Abstract

From a standpoint of phytosociological research, little is known about the phytocoenosis found on the volcanoes of Central America. This paper analyses the distribution of the vegetation on the volcano of San Pedro in terms of its species-richness, composition, structure and abundance, and the possible relationships between these components and the changes in elevation and orientation that occur there. We divided the study area into three altitudinal belts between 2,400 and 3,020 m a.s.l. where carried out 36 inventories, each one in an area of 0.1 ha. We then applied multivariate analysis to classify and order the data in the matrix obtained from the frequency of the sampled plants. Our results lead us to propose two mixed cloud-forest associations within the class Alnetea acuminatae. The first, Saurauio oreophilae-Alnetum acuminatae ass. nova, is found on the more humid western side, while the second, Adianto andicolae-Quercetum peduncularis ass. nova, appears in sunnier and less shady sites, mainly on the east face. As part of this latter association, we also identified the new subassociation festucetosum amplissimae subass. nova. These syntaxa are part of the alliance Oreopanacion xalapensis all. nova, which we have created to embrace the mesophytic montane forests dominated by broad-leaved species.

Key words
Alnetea acuminatae; cloud forest; Phytosociology; vegetation; volcanic mountain

INTRODUCTION

Recent volcanic activity in Mesoamerica has given rise to highly abrupt terrain where pyroclastic materials have helped generate fertile soils. This fact, together with its position at a biogeographical crossroads, has favoured the establishment of great diversity in plant communities (Lauer et al. 2003LAUER W, RAFIQPOOR MD & BENDIX J. 2003. Vergleichende Geoökologie der nördlichen (Mexiko) und südlichen Randtropen sowie der inneren Tropen (Ecuador). Edit. Akademie der Wissenschaften und der Literatur, 154 p.). Besides the important effect that the mountainous terrain has on the region’s climate, the influence of the Pacific and Atlantic (Caribbean and Gulf of Mexico) Oceans must be taken into account as the cause of a certain asymmetry in the rainfall records on the different faces of these mountains. There is a climatic difference between the slopes that drain into the Pacific in Mexico and Guatemala and those that drain into the Caribbean and Gulf of Mexico (Barthlott et al. 2005BARTHLOTT W, MUTKE J, RAFIQPOOR D, KIER G & KREFT H. 2005. Global centers of vascular plant diversity. Nova Acta Leopold 92(342): 61-83. [accessed 2020 Nov]. 06]. https://www.researchgate.net/profile/Holger_Kreft/publication/215672852_Global_centers_of_vascular_plant_diversity/links/0fcfd50472816a625b000000/Global-centers-of-vascular-plant-diversity.pdf.). The Mesoamerican Pacific slope is of great floristic, biogeographical and bioclimatic interest due to the great variation that is found there. Ranges in this territory lies between the subtropical high-pressure belt and the intertropical convergence zone; furthermore, it is exposed to the humid winds off the Pacific and possesses a great altitudinal and latitudinal range. All these factors cause a barrier and rain-shadow effect (Macías-Rodríguez et al. 2014MACÍAS-RODRÍGUEZ MA, PEINADO LORCA M, GIMÉNEZ DE AZCÁRATE J, AGUIRRE MARTÍNEZ JL & DELGADILLO RODRÍGUEZ J. 2014. Clasificación bioclimática de la vertiente del pacífico mexicano y su relación con la vegetación potencial. Acta Bot Mex 109: 133-165., 2017MACÍAS-RODRÍGUEZ MA, GIMÉNEZ DE AZCÁRATE J & GOPAR-MERINO LF. 2017. Bioclimatic systematization of Sierra Madre Occidental (Mexico) and its relationship with vegetation belts. Polibotanica 43(6): 125-163.). There is a dry season that lasts 2–6 months that has a significant effect on the vegetation (Rzedowski 2006RZEDOWSKI J. 2006. Vegetación de México. 1ª Edición digital. México D.F. (MX) Comisión Nacional para el Conocimiento y Uso de la Biodiversidad (CONABIO). [accessed 2020 Dec 02]. https://www.biodiversidad.gob.mx/publicaciones/librosDig/pdf/VegetacionMx_Cont.pdf.
https://www.biodiversidad.gob.mx/publica... ). The vegetation types are related to the plant migration from cooler climates (Holarctic) towards the tropical zones of Central and South America (Stehli & Webb 1985STEHLI FG & WEBB SD (Eds). 1985. The Great American Interchange. New York (NY): Plenum, 532 p.). This phenomenon, along with the uplift of geologically young mountain ranges, has led to high levels of allopatric specification and great numbers of range-restricted endemic species (Gentry 1982aGENTRY AH. 1982a. Neotropical floristic diversity: phytogeographical connections between Central and South America, Pleistocene climatic fluctuations, or an accident of the Andean orogeny? Ann Missouri Bot Garden 69: 557-593., bGENTRY AH. 1982b. Patterns of Neotropical plant species diversity. Evol Biol 15: 1-84.).

Monzón et al. (2000)MONZÓN J, BAILEY A & SCHUSTER J. 2000. Los escarabajos (Cerambycidae y Scarabaeoidea) como indicadores para establecer prioridades en la conservación de bosques nubosos en Guatemala. Revista UVG 10: 13-16. [accessed 2021 Feb 23]. https://res.cloudinary.com/webuvg/image/upload/v1537813245/WEB/Servicios/Editorial%20universitaria/PDF/10/revista10_.pdf. have postulated the existence of a continuous corridor during the Pleistocene glacial periods between the volcanoes that form the volcanic chain that runs parallel to the Pacific coast in Guatemala. These mountains continue northwards as ridges that run along the Pacific coast in Mexico and, somewhat less clearly, along the coast of the Gulf of Mexico. For this reason, these authors have identified this volcanic chain as the most biogeographically diverse area of Guatemala, and the one with the greatest number of endemic species. Numerous studies of the flora and vegetation of the uplands of Guatemala have revealed a great richness and diversity of vascular plants (Véliz 1989VÉLIZ ME. 1989. Caracterización de la comunidad de Canac (Chiranthodendron pentadactylon Larreategui) en el volcán de Acatenango. Guatemala. [degree thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. (Unpublished)., 1998, 2000, Viñals 1993VIÑALS JF. 1993. Estudio de la composición florística de las cimas de los Volcanes Acatenango, Agua, Atitlán, Fuego, Santa María, Santo Tomás (Pecul), Tacaná, Tajumulco y Zunil en la República de Guatemala. [degree thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. [accessed 2020 Nov 11] https://digi.usac.edu.gt/bvirtual/RESUMEN/inf0605.htm. (Unpublished)., Islebe & Cleef 1994ISLEBE GA & CLEEF AM. 1994. Alpine plant communities of Guatemala. Flora 190(1): 79-87., Islebe et al. 1994ISLEBE GA, CLEEF AM & VELÁZQUEZ A. 1994. Especies leñosas de la Sierra de los Cuchumatanes y de la cadena volcánica, Guatemala. Acta Bot Mex 29: 83-92., Spooner et al. 1998SPOONER DM, HOEKSTRA R, VAN DEN BERG RG & MARTÍNEZ V. 1998. Solanum sect. Petota in Guatemala; taxonomy and genetic resources. Amer J Potato Res 75: 3-17., Paiz Merino 2001PAIZ MERINO Y. 2001. Estudio florístico de las comunidades vegetales de la península de Manabique, Izabal. [degree thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. [accessed 2020 Dec 08]. http://biblioteca.usac.edu.gt/tesis/01/01_0942.pdf. (Unpublished)., Véliz et al. 2001VÉLIZ ME, GALLARDO N, VÁZQUEZ M & LUARCA R. 2001. La Vegetación Montana de Guatemala. Ciencia y Tecnología 2001(1): 3-61. [accessed 2020 Nov 06]. http://glifos.senacyt.gob.gt/digital/fodecyt/fodecyt%201999.35.pdf., 2006VÉLIZ ME. 1998. Composición florística de la meseta alta de la Sierra de los Cuchumatanes, Huehuetenango, Guatemala. Revista Ciencia y Tecnología de la USAC 70(1): 29-34., Pineda 2004PINEDA RT. 2004. Estudio florístico de las especies arbóreas y arbustivas en la zona intangible del volcán Ipala, Ipala, Chiquimula y Agua Blanca, Jutiapa. [degree thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. [accessed 2020 Nov 06]. http://biblioteca.usac.edu.gt/tesis/01/01_2142.pdf. (Unpublished)., Véliz & Vargas 2006VÉLIZ ME & VARGAS J. 2006. Helechos arborescentes de Guatemala: diversidad, distribución y usos. Ciudad de Guatemala (GT): FONACON/USAC., Quedensley & Bragg 2007QUEDENSLEY TS & BRAGG TB 2007. The Asteraceae of northwestern Pico Zunil, a cloud forest in western Guatemala. Lundellia 2007(10): 49-70., Martínez-Arévalo 2012MARTÍNEZ-ARÉVALO JV. 2012. Plantas asociadas a los bosques de Abies guatemalensis (Pinaceae) del occidente de Guatemala. Rev Biol Trop 61(1): 321-333.).

The neotropical cloud forests are found between latitudes 23° N and 25° S at an altitude of 1,000–3,000 m a.s.l. (Webster 1995WEBSTER G. 1995. The panorama of Neotropical Cloud Forests. In: CHURCHILL SP, BALSLEV H, FORERO E & LUTEYN JL (Eds), Biodiversity and Conservation of Neotropical Montane Forests. New York (NY): New York Botanical Garden, p. 53-77., Islebe & Véliz 2001ISLEBE GA & VÉLIZ M. 2001. Guatemala. In: KAPPELLE M & BROWN AD (Eds), Bosques Nublados del Neotrópico. Santo Domingo de Heredia (CR): INBio, p. 231-241., Rzedowski 2006RZEDOWSKI J. 2006. Vegetación de México. 1ª Edición digital. México D.F. (MX) Comisión Nacional para el Conocimiento y Uso de la Biodiversidad (CONABIO). [accessed 2020 Dec 02]. https://www.biodiversidad.gob.mx/publicaciones/librosDig/pdf/VegetacionMx_Cont.pdf.
https://www.biodiversidad.gob.mx/publica... , Jiménez-Barrios 2009JIMÉNEZ-BARRIOS JB. 2009. Diversidad de helechos (Monilophyta) en las áreas protegidas del Corredor del Bosque Nuboso, en Purulhá, Baja Verapaz. [doctoral thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. [accessed 2020 Dec 08]. http://biblioteca.usac.edu.gt/tesis/06/06_2844.pdf.), although these limits tend to be smaller in Mexico (Rzedowski 1996RZEDOWSKI J. 1996. Análisis preliminar de la flora vascular de los bosques mesófilos de montaña de México. Acta Bot Mex 35: 25-44., 2006RZEDOWSKI J & CALDERÓN DE RZEDOWSKI G. 2005. Flora fanerogámica del Valle de México. 2nd ed., Michoacán (MX): CONABIO.). They are characterised by vegetation that responds to the continual presence of clouds or mists in movement in which the associated atmospheric humidity – known as ‘horizontal rain’ – can be added to the actual rainfall of the area (Hamilton 2001HAMILTON LS. 2001. Una campaña por los Bosques Nublados: ecosistemas únicos y valiosos en peligro. In: KAPPELLE M & BROWN AD (Eds), Bosques Nublados del Neotrópico. Santo Domingo de Heredia (CR): INBio, p. 41-50.).

These forests have been little studied in Guatemala despite their frequency and presence in a number of the country’s administrative departments (San Marcos, Huehuetenango, Quetzaltenango, Sololá, Chimaltenango and Sacatepéquez). Most of these forests are located in the uplands of the Central Guatemalan Volcanic Belt (CGVB), also referred to as the Chortis Volcanic Front by Marshall (2007)MARSHALL JS. 2007. Geomorphology and Physiographic Provinces of Central America. In: BUNDSCHUH J & ALVARADO G (Eds), Central America: Geology, Resources and Hazards. London (UK): Taylor Francis, p. 75-122. [accessed 2020 Dec 12]. https://www.cpp.edu/~marshall/costa_rica_reading/Marshall_07_Ch3_CenAm_Geomorph.pdf.
https://www.cpp.edu/~marshall/costa_rica... . This author describes these volcanoes as a range consisting of a group of aligned stratovolcanoes and calderas situated along a series of transversal faults. This chain in Guatemala is part of the larger Central American Volcanic Arc. The deforestation processes, together with the forecasts of rising average temperatures, project an unfavourable scenario for the conservation of these forests. According Villers & Trejo (1998)VILLERS L & TREJO I. 1998. El impacto del cambio climático en los bosques y áreas naturales protegidas de México. Interciencia 23: 10-19., these temperate cloud forests will be the most affected by climate change, above all under a scenario in which temperatures rise by more than 2°C and rainfall falls by 10%.

Some studies of vegetation ecology have attempted to disentangle questions relating to the description of plant communities in the CGVB. Of note are the works by Véliz et al. (2001)VÉLIZ ME, GALLARDO N, VÁZQUEZ M & LUARCA R. 2001. La Vegetación Montana de Guatemala. Ciencia y Tecnología 2001(1): 3-61. [accessed 2020 Nov 06]. http://glifos.senacyt.gob.gt/digital/fodecyt/fodecyt%201999.35.pdf., Véliz et al. (2006)VÉLIZ ME, BARRIOS SOLÍS AR & DÁVILA PÉREZ CV. 2006. Actualización taxonómica de la flora de Guatemala, capítulo I. Pinophyta (Coníferas). Ciudad de Guatemala (GT): Universidad San Carlos de Guatemala, DIGI. Proyecto 057-2006. and Pardo et al. (2009)PARDO PD, VÉLIZ ME & MÉNDEZ C. 2009. Estudio de la vegetación del volcán de San Pedro, Reserva de usos múltiples de la cuenca del Lago de Atitlán, Sololá. Revista Cient Insto Invs Quím y Biol 5(1): 65-90., which provide floristics and statistics data that reveal the presence of certain plant communities. However, they fail to provide phytosociological inventories and do not ascribe the syntaxonomy of these plant communities according to the International Code of Phytosociological Nomenclature (ICPN, Theurillat et al. 2021THEURILLAT JP, WILLNER W, FERNÁNDEZ-GONZÁLEZ F, BÜLTMANN H, ČARNI A, GIGANTE D, MUCINA L & WEBER H. 2021. International code of phytosociological nomenclature. 4th ed., Appl Veg Sci 11: 739-768.).

Vegetation studies in the CGVB using the methodology of the Zürich-Montpellier phytosociological school (Braun-Blanquet 1979BRAUN-BLANQUET J. 1979. Fitosociología. Bases para el estudio de las comunidades vegetales. Madrid (Madr.): Ed. Blume, 820 p., Géhu & Rivas-Martínez 1981GÉHU JM & RIVAS-MARTÍNEZ S. 1981. Notions fondamentales de Phytosociologie. In: Dierschke H, editors. Syntaxonomie. Vaduz (LI): J Cramer, p. 5-33.) are few and far between. It is worth highlighting the study of the volcanoes of Tajumulco, Tacaná and Acatenango by Islebe & Cleef (1994)ISLEBE GA & CLEEF AM. 1994. Alpine plant communities of Guatemala. Flora 190(1): 79-87., who describe alpine zacatonal communities but without proper analysis due to the lack of inventories and diagnostic species. Subsequently, Islebe et al. (1995)ISLEBE GA, VELÁZQUEZ A & CLEEF AM. 1995. High elevation coniferous vegetation of Guatemala: A phytosociological approach. Plant Ecol 116: 7-23. described seven herbaceous and woody communities in these volcanoes and in the neighbouring Sierra de los Cuchumatanes. Their work concentrated on community composition, distribution and dynamics without entering into a floristic or syntaxonimical diagnosis of the proposed communities.

Phytosociological studies have been performed in the volcanic uplands of Mexico. It should be pointed out the studies of Velázquez & Cleef (1993)VELÁZQUEZ A & CLEEF A. 1993. The plant communities of the volcanoes Tláloc and Pelado, México. Phytocoenol 22(2): 145-191., Almeida et al. (1994ALMEIDA L, CLEEF AM, HERRERA A, VELÁZQUEZ A & LUNA I. 1994. El zacatonal alpino del volcán Popocatépetl, México, y su posición en las montañas tropicales de América. Phytocoenol 22(3): 391-436., 2004ALMEIDA L, GIMÉNEZ DE AZCÁRATE J, CLEEF AM & GONZÁLEZ-TRÁPAGA A. 2004. Las comunidades vegetales del zacatonal alpino de los volcanes Popocatépetl y Nevado de Toluca, Región Central de México. Phytocoenol 34(1): 91-132., 2007ALMEIDA L, ESCAMILLA M, GIMÉNEZ DE AZCÁRATE J, GONZÁLEZ-TRÁPAGA A & CLEEF AM. 2007. Vegetación alpina de los volcanes Popocatépetl, Iztaccíhuatl y Nevado de Toluca. In: LUNA I, MORRONE JJ & ESPINOSA D (Eds), Biodiversidad de la Faja Volcánica Transmexicana. México (MX): Universidad Nacional de México, p. 179-198.), Islebe & Velázquez (1994)ISLEBE GA & VELÁZQUEZ A. 1994. Affinity among mountain ranges in Megamexico: A phytogeo-graphical scenario. Plant Ecol 115: 1-9., Giménez de Azcárate et al. (1997, 2003, 2009), Giménez de Azcárate & Escamilla (1999)GIMÉNEZ DE AZCÁRATE J & ESCAMILLA M. 1999. Las comunidades edafoxerófilas (enebrales y zacatonales) en las montañas del centro de México. Phytocoenolog 29(4): 449-468., Velázquez et al. (2000)VELÁZQUEZ A, GIMÉNEZ DE AZCÁRATE J, ESCAMILLA WEINMANN M, BOCCO G & VAN DER MAAREL E. 2000. Vegetation dynamics on Paricutín, a recent Mexican volcano. Acta Phytogeogr Suec 85: 71-78., Medina (2016)MEDINA C. 2016. Bases para el conocimiento de los pisos bioclimáticos, la vegetación y flora del occidente de Michoacán (México). [doctoral thesis]. Santiago de Compostela (ES-C): Universidade de Santiago de Compostela, 260 p. and Medina et al. (2020)MEDINA C, GIMÉNEZ DE AZCÁRATE J & VELÁZQUEZ A. 2020. Las comunidades vegetales del bosque de coníferas altimontano en el macizo del Tancítaro, Michoacán, México. Acta Bot Mex 127: 1-20.. Most of these studies perform a geobotanical diagnosis of the studied communities, including bioclimatic, dynamic and catenal analyses.

The aim of this work is thus to use a phytosociological methodology to provide fresh data on the phytocoenosis of the tree species present on the volcano of San Pedro, and to compare the results of the analysis with those from neighbouring areas in the CGVB and other volcanoes. It aims to provide information on the composition, ecology and distribution of the upland forests that thrive in these habitats.

MATERIALS AND METHODS

Study area

The study area was the volcano of de San Pedro in the administrative department of Sololá (western Guatemala). This volcano is south-west of Lake Atitlán, whose surface area of 125.7 km² makes it the second largest lake in Guatemala. Around this lake are located the volcanoes of Atitlán and Tolimán, both of which peak at over 3,000 m a.s.l. They are part of the CGVB, which in all contains over 320 summits and runs parallel to the Pacific coast (Pardo 2007PARDO PD. 2007. Estudio de la vegetación del Volcán San Pedro, Sololá. [degree thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. [accessed 2020 Nov 10]. http://biblioteca.usac.edu.gt/tesis/06/06_2572.pdf. (Unpublished).).

The volcano of San Pedro is covered by cloud forest from around 2,400 m a.s.l. to its summit at 3,020 m a.s.l. The study area covers around 378.23 ha; the coordinates of its summit are 14° 39’ 14.7’’N, 91° 15’ 29.76’’W. Below 2,400 m a.s.l. extends its piedmont, where most of the vegetation has been modified by anthropic and animal activity and so has been excluded from these analyses.

The volcano soils are deep, pale, and well-drained internally, and have formed on steep slopes. They have a horizontal layer (25–30 cm) whose texture is loose to loose-sandy, friable, and slightly acidic. The subsuperficial horizons are a metre deep and their texture is loose-clayey to loose-sandy, lightly acidic and with no quartz (Pardo 2007PARDO PD. 2007. Estudio de la vegetación del Volcán San Pedro, Sololá. [degree thesis]. Ciudad de Guatemala (GT): Universidad de San Carlos de Guatemala. [accessed 2020 Nov 10]. http://biblioteca.usac.edu.gt/tesis/06/06_2572.pdf. (Unpublished).).

Human settlements and tourism around Lake Atitlán have affected the local ecosystems to greater or less degree. Islebe (1993)ISLEBE GA. 1993. Will Guatemala’s Juniperus-Pinus forests survive? Environ Conserv 20: 167-168. has noted that there are very few high mountain environments in Guatemala that are have not been modified in some way.

From a biogeographical standpoint and following Rivas-Martínez et al. (2011a)RIVAS-MARTÍNEZ S, NAVARRO G & PENAS A. 2011a. Biogeographic map of South America. A preliminary survey. Int J of Geobot Res 1: 21-40., the study area lies within the Neotropical-Austroamerican Kingdom, Neotropical Subkingdom, Caribbean-Neogranadian Superregion, Caribbean-Mesoamerican Region, Chiapan-Honduran Province, Chiapaneco Sector, Guatemalan Subsector.

Bioclimatic diagnosis

The bioclimatic characterisation of the study area is based on data from three meteorological stations obtained from the period 2000–2019; belonging to the National Guatemalan Institute of Seismology, Volcanology, Meteorology and Hydrology (INSIVUMEH; https://insivumeh.gob.gt/): Santiago Atitlán (Sololá, Guatemala) at 1,605 m a.s.l., located at the base of the volcano of San Pedro, San Marcos (San Marcos, Guatemala) at 2,447 m a.s.l., and De Todos Santos Cuchumatán (Huehuetenango, Guatemala) at 2,706 m a.s.l.; the latter two stations, although outside the study area, are situated at similar elevations and have analogous climatic conditions with the study area. This information allowed to obtain the averages values for the thermal and ombric parameters of each station, which then served as references for calculating the bioclimatic indices and carrying out the appropriate territorial diagnosis (Rivas-Martínez et al. 2011bRIVAS-MARTÍNEZ S, RIVAS-SÁENZ S & PENAS A. 2011b. World bioclimatic classification system. Global Geobotany 1: 1-634.).

The bioclimatic diagrams have been made in R version 4.2.0 (R Core Team 2022R CORE TEAM. 2022. R: A language and environment for statistical computing. Vienna (AUT): R Foundation for Statistical Computing. https://www.R-project.org/.
https://www.R-project.org/... ) and the CLIMATOL version 3.1.2 package (Guijarro 2019GUIJARRO JA. 2019. Homogeneización de series climáticas con Climatol. [accessed 2021 Jan 11]. https://cran.r-project.org/web/packages/climatol/climatol.pdf.
https://cran.r-project.org/web/packages/... ) was used to make the figures.

Sampling the flora and vegetation

Data were collected using a systematic stratified sampling methodology. The stratification criterion was based on altitudinal variation (A = 2,800–3,020 m a.s.l., 18% total surface area; B = 2,600–2,800 m a.s.l., 31% total surface area; and C = 2,400–2,600 m a.s.l., 51% total surface area) and four cardinal orientations reflecting the observed climatic characteristics, identified in sampling numbering as NE(I), SE(II), SW(III) and NW (IV). The NW- and SW-facing slopes were more humid and less sunny due to convective airflow from the Pacific, whilst the NE- and SE-facing slopes are drier and sunnier. By combining the three altitudinal belts and the four aspects we created 12 sampling subareas in which three replicates were carried out. Sampling units were placed in representative rather homogeneous and well-conserved vegetation plots (Westhoff & van der Maarel 1978WESTHOFF V & VAN DER MAAREL E. 1978. The Braun-Blanquet approach. In: WHITTAKER RH (Ed). Classification of plant communities. Classification of Plant Communities vol 5-1. Dordrecht (ND): Springer, p. 287-374. DOI: 10.1007/978-94-009-9183-5_9., Braun-Blanquet 1979BRAUN-BLANQUET J. 1979. Fitosociología. Bases para el estudio de las comunidades vegetales. Madrid (Madr.): Ed. Blume, 820 p.), covering a surface of 0.1 ha (50 m long x 20 m wide); in this way all relevés were uniform.

Figure 1 shows the location of the study territory and the points of the 36 samplings carried out.

The vegetation study was conducted using the phytosociological method (Braun-Blanquet 1979BRAUN-BLANQUET J. 1979. Fitosociología. Bases para el estudio de las comunidades vegetales. Madrid (Madr.): Ed. Blume, 820 p., Géhu & Rivas-Martínez 1981GÉHU JM & RIVAS-MARTÍNEZ S. 1981. Notions fondamentales de Phytosociologie. In: Dierschke H, editors. Syntaxonomie. Vaduz (LI): J Cramer, p. 5-33.). For the syntaxonomical classification we followed the criteria of Galán de Mera (2005)GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111. and Galán de Mera et al. (2021)GALÁN DE MERA A, CAMPOS DE LA CRUZ J, LINARES-PEREA E, MONTOYA-QUINO J, TORRES-MARQUINA I & VICENTE-ORELLANA JA. 2021. A phytosociological classification of the peruvian vegetation. bioRxiv. [accessed 2022 Apr 12]. https://www.biorxiv.org/content/biorxiv/early/2021/03/17/2021.03.17.435755.full.pdf.
https://www.biorxiv.org/content/biorxiv/... . The nomenclature and description of the new syntaxonomical units are compliant with the International Code of Phytosociological Nomenclature (ICPN) (Theurillat et al. 2021THEURILLAT JP, WILLNER W, FERNÁNDEZ-GONZÁLEZ F, BÜLTMANN H, ČARNI A, GIGANTE D, MUCINA L & WEBER H. 2021. International code of phytosociological nomenclature. 4th ed., Appl Veg Sci 11: 739-768.).

To determine the flora, we used Flora de Guatemala (Standley & Steyermark 1946STANDLEY PC & STEYERMARK JA. 1946-1966. Flora of Guatemala. Fieldiana, Bot 24, Parts I-VI. Chicago (ILL): Field Museum of Natural History.-1966, Standley & Williams 1961STANDLEY PC & WILLIAMS LO. 1961-1975. Flora of Guatemala. Fieldiana, Bot. 24, Parts VII-XI. Chicago (ILL): Field Museum of Natural History.-1975, Gentry & Standley 1974GENTRY JL JR & STANDLEY PC. 1974. Flora of Guatemala. Fieldiana, Bot 24, Part X. Chicago (ILL): Field Museum of Natural History.), and for the taxonomic nomenclature we used the data base from the World Flora Online (WFO 2022WFO. 2022. World Flora Online. Published on the Internet;[accessed 2022 Apr 12] http://www.worldfloraonline.org.
http://www.worldfloraonline.org... ) website (http://www.worldfloraonline.org/).

Statistical analyses

The statistical analyses were performed using the software Ginkgo (De Cáceres et al. 2007DE CÁCERES M, OLIVA F, FONT X & VIVES S. 2007. GINKGO, a program for non-standard multivariate fuzzy analysis. Fuzzy Sets Syst 2: 41-56.). An initial data matrix of 36 inventories x 109 species was employed, which was subjected to an exploratory data analysis using multivariate classification techniques to determine the initial number of general groups. The algorithm K-Means, an iterative technique used to group objects into K partitions or clusters (De Cáceres et al. 2010aDE CÁCERES M, FONT X & OLIVA F. 2010a. The management of vegetation classifications with fuzzy clustering. J Veg Sci 21: 1138-1151. DOI: 10.1111/j.1654-1103.2010.01211.x.), was applied to the matrix generated using Euclidean distance. The initial strategy chosen was seed randomization and the number of proposed routes was 120. The number of initial groups analysed was 2 (K=2) and the test were increased to K=6.

The value used to test for the ideal number of groups was the APS (Average Partition Silhouette) index, which indicates the ideal number of clusters (Rousseeuw 1987ROUSSEEUW PJ. 1987. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. J Comput Appl Math 20: 53-65. DOI: 10.1016/0377-0427(87)90125-7.); the partition with the maximum APS value was retained. The APS indices with values near 1 indicate greater security in the grouping; by contrast, negative values and values close to -1 indicate less reliable groupings.

Thus, we considered as a decisive factor in each of the K-Means analyses that the proposed groups – albeit having a general optimum value – had positive APS values that were far from 0 as in the opposite case the values would indicate that the group was not well defined (Rousseeuw 1987ROUSSEEUW PJ. 1987. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. J Comput Appl Math 20: 53-65. DOI: 10.1016/0377-0427(87)90125-7.).

We also used the model Diagnostic Species Analysis from the program Ginkgo (De Cáceres et al. 2007DE CÁCERES M, OLIVA F, FONT X & VIVES S. 2007. GINKGO, a program for non-standard multivariate fuzzy analysis. Fuzzy Sets Syst 2: 41-56.) to study the characteristic taxa in the selected groups. This model is based on the determination used in the work by Chytrý et al. (2002)CHYTRÝ M, TICHÝ L, HOLT J & BOTTA-DUKÁT Z. 2002. Determination of diagnostic species with statistical fidelity measures. J Veg Sci 13: 79-90., De Cáceres & Legendre (2009)DE CÁCERES M & LEGENDRE P. 2009. Associations between species and groups of sites: indices and statistical inference. Ecology 90(12): 3566-3574. DOI: 10.1890/08-1823.1. and De Cáceres et al. (2010bDE CÁCERES M, LEGENDRE P & MORETTI M. 2010b. Improving indicator species analysis by combining groups of sites. Oikos 119(19): 1674-1684. DOI: 10.1111/j.1600-0706.2010.18334.x.). The significance level used was p≤0.05, with 999 permutations. Only those species with a minimum value of 0.5 in terms of their statistical power were selected.

Finally, we carried out classification (cluster, Ward’s method) and ordination (Principal Component Analysis, PCA) analyses using the program Past v. 4.1 (Hammer et al. 2001HAMMER Ø, HARPER DAT & RYAN PD. 2001. PAST - Paleontological statistics software package for education and data analysis. Palaeontol Electron 4(1): 1-9., https://www.nhm.uio.no/english/research/resources/past/). In the graphical representation of this analysis of principal components, the biplots with the vectors of the floristic variables are observed in green. The length and angle with respect to the principal component axes are important. The more parallel a vector is to the axis of a component, the more it has contributed to the creation of that component. The greater the length of a vector related to a floristic variable, the better its information is represented in the graph.

RESULTS

Floristic species richness and abundance

We identified a total of 415 plant species inside and outside our phytosociological sampling areas, belonging to 102 botanical families. The most represented family was the Asteraceae (17.52%), followed by Orchidaceae (6.57%), Poaceae (4.62%), Solanaceae (3.65%), Fabaceae (3.16%) and Lamiaceae (2.68%). The Asteraceae, Lamiaceae and Poaceae species were most frequent in the sunniest ecological niches (NE- and SE-facing), whilst the Solanaceae, Araliaceae, Piperaceae, Lauraceae, Myrsinaceae and Actinidiaceae tended to be found in shady sites (NW- and SW-facing). A SE aspect had the greatest species richness in all three altitudinal bands due to the greater number of sunshine hours and the disturbance, which favour the presence of heliophile shrubs and herbaceous plants belonging to these families.

There a gradual increase in the number of taxa with greater elevation. The 2,800–3,020 m a.s.l. altitudinal band had more taxa than the others, followed by the 2,600–2,800 m a.s.l. band. This pattern coincides with elevations on the volcano’s slopes that not only have the greatest amount of available habitat but also remain humid throughout the whole year.

Bioclimatic studies

All the stations are influenced by the Tropical Pluviseasonal bioclimate (Io ≥ 3.6; Iod2 ≤ 2.5) (Table I, Figure 2). The meteorological station at Santiago Atitlán is the station at the lowest altitude and is located in the upper Thermotropical thermotype with an upper Subhumid ombrotype. The San Marcos and Todos Santos Cuchumatán stations have an upper Mesotropical thermotype and an upper Humid ombrotype, even though the data they generate for Positive Temperature (Tp) are closer to the horizon of a lower Supratropical thermotype. Nevertheless, the values for their Thermicity Index (It) correspond with the thermotype assigned in Table I. These two stations are thus on the border between these two thermotypes.

Phytosociological studies

Figure 3 shows the values of the Silhouette Index once the successive exploratory K-means analyses had been applied to the different groups (from K=2 to K=6) in the initial data matrix (36 inventories x 109 species). The graph reveals how both indices tend to drop and the data indicate that the group K=2 is the best –– but without ruling out K=4, which had the second highest value. The APS values for each group proposed for K=2 are optimal (G-1=0.2792, G-2=0.4093).

Figure 4 depicts the classification analysis applied to the initial matrix (cophenetic correlation index: 0.8425) in which two groups of inventories appear. On the left (G-1), the inventories are characterized by the presence of Alnus acuminata Kunth, while on the right appear the 12 samples with in which Quercus peduncularis Née was abundant (G-2).

For G-1, the analysis recognizes diagnostic species such as Saurauia oreophila Hemls. (0.888; p-value<0.002), Ardisia venosa Mast. ex Donn.Sm. (0.836; p-value<0.003), Meliosma idiopoda S.F.Blake (0.816; p-value<0.002), Oreopanax echinops (Cham. & Schltdl.) Decne. & Planch. (0.764; p-value<0.002) and Oreopanax xalapensis (Kunth) Decne. & Planch. (0.760; p-value<0.015), amongst others.

In a more detailed analysis of group G-1, a data matrix was prepared using 24 relevés x 86 variables. These variables are related to 85 species and a new variable with data corresponding to the elevation of each inventory is incorporated. Then, a multivariate ordination analysis (PCA, percentage of variance accumulated by the first three axes of the 98.87%) was applied to this matrix, which revealed the existence of up to three subgroups of samples (Figure 5). This result was confirmed by applying to the group G-1 a K-means analysis for this data matrix, which gave K=3 as the maximum value for the APS (mean Silhouette=0.6877; A=0.6657, B=0.6838, C=0.7081). For subgrup A, the Altitude variable is quite significant. In fact, here we find the samplings carried out at higher altitudes in the volcano. The species that best characterize this group are Meliosma idiopoda, Ardisia venosa, Salvia curtiflora Epling, Oreopanax echinops, Roldana heterogamma H.Rob. & Brettell, Arracacia donnell-smithii J.H.Coult. & Rose, Sibthorpia repens (Mutis ex L.) Kuntze and Senecio cobanensis J.M. Coult.. For subgrup B, the floristic variables that best define this set of samples are Buddleja nitida Benth., Quercus peduncularis, Chiranthodendron pentadactylon Larreategui, Fuchsia arborescens Sims, Verbesina apleura S.F. Blake, Cestrum pacayense Francey and Roldana gilgii (Greenm.) H.Rob. & Brettell. Finally, subgrup C would be made up of six samplings and the species that stand out as characteristics of this group are Euonymus enantiophyllus (Donn. Sm.) Lundell, Quercus crispifolia Trel., Oreopanax xalapensis (Kunth) Decne. & Planch., Quercus ocoteifolia Liebm., Neonelsonia acuminata (Benth) J.M. Coult. & Rose ex Drade and Solanum appendiculatum Dunal.

The results of the analyses of the diagnostic species reveal that for G-2 the dominant species is Quercus peduncularis, with a value of 0.994 (p-value<0.001); other characteristic species include Adiantum andicola Liebm. (0.888; p-value<0.001), Festuca amplissima Rupr. (0.816; p-value<0.001), Galium mexicanum Kunth (0.816; p-value<0.001), Salvia laxiantha Benth. (0.816; p-value<0.001), Alnus acuminata (0.808; p-value<0.013) and Arbutus xalapensis Kunth (0.764; p-value<0.002).

To conclude this section, Tables II and III together show the inventories corresponding to the groups G-1 and G-2, respectively. These tables show that the dominant species in both groups of inventories correlated with the results of the analysis of the diagnostic species mentioned above.

The new syntaxa

Four new syntaxa are described.

Saurauio oreophilae-Alnetum acuminatae ass. nov. (Table II, holotypus relevé IV.A.1)

We define this plant association as a cloud forest community, located between 2,400-3,020 m a.s.l., being possible that it is also present in higher altitude volcanoes. It has a clear preference for the more humid orientations of the volcano, with frequent mists and horizontal precipitation phenomena. It sits on volcanic soils and with slopes greater than 30%.

Characteristic or differential species: Alnus acuminata, Saurauia oreophila, endemic to Mexico and Guatemala, Ardisia venosa, Meliosma idiopoda, Clethra mexicana, Cestrum pacayense, Oreopanax xalapensis and O. echinops, the latter species classified as vulnerable (VU) and with distribution only in Mexico, Guatemala and Honduras (González-Espinosa et al. 2011GONZÁLEZ-ESPINOSA M, MEAVE JA, LOREA-HERNÁNDEZ FG, IBARRA-MANRÍQUEZ G & NEWTON AC (Eds) 2011. The Red List of Mexican Cloud Forest Trees. Fauna and Flora International. Cambridge (UK): Fauna & Flora International. [accessed 2020 Dec 02]. https://www.iucn.org/es/content/red-list-mexican-cloud-forest-trees.
https://www.iucn.org/es/content/red-list... ). Lastly, they accompany a series of scandents and lianas such as Maianthemum flexuosum (Bertol.) LaFrankie, Valeriana scandens var. candolleana (Gardner) C.A. Mull., Smilax moranensis M. Martens & Galeotti and Iresine calea (Ibáñez) Standl.

The distribution area of ​​this association would be restricted to the Guatemalan subsector since we do not currently have data to reliably affirm that it is found in the Mexican volcanic chains. From the bioclimatic point of view, the association is optimal in the upper Mesotropical thermotype, and can reach the lower Supratropical, under Humid and Hyperhumid ombrotypes.

Adianto andicolae-Quercetum peduncularis ass. nova (Table III, holotypus relevé II.A.3)

This association develops on volcanic soils and with slopes greater than 30%, and in places with altitudes above 2,400 m a.s.l., located on the most exposed, steep and sunny slopes of the mountain.

Characteristic or differential species: Quercus peduncularis, Adiantum andicola, Alnus acuminata, Chiranthodendron pentadactylon, and to a lesser extent Fuchsia arborescens and F. microphylla Kunth.

The biogeographical distribution of this association are the territories corresponding to the Guatemalan subsector. Bioclimatically it is restricted to lower Mesotropical and lower Supratropical termotypes under an upper Subhumid and lower Humid ombrotypes.

Adiantho andicolae-Quercetum peduncularis festucetosum amplissimae subass. nova (Table III, holotypus relevé II.A.1)

In the open and discontinuous areas of the mature forest, with rocky soils, a set of heliophyte species appear. They occur in the samples of the SE slope and above 2,800 m a.s.l. It can be considered as a transition to the more xeric plant formations.

Characteristic or differential species: Festuca amplissima, Galium mexicanum, Salvia lasiantha Benth. and Brachypodium mexicanum (Roem. & Schult.) Link.

The biogeographical distribution of this subassociation are the territories corresponding to the Guatemalan subsector. It is distributed by the lower Supratropical thermotype under a lower Humid ombrotype.

Oreopanacion xalapensis all. nova [Holotypus: Saurauio oreophilae-Alnetum acuminatae ass. nov.]

Both associations are proposed to be included in the class Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111.. Given the distribution area of ​​these associations (Mexico – to be confirmed – and Guatemala).

The diagnostic species of this alliance are Oreopanax xalapensis, O. echinops, Fuchsia arborescens, Meliosma idiopoda, Clethra mexicana, C. pachecoana Standl. & Steyerm., Verbesina apleura and Chiranthodendron pentadactylon; and shrubs such as Fuchsia microphylla and Roldana gilgii. Most of them distributed exclusively in Mexico and Central America. From the bioclimatic point of view, the alliance is optimal in the lower Mesotropical thermotype, being able to reach the lower Supratropical, under the ombrotypes from Subhumid to Hyperhumid.

This alliance includes associations of cloud forest and mixed forest with a mixture of species of the genera Pinus spp., Abies spp., Alnus spp., Quercus spp., Arbutus spp., Oreopanax spp., Clethra spp., Cleyera spp., Saurauia spp., Tilia spp., Carpinus spp. Ternstroemia spp., Liquidambar spp., etc., along with scandent species and lianas typical of the territory. This alliance of alder forests distributed throughout Mexico and Central America are included in the order Alnetalia acuminatae Galán de Mera & Rosa in Galán de Mera, Rosa & Cáceres 2002.

DISCUSSION

Alnus acuminata (Andean alder) is a species that Table II (G-1) shows is present consistently in the inventories of this group. It is found in Mexico and other parts of Central America, Colombia, Venezuela, Ecuador, Peru, Bolivia and as far south as Argentina. It is one of the most characteristic species of the cloud forests, although it is found in other types of ecosystems. It frequently appears in oak and pine-oak forests that are sufficiently humid above 1,000 m a.s.l. (González-Espinosa et al. 2011GONZÁLEZ-ESPINOSA M, MEAVE JA, LOREA-HERNÁNDEZ FG, IBARRA-MANRÍQUEZ G & NEWTON AC (Eds) 2011. The Red List of Mexican Cloud Forest Trees. Fauna and Flora International. Cambridge (UK): Fauna & Flora International. [accessed 2020 Dec 02]. https://www.iucn.org/es/content/red-list-mexican-cloud-forest-trees.
https://www.iucn.org/es/content/red-list... ). This species gives its name to the phytosociological plant class Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111., which embraces the ombrophilous uplands tropical forests where this alder, along with other trees (Fagaceae, Lauraceae, Clethraceae, Theaceae, etc.) and many climbers and epiphytes, is the dominant species.

A revision of the literature and similar phytocoenosis described from Sierra de los Cuchumatanes (Guatemala) shows that Islebe et al. (1995)ISLEBE GA, VELÁZQUEZ A & CLEEF AM. 1995. High elevation coniferous vegetation of Guatemala: A phytosociological approach. Plant Ecol 116: 7-23. described an alder-pine community from elevations over 3,000 m a.s.l. They named it a ‘mixed disturbed forest’ and it is dominated by Pinus hartwegii Lindl., Alnus jorullensis Kunth (=Alnus firmifolia), Alnus acuminata subsp. arguta (Schltdl.) Furlow (=Alnus arguta) and Alnus acuminata (=A. ferruginea), along with an abundance of the endemic Agave hurteri Trel. These authors indicate that this community is a successional stage of the Abies guatemalensis Rehder fir forests, and consists of an association that substitutes the mixed pine-alder forest dominated by Pinus hartwegii and Agave hurteri, found at greater elevations (upper Supratropical).

In Sierra de Angangueo (states of Michoacán and Mexico) at an altitudinal range of 2,500–3,100 m a.s.l., Giménez de Azcárate et al. (2003)GIMÉNEZ DE AZCÁRATE J, RAMÍREZ MI & PINTO M. 2003. Las comunidades vegetales de la Sierra de Angangueo (estados de Michoacán y México, México): clasificación, composición y distribución. Lazaroa 24: 87-111. describe a ‘mixed forest’ community disturbed by human action that is dominated by Pinus pseudostrobus Lindl., accompanied by Quercus laurina Bonpl., Abies religiosa (Kunth) Schltdl. & Cham., Alnus acuminata subsp. arguta and Clethra mexicana DC. These authors describe it as a pre-climax community that substitutes the climax “oyamel” mixed forest (Cleyero integrifoliae-Abietetum religiosae G. Azcárate & Ramírez 2004). This last one is a mesophyllous and thermophylous association, located in the lower Supratropical and Humid bioclimatic belt. Its composition includes differential species like Pinus pseudostrobus, Clethra mexicana DC., Quercus laurina, Cleyera integrifolia (Benth.) Choisy, Smilax morannensis M. Martens & Galleotti, Cornus disciflora Moc. & Sessé ex DC., Simplocos prionophylla Hemsl. etc. All of them are absent in the association that replaces it in altitude (Sibthorpio repentis-Abietetum religiosae G. Azcárate & Ramírez 2004), and dominated exclusively in the tree layer by Abies religiosa (Giménez de Azcárate & Ramírez 2004GIMÉNEZ DE AZCÁRATE J & RAMÍREZ MI. 2004. Análisis fitosociológico de los bosques de oyamel (Abies religiosa (H.B.K.) Cham. Schlecht.) de la Sierra de Angangueo, Región Central de México. Plant Sociol 41(1): 91-100. [accessed 2020 Dec 07]. http://www.scienzadellavegetazione.it/sisv/documenti/Articolo/pdf/148.pdf.). An analogous situation has been described by Ern (1976)ERN H. 1976. Descripción de la vegetación montañosa de los estados mexicanos de Puebla y Tlaxcala. Santiago de Chile: Ed. Mueller S.A., 70 p. for the volcanoes in the Puebla Valley, and by Rzedowski et al. (1977)RZEDOWSKI J, VELA G & MADRIGAL X. 1977. Algunas consideraciones acerca de la dinámica de los bosques de coníferas en México. Ciencia Forestal 5(2): 15-35. and Rzedowski & Calderón de Rzedowski (2005) for the Sierra Nevada (both in central Mexico). Depending on the interpretation, this mixed forest formation consists of serial heliophilous coniferous forests (“oyamel” fir), with other trees, shrubs and grassland that are more or less resistant to disturbance. These authors differentiate a thermophile variant (variant with Alnus acuminata subsp. arguta), found most commonly on shady east-facing slopes and gullies, where the species finds the edaphic humidity that it requires.

Catalán-Heverástico et al. (2003)CATALÁN-HEVERÁSTICO C, LÓPEZ-MATA L & TERRAZAS T. 2003. Estructura, composición florística y diversidad de especies leñosas de un bosque mesófilo de montaña de Guerrero, México. Inst Biol UNAM Ser Bot 74(2): 209-230. did not perform any phytosociological studies but do provide a list of tree species for the mountains of Guerrero (Mexico) that includes Abies guatemalensis, Quercus salicifolia Née, Oreopanax xalapensis, Ocotea chiapensis (Lundell) Standl. & Steyerm., Persea americana Mill., Meliosma dentata (Liebm.) Urb., Clethra mexicana, Alnus acuminata and Chiranthodendron pentadactylon Larreategui, amongst other species characteristic of the upland mesophile cloud forests. Their study was performed at 2,500–2,800 m a.s.l. and reveals a floristic composition with similarities to the one we describe here. Velázquez et al. (2021)VELÁZQUEZ A ET AL. 2021. Merged phytosociological and geographical approach for multiple scale vegetation mapping as a baseline for public environmental policy in Mexico. Appl Veg Sci 24(3): e12595. in an exercise on the integration of geographic, physiognomic and geobotanical attributes of the plant formations of Michoacán (Mexico) and their relationship at the syntaxonomic level with the units defined in the state, recognize the the class Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111.. This is proposed by temporarily assigning of lower rank syntaxa, pending validation, included in the provisional order Alnetalia acuminate-jorullensae (Takaki et al. 2019TAKAKI F, VICTORIA HERNÁNDEZ A, DÍAZ RÍOS R, MALAQUÍAS GONZÁLEZ S, CARRANZA GONZÁLEZ E & BLANCO-GARCÍA J. 2019. Tipos de vegetación conforme al sistema INEGI. In: CONABIO (Ed). La biodiversidad en Michoacán. Estudio de Estado 2, vol. I. Michoacán (MX): CONABIO, p. 297-318.) and included in the Cloud Forest concept. The absence of types for the syntaxonomic units, as well as reference inventories, prevent their analysis and relationship with those here defined.

Likewise, Medina et al. (2020)MEDINA C, GIMÉNEZ DE AZCÁRATE J & VELÁZQUEZ A. 2020. Las comunidades vegetales del bosque de coníferas altimontano en el macizo del Tancítaro, Michoacán, México. Acta Bot Mex 127: 1-20. describes a number of coniferous forest communities from the volcano Tancitaro, Michoacán (Mexico), including the new subassociation quercetosum laurinae within the association Sibthorpio repentis-Abietetum religiosae G. Azcárate & Ramírez 2003, found in the lower Supratropical horizon with a Humid ombrotype, at 2,600–3,000 m a.s.l. This subassociation is based on the presence of Quercus laurina, Geranium seemannii Peyr., Galium mexicanum, Pinus pseudostrobus and Alnus jorullensis. These authors consider that this subassociation represents a catenal transition from the pine-oak forests in the Mesotropical stage to the “oyamel” forests belonging to the typical association. Further north in the Sierra Madre Occidental, Giménez de Azcárate et al. (2013)GIMÉNEZ DE AZCÁRATE J, MACÍAS-RODRÍGUEZ MA & GOPAR-MERINO F. 2013. Bioclimatic belts of Sierra Madre Occidental México: A preliminary approach. Int J Geobot Res 3: 19-35. have identified the presence of upland mesophile forests (mixed sub-evergreen macro-forests) in the shady gullies on steep slopes at around 2,000 m a.s.l., coinciding basically with the humid Mesotropical stage. The most representative species here include Alnus acuminata, Carpinus caroliniana Walter, Cleyera integrifolia (Benth.) Choisy, Clethra spp., Cornus disciflora, Garrya laurifolia Benth., Ilex quercetorum I.M.Johnst., Magnolia pacifica subsp. tarahumara A. Vázquez, Oreopanax xalapensis, Ostrya virginiana (Mill.) K.Koch, Persea liebmannii Mez, Tilia americana L., in addition to different species of Pinus spp. and Quercus spp. (Pinus herrerae Martínez, P. maximinoi H.E.Moore, P. oocarpa Schiede, Quercus candicans Née, Q. castanea Née, Q. diversifolia Née, Q. obtusata Bonpl., Q. splendens Née and Q. subspathulata Trel.). With greater elevation and proximity to the Supratropical thermotype, the more thermophile species tend to disappear and others such as Clethra rosei Britton, Pinus ayacahuite Ehrenb. ex Schltdl., P. douglasiana Martínez, P. pseudostrobus and Abies durangensis Martínez, appear, and the formations assume an appearance that is more reminiscent of the pine-oak forests.

These mixed forests of Pinus spp., Abies spp., Alnus spp., Quercus spp. and Arbutus spp. are found throughout the volcanic mountain chains in Mexico and Guatemala, at similar elevations and in analogous ecological conditions. They represent particular dynamic, catenal and transitional situations between the Abies spp. forests and the mixed Pinus spp. and Quercus spp. forests.

Based on our statistical results that reveal the existence of at least two groups of inventories with different floristic and ecological compositions, we hereby propose two new phytosociological associations to be included in the class Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111..

We describe the association Saurauio oreophilae-Alnetum acuminatae ass. nova (Table II, typus relevé IV.A.1) as an upland cloud forest found at 2,400–3,020 m a.s.l., that may also be present on volcanoes at greater altitudes. It has a clear preference for humid and slopes (over 30°) of the volcano, where mists and ‘horizontal rain’ are frequent. The characteristic species are Alnus acuminata, which is accompanied by trees and shrubs of 5–25 meters in height such as Saurauia oreophila, endemic to Mexico and Guatemala, Ardisia venosa, Meliosma idiopoda, Clethra mexicana, Cestrum pacayense Francey, Oreopanax xalapensis and O. echinops, the latter species classified as Vulnerable (VU) and found only in Mexico, Guatemala and Honduras (González-Espinosa et al. 2011GONZÁLEZ-ESPINOSA M, MEAVE JA, LOREA-HERNÁNDEZ FG, IBARRA-MANRÍQUEZ G & NEWTON AC (Eds) 2011. The Red List of Mexican Cloud Forest Trees. Fauna and Flora International. Cambridge (UK): Fauna & Flora International. [accessed 2020 Dec 02]. https://www.iucn.org/es/content/red-list-mexican-cloud-forest-trees.
https://www.iucn.org/es/content/red-list... ). Finally, it is accompanied by a series of climbers and lianas including Maianthemum flexuosum (Bertol.) LaFrankie, Valeriana scandens var. candolleana (Gardner) C.A. Mull., Smilax moranensis M. Martens & Galeotti and Iresine calea (Ibáñez) Standl.

This association is found only in the Guatemalan subsector as, for the time being, we have no data that conclusively prove that it is found on the Mexican volcanic chains. From a bioclimatic standpoint, the optimum of this association is the upper Mesotropical thermotype, although it will reach the lower Supratropical horizon in Humid and Hyperhumid ombrotypes.

In Table II, which shows the main group of the 13 inventories (subgrup A, Fig. 5), two other subgroups of inventories are apparent. The first consists of five inventories (Fig. 5, subgroup B) enriched by Quercus peduncularis, Quercus crispifolia, Adiantum andicola Liebm. and Zeugites munroanus Hemsl. The presence of Quercus species suggest that it is a transition to the formations dominated by Q. peduncularis. We thus believe that it is an ecological variant that indicates the transition to oak-dominated formations that do not require such high levels of humidity.

In the other subgroup of inventories (subgrup C, Fig. 5) Prunus salasii Standl and Billia hippocastanum Peyr are frequent. They are found above all in shady gullies below 2,730 m a.s.l., subject to erosion with deep organic soils on north-west facing slopes. The shady humid conditions encourage a proliferation of terrestrial orchids such as Goodyera striata Rchb. f.

Prunus salasii and Billia hippocastanum are two trees that reach 15–25 m height and are founded in mature undisturbed cloud forests (González-Espinosa et al. 2011GONZÁLEZ-ESPINOSA M, MEAVE JA, LOREA-HERNÁNDEZ FG, IBARRA-MANRÍQUEZ G & NEWTON AC (Eds) 2011. The Red List of Mexican Cloud Forest Trees. Fauna and Flora International. Cambridge (UK): Fauna & Flora International. [accessed 2020 Dec 02]. https://www.iucn.org/es/content/red-list-mexican-cloud-forest-trees.
https://www.iucn.org/es/content/red-list... ) in Mexico and Central America (Roskov et al. 2019ROSKOV Y ET AL. (Eds). 2019. Species 2000 ITIS Catalogue of Life: 2019 Annual Checklist. [accessed 2020 Dec 01]. www.catalogueoflife.org/annual-checklist/2019.
www.catalogueoflife.org/annual-checklist... ) on volcanoes at low-to-mid elevations (1,500–2,700 m a.s.l.) (Kappelle & Uffelen 2006KAPPELLE M & UFFELEN VAN G. 2006. Altitudinal zonation of montane oak forests along climate and soil gradients in Costa Rica. In: KAPELLE M (Ed), Ecology and Conservation of Neotropical montane oak forests. Ecological Studies (Analysis and Synthesis) 185. Heidelberg (DE): Springer, p. 39-54. DOI: 10.1007/3-540-28909-7_4.). Both species are abundant in the depressions in the volcano – known as rejoyas – caused by landslips that create large clearings in the forests. This situation, along with the constantly high humidity levels, ensure that this cohort of species appears at elevations of 2,400–2,700 m a.s.l. Thus, we propose the existence of this variant of the association Saurauio-Alnetum acuminatae on disturbed soils, which is found on the lower slopes of the volcanoes where most landslides occur. Their frequency and the natural fall of trees, together with the shady nature of these sites, favour the presence of subnitrophile species of disturbed soils such as Arracacia donnell-smithii J.H.Coult. & Rose, Maianthemum flexuosum (Bertol.) LaFrankie, Roldana heterogama H.Rob. & Brettell, Neonelsonia acuminata, Begonia oaxacana A.DC., Solanum appendiculatum and Polypodium alansmithii R.C. Morán.

The inventories in Table III (G-2, Fig. 4) correspond to mixed forests of trees up to 25 m in height dominated by Quercus peduncularis, Arbutus xalapensis, Alnus acuminata and Adiantum andicola. This latter taxon has been recognised by other authors as characteristic of the mixed oak forests of Mexico (Rodríguez-Romero et al. 2008RODRÍGUEZ-ROMERO L, PACHECO L & ZAVALA-HURTADO JA. 2008. Pteridofitas indicadoras de alteración ambiental en el bosque templado de San Jerónimo Amanalco, Texcoco, México. Rev Biol Trop 56(2): 641-656.). This phytocoenosis is clearly dominated by Quercus peduncularis, Arbutus xalapensis and, to a lesser extent, by Alnus acuminata. In some inventories appear Quercus crispifolia and Quercus ocoteifolia. (Table III).

Quercus peduncularis thrives at 900–3,000 m a.s.l. in Honduras, Guatemala, Nicaragua, Belize and Mexico (Sierra Madre Oriental). In Mexico its distribution has been severely limited by anthropic activity (CONABIO 2020CONABIO - COMISIÓN NACIONAL PARA EL CONOCIMIENTO Y USO DE LA BIODIVERSIDAD (MX). 2020. Encino rojo (Quercus peduncularis Née). [accessed 2020 Nov 30]. https://enciclovida.mx/especies/150704-quercus-peduncularis#cite_note-INECC-1.
https://enciclovida.mx/especies/150704-q... ). Normally, it is found in mixed pine-oak forests but does not form part of the montane mesophile cloud forests (González-Espinosa et al. 2011GONZÁLEZ-ESPINOSA M, MEAVE JA, LOREA-HERNÁNDEZ FG, IBARRA-MANRÍQUEZ G & NEWTON AC (Eds) 2011. The Red List of Mexican Cloud Forest Trees. Fauna and Flora International. Cambridge (UK): Fauna & Flora International. [accessed 2020 Dec 02]. https://www.iucn.org/es/content/red-list-mexican-cloud-forest-trees.
https://www.iucn.org/es/content/red-list... ); Quercus ocoteifolia is, nevertheless, characteristic of these cloud forest ecosystems.

For Chiapas uplands (Mexico), González-Espinosa et al. (1991)GONZÁLEZ-ESPINOSA M, QUINTANA-ASCENSIO PF & RAMÍREZ-MARCIAL N. 1991. Secondary succession in disturbed Pinus-Quercus forests in the highlands of Chiapas, México. J Veg Sci 2(3): 351-360. describe for a mature stage of pine-oak forests with the presence of Pinus oocarpa, Prunus serotina Ehrh., Quercus laurina Bonpl., Q. crassifolia Bonpl., Q. rugosa Née, Myrsine juergensenii (Mez) Ricketson & Pipoly, Oreopanax xalapensis and Alnus acuminata. It’s climatic conditions of that study are similar to those found in the lower parts of the volcano of San Pedro. In a later work (González-Espinosa et al. 2006GONZÁLEZ-ESPINOSA M, RAMÍREZ-MARCIAL N & GALINDO-JAIME L. 2006. Secondary succession in montane pine-oak forests of Chiapas, México. In: KAPELLE M (Ed). Ecology and Conservation of Neotropical montane oak forests. Ecological Studies (Analysis and Synthesis) 185. Heidelberg (DE): Springer, p. 209-221. DOI: 10.1007/3-540-28909-7_16.) they influence the different composition between the pine-oak forests discussed and the tropical cloud forest formations, which are even more biodiverse in terms of the plant species, containing essentially various Quercus species plus 1–3 pine species and dozens of other broad-leaved trees.

Giménez de Azcárate et al. (2003)GIMÉNEZ DE AZCÁRATE J, RAMÍREZ MI & PINTO M. 2003. Las comunidades vegetales de la Sierra de Angangueo (estados de Michoacán y México, México): clasificación, composición y distribución. Lazaroa 24: 87-111. provide a table with 17 samples from mixed pine-oak forests in which the dominant species are Pinus pseudostrobus, Quercus laurina and, to a lesser extent, Abies religiosa, Alnus spp., Arbutus spp., Salix paradoxa Kunth, Quercus rugosa, Clethra mexicana and Ternstroemia lineata DC., amongst others. Of note in this community is the proliferation of the typical shrubs found in preforestry environments with open canopies due to timber extraction. This community is found at 2,500–3,070 m a.s.l. in the upper Mesotropical and lower Supratropical horizons with upper Subhumid-lower Humid horizons. From a dynamic viewpoint, this community replaces in disturbed areas the mesophytic mixed fir formations belonging to the association Cleyero integrifoliae-Abietetum religiosae Giménez de Azcárate & Ramírez 2003GIMÉNEZ DE AZCÁRATE J, ESCAMILLA WEINMANN ME & ALMEIDA-LEÑERO L. 2009. Datos sobre la vegetación higrófila altimontana del Volcán Iztaccíhuatl (México). Lazaroa 30: 109-118.. This is a similar case to that mentioned for the same altitudinal band in Valle de México (Rzedowski et al. 1977RZEDOWSKI J, VELA G & MADRIGAL X. 1977. Algunas consideraciones acerca de la dinámica de los bosques de coníferas en México. Ciencia Forestal 5(2): 15-35., Velázquez & Cleef 1993VELÁZQUEZ A & CLEEF A. 1993. The plant communities of the volcanoes Tláloc and Pelado, México. Phytocoenol 22(2): 145-191.) and in Sierra de los Cuchumatanes (Islebe & Velázquez 1994ISLEBE GA & VELÁZQUEZ A. 1994. Affinity among mountain ranges in Megamexico: A phytogeo-graphical scenario. Plant Ecol 115: 1-9.).

Finally, Medina (2016)MEDINA C. 2016. Bases para el conocimiento de los pisos bioclimáticos, la vegetación y flora del occidente de Michoacán (México). [doctoral thesis]. Santiago de Compostela (ES-C): Universidade de Santiago de Compostela, 260 p. in a phytosociological study of the west of Michoacán (Mexico) proposed provisionally two associations that are very similar to those we detected in Guatemala. Firstly, the association Querco laurinae-Pinetum moztezumae Medina 2016MEDINA C. 2016. Bases para el conocimiento de los pisos bioclimáticos, la vegetación y flora del occidente de Michoacán (México). [doctoral thesis]. Santiago de Compostela (ES-C): Universidade de Santiago de Compostela, 260 p. is described as a mixed coniferous and broad-leaved association of sub-evergreen species that occupies slopes of average gradient, west-facing, at 2,600–2,800 m a.s.l., which contact catenally with the alder formations of Sibthorpio-Abietetum religiosae quercetosum laurinae. This phytocoenosis thrives between the upper Mesotropical-lower Supratropical stages, both of which possess a upper Subhumid-Humid ombrotype. Secondly, the association Clethro mexicanae-Pinetum pseudostrobi Medina 2016MEDINA C. 2016. Bases para el conocimiento de los pisos bioclimáticos, la vegetación y flora del occidente de Michoacán (México). [doctoral thesis]. Santiago de Compostela (ES-C): Universidade de Santiago de Compostela, 260 p. is defined as a mixed polyspecific forest, with broad-leaved and coniferous species, dominated by trees of up to 20–40 m in height, with species such as Pinus pseudostrobus, Clethra mexicana, Quercus laurina, Q. candicans, Oreopanax xalapensis, Tilia americana var. mexicana (Schltdl.) Hardin and Carpinus caroliniana. This latter association is found in gullies deep and steep valleys at 1,900-2,550 m a.s.l. in the Humid Mesotropical stage on the south face of the volcano de Tancítaro. There is a coincidence of species between these two phytocoenosis and our community; nevertheless, in our inventories Quercus peduncularis and Arbutus xalapensis clearly dominate, and there is a residual presence of P. pseudostrobus.

The reiterated presence of these mixed formations in the mountains of Mexico and Guatemala at 2,000–3,300 m a.s.l., linked to the Meso- and Supratropical thermotypes, reveals subtle but important biogeografical, bioclimatic, edaphic and geomorphological differences that enable us to differentiate phytocoenosis, each with particular floristic assemblages. The Quercus, Pinus and Alnus species, along with the other trees commonly found in these environments (e.g. Clethra spp., Cleyera spp., Saurauia spp., Tilia spp., Carpinus spp. Ternstroemia spp. and Liquidambar spp.) are important indicators of the environmental conditions and of the extent – to a greater or lesser degree – that human action has had an impact on these forest in recent times.

The phytocoenosis that we studied are found basically at lower elevations than those similarly studied in Mexico, where the climax communities above the timberline are dominated by Abies spp. and Pinus spp. forests, or by mixed forests of these species and others. The plant communities of the volcano of Sant Pedro in Guatemala is situated at a lower elevation that on the Mexican volcanoes, with the consequence that certain climax forest such as fir (Abies spp.) and pine (Pinus spp.) founded in the upper Supratropical and lower Orotropical horizons, are absent (Velázquez 1994VELÁZQUEZ A. 1994. Multivariate analysis of the vegetation of the volcanoes Tláloc and Pelado, Mexico. J Veg Sci 5: 263-270., Islebe & Velázquez 1994ISLEBE GA & VELÁZQUEZ A. 1994. Affinity among mountain ranges in Megamexico: A phytogeo-graphical scenario. Plant Ecol 115: 1-9., Velázquez & Islebe 1995VELÁZQUEZ A & ISLEBE G. 1995. Comparación fitogeográfica entre las montañas del centro de México y Guatemala. Caldasia 17: 501-508., Islebe et al. 1995ISLEBE GA, VELÁZQUEZ A & CLEEF AM. 1995. High elevation coniferous vegetation of Guatemala: A phytosociological approach. Plant Ecol 116: 7-23., Giménez de Azcárate et al. 2003GIMÉNEZ DE AZCÁRATE J, RAMÍREZ MI & PINTO M. 2003. Las comunidades vegetales de la Sierra de Angangueo (estados de Michoacán y México, México): clasificación, composición y distribución. Lazaroa 24: 87-111., Medina et al. 2020MEDINA C, GIMÉNEZ DE AZCÁRATE J & VELÁZQUEZ A. 2020. Las comunidades vegetales del bosque de coníferas altimontano en el macizo del Tancítaro, Michoacán, México. Acta Bot Mex 127: 1-20.). With the lower temperatures at greater elevations, these climax forest formations tend to lose the polyspecific characteristic that distinguishes the mixed upland forest that are studying here.

We propose Adianto andicolae-Quercetum peduncularis ass. nova (Table III, typus relevé II.A.3) as a new association of mixed forest that develops on volcanic soils with slopes around 30% at elevations over 2,400 m a.s.l. It is present in the lower Mesotropical-lower Supratropical thermotype and upper Subhumid-lower Humid ombrotype, above all on steep, sunny exposed slopes. The characteristic or differential species of this association are: Quercus peduncularis, Adiantum andicola, Alnus acuminata, Chiranthodendron pentadactylon, Fuchsia arborescens and F. microphylla. The biogeographical distribution of this association are the territories corresponding to the Guatemalan subsector.

Biogeographically, belongs the Guatemala subsector; the presence of Quercus peduncularis in other Central American countries and Mexico suggest that this association or similar ones are also present in this subsector.

It is important to add that a number of samples were enriched by the presence of Festuca amplissima, Galium mexicanum, Salvia lasiantha Benth. and Brachypodium mexicanum. All these species except the last one were detected from above 2,800 m a.s.l. as far as the summits. As mentioned in the previous section, the ascendency here of these species is due to the ‘rain-shadow’ effect, together with the more open canopy and increase in sunshine hours. These grasses tend to dry out in summer (November–May), which leads to a proliferation of wildfires and increases the impact of erosion and landslips in winter (the rainy season). In certain areas the soil is completely covered by these grasses and various layers of dry material, whose presence – together with low scrub dominated by Salvia lasiantha – slows down the development of a well-structured mature forest. In these conditions – i.e. an open and discontinuous forest canopy – heliophile plants thrive and so on this side of the volcano wildfires are commonplace. Due to its suberification, the oak Quercus peduncularis resists relatively well the extreme temperatures provoked by such fires, whose frequency mean that soils are exposed to rain and severe run-off, and landslips occur that have led to the loss of up to two metres of soil in some places of the volcano.

In the forests of Sibthorpio repentis-Abietetum religiosae (Giménez de Azcárate & Ramírez 2004GIMÉNEZ DE AZCÁRATE J & RAMÍREZ MI. 2004. Análisis fitosociológico de los bosques de oyamel (Abies religiosa (H.B.K.) Cham. Schlecht.) de la Sierra de Angangueo, Región Central de México. Plant Sociol 41(1): 91-100. [accessed 2020 Dec 07]. http://www.scienzadellavegetazione.it/sisv/documenti/Articolo/pdf/148.pdf.) and Polysticho speciosissimae-Abietetum religiosae (Medina et al. 2020MEDINA C, GIMÉNEZ DE AZCÁRATE J & VELÁZQUEZ A. 2020. Las comunidades vegetales del bosque de coníferas altimontano en el macizo del Tancítaro, Michoacán, México. Acta Bot Mex 127: 1-20.) appear scattered B. mexicanum, F. amplissima and various Salvia spp., being more abundant in their shrub forest fringes.

These species were present regularly in our inventories, above all on the south-east-facing slopes and at the highest elevations above 2,800 m a.s.l. We thus believe that this represents a subassociation within the association Adianto andicolae-Quercetum peduncularis that occurs in higher, rockier and treeless sectors, and where Brachypodium mexicanum, Festuca amplissima and Salvia lasiantha are common (festucetosum amplissimae subass. nova, relevé typus: II.A.1, Table III). This latter species is regarded by Rivera-Hernández et al. (2019)RIVERA-HERNÁNDEZ JE, FLORES-HERNÁNDEZ N, VARGAS-RUEDA AF, ALCÁNTARA-SALINAS G, CHÁZARO-BASÁÑEZ MJ & GARCÍA-ALBARADO JC. 2019. Flora and vegetation from the semiarid region of Acultzingo-Maltrata, Veracruz, Mexico. Acta Bot Mex 126: 1-26. DOI: 10.21829/abm126.2019.1433. as being typical of semi-arid thorny woodland and so is an indication of a transition towards other, more xeric plant formations.

In terms of the syntaxonomy of these two associations, it is clear that they correspond to phytocoenosis of cloud or mixed forests found in very humid mesophile environments, which leads us to link them with the class Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111.. This unit embraces the alder forests dominated by Alnus acuminata and other climbers and lianas (Galán de Mera & Vicente Orellana 2006GALÁN DE MERA A & VICENTE ORELLANA JA. 2006. Aproximación al esquema sintaxonómico de la vegetación de la región Caribe y América del Sur. An Bio 28: 3-27. https://www.biorxiv.org/content/biorxiv/early/2021/03/17/2021.03.17.435755.full.pdf.).

Nevertheless, our inventories contain species restricted to the cloud forests of Mexico and Central America and so at a syntaxonomical level we cannot include this association in the syntaxa created originally for Colombia, Peru, Bolivia or Argentina.

In Colombia the communities belonging to Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111. are described from around Lago de Tota (Boyacá), a locality at above 3,025 m a.s.l., where the characteristic species include Viburnum tinoides L.f., Oreopanax incisus (Willd. ex Schult.) Decne. & Planch., Solanum elaeagnifolium Cav., Weinmannia microphylla Kunth, Hesperomeles heterophylla Hook., Clusia multiflora Kunth, Verbesina elegans Kunth, Morella parvifolia (Benth.) Parra-Os., Passiflora mixta L.f., Bomarea multiflora (L.f.) Mirb. and Clethra fimbriata Kunth., amongst others (Rangel-Churio & Aguirre 1986RANGEL-CHURIO J & AGUIRRE J. 1986. Estudios ecológicos en la Cordillera Oriental Colombiana. III. La vegetación de la Cuenca del Lago de Tota (Boyacá). Caldasia 15: 71-75.). This study contains no syntaxonomical ascription.

The characteristic species of the Andean alder formations include Berberis lutea Ruiz & Pav., Dioscorea stenopetala Hauman, Fuchsia denticulata Ruiz & Pav., Morella pubescens (Humb. & Bonpl. ex Willd.) Wilbur, Passiflora mollissima (Kunth) L.H.Bailey, P. trifoliata Cav., Rubus urticifolius Poir. and Brachypodium mexicanum (Galán de Mera et al. 2002GALÁN DE MERA A, ROSA V & CÁCERES C. 2002. Una aproximación sintaxonómica sobre la vegetación del Perú. Clases, órdenes y alianzas. Acta Bot Malac 27: 75-103., Galán de Mera & Vicente Orellana 2006GALÁN DE MERA A & VICENTE ORELLANA JA. 2006. Aproximación al esquema sintaxonómico de la vegetación de la región Caribe y América del Sur. An Bio 28: 3-27. https://www.biorxiv.org/content/biorxiv/early/2021/03/17/2021.03.17.435755.full.pdf.). The alliance Myrico pubescentis-Alnion acuminatae Galán de Mera, Rosa & Cáceres 2002 includes the alder formations of central sector of the Andean Cordillera (Peru and Bolivia).

For the southern Bolivia and northern Argentina communities belonging to the alliance Pruno tucumanensis-Alnion acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111. have been described, with the characteristic species including Myrsine laetevirens (Mez) Arechav., Clinopodium gilliesii (Benth.) Kuntze, Kaunia lasiophthalma (Griseb.) R.M.King & H.Rob., Boehmeria caudata Sw., Podocarpus parlatorei Pilg., Polystichum montevidense (Spreng.) Rosenst., Sambucus peruviana Kunth, Prunus tucumanensis Lillo, Crinodendron tucumanum Lillo and Duranta serratifolia (Griseb.) Kuntze, amongst others.

The Cyatheo herzogii-Alnion acuminatae Galán, Campos, Linares Montoya Torres & Vicente 2020 alliance includes Andean alder vegetation from the thermotropical montane forest in soils rich in clay and hydromorphism (Galán de Mera et al. 2020GALÁN DE MERA A, CAMPOS DE LA CRUZ J, LINARES-PEREA E, MONTOYA QUINO J, TORRES-MARQUINA I & VICENTE ORELLANA JA. 2020. A phytosociological study on andean rainforest of Peru, and comparison with the surrounding countries. Plants 9(12): 1654.) with dominance of plant species Acanthus ilicifolius L., Asplenium auriculatum Sw., Aulonemia longiaristata L.G. Clark & Londoño, Cyathea herzogii Rosenst., Cyathea lechleri Mett., Geonoma orbignyana Mart., Hedyosmum sprucei Solms, Hypolepis parallelogramma (Kunze) C. Presl, Hypolepis obtusata (C. Presl) Kuhn, Inga acuminata Benth., Miconia aff asperrima Triana, Miconia aggregata, Saccoloma inaequale (Kunze) Mett., and Serpocaulon fraxinifolium.

Finally, the Escallonio pendulae-Alnion acuminatae Galán, Campos, Linares Montoya Torres & Vicente 2020 alliance lincludes the riparian forests and peat bogs of northern Peru. The main characteristic species are Bia alienata Didr., Escallonia pendula (Ruiz & Pav.) Pers., Ladenbergia oblongifolia (Humb. ex Mutis) L.Andersson, Pentacalia reflexa (Kunth) Cuatrec., Thelypteris pennata (Poir.) C.V. Morton, and Styloceras laurifolium (Willd.) Kunth.

The new associations that we describe here, restricted to Mexico (to be confirmed) and Guatemala, are included in a new phytosociological alliance Oreopanacion xalapensis all. nova [Holotypus: Saurauio oreophilae-Alnetum acuminatae], which embraces cloud and mixed forest associations with species from genera such as Pinus spp., Abies spp., Alnus spp., Quercus spp., Arbutus spp., Oreopanax spp., Clethra spp., Cleyera spp., Saurauia spp., Tilia spp., Carpinus spp. Ternstroemia spp., and Liquidambar spp., along with the typical climbers and lianas from this region.

The order Alnetalia acuminatae Galán de Mera & Rosa in Galán de Mera, Rosa & Cáceres 2002 was created to cover the alder formations of the Supratropical belt in an area stretching from Venezuela to Argentina (Andean region) (Galán de Mera et al. 2002GALÁN DE MERA A, ROSA V & CÁCERES C. 2002. Una aproximación sintaxonómica sobre la vegetación del Perú. Clases, órdenes y alianzas. Acta Bot Malac 27: 75-103.). In the hope that future studies in Mexico and Central America will gather more information on the phytocoenosis with these characteristics, we maintain the new alliance in this order, although we propose that the range of this order and of this phytosociological class be extended.

CONCLUSIONS

This study has highlighted the variation of mixed cloud forest communities at mid- and high elevations on Mexican and Guatemalan all of which gathering an ecological identities, such as altitude and orientation. Nevertheless, these phytocoenosis are not identical as there are differences in their floristic composition and elevation ranges, which are determined by their biogeographical locations and their ombroclimatic context, which is much more humid in the mountains of Chiapas sector than in those in the province of Neovolcánica (central Mexico).

Our studies must be complemented in future with other phytosociological work that will allow us to appreciate the full extent of the forest associations presents in this biogeographical complex, as well as their successional dynamics. This knowledge is essential if we are to manage these forests correctly, above all in terms of the extraction of resources and the reforestation with appropriate tree species.

These forests are, moreover, subject to continuous anthropic activity, mainly in areas of easy access and gentler slopes. Uncontrolled cutting and clearing to create agricultural land are the main negative impacts that are currently affecting these ecosystems, although we should not forget that climate change may also seriously modify its distribution and composition. Finally point out, these formations harbour many plant and animal species of great importance, of which it is worth highlighting the resplendent quetzal (Pharomachrus mocinno mocinno), endemic to Central America, and the horned guan (Oreophasis derbianus), both species that are currently classified as Endangered (IUCN Red List of Threatened Species 2020IUCN. 2020. Red List of Threatened Species. [accessed 2022 Apr 13]. https://www.iucnredlist.org/species/22678453/177970135.
https://www.iucnredlist.org/species/2267... , https://www.iucnredlist.org/). Indeed, the forests analysed form part of the diet of the horned guan and are its main habitat. Thus, it is urgent that conservation plans that will manage the exploitation of these forests’ natural resources be drawn up and put in place.

Syntaxonomical scheme

Class Alnetea acuminatae Galán de Mera 2005GALÁN DE MERA A. 2005. Clasificación fitosociológica de la vegetación de la región del Caribe y América del Sur. Arnaldoa 12(1-2): 86-111.

+ Order Alnetalia acuminatae Galán de Mera & Rosa in Galán de Mera, Rosa & Cáceres 2002

• Alliance Oreopanacion xalapensis all. nov. [Holoyypus: Saurauio oreophilae-Alnetum acuminatae, typus ass.: rel. IV.A.1, Table II]

• Saurauio oreophilae-Alnetum acuminatae ass. nov. (Holotypus: rel. IV.A.1, Table II)

• Quercus peduncularis variant

• Prunus salasii variant

• Adianto andicolae-Quercetum peduncularis ass. nova (Holotypus: rel. II.A.3, Table III)

• festucetosum amplissimae subass. nov. (rel. I.A.3 to II.B.3. Holotypus: inv. II.A.1, Table III)

ACKNOWLEDGMENTS

We thank all those people who have helped in the preparation of this work, especially Engineer Mario Véliz from the University of San Carlos of Guatemala for his help in determining the plant taxa. Likewise, we appreciate the suggestions made by the editors and referees of this journal that have managed to improve this paper; and, finally, to Michael Lockwood for proofreading the manuscript.

REFERENCES

Publication Dates

History