuma (Santa Catarina state)and Torres. As a result, a provisory Vertical Datum,called Torres Datum, was established. Thedetermination of Vertical Datum had alreadyattracted the attention of the Brazilian GeodesySystem (SGB). Consequently, the Altimetry HighPrecision Net (or ‘Rede Altimétrica de Alta Precisão– RAAP’), created in 1945, opted to establish theDatum of Torres, which marks the origin of allaltimetric measurements. Thereafter, the GeographicNational Council (presently, Brazilian Statistic andGeography Institute – IBGE) began proceedings toconnect the RAAP datum with the several tidegauges spread along the Brazilian coast. In 1952, thefirst geographic readjustment of the Datum of Torreswas established. By that time, more than 5000RRNNs had been set near the Brazilian tide gauges.In 1959, the last readjustment of the Datum tookplace when it was moved definitely to the city ofImbituba (Santa Catarina state). The new VerticalDatum – Imbituba Datum – was carried out by theInter-American Geodetic Survey after 9 years ofobservations, extending from 1949 to 1957.According to Franco (1988), the minimumrecommended duration is 19 years. In the followingdecades (especially after 1970), there was anorthward and mainland expansion regarding theestablishment of RRNNs referenced to the Datum ofImbituba.By the year 1960, a number of importanttide-gauge stations with long data set records beganto be deactivated. Recognizing the importance ofhaving continuous and detailed records of sea level,in 1976, the IBGE considered the possibility ofreactivating the tide gauge stations that werepreviously under the responsibility of IAGS.Between 1980 and 1986, the IBGE realized some reevaluationof the Imbituba Datum. In 1993, theIBGE began to monitor sea level using tide gauges.This was done experimentally at the ‘EstaçãoMaregráfica Experimental de Copacabana’, andlasted for one year when the gauge was destroyed bya storm surge. After a year of observation, it wasconcluded that sea level variation could be biasedowing to instrumental deviation and vertical motionsof the RRNNs.In 1994, the IBGE took over theconventional tide gauge located at Porto de Imbetibain the city of Macaé (Rio de Janeiro state) fromPetrobras. Based on previous experience, IBGEupgraded the station, that is, employed measurementredundancy by operating two tide gauges withdistinct principles of functioning, to avoid lack ofinformation due to instrument failure. Inaddition, by upgrading the method of data storage, itwas possible to make the information available inreal time. As a result, the Macaé station hasbecome a pilot station for the future PermanentTide Network for Geodesy (PTNG) (or‘Rede Maregráfica Permanente para Geodésia –RMPG’).The PTNG was created in 1997 with a clearpicture regarding the location of tide gauges. Thelocations chosen were in the following cities:Imbituba (Santa Catarina), Macaé (Rio de Janeiro),Salvador (Bahia), Fortaleza (Ceará), and Santana(Pará). The network became operational effectivelyfrom 2001, after the installation of digital equipmentin Macaé and Imbituba.The main goal of this network is to provideinformation to correlate the Datum of Imbituba withother sea level (tidal) references. The demand forsuch information was made clear in a study carriedout by Alencar (1990). In his study, the authorcompared the local MSL with that referenced to theDatum of Imbituba (Table II).Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

336A. T. LEMOS & R. D. GHISOLFITable II. MSL discrepancies between that measuredlocally and the reference level transported mechanicallyby geometric operations from Imbituba (SC). Adaptedfrom Alencar (1990).Tide gauge stationsGapTorres (Rio Grande do Sul) +0.0584Itajaí (Santa Catarina) +0.1399Paranaguá (Paraná) +0.0010Rio de Janeiro (Rio de Janeiro) +0.1237Vitória (Espírito Santo) +0.2840Fortaleza (Ceará) +0.2923Belém (Pará) +0.8808Positive values indicate that the local MSLis above the MSL referenced to Imbituba. Accordingto Alencar (1990), the differences were accountedfor by the errors in measuring sea level locally aswell as by the instrumental, operational andgravimetric errors associated with the procedure oftransporting the reference from Imbituba. Othercauses may be due to meteorological andoceanographic factors, such as sea-surfacetemperature and salinity, and the lack of accurateinformation about MSL.To achieve the main PTNG goal, it isnecessary to repeat the leveling operations betweenstations using gravimetric information, and controlthe horizontal and vertical movements using CGPSestimates. These procedures will allow for ameticulous determination of the tide-gauge RRNNaltitude as well as help in identifying the crustalmovements, because any movement introduces errorin the MSL estimates.The PTNG is also a part of the SIRGASProject (Reference System to the Americas(‘Sistema de Referência para as Américas’), a studythat began in 1993. The project aims to define aunique reference system for the whole of SouthAmerica, to establish and maintain a referencenetwork among the South American countries, aswell as to define a geocentric Datum. The readerscan refer to SIRGAS (1997) for more details aboutthe project.Until 1993, every measurement assigned tothe Datum and to the RRNN were (and most, or all,of them still are) topocentric. According to Freitas etal. (2002), the geocentric positions of the tidegauges serve as an initial condition to associate theMSL with global geoids. Thus, the contribution ofSIRGAS to the studies on MSL comprises producingaccurate positioning of the tide gauges and RRNNs,and their respective referencing to the newgeocentric Vertical Datum.The MSL topography, defined as thedistance between the MSL and the geoid, shouldalways be used to correct the tide gaugeobservations (Luz et al. 2008). The MSL topographyresults from the almost continuous action of manymeteorological and oceanographic factors active onthe sea surface, mainly on the shallow coastalregions. Thus, it is necessary for each tide gaugestation to determine a specific value for the MSLtopography, which was not possible until after 1995,making it difficult to correlate the MSL of theVertical Datum region with that measured by thetide gauge stations along the coast. The technologythat enables accurate estimates of MSL topographyis altimetry satellite. The altimetry information wasavailable since after the launch of a Franco-American mission carrying the TOPEX/Poseidon(T/P) sensor in 1992 (Dunbar & Hardin 1992). In2005, the T/P mission finished its operations, whenit was substituted by the Jason-1 satellite, launchedin 2001. Despite the importance of the altimetrysatellite in determining MSL topography, theestimates along the coastal regions are subject tosome atmospheric and geophysical corrections oflow accuracy. Nevertheless, Bosch & Savcenko(2007) used T/P and Jason-1 observations between2002 and 2005 to estimate MSL topography inglobal coastal areas, using one-dimensional spectralfiltering. Luz et al. (2008) applied the samemethodology to the south-southeast Brazilian coast,with the purpose of solving the problem ofintegrating the RMPG results with the RRNNs. Theresults obtained showed some problems inestimating the MSL topography, mainly due to thespatial extension of the filter used to make thesatellite and geoid model estimates compatible.However, the authors suggested a further studyusing the EGM-2008 (Earth Gravitational Model –US – National Geospatial-Intelligence Agency –NGA) model, which allows for a better spatialresolution of the geoid model in shallow waters.Matching deep-water sea level estimates and in situtide gauge measurements, obtained along thesatellite tracks, is also one of the objectives of thePTNG.Present state of sea level measurements along theBrazilian coastDuring the last century, some effort has beenmade to collect sea level information along the Braziliancoast. This aspect has been briefly reviewed inPan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

Long-term mean sea level measurements along the Brazilian coast337the previous section. Unfortunately, the majority (ifnot all) of the tide gauges, once active, are either notoperational or have been destroyed. An exception isthe Cananéia station, where the time-series is morethan 50 years long. According to Pirazzoli (1986),who analyzed the data of long-term variations ofMSL measurement from a data set available in thePermanente Mean Sea Level Institute (PMSLI), therate of variation of the MSL followed a period of 20years. Hence, the studies on long-term tendencyshould have at least 50 years of data.Presently, according to their own needs forsuch information, universities, private institutions(industries), and public institutions or agencies (e.g.,IBGE, INPE – Space Research National Institute,CHM – Navy Hydrographic Center) are (or arestarting to) conducting long-term in situ sea levelmeasurements. As a result, the effort of measuringsea level seems to be pulverized and uncoordinated,and it is not uncommon to observe gaps ordiscontinuity in the sea level time-series. Table IIIpresents the active or planned (yet to install) tidegauges along the Brazilian coast and oceanic islands,and Figure 3 shows their spatial distribution. It isimportant to point out that a majority of the alreadyactivestations shown in Figure 3 have not beenplanned for sea level change studies owing toclimate change.Table III. GLOSS-Brazil stations (Global Sea Level Observing System). Adapted from CHN (2009). Stations markedby * are part of the PTNG program.ExpectedStation Responsible SituationObservationssituation in 2010Rio Grande (Rio Grandedo Sul)FURG-CHM To be installed Yet to be installed Radar tide gauge*Imbituba (Santa Catarina) IBGE Operating OperationalPressure tide gauge since 2001,CGPS from Dec 2006Cananéia (Santa Catarina) IOUSP Operating Operational Radar tide gaugeIlha Fiscal (Rio de Janeiro) CHM Operating Operational*Macaé (Rio de Janeiro) IBGE Operating OperationalBarra do Riacho (EspíritoSanto) / Transfering toVitória (Espírito Santo)*Salvador (Bahia)*Fortaleza (Ceará)Ponta da Madeira(Maranhão)Ilha da Trindade (EspíritoSanto)Fernando de Noronha(Pernambuco)Arquipélago de São Pedroe São Paulo (Rio Grandedo Norte)PORTOCEL/VALEIBGE (CHM)IBGEValeRadar tide gauge, conventionaltide gauge (backup)Pressure tide gauge, since July2001, no CGPS stationUnder test Operational Pressure tide gaugeOperating(underevaluation)Operating(underevaluation)OperatingOperationalOperationalOperational nearreal-timeautomatic datatransmissionINPE-CHM To be installed Under evaluationINPE-CHM To be installed Under evaluationINPE-CHMInstalled in2008 (undertest)Operational*Santana (Pará) IBGE Operating OperationalRadar tide gauge since Apr 2008;CGPS from Apr 2007; datatransmitting in real time throughsatelliteRadar tide gauge since Apr 2008;CGPS from Oct 2008; datatransmitting in real time throughsatelliteConventional tide gauge, radar tobe installedRadar tide gauge to be installed in2010Radar tide gauge to be installed in2010Radar tide gauge; datatransmitting in real-time throughsatelliteRadar tide gauge since Dec 2007;CGPS from July 2008Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

338A. T. LEMOS & R. D. GHISOLFIFigure 3. Location of the tide gauge stations along theBrazilian coast (red dots) included in the GLOSS-BrazilProgram.DiscussionIn the Brazilian context of sea levelmeasurements, two distinct periods can beestablished. The first basically comprised the setupand maintenance of tide gauges, focusing onobtaining information for harbor purposes, whichdid not require accurate estimates. This period endedby the late 1980s. The second phase that spans todate is marked by an improvement in theestablishment of reference levels (either local orVertical Datum) and the creation of PTNG for moreprecise and accurate estimates using CGPS,gravimeters and altimetry.Although there are many tide gauge stationsdistributed along the Brazilian coast, a majority ofthem do not have an associated time-series of morethan 10 years. Time-series longer than 30 years arerestricted to the stations of Cananéia (São Paulostate) and Ilha Fiscal (Rio de Janeiro state). From the1990s, these stations started implementing atopographic–geodesic control and installed digitalequipment like the CGPS, covering a relatively shortperiod of accurate data compared with the periodwhen tide gauges were operational. The availabletime-series is shorter than that stipulated to study theimpact of climate change on MSL. Douglas (1991)and Pirazzoli (1986) affirmed that the time-seriesshould be of at least 50 years duration, to draw anyspecific conclusion regarding the sea level change.Despite the results obtained by Douglas(1991), there have been some studies focused onevaluating the tendency of MSL along the Braziliancoast. França (1995) used tide gauge informationfrom 1950 to 1990 from some Brazilians stations(Belém-Pará, Salinópolis- Pará,, Fortaleza-Ceará,Recife-Pernambuco, Salvador-Bahia, Canavieiras-Bahia, Rio de Janeiro-Rio de Janeiro, Ubatuba-SãoPaulo, Cananéia-Santa Catarina, and Imbituba-Santa Catarina) to verify the trend of MSLelevations. He found the tendency of elevation tobe approximately 4 mm year -1 or about 50 cmcentury -1 .Furthermore, the work carried out byAubrey et al. (1988) also used the information fromtide gauges to estimate MSL oscillations on theBrazilian coast, and concluded that the nationaltendency is of elevation. Both studies represent thefirst attempt to estimate the MSL changes along thecoast. However, as stated previously, theinformation used in these studies had neithertopographic–geodesic control nor correction fromCGPS to eliminate the possible vertical andhorizontal displacements of the crust.Two problems can be associated with thecurrent implemented system of using tide gauges formeasuring sea level based on climate changes. Thefirst problem is the use of different tide gauges alongthe coast (distinct physical principles ofmeasurement); they might have different samplingrate and/or methods of digital record. Furthermore,the use of CGPS and gravimeter, the errorsassociated with the leveling using differenttechniques and the incompatibility of data fromdifferent types of equipment (cannot be usedstatistically) also pose a problem. Neves (2005)reported that to maintain a long-term MSLmeasurement, a rigorous assessment of the internalmonitoring network, identification of the operationand its responsibilities, and a solid organization tomaintain the network operation to continue dataassimilation along the years are necessary.The second problem is related to themaintenance of the station (which demands longtermfinancial support to maintain the functioning ofthe network), the control and processing of the datacollected and its availability to the general public.These two aspects may be the cause of gaps in thehistoric time-series and the lack of sufficiently longsea level time-series to be used in studies regardingthe impact of climate change on MSL.Despite the efforts made by IBGE, CHN andother universities since the 1990’s to obtain sea levelPan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

Long-term mean sea level measurements along the Brazilian coast339data, Brazil is still in the early stages of developing asafe, precise, accurate and long-term sea-level timeseriesaccording to the standard protocols. Animportant initiative to create a realistic network formonitoring MSL is the Fluminense Tide GaugeNetwork (or ‘Rede Maregráfica Fluminense –RMFlu’). This network was created in 1995 tocontrol and support MSL measurements in Rio deJaneiro state. Currently, several organizations andinstitutions, like COPPE/UFRJ (Federal Universityof Rio de Janeiro), IBGE, Petrobras, DHN, CHM,IEAPM (Marine Research Institution AlmirantePaulo Moreira) and Electronuclear S.A., constitutethe RMFlu.Nevertheless, the most recent and completecollection of MSL information and observations isincluded in the GLOSS-Brazil Program and PTMGproject. According to the Marine HydrographicCenter (CHM-DHN), currently, only the tide gaugestation of Imbituba (since December 2006),Cananéia (since January 2006), Salvador (sinceApril 2007) and Fortaleza (since October 2008) havethe CGPS installed and none of the GLOSS-Brazilstations have a gravimeter installed (Table III). Inaddition, they are mostly of the conventional (float)and pressure type. However, some stations like thestations of Ilha Fiscal, Cananéia, Vitória, Salvador,and Fortaleza, use the radar technology. Only two ofthe sites shown in Figure 3, Salvador and Fortaleza,have their measurements available online (data canbe accessed on www.vliz.be/gauges). Most of theother stations have their information supposedlycontinuously uploaded (gaps in the records arecommon) to the international data centers likeUHSLC (University of Hawaii Sea Level Center)and PSMSL. Finally, the application of altimetrytechnology on coastal regions is fundamental tostudies regarding MSL changes since a geocentricReferencesAlencar, J. C. M. 1990. Datum Altimétrico Brasileiro.Caderno de Geociências, 5: 69-73.Aubrey, D. G., Emery, C. O. & Uchupi, E. 1988.Changing coastal levels of South Americaand the Caribbean region from tide-gaugesrecords. Tectonophysics. 154: 269-284.Bosch, W. & Savcenko, R. 2007. Estimating the seasurface topography - Profile approach witherror examination. In: "Earth, OurChanging Planet". IUGG XXIV GeneralAssembly.CHN - Centro Hidrográfico de Navegação. 2009.Technical Report of Brazil: The GLOSS-Brazil Program, 9 p.Dalazoana, R. 2005. Estudos dirigidos à análiseDatum has to be used as the reference for MSLmeasurement.ConclusionOwing to the inconsistency of the local datadistributed along the coast in relation to theBrazilian Vertical Datum, most of the tide gaugemeasurements carried out along the coast are notstrictly accurate. This is because of the absence ofaccurate MSL topography estimates in shallowwater regions. In addition, the fact that most of theMSL time-series do not have CGPS and gravimetermeasurements (only recently, some stations haveinstalled a CGPS) and the absence of a central organresponsible for managing the implementation,control and maintenance of the tide gauges,contribute to the current scenario where Brazil isstill not able to precisely inform the impact of globalwarming on MSL. Hence, it can be concluded fromthe present work that these two aspects are theprimordial issues that should be addressed.Nevertheless, it is believed that the continuity of thecurrent efforts to improve MSL measurements, theimplementation and maintenance of a geocentricDatum, the use of altimetric information, theincorporation of geodesic measures, as well as thecontrol of crustal movements are approaches thatallow for the development of long time-series datathat can be applied to studies on the effects ofclimate changes on MSL.AcknowledgmentThe authors thank the Conselho Nacional deCiência e Tecnologia (CNPq) for the financialsupport, Capitã de Corveta Rosuita Helena Roso(CHN) and Dr. Roberto Luz from PTNG-IBGE fortheir assistance and valuable contribution during thepreparation of this study.temporal do Datum vertical brasileiro. MSc.Thesis. Universidade Federal do Paraná.202 p.Douglas, B. C. 1991. Global sea level. JournalGeophysical Research, 96: 6981-6992.Dunbar, B. & Hardin, M. 1992. Mission to PlanetEarth: TOPEX/POSEIDON. NASA/CNES,14.FEMAR. 2000. Catálogo de estações maregráficasbrasileiras. Fundação de estudos do Mar,Rio de Janeiro, 280 p.Franco, A. S. 1988. Tides - fundamentals, analysisand prediction. FCTH, São Paulo, 249 p.França, C. A. S. 1995. O litoral brasileiro - Estudossobre o Nível Médio do Mar. InstitutoPan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

340A. T. LEMOS & R. D. GHISOLFIOceanográfico. Universidade de SãoPaulo, 21 p.Freitas, S. R. C., Schwab, S. H. S., Marone, E.,Pires, A. O. & Dalazoana, R. 2002. Localeffects in the Brazilian vertical datum. IAGSymposia 125 - Vistas for Geodesy in theNew Millennium.Houghton, J. 2004. Global Warming: TheComplete Briefing. (3 Ed.). CambridgeUniversity Press.IPCC, 2007. Contribution of Working Groups I,II e III to the Fourth Assessment Reportof the Intergovernmental Panel onClimate Change. Core Writing Team,Pachauri, R. K. and Reisinger, A. (Eds),Geneva, Switzerland, 104 p.Lisitzin, E. 1974. Sea Level Changes. ElsevierScientific Publishing Company, Amsterdam,335 p.Luz, R. T., Bosch, W., Freitas, S. R. C. & Heck, B.2008. Topografia do Nível Médio do Mar noLitoral Sul-Sudeste Brasileiro. II SimpósioBrasileiro de Ciências Geodésicas e Tecnológicasda Geoinformação. Recife-PE.Neves, C. F. 2005. O nível médio do mar: Umarealidade física ou um critério deengenharia? Vetor, 15(2): 19-33.Pirazzoli, P. A. 1986. Secular trends of relative sealevel(RSL) changes indicated by tidegaugesrecords. Journal of CoastalResearch, SI(1): 1-26.Pugh, H. Pugh, H. P. 2004. Changing Sea Levels:Effects of Tides, Weather and Climate.Southampton Oceanography Center, UK.SIRGAS - Sistema de Referência Geocêntrico para aAmérica do Sul. 1997. Relatório Final:grupos de trabalho I e II. IBGE.Departamento de Geodésia - Rio deJaneiro, 99 p.UNESCO. 1985. Manual de Medição e Interpretaçãodo Nível do Mar: Volume II - TecnologiasEmergentes. Comissão OceanográficaIntergovernamental, 52 p.UNESCO. 1994. Manual de Medição e Interpretaçãodo Nível do Mar: Volume II - TecnologiasEmergentes. Comissão Oceanográfica Intergovernamental,52 p.UNESCO. 2002. Manual de Medição e Interpretaçãodo Nível do Mar: Volume III – Reavaliaçõese Recomendações. Comissão OceanográficaIntergovernamental, 55 p.UNESCO. 2006. Manual de Medição e Interpretaçãodo Nível do Mar: Volume IV - An Update to2006. Comissão Oceanográfica Intergovernamental,88 p.Received January 2010Accepted January 2011Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340

Vulnerability and impacts related to the rising sea level in theMetropolitan Center of Recife, Northeast BrazilMIRELLA B. S. F. COSTA 1 , DANIELE L. B. MALLMANN 2 ,PATRÍCIA M. PONTES 2 & MOACYR ARAUJO 1*1 Centro de Estudos e Ensaios em Risco e Modelagem Ambiental – CEERMA, Universidade Federal de Pernambuco(UFPE). Av. Arquitetura, s/n – Cidade Universitária. CEP 50740-550 - Recife - PE - Brasil.2 Laboratório de Oceanografia Geológica do Departamento de Oceanografia – LABOGEO/DOCEAN, UniversidadeFederal de Pernambuco (UFPE). Av. Arquitetura, s/n – Cidade Universitária. CEP 50740-550 - Recife - PE - Brasil.*corresponding author. E-mail: moa@ufpe.brAbstract. The Metropolitan Center of Recife figures as one of the most vulnerable regions to a rise insea level along the Brazilian coast, due to its physical characteristics and to various problems related toflooding and coastal erosion. The analysis of potential flood zones and vulnerability assessments wasbased on an empirical approach, considering the estimates made by the IPCC on sea level rise andextreme scenarios of astronomical ride, storm surge and run up for the region. The results indicate thatfor a 0.5 m (optimistic scenario) rise in sea level, at least 39.32 km 2 of the area of the municipalitieswould become potential flood zones. In a scenario of critical sea level rise (1 m), this figure wouldincrease to 53.69 km 2 . Analysis of the entire coast indicates that 81.8% of urban constructions situatedless than 30 m from de shoreline and located 5 m below ground level would be severely affected bychanges in sea level. Currently 45.7% of the coast is considered a high vulnerability area. In view of thesevere losses predicted by the simulated scenarios, response strategies identifying the most appropriateadaptation options must be developed.Key words: coastal flooding risk, sea level rise, Northeast BrazilResumo: Vulnerabilidade e impactos relacionados a subida do nível do mar no centrometropolitano do Recife, nordeste do Brasil. O Centro Metropolitano do Recife figura como uma dascidades mais vulneráveis ao aumento do nível do mar do litoral brasileiro, devido às suas característicasfísicas e aos diversos problemas referentes a inundações e erosão costeira. A análise das zonaspotencialmente inundáveis e da vulnerabilidade costeira foi baseada numa abordagem empírica,considerando as estimativas realizadas pelo IPCC sobre a elevação do nível do mar e os cenáriosextremos de maré astronômica, ressaca e run up. Os resultados indicam que diante de um aumento donível do mar da ordem de 0,5 m (cenário otimista), é esperado que, pelo menos 39,32 km 2 da área dosmunicípios analisados constituam zonas potencialmente inundadas. Num cenário crítico de elevação donível do mar (1 m), este valor aumentaria para 53,69 km 2 . A análise da costa como um todo indica que os81,8% das construções urbanas, que estão a menos de 30 m da linha de costa e em terrenos abaixo de 5m, deverão ser rapidamente atingidos pela mudança no nível do mar atual e que o litoral possui 45,7% desua extensão sob zona de alta vulnerabilidade. Frente aos cenários simulados, aponta-se a necessidade deum planejamento público para mitigação dos futuros impactos.Palavras-chave: risco a inundações costeiras, elevação do nível do mar, nordeste do BrasilIntroductionThe last report of the IntergovernmentalPanel on Climate Change concluded that the planetclimate is unmistakably experiencing rapid heatingand this is partially due to human activities (IPCC,2007). According to this and other recent studies(Neves & Muehe 1995, Thieler & Hammar-Klose1999, Alfredini et al. 2008, Snoussi et al. 2008,Vargas et al. 2008), the rising sea level scenarios fornear future are of great concern. The acceleration ofthe rate in which the sea level rises not only raisesthe possibility of intensified impacts – such ascoastal erosion, habitat losses and saline intrusion inPan-American Journal of Aquatic Sciences (2010) 5(2): 341-349

342M. B. S. F. COSTA ET ALLIcoastal aquifers and rives – but could also result incomplete suppression of sandy beaches and wetlands(Snoussi et al. 2007).Such an impact, although felt globally,depends on local peculiarities; therefore, happensunevenly across countries, regions, communitiesand individuals as the result of different levels ofexposure and vulnerability (Clark et al. 1998).Moreover, adapting coastal regions represent alarger challenge to developing countries than itdoes to developed ones, because of economiclimitations. As an option for minimizing sucheffects, vulnerability and impact assessmentprovides a starting point for guidance in decisionmaking for effective measures that envisionimpact reduction and reestablishment of initialconditions.In Brazil, sea level trends for differentplaces are sparing. For Recife, specifically, Harariet al. (2008) pointed a sea level rise of about5.6 mm.year -1 . In spite of this gap, some studieshave been dedicated to the coastal predictions infront of future sea level sceneries, as Alfrediniet al. (2008), which describes islands and coastalurban areas as the most vulnerable to floodingin medium and long terms. Another study, madeby Marengo et al. (2007) in national scale,pointed Pernambuco as one of the most affectedstates by the increasing on the sea level, atthis moment and potentially. The same conclusionis showed by two another studies: Neveset al. (2007) and Naccarati (2008). According tothese authors, both, natural and occupationcharacteristics contribute for the vulnerability of thearea.The Metropolitan Center of Recife combineslow topography, intense urbanization, highdemographic density, and elevated ecological,tourist and economic values (Araújo et al. 2009).Furthermore, it presents several conflicts in coastalland and shoreline usage, which is one of thereasons it became one of the first regions in Brazil tobe the subject of integrated studies on theproblems of coastal erosion, with the collaborationof several spheres of public power (FINEP/UFPE,2008).Within this context, the goal of this paperis to present possible scenarios for MetropolitanCenter of Recife, prompted by the rise in theaverage sea level trend forecasted by the IPCCemission scenarios. The aim is to foresee scenariosthat will raise awareness in the decision makers, aswell as to point out possible strategies forminimization of potential impacts of the rise in sealevel in this region.MethodologyDescription of study areaThe Metropolitan Center of Recife, whichcovers the cities of Paulista, Olinda, Recife andJaboatão dos Guararapes, is located on the coast ofthe state of Pernambuco (Fig. 1) and extends over asedimentary plain, with an average altitude of 4 m.The region presents Tropical Atlantic climate withan average annual temperature of 27oC and annualpluviosity of around 2000 mm unequally distributedbetween dry and rainy periods. The wind regime isgoverned by the general atmospheric pressuredistribution pattern of the South Atlantic Ocean,with predominance of SE winds.The tides registered in the region are allsemi-diurnal and classified as mesotidal in terms ofamplitude. The largest and smallest swells occur inthe months of September and January, respectively,with significant wave periods between 5.1 and 6.8seconds. The significant wave height varies betweeneach location, with values of 0.29/0.27 for Paulista;0.6/0.61 for Olinda; 0.97/0.66 for Recife and0.61/0.44 for Jaboatão dos Guararapes (FINEP/UFPE, 2008).The main ecosystems in this region aremangroves, fragments of Atlantic Forest, coral reefs,and restinga. The hydrographic system is welldrained, with several rivers, streams and lakes. TheCapibaribe River is the main watercourse.Establishing Potential Flood ZonesThe methodological steps proposed byHoozemans et al. (1993), Snoussi et al. (2007) andVargas et al. (2008) were followed, withdetermination of the resulting flood levels by thesum of the factors involved in the sea level rise.Therefore, the contributions of the followingextreme water level-induced phenomena weresimulated: maximum high water line in the last 20years; sea level rise due to storm surge(meteorological tide plus wave run up)(FINEP/UFPE, 2008); projected sea level rise for theFigure 1. Location of the studied coastal area in theMetropolitan Center of Recife, Northeast Brazil.Pan-American Journal of Aquatic Sciences (2010) 5(2): 341-349

Vulnerability and impacts related to the rising sea level343next century (IPCC, 2007), which can be optimistic(+0.5 m by 2100, or about 5 mm·year -1 ) or critical (+1.0 m by 2100, about 10 mm·year -1 ). These scenarioswere adapted from the IPCC Emission Scenarios(B1 and A2), considering that the sea level rise forthe next century may be up to roughly twice themaximum projection found in this report (Rohling etal. 2008). Equations (I) and (II) express theempirical approach used to determine the floodlevels in coastal and estuary areas, respectively.NI = PM + MM + SLR (Eq. I)NI = PM + SLR (Eq. II)where NI – Flood Level; PM – maximum high tidewater level; MM –Meteorological Tide (includingwave run up); SLR – Sea Level Rise.For delimitation of potential flood zones, aGeographic Information System (GIS) was used.Rectified aerial photography in a 1:2000 scale(Pernambuco, 1974) was processed in order todigitalize the flood level lines. The entire territorialportion below such levels, therefore, wouldpotentially be flooded in similar conditions to theones in the above modeled scenarios.The rectified aerial photography employs theaverage sea level in Imbituba, South Brazil,established by the Brazilian Institute of Geographyand Statistics (IBGE). According to Neves & Muehe(1995), it is located 1.106 m above the datum usedby the Hydrography and Navigation Board (DHN).The values considered and the resulting levels are inTable I.Attribution of Degrees of Vulnerability to theShorelineBearing in mind that sandy beaches serve asa buffer, reducing the energy that the oceans imposeto the shore, one can say that the closer the firstman-made structures are to the shoreline, the highertheir degree of vulnerability to the impacts oferosion and floods (Lathrop Jr. & Love, 2007).Accordingly, the degree of vulnerability wasassigned based on the current capacity of thebeaches of the Metropolitan Center of Recife toprotect inland resources, be they economic,ecological or of cultural values.To calculate the average distance betweenthe maximum high water line and the closestinfrastructure, a spatial analysis module available ina GIS environment was used. The line of first manmadestructures was taken from UNIBASE (2002) –a cartographic base on a 1:1.000 scale – and theshoreline was obtained in situ using geodesic GPSequipment. The marker used was the High WaterLine, or HWL.Once the distances were calculated, limitvalues were established for each degree ofvulnerability (Table II). In order to fulfill the safetyconditions, beaches should not only present abackshore, but this sector should also be at least 30m wide (FINEP/UFPE, 2008). This value wasestablished through an estimation of the resultsobtained by Pajak & Leatherman (2002) and Costaet al. (2008) on the variation of the HWL positionthrough time.Analysis of the Coastal Resources at RiskIn a GIS environment, a layer containing thepotential flood zones was superimposed to the layercontaining occupation information. The resources atrisk were divided into the following categories:wetland (mangrove and flooded areas), unoccupiedland, beaches, buildings with over three floors,buildings with less than three floors andhistorical/cultural patrimony.Table I. Values considered for the estimation of the levels in risk of flooding.ScenarioMaximum high tidewater level (m)Storm surge effect SLR (m) Flood level (m)Optimistic 2.7 (– 1.106) 1.0 0.5 3.1Critical 2.7 (– 1.106) 1.0 1.0 3.6Table II. Criteria used in the determination of the Degree of Vulnerability.Backshore width (m) Degree of vulnerability Maintenance priorityNull* Conditional Constant maintenance>30 Low Moderate

344M. B. S. F. COSTA ET ALLIResultsEstablishing Potential Flood ZonesJaboatão dos Guararapes presents the largestarea among the studied cities, with 256 km 2 . Thearea considered in this analysis was 15.45 km 2 . Thepotential flood zone is shown in the following figureas polygons for the benign and critical scenarios(Fig. 2). The most important changes in bothscenarios are seen in the areas surrounding thePirapama-Jaboatão Estuarine Complex and also onthe coastal strip positioned further south, on theBarra de Jangadas beach. Attention is drawn toPaiva Beach (Praia do Paiva), a sandy spit extremelyvulnerable to the possibility of a rise in sea level. Ofthe 0.97 km 2 that form the extremity of the sandyspit, 0.30 km² would be left above water in ascenario with a 0.5 m sea level rise. In a criticalscenario – 1m rise in the sea level – only about 0.15km² of the initial area would be left.It is worth while to observe that throughoutpractically the entire coast – except the portion infront of the Candeias beach, protected by abreakwater, and a small section of the Barra deJangadas beach – the beach system is no longer anefficient means of coastal protection since one ormore sectors have been suppressed. Nonetheless,there is still an estimated sandy beach loss of at least13.71 m² as a result of coastal erosionRecife is the city with lowest altitude amongthe ones studied here, with an average of only 4 mabove sea level. As a result, in a scenario of a 0.5 mFigure 2. Potential flood zones in Jaboatão dosGuararapes.Figure 3. Potential Flood Zones in Recife.sea level rise, it is estimated that the flooded areawould amount to about 25.38 km 2 ; while in thecritical scenario, with a 1 m rise in sea level, itwould be approximately 33.71 km 2 (Fig. 3).The city of Olinda is located on the highestaverage altitude among the cities studied, at roughly16 m above sea level. Consequently, the forecast isthat the most affected areas would be those situatednear the city boundary with Recife, in the lowlands,bathed by the Pina Basin (Fig. 4).According to the flood levels isolinesanalysis of the Del Chifre beach, better known as theOlinda isthmus, and considering a progressive floodof the spit, due to a rise in the level of flooding,there is a possibility that this sandy spit may developinsular characteristics, with an estimated areabetween 0.076 km 2 (optimistic scenario) and 0.069km 2 (critical scenario). In light of the predictions forthe other cities of the Metropolitan Center of Recife,this coastal strip does not present itself as an areahighly prone to flooding. However, the coast ofOlinda presents a severe and historical erosiveprocess, with a noteworthy part of its extension(around 59%) impermeabilized by coast defenseprojects, which hinder the limited capacity of itsbeaches to act as protection zones for resourceslocated inland.The city of Paulista presents the mostextensive coast among the cities in this study. It islimited to the North and to the South by the Timbóand Paratibe River estuaries, respectively. With aPan-American Journal of Aquatic Sciences (2010) 5(2): 341-349

Vulnerability and impacts related to the rising sea level345total area of 94 km 2 , approximately 26.53 km 2 wereanalyzed (coastal and estuarine areas). The potentialflood zones in the city are shown in figure 5.Vulnerability AnalysisGenerally speaking, the beaches of theMetropolitan Center of Recife would be severelyaffected in case of a sea level rise. An analysis of theentire coast indicates that 81.8% of urbanconstructions are less than 30 m from the shoreline.Comparatively, only small portions of the coastpresent a beach system little affected by adjacentdevelopment. Jaboatão dos Guararapes stands out inthis sense, presenting 36% of its coastline with lowvariability. Olinda presents the most criticalscenario, where 59% of the coast no longer hasrecreational beaches and the integrity of the urbandevelopment depends exclusively on coastalprotection projects (Fig. 6).damage to historical and cultural patrimony such asan important culture and leisure center (SESC) andthe Nossa Senhora da Piedade Church, both locatedon the Piedade Beach. This patrimony is vulnerablein both scenarios, optimistic and critical.In Recife, the large area with buildings withless than three levels is highly vulnerable to a seaFigure 5. Potential flood zones in Paulista.Figure 4. Potential flood zones in Olinda.Jaboatão Jaboatãodo Guararapes8%8%29%RecifeAnalysis of Resources at RiskTypes of land use and the area (in m 2 ) of theaffected resources in the cities studied arerepresented in figure 7.For the city of Jaboatão dos Guararapes, inan optimistic scenario, the relatively pristine areas –unoccupied and beach areas – would suffer thegreatest impact. On the other hand, in a criticalscenario, the areas with buildings of cultural andhistorical interest would amount to 57% of theaffected resources, exceeding the sum of vulnerablebeach and unoccupied areas.In the case of an overwash, there could be56%56%29%12%Olinda36%36%59%70%18%70%Paulista17%53%13%Conditional Low vulnerability Vulnerability High vulnerabilityVulnerabitlityFigure 6. Areas with high, low and conditionalvulnerability percentages.13%Pan-American Journal of Aquatic Sciences (2010), 5(2): 341-349

346M. B. S. F. COSTA ET ALLIlevel increase. From a historical-cultural perspective,we highlight the potential flooding of part of a greenarea of 33,000 m 2 set apart for leisure and a culturalcenter in the Boa Viagem neighborhood (DonaLindu Park). In the optimistic and critical scenarios,the impacted areas would be of approximately 3,100m 2 and 13,200 m 2 in area, respectively.In terms of coastal resources endangered byflooding in Olinda, in an optimistic scenario of sealevel increase, it was mostly the smallerconstructions that were found to be at risk. In acritical scenario, though, taller buildings (>3 stories)are included in the vulnerable area.Among the vulnerable areas of Paulista,Maria Farinha Headland demands attention since itis an area of great ecological and economicimportance. From a historical and culturalperspective, Pau Amarelo Fort and a water parkrelevant for the tourism industry near the MariaFarinha Beach represent potentially impactedpatrimonies.Area (10 5 m 2 )14121086420Paulista (O)Pau lista (C)Olin da (O)Olin da (C)Recife (O)Recife (C)Jaboatão (O)Jaboatão (C)Unoccupied land Beach Wetland Buildings (< 3 levels) Buildings (> 3 levels)Figure 7. Resources at risk in each city for bothscenarios: optimistic and critical.DiscussionThe present analysis was restricted to coastaland estuarine areas, which are the most affected byan increase in the sea level. For this reason it shouldbe made clear that the numbers and debate herepresented are underestimated and that the potentialflood zones in the studied area are even greater,bearing in mind the low altitude of the terrain andvast drainage network that bathes the region andbranches out to other estuarine areas.In terms of environmental impact, weemphasize that important ecological losses mayoccur not only along the estuary – a rich andimportant ecosystem – but also to the south portion,beyond the area studied, to the Paiva Beach. Thissite, as previously mentioned, is a restinga rich inflowers and with typical morphological aspects, withecological potential that justifies better protection ofthis ecosystem (Sacramento et al. 2007).On the coast of Jaboatão dos Guararapes, theloss of beaches and damage to constructions by theshoreline, due to its exposure to the energy of theocean, are common. For this reason, there arecurrently ten coastal defense construction sites alongthe shore, among them seawalls, groins, revetmentsand breakwaters (FINEP/UFPE, 2008). Asmentioned previously, the shoreline of the citypresents a high degree of verticalization, whichincreases the magnitude of the impact caused bycoastal flooding.For Recife, among the areas prone toflooding in both scenarios, attention is drawn toRecife Antigo (Old Recife). It represents animportant economic and administrative center of thecity with great historical and cultural value since it isthe origin of the city of Recife. It is currently ofgreat significance for cultural and artistic activities.Attention should also be given to the factthat the largest estuarine complex in the studied areais located in the city and includes the Pina Basin, theCapibaribe and Beberibe Rivers. The Port of Recifeis also located in the city. This Port, along with thePort of Suape, handles all cargo movements in theState of Pernambuco. The region has sufferedflooding since the beginning of its occupation, butthe intensification of the phenomenon makes anincrease in the silting rate a possibility throughoutthe port area. This would increase the demand fordredging and would potentially be detrimental toport operations.Although the potential flood area is narrowin the coastal strip, the intensification of the erosiveprocess is notable, considering that there arebuildings located very close to the beach, leavinglittle room for morphodynamic processes. As inJaboatão dos Guararapes, the Recife shorelinepresents a high degree of verticalization, to such anextent that the small area of land lost, or endangeredby erosion, would imply in large economic lossesand social disorganization. Additional disturbancecould be implied by the fact that countlessinhabitants use water extracted from the water tableusing wells, which could become unfeasible becauseof saline intrusion.In the sandy coastal regions, it is now wellestablishedthat sea-level rise leads, in average, toerosion and consequent recession of the shoreline(Snoussi & Niazi 2007). Sea-level rise is the maincause of shoreline retreat in coastal areas underdynamic equilibrium, i.e., where the natural sandsupply allows the potential and the effective littoraldrift to be equal. However, on sandy shores, wherethis supply has been strongly reduced (as alreadydiscussed), the main cause of shoreline retreat isPan-American Journal of Aquatic Sciences (2010) 5(2): 341-349

Vulnerability and impacts related to the rising sea level347sediment deficiency, Ferreira et al. (2008). In thiscase, which is similar to the study area, thisdeficiency plays a larger role in erosion than theaccelerated sea-level rise and coastal morphologyand size will be mainly dependent on the directresult of human actions.In Olinda, the flood problems are ahistorical inconvenience for the population. Floodsare annual events in some of the mainneighborhoods of the city, and are among the mostdistinguished environmental impacts, causing lossand disarray to the population. Some areas in thislocation are flooded from time to time, whichdamages road structures, makes the circulation ofvehicles and people impractical, jeopardizescommerce and causes public health issues (Melo2003). Anthropic causes are added to the geographicfeatures and contribute to the intensification of theproblem, such as the illegitimate occupation offreshwater swamp forests, illegal landfills inmangroves, criminal disposal of solid waste and thenonexistence of an efficient draining network.The city, due to its historical erosion andflood problems, has a large range of coastal defensestructures in place, which may provide efficientprotection from the sea level rise (Neves & Muehe1995). Presently, these structures, in addition tocountless sections with hydraulic landfills, put thecoastal streets in flood zones above sea level,reducing the area vulnerable to flood effects. InPaulista we can highlight the potential flooding of anarea relevant from an ecological and economicalpoint of view, the Maria Farinha Headland. On thissite there is a complex estuarine system of elevatedscenic attributes that have contributed to recentdevelopment of the tourism industry with theinstallation of marinas, hotels, inns and water park.Generally, and in what is referred to asvulnerability assessment, the studies developed byFINEP/UFPE (2008), considering only the currenterosion rate and the rate at which constructionpresses closer to the beach, indicate that the width ofthe beach tends to decrease with the passage of time.This analysis demonstrates that in most of thecoastline analyzed, a significant part of thevulnerability may be related to the acceleratedoccupation of areas immediately inland. With thepredicted sea level rise, a higher demand forshoreline protection structures is expectedthroughout the metropolitan coast of Recife.Regarding the methodology adopted, the useof GIS is justified by the fact that the spatialcomponent of climate risk is critical for buildingknowledge on climate risk, potential managementoptions and challenges in local level. A range ofmethods is available for exploring climate riskacross a landscape, however, as with any scientificassessment process, the appropriate methodologydepends on the needs of stakeholders as well aspotential constraints placed upon a project such asfunding, time, data access and expertise (Preston etal. 2009).Based on worldwide experience in policycreation, on data and recommendations of the MAI-PE Project (FINEP/UFPE 2008), and on the resultsgiven in this paper, the implementation of publicpolicies for the protection of coastal and flood zonesare suggested in two fronts: (I) expansion andconsolidation of scientific knowledge of thephenomenon, since the effects of the change in therelative sea level will differ according to localcharacteristics; and (II) management and establishmentof adaptive measures to minimize its impact.These fronts complement the suggestion of Jallow etal. (1996), which present the following means todeal with the problem of coastal vulnerability to thesea level rise: urban growth planning, publicawareness, wetland preservation and mitigation, andcoastal zone management.ConclusionThe Metropolitan Center of Recife, due toits physical characteristics and its current erosionand flood problems, presents itself as a region highlyvulnerable to an increase in sea level. Additionally,it has unfavorable social charac-teristics forresponses to flooding, including high demographicdensity and intensified vertical growth on the coast,as well as occupation of riverside are-as. The impactassociated with the relative sea level rise mayintensify if relief measures are not taken.These results comprise first an approach tothe impacts caused by the combined projectedchanges in coastal areas in the Metropolitan Centerof Recife. Since the response prediction and scenarioanticipation for these areas are highly complex tasks,it is essential to obtain more funding from researchfoundations, institutions responsible fordevelopment of human resources and from those incharge of public policies in order to perform furtherresearch on the matter. The bigger the knowledgebase and the better the prediction of the impactsresulting from climatic change, the better the plansfor economic, social and environmental riskprevention will be.AcknowledgmentsThe authors wish to thank the agencyFinanciadora de Estudos e Projetos (FINEP/MCT)and the Recife, Jaboatão dos Guararapes, Olinda andPan-American Journal of Aquatic Sciences (2010), 5(2): 341-349

348M. B. S. F. COSTA ET ALLIPaulista City Halls for funding the IntegratedEnvironmental Monitoring Project – MAI-PE(Projeto Monitoramento Ambiental Integrado),ReferencesAlfredini, P., Arasaki, E., Do Amaral, R. F. 2008.Mean sea-level rise impacts on Santos Bay,Southeastern Brazil – physical modellingstudy. Environmental Monitoring andAssessment, 144: 1-3.Araujo, M., Mallmann, D. L. B., Leite, F. S.,Rollnic, M., Mesquita, P. P., Borba, M. &Façanha, P. 2009. Vulnerabilidade eimpactos à elevação do nível do mar doCentro Metropolitano do Recife, PE.Technical report. 168 p.Clark, G. E., Moser, S. C., Ratick, S. J., Dow, K.,Meyer, W. B., Emani, S., Jin, W., Kasperson,J. X., Kasperson, R. E. & Schwarz, H. E.1998. Assessing the vulnerability of coastalcommunities to extreme storms: the case ofRevere, MA, USA. Mitigation andAdaptation Strategies for Global Change,3: 59–82.Costa, M. B. S. F., Pontes, P. M. & Araujo, T. C. M.2008. Monitoramento da Linha de Preamardas Praias de Olinda - PE (Brasil) comoFerramenta à Gestão Costeira. Revista daGestão Costeira Integrada, 8(2): 101-112.Ferreira, O., Dias, J. A. & Taborda, R. 2008.Implications of Sea-Level Rise forContinental Portugal. Journal of CoastalResearch, 24(2): 317-324.FINEP/UFPE, 2008. Monitoramento AmbientalIntegrado – MAI-PE. Tecnichal Report.Financiadora de Estudos e Projetos – FINEP,Recife, 383 pp.Harari, J., França, C. A. S. & Camargo, R. 2008.Variabilidade de longo termo decomponentes de maré e do nível médio domar na costa brasileira. Accessible at:http://migre.me/3aOax.Hoozemans, F. M. J., Stive, M. J. F. & Bijlsma, L.1993. A global vulnerability assessment:vulnerability of coastal áreas to sea-level rise.8th Symposium on Coastal and OceanManagement - Coastal Zone ’93, 8, NewOrleans, 390-404.IPCC – Intergovernmental Panel on ClimateChange. Climate Change. 2007 - ThePhysical Science Basis. Accessible at:http://www.ipcc.un.org.Jallow, B. P., Barrow, M. K. A. & Leatherman, S. P.1996. Vulnerability of the coastal zone of Thewhich ena-bled the acquisition and the detailedanalysis of recent information on the coastaldynamics of these cities.Gambia to sea level rise and development ofresponse strategies and adaptation options.Climate Research, 6:165-177.Lathrop Jr., R. G. & Love, A. 2007. Vulnerability ofNew Jersey’s Coastal Habitats to Sea LevelRise. Grant F. Walton Center for RemoteSensing & Spatial Analysis. RutgersUniversity. Accessible at:http://migre.me/3aTcy.Melo, M. J. V. 2003. A bacia do Rio Fragoso emOlinda, PE: Drenagem e Gestão Ambiental.Master Thesis. Universidade Federal dePernambuco, Recife, 183 p.Naccarati, M. A. 2008. Os cenários de mudançasclimáticas como novo condicionante para agestão urbana: as perspectivas para apopulação da cidade do Rio de Janeiro. XVIEncontro Nacional de EstudosPopulacionais, Caxambu, MG, Brasil.Neves, C. F. & Muehe, D. 1995. Potential impacts ofsea-level rise on the Metropolitan Region ofRecife, Brazil. Journal of Coastal Research,SI 14:116-131.Neves, C. F, Muehe, D. E. & Valentini, E. M.,Rosman P. C. C. 2007. Estudo devulnerabilidades no litoral do Rio deJaneiro devido às mudanças climáticas.Relatório final, 110 p.Pajak, M. J. & Leatherman, S. 2002. The high waterline as shoreline indicator. Journal ofCoastal Research, 18(2): 329-337.Pernambuco. CONDEPE/FIDEM - AgênciaEstadual de Planejamento e Pesquisa dePernambuco. 1974. Rectified aerialphotography numbers: 99/08 to 80/95.[Recife], versão digital. Escala: 1: 2.000.Preston, B.L., Abbs, D., Beveridge, B., Brooke, C.,Gorddard, R., Hunt, G., Justus, M., Kinrade,P., Macadam, I., Measham, T. G., McInnes,K., Morrison, C., O’Grady, J., Smith, T. F.,Withycombe, G. Spatial Approaches forAssessing Vulnerability and Consequencesin Climate Change Assessments. Accessibleat: http://migre.me/3aPfI.Rohling, E. J., Grant, K., Hemleben, C., Siddall, M.,Hoogakker, B. A. A., Bolshaw, M. & Kucera,M. 2008. High rates of sea-level rise duringthe last interglacial period. NatureGeoscience, 1: 38-42.Pan-American Journal of Aquatic Sciences (2010) 5(2): 341-349

Vulnerability and impacts related to the rising sea level349Sacramento, A. C., Zickelli, C. S. & Almeida Jr., E.B. 2007. Aspectos florísticos da vegetação derestinga no litoral de Pernambuco-Brasil.Revista Arvore, 31(6): 1121-1130.Snoussi, M., Ouchani, T. & Niazi, S. 2008.Vulnerability assessment of the impact of sealevelrise and flooding on the Moroccan coast:The case of the Mediterranean eastern zone.Estuarine, Coastal and Shelf Science, 77:206-213.Thieler, E. R., Hammar-Klose, E. S. 1999. Nationalassessment of coastal vulnerability to sealevelrise, Preliminary Results for the U. S.Atlantic Coast. U. S. Geological SurveyOpen-file Report. Massachusetts, 99-593.Accessible at: http://migre.me/3aT87Vargas, C. I. C., Oliveira, F. S. B. F., Oliveira, A. &Charneca, N. 2008. Análise davulnerabilidade de uma praia estuarina àinundação: aplicação à restinga do Alfeite(Estuário do Tejo). Revista da GestãoCosteira Integrada, 8(1): 25-43.Received December 2009Accepted June 2010Published online January 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 341-349

Temporal changes in the seaweed flora in SouthernBrazil and its potential causesCAROLINE DE FAVERI, FERNANDO SCHERNER, JULYANA FARIAS, EURICO C. DEOLIVEIRA & PAULO A. HORTA *Universidade Federal de Santa Catarina, Centro de Ciências Biológicas, Departamento de Botânica, CampusUniversitário, Trindade, Florianópolis, Santa Catarina, CEP: 88040-900, Brasil.*corresponding author. E-mail: pahorta@ccb.ufsc.brAbstract. The anthropogenic activities in recent centuries have led to atmospheric changes that directlyinfluence the climate, resulting in global warming. Coastal ecosystems are subjected to global threats bytheir sensitivity to chemical and physical characteristics of seawater and seaweed communities areconsidered good indicators of environmental changes. This study aimed to evaluate changes in theseaweed flora at Ribanceira Beach (Santa Catarina state, Southern Brazil) by comparing recent to pastdata (30 years apart), motivated by the possible effects of climate change in this subtropical region,dominated by warm-temperate coastal waters. Significant differences between the past and the currentflora were observed. The absence of 17 taxa, observed in the past, and the presence of 16 taxa notreported before in the area are discussed under the perspective of possible global warming effects.Key words: coastal ecosystems, climate change, macroalgal flora, Southern BrazilResumo: Mudanças temporais na flora de macroalgas no Sul do Brasil e suas causas potenciais. Aação antrópica nos últimos séculos vem provocando alterações atmosféricas que influenciam diretamenteo clima, resultando no aquecimento global. Os ecossistemas costeiros estão sujeitos a ameaças globaispela sua sensibilidade a alterações químicas e físicas da água marinha, e as macroalgas são consideradasboas indicadoras de mudanças ambientais. Este trabalho teve como objetivo detectar mudanças na floramacroalgal da Praia da Ribanceira (Santa Catarina, Sul do Brasil) através da comparação de resultados deum inventário recente com um estudo passado, motivados pelos possíveis efeitos das mudançasclimáticas nesta região sub-tropical, dominada por águas costeiras temperada quente. Diferençassignificativas entre a flora atual e de três décadas atrás foram observadas. A ausência de 17 táxons,observados no passado, e a presença de 16 táxons não reportados para área no levantamento anterior sãodiscutidas sob a perspectiva de possíveis efeitos do aquecimento global.Palavras-chave: ecossistemas costeiros, mudanças climáticas, flora macroalgal, Brasil, Região SulIntroductionThe planet has been affected by acceleratedprocesses of global changes in such a way thatprobably no area, worldwide, remains completelyunaffected by human influence (Halpern et al.2008). Coastal ecosystems are one of the mostvulnerable natural environments. Although not wellunderstood, but traditionally used for providinggoods and services, coastal ecosystems have beenbroadly threatened by anthropogenic impacts andwill very likely be severely affected by climatechanges (Vitousek et al. 1997, Orfanidis et al.2001).Some evidences are showing that globalwarming are progressing at a faster rate thanpreviously recorded by IPCC third assessment report(IPCC, 2007), be responsible by recent variation inthe species composition and distribution in marineenvironments (e.g. Stachowicz et al. 2002, Dijkstraet al. 2010). On the other hand, despite of thetemperature increase be one of the most debatedglobal climate change effect, directly or indirectlyrelated factors must be considered. Additionally, thesynergistic action of other physical, chemical andbiological factors should be evaluated (Russel et al.2009). Anthropogenic drivers associated to globalclimate change are distributed widely and arePan-American Journal of Aquatic Sciences (2010) 5(2): 350-357

Temporal changes in the seaweed flora351another important component of global synergisticimpact (Halpern et al. 2008). Macroalgae, beingstationary organisms, can be useful bioindicators todetect environmental changes of various kinds.Therefore, monitoring macroalgae distribution inspace and time may help to anticipate effects ofglobal changes on the biota and guide policiestowards environmental conservation and planning ofmitigation initiatives.Changes in distribution patterns of macroalgaeattributed to pollution have been documentedonly in two restricted areas along the Brazilian coast(Oliveira & Berchez 1978, Oliveira & Qi 2003,Taouil & Yoneshigue-Valentin 2002). However, animportant constraint to detect changes in distributionpatterns along time is related to the non-existence ofprevious reliable floristic surveys of seaweed floras.The publications mentioned above dealt withtemporal changes on polluted tropical bays on theSouth East Brazilian coast and cannot be attributedto climate changes alone. In 1978, the seaweed floraof Imbituba, (Santa Catarina State, South Brazil)was surveyed motivated by the establishment of acarbon-chemistry industrial complex, in order todocument the pre-impact situation (Citadini-Zanetteet al. 1979). However, the company was declaredbankruptcy shortly after opening and the powerplant was not established. Despite the increasedurbanisation and population increment observedalong most of the Brazilian coast, the referred areawas kept with similar pattern of urban occupationobserved 30 years before (IBGE 2005). Here wereport the results of a recent survey in the same area,to look for eventual changes in the seaweed floraafter 30 years. This new survey meets furtherjustification since the macroalgal flora reported forSanta Catarina is considered warm-temperate andtransitional to the more tropical northern flora (Hortaet al. 2001) and, therefore, more prone to yield cluesto spot floristic differences due to environmentalchanges, eventually related with global warmingprocess and anthropogenic ecological footprint.Material and MethodsMacroalgal specimens were collected atRibanceira Beach, Imbituba (Santa Catarina, Brazil).Samplings were made on two rocky shores (28°14’S/ 48°40’ W; 28°11’ S /48°39’ W), which wereselected in order to cover the same area surveyed byCitadini-Zanette et al. (1979). Our floristic list wasbased on collections made in August (winter) andJune (late autumn) of 2007, and March (latesummer) and September (early spring) of 2008. Oneach site, one 10 cm broad transect was placedperpendicularly to the water line and algae wasscraped from the rocky substrate from the supra tothe sublittoral fringe.Specimens were collected during periods oflow water spring tides, sorted and preserved informaline:seawater 4%. After identification,vouchers were deposited at the Herbário RaulinoReitz (CRI, UNESC). Abiotic parameters weremeasured at surface water at each sampling period.Salinity was measured with a portable refractometer(RTS-101 ATC, Meditec, Brazil), pH with aportable pH-meter (pH 1800, Instrutherm, Brazil)and water temperature with a digital thermometer(HT-210, Instrutherm, Brazil).The floristic similarity between this studyand the previous one (Citadini-Zanette et al. 1979)was evaluated through similarity index of Sorensen(S = [2C/(A + B)]×100, in which C is the number ofcommon species in both surveys , A is the totalnumber of species and B is the total number ofspecies in the work B, (Cullen et al. 2003).Comparison between both surveys consider thatCitadini-Zanette et al. (1979) carried their surveyout in the spring with similar methodology andgeneral area. The Feldmann (1937) and Cheney(1977) indexes were utilized as an attempt tocharacterize the biogeographic affinities of the florason the two sampling moments.ResultsSalinity varied from 35 to 39 ppt, pH from7.20 to 7.37 and water temperature from 18.7 to29.5 °C (Tab. I). The non expected low pH valuesmay be due to the dynamics of oceanic CO 2 uptakeon surface water determined by the rate ofdownward transport of CO 2 from the surface tobottom (Siegenthaler & Sarmiento1993) or to therunoff of residual waters from neighboringindustries or urbanized areas. This hypothesis arereinforced by the works of Feely et al. (2010) that,working in estuary complex in the U.S. PacificNorthwest, estimate that part of the acidificationobserved results from remineralization of organicmatter due to natural or anthropogenicallystimulated respiration processes. Therefore, evenconsidering that referred data are punctual and thatpH is a very sensitive parameter, the observed valuescan be resulted from processes related tourbanization and pollution.A total of 62 infrageneric taxa (Tab. II) was found,being nine Phaeophyceae (14.5%), 14 Chlorophyta(22.5%) and 39 Rhodophyta (63%). Among those,43 species were found in the spring of 2008, 32 inthe summer, 28 in the autumn and 50 in the winter(Tab. II, Fig. 1). Besides that, Arthrocardiaflabellata, A. gardneri, Crytopleura ramosa,Pan-American Journal of Aquatic Sciences (2010), 5(2): 350-357

352CAROLINE DE FAVERI ET ALLITable I. Seasonal variation of the chemical and physicalwater parameters.Summer Autumn Winter SpringWater temp (°C) 25.8 29.5 18.7 24.7Salinity (ppt) 39.0 35.0 36.0 35.0pH 7.20 7.30 7.37 7.33Figure 1. Seasonal distribution in number of species ofChlorophyta, Phaeophyceae and Rhodophyta, found atRibanceira Beach, Imbituba, SC.Gelidium floridanum, Codium decorticatum, Ulvafasciata, U. lactuca, Cladophora prolifera,Chondracanthus teedi, and C. elegans were the mostfrequent species in all seasons.Among the Chlorophyta, theCladophoraceae (six species) and the Ulvaceae (fivespecies) were the more diversified, while among theRhodophyta the families with higher diversity wereCorallinaceae, with six species, followed byGelidiaceae, Ceramiaceae and Rhodomelaceae withfour species each. Within the Phaeophyceae,Scytosiphonaceae (three species), Acinetosporaceae(two species) and the other families with only onespecies each. Most of the species belonged to thefilamentous morphological-functional group whichare more likely to colonize physically disturbedenvironments, such as intertidal zones of rockyshores exposed to waves (Littler & Littler 1980).The predominance in number of Rhodophytaspecies over Phaeophyceae and Chlorophyta is acommon pattern for Santa Catarina region, as well asfor other areas along Brazilian coast (Horta et al.2001). However, looking at the physiognomy of thesampled area, the scenario was greenish sinceChlorophyta, represented mainly by Ulva spp. andCladophora spp., dominated over the red and brownseaweeds.Species richness was higher in the coldestperiod, when water temperature was 18.7 °C, whichis in agreement with Yoneshigue-Valentin &Valentin (1992) for an upwelling region in Rio deJaneiro state. However, one should consider that thehighest species richness in winter may be due not totemperature, per se, but to other factors such asnutrient enrichment, what remains to be studied. Thecoastal upwelling around Cape of Santa MartaGrande occurs mainly in spring and summer, whennortheast winds prevail, which facilitate thepenetration of South Atlantic Central Water (ACAS)onto the local continental shelf (Pereira et al.2008).The similarity followed by ANOSIM analysesshows a significant difference (ANOSIN p< 0,05)between our sample and the flora presented byCitadini-Zanette et al. (1979), indicating a change inspecies composition between the two surveys.Results from our analysis showed that the recentspring flora was more similar to summer, autumnand winter flora, surveyed in 2008 (similaritiesbetween 73,9 and 69,5), than to the old spring data(similarities between 64, and 51,2; Tab. III).Yoneshigue-Valentin & Valentin (1992)documented a change in species richness along theyear in areas subjected to upwelling north of Rio deJaneiro, which was attributed to temperature.However, in other instances, the distinction betweenthe effects of temperature and other parameters, suchas pollution, as causing factors of floristic changes isnot clear. Stressful conditions due to the seasonalvariation of different parameters seem to reducespecies richness and favor the dominance ofopportunist algae.Comparing the total number of speciesfound by Citadini-Zanette et al. (1979) in the springand the present survey at the same season, a total of27 taxa previously listed were not found in the studyarea. From these, 16 taxa were Rhodophyta, fourPhaeophyceae and seven Chlorophyta. On the otherhand, we observed the appearance of 15 taxa thatwere not present before: 11 Rhodophyta, onePhaeophyceae and three Chlorophyta. Feldmann andCheney indices for 1978 show values typical ofwarm temperate environment (3.9 and 5.1,respectively). Higher values (5.0 and 6.14,respectively) were observed in spring of 2008,characterizing a tropical environment. The sameresults were found when considering general data of2008 (4.5 and 6.125 respectively). Horta et al.(2001) evaluating the distribution pattern of the floraalong the Brazilian coast, characterized the southerncoast as belonging to a warm temperate province.Considering that temperature is traditionallyconsidered the main controlling factor of seaweeddistribution, the “tropicalization” of the valuesobserved for Ribanceira beach, compared to the past,could be an indication of global warming. Byanalysing surface air and sea surface temperaturetrends in Southern Brazil, Marengo & Camargo(2007) highlighted that the frequency of warmerPan-American Journal of Aquatic Sciences (2010) 5(2): 350-357

Temporal changes in the seaweed flora353Table II. Macroalgal species recorded at the Ribanceira Beach, Santa Catarina, during the periods 2007/2008 andSpring 1978 (Citadini-Zanette et al. 1979). Numbers I, II, III, IV refers to the sampling sites surveyed by Citadini-Zanette et al. (1979). 1 = presence and 0 = absence.Species2007/2008Present studySpring 1978(Citadini-Zanette et al. 1979)Sp Su Au Wi Site I Site II Site III Site IVBryopsis pennata J. V. Lamour. 1 0 0 1 1 1 1 1Bryopsis plumosa (Huds.) C. Agardh 0 0 0 1 0 0 1 0Chaetomorpha antennina (Bory) Kütz. 0 0 0 1 1 1 1 1Cladophora prolifera (Roth) Kütz. 1 1 1 1 1 0 1 0Cladophora montagneana Kütz. 0 0 0 0 1 1 1 1Cladophora sp.1 0 0 0 1 0 0 0 0Cladophora sp.2 1 1 0 1 0 0 0 0Cladophora vagabunda (L.) C. Hoek 0 1 0 1 1 1 1 1Cladophoropsis membranacea (C. Agardh) Borgesen 1 0 1 1 0 0 0 0Codium decorticatum (Woodw.) M. Howe 1 1 1 1 1 1 1 1Codium intertextum Collins & Herv. 0 0 0 0 0 0 0 1Codium taylorii P.C. Silva 1 0 0 1 0 0 0 0Rhizoclonium riparium (Roth) Kütz. ex Harv. 0 0 0 0 1 1 0 0Ulva compressa L. 1 0 0 0 0 0 0 0Ulva fasciata Delile 1 1 1 1 1 1 1 1Ulva lactuta L. 1 1 1 1 1 1 1 1Ulva linza L. 0 0 0 1 1 1 1 1Bachelotia antillarum (Grunow) Gerloff 0 0 0 0 1 0 0 0Chnoospora minima (K. Hering) Papenf. 0 0 0 0 0 1 0 0Colpomenia sinuosa (Roth) Derbès & Solier 1 0 0 1 1 1 1 1Feldmannia irregularis (Kütz.) Hamel 0 1 0 1 1 0 0 0Hincksia mitchelliae (Harv.) P.C. Silva 1 0 1 0 1 0 0 0Levringia brasiliensis (Mont.) A.B. Joly 1 1 1 1 1 1 1 0Padina gymnospora (Kütz.) Sond. 1 1 0 1 1 0 0 0Petalonia fascia (O.F. Müll.) Kuntze 1 0 0 1 1 0 1 1Rosenvingea sanctae-crucis Borgesen 1 0 0 1 0 0 0 0Sargassum cymosum C. Agardh 1 1 1 1 1 1 1 1Scytosiphon lomentaria (Lyngb.) Link nom.cons. 0 0 0 0 0 0 0 1Rhodothamniella codicola Borgesen 0 0 0 1 0 1 0 0Acrochaetium globosum Borgensen 0 0 0 0 1 0 0 0Acrochaetium microscopium (Nägeli ex Kütz.) Nägeli 0 0 0 0 1 0 0 0Aglaothamnion felliponei (M. Howe) N. Aponte, D. 0 0 1 0 1 0 0 0L.Ballant. & J. N. NorrisAglaothamnion uruguayense (W.R. Taylor) N. Aponte, 1 1 0 0 1 1 1 1D. L.Ballant. & J. N. NorrisArthrocardia flabellata (Kütz.) Manza 1 1 1 1 1 0 0 0Arthrocardia gardneri Manza 1 1 1 1 1 1 1 1Bangia fuscopurpurea (Dillw.) Lyngb. 0 0 0 0 1 1 1 0Bostrychia tenella (J.V.Lamour.) J. Agardh 0 0 0 0 0 1 0 0Bryocladia thyrsigera (J. Agardh) F. Schmitz inFalkenb0 0 0 1 0 0 0 0Pan-American Journal of Aquatic Sciences (2010), 5(2): 350-357

354CAROLINE DE FAVERI ET ALLICallithamnion corymbosum (Sm.) Lyngb. 1 1 0 1 0 0 0 0Centroceras clavulatum (C. Agardh in Kunth) Mont. in 1 1 0 1 1 1 1 1Durieu de MaisonneuveCeramium brevizonatum var. caraibicum H.E. Petersen 0 0 0 0 0 1 0 1Ceramium dawsonii A.B. Joly 0 0 0 0 0 1 0 0Ceramuim tenerrimum (G. Martens) Okamura 1 0 1 0 0 0 0 0Champia parvula (C. Agardh) Harv. 1 0 1 1 0 0 0 0Cheilosporum sagittatum (J. Ellis & Sol.) Aresch. 0 0 0 1 0 1 1 1Chondracanthus acicularis (Roth) Fredericq 1 0 0 1 1 1 1 1Chondracanthus elegans (Grev. in J. St.-Hil.) Guiry 1 1 1 1 0 1 1 1Chondracanthus teedei (Mertens ex Roth) Kütz. 1 1 1 1 1 1 0 0Corallina officinalis L. 1 0 0 1 0 0 0 1Crytopleura ramosa (Hudson) Kylin ex L. Newton 1 1 1 1 1 1 1 1Erythrotrichia carnea (Dillwyn) J. Agardh 0 1 0 1 0 0 0 0Gelidium crinale (Turner) Gaillon 0 0 1 1 0 1 1 1Gelidium floridanum W.R. Taylor 1 1 1 1 1 1 1 1Gelidium pusilum (Stackh.) Le Jolis 1 0 0 0 0 1 1 1Gracilaria cf. tepocensis (E.Y. Dawson) E.Y. Dawson 0 0 0 1 0 1 0 0Grateloupia cuneifolia J. Agardh 1 1 1 1 1 0 0 0Grateloupia filiformis Kützing 0 1 0 1 0 0 1 1Gymnogongrus griffithsiae (Turner) Mart. 0 1 0 1 1 1 1 1Herposiphonia secunda (C. Agardh) Ambronn 0 0 0 0 0 1 0 0Hypnea musciformis (Wulfen in Jacquin) J.V. Lamour. 1 1 1 1 1 1 1 1Hypnea spinella (C. Agardh) Kützing 1 1 0 1 0 0 0 0Jania adhaerens J.V. Lamour. 1 0 0 0 1 0 0 0Jania cappillacea Harv. 0 0 0 0 0 0 1 1Jania crassa J.V. Lamour. 1 1 1 1 1 1 1 1Laurencia sp. 0 0 0 0 1 1 0 0Nemalion helminthoides (Velley in With.) Batters 1 0 1 0 1 1 1 0Neosiphonia ferulacea (Suhr ex J. Agardh) S.M. Guim. 0 0 1 0 0 0 0 0& M.T. FujiiNeosiphonia tepida (Hollenb.) S.M. Guim. & M.T. 1 0 1 1 0 1 0 0FujiiPeyssonnelia capensis Mont. 0 0 0 0 1 1 0 0Plocamium brasiliense (Grev. in J. St. -Hil.) M. Howe 1 1 1 1 1 0 1 0& W.R. TaylorPolysiphonia decussata Hollenb. 0 0 0 0 0 0 0 1Polysiphonia scopulorum Harv. 0 0 0 0 1 1 1 0Polysiphonia subtilissima Mont. 0 0 0 1 0 0 0 0Porphyra acanthophora E. C. Oliveira & Coll var. 1 0 1 0 0 0 0 0acanthoporaPorphyra pujalsiae Coll & E.C. Oliveira 0 0 0 1 0 0 0 0Pterocladiella capillacea (S.G. Gmel.) Santel. & 1 1 0 1 1 0 1 1HommersPterosiphonia parasitica var. australis A.B. Joly & 1 1 1 1 1 1 1 1Cord.-MarPterosiphonia pennata (C. Agardh) Falkenb. 1 1 0 1 0 0 0 0Sphacelaria tribuloides Menegh 1 1 1 0 0 0 0 0Pan-American Journal of Aquatic Sciences (2010) 5(2): 350-357

Temporal changes in the seaweed flora355Table III. Sorensen similarity matrix results, with comparison among the recent sampling periods and the four sitessurveyed by Citadini-Zanette et al. (1979).Spring Summer Autumn Winter SiteI SiteII SiteIII SiteIVSpringSummer 71,23Autumn 69,56 62,07Winter 73,91 71,60 54,55SiteI 64,29 63,01 55,07 60,87SiteII 51,22 47,89 47,76 57,78 68,29SiteIII 59,74 60,61 51,61 65,88 72,73 74,67SiteIV 53,33 53,12 40 60,24 58,67 68,49 82,35days increased during both summer and winter,especially during the last two decades. Additionally,Wainer & Venegas (2002) described a possibledisplacement of the Brazilian/ Malvinasconvergence to the south, resulting in a possiblewater warming in the referred region during nextdecades. Our results may represent some precocioussigns of such reported warming. However, thevariation of other parameters, such as eutrophizationand pollution, or even their synergistic effect (Russelet al. 2009), besides species introduction, cannot beruled out. Of course, on this kind of investigation,differences in presence-absence of species alongtime may be due to taxonomic problems, or detailsin sampling methodology.However, in this case we were dealingmostly with conspicuous and easily identified taxawhat makes our comparison more acceptable. Therelative reduction in the number of Phaeophyceaespecies is an indication of impacts related topollution. This is also supported by an increase ofthe Chlorophyta in comparison with other groups asdocumented by several authors (e.g. Reis &Yoneshigue-Valentin 1996; Oliveira & Qi 2003,Taouil & Yoneshigue-Valentin 2002, Lehmkuhl-Bouzon 2005). Conversely, the absence ofScytosiphon lomentaria, a species that needtemperatures below 20 ºC to induce macrothallusformation (Lüning 1980, Orfanidis et al. 1996) mayalso indicate a warming process.The appearance of taxa not present before,such as Rosenvingea sanctae-crucis, Champiaparvula, Erythrotrichia carnea, Callithamnioncorymbosum, Sphacelaria tribuloides may be relatedto biogeographic issues (Oliveira et al. 2001). Thepresence of C. parvula, although not recorded byCitadini-Zanette et al. (1979), was reported earlier inthe area by Cordeiro-Marino in 1966 (Cordeiro-Marino 1978). Considering that R. sanctae-crucis isconsidered a tropical species, recorded to theBrazilian northeastern coast (Oliveira et al. 1983),the extension of its distribution to higher latitudesmay also be an indication of global warming whatremains to be tested. One may hypothesize that thereported changes are responses to climatic change,considering that temperature alterations also alter thepattern of geographic distribution of species.However, among the species not recorded byCitadini-Zanette et al. (1979), stands out species ofCladophora, Chaetomorpha and Ulva (Tab. II),genera that includes typically opportunistic species,evidencing that we cannot discard the interaction offactors related to the eventual increases of theanthropogenic ecological footprint. The humanactivities can change the seawater quality due theeffluent discharge favoring species opportunists(Orfanidis et al. 2001) or even be responsible by thearrival of newcomer species through their transportvia ships hull fouling (Mineur et al. 2007).A decrease of species richness and anelevation of the indices of Feldmann and Cheneywas observed after the discharge of thermal effluentfrom a nuclear plant in the Bay of Ilha Grande, Riode Janeiro state, resulting from the effects oftemperature increase on the seaweed flora (Széchy& Nassar 2005). Schield et al. (2004) observed thata 3.5 o C rise in seawater temperature, induced alsoby the thermal outfall of a power-generating station,resulted in significant community changes in 150species of algae and invertebrates relative toadjacent control areas. However, they did notevidence clear tendencies toward warmer-waterspecies with southern geographic affinities replacingcolder water species with northern affinities. Theseauthors reinforce that responses of these benthiccommunities to ocean warming were stronglycoupled to direct effects of temperature on some keytaxa, as habitat-forming subtidal kelps, and indirecteffects operating through ecological interactionsbetween herbivores and primary producers.In spite of the uncertainties about the causalfactors that produced the differences in the algalcommunity composition, the observed changes arereal. The interaction of factors such as temperatureincrease, variation in salinity, nutrient availabilityand pollution, acting per se, and interacting in aPan-American Journal of Aquatic Sciences (2010), 5(2): 350-357

356CAROLINE DE FAVERI ET ALLIcomplex fashion, will certainly have a broad impacton seaweed floras and biodiversity, and should beevaluated with an experimental approach. Further, ifwe consider ocean acidification and the increase inthe intensity and frequency of storms, biodiversitylosses can be very high in the coming years. Thisscenario reinforces the need for constant monitoringand decision making with regard to coastalReferencesCitadini-Zanette, V., Veiga Neto, A. J. & Veiga, S.G.1979. Algas bentônicas de Imbituba, SantaCatarina, Brasil. Iheringia, Série botânica,25: 111-121.Cheney, D. F. 1977. R+C/P, a new improved ratiofor comparing seaweed floras. Journal ofPhycology 13(supl.): 12.Cordeiro-Marino, M. 1978.Rodofíceas MarinhasBentônicas de Santa Catarina. Instituto deBotânica São Paulo. Rickia, 7: 1-243.Cullen, L., Valladares-Padua, C. & Rudran, R. 2003.Métodos de estudos em biologia da conservaçãoe manejo da vida silvestre. Curitiba,PR: UFPR, Fundação O Boticário, 663 p.Dijkstra, J., Westerman, E. & Harris. L. 2010. Theeffects of climate change on speciescomposition, succession and phenology: acase study. Global Change Biologyhttp://www.esajournals.org/doi/abs/10.1890/03-3107Feldmann, J. 1937. Recherches sur la vegetationmarine de la Méditerranée. La cote dêsAlberes. Revue Algologique, 10: 1-339.Halpern, B. S., Walbridge, S., Kappel, C. V.,Micheli, F., D’Agrosa, C., Bruno, J. F., Casey,K. S., Ebert, C., Fox, H. E., Fujita, R.,Heinemann, D., Lenihan, H. S., Madin, E. M.P., Perry, M. T., Selig, E. R., Spalding, M.,Steneck, R. & Watson, R. 2008. A global mapof human impact on Marine ecosystems.Science, 319: 948-952.Horta, P. A., Amancio, C. E., Coimbra, C. S. &Oliveira, E. C. 2001. Considerações sobre adistribuição e origem da flora de macroalgasmarinhas brasileiras. Hoehnea, 28: 243-265.IBAMA 2005. Censo demográfico. Rio de Janeirohttp://www.ibge.gov.br.IPCC 2007. The physical science basis. Contributionof Working Group I to the Fourth AssessmentReport of the Intergovernmental Panel onClimate Change, Vol. 1. Intergovernmentalpanel on climate change, Geneva,Switzerland.Lehmkuhl-Bouzon, J. 2005. Composição e estruturamanagement, to mitigate environmental impactsderived from human activities.AknowledgementsFinancial support was given by ‘Coordenação deAperfeiçoamento de Pessoal de Nível Superior’(CAPES) and ‘Conselho Nacional deDesenvolvimento Científico e Tecnológico’ (CNPq).espacial da comunidade Macrofitobêntica defundos consolidados das baías da ilha deSanta Catarina (SC): subsídios para aavaliação do impacto da urbanização. Dissertaçãode Mestrado. Universidade Federalde Santa Catarina, Florianópolis, 77 p.Littler, M. M. & Littler, D. S. 1980. The evolutionof thallus form and survival strategies inbenthic macroalgae: field and laboratory testsof a functional-form model. AmericanNatura-list, 116: 25-44.Lüning, K. 1980. Critical levels of light andtemperature regulating the gametogenesis ofthree Laminaria species (Phaeophyceae).Journal of Phycology, 16: 1-15.Marengo, J. & Camargo, C. C. 2007. Surface airtemperature trends in Southern Brazil for1960-2002. International Journal ofClimatology, 28: 893-904.Mineur, F., Johnson, M. P., Maggs, C. & Stegenga,H. 2007. Hull fouling on commercial ships asa vector of macroalgal introduction. MarineBiology, 151: 1299-1307.Oliveira Filho, E. C., Pirani, J. R. & Giulietti, A. M.1983. The Brazilian seagrasses. AquaticBotany, 16: 251-267.Oliveira Filho, E. C. & Qi, Y. 2003. Decadalchanges in a polluted bay as seen from itsseaweed flora: the case of Santos Bay inBrazil. Ambio: A Journal of the HumanEnvironment, 32: 403-405.Oliveira, E. C. & Berchez, F. A. S. 1978. Algasbentônicas da Baía de Santos - alterações daflora no período de 1957-1978. Boletim deBotânica da Universidade de São Paulo, 6:49-59.Oliveira Filho, E. C., Horta, P. A. & Sant’ Anna, C.L. 2001. Algas e angiospermas marinhasbênticas do litoral brasileiro: diversidade,explotação e conservação. Instituto de Biociênciasda Universidade de São Paulo, Institutode Botânica, Seção Ficologia, São Pau-lo,SP, acessible at: http://migre.me/3Yd7r.(acessed 04/10/2008).Pan-American Journal of Aquatic Sciences (2010) 5(2): 350-357

Temporal changes in the seaweed flora357Orfanidis, S., Haditonidis, S. & Tsekos, I. 1996.Temperature requirements of Scytosiphonlomentaria (Scytosiphonales, Phaeophyta)from the Gulf of Thessaloniki, Greece, inrelation to geographic distribution.Helgol/inder Meeresunters, 50: 15-24.Orfanidis, S., Panayotidis, P. & Stamatis, N. 2001.Ecological evaluation of transitional andcoastal waters: A marine benthicmacrophytes-based model. MediterraneanMarine Science, 2: 45-65.Pereira, M. D., Schettini, C. A. F. & Omachi, C. Y.2008. Caracterização de feições Oceanográficasna Plataforma de Santa Catarina atravésde imagens orbitais. Revista Brasileira deGeofísica, 27: 81-93.Reis, R. P. & Yoneshigue-Valentin, Y. 1996.Distribuição das macroalgas na Lagoa deAraruama, estado do Rio de Janeiro, Brasil.Revista Brasileira de Botânica, 19: 77-85.Russel, B. D., Thompson, I. J., Falkenberg, L. J. &Connel, S. D. 2009. Synergistic effects ofclimate change and local stressors: CO 2 andnutrient-driven change in subtidal rockyhabitats. Global Change Biology, 15: 2153-2162.Schiel, D., Steinbeck, J. & Foster, M. S. 2004. Tenyears of Induced Ocean Warming causescomprehensive changes in Marine BenthicCommunities. Ecology, 85: 1833-1839.Siagenthaler, U. & Sarmiento, J. L. 1993.Atmosferic carbon dioxide and the ocean.Nature, 395: 119-125.Stachowicz, J. J., Terwin, J. R., Whitlatch, R. B. &Osman, R. W. 2002. Linking climate changeand biological invasions: ocean warmingfacilitates nonindigenous species invasions.Proceedings of the National Academy ofSciences of the United States of America,99: 15497-15500.Széchy, M. T. M. & Nassar, C. A. G. 2005. Floraficológica bentônica da Baía da Ribeira, suldo estado do Rio de Janeiro: avaliação apósduas décadas de operação da Central NuclearAlmirante Álvaro Alberto. In: SociedadeBrasileira de Ficologia. (Org.). Anais da XReunião Brasileira de Ficologia: Formaçãode ficólogos- um compromisso com a SUStentabilidadedos recursos aquáticos. 1 ed.Rio de Janeiro: Museu Nacional, 1, 373-397.Taouil, A. & Yoneshigue-Valentin, Y. 2002. Alteraçãoflorística na composição das algas daPraia de Boa Viagem (Niterói, RJ). RevistaBrasileira de Botânica, 25: 405-412.Vitousek, P. M., Aber, J. D., Howarth, R. W.,Likens, G. E., Matson, P. A., Schindler, D.W., Schlesinger, W. H. & Tilman, D. G. 1997.Human alteration of the global nitrogen cycle:sources and consequences. Ecology Applied,7: 737-750.Wainer, I., Venegas S. A., 2002. South AtlanticMultidecadal Variability in the ClimateSystem Model. Journal of Climate, 15:1408-1420.Wynne, J. M. 2005. A checklist of benthic marinealgae of the tropical and subtropicalwestern Atlantic: second revision. NovaHedwigia, 1-73.Yoneshigue-Valentin, Y. & Valentin, J. L. 1992.Macroalgae of the Cabo Frio. Pp. 31-50.Upwelling region, Brazil: ordination ofcommunities. In: Seeliger, U. (Ed.). Coastalplant communities of Latin America.Academic Press, San Diego.Received May 2010Accepted February 2011Published online March 2011Pan-American Journal of Aquatic Sciences (2010), 5(2): 350-357

RIO GRANDE DECLARATION1 st BRAZILIAN WORKSHOP ON CLIMATE CHANGES IN COASTALZONES: CURRENT KNOWLEDGE AND RECOMMENDATIONSUNIVERSIDADE FEDERAL DE RIO GRANDE (FURG)Rio Grande, RS, Brazil13 – 16 September 2009The impacts of Global Climate Change on the environment and society represent the biggestchallenge for human civilization in the twenty-first century. Scientists around the world are workingintensely to understand the climatic processes involved and the possible consequences of climate changeat global, regional and local levels. Governments of different countries have initiated studies ofvulnerability to climate change and adopted mitigation and adaptation measures to face this new reality.The establishment of the Brazilian Research Network on Global Climate Change (Rede CLIMA)and the implementation of the National Institute of Science and Technology (INCT) for Climate Change,were important initiatives to adequately address these issues in Brazil; since these organizations involvescientists of different disciplines. One of the main objectives of Rede CLIMA is to significantly increasethe knowledge on the impact of climate change and to identify key vulnerabilities in different Braziliansectors and systems. Coastal Zones stand out as an important system due to its environmental and societalsignificance.Large cities and populations are concentrated near rivers and in low altitude regions (fertilevalleys) within 100 km of the coastline, and population density of the coastal zone is likely to more thandouble by 2050. Impacts of climate change and urban development will triple the number of peopleexposed to coastal flooding by 2070. Goods and services from coastal ecosystems valued by societyrepresent about 33 trillion dollars globally. Coastal zones are therefore, among the most vulnerable areasto global climate change impacts, since they will be directly affected by the increase in the average sealevel, exposure to extreme storms, changes in discharge regimes of rivers, elevation of sea surfacetemperature, ocean acidification and other events. However, potential impacts of climate change, bothphysical and biological, will vary considerably among coastal regions, according to their naturalcharacteristics and the degree of environmental degradation. Therefore, understanding the impacts ofglobal climate change in every region is essential for strategic planning and decision-making by thegovernment and the Brazilian society.During the "First Brazilian Workshop on Climate Changes in Coastal Zones" in Rio Grande (RS),scientists from around the country assessed the current state of knowledge on impacts of climate changeon Brazilian coastal zones and discussed procedures to standardize protocols and strategies for networkingobservational studies. About 200 university professors, graduate and post-graduate students attended theevent. Among these, thirty-five were invited speakers from Rede CLIMA and the National Institute ofScience and Technology (INCT) for Climate Change. Thirty-eight research papers (among oral andpanels) were presented, representing 121 authors from different Brazilian institutions and regions. BasedPan-American Journal of Aquatic Sciences (2010), 5(2): XII-XV

XIIIon the current state of knowledge and the discussions held during the workshop, the participants believethat it is still possible to save the coastal ecosystem and its environmental assets (ecological, social andeconomic) against scenarios of climate change, although urgency and determination are required toachieve this task.In order to adequately assess and monitor the effects of climate change on coastal ecosystems inBrazil, in an objective and regional manner, the following scientific goals have to be achieved urgently:1) Validation of regional climate models based on local observational data;2) Geodetically controlled measurements of sea level accompanied by altimetric surveys,integrating terrestrial and nautical cartography of important coastal regions of Brazil with scenarios for thetwenty-first century;3) Acquisition of long and sustainable time series of physical, chemical and biological processesin coastal waters;4) Greater understanding of factors controlling the processes of erosion and coastal progradation;5) Evaluation of the potential consequences of climate changes on aquatic biogeochemical cycles;6) Analysis of the responses of physiological and ecological populations, marine, estuarine andfreshwater communities and ecosystems on climate change;7) Assessment of variability in fish stocks and other natural resources of economic importance;8) Evaluation of social and economic vulnerability of coastal populations, particularly those thatdirectly depend on coastal resources and traditional activities.The advances in scientific knowledge on coastal ecosystems, with emphasis on the topics outlinedabove, will lead to better insights of the effects of climate change on coastal regions. Investments inenvironmental sciences in coastal areas, with emphasis on climate change are essential, therefore, forgreater understanding of these important ecosystems and their vulnerabilities.In this context, we recommend to the government and organized society that actions be created orstrengthened to promote:• The immediate reduction of emissions of greenhouse gases (GHGs) in order to contribute toslowing down global warming;• The immediate deterrence of deforestation in different regions of the country;• Advances in scientific knowledge on coastal ecosystems, with particular emphasis on topicsalready presented above, through the induction and effective support for research in these subject areas;• The strengthening of the monitoring system of the Brazilian Coastal Zone;• The development and implementation of management plans that promote the use, conservationand restoration of coastal ecosystems, considering climate change scenarios, thus strengthening existingand incidental public policies on this zone (National Coastal Management Program, “Orla” Project, SectorPlan for Sea Resources, National Plan for Water Resources, The Conservation National System, LocalAgenda 21);• The promotion and encouragement innovative solutions and actions that encourage adaptationmeasures in coastal cities and towns facing the new climate scenario;• The expansion of the critical insight and awareness of society regarding the Climate Change,through formal education (via educational institutions) and non-formal (via the media, nongovernmentalorganizations, civil organizations etc.), with dissemination of clear and contextualized information aboutscientific aspects of the topic in the appropriate language.For the implementation of these recommendations to be successful, depend on decision fromlocal, State and Federal public policies engaged to Climate Change. Agility and long-term commitmentare essential requirements to support and encourage the efforts of science and technology to confront andadapt to the challenges posed by climate change.Rio Grande, October 20 th 2009.Declaration approved by 124 scientists linked to 20 federal and state institutions (distributed in eightstates of the Brazilian coast) and three foreign institutions, belonging to leading groups in research onissues related to the sciences of the sea.Pan-American Journal of Aquatic Sciences (2010), 5(2): XII-XV

XIVResearchers from Rede Clima and National Institute of Science and Technology (INCT) for climate changeABDALAH, Patrizia - Universidade Federal do RioGrande, RSALBERTONI, Edélti - Universidade Federal do RioGrande, RSARAÚJO, Francisco Gerson - Universidade Federal Ruraldo Rio de Janeiro, RJCALLIARI, Lauro Júlio - Universidade Federal do RioGrande, RSCAMPOS, Edmo - Universidade de São Paulo, SPCASTELLO, Jorge Pablo - Universidade Federal do RioGrande, RSCOPERTINO, Margareth da Silva - Universidade Federaldo Rio Grande, RSCIOTTI, Áurea Maria Ciotti - Universidade Estadual deSão Paulo, SPDA SILVA, Mário Pereira - Universidade Federal do RioGrande do Norte, RNDOMINGUES, José Maria Landim - UniversidadeFederal da Bahia, BAFARACO, Luís Francisco Ditzel - Instituto ChicoMendes, PRFARIAS, Julyana Nóbrega - Universidade Federal deSanta CatarinaFERNANDES, Elisa Leão - Universidade Federal do RioGrande, RSGARCIA, Alexandre Miranda - Universidade Federal doRio Grande, RSGARCIA, Carlos Alberto Eiras - Universidade Federal doRio Grande, RSGHERARDI, Douglas - Instituto Nacional de PesquisasEspaciais, SPGHISOLFI, Renato - Universidade Federal do EspíritoSanto, ESGIANESELLA, Sônia Maria Flores - Universidade deSão Paulo, SPGONÇALVES, Gláuber Acunha - Universidade Federaldo Rio Grande, RSHELLEBRANDT, Dênis - University of East Anglia, UKHELLEBRANDT, Luceni - Universidade Federal do RioGrande, RSHORTA, Paulo Antunes - Universidade Federal de SantaCatarina, SCKLEIN, Antônio Henrique da Fontoura - Universidade doVale do Itajaí, SCKIKUCHI, Ruy Kenji Papa - Universidade Federal daBahia, BAKRUSCHE, Nísia - Universidade Federal do Rio Grande,RSLANA, Paulo - Universidade Federal do Paraná, UFPRLEÃO, Zelinda Margarida Nery - Universidade Federalda Bahia, BALEMOS, Ângelo Teixeira - Universidade Federal doEspírito SantoLUZ, Roberto Teixeira - Instituto Brasileiro de Geografiae Estatística, RJMACHADO, Arthur Antônio - Universidade Federal doRio Grande, RSMARQUES, William - Universidade Federal do RioGrande, RSPan-American Journal of Aquatic Sciences (2010), 5(2): XII-XVMÖLLER, Osmar Olinto - Universidade Federal doRio Grande, RSMUEHE, Dieter - Universidade Federal do Rio deJaneiro, RSMUELBERT, José Henrique - Universidade Federal doRio Grande, RSNOERNBERG, Maurício Almeida - UniversidadeFederal do Paraná, PRNOBRE, Carlos Afonso - Instituto Nacional dePesquisas Espaciais, SPNOBRE, Paulo - Instituto Nacional de PesquisasEspaciais, SPOLIVEIRA, Marília de Dirceu Machado de -Universidade Federal da Bahia, BAPAES, Eduardo Tavares - Instituto Nacional dePesquisas Espaciais, SPPEREIRA, Natália - Universidade Federal do RioGrande, RSSOARES, Helena Cachanhuk - Instituto Nacional dePesquisas EspaciaisSIEGLE, Eduardo - Universidade de São Paulo, SPSILVA, Cléber Palma - Universidade Federal do RioGrande, RSTURRA, Alexander - Universidade de São Paulo, SPTROTTE-DUÁH, Janice - DHN/ GOOS BRASILVASCONCELLOS, Vivian Soares - UniversidadeFederal da Bahia, BAVIEIRA, João Paes - Universidade Federal do RioGrande, RSVITAL, Helenice - Universidade Federal do RioGrande do Norte, RNGuestsGARCIA, Virgínia Maria Tavano - UniversidadeFederal do Rio Grande, RSMATA, Maurício Magalhães - Universidade Federaldo Rio Grande, RSSOARES, Ivan Dias - Universidade Federal do RioGrande, RSVASCONCELLOS, Marcelo - Universidade Federaldo Rio Grande, RSOther authors and co-authors of presented papersABREU, Paulo César - Universidade Federal do RioGrande, RSALMEIDA, João - Universidade de São Paulo, SPAMADO FILHO, Gilberto - Jardim Botânico do Riode Janeiro, RJAMARAL, Antônia C Z - Universidade Estadual deCampinas, SPAMARAL, Waldemar J A - Universidade Federal doRio Grande, RSARAÚJO, Rafael Sperb - Universidade do Vale doItajaí, SCARAÚJO, Renato - Universidade Federal de SantaCatarina, SCARIGONY-NETO, Jorge - Universidade Federal doRio Grande, RSAZEVEDO, Márcia Cristina Costa - UniversidadeFederal Rural do Rio de Janeiro, RJ

XVBAISCH, Paulo Roberto Martins - Universidade Federaldo Rio Grande, RSBARROS, Marcos P F - Universidade Federal do Rio deJaneiro, RJBEMVENUTTI, Carlos Emílio - Universidade Federal doRio Grande, RSBERCHEZ, Flávio - Universidade de São Paulo, SPBERGESCH, Marli - Universidade Federal do RioGrande, RSBERSANO, José G F - Universidade Federal do RioGrande, RSBIANC, Andre De Pieri - Universidade do Vale do Itajaí,SCBRUNO, Marcelo A - Universidade Federal do RioGrande, RSCABRAL, Débora - Universidade Federal de SantaCatarina, SCCAMARGO, Maurício - Universidade Federal do Paraná,PRCAPPELLETTO, Eliane - Universidade Federal do RioGrande, RSCOIMBRA, Franciane - Universidade Federal do RioGrande, RSCOLEPICOLO, Pio - Universidade de São Paulo, SPCOLLING, Leonir André - Universidade Federal do RioGrande, RSCORRÊA, Iran Carlos S - Universidade Federal do RioGrande do Sul, RSCOSTA, César Serra Bonifácio - Universidade Federal doRio Grande, RSDENADAI, Márcia F - Instituto Costa BrasilisDONNANGELO, Alejandro - Universidade Federal deSanta Catarina, SCESTEVES, Francisco de Assis - Universidade Federal doRio de Janeiro, RJFREITAS, Dominício - Universidade do Vale do Itajaí,SCFUJI, Mutue - Instituto de Botânica do Estado de SãoPaulo, SPFURLANETTO, Leonardo - Universidade Federal do RioGrande, RSGIACOMINI, Yara B - Universidade Federal do RioGrande, RSGIANASI, Bruno Lainetti - Universidade Federal do RioGrande, RSGIANUCA, Dimas - Universidade Federal do RioGrande, RSGIOVANINI, Renata M. Bretz - Universidade Federal doRio Grande, RSGOULART, Elaine - Universidade Federal do RioGrande, RSGUIMARÃES, Sílvia Pita - Instituto de Botânica doEstado de São Paulo, SPHIRATA, Fernando Endo - School of Earth andAtmospheric Sciences, EUAJORGE, Daniel F Schroeder - Universidade EstadualPaulista, SPLESSA, Guilherme - Universidade Federal da Bahia, BAMACHADO, Maria Isabel - Universidade Federal do RioGrande, RSMACHADO, Luis Eduardo - Universidade Federal doRio Grande, RSMAIA, Natan - Universidade Federal do Rio Grande,RSMARANGONI, Juliano C - Universidade Federal doRio Grande, RSMARINHO, Cláudio C - Universidade Federal do Riode Janeiro, RJMARTINS, Aline - Universidade de São Paulo, SPMEDEANIC, Svetlana - Universidade Federal do RioGrande do Sul, RSMENEZES, João Tadeu - Universidade do Vale doItajaí, SCMONTEIRO, Patricia C - Universidade Federal deSanta Catarina, SCMORAES, Leonardo E - Universidade Federal do RioGrande, RSMELO FILHO, Eloi - Universidade Federal do RioGrande, RSODEBRECHT, Clarisse - Universidade Federal do RioGrande, RSOLIVEIRA, Joyce A - Universidade Federal do RioGrande, RSPEREIRA, Pedro - Universidade Federal do RioGrande, RSPEZZI, Luciano - Instituto Nacional de PesquisasEspaciais, SPPINTO, Camila - Universidade do Vale do Itajaí, SCRECHIA, Rafael - Universidade Federal do RioGrande, RSRUGNA, Rafael C - Universidade de São Paulo, SPSALDAÑA-CORRÊA, Flávia M P - Universidade deSão Paulo, SPSANCHES, Paola - Universidade Federal de SantaCatarina, SCSANTOS, Margareth B - Universidade Federal do RioGrande, RSSEELIGER, Ulrich - Universidade Federal do RioGrande, RSSERRA, Sérgio Pastor - Universidade do Vale doItajaí, SCSHROEDER, Fábio - Universidade Federal do RioGrande, RSSILVA, Maira S M - Universidade Federal do RioGrande, RSSILVA, Maristela B - Universidade Federal do RioGrande, RSSISSINI, Marina N - Universidade Federal de SantaCatarina, SCSTIVE, Marcel - Delft University of Technology,HolandaTOLDO, Elírio - Universidade Federal do Rio Grandedo Sul, RSYOSHIMURA, Cristalina - Universidade Federal deSanta Catarina, SCZANANDREA, Ana - Universidade Federal de SantaCatarina, SCZANELLA, Nicolas Paolo - Universidade Federal doRio Grande, RSPan-American Journal of Aquatic Sciences (2010), 5(2): XII-XV

Download full issue - PanamJAS

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?