Species Richness Patterns in Space and Time in the Himalayan Area Chitra Bahadur Baniya Dissertation for the degree of philosophiae doctor (PhD) University of Bergen, Norway 2010  Species Richness Patterns in Space and Time in the Himalayan Area Chitra Bahadur Baniya Dissertation for the Degree of Philosophiae Doctor (PhD) University of Bergen, Norway, 2010 All photographs taken by Chitra Bahadur Baniya Photographs from left to right and first to the last rows: 1. Upper Manang, 3475 m, Nepal showing both active and abandoned fields 2. Buddha Mountain, 4985-5685 m, Tibet: Early snow fall in 2004 3. Bhraka, Upper Manang, 3175 m, Nepal showing both active and abandoned fields 1. Pedicularis roylei, 5010-5260 m, Buddha Mountain, Tibet 2. Meconopsis horridula, 5035-5360 m, Buddha Mountain, Tibet 3. Gentiana urnula, 5260-5460 m, Buddha Mountain, Tibet 1. Ajania rubigena, 5110-5385 m, Buddha Mountain, Tibet 2. Lamiophlomis rotata, 4985-5135 m, Buddha Mountain, Tibet 3. Cremanthodium nanum, 5335-5360 m, Buddha Mountain, Tibet 1. Cremanthodium ellisii, 5210-5435 m, Buddha Mountain, Tibet 2. Delphinium brunonianum, 5385-5435 m, Buddha Mountain, Tibet 3. Saussurea tridactyla, 5035-5310 m, Buddha Mountain, Tibet 1. Arctocetraria nigricascens, 5085-5585 m, Buddha Mountain, Tibet 2. Thylacospermum caespitosum, 5310-5510 m, Buddha Mountain, Tibet 3. Saussurea sp., 5210-5285 m, Buddha Mountain, Tibet II Contents DECLARATION............................................................................................................................................... IV ACKNOWLEDGEMENTS..............................................................................................................................III ABSTRACT.................................................................................................................................................... VIII LIST OF PUBLICATIONS................................................................................................................................X Synthesis INTRODUCTION............................................................................................................................................... 1 AIMS ................................................................................................................................................................. 4 BUDDHA MOUNTAIN, TAR- PAPER I .................................................................................................................. 5 TIBET’S HIGHEST ALPINE ZONE- PAPER II........................................................................................................... 6 NEPAL-PAPER III .............................................................................................................................................. 7 MANANG-PAPER IV ........................................................................................................................................... 9 TERMINOLOGY ............................................................................................................................................. 10 SCALE ............................................................................................................................................................... 10 AREA ................................................................................................................................................................. 14 VARIABLES AT THE LOCAL-SCALE........................................................................................................ 14 VARIABLES AT THE LANDSCAPE-TO-MACRO-SCALES.................................................................... 15 VARIABLES AND MECHANISMS INFLUENCING SPATIO-TEMPORAL RICHNESS PATTERNS17 SAMPLING ARTEFACTS ..................................................................................................................................... 17 MASS EFFECTS ................................................................................................................................................ 18 SPECIES-POOL ................................................................................................................................................. 18 HISTORICAL FACTORS AND LOCAL-SCALE COMMUNITY SUCCESSIONAL PATTERNS............................................... 19 CLIMATE AND ENERGY-RELATED VARIABLES TO EXPLAIN LANDSCAPE AND MACRO-SCALE ELEVATIONAL PATTERNS ....................................................................................................................................................................... 20 MECHANISMS UNDERLYING THE UNIMODAL RICHNESS PATTERN OF LICHENS OF NEPAL AND TAR ...................... 22 STRESS, STABILITY, DISTURBANCE AND WATER-ENERGY DYNAMICS ..................................................................... 23 CONCLUSIONS ............................................................................................................................................... 25 FUTURE PLAN ................................................................................................................................................ 26 REFERENCES.................................................................................................................................................. 27 Papers I-IV III Declaration This dissertation consists of an introduction, two published, and two papers submitted for publication. Three papers are co-authored and one is a single author paper. The publishers of both the published papers granted copyright permission to include them in this thesis. The nature of the contributions by the different authors is outlined below. Paper I: Baniya, C.B., Solhøy, T., Gauslaa, Y. & Palmer, M.W. Richness and Composition of Vascular Plants and Cryptogams along a High Elevational Gradient on Buddha Mountain, Central Tibet. (Submitted). Chitra Bahadur Baniya – Ideas, concept, field design/work, data processing, statistical analysis, writing, editing, and correspondence Torstein Solhøy – Supervision, rechecked field work Yngvar Gauslaa – Editing Michael W. Palmer – Statistical analysis, editing Paper II: Baniya, C.B. Vascular and cryptogam richness in the world’s highest alpine zone, Tibet. (Submitted). Own idea, concept, data collection, processing, writing and corresponding Paper III: Baniya, C.B., Solhøy, T., Gauslaa, Y. & Palmer, M.W. 2010. The elevation gradient of lichen species richness in Nepal. The Lichenologist, 42 (1): 83-96. Chitra Bahadur Baniya – Ideas and concept, literature and herbarium survey, processing, statistical analysis, writing, editing, review, correspondence to the journal Torstein Solhøy – Supervision, editing Yngvar Gauslaa – Editing, making graphs, guiding to correspond the journal Editor Michael W. Palmer – Statistical analysis, editing Paper IV: Baniya, C.B., Solhøy, T. & Vetaas, O.R. 2009. Temporal changes in species diversity and composition in abandoned fields in a trans-Himalayan landscape, Nepal. Plant Ecology, 201: 383-399. Chitra Bahadur Baniya – Ideas, concept, field design, field work, taxonomic identification, data processing, data analysis, writing, and editing Torstein Solhøy – Supervision Ole Reidar Vetaas – Supervision, editing, correspondence to the journal IV Acknowledgements This dissertation could not have reached completition without positive thoughts, significant guidance, practical help and advice. A foreign PhD student such as me who did fieldwork in outside of my own received substantial help from a number of people and organizations. I would like to acknowledge them all. Thanks to the Norwegian State Loan Fund (Lånekassen) for funding this study. I thank the Service Commission, Science Dean Office, and Central Department of Botany, Tribhuvan University, Nepal for a study leave; to the Network for University Co-operation Tibet-Norway, Oslo, and Faculty of Mathematics and Natural Science, Bergen University. I am indebted to Torstein Solhøy for supervising me. I benefitted greatly from many aspects of his academic and social networks. I got a great help from both of my co-supervisors Yngvar Gauslaa, Norwegian University of Life Sciences, Ås and Mike W. Palmer, Oklahoma State University, Oklahoma, USA. I will never forget Yngvar, all his encouragements, and practicalities in strengthening my academic field. Many thanks to Mike, he visited me and my family from such a long distance. I got his great time and insights to be realistic on natural data than model. My sincere thanks also go to Ole Reidar Vetaas and Per Harald Salvesen for helping me to come to Bergen University as their MPhil student. I thank to Walter Obermayer, University Graz for helping me in many ways and also thank to all my co-authors for their valuable comments and time. Academic difficulties during my PhD such as plant identification, statistical analysis, analysing software, books, literature, writing and editing were made possible and easier by John Birks. I got whatever I wanted from him. His valuable comments and remarks after each edition on my manuscripts made me optimistic. I am very much fortunate that he and Hilary invited me to several lovely dinners at their home, got nice time to watch slides and DVDs of his expeditions in High Asia, and alpine areas of the world. My sincere gratitude is always to John Birks and Hilary Birks. Without their help I would not have been able to achieve a PhD. I got both academic and administrative help from all members of the Department of Biology, the EECRG research group, the Science Library, and Centre for Foreign Students, University of Bergen. All are highly acknowledged. Some special names are Tor Tønsberg, Per Magnus V Jørgensen, Dagfinn Moe, Torbjørg Bjelland, Dag Olav Øvstedal, Astri Boten, Solfrid Hjelmtveit, Beate Helle, Peter Emil Kaland, Einar Heegaard, John-Arvid Grytnes, Vigdis Vandvik, Richard Telford, Ingrid Solhøy, Ana Veronica Cordova, Leikny Lavik, Hege Folkestand, Late. Arvid Kleppe and Tommy Strand who helped me in many ways. Also help provided by Knut Helge, Arild Breistøl, Frode Falkenberg, Gaute Bø Grønstøl, and John Skarveit was indispensible. I am grateful to my respected teachers Govind Prasad Sharma Ghimire and Pramod Kumar Jha from the Botany Department, Kirtipur, Nepal and Bijaya Kattel from Florida, USA. I am happy that they motivated me in the field of botany, plant ecology and lichenology. Many thanks to Ram Prasad Choudhary, Krishna Kumar Shrestha, Sanu Devi Joshi, Janak Devi Manandhar, Vimal Narayan Prasad Gupta, all teachers, staff and students of Central Department of Botany who helped me in different ways during my PhD. Many thanks for sharing all my responsibilities of teaching during the time when I was away. I will not forget the time and life I shared with many international students at Fantoft, staffs at Fantoftgård kindergarten. I got help from many international friends. Renuka Priyantha Ilandarige, a student at UoB from SriLanka receives my special thanks for his kindness in sharing with me many aspects as well as time with his family. The Nepalese Society in Bergen (NEBERS), Nepalese students and families at Fantoft always kept me and my family in touch and alive with all things Nepalese and Nepal. All Nepalese at Bergen and particularly Krishna Babu, Keshav Paudel, Prativa, Bishnu, Kapil, Deepak, Rajiv, Buddha, Meeru, Jyoti, Chandana and Anank Bhandari were cooperative in many ways. I enjoyed the company of my colleagues Marianne Presthus Heggen, Arguitxu de la Riva Caballero, Bjørn Arild Hatteland, Brooke Wilkerson, Ingelinn Aarnes, Ronald Semyalo, and La Duo for sharing coffees and lunch with me and many invaluable discussions during my PhD life. My fieldwork in Tibet and Manang, Nepal would not have been possible without great help provided by friends such as Tsering, La Qiong, CaiDong, PuBu, Migmar, Dronba, Joachim Schmidt, Chhong Norupa, Ram Prasad and others. Thank you all. I got time to participate in PhD courses in Sweden and Iceland. Many thank to Kristina Articus, Uppsala University, Sweden and Bjarni E. Guðleifsson, Agricultural University, Iceland for giving me valuable time. My special thanks go to Cathy Jenks for her valuable time in language correction and practical suggestions. VI Finally, I would like to dedicate this PhD to my respected parents who raised me in such a remote village with never-ending love and faith, to my dear brothers and sister who encourage me regularly, my wife Hira who is painlessly helping me with her full support, co-operating with me in my every difficulty of PhD life, especially the final year when I run out of financial support, my dear wise and lovely kids Pratima, Prabhat and Pratik who are patiently looking forward for me to finish my PhD and to rejoin their school back home, and are always eager to look at BUWA’s (father) happy face when I returned home every evening from University and my brothers-in-law, Nanda Bahadur and Shiva who are helping me regularly from Nepal. Bergen, December, 2009 Chitra Bahadur Baniya VII Abstract Baniya, C.B. 2010. Species Richness Patterns in Space and Time in the Himalayan Area. PhD Thesis, Faculty of Science, University of Bergen, Norway. Aims: The Himalayan Mountains are highly sensitive to current global climate changes and their local impacts. Potentially more sensitive are lichens growing throughout the Himalaya and species living in the alpine zones of the world. In this synthesis I examine the variation in species composition and richness patterns at different scales of spatio-temporal gradients from mountainous areas in the Tibet Autonomous Region (TAR) and the Nepalese Himalaya. Methods: This thesis is based on field-survey data as well as secondary data from published floras. The study of the temporal responses of plant species composition, richness and soil heterogeneity is based on direct field sampling in old abandoned fields forming different chronosequences in the trans-Himalayan zone, Nepal. Likewise, the elevational response of species composition, richness patterns and soil nutrient heterogeneity are investigated by direct field observations and sampling in Buddha Mountain, Nyenchentanglha Shan, Central Tibet, TAR. Buddha Mountain elevation gradient study is compared with similar species but based on secondary data, namely published floras over almost a similar elevational range. The elevation gradient of lichen species in Nepal has been studied based on published literature on Nepalese lichens. Major findings: (i) Lichens, both in TAR, on Buddha Mountain and in the Nepalese Himalaya represent the organism group with the highest elevational occurrences and a high degree of adaptations to extreme environments. A unimodal elevational relationship is common in lichen species richness. Maximum richness for Nepalese endemic lichens coincides with the maximum elevation range for Nepalese vascular plant endemics. (ii) Landscape-level elevational species richness patterns (Buddha Mountain) show an early plateau and a sharp decline afterwards that differs from the smooth decling pattern with the interpolated macro-scale elevational ranges. This may indicate that differences in scale of the VIII measurement of species, averaging out of environmental heterogeneities and the influence of climatic variables increasing with increasing spatial scales are all potentially important. (iii) The general pattern for the total richness hides a variety of specific patterns for the subgroups of life-forms and substrate types considered. (iv) A divergence pattern of secondary succession is indicated after increasing beta diversity along the temporal gradient in the trans-Himalayan zone. But the total species richness clearly indicated a unimodal species richness pattern in the secondary succession. Decreased richness with increased beta diversity towards the oldest abandoned fields highlights the potential importance of the high intensity of grazing, and grazing-tolerant ‘nurse’ species protecting future forest species. Conclusions: Species richness and composition patterns are highly scale dependent at all levels of spatio-temporal gradients. Scales at all hierarchical levels are inter-related. Some variables such as soil-related local-scale heterogeneity are highly important at the local-scale species composition and richness patterns. But that type of variables will not be significant or will be averaged out at a broader scale. Climate-related variables are highly influential at the broad-scale but can also be influential at local scales. Species richness patterns depend on the studied group of species, functional group of species, life-forms and their geography. The ecological significance of species depends on elevation. For example, in Nepal an elevation of 500 m is suitable for big wild animals, likewise 1000 m for vascular plants and 3100 to 3400 m for lichens. This indicates an important conservation strategy. Implications: Further investigation on elevational and temporal species richness patterns in other landscapes of both Tibet and Nepal Himalaya will extend the findings of the present study and help to predict the local impact of broad-scale climate change on the biodiversity hotspots of the Himalaya. Keywords: Species richness, lichen, elevation gradient, Nepal, Himalaya, Tibet IX List of publications The dissertation is based on the following four publications, and they will subsequently be referred by their Roman number in the synthesis. Paper I: Baniya, C.B., Solhøy, T., Gauslaa, Y. & Palmer, M.W. Richness and Composition of Vascular Plants and Cryptogams along a High Elevational Gradient on Buddha Mountain, Central Tibet. (Submitted). Paper II: Baniya, C.B. Vascular and cryptogam richness in the world’s highest alpine zone, Tibet. (Submitted). Paper III: Baniya, C.B., Solhøy, T., Gauslaa, Y. & Palmer, M.W. 2010. The elevation gradient of lichen species richness in Nepal. The Lichenologist, 42 (1): 83-96. Paper IV: Baniya, C.B., Solhøy, T. & Vetaas, O.R. 2009. Temporal changes in species diversity and composition in abandoned fields in a trans-Himalayan landscape, Nepal. Plant Ecology, 201: 383-399. X Synthesis Species Richness Patterns in Space and Time in the Himalayan Area Chitra Bahadur Baniya Introduction Every mountain has its own special meaning and symbolism at least in the culture in which I grew up. For example, people often go for pilgrimage to Mount Kailash (in western Tibet) in the belief that the mountain was the birth-place of the God Shiva. Many wonders and questions are associated with each mountain besides its beauty. Explaining differences in species composition at a particular time and space are one of the hardest problems for an ecologist. A great mystery is concerned with how species differ in their number, type and zonation within and between mountains. It has been almost 200 years since biogeographers tried to explain scientifically the latitudinal gradient of species diversity. Willdenow (1805), von Humboldt (1849), Darwin (1859), Wallace (1878), etc. were among the early famous biogeographers, evolutionary biologists and ecologists who initiated studies of diversity gradients. Willdenow (1805) was perhaps the first to present the soil and heat-energy related hypotheses to explain species richness patterns. However, gradients in species diversity are an old concept. It may have been known to Stone Age people because they were familiar about where to find prey after fighting, burning etc. (Lomolino 2001, Lomolino et al. 2006). Different names are given to their daily routine nowadays, such as spatial autocorrelation and non-random distribution of resources for finding game animals after fire, etc. The altitudinal (elevational) gradient is believed to be a mirror of the latitudinal gradient in species richness (Stevens 1992, Rahbek 1995, 1997, 2005). There are, however, some differences such as the condensation of huge latitudinal area into a short, narrow elevation band (Körner 2003, 2007). Many ecologists have strongly utilized both latitude and altitude as the main predictors for species richness in their studies (Pianka 1966, Rohde 1992, Brown 1995, Vazquez & Givnish 1998, Odland & Birks 1999, Gaston 2000, Körner 2000, Ohlemuller & Wilson 2000, Brown 2001, Heaney 2001, Lomolino 2001, Md. Nor 2001, Grytnes 2003, Bhattarai et al. 2004, Carpenter 2005, Herzog et al. 2005, Oommen & Shanker 2005, Rahbek 2005, Wilson et al. 2005, Bruun et al. 2006, Fontaneto & Ricci 2006, Harris 2006, Lomolino et al. 2006, Grau et al. 2007, Mittelbach et al. 2007, Qian et al. 2007, Romdal & Grytnes 2007, Kessler 2009). A review by Rahbek (2005) found 204 studies that concentrated on species richness patterns for different species along the macro-scales of 1 elevational or latitudinal gradients in different parts of the world. More than 120 different plausible hypotheses related to species richness have been compiled (Palmer 1994), and the number is increasing. However, many of them are neither mutually exclusive, nor inclusive, with more than 100 violating the assumptions of the competitive exclusion principle (Palmer 1994). After Brown (1995) laid the foundation of macro-ecology, studies on elevational gradients became popular among ecologists to answer earlier questions related to diversity theory for both latitude and altitude. The altitudinal gradient is believed to be a natural experimental station on each mountain (Körner 2007). Studies on species richness patterns that mainly focus on area (Palmer & White 1994, Rosenzweig 1995), energy (Wright 1983, Palmer & Dixon 1990), water-energy (O' Brien 1993, Hawkins et al. 2003), evolutionary time (Ricklefs 1987, Begon et al. 1990, Mittelbach et al. 2007), competition and disturbance (Connell & Slatyer 1977, Grime 1977, Connell 1978, Huston 1979, Tilman 1982), fire (Auclair et al. 1976), and the species pool concept (Zobel 1997) have contributed much. There is now increasing consensuses about that macro-scale patterns are resulting from available energy, evolutionary time, habitat heterogeneity, area and geometric constraints (Mittelbach et al. 2007). Despite this consensus, debates continue regarding issues on other aspects of species richness patterns. Similarly, the foundation of the temporal gradient in species richness patterns was laid by Cowles (1899) after his study on plant colonization on the sand-dunes of Lake Michigan. It greatly influenced Fredric E. Clements who later presented a challenging holistic theory of succession (Clements 1928) and ecosystem development. This holistic view of succession is hotly debated among ecologists. The contrasting individualistic theory of succession and ecosystem development was proposed by Gleason (1926). Both theories are still critical in studies on succession and community ecology. It has been believed that both these theories may be appropriate but at different scales (e.g., Odum 1971, Horn 1974, Pickett 1976, Glenn-Lewin 1980, Christensen & Peet 1984, Carson & Barrett 1988, Inouye & Tilman 1995, Fukami et al. 2005, del Moral 2007). Questions of scale will be discussed below. North America became one of the highly active areas in research related to secondary succession or old-field succession study followed by work in Europe (Bazzaz 2005). The Initial Floristic Composition (IFC), and Relay Floristic Composition (RFC) models of succession (Horn 1974), and the facilitation, inhibition and tolerance mechanisms of 2 succession (Connell & Slatyer 1977) were well accepted in successional studies. Each theory is not free of debate. The Long Term Research (LTR) at the Park Grass Experimental Station of Rothamsted, England is a good tool to resolve some of these debates (Silvertown et al. 2006). In addition, long-term and chronological short-term temporal responses in species richness patterns on glacier forelands and in palaeoecological research will help to resolve some of these debates. Surprisingly, the Himalaya, one of the global biodiversity hotspots (including its central part, Nepal) is poorly explored in term of its biodiversity. Within the Himalaya, lichens which are known to be a very sensitive group of species to environmental change have been completely ignored. There have been some taxonomic studies done on this group by foreign as well as Nepalese scientists. More than 2000 species of lichens are expected to occur in Nepal (Sharma 1995), but a critical documentation hardly exceeds 525 species (Paper III). When I looked at the history of the botanical exploration of Nepal, it dates back to 1802 during which the first botanical exploration of Nepal was done by Francis BuchananHamilton, a superintendent at the Royal Botanic Garden, Calcutta (Rajbhandari 1976). The second major exploration was done in Eastern Nepal by Sir J. D. Hooker in 1848. The first elevational species gradient paper related to forest-tree species in Nepal was published by Yoda (1967). Almost after three decades an another elevational species richness pattern on animals and mammals of Nepal was published by Hunter & Yonzon (1993). In-between, the vegetation and ecological map of Nepal was published (Dobremez & Jest 1969). Research regarding to Nepalese biodiversity is now increasing in the current decade. Unfortunately, Nepal does not have a flora, and it is thus far from knowing its real diversity status. It is the same in the Tibet Autonomous Region (TAR). Flora Xizangica was published (Cheng-yih 1983-1987) by the Chinese Academic of Sciences but it has not been revised after its initial publication. Both Nepal Himalaya as well as TAR are far from knowing the real status of biodiversity of its flora and fauna. Both areas lie in a highly sensitive and highly vulnerable zones with respect to predicted global climate changes (IPCC 2007). The high mountains of Nepal and Tibet share and store more than 85% of the water necessary for people living further down slopes or on the plains. Most of the biggest Asian rivers originate from here. IPCC (2007) says that almost 82% of the TAR’s glaciers have already vanished as have glaciers in Nepal. Unfortunately, we do not know which species we already have lost, and 3 that we are currently losing. Loss of biodiversity is an inevitable result of change in climate that has been experimentally proved (Silvertown et al. 2006). Such perspectives urged me to start biodiversity research in the Nepalese Himalaya. First I tried to understand the response of species along short chronosequences. Abandonment of agricultural and pastoral land is a common and dominant practice mostly in the high transHimalayan zone in Nepal. Thus I selected this as a main source of landscape change to study short-term temporal changes in species richness patterns in Nepal together with elevational species richness patterns. I chose one mountain in TAR to study the elevational pattern of all species including lichens. Later, I compared this local elevational species richness pattern with the species richness of the whole TAR using data from floras. After comparing the results from different ecological approaches, contrasting spatial and temporal scales of investigations, and rigorous statistical testing, I hope to maximize the scientific soundness of my conclusions and to increase ecological understanding from my work. Aims Climatically harsh and unstable conditions at high elevations may give a decline in species richness but this is not the case on sub-tropical mountains directly, on a macro-scale, or for functional groups. On the climatically more congenial sub-tropical mountains, the Nepalese elevation gradient shows unimodal species richness patterns for vascular plant, fern, moss, liverwort and orchid richness but is not known for one of the most sensitive organisms in nature, namely lichens. The study of species richness patterns in the short temporal change gradient in the trans-Himalayan zone is one of the first studies of its kind in this area. The specific questions that this thesis addresses are as follows: Paper I: Richness and Composition of Vascular Plants and Cryptogams along a High Elevational Gradient on Buddha Mountain, Central Tibet. • How does floristic composition vary within a landscape whose elevation represents the mid- to the high-alpine zone of Buddha Mountain? 4 • How do the richness patterns vary in functional groups, and between a quadrat-based field survey and an interpolation study? Paper II: Vascular and cryptogam richness in the world’s highest alpine zone, Tibet. • How does the elevational species richness pattern vary along TAR’s mid- to highelevation gradient? • How do the functional groups behave along this elevational richness pattern, and which climate variables may influence these patterns? Paper III: The elevation gradient of lichen species richness in Nepal. • How does lichen species richness vary along the elevational gradient in Nepal and do their functional groups follow this overall pattern? Paper IV: Temporal changes in species diversity and composition in abandoned fields in a trans-Himalayan landscape, Nepal. • What is the successional compositional pattern of species in different aged abandoned fields? • Do the changes in species richness and beta diversity match with the temporal gradient? Which local factors influence these successional patterns? I made Paper I is core of this dissertation under which research questions are generated and answered by other papers. These studies are based on the following study areas: Buddha Mountain, TAR- Paper I Buddha Mountain (30° 11’ N, 90° 29’ E) lies 100 km north-west of Lhasa, the capital of TAR. The altitude of this mountain ranges from 4985 to 5700 m above sea level. According to the Meteorological Bureau of Lhasa, Damzhung is the closest climatic station (about 80 km NE, and at 4200 m asl) from the study site and has a mean annual temperature of 1.5°C, mean summer temperature of 14°C, and the mean winter temperature is -7°C. The mean annual precipitation is 442 mm. The vegetation of the studied mountain slope below 5300 m 5 asl represents an alpine steppe type of the mid-alpine zone and nival or desert type above 5300 m asl (Chang 1981, Miehe 1988, Birks et al. 2007). Kobresia spp., Stipa koelzi, cushions of Androsace tapete and Arenaria bryophylla, Astragalus donianus, Oxytropis spp., Ranunculus lobatus, Lancea tibetica, and Thalictrum alpinum together with Saussurea spp. represent species common to alpine steppe vegetation. Sedum spp., Rhodiola spp., Androsace coronata, A. zambalensis, Arenaria gerzensis, Thylacospermum caespitosum, Draba glomerata, Gentiana urnula, Ranunculus involucratus, Saxifraga spp., Koenigia islandica, Anomobryum concinnatum and some saxicolous lichens such as Dimelaena oreina and Rhizoplaca peltata represent alpine type species. Tibet’s highest alpine zone- Paper II Elevation between 4900 to 6000 m asl of TAR is the study area of paper II. The Tibetan Plateau is the largest and highest and the youngest alpine plateau in the world (Chang 1981). It is surrounded by tall mountains with an average elevation between 4500 to 5000 m asl. Tibet Autonomous Region (TAR) or Xizang is the political and official designation by the government of China to represent part of the present Tibetan Plateau. TAR stretches between 78° 25’ to 99° 06’ E and 26° 50’ to 36° 53’ N with altitudinal range between 2700 to 8848 m asl. It covers an area of 1.23 million km2 (Birks et al. 2007). It runs parallel to the main Himalayan arc, over 2500 km long. Its eastern border is with other biodiversity hotspots, such as the Yunnan and Sichuan provinces of China, whereas its north-western border meets the Karakorum Mountain and the Takla Makan desert (Dickoré & Miehe 2002). The Tibetan Plateau was believed to be uplifted over the last 120 million years due to the collision between the Asian plate and continental fragments (Molnar 1989, La Duo 2008). Later, the Himalaya was believed to be uplifted after collision between the Asian and the Indian plate about 60 million years ago (Singh & Singh 1987). Thus, the Tibetan Plateau may have a history almost twice that age of the Himalaya. Both are considered as the youngest and the most fragile mountain areas in the world. Very tall mountains and the Plateau of TAR play a major role in the central Asian climate and its biota. The TAR has 50-60% less atmospheric pressure than sites at the sea level, lies closer to the troposphere than mountains elsewhere, causing great impacts on the 6 atmospheric circulation, heats up land masses more quickly, receives high solar radiation due to low interception by clouds, and receives the highest amount of UV-B radiation compared to mountains elsewhere (Chang 1981). These features made the Plateau a “Hot Island” (Birks et al. 2007). Life-forms adapted to this environment show an unique physiological and ecological adaptation. In addition, the Himalayan arch in the south-east acts as a strong barrier to the Indian monsoon. This creates a rain shadow in the TAR except in some narrow through-valleys in the south and east. The TAR receives an annual precipitation of 200-1000 mm during summer. It shows a spatial trend of decreasing precipitation from south-east to north-west. During the winter, cold and dry air blows out from central Asia due to the Siberian high pressure that creates winter rain towards the north-west. It also shows a spatial pattern of decreasing values towards the east. Lhasa, the capital of TAR which lies in the central part, receives 443 mm annual rainfall, and has an average annual temperature between -16.5°C and 7.7 °C to a 28.3°C extreme (Chang 1981). The present vegetation of the TAR clearly indicates two horizontal (east-west) and vertical (north-south) gradients (Chang 1981, Miehe 1988, Ni 2000, Birks et al. 2007, Miehe et al. 2007, Miehe et al. 2008). TAR has a sub-tropical montane forest to high-cold desert vegetation along the horizontal gradient and sub-tropical rain forest to nival zone along the vertical gradient. Birks et al. (2007) classified the south-central TAR’s vegetation above 3500 m asl into 9 horizontal and 4 vertical zones based on annual precipitation. There is variation in each mountain system as elsewhere. However, TAR seems to have a more complicated vegetation zonation than elsewhere (Chang & Gauch 1986, Schaller 1998, Wang et al. 2007). Nepal-Paper III Paper III covers the whole country of Nepal. Nepal (26˚22’ to 30˚27’ N and 80˚04’ to 88˚ 12’ E) lies in the central Himalayan zone. The average distance between east-west is 885 km and between north-south is between 145-248 km (Kansakar et al. 2004). Both eastern and western Himalayan elements are present here. The country has the highest mountain in the world, Mount Everest (8848 m asl) with TAR. The lowest elevation is at 60 m asl towards the south. The shape of the country looks like a tilted rectangle, bordered by the IndoGangetic plain on three sides and by TAR in the north. Eastern Nepal, the less tilted side on 7 the map, lies closer to the eastern Himalaya, the main entrance for the Indian Monsoon coming from the Bay of Bengal. The monsoon is the main source of precipitation that has a spatial pattern starting from eastern to western Nepal. It begins in May and ends in September. Winter precipitation is also available here but that spreads from western Nepal through westerly winds. It also shows a spatial pattern beginning from western Nepal and ends towards eastern Nepal. Some places in central Nepal, e.g. Lumle and Langtang receive the highest amounts of annual precipitation (above 6000 mm) (Kansakar et al. 2004). Some rain-shadow areas are also present in Nepal such as Upper Manang and Dolpo that lie behind the main mountain range. These areas hardly receive more than 400 mm annual precipitation and resemble the Tibetan Plateau. Weather conditions in Nepal vary from place to place. During summer, the southern part of Nepal which is the flat lowland (Terai) experiences maximum temperatures between 3540°C, while hilly regions have 22-27°C. Likewise, winter temperatures in the flat lowland ranges between 7-23°C, and the hills have around 12°C. There are places at high altitudes that are always covered by snow and have freezing temperature throughout the year. The elevation gradient is a very strong, sharp and important ecological gradient in Nepal. Within the shortest north-south transect of about 200 km distance one can pass from the tropical to the nival zone (Dobremez 1976). The elevation gradient in Nepal is also inseparable from land-use patterns. Nepal is an agricultural country where almost 90% of the population directly dependent on subsistence agriculture and livestock (Menon 2009). All agricultural practices depend directly on the monsoon and the westerlies in the hilly mountains regions except some arable areas in the flat inner valleys, and lowlands such as Terai where irrigation is a common practice. Nepal has been divided horizontally into five physiognomic zones: Terai, Siwalik, Mahabharat, Greater-Himalaya and Trans-Himalaya (Hagen 1969, Upreti 1999). Each zone represents a definite elevation range. Similarly, the vegetation of Nepal has been classified into 8 elevational zones: Tropical, Sub-tropical, Lower temperate, Temperate, Sub-alpine, Open low-alpine, Mid-alpine and Nival zones (Yoda 1967, Dobremez & Jest 1969, Stainton 1972, Dobremez 1976, Miehe 1982, 1989, Stainton 2001). 8 Manang-Paper IV Manang (28°40´N to 84°01´E) is one of the 75 districts of Nepal that lies in the north-central part. The peculiarity for this district is its location behind the Annapurna massif of above 7000 m asl elevation. This Himalayan range creates a rain shadow in this district. Thus Manang falls in the trans-Himalayan zone. It receives annual rainfall of about 400 mm, the temperatures recorded at Jomsom, 30 km south of this U-shaped valley are between -1.8 and 7.9°C during winter versus 14.2 and 23°C in summer (Anonymous 1999). The monsoon begins from the south-east and decreases to the north-west. The Marshyangdi River passes all the way through this district and forms an U-valley in upper Manang and a v-shaped valley towards lower Manang. My study area lies in upper Manang. This part of Manang has two distinct slopes. Part of the landscape towards the Annapurna massif is the north-facing slope with forests of Pinus wallichiana, Abies spectabilis, and Betula utilis. On the drier south-facing side there are settlement areas, some agricultural fields and bushes of Juniperus spp. The main agricultural fields are located on either side of the Marshyangdi River. Agricultural fields are also located near villages as well as closer to the forest far from villages. However, many agricultural fields have been abandoned. Pisang has about 150 ha agricultural land out of which 75% is abandoned. Similarly, Bhraka has about 185 ha out of which 60% is abandoned (Paper IV). Villagers left their fields in different years due to many reasons. I sampled the abandoned fields belonging to two villages, Pisang and Bhraka. Pisang is located at a lower elevation than Bhraka and approximately 12 km away. Elevational difference among all the sampled abandoned fields in Pisang is 300 m asl (3175 to 3475 m asl) and 50 m asl (3400 to 3450 m asl) in Bhraka. A total of 42 abandoned fields are available for this study (paper IV). 9 Terminology Diversity studies use terms that need to be well defined. The concept of species richness is commonly used for the number of species in a sampling unit, and species diversity to denote the species richness and evenness at local scales (Whittaker et al. 2001). Some ecologists apply the terminology ‘species density’ to represent the number of species sampled in each standardized sample unit, e.g. per unit area (Whittaker 1975, Carpenter 2005) while others retain the terms ‘diversity’ or ‘richness’ rather than ‘density’ (O' Brien 1993, Grytnes 2002) for such applications. Different indices of biodiversity such as Shannon-Weaver’s H, Simpson’s Ȝ, etc. are also commonly used. Such indices can be confusing and may not be easy to understand (Peet 1974, Huston 1994). This dissertation uses the term ‘species richness’, because it is self-explanatory and has comparability with other studies. Species richness is the fundamental measure of community and regional diversity (Magurran 1988, Gotelli & Colwell 2001). Thus the species richness as used in this thesis is the total number of species present within the six 1×1 m2 quadrats per 25 m elevation band (Paper I), the total number of species within each 100 m elevation band (Papers II and III), and the total number of species within each 1×1 m2 plot (Paper IV). The smallest sampling units of quadrats and plots are considered synonymous. This thesis focuses on species richness of plants and lichens in Papers I and II, lichens in Paper III and only vascular plants in Paper IV. Species richness also applies to specific functional group richness such as forb, graminoid, cushion, shrub, crustose lichen, foliose lichen, fruticose lichen, green algal lichen, cyanobacterial lichen, saxicolous lichen, terricolous lichen, corticolous lichen etc. Scale The concepts of scale and space in biodiversity studies are highly debated among ecologists. The importance of scale was first realized by R. H. Whittaker (1960) when he documented the biodiversity of Siskiyou Mountain, Oregon, USA. He coined the Greek lettering scheme of Į, ȕ, Ȗ, İ, į, etc. to represent the diversity patterns at different scales. Among them, the first three are widely used. According to Whittaker (1960), Į diversity represents the diversity in an individual stand or sampled unit. His Siskiyou Mountain diversity represented the Ȗ diversity, and the change in species composition along stands or gradient was 10 designated ȕ diversity. The lack of lower and upper limits to this landscape scale of diversity created controversies and confusions among ecologists. Later, Whittaker (1977) introduced a further classification regarding the measurement scale of biodiversity study. The importance and awareness of scale in ecological studies have increased in recent years (e.g., Rahbek 1995, Rahbek 1997, Whittaker et al. 2001, Willis & Whittaker 2002, Rahbek 2005, Willis & Bhagwat 2009), although recognized already by Allen & Starr (1982). To minimize controversies among ecologists regarding scales, Whittaker et al. (2001) proposed the three following intuitive terms: local-scales, landscape scales and macro-scales in their hierarchical theory of biodiversity. Applying the scale concept to this thesis, the species richness inside each six 1× 1 m2 quadrats per 25 m elevation band in Paper I represents Į diversity and belongs to local-scale diversity. The turnover of species between each two elevation bands represents ȕ diversity. It was estimated through the DCA- I axis sample score as a function of elevation, which is a valid method (Lepš & Šmilauer 2003). The whole mountain range of Buddha Mountain represents the landscape scale (Ȗ = 143 species). Similarly, species richness inside each 1×1 m2 plot in Paper IV represents the local-scale or Į diversity, and the turnover between two plots represents ȕ diversity. Here ȕ diversity was estimated through the length of the gradient of the DCA- I axis. If I follow the hierarchical theory of Willis & Whittaker (2002), Paper IV represents the landscape spatial scale with Ȗ = 136 species and the local temporal scale change between 1 to 50 years. Papers II and III belong to a macro-scale study with Ȗ = 642 and 525, respectively. Scale has three aspects in ecology: spatial, temporal and organizational complexity (Levin 1981). Likewise, spatial scale also has three components: sample size (grain or focus or individual size of quadrat or plot), extent and intensity (Wiens 1989, Palmer 1994, Whittaker et al. 2001). The sample size defines the lowermost limit of study that directly correlates with the body size of the organism studied (Brown et al. 2004). Sample extent defines the maximum area (uppermost limits) of study. Sample intensity defines the number of sampling units in the study. Thus in combination these three spatial components quantify the variation in the studied system. Systems are highly hierarchical in nature, dynamics at lower scales affect the dynamics of a system at higher levels and vice versa (Allen & Starr 1982, O'Neill 1989). Thus there is a strong link between grain size and extent of study area. Smaller grain 11 size is used to detect variation at local scale and larger grain size is used to detect larger (regional) variation. The six 1×1 m2 quadrats (Paper I), and 1×1 m2 plots (Paper IV) represent the sampling size or focus for my study. The area in each 100 m elevational band between 4900 to 6000 m of the TAR (Paper II), and 100 m between 200 to 7400 m of Nepal (Paper III) do not represent the actual sampling size. It covers a large geographical area. It is impracticable to sample both TAR and the Nepalese Himalayan gradient completely. Thus, the number of species at each 100 m elevational band was counted for each study. Paper I covers a 2.3 km transect. Paper II studies an average north-south distance of about 950 km and an average east-west distance of about 2150 km comprising an elevational range between 4900 to 6000 m. Paper III deals with an average east-west distance of about 885 km, and an average north-south distance of about 250 km to cover the elevational range of the whole of Nepal between 200 to 7400 m. Paper IV studies approximately a 1 km long transect located in abandoned fields of each village. With respect to sampling intensity, Paper I is based on 174 quadrats per ca. 200 km2. However, the resolution has been reduced to n = 29 to avoid pseudoreplication. Paper II is based on 12 elevation bands between 4900 to 6000 m elevations in TAR of approximately 612 000 km2 area. Paper III is based on 73 elevation bands between 200 to 7400 m elevation in Nepal comprising 140 000 km2. Paper IV is based on 242 plots in ca. 2.2 km2 abandoned fields. Temporal scale cannot always be separated from those of spatial scale. Willis and Whittaker (2002) have classified the ecological temporal scale into three levels confounded with the spatial scale in the hierarchical theory of biodiversity. According to them, the successional change in species composition and richness patterns recorded between 1 to 100 years can only be seen at the local scale, and is defined as the local temporal scale. If the dynamics occur between 100 to 1000 years, this will fall under a landscape temporal scale. Finally, pattern occurring in more than 10,000 years is defined as a regional temporal scale. Most of the community dynamics research falls under the local temporal scale regardless of study area size. Studies of chronosequences formed after abandoned agriculture, fires, volcano eruptions, sand-dune formation, deglaciated moraines or long term research studies are examples of the local temporal scale. Palaeoecological inventories comprising a time scale of 12 10,000 years are an example of regional temporal scale. Ecological research based on temporal scales between 100 to 1000 years is an example of the landscape temporal scale. Paper IV represents purely the local scale temporal change in species composition and richness patterns in between 1 year to 50 years in old abandoned fields. As a short temporal gradient it is hard to separate it from the spatial scale. Thus the first axis sample scores were extracted out through an age-constrained Canonical Correspondence Analysis (CCA) and utilized as a predictor variable as was done by Lepš et al. (2001) and Bartolome et al. (2004). Other papers utilized temporal scales indirectly. Organizational complexity scale may focus on variations within- and/or between populations. Experimental manipulations are needed to test hypotheses related to species diversity. All papers in this thesis are descriptive, dealing with spatial and temporal variation among communities. Papers I and IV are based on observational studies and Papers II and III are based on interpolation approaches. It was not possible to do any experimental manipulations in my thesis research. Some authors such as Shmida & Wilson (1985) have suggested the minimum geographical area to measure biodiversity pattern. They suggested 102 to 103 m2 area to measure terrestrial plants for Į diversity, and 106-108 m2 for Ȗ diversity. Mittelbach et al. (2001) defined the scale of study based on geographical extent and ecological association. They recognized four geographical scales: local scales (0-2×104 m), landscape scale (2×104 -2×105 m), regional scale (2×105 -4×106 m) and continental to global (> 4×106 m). If I look at geographical distance for my samplings, Papers I and IV both cover at least 10100 m distance between each sampling unit. This comes far below the lowest limit as set by Shmida & Wilson (1985). May be this limit applies more towards other ecological communities than sub-alpine or alpine areas which I have been studying. The geographical distance for Papers II and III are rather complicated to know as they are interpolation studies. Papers I (ca. 2×105 m2) comes under the landscape geographical scale, both Papers II and III fall within the continental to global scales but paper IV (22×102 m2) represents the local scale of Mittelbach et al. (2001). 13 Area The space/geography (Area) occupied by species has been a controversial issue in diversity research. It took almost 35 years after the first publication of the Equilibrium Theory of Island Biogeography by MacArthur & Wilson (1967) to reach the general appreciation about the scale dependency of area (Palmer & White 1994, Rosenzweig 1995, Gaston 2000, Whittaker et al. 2001, Lomolino et al. 2006). Data dealing with species richness can only be compared between areas similar in size (Clinebell et al. 1995). This view was clearly highlighted by the differences in patterns between Papers I and II. Models at two different scales of areas of measurement (landscape in Paper I and macro-scale in Paper II) of species richness are used. Difference in patterns was thus directly connected to the scale of area. Paper III allowed a comparison of patterns shown by species richness along the same Nepalese Himalayan gradient. Based on earlier findings, I believe that the unimodal richness patterns for both total and endemic lichen richness documented in Paper III are caused by factors such as habitat diversity and environmental suitability rather than elevational band area. An area effect is also considered insignificant for the higher species diversity at the equator (Terborgh 1973, Rosenzweig 1995, Chown & Gaston 2000). All these authors believed that higher diversity at the equator is due to the larger continuous area with a similar ecoclimate that result in lower probabilities of extinction and higher probabilities of allopatric speciation. Variables at the local-scale Environmental heterogeneity (Palmer 1990), disturbance (Connell & Slatyer 1977, Connell 1978, Huston 1979), fire (Auclair et al. 1976), and species pool (Zobel 1997) are local-scale dependent variables. Variations, tests and predictions in species-richness studies are usually done on local-scale data. Existing hypotheses are tested and theories are developed based on local-scale ecological studies. Plant community dynamics and succession are visible phenomena at a local scale. In addition, changes in biomass and vegetation cover, as well as species richness changes after experimental treatments can also be striking processes. The long-term research at the Park Grass Experimental Station of Rothamsted, England is a good example of a local-scale 14 study. This experimental station, established in 1856 for other purposes than succession and competition studies, has provided a unique data source to resolve debates within local-scale plant ecology (Silvertown et al. 2006). The Resource Ratio Hypothesis of Tilman & Downing (1994) was developed based on this Long Term Research experiment (LTR). Successional changes that were well experimented in the 150 years LTR are still relevant in modern discussions dealing with stability and competition (Silvertown et al. 2006). The first successional study started in the sand-dunes of Lake Michigan (Cowles 1899) which led Fredric E. Clements to develop his holistic view of community development and succession (Clements 1928) that was widely accepted at the time. Successional studies continued in old abandoned fields of the Piedmont Mountain of North Carolina (Oosting 1942) and the Sandhills of South Carolina (Odum 1960). Secondary succession studies resulted in the Initial Floristic Composition (IFC) and Relay Floristic Composition (RFC) models and mechanisms of succession (Horn 1974) and the facilitation, inhibition and tolerance mechanisms of succession (Connell & Slatyer 1977), and tested models of energy flow, biomass production, dominance and competitive replacement mechanisms in the succession from bare-ground in local-scale studies. Odum (1969, 1971) proposed models of succession relevant for different scales of space and time that varied among researchers. In general, a local-scale study, like the one used in Paper IV, is a useful approach to understand ecological processes at the lowest resolution. Variables such as changes in soil pH, bare ground, vegetation cover, and soil moisture index were linked with short-term temporal changes in species richness and composition. Using species cover and their richness in the regression models can be problematic because of circular reasoning (Palmer 1993), thus they do not relate to the observed effect of species richness but they illustrate changes in the field with time. Paper I utilizes change in soil-nutrient related variables such as N, C, and pH from two quadrats at each elevation band. Variables at the landscape- to macro-scales After the foundation of macro-ecological investigation by Brown (1995), new insights resolved questions asked over 200 years ago concerning the general theory to explain latitudinal biogeographical patterns. Present research focusing on elevational gradients may partly apply to latitudinal gradients (Brown 2001). Latitudinal and altitudinal gradients are two macro-scale species richness patterns. Both have been studied in several groups of living 15 plants and animals as well as in fossils (Rohde 1992, Körner 2000, Ohlemuller & Wilson 2000, Lomolino 2001, Harris 2006, Lomolino et al. 2006, Grau et al. 2007, Mittelbach et al. 2007, Birks & Birks 2008, La Duo 2008, Kessler 2009, Miehe et al. 2009, Willis et al. 2009). Elevation and latitude per se have no direct influence on species richness but act through environmental factors correlated with these variables (Grytnes 2002). The controlling factors towards each macro-scale variable (latitude and altitude), are to some extent, similar and yet dissimilar. Macro-scale patterns are mainly controlled by climatic variables (Whittaker et al. 2001) shaping the distribution of each biome (Woodward 1987). Climate maintains as well as generates biodiversity patterns at macro-scales (Wright 1983, O' Brien 1993, Hawkins et al. 2003). Various climatic factors such as temperature, rainfall, moistures, day length, seasonal variation, solar radiations, UV-B radiation, atmospheric pressure, and air humidity influence species composition and species richness patterns at macro-scales (O' Brien 1993, Brown 1995, Odland & Birks 1999, Grytnes 2002, Bhattarai 2003, Birks et al. 2007, Grau et al. 2007, Körner 2007, Birks & Birks 2008). Papers I, II and III try to link elevational macro-scale climate variables to species composition (Paper I) and species richness patterns. Paper IV also shows an indirect effect of climate change. Agricultural fields in the trans-Himalayan sub-alpine zone were abandoned as a result of declining production of crops, and unpredictable seasonal changes faced by local people. Due to the absence of measurements of such climatic factors in this thesis, I predicted climatic factors from similar studies done in nearby regions. For example, the midto high-alpine zone in the Tibet Autonomous Region (TAR) is geographically defined as a rain- shadow region in High Asia. High mountains receive more solar and UV-radiation than mountains elsewhere. Thus patterns derived after such climatic filters may have a high influence. 16 Variables and mechanisms influencing spatiotemporal richness patterns Sampling artefacts Macro-scale studies often collect data from a large grain-size in area. Thus field surveys are hardily feasible. It is a common trend to depend on secondary sources such as biological collections in museums, herbaria, published accounts in floras or faunal monographs, and distribution maps. Indeed, the importance of such information is growing in the present world’s growing concern about the conservation of biodiversity (Sun ZhenHua et al. 2007). Interpolation of species ranges from their observed extremes of the studied gradient is common for both latitudinal and altitudinal patterns (Rahbek 1995, 1997, Patterson et al. 1998, Odland & Birks 1999, Heaney 2001, Md. Nor 2001, Grytnes & Vetaas 2002, Vetaas & Grytnes 2002, Hawkins et al. 2003, Koleff et al. 2003, Bhattarai et al. 2004, Grytnes & Beaman 2006, Grau et al. 2007, Sun ZhenHua et al. 2007). Distribution maps for species are good sources of information for a macro-ecological study, but a detailed investigation for each or at least a target species prior to the completion of a distribution map is required. Common assumptions such as the presence of a species at all elevational bands or cells between the observed extremes and absence of occurrences of the species beyond the observed ranges are made in all interpolation studies. However, these assumptions are easily violated (Birks 1993, Whittaker et al. 2001). There are disjunct distributions of species, endemic species, rare and other threats to biodiversity. Species can be absent from a certain area due to different causes than interpolation and can also be present outside the observed ranges. Papers I, II and III rely on species richness patterns from some of the least studied areas such as the Tibetan Autonomous Region (TAR) and the Nepal Himalaya. Moreover, lichens represent one of the least studied groups in biodiversity analyses. This thesis shares sampling artefacts to other similar studies. Furthermore, the thesis shares the view of Vetaas & Grytnes (2002) that the area of altitudinal bands does not account for significant variation in speciesrichness patterns for the Nepalese Himalaya. Epiphytic lichen species included in Paper III are dependent on vascular species for their host and on the special micro-environments 17 created by them. The dispersal barrier caused by certain landscape elements has been proposed in the interpolation of species range data towards their observed extremes (Colwell & Lees 2000). The generated null-models define the mid-domain effect based on the occurrence of a peak in species richness in the intermediate part of the studied gradient or domain (Whittaker et al. 2001). The argument arises from the consequences of placing datasets of varying species ranges randomly within a bounded domain. Part of the studies in Papers I and II are fully located towards the higher end of the elevation gradient i.e., the midto high-alpine zone of TAR. Thus we are more concerned about the spatio-temporal dynamics of species-richness patterns. We anticipated and acknowledge potential artifacts after interpolation. Mass effects Mass effects are important phenomena in the establishment of a species from its original distribution to a new area. This process has been given various terms such as the rescue effect (Brown & Kodric-Brown 1977, Stevens 1992), mass effect (Shmida & Wilson 1985) and the source and sink effect (Pulliam 1988). These studies propose that the mass effect has an effect on the shape of the elevational richness pattern. Kessler (2000b) found an up-slope mass effect for palm species elevational richness patterns. He also stated that high elevational species would show a down-slope mass effect and that mid-elevational species would respond to both sides (Kessler 2000b, Grytnes 2002). In an empirical study, increased species richness at ecotones may be the cause by mass effect (Shmida & Wilson 1985). My thesis focuses mainly on the mid-to high-alpine zone (Papers I and II) in the high Himalayan Mountain area where a sharp elevation gradient will result in other ecological changes (Körner 2007). A unimodal pattern of lichen species richness could have a mass effect from both sides of the elevational gradients (Paper III). Species-pool The species-pool hypothesis (Eriksson 1993, Zobel 1997) states that the number of species in a certain habitat is dependent upon the commonness of this habitat elsewhere. The speciespool hypothesis is connected more with historical and evolutionary explanations of species 18 diversity than with ecological species-richness patterns (Bhattarai 2003). The environment, size of area, and geology can all determine this relationship. The observed elevational decline in species richness with increasing altitude discussed in Papers I and II may be due to a decline in the regional species-pool of biodiversity in hotspots such as the Himalaya and the Sichuan and Yunnan provinces of China. This phenomenon may be also true for the unimodal richness pattern for lichens in Nepal (Paper III), but the elevation of maximum richness is substantial higher than the maximum richness of vascular plants and other plant types. Thus I believe that local phenomena may be more important for lichens than the species-pool as such. In Paper IV the richness towards the oldest abandoned field is apparently enriched by the species-pool from adjacent forests. Thus the oldest abandoned fields were more likely to be within surrounding vegetation that act locally as a species-pool. Historical factors and local-scale community successional patterns Changes in species composition and in species-richness patterns in different chronosequences are influenced by historical factors and local-scale successions. Although a chronosequence study is a good source of understanding ecological succession, it has the assumption of the repetition of the youngest chronosequence via the old stages (Pickett 1989). The importance of land-use history prior to abandonment in secondary succession has been widely discussed (Odum 1960, 1969, 1971, Drury & Nisbet 1973, Horn 1974, Pickett 1976, Huston & Smith 1987, Pickett 1989, Glenn-Lewin et al. 1992, Pausas 1994, Bazzaz 2005, Pausas et al. 2006, Krebs 2008). Different colonization processes, species richness patterns, hypotheses and theories have been proposed and well established through succession studies. Historical factors often has less probability of re-occurrence and that would be eliminated by macro-scale or dynamic factors related to climate which often have a high probability to re-occur (Whittaker et al. 2001). Paper IV shows that there is a general pattern among colonizing species, unimodal successional species richness patterns and divergence patterns of succession through time. General patterns are explained by scales such as differences in time since abandonment, and specific patterns are explained by differences in local factors such as soil pH, moisture, relative radiation index, and closeness to species-pool (forest). A decreasing trend in soil pH and vegetation cover with time differed from the generally expected trend, but agreed with 19 results from some other studies (Bonet & Pausas 2004, Arbelo et al. 2006) in which old fields were grazed. Grazing may create an intermediate intensity of disturbance through time and give unimodal temporal species richness patterns (Grime 1977). Unimodal species richness patterns in succession with maximum modeled richness between 15 to 25 years is also common in other studies (Horn 1974, Brown & Southwood 1987, Bazzaz 2005), although other responses have also been reported (Carson & Barrett 1988, Silvertown et al. 2006). A high richness in a mid-successional phase may occur when both pioneer (lightdemanding) species and species of mature phase (shade-tolerant) co-exist (Bazzaz 2005). This study shows a decreasing pattern in total richness and vegetation cover towards the oldest temporal gradient with increasing ȕ diversity. The divergence of patterns of succession may be due to surrounding vegetation composition (source-pool), a high grazing pressure interaction with species, and spatial heterogeneity. Decreasing diversity during intensive grazing was observed in Papers I and II. The study sites of the two papers differ in scale and intensity of grazing. Increasing ȕ diversity with decreasing total richness would be a facilitation effect of succession (Connell & Slatyer 1977) through more suitable (heterogeneous) habitats as for example, the nursery effect provided by thorny bushes (Pausas 1994, Pausas et al. 2006). Climate and energy-related variables to explain landscape and macroscale elevational patterns Climate is a main driving factor in influencing landscapes and macro-scale patterns of diversity (Whittaker et al. 2001). The general elevational decline in species richness for both plants and lichens at the quadrat scale for Buddha Mountain (Paper I), interpolated species ranges from the mid- to the high-alpine zone of the Tibet Autonomous Region (Paper II) and the unimodal elevational pattern for Nepalese Himalayan lichens (Paper III) may be caused by climatic variables. However, these general patterns do not apply to all functional and taxonomic subsets such as unimodal patterns for cushions, cryptogams and lichens in Buddha Mountain (Paper I), and unimodal relationship for lichens (Paper II). The shape of the decline relationship found in Paper I was quantified by the type of curve fit between the DCA-I axis sample scores and elevation. I argue that that declining patterns could have been caused by climatic factors as well as by extreme physical and environmental features at high 20 elevations in this geologically young mountain. Specifically, the declining pattern for the total richness with elevation in Buddha Mountain (Paper I) was hidden due to separate but contrasting patterns of individual functional groups. The shrubs start to decline at the lowest elevation, followed upwards by almost similar declining trends of vascular plants, forbs and graminoids. Cushions declined at higher elevations followed by bryophytes and finally lichens at the highest elevation. Similarly, this declining pattern for the total richness (Paper II) followed by similar elevational patterns of functional groups (non-graminoids, graminoids, Asteraceae, total moss, pleurocarpous moss and acrocarpous moss) except for a unimodal relationship found in lichens. The decline is rather gentle compared to other interpolation studies. Differences among adaptational features for each functional group against harsh and unstable environmental factors would influence the response type to the environment (Agakhanyantz & Breckle 1995, Klimes 2003, Körner 2003). Such filters include large differences in ambient temperature, precipitation, cloud formation, short growing season, differences in landscape geometry, high solar and UV-B radiation, etc. An almost similar declining pattern for most species and their functional and taxonomic group except some such as lichens points to climate as the main factor. The initial plateau followed by the sharp decline in the elevational richness pattern in Buddha Mountain (Paper I) differed from the smoothly declining elevational pattern in interpolated species ranges of the flora-based elevational species richness pattern (Paper II) for various causes. The scale-dependency of species along the elevation gradient may be one cause. Environmental heterogeneity, which is minimized in the study of Paper I may vary as a function of elevation (Palmer 2006) and thus accentuate differences from interpolation studies. Fine-scale environmental variation in factors such as soil pH, N, C, moisture, atmospheric humidity, disturbance etc., influence the success of species which may be less important at broader floristic scales (Whittaker et al. 2001). Thus a smoothly declining elevational species richness pattern would have been expected. A general question can be about why there is not a plateau in richness after increasing endemism. Local endemism may increase at higher elevations which may inflate the richness for regions covering many mountains (Paper III). Specifically, in some quadrats at specific elevation bands e.g. 5500 m (Paper I) I found higher species richness for lichens than in the interpolated species ranges (Paper II). That implies that lichens may have been insufficiently studied in TAR. 21 Mechanisms underlying the unimodal richness pattern of lichens of Nepal and TAR A uniqueness in the elevational biodiversity pattern for lichens is confirmed after both interpolation and direct field observation along the elevation gradient in Nepal and TAR (Papers I, II and III). The highly significant unimodal richness pattern for the total lichen richness in Nepal with the highest richness between 3100-3400 m represents the highest elevation maximum compared to that of major taxonomic groups of plants. This general pattern is also evident for the endemic lichen species richness of Nepal with maximum richness between 4000-4100 m that represented the sub-alpine zone in Nepal. Maximum endemic lichen richness lies within the range of the highest endemic richness of vascular plants in Nepal (Vetaas & Grytnes 2002). A unimodal richness pattern with elevation is also commonly found among lichen life-form richness, lichen photobionts types and the main lichen substratum types. Area accounted less for unimodal vascular richness pattern in Nepal (Vetaas & Grytnes 2002). This may also apply to lichen richness. Hot, humid and closed canopies with disturbed forest patches at lower elevation are also less congenial for lichens. The elevation zone of the highest lichen diversity is characterized by maximum variation in rainfall such as annual rainfall (> 4000 mm) in the rainy south side and lowest (< 500 mm) towards the northern, rain-shadow side, with mean summer temperature between 14-17°C in Nepal (Bhattarai et al. 2004). This elevation zone is represented by mixed, broad-leaved and coniferous forests with varied roughness and pH of barks. The canopies formed by these forests may give extreme local variation in the availability of water that facilitates lichen growth. These entire features may support the highest richness for lichens in general, and particularly for the more light-demanding, canopy-inhabiting fruticose form of green lichens. The lower canopy and forest floor receives less light and is suited to shade-tolerant foliose species and flat forms which may capture light from one direction only (Gauslaa et al. 2009). The lower temperate forest between 2400-2500 has the highest number of foliose lichens and open canopies in the sub-alpine zone appear more suitable for crustose species. The high endemic lichen richness in Nepal at high elevations is presumably associated with the history of glaciations that causes phytogeographic isolation and further facilitated by reduced competition after periods of harsh and unstable climatic conditions, with rapid 22 freezing and drying water-cycles. The latter interpretation may also apply for the unimodal richness pattern of lichens in the mid-to high-alpine zones of TAR. High solar and UV-B radiation, high fluctuations in ambient temperature, rapid shifting between moist and dry conditions and frequent cloud formation may be more beneficial to lichens than to other photosynthetically active groups of organisms. Stress, stability, disturbance and water-energy dynamics Considering the general diversity hypotheses formulated to explain macro-scale latitudinal and elevational species richness patterns (Fraser & Currie 1996, Whittaker et al. 2001), ecological factors like available energy (water, solar), disturbance, environmental stress and stability may shape declining elevational diversity patterns for total species richness as well as specific patterns in their functional and taxonomic subsets (Papers I and II) from the midto high-alpine zone of TAR, as well as the unimodal richness patterns for lichens and lichen functional groups (Paper III) from Nepal. One can easily expect a declining elevational pattern due to decreasing temperature and precipitation with increasing elevation. These factors are expected to be strongly influential for the TAR transect, as it is located in the high elevation rain-shadow side, unlike the rainy side of the Himalaya where the Nepalese elevational gradient is studied. Low precipitation is believed to be a major limiting factor for the biodiversity of Tibet (Chang 1981, Birks et al. 2007). Both temperature and precipitation are presumably more influencing factors in Tibet (Chang & Gauch 1986, Ni 2000, Wang et al. 2006) than for the Nepalese Himalayan elevation gradient. In addition to energy, other local factors may facilitate colonization of non-woody species in such harsh and unstable environments. Such hypotheses need to be tested in designed and controlled sets of experiments. Finally, one must not forget the young geological history of the Tibetan Plateau and the Nepalese Himalayas. Thus hypotheses related to disturbance (Huston 1994, Miehe et al. 2009), stress and stability (Begon et al. 1990, Callaway 2002), water-energy dynamics (O' Brien 1993), and the local effect of macro-scale climate change (Birks et al. 2007, La Duo 2008) may be plausible explanations behind this declining pattern. The maximum richness for lichens at 5500 m in Buddha Mountain, 5000-5500 m in the TAR, and 3100-3400 m in Nepal represent the highest elevational maxima among all biological groups studied. The capacity for lichens to survive at both simulated (de Vera et al. 2004) and real (Sancho et al. 2007) outer-space conditions documents an extreme 23 tolerance. A linearly increasing elevational richness pattern for lichens (Körner 2003, Grytnes et al. 2006), unimodal for lichens and linear for bryophytes (Bruun et al. 2006), and a higher and more linear richness pattern for both mosses and lichens than for other vascular plants were also reported from European Alps (Theurillat et al. 2003, Virtanen et al. 2003). Kessler (2000a) from the South America has reported a unimodal richness pattern for lichens. Since my study is confined to the mid-high alpine zones of TAR, lichens represented the main component of cryptogams on Buddha Mountain. This elevation zone may have high stress due to frequent and severe frost, fluctuating ambient and ground temperature, strong winds, snow, poor soils, periods of very low humidity, strong UV-B and solar radiation, short growing season, etc. These factors likely strongly impact and limit the distribution of vascular plants, but are not necessarily destructive to lichens. Cushion plants (e.g., Androsace, Thylacospermum spp.) exhibit more tolerance than other forms of vascular plants. Possibly this modified growth-form helps them in this extreme environment. Limited study in this part of the world limits our understanding. The highly heterogeneous microhabitats of the Tibetan Mountains are poorly known in the science of biodiversity. The declining pattern of total species diversity in the TAR would be explained first by decreasing area with elevation. Also, this alpine zone receives strong solar radiation that warms the landscape and rapidly evaporates water (Ni 2000). Some high-alpine plants have developed special features such as silky hairs such as some Saussurea spp. to avoid overheating and decrease transpiration (Gauslaa 1984), or form cuticles with high transpiration resistances and high reflection of short-wave radiation (Yang et al. 2008). Other adaptational features such as succulent leaves for crassulacean acid metabolism (CAM) photosynthesis (Sedum and Rhodiola spp.) are common in cold deserts (Körner 2003). Particularly high doses of UV-B radiation occur in all high mountains at low latitudes (Willis et al. 2009), which lichens can screen due to their high concentration of melanin, and/or parietin pigments in their outer cortex (Gauslaa & Ustvedt 2003). Cloud formation, strong wind, and high moisture content of air are also common at the mountain summits. Lack of multiple layers in the vertical stratification in the alpine vegetation creates an amount of convective long-wave radiation during night which cools air far below freezing temperature (Körner 2003). Thus there will be a strong selection among species that causes a decline in species richness with increasing elevation in TAR (Paper II). 24 Conclusions From my four papers and review of the pertinent literature, I can conclude that species richness patterns are real and highly scale-dependent at all spatial and temporal scales. Scales at all hierarchical levels are inter-related. Some dominant variables important at local-scales such as soil-related heterogeneity will be averaged out at the macro-scale. Climate-related variables will be highly influential at the macro-scale but will be also influential at the localscale. The species richness pattern depends on the studied group of species, its functional group, life-form, and their geography. Ecological significance of species differs by elevation. For example, in Nepal an elevation of 500 m has been shown suitable for the large wild animals, likewise 1000 m for vascular plants and 3100 to 3400 m for lichens. This highlights an important issue in conservation. The inclusion of the horizontal elevational zone concept in a conservation strategy will avoid species of importance compared to the vertical zonation concept in designing the conservation area. No ecological study is free from errors, which is true for this spatio-temporal study as well. The wisest thing is to evaluate such errors and artefacts and to improve in later studies. One must be vigilant in avoiding pitfalls of earlier works. The work which I have done has been rather challenging. The study of biodiversity study in the Himalaya, especially in Nepal, is rather young compared to Europe, America and elsewhere. It is the same in TAR, with its complex geography, due to many logistical and political reasons. My field-survey data supplement interpolated data. They support theories and hypotheses of species diversity and will help to build new hypotheses. I feel that each of my studies has its own importance for a particular scale of space and time. There is the choice of variables to measure and to quantify the underlying patterns. Behind every ecological pattern there is a great role of nature and its history which we ecologists or biogeographers must simply accept and acknowledge. I highlight the conclusions of my thesis as follows: • Lichens are represented as one of the most stress-tolerant groups of species in the Buddha Mountain, Tibetan Autonomous Region, and the Nepalese Himalaya. A unimodal relationship-elevation relationship is common among these areas. Maximum richness for Nepalese endemic lichens coincides with the maximum 25 elevation range for Nepalese vascular plant endemics. But total lichen species richness in Nepal is at the highest elevation among all studied plant groups. • Landscape-level elevational species richness patterns show an early plateau and a sharp decline afterwards that differed with no early plateau but a smooth decline afterwards at the macro-scale. That may clearly indicate the difference in scale of measurement of species, averaging out of environmental heterogeneities and the influence of climatic variables increasing with increasing scales. • The general pattern for total richness is hidden amongst many specific patterns for their functional, life-form groups. • An increasing trend in ȕ diversity with decreasing species richness and with increasing age of abandoned fields in the trans-Himalaya suggests high grazing pressure and grazing-tolerant species may act as nurseries to protect seedlings for the future generation of trees on this landscape. Future plan Field-based studies in a particular elevational band which has high richness and local-scale studies at a particular mountain will help to supplement this study in the Nepalese Himalaya. Lichens are a highly sensitive form of life. Further studies on their species richness patterns in different environmental gradients will help to understand if there is a local impact of macro-scale climate change. More studies in Buddha Mountain as well as in the mid- to high-alpine elevation zone of TAR will help to understand changes in biodiversity in the face of future climate change. Our present knowledge of elevational and successional patterns in TAR and Nepal will help theoretical and empirical understanding of the science of biodiversity. 26 References Agakhanyantz, O.E. & Breckle, S.W. (1995) Origin and evolution of the mountain flora in middle Asia and neighbouring mountain regions. In: Chapin III, F.S. & Körner, C.(eds) Arctic and alpine biodiversity: patterns, causes and ecosystem consequences. Ecological studies. Springer, Berlin, Heidelberg, New York. Allen, T.F.H. & Starr, T.B. (1982) Hierarchy: prospectives for ecological complexity. University of Chicago Press, Chicago. Anonymous (1999) Climatological records of Nepal 1995-1996. Department of Hydrology and Metrology, Kathmandu, Nepal. Arbelo, C.D., Rodriguez-Rodriguez, A., Guerra, J.A., Mora, J.L., Notario, J.S. & Fuentes, F. (2006) Soil degradation processes and plant colonization in abandoned terraced fields overlying pumice tuffs. Land Degradation & Development, 17, 571-588. Auclair, A.N.D., Bouchard, A. & Pajaczkowski, J. (1976) Productivity relations in a Carexdominated ecosystem. Oecologia, 26, 9-31. Bartolome, J.W., Fehmi, J.S., Jackson, R.D. & Allen-Diaz, B. (2004) Response of a native perennial grass stand to disturbance in California's Coast Range Grassland. Restoration Ecology, 12, 279-289. Bazzaz, F.A. (2005) Plants in changing environments: linking physiological, population, and community ecology. (2nd ed.). Cambridge University Press, Cambridge. Begon, M., Harper, J.L. & Townsend, C.R. (1990) Ecology: individuals, populations and communities. (2nd ed.). Blackwell Scientific Publication, London. Bhattarai, K.R. (2003) Variation in species richness at the fine-scale and at the broad-scale in the Himalayas, Nepal. PhD Thesis, University of Bergen. Bhattarai, K.R., Vetaas, O.R. & Grytnes, J.-A. (2004) Fern species richness along a central Himalayan elevational gradient, Nepal. Journal of Biogeography, 31, 389-400. Birks, H.J.B. (1993) Is the hypothesis of survival on glacial nunataks necessary to explain the present-day distributions of Norwegian mountain plants? Phytocoenologia, 23, 399426. Birks, H.J.B. & Birks, H.H. (2008) Biological responses to rapid climate change at the Younger Dryas-Holocene transition at Kråkenes, western Norway. Holocene, 18, 1930. Birks, H.J.B., Birks, H.H., Everson, J., Jans, H., Thorne, D. & M., T. (2007) The AGS in Tibet 2005. The Alpine Gardener: Tibet Special Issue. The Alpine Garden Society, AGS Centre, Avon Bank, Pershore, UK. pp. 289-349. Bonet, A. & Pausas, J.G. (2004) Species richness and cover along a 60-year chronosequence in old-fields of southeastern Spain. Plant Ecology, 174, 257-270. Brown, J.H. (1995) Macroecology. University of Chicago Press, Chicago. Brown, J.H. (2001) Mammals on mountainsides: elevational patterns of diversity. Global Ecology & Biogeography, 10, 101-109. Brown, J.H., Gillooly, J.F., Allen, A.P., Savage, V.M. & West, G.B. (2004) Toward a metabolic theory of ecology. Ecology, 85, 1771-1789. Brown, J.H. & Kodric-Brown, A. (1977) Turnover rates in insular biogeography: effect of immigration on extinction. Ecology, 58, 445-449. Brown, V.K. & Southwood, T.R.E. (1987) Secondary succession: patterns and strategies. In: Gray, A.J., Crawley, M.J., & Edwards, P.J.(eds) Colonization, Succession and Stability. Blackwell Scientific Publications, Oxford. pp. 315-338. 27 Bruun, H.H., Moen, J., Virtanen, R., Grytnes, J.A., Oksanen, L. & Angerbjörn, A. (2006) Effects of altitude and topography on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal of Vegetation Science, 17, 37-46. Callaway, R.M.R. (2002) Positive interactions among alpine plants increase with stress. Nature, 417, 844. Carpenter, C. (2005) The environmental control of plant species density on a Himalayan elevation gradient. Journal of Biogeography, 32, 999-1018. Carson, W.P. & Barrett, G.W. (1988) Succession in old field plant communities: effects of contrasting types of nutrient enrichment. Ecology, 69, 984-994. Chang, D.H.S. (1981) The vegetation zonation of the Tibetan Plateau. Mountain Research and Development, 1, 29-48. Chang, D.H.S. & Gauch, H.G., Jr. (1986) Multivariate analysis of plant communities and environmental factors in Ngari, Tibet. Ecology, 67, 1568-1575. Cheng-yih, W., ed (1983-1987) Flora Xizangica (Vol. I-V). Science Press, Beijing (In Chinese). Chown, S.L. & Gaston, K.J. (2000) Areas, cradles and museums: the latitudinal gradient in species richness. Trends in Ecology & Evolution, 15, 311-315. Christensen, N.L. & Peet, R.K. (1984) Convergence during secondary forest succession. The Journal of Ecology, 72, 25-36. Clements, F.E. (1928) Plant succession and indicators. H. W. Wilson, New York. Clinebell, R.R., Phillips, O.L., Gentry, A.H., Stark, N. & Zuuring, H. (1995) Prediction of neotropical tree and liana species richness from soil and climatic data. Biodiversity and Conservation, 4, 56-90. Colwell, R.K. & Lees, D.C. (2000) The mid-domain effect: geometric constraints on the geography of species richness. Trends in Ecology & Evolution, 15, 70-76. Connell, J.H. (1978) Diversity in tropical rain forests and coral reef. Science, 199, 13021310. Connell, J.H. & Slatyer, R.O. (1977) Mechanisms of succession in natural communities and their role in community stability and organization. American Naturalist, 111, 11191144. Cowles, H.C. (1899) The ecological relations of the vegetation on the sand dunes of Lake Michigan. 1. Geographical relations of the dune flora. Botanical Gazette, 27, 95-117, 167-202, 281-308, 361-391. Darwin, C. (1859) On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. John Murray, London, UK. de Vera, J.P., Horneck, G., Rettberg, P. & Ott, S. (2004) The potential of the lichen symbiosis to cope with the extreme conditions of outer space II: germination capacity of lichen ascospores in response to simulated space conditions. Advances in Space Research, 33, 1236-1243. del Moral, R. (2007) Limits to convergence of vegetation during early primary succession. Journal of Vegetation Science, 18, 479-488. Dickoré, W.B. & Miehe, G. (2002) Cold spots in the highest mountains of the world diversity patterns and gradients in the flora of the Karakorum. In: Körner, C. & Sphen, E.M.(eds) Mountain biodiversity a global assessment. Parthenon, New York. pp. 129-147. Dobremez, J.F. (1976) Le Népal, Écologie et biogéographie. Centre National de la Rechereche Scientifique, Paris. Dobremez, J.F. & Jest, C. (1969) Carte écologique de la region Annapurna Dhaulagiri. Centre National de la Recherche Scientifque, Paris. 28 Drury, W.H. & Nisbet, I.C.T. (1973) Succession: contribution no. 90 from the scientific staff, Massachusetts Audubon Society. Journal of the Arnold arboretum, 54, 331368. Eriksson, O. (1993) The species-pool hypothesis and plant community diversity. Oikos, 68, 371-374. Fontaneto, D. & Ricci, C. (2006) Spatial gradients in species diversity of microscopic animals: the case of bdelloid rotifers at high altitude. Journal of Biogeography, 33, 1305-1313. Fraser, R.H. & Currie, D.J. (1996) The species richness-energy hypothesis in a system where historical factors are thought to prevail: coral reefs. The American Naturalist, 148, 138-159. Fukami, T., Bezemer, T.M., Mortimer, S.R. & van der Putten, W.H. (2005) Species divergence and trait convergence in experimental plant community assembly. Ecology Letters, 8, 1283-1290. Gaston, K.J. (2000) Global patterns in biodiversity. Nature, 405, 220-227. Gauslaa, Y. (1984) Heat resistance and energy budget in different Scandinavian plants. Holarctic Ecology, 7, 5-78. Gauslaa, Y., Palmqvist, K., Solhaug, K.A., Hilmo, O., Holien, H., Nybakken, L. & Ohlson, M. (2009) Size dependent growth in two old-growth associated macrolichen species. New Phytologist, 18, 683-692. Gauslaa, Y. & Ustvedt, E.M. (2003) Is parietin a UV-B or a blue-light screening pigment in the lichen Xanthoria parietina? Photochemical & Photobiological Sciences, 2, 424432. Gleason, H.A. (1926) The individualistic concept of the plant association. Bulletin of the Torrey Botanical Club, 53, 7-26. Glenn-Lewin, D.C. (1980) The individualistic nature of plant community development. Vegetatio, 43, 141-146. Glenn-Lewin, D.C., Peet, R.K. & Veblen, T.T., eds (1992) Plant Succession: theory and prediction. Chapman and Hall, London. Gotelli, N.J. & Colwell, R.K. (2001) Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters, 4, 379-391. Grau, O., Grytnes, J.-A. & Birks, H.J.B. (2007) A comparison of altitudinal species richness patterns of bryophytes with other plant groups in Nepal, Central Himalaya. Journal of Biogeography, 34, 1907-1915. Grime, J.P. (1977) Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. The American Naturalist, 11, 11691194. Grytnes, J.-A. (2002) Patterns of vascular plant species richness at range of spatial scales. PhD Thesis, University of Bergen. Grytnes, J.A. (2003) Species-richness patterns of vascular plants along seven altitudinal transects in Norway. Ecography, 26, 291-300. Grytnes, J.A. & Beaman, J.H. (2006) Elevational species richness patterns for vascular plants on Mount Kinabalu, Borneo. Journal of Biogeography, 33, 1838-1849. Grytnes, J.A., Heegaard, E. & Ihlen, P.G. (2006) Species richness of vascular plants, bryophytes, and lichens along an altitudinal gradient in western Norway. Acta Oecologica, 29, 241-246. Grytnes, J.A. & Vetaas, O.R. (2002) Species richness and altitude: a comparison between null models and interpolated plant species richness along the Himalayan altitudinal gradient, Nepal. American Naturalist, 159, 294-304. 29 Hagen, T. (1969) Report on the Geological Survey of Nepal: Preliminary reconnaissance (vol. LXXXVI). Orell Füssli Arts Graphiques, Zurich. Harris, N. (2006) The elevation history of the Tibetan Plateau and its implications for the Asian monsoon. Palaeogeography, Palaeoclimatology, Palaeoecology, 241, 4-15. Hawkins, B.A., Field, R., Cornell, H.V., Currie, D.J., Guegan, J.-F., Kaufman, D.M., Kerr, J.T., Mittelbach, G.G., Oberdorff, T., O'Brien, E.M., Porter, E.E. & Turner, J.R.G. (2003) Energy, water, and broad-scale geographic patterns of species richness. Ecology, 84, 3105-3117. Heaney, L.R. (2001) Small mammal diversity along elevational gradients in the Philippines: an assessment of patterns and hypotheses. Global Ecology & Biogeography, 10, 1539. Herzog, S.K., Kessler, M. & Bach, K. (2005) The elevational gradient in Andean bird species richness at the local scale: a foothill peak and a high-elevation plateau. Ecography, 28, 209-222. Horn, H. (1974) The ecology of secondary succession. Annual Review of Ecology and Systematic, 5, 25-37. Hunter, M.L. & Yonzon, P. (1993) Altitudinal distributions of birds, mammals, people, forests, and parks in Nepal. Conservation Biology, 7, 420-423. Huston, M. (1979) A general hypothesis of species diversity. The American Naturalist, 113, 81-101. Huston, M. & Smith, T. (1987) Plant succession-life-history and competition. American Naturalist, 130, 168-198. Huston, M.A. (1994) Biological diversity: the coexistence of species on changing landscapes. Cambridge University Press, Cambridge. Inouye, R.S. & Tilman, D. (1995) Convergence and divergence of old-field vegetation after 11 yr of nitrogen addition. Ecology, 76, 1872-1887. IPCC (2007) Climate change 2007: the physical sciences basis. In summary for policy makers. IPCC. Kansakar, S., R., Hannah, D., M., Gerrard, J. & Rees, G. (2004) Spatial pattern in the precipitation regime of Nepal. International Journal of Climatology, 24, 1645-1659. Kessler, M. (2000a) Altitudinal zonation of Andean cryptogam communities. Journal of Biogeography, 27, 275-282. Kessler, M. (2000b) Upslope-directed mass effect in palms along an Andean elevational gradient: a cause for high diversity at mid-elevations? Biotropica, 32, 756-759. Kessler, M. (2009) The impact of population processes on patterns of species richness: Lessons from elevational gradients. Basic and Applied Ecology, 10, 295-299. Klimes, L. (2003) Life-forms and clonality of vascular plants along an altitudinal gradient in East Ladakh (NW Himalayas). Basic and Applied Ecology, 4, 317-328. Koleff, P., Gaston, K.J. & Lennon, J.J. (2003) Measuring beta diversity for presence-absence data. Journal of Animal Ecology, 72, 367-382. Körner, C. (2000) Why are there global gradients in species richness? Mountains might hold the answer. Trends in Ecology & Evolution, 15, 513-514. Körner, C. (2003) Alpine plant life: functional plant ecology of high mountain ecosystems. (2nd ed.). Springer Verlag, Berlin. Körner, C. (2007) The use of `altitude' in ecological research. Trends in Ecology & Evolution, 22, 569-574. Krebs, C. (2008) The ecological world view. California University Press, Berkeley and Los Angeles, California. 30 La Duo (2008) Studies in the ecology and palaeoecology of the Lhasa valley, Tibet Autonomous Region, China. PhD Thesis, University of Bergen. Lepš, J., Novotny´, V. & Basset, Y. (2001) Habitat and successional status of plants in relation to the communities of their leaf-chewing herbivores in Papua New Guinea. Journal of Environmental Biology, 89, 186-199. Lepš, J. & Šmilauer, P. (2003) Multivariate analysis of ecological data using CANOCO. Cambridge University Press, Cambridge. Levin, S.A. (1981) Mechanism for generation and maintenance of diversity. In: Hiorns, R.H. & Cooke, D.(eds) The mathematical theory of the dynamics of biological populations. Academy Press, London, UK. pp. 176-194. Lomolino, M.V. (2001) Elevation gradients of species-density: historical and prospective views. Global Ecology & Biogeography, 10, 3-13. Lomolino, M.V., Riddle, B.R. & Brown, J.H. (2006) Biogeography. (3rd ed.). Sinauer Associates, Inc. Sunderland, Massachusetts. MacArthur, J.W. & Wilson, E.O. (1967) The theory of island biogeography. Princeton University Press, Princeton. Magurran, A.E. (1988) Ecological diversity and its measurement. Princeton University Press., Princeton. Md. Nor, S. (2001) Elevational diversity patterns of small mammals on Mount Kinabalu, Sabah, Malaysia. Global Ecology & Biogeography, 10, 41-62. Menon, N. (2009) Rainfall Uncertainty and Occupational Choice in Agricultural Households of Rural Nepal. Journal of Development Studies, 45, 864-888. Miehe, G. (1982) Vegetationsgeographische Untersuchungen im Dhaulgiri und Annapurna Himalaya, Dissertationes Botanicæ. Journal Cramer, Vaduz, 66.2, 1-152. Miehe, G. (1988) Geoecological reconnaissance in the alpine belt of southern Tibet. Geojournal, 17, 635-648. Miehe, G. (1989) Vegetation patterns in Mount Everest as influenced by monsoon and föhn. Vegetatio, 79, 21-32. Miehe, G., Miehe, S., Kaiser, K., Reudenbach, C., Behrendes, L., La Duo & Schlütz, F. (2009) How old is pastoralism in Tibet? An ecological approach to the making of a Tibetan landscape. Palaeogeography, Palaeoclimatology, Palaeoecology, 276, 130147. Miehe, G., Miehe, S., Vogel, J., Co, S. & La Duo (2007) Highest treeline in the northern hemisphere found in southern Tibet. Mountain Research and Development, 27, 169173. Miehe, G., Miehe, S., Will, M., Opgenoorth, L., Duo, L., Dorgeh, T. & Liu, J. (2008) An inventory of forest relicts in the pastures of Southern Tibet (Xizang A.R., China). Plant Ecology, 194, 157-177. Mittelbach, G., G., Douglas, W.S., Howard, V.C., Andrew, P.A., Jonathan, M.B., Mark, B.B., Susan, P.H., Allen, H.H., Nancy, K., Harilaos, A.L., Christy, M.M., Amy, R.M., Lucinda, A.M., Mark, A.M., Thomas, J.N., Trevor, D.P., Robert, E.R., Kaustuv, R., Dov, F.S., Dolph, S., James, M.S. & Michael, T. (2007) Evolution and the latitudinal diversity gradient: speciation, extinction and biogeography. Ecology Letters, 10, 315-331. Mittelbach, G.G., Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Willig, M.R., Dodson, S.I. & Gough, L. (2001) What is the observed relationship between species richness and productivity? Ecology, 82, 2381-2396. Molnar, P. (1989) The geologic evolution of the Tibetan Plateau. American Scientist, 77, 350. 31 Ni, J. (2000) A simulation of biomes on the Tibetan Plateau and their responses to global climate change. Mountain Research and Development, 20, 80-89. O' Brien, E.M. (1993) Climatic gradients in woody plant species richness: towards an explanation based on an analysis of Southern Africa's woody flora. Journal of Biogeography, 20, 181-198. O'Neill, R.V. (1989) Hierarchy theory and global change. In: Rosswall, T., Woodmansee, R.G., & Risser, P.G.(eds) Scales and global change. John Wiley and Sons, New York. pp. 29-45. Odland, A. & Birks, H.J.B. (1999) The altitudinal gradient of vascular plant species richness in Aurland, western Norway. Ecography, 22, 548-566. Odum, E.P. (1960) Organic production and turnover in old field succession. Ecology, 41, 3449. Odum, E.P. (1969) The strategy of ecosystem development. Science, 164, 262-270. Odum, E.P. (1971) Fundamentals of ecology. (3rd ed.). Saunders company, Phiadelphia, London, Toronto. Ohlemuller, R. & Wilson, J.B. (2000) Vascular plant species richness along latitudinal and altitudinal gradients: a contribution from New Zealand temperate rainforests. Ecology Letters, 3, 262-266. Oommen, M.A. & Shanker, K. (2005) Elevational species richness patterns emerge from multiple local mechanisms in Himalayan woody plants. Ecology, 86, 3039-3047. Oosting, H.J. (1942) An ecological analysis of the plant communities of Piedmont, North Carolina. The Americian Midland Naturalist, 28, 1-126. Palmer, M.W. (1990) Spatial scale and patterns of species-environment relationships in hardwood forest of North Carolina Piedmont. Coenoses, 5, 79-87. Palmer, M.W. (1993) Putting things in even better order: the advantages of Canonical Correspondence Analysis. Ecology, 74, 2215-2230. Palmer, M.W. (1994) Variation in species richness: towards a unification of hypothesis. Folia Geobotanica, 29, 511-530. Palmer, M.W. (2006) Scale dependence of native and alien species richness in North American floras. Preslia, 78, 427-436. Palmer, M.W. & Dixon, P.M. (1990) Small-scale environmental heterogeneity and the analysis of species distributions along gradients. Journal of Vegetation Science, 1, 57-65. Palmer, M.W. & White, P.S. (1994) Scale dependence and the species-area relationship. The American Naturalist, 144, 717-740. Patterson, B.D., Stotz, D.F., Solari, S., Fitzpatrick, J.W. & Pacheco, V. (1998) Contrasting patterns of elevational zonation for birds and mammals in the Andes of Southeastern Peru. Journal of Biogeography, 25, 593-607. Pausas, J.G. (1994) Species richness patterns in the understorey of Pyrenean Pinus sylvestris forest. Journal of Vegetation Science, 5, 517-524. Pausas, J.G., Bonet, A., Maestre, F. & Climent, A. (2006) The role of the perch on the nucleation process in Mediterranean semi-arid oldfield. Acta Oecologica, 29, 346352. Peet, R.K. (1974) The measurement of species diversity. Annual Review of Ecology and Systematic, 5, 285-307. Pianka, E.R. (1966) Latitudinal gradients in species diversity: a review of concepts. The American Naturalist, 100, 33-46. Pickett, S.T.A. (1976) Succession: an evolutionary interpretation. The American Naturalist, 110, 107-119. 32 Pickett, S.T.A. (1989) Space-for-time substitution as an alternative to long-term studies. In: Likens, G.E.(ed) Long-term studies in ecology: approaches and alternatives. Springer-Verlag, New York. pp. 110-135. Pulliam, H.R. (1988) Sources, sinks and population regulation. American Naturalist, 132, 652-661. Qian, H., Fridley, J.D. & Palmer, M.W. (2007) The latitudinal gradient of species-area relationships for vascular plants of North America. American Naturalist, 170, 690701. Rahbek, C. (1995) The elevational gradient of species richness: a uniform pattern? Ecography, 18, 200-205. Rahbek, C. (1997) The relationship among area, elevation, and regional species richness in Neotropical birds. American Naturalist, 149, 875-902. Rahbek, C. (2005) The role of spatial scale and the perception of large-scale species-richness patterns. Ecology Letters, 8, 224-239. Rajbhandari, K.R. (1976) History of botanical explorations in Nepal. Journal of Bombay Natural History Society, 73, 468-481. Ricklefs, R.E. (1987) Community diversity: relative roles of local and regional processes. Science, 235, 167-171. Rohde, K. (1992) Latitudinal gradients in species diversity: the search for the primary cause. Oikos, 65, 514-527. Romdal, T.S. & Grytnes, J.-A. (2007) An indirect area effect on elevational species richness patterns. Ecography, 30, 440-448. Rosenzweig, M.L. (1995) Species diversity in space and time. Cambridge University Press, Cambridge. Sancho, L.G., de la Torre, R., Horneck, G., Ascaso, C., de los Rios, A., Pintado, A., Wierzchos, J. & Schuster, M. (2007) Lichens survive in space: results from the 2005 lichen experiment. Astrobiology, 7, 443-454. Schaller, G.B. (1998) Wildlife of the Tibetan steppe. University of Chicago Press, Chicago. Sharma, L.R. (1995) Enumeration of Lichens of Nepal (Publication No. 3). Nepal Government. Shmida, A. & Wilson, M.W. (1985) Biological determinants of species diversity. Journal of Biogeography, 12, 1-20. Silvertown, J., Poulton, P., Johnston, E., Edwards, G., Heard, M. & Bliss, P. (2006) The park grass experiment 1856–2006: its contribution to ecology. Journal of Ecology, 94, 801–814. Singh, J. & Singh, S. (1987) Forest vegetation of the Himalaya. The Botanical Review, 53, 80-192. Stainton, J.D.A. (1972) Forest of Nepal. John Murray Publisher, Ltd., London. Stainton, J.D.A. (2001) Flowers of the Himalaya: a supplement. Oxford University Press, Oxford. Stevens, G.C. (1992) The elevational gradient in altitudinal range: an extension of Rapoport's latitudinal rule to altitude. The American Naturalist, 140, 893-911. Sun ZhenHua, Peng ShengJing & Ou XiaoKun (2007) Rapid assessment and explanation of tree species abundance along the elevation gradient in Gaoligong Mountains, Yunnan, China. Chinese Science Bulletin, 52, 225-231. Terborgh, J. (1973) On the notion of favorableness in plant ecology. The American Naturalist, 107, 481-501. Theurillat, J.P., Schlüssel, A., Geissler, P., Guisan, A., Velluti, C. & Wiget, L. (2003) Vascular plant and bryophyte diversity along elevational gradients in the Alps. In: 33 Nagy, L., Grabherr, G., Körner, C., & Thompson, D.B.A.(eds) Alpine Biodiversity in Europe. Springer, Germany, Berlin. pp. 185-193. Tilman, D. (1982) Resource competition and community structure. Princeton University Press, Princeton, NJ. Tilman, D. & Downing, J.A. (1994) Biodiversity and stability in grasslands. Nature, 367, 363-365. Upreti, B.N. (1999) An overview of the stratigraphy and tectonics of the Nepal Himalaya. Journal of Asian Earth Sciences, 17, 577-606. Vazquez, G.J.A. & Givnish, T.J. (1998) Altitudinal gradients in tropical forest composition, structure, and diversity in the Sierra de Manantlan. Journal of Ecology, 86, 999-1020. Vetaas, O.R. & Grytnes, J.-A. (2002) Distribution of vascular plant species richness and endemic richness along the Himalayan elevation gradient in Nepal. Global Ecology and Biogeography, 11, 291-301. Virtanen, R., Dirnböck, T., Dullinger, S., Grabherr, G., Pauli, H., Staudinger, M. & Villar, L. (2003) Patterns in the plant species richness of European high mountain vegetation. In: Nagy, L., Grabherr, G., Körner, C., & Thompson, D.B.A.(eds) Alpine Biodiversity in Europe. Ecological Studies 167. Springer, Berlin. pp. 149–172. von Humboldt (1849) Aspects of nature in different lands and different climates, with scientific elucidations (Miehe, S., trans.). Longman, Brown, Green, London. Wallace, A.R. (1878) Tropical nature and other essays. (New ed.). MacMillan and Co, London. Wang, W.Y., Wang, Q.J., Li, S.X. & Wang, G. (2006) Distribution and species diversity of plant communities along transect on the Northeastern Tibetan plateau. Biodiversity and Conservation, 15, 1811-1828. Wang, Z., Tang, Z. & Fang, J. (2007) Altitudinal patterns of seed plant richness in the Gaoligong Mountains, south-east Tibet, China. Diversity and Distributions, 13, 845854. Whittaker, R.H. (1960) Vegetation of the Siskiyou mountains, Oregon and California. Ecological Monographs, 30, 279-338. Whittaker, R.H. (1975) Communities and ecosystems. (2nd ed.). MacMillion, New York. Whittaker, R.H. (1977) Evolution of species diversity in land communities. Evolutionary Biology, 10, 1-67. Whittaker, R.J., Willis, K.J. & Field, R. (2001) Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography, 28, 453470. Wiens, J.A. (1989) Spatial scaling in ecology. Functional Ecology, 3, 385-397. Willdenow, C.L. (1805) The principles of botany, and vegetable physiology. Blackwood, Cadell and Davies, London. Willis, K.J., Bennett, K.D. & Birks, H.J.B. (2009) Variability in thermal and UV-B energy fluxes through time and their influence on plant diversity and speciation. Journal of Biogeography, 36, 1630-1644. Willis, K.J. & Bhagwat, S.A. (2009) Biodiversity and climate change. Science, 326, 806-807. Willis, K.J. & Whittaker, R.J. (2002) Species diversity-scale matters. Science, 295, 12451248. Wilson, R.J., Gutiérrez, D., Gutiérrez, J., Martínez, D., Agudo, R. & Monserrat, V.J. (2005) Changes to elevational limits and extent of species ranges associated with climate change. Ecology Letters, 8, 1138–1146. Woodward, F.I. (1987) Climate and plant distribution. Cambridge University Press, Cambridge. 34 Wright, D.H. (1983) Species-energy theory: an extension of species-area theory. Oikos, 41, 496-506. Yang, Y., Körner, C. & Sun, H. (2008) The ecological significance of pubescence in Saussurea medusa, a high-elevation Himalayan woolly plant. Arctic, Antarctic, and Alpine Research, 40, 250-255. Yoda, K. (1967) A preliminary survey of the forest vegetation of eastern Nepal II. General description, structure and floristic composition of sample plots chosen from different vegetation zones. Journal of the College of Arts and Sciences, Chiba University National Science series, 5, 99-140. Zobel, M. (1997) The relative of species pools in determining plant species richness: an alternative explanation of species coexistence? Trends in Ecology & Evolution, 12, 266-269. 35 Paper I Baniya, C.B., Solhøy, T., Gauslaa, Y. & Palmer, M.W. Richness and Composition of Vascular Plants and Cryptogams along a High Elevational Gradient on Buddha Mountain, Central Tibet. (Submitted) I Richness and Composition of Vascular Plants and Cryptogams along a High Elevational Gradient on Buddha Mountain, Central Tibet Chitra B. Baniya · Torstein Solhøy · Yngvar Gauslaa · Michael W. Palmer Abstract We explored patterns of species richness and composition along an elevational gradient (4985-5685 m) on Buddha Mountain, 100 km northwest of Lhasa, Tibet, and compared our results to those of models derived from species ranges interpolated from the literature. We recorded presence of plants, lichens, and stone cover in bands of six 1 m2 quadrats, separated horizontally by about 10 m. There was a total of 29 bands, by 25 m apart in elevation along a vertical transect with SE aspect. We recorded 143 species, including 107 angiosperms, 2 gymnosperms, 27 lichens, and 7 mosses. pH, C, N and C/N ratio were measured from two randomly located soil samples collected from 10 cm depth in each band. We applied Detrended Correspondence Analysis (DCA), Canonical Correspondence Analysis (CCA), and Generalized Linear Models (GLMs) to assess relationships between species richness, species composition, and the environment. Substrate and soil related variables showed significant but negative relationships with elevation, except pH which had no significant relationship. CCA and DCA demonstrated a pronounced elevational gradient with continuous change in species composition from alpine steppe to desert steppe vegetation. Total species richness was fairly constant between 4985 and 5400 m and continuously declined above 5400 m. We observed similar patterns for forbs and graminoids. Cushion plants exhibited a unimodal relationship with a peak between 5360 and 5385 m. Shrub richness declined monotonically. Lichen richness was unimodal and peaked at 5485-5510 m. Elevational richness gradients interpolated from published species ranges were similar to our observed patterns. However, the range-based data did not display the richness plateau at lower 1 elevations found for the quadrats. This pronounced scale dependence means that data from published floras have limited ability to predict richness patterns on individual mountains. Keywords: Tibetan Plateau, Life-form, Declining, Diversity, Hump pattern, Observed, Interpolated, High-alpine 2 C. B. Baniya ( ) Department of Biology, University of Bergen, AlOpgaten, N-5007 Bergen, Norway Department of Botany, Tribhuvan University, Kirtipur, Kathmandu, Nepal e-mail: chitra.baniya@bio.uib.no Fax: +47 55 58 96 67 Phone: +47 55 58 22 34 T. Solhøy Department of Biology, University of Bergen, $OOpgaten, N-5007 Bergen, Norway e-mail: torstein.solhoy@bio.uib.no Y. Gauslaa Department of Ecology and Natural Resource Management, Norwegian University of Life Sciences, NO-1432 Ås, Norway e-mail: yngvar.gauslaa@umb.no M. W. Palmer Department of Botany, Oklahoma State University, 104 LSE Stillwater OK 74078 USA, 405744-7717 e-mail: mike.palmer@okstate.edu 3 Introduction The elevational pattern of species richness has stimulated much interest among naturalists, ecologists, and evolutionary biologists. This pattern comprises one of the most spectacular diversity gradients in the world, but is highly scale dependent (Körner 2000; Whittaker et al. 2001; Palmer 2006; Nogués-Bravo et al. 2008). In a review of 204 published elevational species richness pattern studies, Rahbek (2005) found a unimodal pattern in almost 50%, a monotonic pattern in 25% and a mixed pattern in the remaining 25% of studies. The mechanisms behind these patterns are a matter of ongoing debate (Rahbek 1995). Studies that employ niche modelling and interpolation for vascular plants almost invariably reveal unimodal elevational richness patterns (e.g. Grytnes and Vetaas 2002; Wang et al. 2007). The monotonic decrease often found above 2500 m asl may arise due to various non-biological reasons such as undersampling, niche modelling, interpolation errors, etc.; thus inferred patterns must be validated in the field before they can be accepted (Grytnes and Vetaas 2002). A general theory of the elevational gradient must cover many disparate phenomena (Whittaker et al. 2001). Past studies on local richness that have been undertaken to test or verify the general pattern were mainly done in tropical and temperate regions at lower (< 4800 m asl) elevation ranges (Odland and Birks 1999; Kessler 2000; Bruun et al. 2006; Grytnes et al. 2006). Here, we extend the study of local species richness to subtropical mountains with substantially higher elevations (> 4900 m asl). The Tibetan plateau is the highest, largest, and youngest plateau of the world (Chang 1981). This plateau has an average elevation of 4500, high mountains of average elevation 4900 m asl (Ni 2000) and the average snowline lies between 4800 to 5700 m (Birks et al. 2007). This plateau, presumably uplifted during the late Cenozoic period (Molnar 1989), influences monsoon intensity and global climate (Spicer et al. 2003). The Tibetan climate is determined mainly by topography and atmospheric circulation; it is warm and humid in the 4 south-east and cold and arid in the north-west (Chang 1981; Miehe 1988). Temperature and precipitation decline towards the north-west causing a reduction in plant diversity (Ni 2000). The first elevational gradient study in the region investigated vascular plants between 3255 to 4340 m asl (Wang et al. 2002). A reconnaissance (Miehe 1988) and classification (Chang 1981) of Tibetan vegetation started earlier. Several expeditions culminated with the one organised by the Alpine Garden Society (AGS), resulting in a book (Birks et al. 2007). Floristic works include a bryophyte flora (Li 1985), a lichen flora (Wei and Jiang 1986) and five volumes of vascular plant flora (Wu 1983-1987). The Tibetan Plateau lies between biodiversity hotspots of the Himalaya (Nepal, Bhutan, Sikkim) and western China (Yunnan, Sichuan, Qinghai). The declining pattern of diversity at high elevations is particularly marked when conditions supporting life reach their absolute limits. Many species from 5000 m asl and above experience both extremes of their distributions, with freezing conditions towards the upper end and higher temperature towards the lower end (4.2°C higher at 5000 m than at 5700 m when assuming a 0.6°C lapse rate for each 100 m elevation). Plants at this elevation show physical and physiological adaptations to survive freezing air and ground temperatures, low humidity and desiccation (Körner 2000). Birks et al. (2007) and Miehe (1991) reported Saussurea gnaphalodes (at 6400m) as the highest flowering plant in the Nepalese Himalaya, followed by Ermania himalayensis (6300 m), Arenaria bryophylla (6200 m), and Stellaria decumbens (6100 m). Miehe (1991) further reported an upper limit of continuous vegetation between 4600 and 5500 m and patches between 5700 and 6000 m in the central Himalaya and East Asia. Lichens exceed the maximum elevation of phanerogams and mosses (Körner 2003; Feuerer and Hawksworth 2007). Several species of lichens are desiccation tolerant (Miehe 1988; 1991), and achieve an 5 elevation of 5900 m in Mt Kilimanjaro (Beck 1988) and 6700 m in the Andes (Halloy 1991). The absolute record of lichens is at 7400 m from the Himalaya (Hertel 1977). It is possible that patterns found in the well-studied temperate zone are different from those found in the subtropics, because weaker seasonality might lead to different life forms expressing themselves differently. A qualitative change in species composition and turnover is expected with forbs dominating low elevations followed by cushions and crust lichens dominating the highest elevations, as well as major elevational changes in the species richness pattern (Odland and Birks 1999). The purpose of this study is to document the richnesselevation relationship on a high-elevation subtropical mountain, compare a quadrat based richness pattern to models based on interpolated elevational species ranges (Baniya 2009 manuscript submitted) and to evaluate the elevational progression of life forms. Materials and Methods Study Area Buddha Mountain is situated at 30° 11’ N, 90° 29’ E, 100 km northwest of Lhasa, central Tibet. The study area lies from 4985 to 5685 m with an adjacent permanently snow-covered mountain reaching 7000 m. The lowest part of our elevation gradient was at a small river valley at 4985 m. A substrate consisting of small (< 5 cm3) granite stones and fine soil particles occurred up to 5300 m. A zone of purely stone scree started at 5200 m and dominated after 5300 m to the permanent snow-capped summit (> 6000 m). A zone of medium-sized stones (5 to 10 cm3) lay between 5300 and 5500 m along with sorted stripes of parallel vertical bands of stones (> 10 cm3) with plant communities locally in between. This was most likely the result of a longterm complex process of periglacial soil movement and stone sorting. Above this zone there were larger (> 10 cm3) stones with some sand extending to the permanent snow at 6000 m. 6 The bedrock was granite. The soil on the slope was dark black loam towards the lower elevation (<5300 m) and a faint brown or reddish brown sandy solifluction soil from 5300 m and above. The average soil depth in a Kobresia-dominated pasture at 5100 m was about 40 cm (direct measurement) but thin (about 10 cm deep) at higher elevations. According to the Meteorological Bureau of Lhasa, Damzhung climatological station located at 4200 m, and ca. 80 km north of our study area, is the nearest climatological station, has an annual mean temperature of 1.5°C, with 14°C summer average and -7°C winter average temperatures, high fluctuation between diurnal and nocturnal temperature from October to March (10°C and -25°C) and a mean annual precipitation of 442 mm. We studied the SE-facing slope, which probably has a locally drier climate. The rain primarily falls from June to September. Winter precipitation is erratic in Tibet (Harris 2006). Precipitation occurs in the form of snow and hail during October to March. A thin layer of snow often covers the vegetation during winter. Aridity increases with elevation (above 5000 m) together with increasing solar radiation and wind velocity (Ni 2000). The vegetation of this area below 5300 m is alpine steppe of the mid-alpine zone followed by nival or desert steppe at higher altitudes (Chang 1981; Miehe 1988; Birks et al. 2007). The alpine steppe had a sparse plant cover, poor soil with gravel, sand, and silt, and lacked a humus layer. The vegetation comprised patches of Kobresia spp., Carex spp., Stipa koelzi, cushions of Androsace tapete and Arenaria bryophylla, Potentilla spp., Astragalus donianus, Oxytropis spp., Ranunculus lobatus, Lancea tibetica, Thalictrum alpinum, Delphinium spp. together with Saussurea spp. Plants in the desert steppe were mixtures of succulents, such as Sedum spp. and Rhodiola spp., and cushion plants, such as Androsace coronata, A. zambalensis, Arenaria gerzensis, and Thylacospermum caespitosum. There were also stone-protected species such as Draba glomerata, Gentiana urnula, Ranunculus involucratus, and Saxifraga spp. along with the moisture-demanding annual Koenigia 7 islandica and the moss Anomobryum concinnatum. Saxicolous lichens such as Dimelaena oreina and Rhizoplaca peltata occurred at high elevations. Sampling The vegetation was sampled in the summer of 2006 using a fixed-area technique. We obtained additional data in the summer of 2007. We followed the approach of Kessler (2000) and Bruun et al. (2006) to study community composition and diversity along elevational transects by subdividing the gradient into smaller elevational bands, each separated by 25 m vertical elevation. Our study started at 4985 m asl. Six 1 m2 quadrats were placed horizontally at intervals of almost 10 m in each band. All vascular plant, bryophyte, and lichen species were recorded inside each quadrat. No liverworts or pteridophytes were found. The horizontal distance between the highest elevation band and permanent snow cover was 160 m. The % stone cover per quadrat was estimated being the most dominant substrate throughout the gradient. Field identification of vascular plants was done using Polunin and Stainton (2001) and Stainton (2001). Unidentified taxa were collected, and final identification and confirmation were done after comparison with authenticated specimens previously deposited in the Bergen University Herbarium (BG) and by consulting experts. Nomenclature for vascular plants followed “Flora Xizangica” (Wu 1983-1987), Li (1985) for mosses and Wei and Jiang (1986) for lichens. All specimens were deposited at BG. A number of taxa remained unresolved due to the difficult taxonomy and difficulties with the Chinese language. Thus, some of the taxa were listed as Oxytropis 1 and 2, and Saussurea 1, 2 and 3 for vascular plants, and the mosses Thuidium sp. and Funaria sp. However, we are fairly confident that these unresolved taxa represent true species, and thus that the lack of species epithets will not affect our measures of the elevational richness pattern. 8 Interpolated species ranges from floras Baniya (2009, manuscript submitted) derived regression models for species richness as a function of elevation, by interpolating elevation ranges of species between 4900 and 6000 m. Reported ranges were obtained from published Tibetan floras (Wu 1983-1987; Li 1985; Wei and Jiang 1986). The interpolated richness can be subcategorized into total richness, vascular richness, graminoid richness and lichen richness. Soil analysis Two soil samples were taken randomly from two of six quadrats in each elevation band at a depth of 10 cm. We minimised the number of soil samples to avoid difficulties of crossborder transportation. Soil pH was measured by a digital pH meter model 131E in a soil/distilled water suspension (2:5). Total nitrogen (N) was quantified by the Kjeldahl digestion process and total carbon (C) by the Walkley-Black method. All these analyses were done in the soil laboratory, Lalitpur, Nepal. The methods are described by Sumner (1999). Numerical analyses Ordination Environmental variables were first averaged over all the quadrats in each band. Soil C and N were log transformed. We applied ordination to analyse species composition and turnover as a function of environmental variables. We applied default settings of Detrended Correspondence Analysis (DCA) but downweighting of rare species (Lepš and Šmilauer 2003, p. 50) to explore the length of the gradient and rate of change in species composition along the elevational gradient. The length of the first DCA axis was 4.7 SD units, thus justifying use of unimodal methods such as CCA (Lepš and Šmilauer 2003). We used manual forward selection with 499 permutations in CCA to select variables explaining species 9 composition. Ordinations were implemented using CANOCO for Windows 4.5 and CanoDraw 4.0 (ter Braak 2002; ter Braak and Šmilauer 2002). Correlation and regression We analysed interrelationships among variables using Spearman’s rank correlation. We analysed total species richness per band, as well as the richness of vascular and non-vascular plants (cryptogams). The vascular richness was further subdivided into forb, shrub, cushion, and graminoid (Cyperaceae, Juncaceae, Gramineae) richness, and similarly the cryptogam species richness into moss and lichen richness. We employed Generalized Linear Models (GLMs, McCullagh and Nelder 1989) to relate richness to elevation. Since response variables were counts, we tested the dispersion in our data and they showed overdispersion. Thus a quasi-Poisson distribution and a logarithmic link were employed. Inspection of diagnostic plots between a logarithmic link and an identity link (assuming a normal distribution of errors) confirmed that a quasi-Poisson distribution with a logarithmic link function was better than a normal distribution and an identity link. We tested up to third-order linear models to describe the relationship between species richness and elevation. GLMs using linear, quadratic or cubic polynomials were first tested against each other and then with the null model if the previous was statistically significant. An F-test was used to select the best model (the best model is with the highest F-value among the significant models). Similar regression methods were derived for interpolated species range data (Baniya 2009 manuscript submitted). The best significant model was selected and compared with the best significant model of our quadrat data for the similar subcategory species richness. The final graphs were based on the best selected model. We used R 2.8.1 statistical package (R Development Core Team 2008) for all regressions. 10 Results A total of 143 species was recorded, including 107 angiosperms in 68 genera, 2 gymnosperms in 2 genera, 27 lichens in 23 genera, and 7 mosses in 7 genera (Appendix 1). Our vascular plants represented four life forms, among which forb was the most common with 88 species including succulents (6 species) followed by graminoids (8 species), cushions (7 species) and shrubs (6 species). Environmental correlation Elevation had significant correlations with measured soil and substrate related environmental variables except pH (Table 1). Soil N, C, and C/N ratio showed negative correlations with elevation, whereas average % stone cover per band increased (r = 0.78, Table 1). Soil pH did not have significant relationships with other measured environmental variables (Table 1). Elevation % stone pH 1 0.78 -0.31 1 -0.04 1 N C -0.82 -0.86 -0.84 -0.89 0.16 0.16 1 0.99 1 C/N -0.67 Elevation -0.61 % stone 0.16 pH 0.54 N 0.59 C 1 C/N Table 1 Spearman’s rank correlation coefficient matrix for the environmental variables other than elevation measured along an elevation gradient of Buddha Mountain. The critical tabulated value for the Spearman’s rank correlation coefficient is: n ” 29, PĮ(2) ” 0.05 • |0.36|. % stone = average percentage of stone cover per band, pH = concentration of H+ ion of soil, N = % of total nitrogen, C = % soil organic carbon, and C/N- ratio = ratio of soil organic carbon and nitrogen. 11 Species composition and distribution A CCA biplot (Fig. 1) derived from forward selection of explanatory variables revealed a strong relationship between species composition and elevation. The first axis eigenvalue of the CCA is 0.638, which is almost as high as the first eigenvalue of Correspondence Analysis (0.659). This means the measured variables successfully explained the strongest gradient. The first axis CCA eigenvalues is much stronger than the second (0.213), implying that elevation dominates the species-environment relationship (Jongman et al. 1995). The species towards the left hand side of the CCA biplot (Fig. 1) are generally vascular plants (largely forbs) that have higher abundance at lower elevations. Species showing their highest abundance towards higher elevations (right side of the biplot, Fig. 1) were mostly lichens. Forbs such as Astragalus donianus, Kobresia pygmaea, Stellaria subumbellata, Potentilla anserina, Polygonum campanulatum favoured moderate soil pH at the lower elevation (Fig. 1). They inhabited stream-bed sediments, with moderately less % stone cover than at high elevation. Species occurring between 5100 and 5300 m were scree dwellers. Potentilla fruticosa shrubs as well stone-created microhabitats that may have facilitated the nucleation of herbs such as Bistorta vivipara, Ranunculus lobatus, Crepis flexuosa, Carex atrofusca and the moss Racomitrium lanuginosum. These species have their optima in abundance towards lower elevations and less % stone cover at the negative end of the CCA axis I (Fig. 1). These species compositions are characteristic of the alpine steppe vegetation or the vegetation of the mid-alpine zone. 12 Fig. 1 CCA species-environment biplot for the Buddha Mountain data. The first two axes explained 87.7% variation in the species-environment data. Scaling based on inter-species biplot. The three significant (p ” 0.05) environmental variables were selected after manual forward selection and 499 permutations (Monte Carlo permutations tests). Full name for each species is given in Appendix 1. For clarity, the positions of some species were changed slightly to avoid overlap and only species with abundances greater than 10% of the maximum abundance are displayed. The forb Ranunculus involucratus, the lichens Frutidella caesioatra, Rhizoplaca chrysoleuca, Psora decipiens, Solorina crocea, Stereocaulon alpinum, and the bryophyte Campylophyllum halleri have optima towards the higher elevation (right) end of CCA axis I (Fig. 1). Elevations above 5300 m had the highest % stone cover but sparse vegetation, and were mostly dominated by crustose lichens and a few mosses. Cushion plants such as Androsace tapete, Arenaria bryophylla, Androsace coronata, Androsace zambalensis, and Thylacospermum caespitosum occur towards the middle of CCA axis I (Fig. 1). 13 Rhodiola himalensis (5000-5500 m), Draba glomerata (5000-5500 m), and the dark orange-coloured saxicolous lichen Xanthoria elegans (5000-5700 m) were the most dominant species along our elevational gradient and had a wide elevational range. CCA revealed the strong effect of elevation. DCA is more effective at showing the rate of change of species composition along gradients (beta diversity). The total beta diversity is 4.7 SD units, indicating a complete turnover of species along the elevational gradient. Below 5500 m, species composition varied continuously, without breaks or thresholds, as a function of elevation (Fig. 2). This continuity implies that classification of alpine vegetation is arbitrary, and communities grade into each other. Above 5500 m, communities were dominated by a few species of lichen, and there was no clear elevational gradient in composition. Species richness Total species richness ranged from 8 to 45 species per band. Similarly, vascular plants ranged from 0 to 40, lichen species from 0 to 16. The average number of species per band was 29 for total species, and 21 for vascular plant species. Fig. 2 DCA axis I sample score of elevational bands as a function of elevation. 14 Forbs and lichens had opposing trends in the % share of their richness in the total dataset. The lowermost elevation band had about 80% forbs (Fig. 3). Moss and cushion species were never in the majority, and have a similar % share of species, except that the highest four bands were devoid of cushions. Graminoids had the third largest share of species and did not occur in the six highest elevation bands. Shrubs were mostly limited to the lower half of the measured gradient (Fig. 3). Fig. 3 Percentage of species representing different life forms as a function of elevation. Total species richness was relatively constant at about 36 species up to about 5400m, above which it declined continuously (Fig. 4A and Table 2). This pattern held for vascular plant richness (Fig. 4B), forb richness (Fig. 4D) and graminoid richness (Fig. 4G). However, shrub richness (Fig. 4E) exhibited a monotonic decline. Although the fitted curves for vascular plants, forbs, and graminoids seem to indicate unimodality, the data themselves do not (Fig. 4B, 4D, and 4G). Cushion plant richness was clearly unimodal with a peak between 5360 and 5385m (Fig. 4F). 15 Fig. 4 The relationship between species richness (of six 1m2 quadrats) and elevation, with polynomial regression functions through GLM superimposed: (A) All species; (B) All vascular plants; (C) All Cryptogams; (D) Forbs; (E) Shrubs; (F) Vascular cushion plants; (G) Graminoids; (H) Lichens. Note the differences in the scale of the ordinate. Like richness of vascular plants, cryptogam species richness also showed marked and differential trends with elevation (Fig. 4C and Table 2). Lichen richness (Fig. 4H) behaved much like cryptogam richness (Fig. 4C) which is not surprising since lichens constituted the majority of cryptogams. Mosses (not shown) were represented by only a few species and had 16 no clear elevational pattern. Richness of cryptogams and lichens increased at elevations where vascular species richness gradually declined trend (Fig. 4A and C). Both cryptogam and lichen richness had their maximum mean richness at 5500m. The most frequent species at this elevation were a forb, Ranunculus involucratus; lichens, Frutidella caesioatra, Rhizoplaca chrysoleuca, Psora decipiens, Solorina crocea, Stereocaulon alpinum; and a bryophyte Vascular richness Cryptogam species richness Forb richness Shrub richness Vascular cushion richness Graminoid richness Lichen richness Null 28 Elev 27 39.06 *** 52.32 *** 13.15 ** 54.11 *** 49.31 *** 0.67 n. s. 73.76 *** 15.52 *** Elev2 26 61.26 *** 97.6 *** 9.76 *** 96.03 *** 28.46 *** 15.37 *** 102.77 *** 13.45 *** Elev3 25 67.69 *** 94.01 *** 12.06 *** 78.22 *** 17.7 *** 19.61 *** 77.99 *** 18.35 *** df Total species richness Campylophyllum halleri (Fig. 1). Table 2 Generalized linear models with F-values. Different species richness has been taken as response and elevation as predictor variables. Model applied up to polynomial third order. Each model was tested against each other and a null model. The statistically significant model with the higher F value was selected as the best explanation of the relationship between the response variable and elevation as a predictor variable. df = residual degrees of freedom, n. s. = non significant; * = p ” 0.01; ** = ” 0.001; *** = ” 0.0001. Comparison with interpolated species richness When interpolated from species ranges, total richness, total vascular richness and graminoid richness all demonstrated significant, continuous declines in species richness as a function of elevation (Fig. 5 A, B and C). In contrast, interpolated lichen richness had a marked unimodal 17 relationship (Fig. 5D). Despite superficial similarities between interpolated models and quadrat-based models, there are pronounced differences. Naturally, the interpolated ranges (which could in some senses be considered to represent a “species pool”) generally had higher richness than the quadrat richness. Lichen richness (Fig. 5D) is an exception: we found more lichen species in our quadrats than were reported for those elevations. For other groups (Fig. 5A, B, C), in contrast to the interpolated ranges, we found relatively constant richness at lower elevations, followed by a decline at a certain elevational threshold. Although both models for lichen richness exhibited unimodal patterns, the peak for the quadrat data was much narrower and shifted towards higher elevation (Fig. 5D). Fig. 5 Polynomial GLM regression models of species richness as a function of area for quadrat richness (solid lines; these models are the same as in Fig. 4) and interpolated species ranges (dotted lines). Note differences in scale in the ordinates. A. Total richness; B. Total vascular plant richness; C. Graminoid species richness, and D. Lichen species richness. 18 Discussion We confirmed an elevational decline in the total species richness both at the quadrat scale, and using interpolated species ranges as predicted. However, this pattern did not equally apply for all functional and taxonomic subsets of species. In particular, quadrat richness of cryptogams, cushions, and lichens displayed unimodal relationships with elevation. This elevational decline of total species richness in quadrats is consistent with other alpine studies both for the Tibetan Plateau and elsewhere (Grabherr et al. 1995; Pavon et al. 2000; Körner 2003; Bruun et al. 2006; Wang et al. 2006; Birks et al. 2007). However, there are differences in the shape of the elevational declines. For example, we observed fairly constant richness up to 5435 m and then a continuous decline, in contrast to the stepwise decline as found in the Alps by Grabherr et al. (1995). The importance of habitats such as springs, fens, rock and scree communities, pioneer vegetation on moraines, snow beds, and avalanche paths which are common to our studied sites are likely to create azonal communities that give continuously declining patterns (Grabherr et al. 1995) instead of stepwise declines. However, the latter case may be true for lower alpine vegetation not studied here and this case may be common to plateau like Tibet. A continuous relationship between the DCA first axis sample score and elevation further supports an interpretation of continuous change. Differences in species richness patterns between vascular and cryptogams have been reported previously (Bruun et al. 2006; Grytnes et al. 2006), but between group variation was much stronger and diverse than we would have anticipated from the existing literature. The difference in the trends in models between the quadrat and interpolated species range richness is likely due to the scale-dependence of species along the elevation gradient. The differences between quadrat-based and flora-based elevational richness patterns probably have multiple causes. Larger areas at lower elevations may inflate species richness simply due 19 to the species-area relationship (Kessler 2000; Jürgen et al. 2006) thus eliminating the plateau we found in our quadrat data. Environmental heterogeneity, largely minimised in our study, may vary as a function of elevation (Palmer 2006) and thus accentuate differences from interpolations studies. Fine-scale environmental heterogeneity in factors such as soil pH, N, C, moisture, atmospheric humidity, disturbance, etc. may be effectively ‘averaged out’ at broader floristic scales. Also, local endemism, which may increase at higher elevations for biogeographic regions (Jürgen et al. 2006) will inflate the high-elevation richness in floras covering many mountains, but not necessarily in quadrat-based studies. Interestingly, we found higher richness for lichens at some elevation bands (e.g. 5500 m) in the quadrats than in the interpolated species ranges. This implies that for lichens, a vastly understudied taxonomic group, data quality may strongly influence inferred species ranges. Our observed patterns for the total species richness and its functional groups are consistent with general hypotheses of diversity such as available energy, disturbance, and environmental stress and stability (Whittaker et al. 2001). Both temperature and precipitation decreased with increasing elevation, and limited the distribution of all our species. Thus normally we can expect high species richness at lower elevations with high energy than towards higher elevations with low energy, as predicted by the energy hypothesis (Brown 1981; Wright 1983). This hypothesis was mainly proposed for macro-scale, woody species richness in the tropics and subtropics (Brown 1981; Wright 1983; O' Brien 1993). Coincidence among these patterns would have other causal factors than energy due to colonisation of non-woody species in such harsh and stressful environments. Other general hypotheses related to disturbance (Huston 1994) and stress and stability (Callaway 2002) are plausible explanations for local-scale patterns. The elevation of maximum richness of lichens (5500 m) was the highest among all biological groups along this elevational gradient. In general, lichens reach higher maximum 20 elevations than phanerogams and mosses (Hertel 1977; Körner 2003; Feuerer and Hawksworth 2007). Since lichens were the main component of our cryptogam richness, cryptogams also have their maximum richness at the same elevation. The environment at high elevations is highly stressed due to frost action, below-freezing air and ground temperature, strong wind, snow, poor soil, periods with lack of humidity, strong UV-B and solar radiation, short growing season, etc. Such factors may strongly impact the distribution of vascular plants, but not necessarily destructive to lichens that have a capacity to survive even simulated (de Vera et al. 2004) and real (Sancho et al. 2007) outer space conditions. A linearly increasing elevational richness pattern for lichens has also been reported (Körner 2003; Grytnes et al. 2006). Higher richness for mosses and lichens than vascular plants was reported from European Alps (Theurillat et al. 2003; Virtanen et al. 2003). These studies represented lower alpine zones than ours, and thus did not capture the descending limb of the unimodal richness curve. Heterogeneous microhabitats of Tibetan mountains which were known less would also be likely cause. Our ordination clearly indicates a transition from alpine steppe to desert steppe vegetation. We found a dominant elevational gradient in the species composition that was only slightly modified by other environmental factors. The Tibetan Plateau is unique because of its very high elevation and high solar radiation. Elevation is directly related to temperature and precipitation (not measured) and it is hard to separate between them but precipitation is assumed to be a limiting factor by Chang (1981) and Birks et al (2007). The importance of both is also highlighted by Chang and Gauch (1986), Ni (2000), and Wang et al. (2006). We conclude that there are similarities and differences in high elevational species richness patterns of vascular and cryptogam species and their groups from the both observed and interpolated species range. Total, vascular and graminoid richness exhibited a general declining pattern with increasing elevation, whereas a unimodal pattern was found for cushion 21 plant and lichen richness. Range-based data did not display the richness plateau at lower elevation found for quadrats. The peak for lichens was broader and at a lower elevation for the interpolated data. Our observed richness-elevation relationships resemble those of interpolated species ranges within the Tibetan Plateau, but differ in important respects. These differences imply that species-range based data have limited value in predicting patterns found on individual mountains. Acknowledgements Thanks to the Norwegian State Education Loan Fund (Lånekassen) for funding, the Norway-Tibet Network for partial travel support, and the Central Department of Botany, Tribhuvan University, Kathmandu, Nepal for leave to Baniya CB. Thanks also due to John Birks for his help in plant identifications and valuable suggestions after going through some earlier versions of this manuscript. Einar Heegard and John-Arvid Grytnes are acknowledged for their help. Tsering, Cai Dong, Pubu, La Qiong, LaDuo, Droba, and Frode Falkenberg helped to complete this study. Thanks are also due to Kristian Hassel for helping mosses determinations and Per Magnus Jørgensen for helping identifying lichens. Thanks to Cathy Jenks for language correction. References Baniya CB (2009) Vascular and cryptogam richness in the world's highest alpine zone, Tibet. Manuscript submitted Beck E (1988) Plant life on top of Mt. Kilimanjaro (Tanzania). Flora 181:379-381 Birks HJB, Birks HH, Everson J, Jans H, Thorne D, M. T (2007) The AGS in Tibet 2005. The Alpine Gardener: Tibet Special Issue. The Alpine Garden Society, AGS Centre, Avon Bank, Pershore, UK pp 289-349 22 Brown JH (1981) Two decades of homage to Santa Rosalia: toward a general theory of diversity. Am Zool 21:877-888 Bruun HH, Moen J, Virtanen R, Grytnes JA, Oksanen L, Angerbjörn A (2006) Effects of altitude and topography on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal of Vegetation Science 17:37-46 Callaway RMR (2002) Positive interactions among alpine plants increase with stress. Nature 417:844 Chang DHS (1981) The vegetation zonation of the Tibetan Plateau. Mountain Research and Development 1:29-48 Chang DHS, Gauch HG, Jr. (1986) Multivariate analysis of plant communities and environmental factors in Ngari, Tibet. Ecology 67:1568-1575 de Vera JP, Horneck G, Rettberg P, Ott S (2004) The potential of the lichen symbiosis to cope with the extreme conditions of outer space II: germination capacity of lichen ascospores in response to simulated space conditions. Advances in Space Research 33:1236-1243 Feuerer T, Hawksworth DL (2007) Biodiversity of lichens, including a world-wide analysis of checklist data based on Takhtajan's floristic regions. Biodivers Conserv 16:85-98 Grabherr G, Gottfried M, Gruber A, Pauli H (1995) Patterns and current changes in alpine plant diversity. In Chapin III FS, Körner C (eds) Arctic and alpine biodiversity: patterns, causes and ecosystem consequences. Ecological studies 113. SpringerVerlag, Berlin pp 167-181 Grytnes JA, Heegaard E, Ihlen PG (2006) Species richness of vascular plants, bryophytes, and lichens along an altitudinal gradient in western Norway. Acta Oecol 29:241-246 23 Grytnes JA, Vetaas OR (2002) Species richness and altitude: a comparison between null models and interpolated plant species richness along the Himalayan altitudinal gradient, Nepal. Am Nat 159:294-304 Halloy SRP (1991) Islands of life at 6000 m altitude: the environment of the highest autotrophic communities on earth (Socompa Volcano, Andes). Arct Alp Res 23:247262 Harris N (2006) The elevation history of the Tibetan Plateau and its implications for the Asian monsoon. Palaeogeogr, Palaeoclimatol, Palaeoecol 241:4-15 Hertel H (1977) Gesteinsbewohnende Arten der Sammclgattung Lecidea (Lichenes) aus Zentral-, Ost- und Südasien. Eine erste Übersicht. Khumbu Himal 6:145-378 Huston MA (1994) Biological diversity: the coexistence of species on changing landscapes. Cambridge University Press, Cambridge Jongman RHG, ter Braak JFT, Van Tongeren OFR (1995) Data Analysis in community and landscape ecology. Cambridge University Press, Cambridge Jürgen K, Kessler M, Robert RD (2006) What drives elevational patterns of diversity? A test of geometric constraints, climate and species pool effects for pteridophytes on an elevational gradient in Costa Rica. Global Ecol Biogeogr 15:358-371 Kessler M (2000) Altitudinal zonation of Andean cryptogam communities. J Biogeogr 27:275-282 Körner C (2000) Why are there global gradients in species richness? Mountains might hold the answer. Trends Ecol Evol 15:513-514 Körner C (2003) Alpine plant life: functional plant ecology of high mountain ecosystems. Springer Verlag, Berlin Lepš J, Šmilauer P (2003) Multivariate analysis of ecological data using CANOCO. Press syndicate of the University of Cambridge, Cambridge, United Kingdom 24 Li XJ (1985) Bryoflora of Xizang. Science Press, Beijing (In Chinese) McCullagh P, Nelder JA (1989) Generalized linear models. Chapman & Hall, London Miehe G (1988) Geoecological reconnaissance in the alpine belt of southern Tibet. GeoJournal 17:635-648 Miehe G (1991) Der Himalaya, eine multizonale Gebirgsregion. In Walter H, Breckle S (eds) Ökologie der gemäȕigten und arktischen Zonen auȕerhalb Euro-Nordasiens. Fischer, Stuttgart pp 181-230 Molnar P (1989) The Geologic Evolution of the Tibetan Plateau. Am Sci 77:350 Ni J (2000) A simulation of biomes on the Tibetan Plateau and their responses to global climate change. Mountain Research and Development 20:80-89 Nogués-Bravo D, Araujo MB, Romdal T, Rahbek C (2008) Scale effects and human impact on the elevational species richness gradients. Nature 453:216-219 O' Brien EM (1993) Climatic gradients in woody plant species richness: towards an explanation based on an analysis of Southern Africa's woody flora. J Biogeogr 20:181-198 Odland A, Birks HJB (1999) The altitudinal gradient of vascular plant species richness in Aurland, western Norway. Ecography 22:548-566 Palmer MW (2006) Scale dependence of native and alien species richness in North American floras. Preslia 78:427-436 Pavon NP, Hernandez-Trejo H, Rico-Gray V (2000) Distribution of plant life forms along an altitudinal gradient in the semi-arid valley of Zapotitlan, Mexico. Journal of Vegetation Science 11:39-42 Polunin O, Stainton A (2001) Concise flowers of the Himalaya. Oxford University Press, R Development Core Team (2008) R: A language and environment for statistical computing version 2.8.1. R Foundation for Statistical Computing, Vienna, Austria 25 Rahbek C (1995) The elevational gradient of species richness: a uniform pattern? Ecography 18:200-205 Rahbek C (2005) The role of spatial scale and the perception of large-scale species-richness patterns. Ecol Lett 8:224-239 Sancho LG, de la Torre R, Horneck G, Ascaso C, de los Rios A, Pintado A, Wierzchos J, Schuster M (2007) Lichens survive in space: results from the 2005 LICHENS Experiment. Astrobiology 7:443-454 Spicer RA, Harris NBW, Widdowson M, Herman AB, Guo S, Valdes PJ, Wolfe JA, Kelley SP (2003) Constant elevation of southern Tibet over the past 15 million years. Nature 421:622-624 Stainton JDA (2001) Flowers of the Himalaya: a supplement. Oxford University Press, Oxford Sumner ME (1999) Handbook of soil science. CRC Press LLC, ter Braak CJF (2002) CANOCO-A FORTRAN program for canonical community ordination by (partial) (detrended) (canonical) correspondence analysis, principal component analysis and redundancy analysis, Version 4.5 ed. Wageningen ter Braak CJF, Šmilauer P (2002) CANOCO reference manual and CanoDraw for Windows user's guide: software for canonical community ordination (version 4.5). Ithaca NY USA Microcomputer Power Theurillat JP, Schlüssel A, Geissler P, Guisan A, Velluti C, Wiget L (2003) Vascular plant and bryophyte diversity along elevational gradients in the Alps. In Nagy L, Grabherr G, Körner C, Thompson DBA (eds) Alpine Biodiversity in Europe. Ecological Studies 167. Springer, Germany, Berlin pp 185-193 Virtanen R, Dirnböck T, Dullinger S, Grabherr G, Pauli H, Staudinger M, Villar L (2003) Patterns in the plant species richness of European high mountain 26 vegetation. In Nagy L, Grabherr G, Körner C, Thompson DBA (eds) Alpine Biodiversity in Europe Ecological Studies 167. Springer, Berlin, Germany pp 149–172 Wang QJ, Liu JQ, Zhao XQ (2002) Patterns of plant species diversity in the Northeastern Tibetan Plateau, Qinghai, China. In Körner C, Spehn E (eds) Mountain biodiversity - a global assessment. Parthenon, New York pp 149-153 Wang WY, Wang QJ, Li SX, Wang G (2006) Distribution and species diversity of plant communities along transect on the Northeastern Tibetan plateau. Biodivers Conserv 15:1811-1828 Wang Z, Tang Z, Fang J (2007) Altitudinal patterns of seed plant richness in the Gaoligong Mountains, south-east Tibet, China. Divers Distrib 13:845-854 Wei J-C, Jiang Y-M (1986) Lichens of Xizang. Science Press, Beijing (In Chinese) Whittaker RJ, Willis KJ, Field R (2001) Scale and species richness: towards a general, hierarchical theory of species diversity. J Biogeogr 28:453-470 Wright DH (1983) Species-energy theory: an extension of species-area theory. Oikos 41:496506 Wu C-Y (1983-1987) Flora Xizangica. Vol. I-V. Science Press, Beijing (In Chinese) 27 Appendix 1 Scientific name, abbreviation, family, lifeform subgroup1 and 2 of species from the Buddha Mountain, Central Tibet. The abbreviation is the name of the species that is used for the CANOCO analysis. Names are arranged alphabetically among their taxonomic group. Nomenclature for vascular plants follow the “Flora Xizangica”, volume I-V (Wu 1983-1987), Li (1985) for mosses and Wei and Jiang (1986) for lichens. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Scientific name of species Biatora sphaeroides (Dicks.) Körb. Frutidella caesioatra (Schaer.) Kalb Sporastatia testudinea (Ach.) A. Massal. Cladonia ustulata (Hook.f. & Taylor) Leight. Leptogium saturninoides H. Magn. Aspicilia lesleyana Darb. Aspicilia maculata (H.Magn.) Oksner Lecanora polytropa (Ehrh.) Rabenh. Rhizoplaca chrysoleuca (Sm.) Zopf Rhizoplaca peltata (Ramond) Leuckert & Poelt Lecidea auriculata Th. Fr. Arctocetraria nigricascens (Nyl.) Kärnefelt & A. Thell Vulpicida pinastri (Scop.) J.-E. Mattsson Peltigera membranacea (Ach.) Nyl. Solorina bispora Nyl. Solorina crocea (L.) Ach. Ochrolechia pallescens (L.) A. Massal. Dimelaena oreina (Ach.) Norman Physconia muscigena (Ach.) Poelt Physconia distorta (With.) J.R. Laundon Psora decipiens (Hedw.) Hoffm. Rhizocarpon geographicum (L.) Dc. Thamnolia vermicularis (Sw.) Ach. ex Schaer. Stereocaulon alpinum Laurer Xanthoria elegans (Link) Th. Fr. Diploschistes muscorum (Scop.) R. Sant. Umbilicaria yunnana (Nyl.) Hue. Campylophyllum halleri Fleisch. Andreaea rupestris Hedwig Anomobryum concinnatum (Spruce) Lindb. Bryum argenteum Hedw. Funaria sp. Racomitrium lanuginosum Bridel Thuidium sp. Silene himalayensis (Edgew.) Majumdar Thylacospermum caespitosum (Cambessèdes) Schischkin Arenaria bryophylla Fernald Arenaria gerzensis L.H. Chou Androsace coronata Hand.-Mazz. Androsace tapete Maxim. Androsace zambalensis Hand.-Mazz. Cortiella hookeri (C. B.Clarke) C.Norman Ajania nubigena (Wall.) C.Shih Anaphalis xylorhiza Sch.Bip. ex Hook.f. Artemisia stricta Heyne ex DC. Abbr. Biat sph Frut cae Spor tes Clad ust Lept sat Aspi les Aspi mac Leca pol Rhiz chr Rhiz pel Leci aur Family Bacidiaceae Bacidiaceae Catillariaceae Cladoniaceae Collemataceae Hymeneliaceae Hymeneliaceae Lecanoraceae Lecanoraceae Lecanoraceae Lecideaceae Subgroup1 cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam Subgroup2 lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen Acrt nig Vulp pin Pelt mem Solo bis Solo cro Ochr pal Dime ore Phys mus Phys dis Psor dec Rhiz geo Tham ver Ster alp Xant ele Dipl mus Umbi yun Camp hal Andr rup Anob con Bryu arg Funa sp Raco lan Thui sp Sile him Parmeliaceae Parmeliaceae Peltigeraceae Peltigeraceae Peltigeraceae Pertusariaceae Physciaceae Physciaceae Physciaceae Psoraceae Rhizocarpaceae Siphulaceae Stereocaulaceae Teloschistaceae Thelotremataceae Umbilicariaceae Amblystegiaceae Andreaeaceae Bryaceae Bryaceae Funariaceae Grimmiaceae Thuidiaceae Caryophyllaceae cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam cryptogam vascular lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen lichen moss moss moss moss moss moss moss cushion Thyl cae Aren bry Aren ger Andr cor Andr tap Andr zam Cort hoo Ajan nub Anap xyl Arte str Caryophyllaceae Caryophyllaceae Caryophyllaceae Primulaceae Primulaceae Primulaceae Apiaceae Asteraceae Asteraceae Asteraceae vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular cushion cushion cushion cushion cushion cushion forb forb forb forb 28 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 Aster himalaicus C.B.Clarke Cremanthodium ellisii (Hook.f.) S.Kitamura Cremanthodium nanum (Decne.) W.W.Smith Cremanthodium thomsonii C.B.Clarke Cremanthodium rhodocephalum Diels Crepis flexuosa Bernh. ex Steud. Helichrysum arenarium Moench Leibnitzia anandria (L.) Turcz. Leontopodium nanum Hand.-Mazz. Ligularia rumicifolia (Drumm.) S.W.Liu Saussurea tridactyla Sch.-Bip. ex Hook.f. Saussurea gnaphalodes Ostenf. Saussurea gossypiphora D. Don Saussurea graminifolia Wall. Saussurea sp1 Saussurea sp2 Syncalathium kawaguchii (Kitam.) Ling Taraxacum tibeticum Hand.-Mazz. Eritrichium laxum I.M.Johnston Microula sikkimensis (C.B.Clarke) Hemsley Microula tangutica Maxim. Trigonotis rockii I.M.Johnston Arabidopsis himalaica (Edgew.) O.E.Schulz Dontostemon glandulosus (Kar. & Kir.) O.E.Schulz Draba glomerata Royle Solms-laubachia platycarpa (Hook.f. & Thomson) Botschantzev Pegaeophyton scapiflorum (Hook.f. & Thomson) C.Marquand & Airy Shaw Cyananthus sherriffii Cowan Stellaria decumbens Edgewo. Stellaria subumbellata Edgew. Stellaria tibetica Kurz Rosularia alpestris (Kar. & Kir.) Boriss. Rhodiola crenulata (Hook.f. & Thomson) H.Ohba Rhodiola himalensis (D.Don) Fu Rosularia alpestris (Kar. & Kir.) Boriss. Sedum erici-magnusii Fröderstr. Sedum fischeri Raym.-Hamet Euphorbia stracheyi Boiss. Astragalus donianus DC. Astragalus yunnanensis Franch Cicer microphyllum Benth. Oxytropis sp1 Oxytropis sp2 Oxytropis sp3 Gentiana simulatrix C.Marquand Gentiana phyllocalyx C.B.Clarke Gentiana tetrasticha C.Marquand Gentiana urnula Harry Sm. Aste him Crem ell Crem nan Crem tho Crem rho Crep fle Heli are Leib ana Leon nan Ligu rum Saus tri Saus gna Saus gos Saus gra Saus sp1 Saus sp2 Sync kaw Tara tib Erit lax Micr sik Micr tan Trig roc Arab him Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Brassicaceae vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb Dont gla Drab glo Brassicaceae Brassicaceae vascular vascular forb forb Solm pla Brassicaceae vascular forb Pega sca Cyan she Stel dec Stel sub Stel tib Rosu alp Brassicaceae Campanulaceae Caryophyllaceae Caryophyllaceae Caryophyllaceae Crassulaceae vascular vascular vascular vascular vascular vascular forb forb forb forb forb forb Rhod cre Rhod him Rosu alp Sedu eri Sedu fis Euph str Astr don Astr yun Cice mic Oxyt sp1 Oxyt sp2 Oxyt sp3 Gent sim Gent phy Gent tet Gent urn Crassulaceae Crassulaceae Crassulaceae Crassulaceae Crassulaceae Euphorbiaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Fabaceae Gentianaceae Gentianaceae Gentianaceae Gentianaceae vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb 29 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 Geranium robertianum L. Eriophyton wallichii Bentham Lamiophlomis rotata (Benth.) Kudô Marmoritis nivalis (Jacq. ex Benth.) Hedge Allium sikkimense Baker Corydalis cashmeriana Royle Corydalis trifoliolata Franch. Corydalis hendersonii Hemsl. Corydalis inopinata Prain ex Fedde Corydalis melanochlora Maxim. Meconopsis horridula Hook.f. & Thomson Polygala sibirica L. Bistorta vivipara (L.) S.F. Gray Koenigia islandica L. Polygonum campanulatum Hook.f. Rheum spiciforme Royle Primula nivalis Pallas Primula sikkimensis Hook.f. Primula tibetica Watt Aconitum fletcheranum G.Taylor Aconitum orochryseum Stapf. Delphinium brunonianum Royle Ranunculus involucratus Maxim. Ranunculus lobatus Moench Thalictrum alpinum L. Delphinium chungbaense W.T.Wang Potentilla anserina L. Potentilla saundersiana Royle Galium pauciflorum Bunge Saxifraga granulifera H. Sm. Saxifraga punctulata Engl. Lancea tibetica Hooker.f. & Thomson Pedicularis roylei Maxim. Stellera chamaejasme L. Urtica hyperborea Jacquem. ex Wedd. Nardostachys jatamansi DC. Carex atrofusca Sch. Carex moorcroftii Falc Carex praeclara Nelmes Kobresia humilis (C.A.Mey. ex Trautv.) Serg. Kobresia pygmaea C.B.Clarke ex Hook.f. Juncus thomsonii Buchenau Poa pagophila Bor. Stipa koelzii R.R.Stewart Lonicera myrtillus Hook.f. & Thomson Juniperus pingii W.C. Cheng ex Ferre Ephedra gerardiana Wallich ex Stapf. Potentilla eriocarpa Wallich ex Lehmann Potentilla fruticosa L. Potentilla parvifolia Fisch. ex Lehm. Gera rob Erio wal Lami rot Marm niv Alli sik Cory cas Cory tri Cory hen Cory ino Cory mel Meco hor Poly sib Bist viv Koen isl Poly cam Rheu spi Prim niv Prim sik Prim tib Acon fle Acon oro Delp bru Ranu inv Ranu lob Thal alp Delp chu Pote ans Pote sau Gali pau Saxi gra Saxi pun Lanc tib Pedi roy Stel cha Urti hyp Nard jat Care atr Care moo Care pra Kobr hum Kobr pyg Junc tho Poa pago Stip koe Loni myr Juni pin Ephe ger Pote eri Pote fru Pote par 30 Geraniaceae Lamiaceae Lamiaceae Lamiaceae Liliaceae Papaveraceae Papaveraceae Papaveraceae Papaveraceae Papaveraceae Papaveraceae Polygalaceae Polygonaceae Polygonaceae Polygonaceae Polygonaceae Primulaceae Primulaceae Primulaceae Ranunculaceae Ranunculaceae Ranunculaceae Ranunculaceae Ranunculaceae Ranunculaceae Ranunculaceae Rosaceae Rosaceae Rubiaceae Saxifragaceae Saxifragaceae Scrophulariaceae Scrophulariaceae Thymelaeaceae Urticaceae Valerianaceae Cyperaceae Cyperaceae Cyperaceae Cyperaceae Cyperaceae Juncaceae Poaceae Poaceae Caprifoliaceae Cupressaceae Ephedraceae Rosaceae Rosaceae Rosaceae vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular vascular forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb forb graminoid graminoid graminoid graminoid graminoid graminoid graminoid graminoid shrub shrub shrub shrub shrub shrub Paper II Baniya, C.B. Vascular and cryptogam richness in the world’s highest alpine zone, Tibet. (Submitted) II 1 Vascular and cryptogam richness in the world’s highest alpine 2 zone, Tibet 3 Chitra Bahadur Baniya1,2 4 chitra.baniya@bio.uib.no 5 1 Department of Biology, University of Bergen, Allégaten 41, N-5007 Bergen, Norway. 6 2 Central Department of Botany, Tribhuvan University, Kirtipur, Kathmandu, Nepal 7 1 8 9 10 Abstract 11 highest alpine zone in the world, namely in the Tibet/Xizang Autonomous Region (78° 25’ to 12 99° 06’ E, 26° 50’ to 36° 53’ N). The data are based on the published floras for vascular 13 plants, bryophytes and lichens. I have interpolated elevational ranges of each species recorded 14 in the flora between 4900 to 6000 m above sea level into 12 elevational bands of 100m each. 15 A species is assumed to be present at all elevational bands between its lower and upper limits 16 as recorded in the flora. Total richness has been further subcategorized into non-graminoid 17 vascular, graminoid, lichen, total moss, pleurocarpous moss, and acrocarpous moss richness. I 18 applied Generalized Linear Models (GLMs) up to three orders to assess the relationship 19 between species richness and elevation and selected the statistically most appropriate model 20 based on the highest F-value among the significant models. A total of 642 species, 209 21 genera, and 70 families are recorded in the floras from this part of the alpine zone. Flowering 22 plants are represented by 533 species, 153 genera, and 37 families. Total, non-graminoid 23 vascular plant, graminoid, and Compositae species richness showed a significant decreasing 24 quadratic relation with increasing elevation. Total moss, pleurocarpous, and acrocarpous 25 mosses showed a linear declining relationship. Lichen richness showed a unimodal relation 26 with highest richness between 5000-5500 m. The patterns found are both similar and 27 dissimilar to published results from studies using interpolation or direct observations. Scale, 28 environmental heterogeneities, stress, disturbance, and tolerance by individual species may be 29 the likely causes for these patterns. This study explores the elevational richness patterns of vascular and cryptogam species in the 30 31 Keywords: Alpine Zone, Declining Pattern, Tibet, Interpolation, Himalayas, Elevation 32 33 2 34 35 Introduction 36 Alpine life and its biodiversity are of considerable international interest (Körner 2003, Nagy 37 & Grabherr 2009). How life adapted to alpine environments is a very challenging question 38 and probably a key problem after the basic question as to why there are so many species in the 39 tropics. Survival in an alpine zone involves adaptation to severe cold and a short growing 40 season. It is a habitat from where many new species are yet to be discovered and described, 41 and where a greater rate of speciation may take place due to long term isolation, which is 42 generally believed to increase with increasing elevation (Kluge et al. 2006). Latitudinal and 43 altitudinal declines of species richness are established patterns of nature first noticed by 44 Alfred Russel Wallace (1878). However, richness studies at high elevations and high latitudes 45 are comparatively less common than elsewhere (Körner 2003), partly because they are 46 difficult to access and generally have a low number of species. 47 Present-days rates of global warming and associated melting of alpine glaciers are 48 creating open habitats for colonization by different life-forms. These changes are more 49 pronounced in arctic and alpine zones than in other areas. The upward shift of species from 50 lower elevations and the disappearance of alpine species with a narrow niche in lower 51 elevations have already been observed (Klanderud & Birks 2003). There are records of a rapid 52 increase in diversity of alpine zones (Klanderud & Birks 2003, Salick et al. 2009). 53 Colonization normally takes place in such open habitats by dynamic or static equilibrium 54 mechanisms (Brown 1981, Huston 1994). One common belief in community ecology is that 55 there is a certain number of species within a particular space and time, but that the number 56 varies according to the scale of measurement (Palmer 2006). Hence a pattern exists 57 (MacArthur 1965) that changes with space, time, and scale of measurement. One can 58 commonly expect therefore, an elevational decline in species richness when the environment 3 59 is harsh and difficult to colonize. Species richness will be zero (except for some bacteria) if 60 the environment is too harsh for the life-form of interest. 61 The alpine zones of the world have 5% of the terrestrial world’s land area, and support 62 10,000 flowering plant species, 2000 genera, and 100 families (Chapin III et al. 1997, Körner 63 2003). Absent of tree is one of main characteristics of alpine zone but with many definitions 64 of trees, and causes of forming treeline are highly discussed (Körner 2000, 2003, Körner & 65 Paulsen 2004, Miehe et al. 2007, Miehe & Miehe 2000). The Tibet/Xizang Autonomous 66 Region (TAR) is a unique area where the highest known treeline elevation in the Northern 67 Hemisphere (4850 m) occurs (Miehe et al. 2007). The high Asia represents the highest known 68 elevation for treeline at about 4850 m by Juniperus tibetica (Miehe & Miehe 2000) is the 69 highest and the youngest high-altitude Plateau in the world with and average elevation of 70 4900 m (Ni 2000). According to Körner (2003), the alpine area is a life zone between the 71 treeline and the snowline. Thus the Tibetan alpine zone is the highest alpine bioclimatic zone 72 in the Northern Hemisphere. The lower and upper elevation limits of TAR’s alpine zones are 73 debated in the literature. The elevational range of the ‘alpine zone’ in TAR is defined here 74 between 4900-6000 m elevation, the highest position of the treeline (Miehe et al. 2007, Miehe 75 & Miehe 2000) as its lower and the upper limit of alpine zone (5900 m, Birks et al. 2007), to 76 which a further 100 m was added to ensure any change in species richness was captured. 77 TAR consists of about 5766 flowering plant species which accounts about 19% of the 78 flora of China and 2% of the world (Cheng-yih 1983-1987, Barthlott et al. 2005, Birks et al. 79 2007). Besides its highest elevation range, the TAR also has a varied topography which has a 80 great influence on the stratospheric circulation that in turn greatly influences global climate. 81 The TAR has high solar insolation, high UV-B radiation, and is geologically young due to 82 recent tectonic uplift. The present-day vegetation of the TAR has been the result of severe 4 83 human disturbances in the past (Miehe et al. 2009), and/or the result of climate change (Birks 84 et al. 2007). 85 The TAR lies within two big biodiversity ‘hotspots’, the Himalaya (India, Nepal and 86 Bhutan) to the south and western China (Yunnan and Sichuan) (Birks et al. 2007) towards the 87 east and two big ‘coldspots’, the Karakorum Mountain and the Takla Makan desert (Dickoré 88 & Miehe 2002). No biodiversity studies have been published about the alpine zone of the 89 TAR concerning elevational species richness patterns between 4900-6000 m, namely the mid- 90 to high-alpine zone of the TAR. 91 Floras provide the best available documentation of diversity. Thus this study is based 92 on previously documented floras to analyze the elevational species richness patterns of 93 different plant and lichen groups. I aim to evaluate whether the total richness as well as 94 richness for the biological subgroups show a simple monotonically decreasing pattern with 95 increasing elevation. 96 97 98 Material and methods 99 The TAR is located between 78° 25’ to 99° 06’ E and 26° 50’ to 36° 53’ N. The elevation 100 range of TAR lies between 700 m to 8848 m, the so-called ‘roof of the world’. The vegetation 101 of TAR ranges from montane forests, high-cold meadows, high-cold steppes, semi-deserts 102 and high-cold deserts of the subtropical south-east to the areas of the nival north-west (Chang 103 1981). The mean July temperature is 9°C in the west Tibet and 11°C in the east Tibet, the 104 boundary between meadow and steppe closely coincided with the isohyet of 400 mm annual 105 precipitation in the north Tibet and the 500 mm isohyet in the south Tibet. The transition 106 between steppe and desert is approximately at the 100 mm annual precipitation isohyet 5 107 (Chang 1981). The climate of the south-eastern area is mainly controlled by the Asian 108 monsoon and north-western area by the westerlies (Chang 1981). 109 Flora Xizangica (five volumes) (Cheng-yih 1983-1987) for vascular plants, Li (1985) 110 for bryophytes, and Wei and Jiang (1986) for lichens are the main data sources for this study. 111 I interpolated elevational ranges of each species between 4900 to 6000 m above sea level. I 112 divided this elevational range into 12 bands of 100 m each. Data matrices for all alpine 113 species between 4900-6000 m were prepared. Each matrix represents a separate taxonomic 114 group. Presence of a species in the data matrix records the known occurrence of that species, 115 usually documented by collections. A species is assumed to be present at all 100 m bands 116 between its upper and lower elevation limit. For example, if a species, say Arenaria 117 melandrioides is reported as 4200-5020 m asl in the flora, it is assumed that it occurs at all 118 elevation bands considered in this study, namely between the 4900 and 5000-m bands. This 119 method is same as in other interpolation studies from elsewhere in the Himalaya (Grytnes & 120 Vetaas 2002). 121 Species richness here is an estimate of the total number of species or a particular 122 group of species in an elevation band. It is a macro-scale study (gamma diversity, sensu 123 Whittaker 1972) that covers the area between 4900-6000 m of an entire region. Taxonomic 124 ranks below species level were also treated as species as in Grytnes & Vetaas (2002). 125 There are some limitations in this study. I do not discuss endemism, any other 126 diversity status, or particular geographic distributions, because the information regarding to 127 species richness has been taken from published floras that are written in Chinese, of which I 128 have limited understanding. I relay on what generally literature said about east Tibet is wet, 129 well vegetated than dry west Tibet but the centre is mix of between. 130 I first analyzed the data using total species richness as a response variable and 131 elevation as a predictor variable. Total species richness was further subdivided into non- 6 132 graminoid vascular richness, graminoid richness, Compositae (=Asteraceae) richness, lichen 133 richness, total moss richness, pleurocarpous moss richness; acrocarpous moss richness, and 134 liverwort richness. Vascular plants (Pteridophytes, Gymnosperms, and Angiosperms) other 135 than Gramineae, Cyperaceae, and Juncaceae were grouped as non-graminoid vascular 136 richness and latter three were grouped as graminoid richness. I also analyzed the richness 137 pattern shown by species in the most dominant family of vascular plants, namely Compositae, 138 and four groups of bryophytes: total mosses, pleurocarpous mosses, acrocarpous mosses, and 139 liverworts. Liverworts contained very few numbers of species (four) thus did not analyze 140 them separately but they are included in the total richness analysis. I used Generalized Linear 141 Models (GLMs, McCullagh & Nelder 1989) to relate richness to elevation. Since the response 142 variables are counts, I tested the dispersion in the data and they showed overdispersion. Thus, 143 a quasi-Poisson distribution and a logarithmic link were employed following Crawley (2006). 144 Inspection of diagnostic plots between a logarithmic link and an identity link (assuming a 145 normal distribution of errors) also confirmed that a quasi-Poisson with a logarithmic link 146 function was better than a normal distribution and an identity link. I tested up to third-order 147 linear models to model the relationship between species richness and elevation. GLMs using 148 linear, quadratic or cubic polynomials were first tested against each other and then with the 149 null model if the previous was statistically significant. An F- test was used to select the best 150 model (the best model is with the highest F-value among the significant models). The final 151 graphs were based on the best selected model. I used R 2.8.1 statistical package (R 152 Development Core Team 2008) for all analyses. 153 7 154 155 Results 156 A total of 642 species, from 209 genera and 70 families were represented in the flora in the 157 elevation range 4900-6000 m of the alpine zone. The flowering plants contained 533 species, 158 153 genera and 37 families which is 9% of the total flowering plants of Tibet. 159 As dataset, Delphinium brunonianum Royle is the flowering plant at the highest 160 elevation (6000 m) and the lichen Lecidea diducens (Nyl.) Th. Fr. occurs at 6100 m. 161 Compositae is the largest angiosperm family with 87 species, followed by Gramineae with 43 162 species and Caryophyllaceae with 39 species. Saussurea simpsoniana (Field. & Gardn.) 163 Lipsch. (5750 m) and S. gnaphalodes Ostenf. (5700 m) were the highest growing Compositae 164 in Tibet. Interestingly S. gnaphalodes is recorded at the highest elevation (6400 m) of any 165 vascular plant in the world (Körner 2003), on the north flank of Mount Everest, i.e., Nepal. 166 Similarly, Kobresia prainii Kük. (5600 m) and Littledalea przevalskyi Tzvelev (5700 m) are 167 the two highest recorded graminoids in Tibet. 168 Lichens are represented by 21 species, 13 genera and 13 families. Sporastatia asiatica 169 H. Magn. (6000 m) and Lecidea auriculata var. diducens (6100 m) are the two highest 170 recorded crustose lichen and the highest fruticose lichens are the rock dwelling Lethariella 171 flexuosa (Nyl.) Wei and Jiang (1986) (5800 m) and a soil dweller Thamnolia vermicularis f. 172 qomolangmana Wei & Jiang (5510-5700 m). 173 Mosses are represented by 80 species, 35 genera and 14 families. Among them 70 174 species are acrocarpous, nine are pleurocarpous, and one is a Sphagnum. Tortula desertorum 175 Broth. (5800 m) and Bryum gossypinum Gao Chien and Zhang Guang-Chu (5800 m) are the 176 two highest recorded acrocarps; Brachythecium brotheri Paris (5600 m) and Haplocladium 177 microphyllum Brotherus (5550 m) are the two highest recorded pleurocarps. 8 178 179 180 181 Liverworts (not shown) are represented only by four species, four genera and two families. Anastrepta orcadensis (Hook.) Schiffn. (5100 m) is the highest recorded hepatic. Pteridophytes and Gymnosperms are each represented by two genera, species and families. 182 Total species richness ranged from three to 540 species with an average of 195 species 183 per 100 m elevational band. All showed a significant quadratic relationship with elevation 184 (Figures 1A, B, C, D, and E) except for total moss, pleurocarpous moss, and acrocarpous 185 moss richness (Figures 1F, G and H) which showed a significant linear relationship with 186 elevation. The lowest two elevation bands (4900-5000 m and 5000-5100 m) had an almost 187 equal and the highest number of species and there is a continuous decline thereafter in total 188 richness (Figure 1A): a pattern seen in other groups such as non-graminoid vascular species 189 richness (Figure 1B), graminoid species richness (Figure 1D), moss richness (Figure 1F), 190 pleurocarpous moss richness (Figure 1G), and acrocarpous moss richness (Figure 1H). A 191 continuous decline from 4900 m was found for Compositae species richness (Figure 1C). A 192 unimodal relation with highest richness between 5000-5500 m was found for lichen richness 193 (Figure 1E). 194 9 195 196 197 198 199 200 Figure1. The relationships between species richness and elevation, with the fitted polynomial GLM regression model superimposed: A. All species; B. All non-graminoid vascular plant species; C. Compositae species; D. Graminoids species; E. All lichens; F. All mosses; G. Pleurocarpous mosses; and H. Acrocarpous mosses. Note the major differences in the scale of the ordinate. 201 202 10 203 204 Discussion 205 The patterns showed by the alpine flora of the TAR are in line with the general patterns as 206 expected (Körner 2003), as well as with results from other high elevation interpolation studies 207 (Grytnes & Vetaas 2002, Bhattarai et al. 2004). However, these patterns are not the same for 208 all biological groups and nor is the shape of their declines. Non-graminoid, graminoid, 209 Compositae, total moss, pleurocarpous moss and acrocarpous moss richness show a similar 210 elevational declining pattern but the lichen richness pattern differs. This latter one diverges 211 from the general pattern and shows a unimodal quadratic relationship. The pattern of decline 212 especially towards higher elevations is gentle like an asymptote rather than a sharp decline 213 that has been found in interpolation studies elsewhere (e.g., Grytnes & Vetaas 2002, Bhattarai 214 et al. 2004) or a step-wise decline in an empirical study (Grabherr et al. 1995). 215 One possible reason behind these different elevational patterns among biological 216 groups could be due to the differential adaptational features of individual species against very 217 harsh and strong environmental filters (Agakhanyantz & Breckle 1995, Klimes 2003, Körner 218 2003) that almost certainly apply to alpine species in the TAR. Such environmental filters 219 may represent very high differences in ambient temperature (far below to freezing 220 temperature at night and relatively high during day), short growing season, different geometry 221 of each landscape, steep environmental gradients, below-freezing soil temperature, low soil 222 nutrients, low precipitation, periods of lack of humidity, substrate differences, and increased 223 soil disturbances. Besides these ecological factors, artifacts caused by interpolation: such as 224 mid-domain effect (Grytnes & Vetaas 2002), poor collection and low representation of 225 samples from high-elevation, inaccessible landscapes may also cause the observed 226 relationships (Colwell & Hurtt 1994, Grytnes & Vetaas 2002). The finding of similar patterns 227 for total, non-graminoid vascular, graminoid, Compositae, total moss, pleurocarpous moss, 228 and acrocarpous moss suggests that the patterns are less likely to result from a poor 11 229 representation of collections or be artifacts from interpolation. However, the unimodal lichen 230 richness pattern is probably due to many reasons. Biologically, some lichens proved their 231 capacity to survive at both simulated (de Vera et al. 2004) and real (Sancho et al. 2007) outer 232 space conditions. Non-biological stress factors such as periods of lack of humidity, strong 233 UV-B and solar radiation, below freezing air and soil temperature strongly impact the 234 distribution of vascular plant life but may not necessarily be destructive to lichens. A 235 unimodal pattern may also arise after artefacts such as interpolation and/or biased collections. 236 The range of the maximum lichen richness in this model lies in wider (5000-5500 m) scale 237 thus would be less likely of having mid-domain effect or artefact caused by interpolation 238 (Grytnes & Vetaas 2002). Similarly, there were only four species of liverworts collected up to 239 5100 m (pattern not shown). Liverworts are extremely rare at high elevations above 4800 m in 240 the Tibetan Himalaya (Birks HJB unpublished observation) but the other taxa show a fairly 241 good representation of both lower and higher elevation. This signifies the sharp decrease in 242 moisture with increasing elevation. The finding of Proctor et al. (2009) also support that 243 alpine hepatics distribution depend not on temperature but on soil humidity or snow cover. 244 The declining richness pattern for TAR can be explained in term of an energy-related 245 hypothesis (Brown 1981, Wright 1983). This hypothesis is mainly proposed for woody plant 246 richness towards the tropics, and subtropics where there are generally more stable 247 environments. Both temperature and precipitation decreased with elevation, but would be as 248 defined in the TAR’s high alpine zone due to high instability and harshness. Other hypotheses 249 related to disturbance (Huston 1994), stress and stability (Begon et al. 1990) are more 250 plausible explanations for this decline in richness than energy hypothesis. This alpine zone 251 receives strong solar radiation that warms the landscape and evaporates water in high 252 amounts. Plants occupy microhabitats in the high alpine zone which may have relatively flat 253 slope, more soil or shelter amongst rocks. Some high-alpine plants have special structures 12 254 such as silky hairs all over the plant such as with Compositae (Saussurea spp.) to avoid 255 overheating and decrease transpiration (Gauslaa 1984) or water repellence and reflection of 256 short peaks of high radiation (Yang et al. 2008). Other adaptational features such as succulent 257 leaves for CAM photosynthesis (Sedum and Rhodiola spp.) are common in desert plants 258 (Körner 2003) as are cushion growth forms (e.g., Androsace, and Thylacospermum spp.). 259 High doses of UV-B radiation are received by these low-latitude mountains (Willis et al. 260 2009), which lichens can screen by having a high concentration of melanin, and/or parietin in 261 the outermost body cells (Gauslaa & Ustvedt 2003). Cloud formation, strong wind, and high 262 moisture content in the air are common towards mountain summits that may create low air 263 pressure. Lack of multiple layers in the vertical stratification in the alpine vegetation creates a 264 high amount of convective long-wave radiation during the night which cools air far below 265 freezing temperature (Körner 2003). Thus strong selection will take place among species and 266 species not adapted to this high harsh environment will be eliminated, thereby causing a 267 decline in species richness with increased elevation. The monotonic declines of Compositae, 268 total moss, pleurocarpous moss, and acrocarpous moss richness adds further support for 269 selective microhabitats in this extreme single-layered ecological zone. Damp ground with 270 scattered Kobresia prainii would be suitable habitats for low-growing pleurocarpous mosses, 271 whereas stone outcrops surrounded by humid air will be appropriate habitats for acrocarpous 272 mosses and lichens. Lichens show a higher resistance to extreme environments than other 273 species and thus they have a higher richness at higher elevations than other species. Fast 274 drying and wetting cycles at this elevation may also favour their high richness. 275 There is a long tradition of pastoralism in Tibet and there could easily be grazing 276 pressure from wild and domestic animals (yak, sheep, horse, goat) in the Tibetan highlands 277 (Miehe et al. 2009). But species richness can increase after foraging (Smith & Foggin 1999) 278 depending on the intensity of grazing, environmental conditions, herbivore types, and 13 279 elevation. At high elevations, a small amount of grazing may have a high impact due to the 280 harsh conditions, but would not be such a problem for plants growing in lower, more 281 congenial environments. Herbivores shape plant communities through their selective foraging 282 behavior (Evju et al. 2009). Alpine rangeland dominated by grasslands and steppes may 283 indicate strong selection. Also the harvesting of large amounts of high-altitude medicinal 284 plants from high elevations of Tibet is important (Salick et al. 2009). However, these factors 285 are associated with local to landscape scales of disturbances which have no or very little role 286 at macro-scale species richness patterns. Climate related large scale variables such as 287 glaciation history, and/or present climate change may define the present pattern which is 288 consistent with generalization made by Whittaker et al. (2001). 289 The alpine flora of the world is nested within the regional flora (Klimes 2003, Körner 290 2003). This mid- to high-alpine Tibetan flora has far more (533 species) than the expected 291 number of higher plant species (200-280) for the individual regional alpine floras (Körner 292 2003) but below (37) the average number of families (40). This number would be even much 293 higher if the lower alpine zone is included, because of richness increased with increase of 294 area. Results for the number of angiosperm species and their families for the Hokkaido alpine 295 zone of Japan are 225 species and 45 families (Tatewaki 1968). The 404 species of vascular 296 alpine plants in Ladakh, north-west Himalaya (Klimes 2003) is closer to the Tibet figure. The 297 alpine flora of the TAR represents an accumulated flora (large area) of different regions, each 298 of which may differ at the species level and may be similar at their family level, which may 299 account for the high number of species. Another likely explanation for this high diversity is 300 that the TAR species pool is high due to the geographical proximity and similar environment 301 to the alpine zones of two large biodiversity ‘hotspots’: the Himalaya and the Yunnan and 302 Sichuan provinces (Barthlott et al. 2001, Barthlott et al. 2005, Birks et al. 2007). 14 303 The rate of reduction in total richness per 100 m elevation is 49 species and for higher 304 plants is 43 species per 100 m, which are both higher than the mean value (40 spp. per 100 m) 305 for individual alpine regions (Körner 2003). This higher reduction rate may be a feature of 306 this unique region with its steep elevational gradient of environmental harshness which gives 307 higher instability but also provides more opportunities for colonization. The increasing plant 308 diversity found in the alpine zone of the Scandes (Klanderud & Birks 2003) and within 309 eastern Tibet (Salick et al. 2009) were derived from a multi-temporal comparison, whereas 310 the declining species richness in the TAR alpine zone is only the species number difference 311 between 100 m elevational bands. 312 These smooth decline richness patterns may also be connected to species-area 313 relationships. Higher areas at lower elevation may inflate species richness simply due to the 314 species-area relationship (Kluge et al. 2006). Environmental heterogeneity may vary as a 315 function of elevation (Palmer 2006) and local endemism may increase at higher elevations 316 specially at landscape scale mountain range (Birks et al. 2007). Both environmental 317 heterogeneity and local endemism will be average out and climate influences strongly at 318 macro-scale. Thus species richness declined. This is most likely true to all groups besides 319 lichen richness that showed high richness at intermediate elevation, and almost similar for all 320 groups towards higher elevation. 321 322 323 324 Conclusion 325 species richness pattern of vascular and cryptogam species. A continuous monotonic decline 326 with total, all groups of vascular, and moss richness is as predicted but unimodal with lichen 327 richness is different. The richness patterns found differ from earlier interpolation studies by 328 having gentler and continuous declines with increasing elevation as well as from observed I conclude that there are more similarities than dissimilarities in this mid- to high-alpine 15 329 species-richness studies done elsewhere. Besides the spatial scale of study, environmental 330 heterogeneities such as stress, disturbance, and tolerance by individual species are possible 331 causes of these patterns. The TAR is rich in biodiversity and there is much scope for research 332 in this high elevation range. This study can provide a basis for future climate-change research 333 such as how climate change can impact the flora in the highest alpine zone in the world. 334 335 336 Acknowledgements 337 I am very grateful to John Birks for providing me with the necessary literature and for 338 substantial help with the manuscript. Thanks also Torstein Solhøy, Mike W. Palmer, Yngvar 339 Gauslaa and Bjørn Arild Hatteland for comments and valuable suggestions. I am also thankful 340 to Cathy Jenks for language correction and comments. 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 References Agakhanyantz, O.E. & Breckle, S.W. (1995) Origin and evolution of the mountain flora in middle Asia and neighbouring mountain regions. In: Chapin III, F.S. & Körner, C.(eds) Arctic and alpine biodiversity: patterns, causes and ecosystem consequences. Ecological studies. Springer, Berlin, Heidelberg, New York. Barthlott, W., Mutke, J., Rafiqpoor, D., Kier, G. & Kreft, H. (2005) Global centers of vascular plant diversity. Deutsche Akad Naturforscher Leopold. Barthlott, W., Schmit-Neuerburg, V., Nieder, J. & Engwald, S. (2001) Diversity and abundance of vascular epiphytes: a comparison of secondary vegetation and primary montane rain forest in the Venezuelan Andes. Plant Ecology, 152, 145-156. Begon, M., Harper, J.L. & Townsend, C.R. (1990) Ecology: individuals, populations and communities. (2nd ed.). Blackwell Scientific Publication, London. Bhattarai, K.R., Vetaas, O.R. & Grytnes, J.-A. (2004) Fern species richness along a central Himalayan elevational gradient, Nepal. Journal of Biogeography, 31, 389-400. Birks, H.J.B., Birks, H.H., Everson, J., Jans, H., Thorne, D. & M., T. (2007) The AGS in Tibet 2005. The Alpine Gardener: Tibet Special Issue. The Alpine Garden Society, AGS Centre, Avon Bank, Pershore, UK. pp. 289-349. Brown, J.H. (1981) Two decades of homage to Santa Rosalia: toward a general theory of diversity. American Zoologist, 21, 877-888. Chang, D.H.S. (1981) The vegetation zonation of the Tibetan Plateau. Mountain Research and Development, 1, 29-48. 16 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 Chapin III, F.S., Walker, B.H., Hobbs, R.J., Hooper, D.U., Lawton, J.H., Sala, O.E. & Tilman, D. (1997) Biotic control over the functioning of ecosystems. Science, 277, 500-504. Cheng-yih, W., ed (1983-1987) Flora Xizangica (Vol. I-V). Science Press, Beijing (In Chinese). Colwell, R.K. & Hurtt, G.C. (1994) Nonbiological gradients in species richness and a spurious Rapoport effect. The American Naturalist, 144, 570-595. Crawley, M.J. (2006) Statistics: an introduction using R. John Wiley & Sons, Ltd, London. de Vera, J.P., Horneck, G., Rettberg, P. & Ott, S. (2004) The potential of the lichen symbiosis to cope with the extreme conditions of outer space II: germination capacity of lichen ascospores in response to simulated space conditions. Advances in Space Research, 33, 1236-1243. Dickoré, W.B. & Miehe, G. (2002) Cold spots in the highest mountains of the world diversity patterns and gradients in the flora of the Karakorum. In: Körner, C. & Sphen, E.M.(eds) Mountain biodiversity a global assessment. Parthenon, New York. pp. 129147. Evju, M., Austrheim, G., Halvorsen, R. & Mysterud, A. (2009) Grazing responses in herbs in relation to herbivore selectivity and plant traits in an alpine ecosystem. Oecologia, 161, 77-85. Gauslaa, Y. (1984) Heat resistance and energy budget in different Scandinavian plants. Holarctic Ecology, 7, 5-78. Gauslaa, Y. & Ustvedt, E.M. (2003) Is parietin a UV-B or a blue-light screening pigment in the lichen Xanthoria parietina? Photochemical & Photobiological Sciences, 2, 424432. Grabherr, G., Gottfried, M., Gruber, A. & Pauli, H. (1995) Patterns and current changes in alpine plant diversity. In: Chapin III, F.S. & Körner, C.(eds) Arctic and alpine biodiversity: patterns, causes and ecosystem consequences. Springer-Verlag, Berlin. pp. 167-181. Grytnes, J.A. & Vetaas, O.R. (2002) Species richness and altitude: a comparison between null models and interpolated plant species richness along the Himalayan altitudinal gradient, Nepal. American Naturalist, 159, 294-304. Huston, M.A. (1994) Biological diversity: the coexistence of species on changing landscapes. Cambridge University Press, Cambridge. Jiang-Chun, W. & Yu-Mei, J., eds (1986) Lichens of Xizang. Science Press, Beijing (In Chinese). Klanderud, K. & Birks, H.J.B. (2003) Recent increases in species richness and shifts in altitudinal distributions of Norwegian mountain plants. The Holocene, 13, 1-6. Klimes, L. (2003) Life-forms and clonality of vascular plants along an altitudinal gradient in East Ladakh (NW Himalayas). Basic and Applied Ecology, 4, 317-328. Kluge, J., Kessler, M. & Robert, R.D. (2006) What drives elevational patterns of diversity? A test of geometric constraints, climate and species pool effects for pteridophytes on an elevational gradient in Costa Rica. Global Ecology and Biogeography, 15, 358-371. Körner, C. (2000) Why are there global gradients in species richness? Mountains might hold the answer. Trends in Ecology & Evolution, 15, 513-514. Körner, C. (2003) Alpine plant life: functional plant ecology of high mountain ecosystems. (2nd ed.). Springer Verlag, Berlin. Körner, C. & Paulsen, J. (2004) A world-wide study of high altitude treeline temperatures. Journal of Biogeography, 31, 713–732. Li, X.J., ed (1985) Bryoflora of Xizang. Science Press, Beijing (In Chinese). MacArthur, R.H. (1965) Patterns of species diversity. Biological Reviews, 40, 510-533. 17 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 McCullagh, P. & Nelder, J.A. (1989) Generalized linear models. Chapman & Hall, London. Miehe, G. & Miehe, S. ( 2000) Comparative high mountain research on the treeline ecotone under human impact. Erdkunde, 54, 34-50. Miehe, G., Miehe, S., Kaiser, K., Reudenbach, C., Behrendes, L., La Duo & Schlütz, F. (2009) How old is pastoralism in Tibet? An ecological approach to the making of a Tibetan landscape. Palaeogeography, Palaeoclimatology, Palaeoecology, 276, 130147. Miehe, G., Miehe, S., Vogel, J., Co, S. & La Duo (2007) Highest treeline in the northern hemisphere found in southern Tibet. Mountain Research and Development, 27, 169173. Nagy, L. & Grabherr, G. (2009) The biology of alpine habitats. Oxford University Press. Ni, J. (2000) A simulation of biomes on the Tibetan Plateau and their responses to global climate change. Mountain Research and Development, 20, 80-89. Palmer, M.W. (2006) Scale dependence of native and alien species richness in North American floras. Preslia, 78, 427-436. Proctor, M.C.F., Oliver, M.J., Wood, A.J., Alpert, P., Stark, L.R., Cleavitt, N.L. & Mishler, B.D. (2009) Desiccation-tolerance in bryophytes: a review. The Bryologist, 110, 595621. R Development Core Team (2008) R: A language and environment for statistical computing version 2.8.1. R Foundation for Statistical Computing, Vienna, Austria. Salick, J., Fang, Z. & Byg, A. (2009) Eastern Himalayan alpine plant ecology, Tibetan ethnobotany, and climate change. Global Environmental Change, 19, 147-155. Sancho, L.G., de la Torre, R., Horneck, G., Ascaso, C., de los Rios, A., Pintado, A., Wierzchos, J. & Schuster, M. (2007) Lichens survive in space: results from the 2005 LICHENS Experiment. Astrobiology, 7, 443-454. Smith, A.T. & Foggin, M.J. (1999) The plateau pika (Ochotona curzoniae) is a keystone species for biodiversity on the Tibetan plateau. Animal Conservation, 2, 235-240. Tatewaki, M. (1968) Distribution of alpine plants in the world. In: Wright, H.J. & Osburn, W.H.(eds) Arctic and alpine environments. Indiana University Press, Bloomington. pp. 119-136. Wallace, A.R. (1878) Tropical nature and other essays. (New ed.). MacMillan and Co, London. Whittaker, R.H. (1972) Evolution and measurement of species diversity. Taxon, 21, 213-251. Whittaker, R.J., Willis, K.J. & Field, R. (2001) Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography, 28, 453-470. Willis, K.J., Bennett, K.D. & Birks, H.J.B. (2009) Variability in thermal and UV-B energy fluxes through time and their influence on plant diversity and speciation. Journal of Biogeography, 36, 1630-1644. Wright, D.H. (1983) Species-energy theory: an extension of species-area theory. Oikos, 41, 496-506. Yang, Y., Körner, C. & Sun, H. (2008) The ecological significance of pubescence in Saussurea medusa, a high-elevation Himalayan woolly plant. Arctic, Antarctic, and Alpine Research, 40, 250-255. 461 18 Paper III Baniya, C.B., Solhøy, T., Gauslaa, Y. & Palmer, M.W. 2010. The elevation gradient of lichen species richness in Nepal. The Lichenologist, 42 (1): 83-96. III The Lichenologist 42(1): 83–96 (2010) doi:10.1017/S0024282909008627 © British Lichen Society, 2009 The elevation gradient of lichen species richness in Nepal Chitra Bahadur BANIYA, Torstein SOLHØY, Yngvar GAUSLAA and Michael W. PALMER Abstract: This study of elevation gradients of lichen species richness in Nepal aimed to compare distribution patterns of different life-forms, substratum affinities, photobiont types, and Nepalese endemism. Distribution patterns of lichens were compared with elevational patterns shown by a wide range of taxonomic groups of plants along the Nepalese Himalayan elevational gradient between 200–7400m. We used published data on the elevation records of 525 Nepalese lichen species to interpolate presence between the maximum and minimum recorded elevations, thereby giving estimates of lichen species richness at each 100-m elevational band. The observed patterns were compared with previously published patterns for other taxonomic groups. The total number of lichens as well as the number of endemic species (55 spp.) showed humped relationships with elevation. Their highest richness was observed between 3100–3400 and 4000–4100m, respectively. Almost 33% of the total lichens and 53% of the endemic species occurred above the treeline (>4300m). Non-endemic richness had the same response as the total richness. All growth forms showed a unimodal relationship of richness with elevation, with crustose lichens having a peak at higher elevations (4100–4200m) than fruticose and foliose lichens. Algal and cyanobacterial lichen richness, as well as corticolous lichen richness, all exhibited unimodal patterns, whereas saxicolous and terricolous lichen richness exhibited slightly bimodal relationships with elevation. The highest lichen richness at mid altitudes concurred with the highest diversity of ecological niches in terms of spatial heterogeneity in rainfall, temperature, cloud formation, as well as high phorophyte abundance and diversity implying large variation in bark roughness, moisture retention capacity, and pH. The slightly bimodal distributions of saxicolous and terricolous lichens were depressed at the elevational maximum of corticolous lichens. Key words: altitude, endemism, Himalaya Introduction The Himalayas are the highest mountain range in the world and show a rich diversity of eco-climatic zones. In such a landscape, elevational gradients are of particular interest in applied as well as in theoretical ecology (Wolf 1993). The pattern of changes in species-richness with elevation characterizes C. B. Baniya (corresponding author) and T. Solhøy: Department of Biology, University of Bergen, Allégaten 41, P.O. Box 7803, N-5020 Bergen, Norway. Email: chitra.baniya@bio.uib.no Current address: Central Department of Botany, Tribhuvan University Kirtipur, Kathmandu, Nepal. Y. Gauslaa: Department of Ecology and Natural Resource Management, Norwegian University of Life Sciences, P.O. Box 5003, NO-1432 Ås, Norway. M. W. Palmer: Department of Botany, Oklahoma State University, 104 LSE Stillwater OK 74078 USA 405-744-7717. the vegetation in a simple but powerful way. For example, Yoda (1967) found gradually decreasing tree species richness with increasing altitude. Later, Hunter & Yonzon (1993) reported a similar pattern for Nepalese birds and mammals. A number of studies have produced macro-scale elevational gradients for the whole of Nepal, using interpolation methods on data from secondary sources. Several groups of organisms have been studied and exhibit maximum richness at group-specific intermediate elevations, referred to as a unimodal, or humped, elevation pattern. For example, Himalayan vascular plants showed their highest total richness at altitudes between 1500 to 2500m and their highest endemic richness at 4000m (Grytnes & Vetaas 2002; Vetaas & Grytnes 2002), while fern richness peaked at 2000m (Bhattarai et al. 2004). However, orchids 84 THE LICHENOLOGIST showed their maximum richness at 1600m whereas the highest orchid endemic richness was bimodal with two separate peaks at 1200 and 3300m (Acharya 2008). Liverworts and mosses had their maximum richness at 2800 and 2500m, respectively, and the peak for endemic liverwort richness occurred at 3300m (Grau et al. 2007). Relatively few studies on elevational richness patterns for lichens have been undertaken (Kessler 2000; Bruun et al. 2006; Grytnes et al. 2006) and among terrestrial photosynthetic organisms they are the only major group that has not yet been investigated in this respect in Nepal. They are among the most successful organisms in extreme environments such as cold arctic and alpine environments where few other plants can grow (e.g. Schroeter et al. 1994; Kappen et al. 1996). Lichens also show a high diversity as eipiphytes and thus may benefit from high diversity of trees and shrubs with species-specific bark chemistry, texture and stability (Kessler 2000; Bruun et al. 2006; Grytnes et al. 2006). This paper describes and interprets the Himalayan elevational distribution of lichen species richness in Nepal. Material and Methods Study area Nepal is a mountainous country in the central Himalayas between 26°22# to 30°27# N and 80°04# to 88°12# E with an area of 147 181 km2. The elevation ranges from 60m above sea level to 8848m (Mt. Everest). The country is surrounded by the Indo-Gangetic Plain in the east, west and south, and the Tibetan Plateau towards the north. Parallel mountain ranges run southeast and north-west dissecting the country into northsouth running river valleys that are associated with fertile plains, inner valleys, lakes, and the outer flood plains. The Indian monsoon beginning in May and ending in September is the main source of precipitation (more than 80%). Monsoon rain forms after condensation of water vapour arising from the Bay of Bengal, and consequently there is a spatial gradient of monsoon rainfall decreasing from east to west. Precipitation also occurs in winter formed from central Asian water vapour and diminishes eastwards. Physiographically, Nepal has been divided into five zones (Hagen 1969; Upreti 1999). 1) A 30–40km wide southernmost plain is called Terai between 60 and 300m. It is the most fertile, densely populated zone with a tropical climate (between 22–27°C during winter and above Vol. 42 35°C during the warmest summer months); it is mainly cultivated but has still some tropical forests. 2) The Siwalik located from 700 to 1500m in the foothills of the Mahabharat Range. 3) The Mahabharat Range lies between 1500 and 3000m; it belongs to the Midlands or lesser Himalaya and includes the principal inner valleys such as Kathmandu and Pokhara. The climate is mild throughout the year and receives the highest annual rainfall (above 4000mm) south of Annapurna, central Nepal. It includes the lower cloud zone at 2000 to 2500m (Miehe 1989, 1990). 4) The Greater-Himalaya extends above 2700 m. Here the climate varies from dry summers to cold winters with snow, and subzero temperatures on the mountain tops. This zone includes the uppermost limit of cloud-zone forest in Nepal at 4000m (Miehe 1989, 1990). 5) Trans-Himalaya lies towards the northern side of the Greater-Himalaya with elevations above 3000m. It is in the rain-shadow zone with an average annual rainfall between 200 and 400mm. Vegetation is one of the prime determinants of lichen diversity and distribution (Awasthi 2007). In Nepal, vegetation has been studied and classified into 8 elevational forest zones (Yoda 1967; Stainton 1972, 2001; Dobremez & Jest 1969; Dobremez 1976; Miehe 1982, 1989). 1) Tropical forest between 60 and 1000 m (Stainton 1972) dominated by Shorea robusta Gaertn., Dalbergia sissoo Roxb., Adina cordifolia (Willd. ex Roxb.) Benth. & Hook. f. ex Brandis, Terminalia spp., Lagerstroemia spp., Michelia champaca L., and Bombax ceiba L. 2) Subtropical forest occurring between 1000 to 2000m with Pinus roxburghii Sarg. as the dominant species on south-facing slopes in the Midlands. 3) Lower temperate broad-leaved forest at 1700–2400m in the east and 2000–2700m in the west. Dominant trees include Alnus nitida (Spach.) Endel., Castanopsis tribuloides A. DC., Castanopsis hystrix A. DC., Lithocarpus pachyphylla (Kurz.) Rehder., and Quercus spp. Moister slopes between 1700–2200m are dominated by Cinnamomum spp. and represent the lower temperate mixed broadleaved forest. Upper temperate broad-leaved forests are represented by various types. Drier southern slopes at 2200–3000m are dominated by Schima wallichii (DC.) Korth., Castanopsis indica (Roxb.) Miq., and Quercus semecarpifolia J. E. Smith. Moist, north and west slopes between 2500–3500m from central to east Nepal are dominated by Acer spp., Rhododendron spp., Aesculus spp. and Juglans spp. 4)Temperate coniferous forests between 2000 and 3700m dominated by Pinus wallichiana A. B. Jack. Other conifers include Cedrus deodara (Roxb.) G. Don, Picea smithiana (Wall.) Boiss., Juniperus indica Bertol., Larix himalaica W. C. Cheng & L. K. Fu, Larix griffithiana Carriere, Cupressus torulosa D. Don, and Tsuga dumosa (D. Don) Eichler. 5) Sub-alpine forests occurring between 3000 and 4100m (Stainton 1972) dominated by Betula utilis D. Don, Abies spectabilis (D. Don) Mirb., and Rhododendron spp. This forest includes the climatic treeline at 4100–4300m. 6) Open, low-alpine shrub communities including Caragana spp., Lonicera spp., Rosa spp., Sophora spp., Rhododendron anthopogon D. Don, R. lepidotum Wall. ex G. Don, Ephedra gerardiana Wall. ex C. A. Meyer, and Hippophae tibetana Schldl. 7) Mid-alpine zone with dwarf-shrubs 2010 Elevation gradients of lichens in Nepal—Baniya et al. but largely dominated by herb communities. 8) Nival zone begins above the mid-alpine zone with more lichens and fewer mosses and angiosperms. Within the nival zone, there are vascular plants of open wind exposed areas as well as those tolerant of snow-lie. Data sources The main source of data for this study is the Lichens of Nepal (Sharma 1995). However, many species reported in Sharma (1995) lack elevation data and since then many new species have been described. Other relevant and easily accessible literature as well as the recent keys for both microlichens and macrolichens of Nepal, India and Sri Lanka by Awasthi (1991; 2007) have also been checked. In addition, the following literature on Nepalese lichens (Lamb 1966; Poelt 1966a, b, 1974; Bystrek 1969; Poelt & Reddi 1969; Abbayes 1974; Jahns & Seelen 1974; Kurokawa 1974; Mitchell 1974; Schmidt 1974; Vězda & Poelt 1975; Hellmich & Poelt 1977; Awasthi 1986; Vitikainen 1986; Awasthi & Mathur 1987; Miehe 1990; Awasthi 1991, 2007; Esslinger & Poelt 1991; Poelt & Hinteregger 1993; Sharma 1995; Baniya 1996; Jørgensen 2001) have provided information on the elevation ranges of Nepalese lichens. These sources have provided elevational distribution data for 525 species, including 55 Nepalese endemics (Appendix 1). Among the 525 species, 172, including 29 endemics, are reported from above the treeline (> 4300m). In addition, data about life-forms (crustose, foliose and fruticose), photobiont types (cyanobacteria and green algae) and five substratum types (tree bark, wood, mosses, rocks and soil) were obtained. Tripartite lichens were treated as cyanobacterial. There were 50 species of lichens for which elevational data or other important information are either incomplete or lacking; these species were excluded from the analysis. Lignicolous and muscicolous substrata had been reported for a few species (5 and 17, respectively) and were thus also excluded during the analyses involving substrata. Lichen nomenclature follows Awasthi (1991; 2007). The altitudinal range of lichens in Nepal, 200– 7400m, was divided into 73 bands of 100m each and a complete data matrix for all species was assembled. Presence of a species indicates that the species occurs in, or has been collected in the past, from that elevation band and absence means either that the species does not occur or has previously not been collected from that elevation. A species is assumed to be present in all possible 100m bands between its upper and lower elevation limits. For example, a lichen that has elevational occurrences between 210 to 451m in the literature falls between the 200- and 500-m bands. A list of the lichens included in this study together with their elevation ranges is provided in Appendix 1. Species richness is an estimate of the total number of lichen species occurring in each 100m elevation band. This is a macro-scale study (gamma diversity, sensu Whittaker 1972) that covers the entire elevational range of Nepal. Fifty-six taxa, varieties, forms or subspecies, were treated as species (Appendix 1). 85 Data analysis Patterns related to total lichen species richness and their sub-sets (life forms, algal partners, and substrata) as responses and their elevations as a predictor variable were extracted by using a cubic smooth spline (s) within the framework of Generalized Additive Models (GAM, Hastie & Tibshirani 1990; Heegaard 2004) with a default of c. 4 degrees of freedom. Response variables are counts; thus, the variance changes with the mean and negative predictions are meaningless. We found overdispersion in our data. We thus applied a Quasi-poisson family error distribution with a logarithmic link function (Crawley 2006). We confirmed our assumption of normal distribution of error after the Q-Q diagnostic plots plotted against residuals. The change in deviance follows the F-distribution. We used R 2.7.0 (R Development Core Team 2008) to analyse our data and smoothers were fitted with library GAM (Hastie & Tibshirani 1990). GAM was used because it is a non-parametric approach that does not make a priori assumptions about the species-elevational relationship. One of the main purposes of this study is to contrast observed lichen richness patterns with other studies. Biogeographic data have biases and constraints, such as a focus on specific taxa and/or a restriction to easily accessible landscapes. However, these problems are shared among data sets with various taxonomic groups (Grytnes & Vetaas 2002; Vetaas & Grytnes 2002; Bhattarai et al. 2004; Grau et al. 2007). Thus, a comparison of empirical patterns among available studies is possible. Results The 525 lichen species recorded (Appendix 1) represented 40 families and 121 genera. Among these lichens, 35·4% were crustose, 46·3% were foliose and 18·3% were fruticose. A total of 12% had a cyanobacterial photobiont and 88% had a green algal photobiont. Lichens endemic to Nepal (n=55) represented 10% of the total Nepalese lichen flora. There were 172 species reported from above the treeline (R 4300m) and 29 of these were endemics. Carbonea vorticosa in Nepalese Himalaya at 7400m was the world’s highest reported lichen. Heterodermia pseudospeciosa represented one of the lichens occurring at the lowest elevation ranges in Nepal (150–2100m). The total lichen species richness showed a unimodal relationship with elevation. The maximum modelled total richness occurred at 3100–3400m (Fig. 1A), whereas the observed maximum richness (144 species) 86 THE LICHENOLOGIST Vol. 42 F. 1. Relationship between elevation and lichen species richness in Nepal. A, total lichen species richness; B, total endemic lichen species richness. The fitted regression lines represent statistically significant (P % 0·001) smooth spline (s) after using GAM with approximately 4 degrees of freedom. was at approximately 4000m. Endemic lichens also had a unimodal relationship with elevation, but their maximum modelled richness occurred at 3900–4400 m (Fig. 1B) approximately 800m higher than the peak for total species. Non-endemic richness matched closely the elevational patterns of the total species richness (data not shown). Species richness of all three growth forms of lichens had unimodal responses to elevation (Fig. 2A–C). Among the three morphological types, crustose lichens peaked at the highest altitude (4100–4200m) while foliose lichens had their maximum richness at consistently lower altitudes (2400–2500m). Fruticose lichens peaked at an intermediate elevation (3200m). With respect to photobiont type, cyanolichens exhibited maximum richness at lower altitudes (2900–3000m; Fig. 3A) than green algal lichens (3300– 3500m; Fig. 3B). Among the five substratum categories studied, corticolous lichens had a clear unimodal relationship to elevation, with the highest species richness occurring between 2500 and 2700m (Fig. 4A) in the lower part of the temperate forests. Saxicolous and terricolous lichens had slightly bimodal patterns F. 2. Relationship between elevation and lichen species life-form richness in Nepal. A, foliose lichen species richness; B, fruticose lichen species richness; C, crustose lichen species richness. The fitted regression lines represent statistically significant (P % 0·001) smooth spline (s) after using GAM with approximately 4 degrees of freedom. with a prominent hump at 3900–4200m (Fig. 4B & C). The regression analysis results showing the best selected model for each response variable are recorded in Appendix 2. Discussion Total lichen species richness in Nepal varies strongly with elevation (Figs 1A & B) in line with previous findings for vascular plants (Grytnes & Vetaas 2002; Vetaas & Grytnes 2002; Bhattarai & Vetaas 2003; Bhattarai et al. 2004) including orchids (Acharya 2008), and for bryophytes (Grau et al. 2007). In all these studies, as well as studies dealing with lichens from other countries (Wolseley & Aguirre-Hudson 1997; Negi 2000, 2003; Wolf & Alejandro 2003; Pinokiyo et al. 2010 Elevation gradients of lichens in Nepal—Baniya et al. 87 F. 3. The relationship between elevation and lichen richness for lichens with different photobionts. A, cyanolichen species richness; B, green algal lichen species richness. The fitted regression lines represent statistically significant (P % 0·001) smooth splines (s) after using GAM with approximately 4 degrees of freedom. 2008), species richness tends to peak at intermediate altitudes. The various major taxonomic groups exhibit maximum richness at different altitudes: vascular plants (1500 to 2500m; orchids 1600m, ferns 2000m), liverworts 1800m, mosses 2500m and as high as 3100 to 3400m for lichens. Thus, total richness for lichens occurs at higher elevations than for any of the other groups studied. The total lichen species richness in the Himalayas peaked in the upper cloud-forest zone (following the vegetation classification of Hagen 1969; as modified by Miehe 1982; 1989). This is the zone of the highest rainfall on southernmost slopes (> 4000mm), and the lowest rainfall (< 300mm) on northernmost slopes where the Himalayan föhn causes local drying (Miehe 1989). The zone with maximum lichen richness represents the temperate zone with extremely large local variations in water availability and the accompanying gradients in vegetation cover. According to Bhattarai et al. (2004), the high annual rainfall (> 4000mm) and cool summer temperatures (14–17°C) in this zone are likely to favour forests.The large number of temperate broad-leaved and coniferous trees F. 4. The relationship between elevation and lichens preferring specific substrata. A, corticolous lichen species richness; B, saxicolous lichen species richness; C, terricolous lichen species richness. The fitted regression lines represent statistically significant (P % 0·001) smooth splines (s) after using GAM with approximately 4 degrees of freedom. with bark differing in roughness, moisture retention capacity and pH present a wide variety of habitats for lichens. The slightly bimodal peaks for saxicolous (Fig. 4B) and terricolous lichens (Fig. 4C) tend to show the lower peak in species richness occurring at the same altitude at which corticolous species reach their maximum richness (Fig. 4A). These patterns may be determined by habitat availability and specificity. Corticolous lichens apparently peak at elevations with a high abundance of forests, implying a reduced occurrence of natural and well-lit terricolous and saxicolous habitats. Fruticose lichens have their maximum richness at a substantially higher altitude than the foliose lichens (Figs 2A & B). A similar relationship, but on a very different scale, can be seen within forest canopies 88 THE LICHENOLOGIST where the biomass of fruticose lichens increases and that of foliose lichens decreases with height above the ground (Goward 1998; Campbell & Coxson 2001; Gauslaa et al. 2008). In well-lit and open high altitude forests and the upper canopies of trees, fruticose lichens may have the advantage in being able to utilise light from all directions, whereas many foliose and flat lichens maximize the harvest of more or less unidirectional light in shady positions in dense forests and on low canopy branches (Gauslaa et al. 2009). The foliose lichen richness peaks in the same zone as the highest richness for mosses (Grau et al. 2007), taxa that are also relatively shade tolerant. Furthermore, finely dissected fruticose lichens on twigs lack thick boundary layers. Thus, they are more closely coupled to ambient air than flat foliose lichens and absorb water vapour more readily from the air as discussed by Jonsson et al. (2008). By contrast, flat lichens and bryophyte carpets are more dependent on rainwater. The elevation range between 4000 to 4300m is the sub-alpine zone of Nepal (Stainton 1972). This humid upper-cloud zone (Miehe 1989) is densely colonized by epiphytic mosses reflecting the high humidity (Grau et al. 2007). This humidity may also facilitate high lichen richness associated with open canopies with more light facilitating lichen growth (Gauslaa et al. 2007). A high degree of isolation creates endemism (Cox & Moore 2000), and isolation might increase with elevation. High mountain peaks with an exceptionally cold and harsh climate act as islands (Grau et al. 2007), and phytogeographical isolation can lead to localized speciation events. The occurrence of 29 endemic species (17%) out of the 172 total species above the treeline (R 4300 m) emphasises the link between elevation and endemism. The Nepalese lichen endemism rate is higher than that for mosses (3 out of 480 spp. i.e., < 1%) and liverworts (33 out of 368 spp. i.e., 9%), but slightly less than that for vascular plants (303 out of 1957 spp. i.e., 16%). Above 4500m both total and endemic lichen diversity (Figs 1A & B) decline until 6100m, which is the highest reported elevation for lichens en- Vol. 42 demic to Nepal. Cold winds leading to rapid freezing and drying may reduce competition from surviving vascular plant species (Grime 1977). The harsh climate might lead to various specializations and thus facilitate formation of endemic species (Aptroot & Bungartz 2007). Land area might impose a unimodal richness pattern. In general, the larger the area sampled, the greater will be the number of species encountered (Rosenzweig 1995; Qian et al. 2007). A decreasing trend in land area in each 500m band occurs with increasing elevation (Vetaas & Grytnes 2002) as estimated from the Digitized Map by the International Centre for Integrated Mountain Development (ICIMOD, Kathmandu). This is not always a common pattern (Körner 2007). If we consider species-area relationships, more species should be expected towards lower elevations than we observed. However, lichen diversity can be greatly limited by lack of long ecological continuity in forests (Rose 1976) and lichens may be less prevalent at low altitudes because of dense forest canopies and intensive land-use. Furthermore, as lichen species richness also tends to be universally greater in cool or cold climates (Mattick 1953), we believe that the strong reduction in lichen richness at low elevations is real and not a sampling artefact. In conclusion, lichens and various subgroups of lichens exhibit unimodal patterns similar to those found in other major taxonomic groups, but the highest total lichen richness peaks at higher altitudes than in any other group. The maximum lichen richness occurred at the altitude with the highest diversity of ecological niches in terms of spatial heterogeneity, rainfall, temperature, cloud formation, as well as high phorophyte abundance and diversity implying large variation in bark roughness, moisture retention capacity, and pH. We are grateful to the Norwegian State Education Loan Fund (Lånekassen) for providing funding and the Central Department of Botany, Tribhuvan University, Nepal, for providing study leave for CBB. Thanks are due to John Birks for his valuable comments, Walter Obermayer and Per Magnus Jørgensen for provision of literature on lichens and to Govind Ghimire and Bijaya 2010 Elevation gradients of lichens in Nepal—Baniya et al. Kattel for their feedback. Thanks are also due to Louise Olley, Teuvo Ahti and Tor Tönsberg for their help in revising the species list and to the Senior Editor, Peter Crittenden and two anonymous referees for their useful comments and suggestions. R Abbayes, H. D. (1974) Cladonia du Nepal. In Khumbu Himal 6 (W. Hellmich & J. Poelt, eds): 111–116. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Acharya, K. P. (2008) Orchid species richness along a Himalayan elevation gradient. M. Sc. thesis, Bergen University. Aptroot, A. & Bungartz, F. (2007) The lichen genus Ramalina on the Galapagos. Lichenologist 39: 519– 542. Awasthi, G. (1986) Lichen genus Usnea in India. Journal of the Hattori Botanical Laboratory 61: 333–421. Awasthi, D. D. (1991) A key to the microlichens of India, Nepal and Sri Lanka. Bibliotheca Lichenologica 40: 1–337. Awasthi, D. D. (2007) A Compendium of the Macrolichens from India, Nepal and Sri Lanka. Dehra Dun: Bishen Singh & Mahendra Pal Singh. Awasthi, D. D. & Mathur, R. (1987) Species of the lichen genera Bacidia, Badimia, Fellhanera and Mycobilimbia from India. Proceedings of Indian Academy of Sciences, Bangalore. 97: 481–503. Baniya, C. B. (1996) Floristic composition of lichens of Sikles (Kaski) and Shivapuri (Kathmandu) and their ecology. M. Sc. thesis, Tribhuvan University, Kathmandu, Nepal. Bhattarai, K. R. & Vetaas, O. R. (2003) Variation in plant species richness of different life forms along a subtropical elevation gradient in the Himalayas, east Nepal. Global Ecology and Biogeography 12: 327–340. Bhattarai, K. R., Vetaas, O. R. & Grytnes, J. A. (2004) Fern species richness along a central Himalayan elevational gradient, Nepal. Journal of Biogeography 31: 389–400. Bruun, H. H., Moen, J., Virtanen, R., Grytnes, J. A., Oksanen, L. & Angerbjörn, A. (2006) Effects of altitude and topography on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal of Vegetation Science 17: 37–46. Bystrek, J. L. (1969) Die Gattung Alectoria. In Khumbu Himal 6 (W. Hellmich & J. Poelt, eds): 17–24. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Campbell, J. & Coxson, D. S. (2001) Canopy microclimate and arboreal lichen loading in subalpine spruce-fir forest. Canadian Journal of Botany 79: 537–555. Cox, C. B. & Moore, P. D. (2000) Biogeography: an Ecological and Evolutionary Approach. London: Blackwells. Crawley, M. J. (2006) Statistics: an Introduction using R. London: John Wiley & Sons, Ltd. 89 Dobremez, J. F. (1976) Le Nepal: Écologie et Biogeographie. Paris: Centre National de la Recherche Scientifique. Dobremez, J. F. & Jest, C. (1969) Carte Écologique de la Region Annapurna Dhaulagiri. Paris: Centre National de la Recherche Scientifque. Esslinger, T. L. & Poelt, J. (1991) Parmelia masonii, a new lichen species (Ascomycota) from the Himalayas. Bryologist 94: 203–206. Gauslaa, Y., Palmqvist, K., Solhaug, K. A., Holien, H., Hilmo, O., Nybakken, L., Myhre, L. C. & Ohlson, M. (2007) Growth of epiphytic old forest lichens across climatic and successional gradients. Canadian Journal of Forest Research 37: 1832–1845. Gauslaa, Y., Lie, M. & Ohlson, M. (2008) Epiphytic lichen biomass in a boreal Norway spruce forest. Lichenologist 40: 257–266. Gauslaa, Y., Palmqvist, K., Solhaug, K. A., Hilmo, O., Holien, H., Nybakken, L. & Ohlson, M. (2009) Size dependent growth in two old-growth associated macrolichen species. New Phytologist 18: 683–692. Goward, T. (1998) Observations on the ecology of the lichen genus Bryoria in high elevation conifer forests. Canadian Field Naturalist 112: 496–501. Grau, O., Grytnes, J.-A. & Birks, H. J. B. (2007) A comparison of altitudinal species richness patterns of bryophytes with other plant groups in Nepal, Central Himalaya. Journal of Biogeography 34: 1907–1915. Grime, J. P. (1977) Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 11: 1169–1194. Grytnes, J. A., Heegaard, E. & Ihlen, P. G. (2006) Species richness of vascular plants, bryophytes, and lichens along an altitudinal gradient in western Norway. Acta Oecologica 29: 241–246. Grytnes, J. A. & Vetaas, O. R. (2002) Species richness and altitude: a comparison between null models and interpolated plant species richness along the Himalayan altitudinal gradient, Nepal. American Naturalist 159: 294–304. Hagen, T. (1969) Report on the Geological Survey of Nepal: Preliminary Reconnaissance. Zurich: Orell Füssli Arts Graphiques. Hastie, T. J. & Tibshirani, R. J. (1990) Generalised Additive Models. London: Chapman & Hall. Heegaard, E. (2004) Trends in aquatic macrophyte species turnover in Northern Ireland – which factors determine the spatial distribution of local species turnover? Global Ecology and Biogeography 13: 397–408. Hellmich, W. & Poelt, J. (eds) (1977). Khumbu Himal 6. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Hunter, M. L. & Yonzon, P. (1993) Altitudinal distributions of birds, mammals, people, forests, and parks in Nepal. Conservation Biology 7: 420–423. Jahns, H. M. & Seelen, E. J. R. (1974) Baeomyces-Funde aus dem Himalaya. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 101–108. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. 90 THE LICHENOLOGIST Jonsson, A. V., Moen, J. & Palmqvist, K. (2008) Predicting lichen hydration using biophysical models. Oecologia 156: 259–273. Jørgensen, P. M. (2001) The lichen genus Erioderma (Pannariaceae) in China and Japan. Annales Botanici Fennici 38: 259–264. Kappen, L., Schroeter, B., Scheidegger, C., Sommerkorn, M. & Hestmark, G. (1996) Cold resistance and metabolic activity of lichens below 0°C. In Life Sciences: Space and Mars Recent Results 18: 119–128. Oxford: Pergamon Press Ltd. Kessler, M. (2000) Altitudinal zonation of Andean cryptogam communities. Journal of Biogeography 27: 275–282. Körner, C. (2007) The use of ‘altitude’ in ecological research. Trends in Ecology and Evolution 22: 569–574. Kurokawa, S. (1974) Anaptychia of the NepalHimalaya. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 109–110. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Kuusinen, M. (1996) Epiphyte flora and diversity on basal trunks of six old-growth forest tree species in southern and middle boreal Finland. Lichenologist 28: 443–463. Lamb, L. M. (1966) Die Gattung Stereocaulon. In Khumbu Himal 1 (W. Hellmich, ed): 349–352. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Mattick, F. (1953) Lichenologisches Notizen. Berichte der Deutschen Botanischen Gesellschaft 66: 263–276. Miehe, G. (1982) Vegetationsgeographische Untersuchungen im Dhaulgiri und Annapurna Himalaya, Dissertationes Botanicæ. Journal Cramer, Vaduz 66: 66. Miehe, G. (1989) Vegetation patterns in Mount Everest as influenced by monsoon and föhn. Vegetatio 79: 21–32. Miehe, G. (1990) Langtang himal-flora und vegetation als klimazeiger und-zeugen im Himalaya. Ph. D. thesis, Universität Göttingen. Mitchell, M. (1974) Die Gattung Leptogium sect. Mallotium im Himalaya. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 121–126. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Negi, H. (2000) On the patterns of abundance and diversity of macrolichens of Chopta-Tunganath in the Garhwal Himalaya. Journal of Biosciences 25: 367–378. Negi, H. (2003) Lichens: a valuable bioresource for environmental monitoring and sustainable development. Resonance 8: 51–58. Pant, G. & Awasthi, D. D. (1989a) Caliciales from India and Nepal. Biovigyanam 15: 3–27. Pant, G. & Awasthi, D. D. (1989b) Lichen genus Catillaria s. lat. in India. Proceedings of Indian Academy of Sciences (Plant Science). 99: 369–384. Pinokiyo, A., Singh, K. P. & Singh, J. S. (2008) Diversity and distribution of lichens in relation to altitude within a protected biodiversity hotspot, north-east India. Lichenologist 40: 47–62. Vol. 42 Poelt, J. (1966a) Die Gattung Ochrolechia. In Khumbu Himal 1 (W. Hellmich, ed): 251–261. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Poelt, J. (1966b) Die Lobaten Arten der Sammelgattung Lecanora, Lecanoraceae. In Khumbu Himal 3 (W. Hellmich, ed): 187–202. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Poelt, J. (1974) Die Gattungen Physcia, Physciopsis und Physconia. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 57–99. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Poelt, J. & Hinteregger, E. (1993) Beiträge zur Kenntnis der Flechtenflora des Himalaya. Bibliotheca Lichenologica 50: 1–247. Poelt, J. & Reddi, B. V. (1969) Candelaria und Candelariella. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 1–16. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Qian, H., Fridley, J. D. & Palmer, M. W. (2007) The latitudinal gradient of species-area relationships for vascular plants of North America. American Naturalist 170: 690–701. R Development Core Team (2008) R: A language and environment for statistical computing version 2.7.0. Rose, F. (1976) Lichenological indicators of age and environmental continuity in woodlands. In Lichenology: Progress and Problems (D. H. Brown, D. L Hawksworth & R. H. Bailey, eds): 279–307. London: Academic Press. Rosenzweig, M. L. M. L. (1995) Species Diversity in Space and Time. Cambridge: Cambridge University Press. Schmidt, A. (1974) Chaenotheca und Coniocybe. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 133–143. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. Schroeter, B., Green, T. G. A., Kappen, L. & Seppelt, R. D. (1994) Carbon dioxide exchange rate at subzero temperatures. Field measurements on Umbilicaria aprina in Antarctica. Cryptogamic Botany 4: 233–241. Sharma, L. R. (1995). Enumeration of Lichens of Nepal. Euroconsult, Biodiversity Profiles Project, Publication No. 3. Nepal Government, 1–109. Stainton, J. D. A. (1972) Forests of Nepal. London: John Murray Publisher, Ltd. Stainton, A. (2001) Flowers of the Himalaya: a Supplement. Oxford: Oxford University Press. Upreti, B. N. (1999) An overview of the stratigraphy and tectonics of the Nepal Himalaya. Journal of Asian Earth Sciences 17: 577–606. Vetaas, O. R. & Grytnes, J. A. (2002) Distribution of vascular plant species richness and endemic richness along the Himalayan elevation gradient in Nepal. Global Ecology and Biogeography 11: 291–301. Vězda, A. & Poelt, J. (1975) Die Gattungen Dimerella und Pachyphiale. In Khumbu Himal 6 (W. Hellmich, & J. Poelt, eds): 127–132. München: Universitätsverlag Wagner Ges. M. B. H. Innsbruck. 2010 Elevation gradients of lichens in Nepal—Baniya et al. Vitikainen, O. (1986) Peltigera dolichospora, a new Himalayan-western Chinese lichen. Lichenologist 18: 387–390. Whittaker, R. H. (1972) Evolution and measurement of species diversity. Taxon 21: 213–251. Wolf, J. H. D. (1993) Diversity patterns and biomass of epiphytic bryophytes and lichens along an altitudinal gradient in the Northern Andes. Annals of the Missouri Botanical Garden 80: 928–960. Wolf, J. H. D. & Alejandro, F.-S. (2003) Patterns in species richness and distribution of vascular epiphytes in Chiapas, Mexico. Journal of Biogeography 30: 1689–1707. 91 Wolseley, P. A. & Aguirre-Hudson, B. (1997) The ecology and distribution of lichens in tropical deciduous and evergreen forests of northern Thailand. Journal of Biogeography 24: 327–343. Yoda, K. (1967) A preliminary survey of the forest vegetation of eastern Nepal II. General description, structure and floristic composition of sample plots chosen from different vegetation zones. Journal of the College of Arts and Sciences, Chiba University National Science Series 5: 99–140. Accepted for publication 28 March 2009 Appendix 1. Elevation ranges of lichen species of Nepal. Name of lichen species Altitudinal range (m) Alectoria ochroleuca Allocetraria flavonigrescens* A. globulans A. oakesiana A. sinensis A. stracheyi Anthracothecium himalayense A. leucostomum Arctocetraria nigricascens Arctoparmelia subcentrifuga Aspicilia cinerea Awasthia melanotricha* Bacidia millegrana B. nigrofusca B. personata B. rubella B. spadicea Baeomyces pachypus B. roseus Bryonora castanea var. castanea B. castanea var. euryspora B. curvescens B. pulvinar var. microspora* B. pulvinar var. pulvinar* B. pulvinar* B. reducta B. rhypariza var. cyanotropha B. rhypariza var. lamaina B. rhypariza var. rhypariza B. selenospora* B. stipitata* B. yeti Bryoria bicolor B. confusa 4600–5100 4600–4800 3700–4700 3900–4200 4830 3000–5000 1650 2400–2600 3800 4600 4300 4350–4510 1800–3200 3100 300–1800 300 2600 2000–4200 2000–2500 3750 4420 3500–5000 4720–5070 4830–5100 4720–5390 3200 3750–3900 5100–5200 4000–4100 3800–4400 3000–5080 4190–5200 3000–5050 3450 Appendix 1. Continued Name of lichen species Altitudinal range (m) B. furcellata B. himalayana B. implexa B. lactinea B. lanestris B. levis B. nadvornikiana B. nepalensis* B. nitidula B. perspinosa B. poeltii B. smithii B. variabilis Buellia geophila B. granularis B. inornata B. papillata Bulbothrix isidiza B. meizospora B. setschwanensis Calicium abietinum C. lenticulare C. viride Caloplaca arnoldii C. aureosora C. borealis var. borealis C. borealis var. oligosperma C. castellana C. cerina var. cerina C. cerina var. chloroleuca C. cerinopsis* C. cirrochroa C. cirrochroopsis* C. citrina C. cupreobrunnea* 3000 3900–4000 3700–4000 3300–3450 3800–3900 3450 3800–3900 3900–4000 3900 3800–4500 4400 3000–4511 3800–4000 4200 2100 700–900 4650 900–2180 2100–2300 1260–2300 3000–3300 2250 3000–3300 3500–4000 2500–3000 3650 4000 4000–5540 3200 4950–5050 1800–3200 3900–4500 1480 3750 2700 92 THE LICHENOLOGIST Appendix 1. Continued Vol. 42 Appendix 1. Continued Name of lichen species Altitudinal range (m) Name of lichen species C. cupulata* C. epiphyta C. epithallina C. exsecuta var. aphanes* C. farinosa* C. grimmiae C. holocarpa C. holochracea C. insularis C. isabellina* C. leptocheila C. lithophila C. lobulascens* C. lypera* C. maura* C. obliterans C. ochroplaca C. phoenicopta* C. praeruptorum* C. procerispora* C. rinodinopsis C. sancta* C. saxicola var. chamaeleon C. saxicola var. saxicola C. saxifragarum C. tetraspora C. ulcerata* C. variabilis Calvitimela aglaea C. armeniaca Candelaria crawfordii C. sphaerobola Candelariella aurella C. coralliza C. grimmiae C. himalayana C. nepalensis C. sorediosa C. vitellina var. glacialis C. vitellina var. vitellina Canomaculina subsumpta C. subtinctoria Canoparmelia aptata C. ecaperata C. eruptens Carbonea vorticosa Catapyrenium cinereum C. daedaleum Catillaria leptocheiloides Catolechia wahlenbergii Cetraria ambigua C. nepalensis* Cetrelia braunsiana 3200–3800 4400–4500 4350–4450 2060 3000–3500 3500 3700–5000 1700–2300 4600–5540 4100–5200 5000 3200–4850 1500–1800 2000–3000 4950–5000 3200–3700 1800–3000 3200 3200–3650 4100–5000 2000–3000 4000–4340 3200–4000 2900–5000 4950–5000 4400–5000 3500–3750 2000–3500 5000 5000 1300–3000 3000–3900 1600–3500 3800–5000 4250–5000 3700–5100 5000–5400 3700–5200 4900–5400 1600–5540 1200–1500 1500 1500–2000 1300–2350 5000 5000–7400 4300–5080 3900–5080 2100 4500 3400–5390 4500 3150 C. cetrarioides C. olivetorum Cetreliopsis rhytidocarpa ssp. langtangi Chaenotheca brunneola C. chrysocephala C. furfuracea C. hispidula C. phaeocephala C. stemonea Chrysothrix chlorina Cladia aggregata Cladonia amaurocraea C. awasthiana C. calyciformis C. cariosa C. carneola C. cartilaginea C. ceratophyllina C. chlorophaea C. ciliata C. coccifera C. corniculata C. corymbescens C. delavayi C. fenestralis C. fimbriata C. fruticulosa C. furcata C. humilis C. laii C. luteoalba C. macilenta C. macroptera C. mongolica C. nitida C. ochrochlora C. pocillum C. pyxidata C. ramulosa C. rangiferina C. scabriuscula C. singhii C. squamosa C. stellaris C. stricta C. subconistea C. subulata C. yunnana Coccocarpia erythroxyli C. palmicola Collema nepalense* C. poeltii* C. pulcellum Altitudinal range (m) 2000–3000 2850 2880 3000–3556† 2700–3879† 3000 †2947–3250 3000–3869† 3000 3100–3200 3900–4000 4530–5230 950–5100 1500–2700 3100 3900–4100 1100–2000 3000–4600 2000–5350 4480 2000–4000 1650–2300 1500–5100 2200–5230 3900–5100 3900–4250 1500–5100 1700–3900 1200–4250 2700–4600 3200–3660 1500–3800 2700–3900 1600–4500 3900 1500–3500 2000–3600 4000–4100 2000–3366 3400–4500 2845–3000 750–2700 1800–4000 4400–4500 3300–4000 2500–3500 2200–2700 3300–4000 3150 1800 3900–4000 3900–4000 1350–1500 2010 Elevation gradients of lichens in Nepal—Baniya et al. Appendix 1. Continued Name of lichen species C. rugosum C. subconveniens C. substipitatum Dermatocarpon miniatum var. miniatum D. vellereum Dibaeis sorediata Dimelaena oreina Dimerella lutea Diploicia canescens Diploschistes muscorum D. muscorum subsp. bartlettii D. muscorum subsp. muscorum D. nepalensis D. scruposus Dirinaria aegialita D. applanata D. consimilis Erioderma meiocarpum Eumitria pectinata Evernia mesomorpha Everniastrum cirrhatum E. nepalense E. rhizodendroideum Flavocetraria cucullata F. nivalis Flavocetrariella leucostigma F. melaloma Flavoparmelia caperata Flavopunctelia flaventior Fuscopannaria poeltii* F. praetermissa Glyphis cicatricosa Graphis scripta G. subglauconigra Haematomma puniceum Heterodermia angustiloba H. awasthii H. boryi H. comosa H. dactyliza H. dactyliza f. serpens H. dendritica H. diademata H. dissecta var. dissecta H. firmula H. flabellata H. himalayensis H. incana H. isidiophora H. japonica H. obscurata H. pellucida H. propagulifera 93 Appendix 1. Continued Altitudinal range (m) Name of lichen species Altitudinal range (m) 1800 1260 3000–4000 3200 2000–3600 2500 3200–4884 3000–4000 4650 900–2160 900–2100 1500–2160 900 2000–3000 1800 600 600 3000–3050 400–2000 3962–4572 1900–3300 1410–3600 2490–3800 3600–4500 3800–5300 3500–4500 3900–4200 2250–2743 2250–2800 3500 †3805–4000 1600 1600–3366 2000–3200 1800 1800–2300 1900 1500–3000 1500–2500 2100 3800–5150 1800 410–3807† 1500–3000 1200–2200 3100 1500–2000 1800–2100 2100–2250 3000–4000 1400–3000 2500–3992 1500–3000 H. pseudospeciosa H. punctifera H. rubescens H. speciosa H. togashii H. tremulans Hyperphyscia granulata H. minor Hypogymnia delavayi H. hypotrypa H. vittata Hypotrachyna adducta H. crenata H. exsecta H. flexilis H. imbricatula H. infirma H. koyaensis H. majoris H. neodissecta H. osseoalba H. revoluta H. rhabdiformis H. scytophylla H. sinuosa H. sublaevigata Immersaria athroocarpa Ingvariella bisporus Ioplaca pindarensis Lasallia freyana L. pertusa Lecanora adolfii L. amorpha* L. chlarotera L. chondroderma L. demissa L. emodi* L. formosa L. garovaglii L. hellmichiana* L. himalayae* L. kirra* L. lesleyana* L. meridionalis L. muralis var. dubyi L. muralis var. muralis L. phaeodrophthalma L. rubina var. australis L. rugosella L. sherparum* L. somervellii L. sulphurea L. terestiuscula 150–2100 2250 1800–2800 1650–2100 3000–4000 1500 1200 1200–1400 1800–4080 3600–4050 2800–4200 2100 2500–2800 1800–2300 2160 1400–2600 1000–2100 1500–2100 1650 2250–2550 1100–2200 2000–2250 2500–2800 2250–3250 3500 2250 3900–4340 3750 2900–5200 3800–3900 3800–5050 5540 5000 200 3800–5400 2700–4000 5540 4850–5860 3000–5050 1800–3500 3900–4850 4400–4500 5029–5334 1600–3366 4930 3500–4340 4100–4500 3300–4250 2732–3200 3300–4600 3700–5500 4300–5639 4500–5200 94 THE LICHENOLOGIST Appendix 1. Continued Vol. 42 Appendix 1. Continued Name of lichen species Altitudinal range (m) Name of lichen species Altitudinal range (m) L. tschomolongmae* Lecidea advena L. auriculata L. bella* L. brachyspora L. bucculenta L. diducens L. epiiodiza L. fuscoatra var. indecora L. haerjedalica var. gyrodisca L. himalaica L. khumbuensis L. lactea L. leptoboloides L. molybdochroa L. poeltii L. secernens L. silacea L. steineri L. tessellata Lecidella carpathica L. dimelaenophila L. stigmatea Leprocaulon arbuscula Leproplaca chrysodeta Leptogium asiaticum L. askotense L. azureum L. brebissonii L. burnetiae var. burnetiae L. cochleatum L. delavayi L. delavayi f. fuliginosulum L. isidiosellum L. javanicum L. pedicellatum L. phyllocarpum L. resupinas L. saturninum L. trichophorum Lethariella cladonioides Letrouitia domingensis Lobaria discolor L. isidiosa L. kurokawae L. pindarensis L. pseudopulmonaria L. retigera L. subretigera Lobothallia alphoplaca L. praeradiosa Melanelia tominii Melanohalea poeltii* 3000–5000 5000–5540 5000–5200 4950–5000 4300–5900 5000–5540 4720 4850 4340 5150–5200 5150–5200 5540 5460 4340–5200 3900–4000 4500–5000 5000 5000 5150–5200 5080–5639 3900–4000 3800–3900 4520–5200 1900–3100 3300–4850 3900–4100 1500–1800 1410 1800–3200 1500–2250 1800 3000–4100 3900–4000 1500 1410–1800 1500–3000 2000–3100 1600 1500–2100 1450–1500 4724 240 3150 2800–3962 1800–3400 2700–4000 2550–4050 1600–3650 2800 4400 4000–4500 3200–3600 4500–4600 Menegazzia terebrata Mycobilimbia hunana Myelochroa aurulenta M. entotheiochroa M. subaurulenta Nephroma helveticum var. helveticum N. nakaoi Nephromopsis ahtii N. isidioidea N. nephromoides N. pallescens N. stracheyi Ochrolechia bryophaga O. glacialis O. margarita O. rosella f. sorediascens O. subviridis O. trochophora Pachyphiale himalayensis Parmelaria subthomsonii P. thomsonii Parmelia adaugescens P. erumpens P. latissima var. marmariza P. masonii* P. meiophora P. omphalodes P. ricasolioides P. squarrosa P. submutata P. sulcata Parmelina tiliacea Parmelinella simplicior P. wallichiana Parmelinopsis expallida P. minarum Parmotrema austrosinense P. cooperi P. hababianum P. maclayanum P. melanothrix P. mellissii P. nilgherrensis P. praesorediosum P. pseudonilgherrense P. pseudotinctorum P. rampoddense P. ravum P. reticulatum P. sancti–angelii P. stuppeum P. tinctorum P. ultralucens 1300–4000 1650 1410–2250 2000–2200 1950 2160–2800 2800 4090–4530 2900–3100 2400–3600 2400–3048 2700–3300 3900–4000 5000–5200 5000 3500 2900 3000–4000 3900–4000 1800–3200 1800–3366 3300–3600 1900–2200 2900 3000–6100 1800–3350 3500–4500 2700 3000–3800 2800 3000–3366 2100 2000–2900 1800–2400 1800 1500 750–3000 1500–2100 1500–2000 1500 2400–3200 1500 1080–2800 1410–1800 1800–3530 2900–3366 900–2800 1500–2500 1500–2500 1410–2100 1800 240–2250 1500 2010 Elevation gradients of lichens in Nepal—Baniya et al. Appendix 1. Continued 95 Appendix 1. Continued Name of lichen species Altitudinal range (m) Name of lichen species Altitudinal range (m) P. yodae* Peltigera canina P. dolichorrhiza f. dolichorrhiza P. dolichospora* P. elisabethae P. malacea P. membranacea P. polydactylon P. pruinosa P. rufescens P. scabrosa Pertusaria hemisphaerica Phaeographina pyrrhochroa Phaeophyscia endococcina P. endococcina var. khumbuensis P. endococcinoides var. megalospora P. hispidula var. exornatula P. hispidula var. hispidula P. lygaea* P. primaria* P. pyrrhophora* P. sciastra Phlyctella indica Physcia aipolia P. caesia P. clementei P. dilatata P. dubia P. phaea P. stellaris ssp. intestiniformis* P. tribacia P. tribacioides Physciella nepalensis* Physconia distorta P. enteroxantha P. grisea P. muscigena Physma byrsaenum Placidium squamulosum Pleopsidium chlorophanum* Porpidia aerolotera P. crustulata P. elegantior P. hydrophila P. macrocarpa Protoparmelia badia var. badia P. effigurans* Punctelia borreri P. rudecta P. subrudecta Pyrenula cayennensis P. immersa Pyxine coccifera 2300 3150–3300 1800–2250 3000–4100 1350 2700–3300 1900–2400 1950–2920 1800 2100 3800 2400–3366 2100 3800–3900 3900 3900–4000 365 3800–3900 4500–4600 3800–3900 1420 3900–5200 1800 1800–2200 3900–5000 3500–4000 1600–5000 5000 4340–5000 4250–4340 3800–4600 1400 1400 3000–4000 2740 1400–4200 4250–5540 1350 4400–4500 3500–4000 5150–5200 2800–5000 4700 3900–4000 2950–4000 3750–4000 3750 1600–3200 1500–3200 1500–2500 3000–3200 2000 750 P. meissnerina P. philippina P. sorediata Ramalina conduplicans R. farinacea R. flabelliformis* R. hossei var. hossei R. sinensis R. subfarinacea Rhizocarpon geographicum Rhizoplaca chrysoleuca var. chrysoleuca R. melanophthalma var. obscura R. peltata Sagema potentillae Sclerophora coniophaea Solorina bispora Stereocaulon claviceps S. foliolosum var. foliolosum S. foliolosum var. botryophorum S. foliolosum var. strictum S. glareosum S. himalayense S. myriocarpum S. paradoxum S. piluliferum S. pomiferum S. sasaki var. sasaki Sticta henryana S. nylanderiana S. platyphylloides S. praetextata S. weigelii var. weigelii Sulcaria sulcata S. virens Tephromela aglaea T. armeniaca T. glacialis* T. siphulodes var. siphulodes* Thamnolia vermicularis Thelenella luridella Tremolecia atrata Tuckneraria laureri Tylophoron moderatum Umbilicaria badia U. cinereorufescens U. decussata var. decussata U. decussata var. rhizinata* U. indica var. indica U. indica var. nana* U. krascheninnikovii U. leiocarpa U. nenella U. nepalensis 300 900–1800 1410–1800 2133 2700 3200 1200–2500 2700–4267 1610 4500 2100–5150 4270 3660 4000 3000 3600–4200 4000 3300–4000 3600–4600 2400–3600 4200–4400 2500–5400 3900–5303 2000–3366 2160–5150 2500–4700 3600–4200 3657–3962 1800–3600 1900–3150 2100–3500 800–2250 3150–3300 3000–3600 5000–5200 5000–5200 4500 4830 3869–5455 1410 5000–5860 3200–4900 1800 2550 4250–5200 5100–5500 4950–5000 1800–3150 4200–4300 3700–5100 5150 5100 3600–5100 96 THE LICHENOLOGIST Vol. 42 Appendix 1. Continued Appendix 1. Continued Name of lichen species Altitudinal range (m) Name of lichen species Altitudinal range (m) U. thamnodes U. trabeculata U. vellea U. yunnana Usnea aciculifera U. baileyi U. compressa U. dendritica U. galbinifera var. subfibrillosa* U. himalayana U. longissima U. montisfuji U. nepalensis U. norkettii* U. pectinata U. pseudomontisfuji U. robusta 3600–5050 5000–5100 3600–5050 2550 1200–2250 1500–2400 2400–2700 1200–3000 2200 1800–3300 2700–3750 2200–3900 3800–4000 3000–3500 3962 3200 2500–3000 U. rubicunda U. splendeus U. thomsonii Xanthoparmelia coreana X. dentata Xanthoparmelia isidiosa X. mexicana X. nepalensis* Xanthoria borealis X. elegans X. fallax X. fulva X. sorediata X. ulophyllodes var. ulophyllodes 2700 1800–3300 1800–3150 3400–3850 4480–4572 2800–3200 3070–3700 3900 3900–4000 3750–6000 3200–3400 2700–3200 3200–5000 3200–4850† * species endemic to Nepal †Louise Olley’s unpublished personal observation Appendix 2. The elevational gradient of lichen species richness regression analysis results modelled after different species richness as response variables and their elevation as predictor variable. The Quasi-poisson family of error fitted in the GAM model after the cubic smooth spline (s) with approximately 4 degrees of freedom. (P%0·05) Response variables Null df Res. df Total species Endemic species Crustose species Foliose species Fruticose species Cyanolichens Green algal lichens Corticolous Saxicolous Terricolous 72 72 72 72 72 72 72 72 72 72 68 68 68 68 68 68 68 68 68 68 D2 0·945 0·915 0·914 0·941 0·961 0·944 0·941 0·970 0·908 0·895 Deviance 3598·1 362·3 1008 1764·9 1252·2 632·2 3047·7 2374·3 1075·7 422·3 Df = degree of freedom, Res. = residual, D2 = regression coefficient of determination F Pr (>F) 304·5 205 181·4 288·4 560·9 338·4 280·9 588 190·2 192·8 <0·001 <0·001 <0·001 <0·001 <0·001 <0·001 <0·001 <0·001 <0·001 <0·001 Paper IV Baniya, C.B., Solhøy, T. & Vetaas, O.R. 2009. Temporal changes in species diversity and composition in abandoned fields in a trans-Himalayan landscape, Nepal. Plant Ecology, 201: 383-399. IV Plant Ecol (2009) 201:383–399 DOI 10.1007/s11258-008-9473-3 Temporal changes in species diversity and composition in abandoned fields in a trans-Himalayan landscape, Nepal Chitra Bahadur Baniya Æ Torstein Solhøy Æ Ole R. Vetaas Received: 14 June 2007 / Accepted: 21 July 2008 / Published online: 12 August 2008 Ó Springer Science+Business Media B.V. 2008 Abstract Secondary succession is an increasing phenomenon due to global changes in agriculture policies and practices. The empirical findings are biased towards the temperate zone. Abandonment of agriculture fields is less frequent in the subtropical and tropical zones where agriculture areas are, in general, expanding. But there are exceptions; a rapid rate of abandonment of agricultural fields have taken place in the arid trans-Himalayan region, due to today’s globalization of economy. We analysed agriculture fields that were abandoned between 1950 and 2003 in a large u-valley in central Nepal (3400 m a.s.l.). The potential forest vegetation is dominated by Pinus wallichina and shrubs of junipers and cotoneaster species. We tested the intermediate richness hypothesis in relation to vegetation cover, soil development and whether old-field succession is convergent or divergent with species data from 242 1 m2 plots in 5 age-classes. The main species compositional turnover expressed by Detrended C. B. Baniya
T. Solhøy Department of Biology, University of Bergen, Allégaten 41, 5007 Bergen, Norway Correspondence Analyses (DCA) correlated, as expected, with time after abandonment. Fields that were abandoned a long time ago are closer to forest at the periphery of the agricultural landscape. Moisture of the soil significantly increased with age of abandonment, but total vegetation cover and pH were negatively related to age. Beta diversity expressed in DCA SD-units showed an increasing trend with age of abandonment, supporting the divergence pattern in old-field succession. The reason why the succession is not converging may be due to browsing by domestic animals that prevent a closed canopy of pines and juniper to develop. There was a significant hump-shaped pattern in species richness along the temporal gradient, which agrees with the intermediate species-richness hypothesis. There was a rapid increase in species richness in plots close to the villages that were used for haymaking which increased the seed input significantly. Keywords GLM
Himalaya
Manang
Multivariate analyses
Old-field succession
Secondary succession
Species richness C. B. Baniya Central Department of Botany, Tribhuvan University Kirtipur, Kathmandu, Nepal Introduction O. R. Vetaas (&) UNIFOB-Global, University of Bergen, Nygaardsgaten 5, 5015 Bergen, Norway e-mail: Ole.Vetaas@global.uib.no Secondary succession has been a key issue in vegetation ecology (Drury and Nisbet 1973; Egler 1952; Horn 1974; Pickett et al. 1987), and its 123 384 importance is increasing due to accelerating land-use changes in most parts of the world (Bazzaz 1996). Old-field succession is one type of secondary succession, which has become more common due to global changes in agricultural policies and practices. Research on abandoned crop fields was started by Oosting (1942), in North Carolina, USA, and since has been used to elucidate processes in succession (e.g. Bard 1952; Bazzaz 1996; Odum 1960, 1969; Quarterman 1957). Several patterns in species richness and turnover are expected in the course of secondary succession, but these have been debated for a long time. The classical view was that species diversity increase during the course of succession and reach a maximum in the later phases (Clements 1936; Margalef 1968; Odum 1969, 1971; Slobodkin and Sanders 1969), but the current view is that diversity is higher in the intermediate phases than in the late phases of succession (Horn 1974; Bazzaz 1996 and others). However, diversity is a loose composite concept, and has to be more operational if one is to test the above predictions. Two types of diversity changes are found in succession; richness within a community (alpha) and difference in species composition between communities (beta). Since community is also a loose concept, this division has some limitations, especially as they are not scaled to unit area (Palmer 1990; Whittaker 1987). If they are scaled to unit area by substituting the community by sampling units, the concept is standardised and can be used to estimate different aspects of diversity along a temporal gradient. In the context of successional changes, beta diversity implies divergence or convergence. Early ecologists considered succession as a convergent development, where different communities, or stages, replaced one another in a predictable pattern, and converged into a relatively stable long-term community, or climax. Studies in recent decades claim that succession is a multidirectional probabilistic process, which may have several different endpoints (Bazzaz 1996; Collins and Adams 1983; Glenn-Lewin 1980; Matthews 1979). This allows divergence and convergence, depending on site conditions, e.g., development of tree layer or not in the mature phases. Many studies have found that successional development into forest may result in convergence since the dominant canopy reduces the micro scale variability on the ground (Bard 1952; 123 Plant Ecol (2009) 201:383–399 Bazzaz 1996; Horn 1974; Odum 1969; Pickett 1989; Pickett and McDonnell 1989; Quarterman 1957; Vetaas 1997). In a recent study in the high Andes, Sarmiento et al. (2003) found that abandoned agricultural fields above the forest line had a divergent development. A temporal change in the number of species per sampling unit (alpha diversity) is expected. Patterns of species richness have been justified both by longterm studies on fixed abandoned agricultural fields, for e.g. Tilman (1986); Pickett (1989); Tunnell et al. (2004) and by chronological studies based on research after certain years of abandonment, such as Bazzaz (1996); Bonet and Pausas (2004); Sarmiento et al. (2003); Ruprecht (2005), Arbelo et al. (2006) and Otto et al. (2006). A low number of species in the initial phase is an inevitable phenomenon in primary succession but also in old-field succession. The controversy is whether the maximum richness is found in the mature phases or in the intermediate phases. The former view is related to the classical view of convergent development into a stable climax with high species richness (Clements 1936; Margalef 1968; Odum 1969, 1971; Slobodkin and Sanders 1969). However, in recent decades several authors have found maximum richness in intermediate phases (Bazzaz 1996; Brown and Southwood 1987; Horn 1974). The intermediate phases may have a mixture of light-demanding early-phase species and shade-tolerant late-phase species. If one assumes that biomass and vegetation cover increase with time, then maximum richness in the intermediate phases could be deduced from Grime’s model on species-richness and biomass. This model predicts higher richness at intermediate levels of biomass, which may correspond to the intermediate phases of succession. If disturbance is defined as removal of biomass (sensu Grime) then the intermediate disturbance model (Connell 1978) will also predict higher richness at intermediate biomass levels, i.e. intermediate succession phases. Thus even if the old-field succession does not develop into a forest, one may expect higher richness in the intermediate phases compared to the oldest fields. One may also expect decrease in soil pH as vegetation cover increases. This development will be enhanced by acidic litter from pine and junipers species. However, grazing may prevent this development and facilitate open calcareous Plant Ecol (2009) 201:383–399 grassland, which in the temperate zone is known to be rich in species (Butaye et al. 2005; Piqueray et al. 2007 and references therein). The variation in the total number of species within each succession phase will depend on variation in beta diversity and the species richness through the course of succession. Even with a very low alpha diversity in the mature phases, the total number of species may be high if beta diversity is high. The variation in total number of species in each phase must be seen in relation to convergence versus divergence and the trend in species richness. We report from an old-field succession in abandoned sub-alpine wheat and buckwheat fields located in the arid trans-Himalayan (dry inner valleys) zone in Nepal. Here we attempt to test three hypotheses on species diversity. The study of old-field succession has mostly been developed in temperate lowland vegetation (Bazzaz 1996; and references therein); considerably less information is available for high altitude old-fields (Sarmiento et al. 2003). We hypothesise that changes in species composition will have a divergent development. The area is used for grazing and it is not certain if the old-field will develop into Pinus wallichina A. B. Jack. forest or Juniperus shrubs, which surround the agricultural landscape. We hypothesised that number of species would show a maximum at the mid-successional 385 phase; this is tested against the null hypothesis of no trend. If one uses the sampling plots as units, alpha diversity means species richness per plot and beta diversity means the species turnover between a set of plots. The aims of the study are to (1) quantitatively describe the temporal changes in species composition, vegetation cover and soil properties by means of ordination, (2) evaluate if the succession is divergent or convergent and (3) test the hypothesis of enhanced species richness in intermediate phases. Materials and methods Study area The study area is located in the Manang district, within Annapurna Conservation Area, in north-central Nepal, latitude 28°400 N and longitude 84°010 E (Fig. 1). It is situated between 3175 and 3475 meters above sea level (m a.s.l.) in the trans-Himalayan region between the Himalayan range and the Tibetan Plateau. The area is north of the massive Annapurna range which has a maximum elevation above 7000 m a.s.l. Thus it receives little of the monsoon rain, which comes from the south-east. The mean annual precipitation during the year is ca. 400 mm (Anonymous CHINA CHINA INDIA 500 km Fig. 1 Map of study area and the location of the two villages Bhraka and Pisang 123 386 1995). The mean maximum/minimum temperatures recorded at Jomsom (nearest comparable station) are 7.9°C/-1.75°C in winter and 22.6°C/14.15°C in summer during 1995 (Anonymous 1999). Snow is common during winter. There is decreasing moisture from east to west in the upper Manang valley, and the south-facing slopes are significantly drier and warmer than those facing north. All agricultural fields are found on the south exposed slopes or at the valley bottom. This is also reflected in the vegetation and forest formation. At elevations above 3000 m a.s.l. there is forest of Pinus wallichiana A. B. Jack., Betula utilis D. Don and Abies spectabilis (D. Don) Mirb., on the north-facing slopes, and some forest of P. wallichiana on the dry south-facing slopes. Here shrubs of Juniperus indica Bertol, Rosa spp. and Caragana spp. dominate the nearby landscape. The valley bottom is in an intermediate situation, and forest of P. wallichiana is the mature vegetation where grazing and forest cutting is restricted. Above the timberline (4000– 4300 m a.s.l.) there is a harsh continental climate, which allows only steppe vegetation similar to that of the Tibetan plateau (Miehe 1982). The dominant land-use is agriculture and domestic animals with yak, horse, mule, sheep and goats graze in the alpine pastures above the timberline (4300 m a.s.l.) during the summer. In the winter the animals are brought down to graze the grasslands of the lower slopes and valley floor. All the abandoned fields and active fields are open for grazing after the harvest in late September. Field methods Two study sites, Pisang and Bhraka (Fig. 1), were selected as suitable sites for sampling. The total agricultural area in Pisang is 150 ha and almost 75% is abandoned. In Bhraka 60% of the agricultural land (total 185 ha) is abandoned. The spatial pattern of the abandonment is similar in the two sites (and all villages in the valley); the oldest abandoned fields are located in the periphery (0.5–2 km away from the village) and the active fields are closest to the concentrated village settlement. The two villages are located 12 km apart with an elevation range of 300 m (3175 to 3475 m a.s.l.) at Pisang and 50 m (3400 to 3450 m a.s.l.) at Bhraka. About 42 abandoned fields were selected to assure that different age classes (see 123 Plant Ecol (2009) 201:383–399 below) were represented as equally as possible. The selected fields represent ages from 1 to 50 years of abandonment. There was some uncertainty about the age of some old-fields, thus the fields were grouped into age-classes of 10-year intervals, except age-class I (1 to 5 years). The border of the oldest abandoned fields ([35 years, age-class V) were determined by fallen stonewalls, terraces and heaps of stones. The fields are by and large evenly grazed, but not mowed except for some cutting of hay during the week before harvesting of the active fields and the return of the animals from the summer farms (Aase and Vetaas 2007). A regularly spaced systematic sampling method (Kershaw and Looney 1985) was used in each field and the longest diagonal was chosen as a transect line. Another line was made perpendicular to this, which crossed the midpoint of the longest diagonal. Plots of 1 9 1 m were regularly placed along both lines. The plot closest to the border of the old-field was always 10 m away from the edge, and the distance between plots was 10 m. Each plot was divided into four subplots and presence/absence of all rooted species inside the plot was noted. This gave a relative abundance estimate from 1 (i.e. present in 1 subplot) to 4 (present in all 4 subplots), which is a rough estimate suitable for ordination analyses. A total of 242 plots were sampled, and each age-class has between 44 and 52 plots. In each plot, total vegetation and stone cover percentage (%) were estimated visually. Soil moisture and pH were measured with a soil pH and Moisture Tester (Model DM 15; Takemura Electric Works Ltd., Japan). In addition to elevation, a clinometer-compass was used to measure the slope and aspect that were then used to calculate the relative radiation index (RRI) for each plot (Oke 1987; Vetaas 1993). The adjacent vegetation of each sampled old-field was recorded, where ‘1’ indicated presence of shrubs or trees and ‘0’ indicated other fields. All the environmental variables, their abbreviations, and their units of measurement are given in Table 1. All plant species occurring inside each 1-m2 plot (i.e. species richness) were identified following the nomenclature of Hara et al. (1978), Hara and Williams (1979), Hara et al. (1982), Press et al. (2000). The growth form of each plant recorded was categorised into five types; forbs, graminoid, fern, shrub and tree (Appendix Table A1) using the method of Sarmiento et al. (2003). Plant Ecol (2009) 201:383–399 387 Table 1 List of abbreviated environmental variables, their full form, statistical summary and criteria of measurement followed Short form Full form Minimum Temgra Temporal gradient Elev Elevation Forest Forest 0.0 0.6 ± 0.0 1.0 0.5 Closeness of each plot, ‘1’ if close to forest, ‘0’ other Probe estimation (1 dry to 9 moist) 0.2 3175.0 Mean ± S.E. 2.1 ± 0.1 3370.0 Maximum 4.5 SD 1.0 3475.0 Measurement criteria Age constrained CCA first axis sample score Altimeter value, meter above sea level (m a.s.l.) Mois Moisture 0.5 2.9 ± 0.1 9.5 1.9 pH % Of H+ concentration 5.0 6.5 ± 0.0 7.7 0.3 Probe estimation RRI Relative radiation index 0.5 0.9 ± 0.0 1.0 0.1 Composit value of aspect, slope and latitude Spp. no Total species number 6.0 14.4 ± 0.2 25.0 3.5 Ston cov Stone cover % 0.0 37.9 ± 2.2 99.0 34.1 Agecl Age class 1.0 2.9 ± 0.1 5.0 1.4 Veg cov Vegetation cover % 5.0 70.6 ± 1.4 99.0 22.2 Count Visual estimate % Temporal data with 10 years interval Visual estimate % SD stands for standard deviation, S.E. stands for standard error for mean, H+ is for hydrogen ion, and % is the percentage Numerical data analysis The relationship among the explanatory variables was explored by correlations. Detrended Correspondence Analysis (DCA) was used to explore the species composition of all plots. Default options, such as detrending by segments, Hill’s scaling and downweighting of rare species were used. The first two axes were then correlated with the explanatory variables. DCA was used to measure beta diversity within each age class by estimating the length of the ordination axes, i.e. species turnover in standard deviation units (SD) (Hill and Gauch 1980). We used SD-units of axis-I and II, together with the total inertia (eigenvalue) to evaluate if there was divergent or convergent development pattern in the succession. Species richness: alpha diversity Age is defined here as the real observed age of abandonment of each old-field after interviewing local people. The age-constrained Canonical Correspondence Analysis (CCA) axis-I sample score from ordination is called the temporal gradient hereafter. Age is classified into 5 age-classes for simplicity. Species richness was related to the temporal gradient by means of a Generalised Linear Model (GLM) (McCullagh and Nelder 1989; Nelder and Wedderburm 1972). When the response variable is expected to have a Poisson distribution, e.g., species count data, GLM is able to link the expected response to the explanatory variables with a log-link function. The main explanatory variable used here is the temporal gradient. This method of using age-constrained CCA-axis-I sample scores is similar to Lepš et al. (2001); Bartolome et al. (2004). Species richness was regressed against the temporal gradient using GLM. The temporal gradient is the main environmental variable and was changed into a continuous variable. This constrained continuous temporal variable is more reliable than the age interval variable, i.e. age class. A preliminary analysis shows the high correlation between the real age of the plots and the age constrained CCA-axis-I sample score which was higher than 0.88. The age of abandonment increases with the distance to the village centre, thus the successional change in species composition is spatially structured where plots close to each other in space and time of abandonment have more similar species composition than those that are more distance in space and time. Thus autocorrelation is part of the phenomenon under study. We view this analyse as test of three different patterns; no relationships, linear (positive or 123 388 Plant Ecol (2009) 201:383–399 negative) and unimodal. It is not the temporal gradient as such that cause the pattern, thus this is not an explicit inferential statically test of causality. Due to this we have not used any spatial models to check for spatial autocorrelation. To evaluate monotonic versus unimodal trend along the temporal gradient, we tested the significance of the additional deviance explained by a second order polynomial term against a linear term. Regression was done between all measured variables against species richness to evaluate the strength of the explanatory variable. We also checked the relation with % vegetation cover, which will indicate a potential sampling effect; i.e. high cover yields many individuals thus enhanced probabilities for more species (May 1975). The adequacy of the fitted models was confirmed by plotting standardised residuals against fitted values, and by the normal probability plots of the fitted values (Crawley 1993). An F-test statistics was used, as the deviance was under dispersed (Hastie and Pregibon 1993). The CCA-axis-I species score was also used to measure the relative successional optimum for each plant species occurring in the dataset (Lepš et al. 2001). CANOCO Version 4.5 (ter Braak 2002) and its graphical program CANODRAW (Smilauer 2002) were used to analyse the relationship between variation in species composition and predictor variables. Regression analysis between species richness and different environmental variables of old abandoned fields was done with the statistical program S-PLUS (Anonymous 2002). Results Environmental correlation There is a high significant correlation (r = 0.875) between the age class and temporal gradient (Table 2). Similarly, % vegetation cover, % stone cover and pH show significant negative correlation with age class and temporal gradient, while forestproximity (plots that were close to forest or woody species) and soil moisture showed significant positive correlation with both age class and temporal gradient. Moisture showed a negative correlation with % stone cover, pH and elevation (Table 2). Environmental correlation with DCA axes The summary of DCA results (Table 3) showed that axis-I have a high eigenvalue (0.55) and is correlated with the temporal gradient. The length of the temporal gradient was 4.8 SD-units, which means that most species showed a complete turnover and a unimodal response with the measured environmental variables. Axis-I represents a complex gradient and correlates with most of the environmental variables (Table 3). Axis-II separated the two sites due to the significant correlation with relative radiation index, elevation and moisture (Fig. 2). On an average the elevation is higher at Bhraka, and the species at upper part of the ordination diagram are more common in this site. The successional changes along axis-I take place in parallel in the two sites. The most pronounced Table 2 Environmental correlation coefficient matrix of explanatory variables (n [ 100, P B 0.05, r C |0.195|) Variables Age class Age class 1.0 Temgra 0.8757 Temgra Veg cov Ston cov RRI pH Elev Mois 1.0 Veg cov -0.3232 -0.4129 1.0 Ston cov -0.214 -0.2202 0.188 RRI -0.1285 -0.2317 0.1316 pH -0.3079 -0.3035 0.028 0.4089 0.0359 Elev -0.0687 -0.2199 -0.0104 0.1496 -0.0598 0.1742 Forest 0.3437 0.4144 -0.2661 -0.1309 -0.1802 0.0003 -0.0839 Mois 0.214 0.2639 0.0418 -0.2729 0.0914 -0.5622 -0.2195 Bold entries are statistically significant coefficients Full forms of the variables are given in the Table 1 123 Forest 1.0 -0.0806 1.0 1.0 1.0 1.0 -0.0127 1.0 Plant Ecol (2009) 201:383–399 389 Table 3 Environmental correlation (weighted) and summary of DCA in the major three axes I II III Environmental correlation with DCA axes Temporal gradient Elevation Forest (closeness) Moisture 1.00 -0.26 0.02 -0.50 0.48 0.10 0.45 -0.10 -0.27 0.24 -0.54 -0.49 pH -0.25 Relative radiation index -0.20 -0.57 -0.16 Stone cover (%) -0.18 Age class Vegetation cover (%) 0.80 0.30 0.16 0.45 0.50 0.05 -0.30 -0.38 -0.39 0.39 DCA summary from dataset Eigenvalues 0.55 0.23 0.15 Lengths of gradient 4.76 2.65 2.58 1.00 0.61 0.40 Species–environment correlations Cumulative percentage variance of species data 14.30 20.20 24.10 Cumulative percentage variance of species–environment relation 48.80 56.80 0.00 Bold entries at the environmental correlation are statistically significant coefficients where, n [ 100, P B 0.05, r C |0.195| difference between the two sites is that plots in Bhraka are located in the valley bottom and do not have high radiation index on average such as the other site, Pisang. Analyses of species composition and beta diversity A total of 136 species was recorded in the study area (Appendix A1). They belonged to 40 different families among which Asteraceae was the most dominant, and grasses were the second largest family. Annual herbs, such as, Malva neglecta Wall. ex Sweet, Capsella bursa-pastoris (L.) Madik., Chenopodium album L., Convolvulus arvensis L., Fagopyrum esculentum Moench., Medicago lupunia L. and Brassica rapa Roxb., (Fig. 2) are dominant in the youngest fields, found towards the negative end of the DCA axis-I. Woody perennials such as Rosa sericea Lindl., Lonicera obovata Royle ex Hook. f. & Thom., Berberis ceratophylla G. Don, Juniperus indica Bertol. and Pinus wallichiana A. B. Jack. are more prominent in the oldest fields. They occurred towards the right end of the DCA axis-I (Fig. 2). Trifolium pratense L., Atremisia gmelinii Weber ex Stechm., Artemisia caruifolia Buch.-Ham., Cyanoglossum zeylanicum (Vahl ex Hornem.) Thunb. ex Lehm. and Taraxacum eriopodum DC., are found at the middle part of the DCA space and are common species along the entire temporal gradient. Graminoids are quite dominant in some agriculture fields. Phelum alpinum L. is abundant in early pioneer stages, whereas Brizia medica L. and Brachypodium sylvaticum (Huds.) P. Beauv are dominant in later phases. The most common grass, Pennisetum flaccidum Griseb., has its optimum in the middle of the temporal gradient, but is found all along the temporal gradient. The beta diversity (length of gradient in SD-units) along the temporal gradient is shown in Fig. 3. There is a clear increasing trend in the length of gradients with increasing age class of the abandoned fields along the DCA axis-I, but the trend is not so clear along axis-II. Total inertia, i.e. sum of all eigenvalues, shows an increasing trend with age class. This clearly says the degree of divergence and heterogeneity increased with the age of abandoned fields as indicated by the length of the gradients of the DCA axis-I and their eigenvalues. Patterns of species richness in abandoned fields and their environmental characters Percentage vegetation cover had a surprisingly negative correlation with age class (Table 2). There was high vegetation cover (almost 100%) in some recently abandoned fields (age-class I), and some of the oldest abandoned fields of age-class V had vegetation cover of only 5%. Plots that were in the oldest abandoned fields had some plots with maximum % stone cover but some recently abandoned plots also had almost 90% cover. pH of the plots was between 5 and 7.7, with the moisture gradient between 1 (driest) and 9.5 (wettest) (Table 1). These plots belonged to two sites almost 12 km apart (Fig. 1), at an elevation gradient of 3175 to 3475 m a.s.l. The relative radiation index (RRI) values are between 0.1 to 1 (Table 1). Total number of species present in each age class and their mean and maximum values obtained after summing in SPLUS are given in Table 4. There is a clear unimodal response in species richness along the temporal gradient, with highest 123 390 Plant Ecol (2009) 201:383–399 Fig. 2 DCA biplot of environmental variables and species (black dots). Species weight ranges between 1 and 100% in the inclusion rule passed by 60 species only. Full name and author of each plant species is given in Appendix Table A1. Abbreviation of environmental variables represented by ‘bold arrow’ is given in Table 1 Fig. 3 Beta diversity patterns among age classes and total dataset estimated by DCA SD-units for axes I, II and total inertia. Symbols ‘1–5’ and ‘Total’ inside the figure box represent age classes and total dataset. There is a clear pattern of increasing degree of divergence as well as heterogeneity in this old-field succession species richness found at the mid successional phase (Table 5 and Fig. 4). After testing this significant model with the null model it showed a significant result (Table 5). Among other environmental variables, age class and temporal gradient showed a significant non-linear dependence with species 123 Plant Ecol (2009) 201:383–399 Table 4 Age classes, their interval, number of plots and range in species number per plot and mean value 391 Classes Interval (years) Plots number 1 1–5 52 2 6–15 52 3 16–25 4 Total no. of species in all plots Minimum no. of species per plot Maximum no. of species per plot Mean no. of species per plot 68 6 25 13 66 9 23 15 48 64 7 22 14 26–35 44 87 8 20 15 5 [36 46 95 6 23 15 Total 1–50 242 136 6 25 14 Table 5 Regression statistics for different explanatory variables regressed against species richness using generalised linear models Models Resid. Df Resid. Dev Test Df Expl. Dev F-value Pr(F) Temporal gradient 240 209 1.0 3.6 4.3 \0.05 Age class 240 209 1.0 4.0 4.1 \0.05 Vegetation cover (%) Stone cover (%) 240 240 198 212 1.0 1.0 15.0 0.7 18.0 0.8 \0.001 ns RRI 240 213 1.0 0.1 0.1 ns pH 240 209 1.0 4.0 4.2 \0.05 Elevation 240 209 1.0 3.4 4.0 \0.05 Forest 240 209 1.0 4.0 4.3 \0.05 Moisture 240 194 1.0 18.2 23.1 \0.001 poly(temporal gradient, 2) 239 184 2.0 29.0 19.0 \0.001 poly(Age class, 2) 239 206 2.0 7.0 4.0 \0.05 poly(Vegetation cover, 2) 239 198 2.0 15.0 9.0 \0.001 poly(Moisture, 2) 239 193 2.0 20.0 13.0 \0.001 Second order polynomial Total degrees of freedom (Df) is 242 and null deviance (Dev) is 213. Resid. = residual, Expl. = explained, for other abbreviations see Table 1 richness. However, vegetation cover, pH, elevation and moisture showed linear relations. Stone cover and RRI did not show any significant relationships (Table 5). Discussion Landscape and vegetation pattern Fig. 4 Plant species-richness pattern along the continuous temporal gradient is modelled by canonical correspondence analyses. The unimodal response is fitted by generalised linear model, see Table 5 The main factor that explains the trends in species composition was the differences in time of abandonment, whereas secondary factors are difference in radiation index, moisture and pH. Thus the results confirm the expected pattern for a succession development found in the literature (Bazzaz 1996), however, some patterns have site-specific explanation 123 392 and some may relate to scale both in time and space. For instance, the positive correlation between fields close to forest and the temporal gradient (Table 2) is because fields that were abandoned a long time ago are closer to forest at the periphery of the agriculture landscape, whereas those fields still in practice are closer to the concentrated village settlement. Elevation was not significantly correlated with age class (Table 2). Research that evaluates temporal patterns in species richness in steep terrace cultivation (average inclination of slope is 30°) (Dobremez 1976; Vetaas and Grytnes 2002; Vetaas 2002) has to take into account the effect of elevation, especially when a space-for-time substitution approach is used (Bazzaz 1996; Matthews 1979; Pickett 1989). Since elevation is rejected as a factor in this study we interpret the temporal pattern in species richness and in species composition in the two sites as one general pattern. There is a significant negative correlation between soil pH and moisture (Table 2). Positive correlations are found among the soil moisture, age class and temporal gradient, but pH decreases with abandonment. This pattern of soil pH matches with Arbelo et al. (2006) but not with Blatt et al. (2005), and many others such as Prach and Rehounkova (2006); Dickie et al. (2007). Most of them have found a positive correlation between pH and age of abandonment and, similarly, with soil moisture. Soil in the study area has a very high pH (Mong and Vetaas 2006) and a decrease in pH as the vegetation cover increased is to be expected due to more litter and humus. Vegetation cover does not increase with the temporal gradient even though stone cover is higher in young old-fields. In fact, the temporal gradient is negatively correlated with vegetation cover (Table 2). This is similar to Bonet and Pausas (2004) but contrasts with the general pattern of development in secondary succession (Bazzaz 1996; Brown et al. 2006; Drury and Nisbet 1973; Horn 1974). However, old-field succession is sometimes different in this respect. Production in some old-field sites is found to increase drastically in the first years after abandonment (Odum 1960, 1969), because of heavy fertilization in the time period before abandonment. High cover ([80%) in abandoned fields of age-class I is not related to the use of chemical fertilizers in our study, but may be due to the decomposition of previously added farmyard manure. 123 Plant Ecol (2009) 201:383–399 High cover of herbs and annual grasses in the younger abandoned fields has also been found by other authors (Bartha et al. 2003; Bonet and Pausas 2004; Otto et al. 2006; Pickett 1982; Ruprecht 2005). The temporal gradient is highly correlated to the first DCA axis (Table 3), as is expected in old-field succession studies and has been documented by Sarmiento et al. (2003) and Arbelo et al. (2006). The successional sequence along the temporal gradient (CCA axis-I) shows a pattern where herbs and grasses are common in the early phase, whereas shrubs, trees and perennial herbs dominate in the later phases (Appendix Table A1). This change in growth form dominance is a general pattern in secondary succession and old-field succession (Bazzaz 1996; Ruprecht 2005). Plants in the very early stages are often common weeds from the agricultural fields, which agrees with the findings of Teira and Peco (2003); Bazzaz (1996); Sarmiento et al. (2003); Bonet and Pausas (2004) and Ruprecht (2005). Diversity pattern We found a significant unimodal pattern between species richness and the temporal gradient (Fig. 4). A unimodal pattern of species richness during old-field succession was also found by Horn (1974); Bazzaz (1996); Pickett (1982); Brown and Southwood (1987); Teira and Peco (2003); Bonet (2003) and Bonet and Pausas (2004). However, there are also studies that show a linear increasing pattern (Carson and Barrett 1988; Collins et al. 1995; Odum 1969; Otto et al. 2006; Ruprecht 2005; Sarmiento et al. 2003; Tramer 1975). High species richness at mid succession is commonly explained by the overlap of pioneer species (light demanding) and the species in the mature phase (shade tolerant) (Bazzaz 1996). High species richness at the intermediate phase may also be due to intermixing of the different growth forms as explained above (annual herbs versus perennial woody plants). Bonet (2003) argued that the mid-successional phase has more functional groups such as annual, biennials forbs and grasses which give high species richness. According to Bazzaz (1996) species richness is positively correlated with structural heterogeneity created by irregular substrate (e.g., stones and pebbles) and the beta diversity of the surrounding vegetation. The increase of species richness in the first phase of Plant Ecol (2009) 201:383–399 succession (age-class I) is inevitable as the biomass and number of individuals increase (sampling effect). This is enhanced by local factors such as high seed input from land use practices. Hay is collected every year (Aase and Vetaas 2007) from different places during late summer before the animals return from the summer grazing. This hay is dried for the winter fodder on the youngest abandoned fields close to the village. This will be supplying a heavy load of seeds to the ground, and may cause a very high variation in number of species in the plots from age-class I. High richness at early stages has also been found in other studies such as Sarmiento et al. (2003) and Teira and Peco (2003). Explanations for the high within-ageclass variation of species richness could be local factors such as percent stone cover (used to prevent erosion when the field was active) and seed input from hay drying. The decrease in the oldest abandoned fields is more of a puzzle with respect to the local site factors. First, there is no proper forest cover with shady habitats and second, these sites are closer to the surrounding forest with a more diverse influx of seed; both factors would be expected to enhance the number of species. Grazing is presumably the main factor in inhibiting a forest cover to develop. The pH is thus decreasing and it does not develop into calcareous grassland known for its high richness in temperate zone of Europe (Butaye et al. 2005). However, the nutrient level may be lower in these fields since they have not been recently manured by compost. The degree of beta diversity in different age classes (Fig. 3), indicate a divergent pattern of secondary succession. This pattern of succession matches with the findings of several other studies (del Moral 1995; Ferweda 1987; Herben et al. 1993; Rikhari et al. 1993; Sarmiento et al. 2003). All these studies indicated that surrounding vegetation plays an important role for the composition and structure of the mature vegetation. We argue that in addition to the surrounding vegetation, high grazing pressure and interactions between species and spatial heterogeneity among the fields will all contribute to a divergent pattern. The homogenization function by canopy cover is not present since pine trees and junipers are present only as scattered individuals. Lack of a closed canopy of trees or 393 shrubs is most likely due to grazing since both goats and sheep eat the seedlings of Juniperus spp. and Pinus wallichiana (Mong and Vetaas 2006). Shrubs such as Berberis spp., Cotoneaster spp. and Rosa spp. seem to represent different developmental stages on more arid locations with higher pH. Seedlings of both Pinus wallichiana and Juniperus spp. were found under bushes of Berberis ceratophylla G. Don, Lonicera obovata Royle ex Hook. f. & Thom., Astragalus rhizanthus Royle ex Benth., Rosa sericea L. and Hippophae tibetana Schltdl. in these locations. These bushes may act as nursery bushes for the Pinus spp. seedlings (cf. Bonet and Pausas 2004). A contrasting development is found in more moist and assumingly more nutrient-rich locations with lower pH, where herbs such as Anaphalis trinervis, Aster himalaicus, Silene gonosperma and Campanulatum pallidia are dominant. We conclude that there is a high number of species at the mid succession, related to an intermingling of different growth forms. Local factors such as haymaking enrich the seedbank which may give an enhanced amount of species richness in the early to mid successional phase. The reason for the decrease in the older phase is not related to competitive exclusion from canopy dominates. The divergent development seems to be related to differences in soil moisture and related pH-values. The area is semi-arid and the ability to retain soil moisture is an important factor behind differences in species composition. We expect a pine forest to develop if grazing is reduced, which may be likely since reduced agriculture production will reduce the demand for manure made from pine and animal dung. Acknowledgements We would like to thank to John Arvid Grytnes, Einar Heegard, Richard Telford, Department of Biology, University of Bergen for their valuable and inspiring suggestion related to statistical analyses and the Norwegian State Education Loan Fund (Lånekassen) for providing funding. Professor Pramod Kumar Jha, Head, Central Department of Botany, Professor Ram Prasad Chaudhary, Central Department of Botany and Govind Prasad Sharma Ghimire, Professor, Central Department of Botany and former Dean, Institute of Science and Technology, Tribhuvan University, Kirtipur, Kathmandu, Nepal are highly acknowledged due to their timely and valuable guidance. Thanks to Cathy Jenks for English correction. This study was part of a project led by O. R. Vetaas which was funded by Norwegian Research Council (project no. 148910/730). 123 394 Plant Ecol (2009) 201:383–399 Appendix Table A1 List of all plant species found on the studied plots, their family, abbreviation, growth form, CCA axis-I species score and frequency (Freq.) of occurrence of individual species S. no. Species name Family Abbreviation Growth form CCAI score Freq. 1 Brassica rapa Roxb. Brassicaeae Bras rap Forb -2.8 3 2 Capsella bursa-pastoris (L.) Madik. Brassicaeae Caps bur Forb -2.8 5 3 Chesneya cuneata (Benth.) Ali. Fabaceae Ches cun Forb -2.8 1 4 Dactylis glomerata L. Poaceae Dact glo Grassa -2.8 1 5 Galium acutum Edgew. Rubiaceae Gali acu Forb -2.8 3 6 Oxytropis microphylla (Pall.) DC. Fabaceae Oxyt mic Forb -2.8 1 7 Poa annua L. Poaceae Poa ann Grass -2.8 2 8 Trigonella pubescens Edgew. ex Baker Fabaceae Trig pub Forb -2.8 1 9 10 Triticum aestivum L. Elymus thomsonii (Hook. f.) Melderis. Poaceae Poaceae Trit aes Elym tho Grass Grass -2.8 -2.6 3 6 11 Fagopyrum esculentum Moench. Polygonaceae Fago esc Forb -2.5 31 12 Malva neglecta Wallr. Malvaceae Malv neg Forb -2.3 53 13 Chenopodium ambrosioides L. Chenopodiaceae Chen amb Forb -2.3 8 14 Elsholtzia eriostachya (Benth.) Benth. Lamiaceae Elsh eri Grass -2.1 14 15 Arisaema flavum (Forssk.) Schott Araceae Aris fla Forb -2.0 13 16 Themeda triandra Forssk. Poaceae Them tri Grass -2.0 2 17 Phleum alpinum L. Poaceae Phle alp Grass -2.0 66 18 Chenopodium album L. Chenopodiaceae Chen alb Forb -1.9 78 19 Cirsium falconeri (Hook. f.) Petr. Asteraceae Cirs fal Forb -1.9 41 20 Cannabis sativa L. Cannabaceae Cana sat Forb -1.8 14 21 Erysimum benthamii Monnet. Brassicaeae Erys ben Forb -1.7 3 22 Medicago lupulina L. Fabaceae Medi lup Forb -1.7 6 23 Axyris hybrida L. Chenopodiaceae Axyr hyb Forb -1.5 115 24 Picris hieracioides L. Asteraceae Picr hie Forb -1.5 25 Fagopyrum dibotrys (D. Don) H. Hara Polygonaceae Fago dib Forb -1.2 44 2 26 Miscanthus nepalensis (Trin.) Hack. Poaceae Misc nep Grass -1.2 1 27 Solsola nepalensis Grubov. Chenopodiaceae Sols nep Forb -1.2 6 28 Convolvulus arvensis L. Convolvulaceae Conv arv Forb -1.1 69 86 29 Bromus himalaicus Stapf. Poaceae Brom him Grass -1.1 30 Eritrichium minimum (Brand) H. Hara Boraginaceae Erit min Forb -1.0 15 31 Artemisia gmelinii Weber ex Stechm. Asteraceae Arte gme Forb -1.0 152 32 Medicago falcata L. Fabaceae Medi fal Forb -0.9 56 33 Rumex nepalensis Spreng. Polygonaceae Rume nep Forb -0.9 4 34 Polygonum tubulosum Boiss. Polygonaceae Poly tub Forb -0.7 58 35 Gerbera nivea (DC.) Sch.Bip. Asteraceae Gerb niv Forb -0.6 68 36 Nepeta leucophylla Benth. Lamiaceae Nepe leu Forb -0.6 45 37 Equisetum arvense L. Equisetaceae Equi arv Fern -0.5 5 38 Erodium stephanianum Willd. Geraniaceae Erod ste Forb -0.5 36 39 40 Morina polyphylla Wall. ex Dc. Cynoglossum zeylanicum (Vahl ex Hornem.) Thunb. ex Lehm. Dipsacaceae Boraginaceae Mori pol Cyno zey Forb Forb -0.4 -0.4 9 180 123 Plant Ecol (2009) 201:383–399 395 Table A1 continued S. no. Species name Family Abbreviation Growth form CCA-I score Freq. 41 Lotus corniculatus L. Fabaceae Lotu cor Forb -0.2 14 42 Geranium donianum Sweet Geraniaceae Gera don Forb -0.2 21 43 Verbascum thapsus L. Scrophulariaceae Verb tha Forb -0.2 42 44 Acronema nervosum H. Woff. Apiaceae Acro ner Forb -0.1 10 45 Pennisetum flaccidum Griseb. Poaceae Penn fla Grass -0.1 115 46 Trifolium pratense L. Fabaceae Trif pra Forb 0.0 201 47 Artemisia caruifolia Buch.-Ham. Asteraceae Arte car Forb 0.0 132 48 Galium aparine L. Rubiaceae Gali apa Forb 0.1 53 49 Arabidopsis himalaica (Edgew.) O. E. Schulz. Brassicaeae Arab him Forb 0.3 9 50 Crepis tibetica Babc. Asteraceae Crep tib Forb 0.3 1 51 Gnaphalium affine D. Don Asteraceae Gnap aff Forb 0.3 9 52 Rubus pungens Cambess. Rosaceae Rubu pun Forb 0.3 2 28 53 Salvia hians Royle ex Benth Lamiaceae Salv hia Forb 0.3 54 Plantago erosa Wall. Plantaginaceae Plan ero Forb 0.3 95 55 56 Stellaria patens D. Don Taraxacum eriopodum DC. Caryophyllaceae Asteraceae Stel pat Tara eri Forb Forb 0.5 0.5 24 118 57 Persicaria polystachya (Wall. ex Meisn.) H. Gross Polygonaceae Pers pol Forb 0.6 42 58 Deschampsia caespitosa (L.) P. Beauv. Poaceae Desc cae Grass 0.6 110 18 59 Pedicularis nodosa Pennell Scrophulariaceae Pedi nod Forb 0.7 60 Dipsacus inermis var. mitis Wall. Dipsacaceae Dips ine Forb 0.7 25 61 Swertia ciliata (D. Don ex G. Don) B. L. Burtt. Gentianaceae Swer cil Forb 0.7 73 62 Erianthus rufipilus (Stueud.) Griseb. Poaceae Eria ruf Grass 0.8 84 63 Halenia elliptica D. Don. Gentianaceae Hale eli Forb 0.8 5 64 Lepidium apetalum Willd. Brassicaeae Lepi ape Forb 0.9 5 65 Bupleurum hamiltonii N. P. Balakr Apiaceae Bupl ham Forb 0.9 36 66 Cotoneaster integrifolius (Roxb.) G. Klotz. Rosaceae Coto inr Shrub 1.0 13 67 Silene gonosperma var. himalayensis (Rohrb.) Bocq. Caryophyllaceae Sile gon Forb 1.1 40 68 Potentilla sericea L. Rosaceae Pote ser Shrub 1.3 21 69 Anemone rivularis Buch.-Ham.ex DC. Ranunculaceae Anem riv Forb 1.3 12 70 71 Lonicera obovata Royle ex Hook. f. & Thom. Selinum wallichianum (DC.) Raizada & Saxena Caprifoliaceae Apiaceae Loni obo Seli wal Shrub Forb 1.5 1.5 9 5 72 Epipactis royleana Lindl. Orchidaceae Epip roy Forb 1.5 9 73 Aster himalaicus C. B. Clarke Asteraceae Aste him Forb 1.6 28 74 Heracleum nepalense D. Don Apiaceae Hera nep Forb 1.7 6 75 Campanula pallida Wall. Campanulaceae Camp pal Forb 1.7 15 76 Artemisia biennis Willd. Asteraceae Arte bie Forb 1.7 8 77 Clematis graveolens Lindl. Ranunculaceae Clem gra Forb 1.7 44 78 Lobelia seguinii var.doniana (Skottsb.) E. Wimm. Campanulaceae Lobe seg Forb 1.8 46 79 Gaultheria trichophylla Royle Ericaceae Gaul tri Forb 1.8 1 80 Potentilla anseriana L. Rosaceae Poti ans Forb 1.8 1 81 Potentilla cuneata Wall. ex Lehm. Rosaceae Pote cun Forb 1.8 2 82 Juniperus indica Bertol. Cupressaceae Juni ind Shrub 1.9 11 83 Euphrasia platyphylla Pennell Scrophulariaceae Euph pla Forb 2.0 28 123 396 Plant Ecol (2009) 201:383–399 Table A1 continued S. no. Species name Family Abbreviation 84 Arabis pterosperma Edgew. Brassicaeae Arab pte 85 Persicaria nepalensis (Meisn.) H. Gross Polygonaceae Pers nep 86 Taraxacum nepalense Soest. Asteraceae Tara nep Growth form CCA-I score Freq. Forb 2.0 43 Forb 2.0 2 Forb 2.1 40 11 87 Andropogon munroi C. B. Clarke Poaceae Andr mun Grass 2.1 88 Briza media L. Poaceae Briz med Grass 2.1 57 89 Erigeron uniflorus L. Asteraceae Erig uni Forb 2.1 12 90 Dicranostigma lactucoides Hook. f. & Thomson Papaveraceae Dicr lac Forb 2.1 4 91 Hedysarum campylocarpon H. Ohashi Fabaceae Hedy cam Forb 2.2 3 92 Malaxis muscifera (Lindl.) Kuntze Orchidaceae Mala mus Forb 2.2 27 93 Thymus linearis Benth. Lamiaceae Thym lin Forb 2.2 66 94 Gentiana pedicellata (D. Don.) Griseb. Gentianaceae Gent ped Forb 2.3 15 95 Oxytropis williamsii Vassilcz. Fabaceae Oxyt wil Shrub 2.3 7 96 Gentiana crassuloides Bureau & Franch. Gentianaceae Gent cra Forb 2.3 9 97 Aster indamellus Grier. Asteraceae Aste ind Forb 2.3 30 98 99 Origanum vulgare L. Danthonia cumminsii Hook. f. Lamiaceae Poaceae Orig vul Dant cum Forb Grass 2.3 2.4 2 4 100 Potentilla lineata Trev. Rosaceae Pote lin Forb 2.4 2 101 Valeriana jatamansii Jones Valerianaceae Vale jat Forb 2.5 7 102 Anaphalis triplinervis (Sims.) C. B. Clarke Asteraceae Anap tri Forb 2.5 15 103 Ajuga bracteosa Wall. ex Benth. Lamiaceae Ajug bra Forb 2.6 4 104 Leontopodium jacotianum Beauv. Asteraceae Leon jac Forb 2.6 2 105 Rosa sericea Lindl. Rosaceae Rosa ser Shrub 2.6 8 106 Pinus wallichiana A. B. Jackson Pinaceae Pinu wal Tree 2.6 18 107 Caltha palustris Tamura Ranunculaceae Calt pal Forb 2.7 6 108 Hippophae tibetana Schltdl. Elaeagnaceae Hipp tib Shrub 2.7 20 109 Eragrostis nigra Nees ex Steud. Poaceae Erag nig Grass 2.7 27 110 Potentilla fructicosa L. Rosaceae Pote fru Shrub 2.7 9 111 Tanacetum gossypinum Hook. f. Asteraceae Tana gos Forb 2.8 31 112 Brachypodium sylvaticum (Huds.) P. Beauv. Poaceae Brac syl Grass 2.9 33 113 114 Carex orbicularis Boott. Salvia nubicola Wall. ex Sweet Cyperaceae Lamiaceae Care orb Salv nub Grass Forb 2.9 3.0 13 3 115 Stellera chamaejasme L. Thymelaeaceae Stel cha Cushion 3.0 3 116 Androsace globifera Duby Primulaceae Andr glo Cushion 3.0 2 117 Berberis ceratophylla G. Don Berberidaceae Berb cer Shrub 3.1 9 118 Thalictrum alpinum L. Ranunculaceae Thal alp Shrub 3.1 16 119 Ephedra gerardiana Wall ex. Stapf. Ephedraceae Ephe ger Shrub 3.1 3 120 Pterocephalus hookeri (C. B. Clarke) Diels Dipsacaceae Pter hoo Shrub 3.1 4 121 Polygonatum verticillatum (L.) All. Polygonaceae Poly ver Forb 3.2 4 122 Gentiana robusta King ex Hook. F. Gentianaceae Gent rob Forb 3.3 11 123 Allium hypsistum Stern. Alliaceae Alli hyp Forb 3.4 2 124 Astragalus rhizanthus Royle ex Benth. Fabaceae Astr rhi Shrub 3.4 5 125 Chenopodium foliosum (Moench) Asch. Chenopodiaceae Chen fol Forb 3.4 1 126 Cuscuta europaea L. Convolvulaceae Cusc eur Forb 3.4 1 123 Plant Ecol (2009) 201:383–399 397 Table A1 continued S. no. Species name Family Abbreviation Growth form CCA-I score Freq. 127 Delphinium williamsii Munz. Ranunculaceae Delp wil Forb 3.4 2 128 Deyeuxia pulchella (Griseb.) Hook. f. Poaceae Deye pul Grass 3.4 3 129 Elephantopus scaber L. Asteraceae Elep sca Forb 3.4 1 130 Herminium macrophyllum (D. Don) Dandy Orchidaceae Herm mac Forb 3.4 2 131 Juncus triglumis L. Juncaceae Junc tri Grass 3.4 5 132 Krascheninnikovia ceratoides (L.) Gueldenst. Chenopodiaceae Kras cer Forb 3.4 1 133 Lilium nepalense D. Don Liliaceae Lili nep Forb 3.4 2 134 Saussurea stracheyana (Kuntze) Lipsch. Asteraceae Saus str Forb 3.4 2 135 Sibbaldia cuneata Hornem. ex Kuntze Rosaceae Siba cun Forb 3.4 1 136 Viola pilosa Blume Violaceae Viol pil Forb 3.4 4 Nomenclature follows Hara et al. (1978, 1982); Hara and Williams (1979) and Press et al. (2000). The table is arranged according to the values of successional optimum. This successional optimum is the CCA axis-I species score. Each value is on a relative scale, high negative values signify the species of the youngest (age-class I) abandoned field and the highest positive values signify the species of the oldest abandoned field (Lepš et al. 2001) a Grass means graminoid References Aase TH, Vetaas OR (2007) Risk management by communal decision in trans-himalayan farming: Manang valley in central Nepal. Hum Ecol 35:453–460. doi:10.1007/s10745006-9057-6 Anonymous (1995) Iso-climatic map of mean annual precipitation. ICIMOD/MENRIS, Kathmandu, Nepal Anonymous (1999) Climatological records of Nepal 1995– 1996. Department of Hydrology and Metrology, Kathmandu, Nepal Anonymous (2002) S-PLUS 6.1 professional release 1. Copyright(c) 1988–2002. Insightful Corporation Arbelo CD, Rodriguez-Rodriguez A, Guerra JA, Mora JL, Notario JS, Fuentes F (2006) Soil degradation processes and plant colonization in abandoned terraced fields overlying pumice tuffs. Land Degrad Dev 17:571–588. doi:10.1002/ ldr.735 Bard GE (1952) Secondary succession on the Piedmont of New Jersey. Ecol Monogr 22:195–216. doi:10.2307/1943565 Bartha S, Meiners SJ, Pickett STA, Cadenasso ML (2003) Plant colonization windows in a mesic oldfield succession. Appl Veg Sci 6:205–212. doi:10.1658/1402-2001(2003)006[020 5:PCWIAM]2.0.CO;2 Bartolome JW, Fehmi JS, Jackson RD, Allen-Diaz B (2004) Response of a native perennial grass stand to disturbance in California’s Coast Range Grassland. Restor Ecol 12:279–289. doi:10.1111/j.1061-2971.2004.00355.x Bazzaz FA (1996) Plants in changing environments: linking physiological, population, and community ecology. Cambridge University Press, Cambridge Blatt SE, Crowder A, Harmsen R (2005) Secondary succession in two south-eastern Ontario old-fields. Plant Ecol 177:25–41. doi:10.1007/s11258-005-2018-0 Bonet A (2003) Secondary succession of semi-arid Mediterranean old-fields in south-eastern Spain: insights for conservation and restoration of degraded lands. J Veg Sci 56:213–233 Bonet A, Pausas JG (2004) Species richness and cover along a 60year chronosequence in old-fields of southeastern Spain. Plant Ecol 174:257–270. doi:10.1023/B:VEGE.0000 049106.96330.9c Brown VK, Southwood TRE (1987) Secondary succession: patterns and strategies. In: Gray AJ (ed) Colonization, succession and stability. Blackwell Scientific Publications, Oxford, pp 315–338 Brown CS, Mark AF, Kershaw GP, Dickinson KM (2006) Secondary succession 24 years after disturbance of a New Zealand high-alpine cushionfield. Arct Antarct Alp Res 38:325–334 Butaye J, Honnay O, Adriaens D, Delescaille LM, Hermy M (2005) Phytosociology and phytogeography of the calcareous grasslands on Devonian limestone in Southwest Belgium. Belg J Bot 138:24–38 Carson WP, Barrett GW (1988) Succession in old field plant communities: effects of contrasting types of nutrient enrichment. Ecology 69:984–994. doi:10.2307/1941253 Clements FE (1936) Nature and structure of the climax. J Ecol 24:252–284. doi:10.2307/2256278 Collins SL, Adams DE (1983) Succession in grasslands: thirtytwo years of change in a central Oklahoma tallgrass prairie. Vegetatio 51:181–190. doi:10.1007/BF00129437 Collins SL, Glenn SM, Gibson DJ (1995) Experimental analysis of intermediate disturbance and initial floristic composition: decoupling cause and effect. Ecology 76:486–492. doi:10.2307/1941207 Connell JH (1978) Diversity in tropical rain forests and coral reef. Science 199:1302–1310. doi:10.1126/science.199. 4335.1302 123 398 Crawley MJ (1993) GLIM for ecologists. Blackwell Scientific Publications, Oxford del Moral R (1995) Caprice and predictability during early primary succession on Mount St. Helens. Am J Bot 82:33 Dickie IA, Schnitzer SA, Reich PB, Hobbie SE (2007) Is oak establishment in old-fields and savanna openings context dependent? J Ecol 95:309–320. doi:10.1111/j.1365-2745. 2006.01202.x Dobremez JF (1976) Le Nepal: ecologie et biogeographie. Centre National de la Recherche Scientifique, Paris Drury WH, Nisbet ICT (1973) Succession: contribution no. 90 from the scientific staff, Massachusetts Audubon Society. J Arnold Arbor 54:331–368 Egler F (1952) Vegetation science concepts I. Initial floristic composition, a factor in old-field vegetation development. Vegetatio 4:412–417. doi:10.1007/BF00275587 Ferweda W (1987) The influence of potato cultivation on the natural bunchgrass páramo in the Colombian Cordillera Oriental. Internal report no. 220. Hugo de Vries Laboratory, University of Amsterdam Glenn-Lewin DC (1980) The individualistic nature of plant community development. Vegetatio 43:141–146. doi: 10.1007/BF00121026 Hara H, Williams HJ (1979) An enumeration of the flowering plants of Nepal. British Museum Natural History, London Hara H, Stearn WT, Williams HJ (1978) An enumeration of the flowering plants of Nepal. British Museum Natural History, London Hara H, Chater AO, Williams HJ (1982) An enumeration of the flowering plants of Nepal. British Museum Natural History, London Hastie TJ, Pregibon D (1993) Generalized linear models. In: Chambers JM, Hastie TJ (eds) Statistical models in S. Chapman & Hall, London, pp 195–247 Herben T, Krahulec F, Hadibcova V, Skalova H (1993) Small scale variability as a mechanism for large-scale stability in mountain grasslands. J Veg Sci 4:163–170. doi:10.2307/3236101 Hill MO, Gauch HG (1980) Detrended correspondence analysis: an improved ordination technique. Vegetatio 42:47– 58. doi:10.1007/BF00048870 Horn H (1974) The ecology of secondary succession. Annu Rev Ecol Syst 5:25–37. doi:10.1146/annurev.es.05. 110174.000325 Kershaw KA, Looney JHH (1985) Quantitative and dynamic plant ecology. Edward Arnold, Baltimore Lepš J, Novotný V, Basset Y (2001) Habitat and successional status of plants in relation to the communities of their leafchewing herbivores in Papua New Guinea. J Environ Biol 89:186–199 Margalef R (1968) Perspectives in ecological theory. University of Chicago Press, Chicago Matthews JA (1979) Refutation of convergence in a vegetation succession. Naturwissenschaften 66:47–49. doi:10.1007/ BF00369358 May RM (1975) Patterns of species abundance and diversity. In: Cody M, Dimond JM (eds) Ecology and evolution of communities. Belknap Press, Cambridge, pp 81–120 McCullagh P, Nelder JA (1989) Generalized linear models. Chapman & Hall, London 123 Plant Ecol (2009) 201:383–399 Miehe G (1982) Vegetationsgeographische untersuchungen im Dhaulgiri ind Annapurna Himalaya, Dissertationes Botanicæ. J Cramer Vaduz 66:66 Mong CE, Vetaas OR (2006) Establishment of Pinus wallichiana on a Himalayan glacier foreland: stochastic distribution or safe sites? Arct Antarct Alp Res 38:584–592. doi: 10.1657/1523-0430(2006)38[584:EOPWOA]2.0.CO;2 Nelder JA, Wedderburm RWM (1972) Generalized linear models. J R Stat Soc [Ser A] 135:370–380. doi: 10.2307/2344614 Odum EP (1960) Organic production and turnover in old field succession. Ecology 41:34–49. doi:10.2307/1931937 Odum EP (1969) The strategy of ecosystem development. Science 164:262–270. doi:10.1126/science.164.3877.262 Odum EP (1971) Fundamentals of ecology. Saunders Company, Philadelphia Oke TR (1987) Boundary layer climates. Metheuen, London Oosting HJ (1942) An ecological analysis of the plant communities of Piedmont, North Carolina. Am Midl Nat 28:1–126. doi:10.2307/2420696 Otto R, Krusi BO, Burga CA, Fernandez-Palacios JM (2006) Old-field succession along a precipitation gradient in the semi-arid coastal region of Tenerife. J Arid Environ 65:156–178. doi:10.1016/j.jaridenv.2005.07.005 Palmer MW (1990) Spatial scale and patterns of species– environment relationships in hardwood forest of North Carolina Piedmont. Coenoses 5:79–87 Pickett STA (1982) Population patterns through twenty years of oldfield succession. Vegetatio 49:45–49. doi: 10.1007/BF00051566 Pickett STA (1989) Space-for-time substitution as an alternative to long-term studies. In: Likens GE (ed) Long-term studies in ecology: approaches and alternatives. SpringerVerlag, New York, pp 110–135 Pickett STA, McDonnell MJ (1989) Changing perspectives in community dynamics: a theory of successional forces. Trends Ecol Evol 42:241–245. doi:10.1016/0169-5347(89) 90170-5 Pickett STA, Collins SL, Armesto JJ (1987) Models, mechanisms and pathways of succession. Bot Rev 53:335–371. doi:10.1007/BF02858321 Piqueray J, Bisteau E, Botun G, Mahy G (2007) Plant communities and species richness of the calcareous grasslands in southeast Belgium. Belg J Bot 140:157–173 Prach K, Rehounkova K (2006) Vegetation succession over broad geographical scales: which factors determine the patterns? Preslia 78:469–480 Press JR, Shrestha KK, Sutton DA (2000) Annotated checklist of the flowering plants of Nepal. The Natural History Museum, London Quarterman E (1957) Early plant succession on abandoned cropland in the central basin of Tennessee. Ecology 38:300–309. doi:10.2307/1931689 Rikhari HC, Negi GCS, Ram J, Singh SP (1993) Humaninduced secondary succession in an Alpine Meadow of Central Himalaya, India. Arct Antarct Alp Res 25:8–14. doi:10.2307/1551474 Ruprecht E (2005) Secondary succession in old-fields in the Transylvanian Lowland (Romania). Preslia 77:145–157 Plant Ecol (2009) 201:383–399 Sarmiento L, Llambi LD, Escalona A, Marquez N (2003) Vegetation patterns, regeneration rates and divergence in an old-field succession of the high tropical Andes. Plant Ecol 166:63–74. doi:10.1023/A:1023262724696 Slobodkin LB, Sanders H (1969) On the contribution of environmental predictability to species diversity. In: Woodwell GM, Smith HH (eds) Diversity and stability in ecological systems. Brookhaven symposia on biology, vol 22. pp 82–95 Smilauer P (2002) CanoDraw for Windows 4.0 Teira AG, Peco B (2003) Modelling oldfield species richness in a mountain area. Plant Ecol 166:249–261. doi:10.1023/ A:1023200517564 ter Braak CJF (2002) CANOCO-A FORTRAN program for canonical community ordination by (partial) (detrended) (canonical) correspondance analysis, principal component analysis and redundancy analysis. Wageningen Tilman D (1986) Evolution and differentiation in terrestrial plant communities: the importance of the soil resource: light gradients. In: Diamond J, Case TJ (eds) Community ecology. Harper and Row, New York, pp 359–380 Tramer EJ (1975) The regulation of plant species diversity on an early successional old field. Ecology 56:905–914. doi: 10.2307/1936300 399 Tunnell SJ, Engle DM, Jorgensen EE (2004) Old-field grassland successional dynamics following cessation of chronic disturbance. J Veg Sci 15:431–436. doi:10.1658/1100-9233 (2004)015[0431:OGSDFC]2.0.CO;2 Vetaas OR (1993) Spatial and temporal vegetation changes along a moisture gradient in north-eastern Sudan. Biotropica 25:164–175. doi:10.2307/2389180 Vetaas OR (1997) The relationships between floristic gradients in a primary succession. J Veg Sci 8:665–676. doi: 10.2307/3237371 Vetaas OR (2002) Global changes and local effects: research agenda for the future research. In: Chaudhary RP (ed) Vegetation and Society: their interaction in the Himalayas. Tribhuvan University, Nepal and Bergen University, Norway, pp 1–10 Vetaas OR, Grytnes JA (2002) Distribution of vascular plant species richness and endemic richness along the Himalayan elevation gradient in Nepal. Glob Ecol Biogeogr 11:291–301. doi:10.1046/j.1466-822X.2002.00297.x Whittaker RJ (1987) An application of detrended correspondence analysis and non-metric multidimensional scaling to the identification and analysis of environmental factor complexes and vegetation structures. J Ecol 75:363–376. doi:10.2307/2260424 123