UNIVERSIDAD DE LOS ANDES

FACULTAD DE CIENCIAS

DEPARTAMENTO DE CIENCIAS BIOLÓGICAS

LABORATORIO DE ECOLOGÍA DE BOSQUES TROPIACLES Y PRIMATOLOGÍA

TESIS PARA OPTAR AL TITULO DE DOCTOR EN CIENCIAS

BIOLÓGICAS

Diversidad, carbono y dinámica de las comunidades de árboles en Bosques Húmedos Tropicales de Colombia

Candidata

Ana María Aldana Serrano

Defendida públicamente el 26 de enero de 2017, frente al jurado:

Dr. Pablo R. Stevenson, Profesor Titular, Universidad de Los Andes Director Dr. Timothy R. Baker, Profesor Asociado, Universidad de Leeds Evaluador Externo Dra. Eloisa Lasso, Profesora Asociada, Universidad de Los Andes Evaluadora Interna

i UNIVERSIDAD DE LOS ANDES

FACULTAD DE CIENCIAS

DEPARTAMENTO DE CIENCIAS BIOLÓGICAS

LABORATORIO DE ECOLOGÍA DE BOSQUES TROPIACLES Y PRIMATOLOGÍA

TESIS PARA OPTAR AL TITULO DE DOCTOR EN CIENCIAS

BIOLÓGICAS

Diversity, Carbon and Dynamics of Tree Communities in Colombia´s Rainforests

Candidata

Ana María Aldana Serrano

Defendida públicamente el 26 de enero de 2017, frente al jurado:

Dr. Pablo R. Stevenson, Profesor Titular, Universidad de Los Andes Director Dr. Timothy R. Baker, Profesor Asociado, Universidad de Leeds Evaluador Externo Dra. Eloisa Lasso, Profesora Asociada, Universidad de Los Andes Evaluadora Interna

ii

Con amor infinito para Alf, Sara y Lucas

iii AGRADECIMIENTOS

Una cosa me quedo clara después de los cinco años que trabajé en éste proyecto: la ciencia no es un trabajo en solitario, la ciencia es el producto de la colaboración y el trabajo en equipo. Éste trabajo, en particular, lo he realizado gracias al esfuerzo y la contribución de muchos investigadores, profesores, voluntarios y amigos. A Alf le agradezco por ser mi cómplice, mi compañero, mi confidente, por apoyarme de todas las maneras posibles durante este proceso y por ser un padre de familia ejemplar. Gracias a su visión igualitaria de las responsabilidades de la familia, permitió que Sara y Lucas no sintieran tan fuerte la ausencia de una madre apasionada por su trabajo. A Pablo gracias por tanto aprendizaje durante éstos 12 años de trabajo conjunto y amistad. Puedo decir que con nuestra amistad creció mi pasión por conocer los bosques de tierras bajas de Colombia. Ojalá algún día, no muy lejano, podamos cumplir nuestro sueño de producir la guía de plantas los bosques de Colombia. El Departamento de Ciencias Biológicas, la Facultad de Ciencias y la Vicerrectoría de Investigaciones de la Universidad de Los Andes propiciaron las condiciones laborales y apoyos económicos para el desarrollo de ésta investigación con las asistencias graduada y de investigación, así como con los proyectos semilla y la financiación de la pasantía en el exterior. La beca para mujeres en ciencia de L´Oréal – Unesco Colombia, versión 2014, fue una fuente de financiación importante para éste proyecto. Agradezco a James Richardson por su apoyo incluso desde antes que iniciara el doctorado, sus concejos y apoyo han sido invaluables. Mailyn González y Catalina González quienes al ser parte del comité de doctorado me dieron su apoyo, no sólo académico si no también anímico. Gracias a las profesoras, investigadoras y excelentes jefas Mónica Pachón, Silvia Restrepo, Camila González por confiar en mi trabajo y en mis capacidades, incluso fuera del ámbito académico. Quiero agradecer especialmente a los estudiantes de pregrado que quisieron trabajar bajo mi co-dirección y me apoyaron en la toma y el análisis de los

iv datos: Juan Sebastián González, Isabel Cristina Restrepo, Andrea Gómez, Melissa Martínez y Guillermo Rivas. Gracias a los muchos voluntarios que me acompañaron y me ayudaron en el trabajo de campo: Vanessa Rubio, Erika Rodríguez, Marcela Córdoba, Indira León, Sasha Cárdenas, Diana Pizano, Alejandra Jiménez, Camilo Quiroga, Eduardo Pinel, Edna Beltrán, Diana Acosta, Felipe Aramburo, Ángela Sánchez, Camila Monje, Ángela Perilla, Maria Juliana Pardo, Nathali Rodriguez, Maria Fernanda Torres, Juan Camilo Muñoz, Sandro Figueroa, Mónica Cardozo, Jaime Cabezas, Andrés Montes y Natalia Alvis. Todos los estudiantes y miembros del Laboratorio de Ecología de Bosques Tropicales y Primatología fueron compañía y apoyo constante durante éste tiempo. A mis colegas, colaboradores y coautores, especialmente a Luis Francisco Henao-Díaz, Sebastián González, Luisa Fernanda Casas, Sasha Cárdenas, Diego Felipe Correa, Boris Villanueva, Ángela Cano y María Natalia Umaña, su trabajo fue el cimiento para desarrollar mi proyecto. A los amigos que hice y reencontré en la Universidad de Berkeley y en el Jardín Botánico de Edimburgo, fue poco tiempo, pero la pasantía fue realmente una experiencia muy valiosa. A los evaluadores de la tesis, Eloisa Lasso y Tim Baker, quienes con sus comentarios ayudaron a mejorar la calidad de éste documento. Finalmente, a mis compañeros del doctorado, que se convirtieron en amigos y cómplices; gracias por las charlas, los cafés, los almuerzos y todos los momentos que aportaron a mi salud mental y a que siempre me sintiera acompañada en ésta hazaña.

v CONTENIDO

RESUMEN ...... 3

ABSTRACT ...... 4

INTRODUCCIÓN GENERAL ...... 5

GENERAL INTRODUCTION ...... 8

CAPÍTULO I – AVANCES SOBRE LA HISTORIA ECOLÓGICA Y EVOLUTIVA DE LOS BOSQUES HÚMEDOS TROPICALES DEL NORTE DE SUR AMÉRICA ...... 11 Resumen ...... 12 Abstract ...... 13 ENVIRONMENTAL FILTERING OF EUDICOT LINEAGES UNDERLIES PHYLOGENETIC CLUSTERING IN TROPICAL SOUTH AMERICAN FLOODED FORESTS ...... 14 Appendix 1 ...... 49 Appendix 2 ...... 50 FLOODING AND DIFFERENCES IN SOIL COMPOSITION DETERMINE BETA- DIVERSITY OF LOWLAND FORESTS IN NORTHERN SOUTH AMERICA ...... 51

CAPÍTULO II – DETERMINANTES DE LAS RESERVAS DE CARBONO DE LOS BOSQUES EN EL NORTE DE SURAMÉRICA ...... 72 Resumen ...... 73 Abstract ...... 74 DRIVERS OF BIOMASS STOCKS IN NORTHWESTERN SOUTH AMERICAN FORESTS: CONTRIBUTING NEW INFORMATION ON THE NEOTROPICS ...... 75

CAPÍTULO III – DINAMICA DE COMUNIDADES Y DE CARBONO DE LOS BOSQUES DE TIERRAS BAJAS EN EL NORTE DE SUR AMÉRICA ...... 89 TREE TURNOVER AND CARBON DYNAMICS OF SEASONALLY-FLOODED AND TERRA FIRME FORESTS OF COLOMBIA ...... 90 FOREST FRAGMENTS OF THE ANDEAN PIEDMONT AS CARBON SINKS: SHORT- TERM GAIN OF ABOVE GROUND BIOMASS IN FRAGMENTS USED BY CATTLE RANCHES ...... 115

CONCLUSIONES GENERALES ...... 125

GENERAL CONCLUSIONS ...... 128

LITERATURA CITADA ...... 130

1 ANEXOS ...... 134 ANEXO 1 – Dinámica de bosques en diferentes escenarios de tala selectiva en el magdalena medio (colombia)...... 135 ANEXO 2 – Forest biomass density across large climate gradients in northern south america is related to water availability but not with temperature ...... 148 ANEXO 3 – Dinámica, estructura y diversidad de los bosques de galería en la región de los llanos, colombia ...... 180 ANEXO 4 – Specific gravity of woody tissue from lowland Neotropical plants: variance among forest types ...... 215

2 RESUMEN

Los servicios ecosistémicos se han convertido en una importante herramienta para enfatizar la necesidad de conservar los diversos ecosistemas del mundo, ya que estos ayudan a mantener el bienestar de la humanidad. Por esta razón, el servicio de captura de carbono ha sido el centro de atención de grupos de investigación que pretenden conocer y describir la ecología de los bosques amazónicos, argumentando que la conservación de estos bosques ayuda a mitigar y disminuir los efectos del cambio climático global. El objetivo principal de ésta investigación fue estimar la capacidad de captura de carbono de bosques de tierras bajas con diferentes regímenes de disturbio y determinar en qué medida este servicio ecosistémico está influenciado por el ambiente y la historia evolutiva de los grupos de plantas que conforman estas comunidades. El sistema de estudio incluye parcelas permanentes de vegetación, de una hectárea, en bosques estacionalmente inundables y de tierra firme, bajo un gradiente de intervención humana, en las cuencas de los ríos Magdalena, Orinoco y Amazonas, lo que permitió realizar comparaciones a nivel de las condiciones ambientales y del ensamblaje filogenético de las comunidades de árboles. Los bosques estudiados se diferencian en la composición florística, diversidad filogenética, reservas de biomasa aérea y dinámica de corto plazo de las comunidades. El principal resultado de ésta investigación es que se evidencian los efectos de la inundación estacional, y, en menor medida, el clima y la historia biogeográfica sobre la estructura, composición y reservas de biomasa (carbono) de los bosques. Este proyecto provee información relevante para la toma de decisiones informadas con respecto al manejo y conservación de los bosques húmedos en Colombia, algunos de los cuales han sido muy poco estudiados.

3 ABSTRACT Ecosystem services have become an important tool to stress the need for conserving the diverse ecosystems of the world, as they help in the maintenance of human wellbeing. This is why, carbon accumulation service has been the focus of the research groups that aim at studying and uncover the ecology of the Amazon rainforest, arguing that the conservation of these forests helps to mitigate and diminish the effects of global climate change. The main objective of this research was to estimate the carbon accumulation capacity of the wet lowland forests, under different disturbance regimes and to determine to what extent this ecosystem service is influenced by environmental variables, and the evolutionary history of the groups of plants that make up the communities. Our study system includes permanent one- hectare vegetation plots, in seasonally flooded and terra firme forests, in a human intervention gradient, in three important river basins: Magdalena, Orinoco and Amazonas. This design allowed us to make comparisons on environmental variables and the phylogenetic community assemblage. The forests we studied differ in their floristic composition, phylogenetic diversity estimates, aboveground biomass stocks and the short term community dynamics. The most important result is that the strongest effects, determining these differences, come from the seasonal flooding of the plots, and the origin of the water that inundated them, and in a lesser extent, climate and biogeographic history. This research provides useful insights on Colombian rainforest ecology, for informed decision making on forest management and conservation, some of which have been poorly studied.

4 INTRODUCCIÓN GENERAL

Los bosques del Neotrópico son de los más diversos en especies de plantas (Gentry 1982), pero a la vez afrontan las mayores amenazas por la expansión de la frontera agrícola, la explotación de minerales y combustibles fósiles, entre otros (Pan et al. 2013). En las últimas décadas, Sur América ha perdido aproximadamente tres millones de hectáreas de bosque al año (FAO and JRC 2012); mientras tanto los científicos intentan conocer y entender los mecanismos que dieron origen a tan extrema riqueza (Pennington and Dick 2004, Antonelli et al. 2009, Ter Steege et al. 2013, Baker et al. 2014). Conocer estos mecanismos les permite a los tomadores de decisiones plantear estrategias de conservación que prevean el impacto de cambios locales, como el cambio de uso de la tierra, así como el cambio climático global (Ackerly et al. 2010), y que a la vez no pongan en riesgo los beneficios que la humanidad recibe de estos ecosistemas (Bunker et al. 2005).

Uno de los servicios ecosistémicos que más se ha utilizado para reforzar la necesidad de conservar los bosques del Neotrópico ha sido la capacidad que estos tienen para capturar carbono, de allí se derivan iniciativas de deforestación evitada como REDD+ (Ortega-P et al. 2010). A pesar de que esta información es relevante para el bienestar de la humanidad, también es necesario continuar estudiando los bosques en términos del valor que albergan en cuanto a las interacciones inter- específicas y la historia evolutiva de los linajes de plantas que co-existen allí (Whitfeld et al. 2012, Werner et al. 2014). Por ejemplo (Ter Steege et al. 2013), en un gran esfuerzo colaborativo de la comunidad científica, estimaron que es posible que en la actualidad sólo se conozca el 30% de las especies de árboles que existen en la cuenca amazónica.

En Colombia el estudio de los bosques de tierras bajas se hace particularmente interesante dada la división geográfica de las cordilleras y la existencia de los valles interandinos (Alizadeh et al. 2015). La riqueza hídrica de la región también ha permitido la existencia de planicies inundables que albergan bosques estacionalmente inundados por ríos con distintos orígenes edáficos (Lewis et al. 1995). Es así que en la Amazonia y en la Orinoquia se pueden encontrar bosques estacionales de Várzea, inundados por ríos con grandes cargas de sedimentos de

5 las cordilleras, y bosques estacionales de Igapó, inundados por ríos con muy pocas cargas de sedimentos, originados en la cuenca misma (Prance 1989). En el Valle del río Magdalena, los bosques de planos inundables son primordialmente Várzeas, aunque la mayoría de éstas han sido transformadas para uso agrícola (Prance 1989).

Indudablemente, algunos de los bosques de tierras bajas en Colombia tienen una historia de intervención humana que se debe tener en cuenta a la hora de realizar estudios sobre su ecología (Etter et al. 2006, Etter 2013). Algunos estudios sobre la diversidad florística y la similitud de estos bosques nos ha mostrado que la intervención humana y los factores ambientales como la inundación y el tipo de suelo, son determinantes de la diversidad y la composición florística (Stevenson et al. 2011).

En cuanto al servicio ecosistémico de captura de carbono, en Colombia se han estimado las reservas de carbono de los bosques maduros a niveles locales (Sierra et al. 2007), y a nivel nacional (Phillips et al. 2011, 2016, Alvarez et al. 2012), pero existen muy pocos estudios sobre los factores determinan las reservas de carbono y su dinámica. Por ejemplo, Peña & Duque (2013) reportaron que el tipo de suelo es un importante determinante de la cantidad de carbono que se acumula anualmente en los bosques andinos secundarios de la cordillera occidental, pero es todavía incierto si la dinámica del carbono en los bosques de tierras bajas en Colombia está mayormente determinado por factores ambientales (Feldpausch et al. 2016) o por factores como la diversidad taxonómica y funcional (Cavanaugh et al. 2014); y si los bosques de la Orinoquía y el Valle del Magdalena tienen determinantes y dinámicas similares a los bosques de la Amazonía (Phillips et al. 2008, Quesada et al. 2012, Brienen et al. 2015, Fauset et al. 2015, Honorio Coronado et al. 2015).

Esta investigación pretende estudiar los determinantes de la diversidad florística y filogenética de los bosques húmedos de tierras bajas de Colombia. Para este fin se utilizaron los métodos estandarizados de monitoreo de comunidades de árboles, como las parcelas permanentes de vegetación. Adicionalmente, se espera establecer que factores ambientales influyen en las reservas de carbono (acumulado en la biomasa aérea de los árboles) y las dinámicas del carbono en

6 éstos bosques; Igualmente, se pretende determinar cómo están relacionadas la dinámica de las poblaciones y del carbono con la historia evolutiva de las distintas comunidades y con factores ambientales. Para este fin se realizarán comparaciones entre bosques estacionalmente inundables y de tierra firme en tres regiones biogeográficas (cuencas de los ríos Amazonas, Magdalena y Orinoco) utilizando los métodos de estudio de diversidad filogenética y ensamblaje filogenético de comunidades. También se observarán los efectos de la intervención humana, dado que en estas regiones los bosques tienen diversas historias de uso y grados de fragmentación.

Las conclusiones de este estudio son de gran utilidad para tomadores de decisiones involucrados en el planteamiento de estrategias de uso sostenible de los bosques de tierras bajas en Colombia ante futuros escenarios de cambio climático global.

7 GENERAL INTRODUCTION The forests of the Neotropics are the most diverse in plant species (Gentry 1982), but, at the same time these forests face the greatest threats with the expansion of the agricultural boundary, mineral and fossil fuel exploitation, amongst other threats (Pan et al. 2013). During the last decades, South America has lost approximately three million forest hectares per year (FAO and JRC 2012); meanwhile scientists are trying to identify and understand the mechanisms that gave rise to such extreme species diversity (Pennington and Dick 2004, Antonelli et al. 2009, Ter Steege et al. 2013, Baker et al. 2014). Identifying such mechanisms helps policy makers and forest managers to outline conservation strategies that foresee the impact of local change, such as land use change, as well as global climate change (Ackerly et al. 2010). And, at the same time, do not jeopardize the benefits that humanity receives from these ecosystems (Bunker et al. 2005).

One of the ecosystem services that has been used more frequently to reinforce the need to conserve Neotropical forests has been their capacity to store carbon, initiatives such as Reducing Emissions from Deforestation and Forest Degradation – REDD + (Ortega-P et al. 2010) derive from this idea. Although this information is relevant for the human wellbeing, it is also necessary to continue studying forests in terms of the value they hold in the inter-specific interactions and the evolutionary history of the plant lineages that co-exist there (Whitfeld et al. 2012, Werner et al. 2014). For example, Ter Steege et al. (2013), in a great collaborative effort from the international cientific community, estimated that currently we only know about 30% of the tree species present in the Amazon basin.

In Colombia, the study of the rainforests is particularly interesting given the geographic division of the Andean ridges and the existence of the inter-Andean valleys (Alizadeh et al. 2015). The hydrographic richness of the region has also allowed the existence of flood plains which hold seasonally flooded forests inundated by rivers with diverse edaphic origins (Lewis et al. 1995). Consequently in the Amazon and Orinoco basins there are váreza forests, flooded by rivers with heavy loads of sediments from the Andean ridges, and igapó forests, flooded by rivers with very low sediment loads, originated in the same basin (Prance 1989). In the Magdalena river valley, forests in the flood plains are mainly várzeas, although the

8 great majority of these have been transformed for agricultural proposes (Prance 1989).

Undoubtedly, some lowland forests in Colombia have a history of human intervention that has to be taken into account when developing studies about their ecology (Etter et al. 2006, Etter 2013). Recent research on floristic diversity and similarity in these forests have shown that human intervention and environmental factors such as flooding and soil characteristics are determining on floristic composition and diversity (Stevenson et al. 2011).

Regarding the ecosystem service of carbon accumulation, in Colombia there have been etimations of carbon stocks in mature forests at the local level (Sierra et al. 2007), and at the national level (Phillips et al. 2011, 2016, Alvarez et al. 2012), but there are very few studies on the variables that determine these stocks and their dynamics. For example, Peña & Duque (2013) reported that the soil type is an important factor for the ammount of carbon accumulated annually in secondary Andean forests of the western Andean ridge, but it is still uncertain if the carbon dynamic of lowland forests in Colombia is mainly determined by environmental factors (Feldpausch et al. 2016) or by factors such as taxonomic and functional diversity (Cavanaugh et al. 2014); And if the forests of the Orinoco and Magdalena Basins are have determinants and dynamics similar to the forests in the Amazon basin (Phillips et al. 2008, Quesada et al. 2012, Brienen et al. 2015, Fauset et al. 2015, Honorio Coronado et al. 2015).

This research aims to study the determinants of the floristic and phylogenetic diversity of the rainforests in Colombia. For this purpose, we used standardized methods for tree community monitoring such as permanent forest plots. We also expect to establish which environmental variables are important for the carbon stocks (accumulated in the aboveground biomass of trees) and the carbon dynamics in these forests; Furthermore, we want to determine how the tree community dynamics and the carbon dynamics are related to the evolutionary history of the forests and the environmental factors. Accordingly, we will compare between seasonally flooded and terra firme forests in three biogeographic regions (Amazon, Magdalena and Orinoco basins), using the research methods from phylogenetic diversity and phylogenetic community assemblage. We will also evaluate the effects

9 of human intervention, as in these regions the forests have diverse histories of land use and levels of fragmentation.

The conclusions derived from this research are of great interest to decision makers involved in sustainable use strategy planning for the lowland forests in Colombia, under future scenarios of global climate change.

10

CAPÍTULO I – AVANCES SOBRE LA HISTORIA ECOLÓGICA Y EVOLUTIVA DE LOS BOSQUES HÚMEDOS TROPICALES DEL NORTE DE SUR AMÉRICA

11 Resumen

A pesar de ser ecosistemas que afrontan la inminente amenaza de la elevada tasa de deforestación en Colombia, los bosques de tierras bajas en el país no han sido suficientemente estudiados. Sabemos que existe una gran diversidad de especies de plantas, pero hasta el momento el número de especies que componen estos ecosistemas permanece incierto. Dada la escasez de estudios al respecto aún ignoramos cuales son los mecanismos que permitieron la evolución, y mantienen tan alta diversidad de especies de plantas. Con los estudios que presentamos a continuación pretendemos aportar al conocimiento sobre la diversidad beta y dar luces sobre la historia evolutiva de las especies de árboles de los bosques húmedos de tres regiones importantes de Colombia como son la Amazonía, Magdalena Medio y Orinoquía. Los dos estudios se realizaron con base en información de la abundancia de especies de árboles de más de 10cm de Diámetro a la Altura del Pecho – DAP de 32 parcelas de vegetación de una hectárea establecidas en bosques de tierra firme y estacionalmente inundables. La diversidad beta de los bosques húmedos de las regiones estudiadas está determinada principalmente por variables ambientales, resultado que le da soporte a la teoría de nichos, sin embargo, una gran proporción de la variación permanece sin ser explicada por las variables estudiadas. En cuanto a la diversidad filogenética beta, los bosques estacionalmente inundables tienen mayor representación de algunos linajes de plantas (eudicotiledoneas), lo que demuestra que el filtro ambiental es fuerte, no sólo a nivel de especies, sino también a nivel de grupos de plantas.

12 Abstract

Regardless of facing an imminent threat due to the elevated deforestation rate in Colombia, the lowland forests of the country remain understudied. We know there is great plant diversity, but the number of species that make up these ecosystems are still unknown. Given the lack of studies on the subject we ignore the mechanisms that have allowed for the evolution and maintenance of such high plant diversity. With the works we present in this chapter we aim to broaden the knowledge about the beta-diversity and the evolutionary history of the tree species of rainforests in three important regions in Colombia as are the Amazon, Magdalena and Orinoco river basins. Both studies were done using information of the abundance of tree species with Diameter at Breast Height – DBH greater than 10cm from thirty-two, one-hectare, permanent vegetation plots that are located in terra firme and seasonally flooded forests. Beta-diversity in these rainforests seems to be determined mostly by environmental variables, this result gives suppor to the Niche Theory, however, a great proportion of the variance remains unexplained by the variables studied. Regarding phylobetadiversity, the seasonally flooded forests exhibit an overrepresentation of some plant lineages (eudicots), which shows that environmental filtering is strong, not only at the species level, but also at the lineage level.

13

ENVIRONMENTAL FILTERING OF EUDICOT LINEAGES UNDERLIES PHYLOGENETIC CLUSTERING IN TROPICAL SOUTH AMERICAN FLOODED FORESTS

Aldana, A. M., Carlucci, M. B., Fine, P. V. A. & Stevenson, P. R. 2016. Environmental filtering of eudicot lineages underlies phylogenetic clustering in tropical South American flooded forests. Oecologia. Published on-line on September 26th 2016. doi:10.1007/s00442-016-3734-y

14 Environmental filtering of eudicot lineages underlies phylogenetic clustering in tropical South American flooded forests

Ana M. Aldanaa1, Marcos B. Carluccib,c, Paul V.A. Fined, Pablo R. Stevensona

a Departamento de Ciencias Biológicas, Universidad de los Andes, carrera 1 No. 18A – 10 Bogotá D.C., Colombia. b Programa de Pós-Graduação em Ecologia e Evolução, Instituto de Ciências Biológicas, Universidade Federal de Goiás, Goiânia, GO 74690-900, Brazil. c CAPES Foundation, Ministry of Education of Brazil, Brasília, DF 70040-020, Brazil. d Department of Integrative Biology and University and Jepson Herbaria, University of California, Berkeley, CA, 94720, U.S.A.

1 Corresponding author. Departamento de Ciencias Biológicas, Universidad de los Andes, carrera 1 No. 18A – 10 Bogotá, Colombia. Telephone number: 57-1-3394949 ext. 3770, e-mail address: [email protected]

1 Highlighted Student Research

1 Evaluation of the phylogenetic diversity of forest communities in the Neotropics is needed to understand the evolutionary process that have led to high species diversity in these forests. We explored the phylogenetic patterns of flooded and terra firme forests in Northern South America and found that flooded forests are low in species diversity as well as in phylogenetic diversity, due to the fact that only a few lineages of the eudicot clade are able to succeed in these extreme conditions. As such, this research helps elucidate a fragment of the evolutionary history of Neotropical forests, and should stimulate tests of new hypotheses.

15 Abstract The phylogenetic community assembly approach has been used to elucidate the role of ecological and historical processes in shaping tropical tree communities. Recent studies have shown that stressful environments, such as seasonally-dry, white-sand and flooded forests tend to be phylogenetically clustered, arguing for niche conservatism as the main driver for this pattern. Very few studies have attempted to identify the lineages that contribute to such assembly patterns. We aimed to improve our understanding of the assembly of flooded forest tree communities in Northern South America by asking the following questions: are seasonally-flooded forests phylogenetically clustered? If so, which angiosperm lineages are over-represented in seasonally-flooded forests? To assess our hypotheses, we investigated seasonally-flooded and terra firme forests from the Magdalena, Orinoco and Amazon Basins, in Colombia. Our results show that, regardless of the river basin in which they are located, seasonally-flooded forests of Northern South America tend to be phylogenetically clustered, which means that the more abundant taxa in these forests are more closely related to each other than expected by chance. Based on our alpha and beta phylodiversity analyses we interpret that eudicots are more likely to adapt to extreme environments such as seasonally- flooded forests, which indicates the importance of environmental filtering in the assembly of the Neotropical flora.

Keywords: Floodplains, várzea, igapó, phylobetadiversity, phylogenetic structure.

Declaration of authorship: PRS and AMA formulated the idea. AMA, MBC and PVAF developed the methodology.

AMA and MBC wrote the manuscript; other authors provided advice and edited the manuscript.

16 Introduction In tropical forests, the phylogenetic community assembly approach has been used to explain the role of ecological and historical processes in shaping these highly diverse tree communities (Eiserhardt et al. 2013; Gerhold et al. 2015; Carlucci et al. 2016). Different historical processes such as time-integrated species-area effect (Fine and Ree 2006; Fine 2015) and extinctions due to major climatic changes affect diversification rates within regional pools and leave imprints in current ecological communities (Parmentier et al. 2007; Kissling et al. 2012). It is possible to hypothesize which historical processes have influenced current community assembly through their effect on the formation of regional pools of species by evaluating whether community phylogenetic structure deviates from random patterns (Webb et al. 2002; Emerson and Gillespie 2008; Cavender-Bares et al. 2009). For instance, a community may be phylogenetically clustered when it displays lower phylogenetic diversity than expected by chance from the species pool, or over-dispersed when it presents higher phylogenetic diversity than expected by chance from the species pool. If adaptations to severe environments involve complex traits, closely-related species would be likely to be more similar than expected under neutral evolution, leading to a phylogenetic signal in functional traits linked to such adaptations (Crisp and Cook 2012). In contrast, if adaptation to severe environments involves simple changes that are possible in many unrelated plant lineages, this convergent evolution of traits would result in phylogenetically over-dispersed communities (Emerson and Gillespie 2008).

A number of recent studies have argued that stressful environments, such as seasonally-dry, white-sand and seasonally-flooded forests tend to be phylogenetically clustered (low phylogenetic diversity) with respect to terra firme forests (Pennington 2009; Gonzalez-Caro et al. 2014; Honorio Coronado et al. 2015; Guevara et al. 2016; but see Fine and Baraloto 2016). Dissimilarities in species composition and richness among South American terra firme and seasonally-flooded forests have been attributed to the differences in environmental conditions, mainly soil texture, nutrient content and hydrological stress (Haugaasen and Peres 2006). Recent studies have argued that, in Amazonian rainforests, some closely related tree species have been able to adapt to the relatively stable conditions of floodplains that can be traced as far back to as the

17 Paleocene, and propose phylogenetic niche conservatism (PNC) as a possible explanation for the distribution patterns of the most important species in Amazonian várzea forests (Wittmann et al. 2011; Wittmann et al. 2013). Responses of tree species to flooding in the Amazon basin include a range of anatomical, morphological and physiological strategies, which can involve xeromorphic leaves, reduction of CO2 uptake, adventitious roots and development of aerenchyma, among other adaptations (Parolin 2008; Parolin and Wittmann 2010).

In the Neotropics, there have been very few studies that have investigated the phylogenetic patterns of seasonally-flooded forests. Umaña et al. (2012) reported phylogenetic clustering in Colombian seasonally-flooded forests and argued that the differences in species composition and community phylogenetic patterns between terra firme and igapó forests in the Colombian Amazon region are due to habitat specialization to edaphic conditions. Nevertheless, the spatial limitation of the study to a single locality did not allow the authors to make further conclusions about the possible biogeographic causes of the patterns they described. Previous studies have investigated the phylogenetic community structure of Neotropical forests, but only very few have attempted to identify the lineages that are involved in particular assembly patterns (Fine and Kembel 2011; Carlucci et al. 2016). The method used here, principal coordinates of phylogenetic structure (PCPS; Duarte 2011; Duarte et al. 2012) is a phylobetadiversity approach based on ordination that aids in the identification of the major lineages that are better represented in different communities. While PCPS evaluates shifts in phylogenetic composition across communities, the net relatedness index (NRI; Webb et al. 2002), measures local phylogenetic structure of communities (clustering or overdispersion) relative to species pools using null models. Using such an approach, it is possible to relate PCPS scores with NRI values, thereby identifying lineages that are phylogenetically clustered or over-dispersed in different communities.

We aimed to improve knowledge on the assembly of flooded forest communities in Northern South America, by asking the following questions: Are seasonally-flooded forests phylogenetically clustered relative to a species pool encompassing both flooded and terra firme forests? If so, which angiosperm lineages are better represented in

18 seasonally-flooded forests? To answer these questions, we proposed three hypotheses. First, we tested the null hypothesis of a sampling artifact; seasonally-flooded forests may simply have fewer individuals per sampled area, and consequently contain fewer evolutionary lineages (Fine and Kembel 2011). Second, due to environmental filtering, particular lineages may have a higher potential to evolve the traits necessary to survive and reproduce in seasonally-flooded forests; thus these forests would have lower phylogenetic diversity than terra firme forests and a clustering of closely related lineages, thereby resembling what has been observed for white-sand and dry forests (Pennington 2009; Fine and Kembel 2011; Guevara et al. 2016). The third hypothesis posits that flood tolerance has independently evolved in many lineages, consistent with convergent evolution, so that seasonally-flooded forests have similar phylogenetic structure to terra firme forests and plant lineages in both environments show overdispersion, similar to what was observed when comparing phylogenetic community structure and habitat heterogeneity in the Korup forest in Cameroon (Baldeck et al. 2016).

To test these hypotheses, we studied seasonally-flooded (várzea and igapó sensu Prance 1989) and terra firme forests from the Magdalena, Orinoco and Amazon Basins, in Colombia. In this research, we present the most comprehensive attempt to study the phylogenetic community structure of seasonally-flooded forests in Northern South America, investigating the potential general importance of environmental filtering as a major factor shaping the evolution of Neotropical floras.

Materials and Methods Study system We used information on the abundance of woody species (DBH>10cm) recorded in 32 1-ha vegetation plots (Stevenson et al. 2004; Aldana et al. 2008; Stevenson and Aldana 2008; Cano and Stevenson 2008; Correa-Gómez and Stevenson 2010; Umaña et al. 2012) from seven lowland forest sites in the Amazon, Magdalena and Orinoco Basins (Fig.1). These three river basins have undergone radically different geologic processes (Antonelli et al. 2009; Hoorn et al. 2010; Montes et al. 2015). In each site, the plots were classified as terra firme (22), or seasonally-flooded (10) if these were submerged

19 for at least one month during the rainy season (ESM1). Vegetation plots in lowland forests sometimes include tree ferns and gymnosperms; in our sites these species represent only an extremely low percentage of the total stems (0.06%). Taking into account the great effect these basal clades can have in the calculations of phylogenetic metrics, we excluded them from the analyses (Honorio-Coronado et al. 2015).

Fig. 1 Approximate location of the 1-ha vegetation plots in various localities of Northern South America used in this study. All sites are located in Colombia. Gray lines represent the boundaries of the three major river basins in the region. Squares represent seasonally-flooded forests (2 plots per square) and triangles represent terra firme forests (3-5 plots per triangle). Detailed plot information can be found in Electronic Supplementary Material Table S1. Color version is available online

Phylogenetic tree In order to build a phylogenetic tree containing the 1,432 species recorded in the study sites, we used the megatree R20120829 (available at https://github.com/camwebb/tree-

20 of-trees/blob/master/megatrees/R20120829.new), which is based on the phylogenetic backbone proposed by APG III (APG 2009) and on relationships among families according to Stevens (2013). The APG III is a consensus tree built from phylogenetic relationships based on several molecular studies that accumulated since the late 1980s, mostly based on analysis of sequences of chloroplast markers and is well resolved with respect to deep phylogenetic relationships (Stevens 2013). The branch lengths of the tree were adjusted through the BLADJ algorithm in Phylocom 4.2 software (Webb et al. 2008) following clade age estimates by (Bell et al. 2010). We used the module Phylomatic in the software Phylocom 4.2 (Webb et al. 2008) to prune the megatree for the species present in our total species pool. The resulting phylogenetic tree (ESM2) was well resolved for deep nodes and mostly contained polytomies linking species within genera and genera within families. Since we were interested in evaluating major differences in phylogenetic composition related to deep relationships in the phylogeny, the resulting phylogenetic tree was appropriate to test our hypotheses.

Phylogenetic Community Structure and Composition To test our first hypothesis, a possible sampling artifact due to a lower number of individuals in seasonally-flooded forests, we first evaluated the difference in the number of stems between the two categories with a t-test. We then applied a rarefaction method (Gotelli and Colwell 2001) to all plots considering 344 individuals, the lowest number of stems registered in a plot, using the function rrarify in the package vegan (Oksanen et al. 2015) for R (R Core Team, 2015). Finally, we calculated different phylogenetic metrics for the rarefied communities to compare them to the values of the original plots.

To assess the phylogenetic structure of communities in terms of phylogenetic clustering or overdispersion, we used NRI (Webb et al. 2002). NRI is the standardized effect size of mean phylogenetic distances (MPD) between pairs of co-occurring species (multiplied by -1 to be interpreted in terms of degrees of phylogenetic relatedness). NRI was computed using species abundances and a regional species pool defined as the total species list of all the 32 plots, thereby including both terra firme and seasonally- flooded forests. The null model used was phylogeny.pool, which draws species at

21 random with equal probability from the species pool while maintaining plot species richness (Kembel 2009; Kembel et al. 2010). NRI for each community (plot) was also calculated using the basin level species pools, to evaluate the effect of these regional pools in the resulting NRI values.

To assess the phylogenetic composition of lineages across plots, we computed principal coordinates of phylogenetic structure (PCPS, Duarte 2011), which are phylogenetic eigenvectors of an ordination (PCoA) applied to a phylogenetic fuzzy-weighted matrix of community composition (Pillar and Duarte 2010). PCPS vectors describe changes in the phylogenetic composition across plots, and have been used to interpret the dominance of different lineages in different communities (Duarte et al. 2014). NRI and PCPS can be used in tandem to interpret which lineages are related to different degrees of phylogenetic clustering or overdispersion (Pérez-Valera et al. 2015; Carlucci et al. 2016). Studies have shown that NRI and the main vectors of PCPS (first and second) are not strongly affected by phylogenies with low terminal resolution, especially those with a great number of species (Swenson 2009; Maestri et al. 2016), such as ours.

Data analyses In order to assess the differences in NRI values between forest types and between river basins we used Analysis of Variance (ANOVA). We also assessed the difference in the values of the first PCPS axes between forest types by using ANOVA. We used a correlation to evaluate the relationship between NRI values of rarefied and sampled communities, and to evaluate the relationship between NRI values and PCPS axis I and II. All analysis were performed in the R environment (R Core Team, 2015).

Results Seasonally-flooded forests tend to be less diverse than terra firme forests (Table 1). As expected, seasonally-flooded forests in the Amazon basin are richer in species than seasonally-flooded forests in the other basins studied; although, comparatively they are less diverse than terra firme forests of the same basin. Even though floristic composition

22 of the várzea and igapó communities included in this study are very different at the family, genus and species levels (Appendix 1), the Index of Importance shows that the most important species of seasonally-flooded forests mostly belong to the eudicot clade. The most important species of the terra firme forests belong to the eudicot, as well as the monocot and magnoliid clades (ESM3).

Table 1. Number of plots per forest type and river basin for the 32 1-ha vegetation plots included in this study. Number of species and number of individuals are shown as the range among the plots. Fisher´s Alpha is given as the average and standard deviation is shown in parenthesis

Forest type River basin No. of plots No. of spp. No. of ind. Fisher´s Alpha

Flooded Amazon 2 111 - 135 501 - 548 50.70 (9.24)

Magdalena 2 59 - 63 350 - 484 19.82 (0.70)

Orinoco 6 26 - 77 399 - 805 11.19 (7.83)

Terra firme Amazon 4 171 - 255 599 - 629 119.27 (32.57)

Magdalena 7 78 - 187 425 - 606 59.10 (27.11)

Orinoco 11 60 - 165 423 - 710 40.41 (19.31)

In order test our first hypothesis, the sampling artifact, under which we expected seasonally-flooded forests to be phylogenetically clustered due to a lower number of individuals; we compared the number of stems between forests. We found no differences (t = -0.91, P= 0.38) in the number of trees surveyed in flooded and terra firme 1-ha vegetation plots (fl=504.1, tf=545.1;Table 1). Additionally, NRI values of rarified communities and NRI values calculated for the original communities were highly correlated (Spearman r=0.986, P<0.001), indicating that phylogenetic metrics of seasonally-flooded forests are not affected by sample size in this study system.

The calculated NRI values for the original communities of seasonally-flooded forests using the total species are significantly higher than terra firme communities and positive,

23 which means that seasonally-flooded forests are more phylogenetically clustered and less diverse than terra firme forests in Northern South America (Fig.2, ESM4). The majority of terra firme forests had negative NRI values, which means that these communities are more phylogenetically diverse and species tend to be over-dispersed in the phylogeny. When comparing NRI values among river basins we found no significant differences between them (ANOVA, d.f=2, F=0.03, p=0.971).

Fig. 2 Distribution of Net Relatedness Index (NRI) values for flooded and terra firme forests in Northern South America. NRI values of seasonally-flooded forests are significantly higher (ANOVA, d.f=1, F= 33.95, p<0.001) We also examined the effect of the differences in species pools among basins and found that the calculated NRI values for each community are very similar when they are calculated using a phylogenetic tree that only contains the regional species pool from each basin (ESM1). There was a very strong correlation between NRI calculated with a regional species pool (all 32 plots) and the NRI calculated with basin species pools (0.99, Pearson t= 56.3, P<0.001).

24 The PCPS analysis showed that seasonally-flooded forests are mostly represented by eudicots, whereas terra firme forests are well represented by magnoliids, monocots and eudicots (Fig.3, ESM3). We evaluated the relationship between the first and second axes of the PCPS and the NRI values of the plots and we found a high correlation between PCPS I and NRI (Spearman r=0.765, p<0.001) and a low correlation between PCPS II and NRI (Spearman r=-0.321, p>0.05).

Fig. 3 Scatter plot showing the variation in phylogenetic composition across flooded (squares) and terra firme (triangles) rainforest tree communities in Northern South America. Net Relatedness Index (NRI) values representing clustering or overdispersion are shown in different colors: communities in reddish colors tend to be phylogenetically clustered, while communities in blueish colors tend to be phylogenetically overdispersed. Phylogenetic composition was assessed using principal coordinates of phylogenetic structure, while phylogenetic structure values (clustering and overdispersion) were computed using the net relatedness index (Webb et al. 2002). Species correlations with PCoA axes located species along PCPS I and II. The spider-like diagrams show the position of species relative to the centroid of their clades in the multivariate space. Note that a PCPS axis shows where clades have more relative importance across the phylogenetic gradient in terms of their relative contribution in abundance and pairwise phylogenetic distances between co-occurring species to communities. Color version is available online

25 Integrating the PCPS and NRI results, we observed that the tendency toward phylogenetic clustering (higher NRI values) in lowland seasonally-flooded forests is explained by the high abundance of the eudicots clade, while the tendency towards phylogenetic overdispersion (lower NRI values) in terra firme forests was associated with a higher relative abundance of magnoliids. This result supports our second hypothesis under which we expected more closely related species to be over- represented due to the stressful environment of floodplains. When comparing PCPS I scores among forest types we found that the phylogenetic composition between flooded and terra firme forests was significantly different (Fig.4).

Fig. 4 Distribution of scores of PCPS I for flooded and terra firme forests in Northern South America. The differences between PCPS scores can be interpreted in terms of phylobetadiversity. Values of seasonally-flooded forests are significantly higher (ANOVA, d.f=1, F= 23.98, p<0.001)

26 Discussion Our results showed widespread phylogenetic clustering in seasonally-flooded forests in Northern South America in relation to a species pool that also contained species from terra firme forests of the region. This pattern was consistent when phylogenetic metrics are calculated for each of the three basins separately. Seasonally-flooded forests were phylogenetically clustered regardless of the type of river flooding them (várzea or igapó) or the river basin in which are located (Amazon, Magdalena or Orinoco). Therefore, the most abundant taxa in these seasonally-flooded forests are more closely related than expected by chance. Moreover, phylobetadiversity analyses revealed that the dominant taxa in seasonally-flooded forests belong to the eudicot clade, mainly from families within the Malpighiales and Fabales orders (ESM3).

Species diversity in local seasonally-flooded communities is low in comparison to terra firme forests, with a mean of 60.5 species per plot (Table 1), which is expected to yield high NRI values because it has been previously reported that low species diversity correlates with low phylogenetic diversity (ESM4) (Honorio-Coronado et al. 2015). Additionally, total species diversity of all the seasonally-flooded forests plots (10) is 443, which is significantly lower than the 1,212 total species for all terra firme plots (22).

Our results indicate a role for habitat filtering (Wiens and Graham 2005; Losos 2008; Crisp and Cook 2012) in flooded forest community assembly. It seems that despite the great diversity of lineages that can be found in the study forests, few magnoliids have evolved traits necessary to survive the environmental pressure of flooding. Wittmann et al. (2013) found similar floristic composition within seasonally-flooded forests from the Amazon and Orinoco regions, and argued that the prolonged stable conditions of flooded environments since the Early Miocene (Hoorn et al. 2010) in Northern South America has allowed for taxa to specialize and adapt to these habitats (Wittmann et al. 2013). We agree with this interpretation and suggest that specialization to different types of substrates and nutrient contents has occurred since the uplift of the Andes, mainly during the late Miocene when the drainage systems in the region radically changed (Mora et al. 2011). As expected, species diversity of seasonally-flooded forests in the Amazon basin is higher than in the Orinoco and Magdalena basins. This pattern

27 could be explained by the time-integrated species-area effect (Fine and Ree 2006) where taxa have had longer to diversify in prolonged flooding conditions in the Amazon basin (Hoorn et al. 2010) and less time to diversify in the river systems that resulted from the Andean orogeny (Antonelli et al. 2009).

We assessed alpha and phylobetadiversity in tropical forest communities using an integrated framework that permitted us to evaluate which angiosperm lineages contributed the most to local phylogenetic clustering or overdispersion patterns. Examining the relationship between NRI values and PCPS scores is a valuable way to integrate alpha phylogenetic structure and phylobetadiversity (Carlucci et al. 2016). Previous studies have used the tool nodesig (Webb et al. 2008) to identify lineages overrepresented in sets of communities, e.g. delimited by a habitat type (Fine and Kembel 2011), an objective similar to ours in the present study. However, in nodesig information on non-random patterns of taxa abundance in communities is interpreted with respect to the phylogeny; whereas, in PCPS phylogenetic information is interpreted at the community scale because an ordination score of phylogenetic weighted community composition is attributed to each community. Consequently, although PCPS and nodesig may be used to answer similar questions, PCPS enables identifying lineages that are more common in sets of communities while assessing phylobetadiversity across communities. Moreover, PCPS enables a continuous evaluation across communities, with no need to categorize communities into different habitat types.

While investigating the traits necessary for trees to adapt to seasonal flooding in the Amazon, Parolin (2008) found that the main strategy used by many species was that of general flood tolerance, as opposed to survival or escape, which is accomplished by a great diversity of different metabolic and morphological adaptations. Similar to our results, Pizano and Garcia (2014) found the composition of Neotropical dry forests included a high dominance of eudicots (Pizano and García 2014). We thus postulate that adaptations to extreme environments could be important in promoting eudicot dominance in dry and seasonally-flooded forests. Field and experimental studies

28 assessing metabolic and morphological adaptations of plant species to life in flooded and dry forests can be used in the future to test this idea.

We conclude that seasonally-flooded forests of Northern South America have low taxonomic and phylogenetic diversity compared to surrounding terra firme forests due to the effect of historical environmental filtering leading to adaptive radiations within lineages of eudicots. Our results suggest that eudicots possess some advantage in relation to magnoliids that has allowed them to evolve the necessary traits to adapt to extreme environments such as seasonal floodplains. We propose that further research should focus on investigating the traits that confer flood tolerance to plants of this important clade of angiosperms.

Acknowledgements We thank Ángela Cano, Sasha Cárdenas, Luisa Fernanda Casas, Diego Felipe Correa, Mabel Suescún, María Natalia Umaña, Boris Villanueva for the information provided from the plots they established. AMA would like to thank James Richardson and Toby Pennington for helpful discussions on the results reported here and the Universidad de Los Andes for providing funding to visit the University of California, Berkeley where this research was envisioned. MBC received fellowships from CAPES-Brazil (grants BEX7913/13-3 and PNPD #1454013). We are thankful to Dr. Ethan Householder and two anonymous reviewers who helped improve an earlier version of this manuscript.

29 References Aldana AM, Beltrán M, Torres-Neira J, Stevenson PR (2008) Habitat Characterization and Population Density of Brown Spider Monkeys (Ateles hybridus) in Magdalena Valley, Colombia. Neotrop Primates 15:46–50. doi: 10.1896/044.015.0203

Antonelli A, Nylander J a a, Persson C, Sanmartín I (2009) Tracing the impact of the Andean uplift on Neotropical plant evolution. Proc … 106:9749–9754

APG (2009) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG III. Bot J Linn Soc 161:105–121

Baldeck CA, Kembel SW, Harms KE, et al (2016) Phylogenetic turnover along local environmental gradients in tropical forest communities. Oecologia. doi: 10.1007/s00442-016-3686-2

Bell CD, Soltis DE, Soltis PS (2010) The age and diversification of the angiosperms re-revisited. Am J Bot 97:1296–1303. doi: 10.3732/ajb.0900346

Cano A, Stevenson PR (2008) Diversidad y composición florística de tres tipos de bosque en la Estación Biológica Caparú, Vaupés. Colomb For 12:63. doi: 10.14483/udistrital.jour.colomb.for.2009.1.a06

Carlucci MB, Seger GDS, Sheil D, et al (2016) Phylogenetic composition and structure of tree communities shed light on historical processes influencing tropical rainforest diversity. Ecography (Cop) n/a–n/a. doi: 10.1111/ecog.02104

Cavender-Bares J, Kozak KH, Fine PVA, Kembel SW (2009) The merging of community ecology and phylogenetic biology. Ecol Lett 12:693–715. doi: 10.1111/j.1461-0248.2009.01314.x

Correa-Gómez DF, Stevenson PR (2010) Estructura y diversidad de bosques de los llanos orientales colombianos (Reserva Tomo Grande, Vichada). Revista Orinoquia 14:31–48

Crisp MD, Cook LG (2012) Phylogenetic niche conservatism: what are the underlying evolutionary and ecological causes? New Phytol 196:681–694

Duarte LDS (2011) Phylogenetic habitat filtering influences forest nucleation in grasslands. Oikos 120:208–215. doi: 10.1111/j.1600-0706.2010.18898.x

Duarte LDS, Both C, Debastiani VJ, et al (2014) Climate effects on amphibian distributions depend on phylogenetic resolution and the biogeographical history of taxa. Glob Ecol Biogeogr 23:213–222. doi: 10.1111/geb.12089

Duarte LDS, Prieto P V., Pillar VD (2012) Assessing spatial and environmental drivers of phylogenetic structure in Brazilian Araucaria forests. Ecography (Cop) 35:952–960. doi: 10.1111/j.1600- 0587.2011.07193.x

Eiserhardt WL, Svenning J-C, Baker WJ, et al (2013) Dispersal and niche evolution jointly shape the geographic turnover of phylogenetic clades across continents. Sci Rep 3:1164. doi: 10.1038/srep01164

Emerson BC, Gillespie RG (2008) Phylogenetic analysis of community assembly and structure over space and time. Trends Ecol Evol 23:619–630. doi: 10.1016/j.tree.2008.07.005

Fine P (2015) Ecological and Evolutionary Drivers of Geographic Variation in Species Diversity. Annu Rev Ecol Evol Syst 46:369–392. doi: 10.1146/annurev-ecolsys-112414-054102

Fine PVA, Kembel SW (2011) Phylogenetic community structure and phylogenetic turnover across space and edaphic gradients in western Amazonian tree communities. Ecography (Cop) 34:552–565. doi: 10.1111/j.1600-0587.2010.06548.x

30 Fine PVA, Baraloto C (2016) Habitat Endemism in White-sand Forests : Insights into the Mechanisms of Lineage Diversification and Community Assembly of the Neotropical Flora. Biotropica 48:24–33. doi: 10.1111/btp.12301

Fine PVA, Ree RH (2006) Evidence for a time-integrated species-area effect on the latitudinal gradient in tree diversity. Am Nat 168:796–804. doi: 10.1086/508635

Gerhold P, Cahill JF, Winter M, et al (2015) Phylogenetic patterns are not proxies of community assembly mechanisms (they are far better). Functional Ecology 29:600–614. doi: 10.1111/1365-2435.12425

Gonzalez-Caro S, Umana MN, Alvarez E, et al (2014) Phylogenetic alpha and beta diversity in tropical tree assemblages along regional-scale environmental gradients in northwest South America. J Plant Ecol 7:145–153. doi: 10.1093/jpe/rtt076

Gotelli NJ, Colwell RK (2001) Quantifying biodiversity: Procedures and pitfalls in the measurement and comparison of species richness. Ecol. Lett. 4:379–391

Guevara JE, Damasco G, Baraloto C, et al (2016) Low Phylogenetic Beta Diversity and Geographic Neo- endemism in Amazonian White-sand Forests. Biotropica 48:34–46. doi: 10.1111/btp.12298

Haugaasen T, Peres CA (2006) Floristic, edaphic and structural characteristics of flooded and unflooded forests in the lower Rio Purús region of central Amazonia, Brazil. Acta Amaz 36:25–36.

Honorio-Coronado EN, Dexter KG, Pennington RT, et al (2015) Phylogenetic diversity of Amazonian tree communities. Divers Distrib 21:1295–1307. doi: 10.1111/ddi.12357

Hoorn C, Roddaz M, Dino R, et al (2010) The Amazonian Craton and its Influence on Past Fluvial Systems (Mesozoic-Cenozoic, Amazonia). In: Hoorn C, Wesselingh FP (eds) Amazonia: Landscape and Species Evolution. Wiley-Blackwell Publishing Ltd., Oxford, UK, pp 101–122

Kembel SW (2009) Disentangling niche and neutral influences on community assembly: assessing the performance of community phylogenetic structure tests. Ecol Lett 12:949–60. doi: 10.1111/j.1461- 0248.2009.01354.x

Kembel SW, Cowan PD, Helmus MR, et al (2010) Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26:1463–1464. doi: 10.1093/bioinformatics/btq166

Kissling WD, Eiserhardt WL, Baker WJ, et al (2012) Cenozoic imprints on the phylogenetic structure of palm species assemblages worldwide. Proc Natl Acad Sci USA 109:7379–7384. doi: 10.1073/pnas.1120467109

Losos J (2008) Phylogenetic niche conservatism, phylogenetic signal and the relationship between phylogenetic relatedness and ecological similarity among species. Ecol Lett 995–1003. doi: 10.1111/j.1461-0248.2008.01229.x

Maestri R, Luza AL, Barros LD de, et al (2016) Geographical variation of body size in sigmodontine rodents depends on both environment and phylogenetic composition of communities.

Montes C, Cardona A, Jaramillo C, et al (2015) Middle Miocene closure of the Central American Seaway. Science (80- ) 348:226–229. doi: 10.1126/science.aaa2815

Mora A, Baby P, Roddaz M, et al (2011) Tectonic History of the Andes and Sub-Andean Zones: Implications for the Development of the Amazon Drainage Basin. In: Hoorn C, Wesselingh FP (eds) Amazonia: Landscape and Species Evolution. Wiley-Blackwell Publishing Ltd., Oxford, UK, pp 38–60

Oksanen J, Guillaume Blanchet F, Kindt R, et al (2015) vegan: Community Ecology Package. R package version 2.3-0

31 Parmentier I, Malhi Y, Senterre B, et al (2007) The odd man out? Might climate explain the lower tree α- diversity of African rain forests relative to Amazonian rain forests? J Ecol 95:1058–1071. doi: 10.1111/j.1365-2745.2007.01273.x

Parolin P (2008) Submerged in darkness: adaptations to prolonged submergence by woody species of the Amazonian floodplains. Ann Bot 103:359–376. doi: 10.1093/aob/mcn216

Parolin P, Wittmann F (2010) Struggle in the flood: tree responses to flooding stress in four tropical floodplain systems. AoB Plants 2010:plq003–plq003. doi: 10.1093/aobpla/plq003

Pennington R (2009) Woody plant diversity, evolution, and ecology in the tropics: perspectives from seasonally dry tropical forests. Annu Rev …. doi: 10.1146/annurev.ecolsys.110308.120327

Pérez-Valera E, Goberna M, Verdú M (2015) Phylogenetic structure of soil bacterial communities predicts ecosystem functioning. FEMS Microbiol Ecol 91:fiv031. doi: 10.1093/femsec/fiv031

Pillar VD, Duarte LDS (2010) A framework for metacommunity analysis of phylogenetic structure. Ecol Lett 13:587–596. doi: 10.1111/j.1461-0248.2010.01456.x

Pizano C, García H (eds) (2014) El Bosque Seco Tropical en Colombia. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt, Bogotá, D. C.

Prance G (1989) American tropical forests. Trop rain For Ecosyst Ecol Stud 99–132. doi: 10.1016/B978- 0-444-42755-7.50012-2

R Core Team (2015) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria

Stevens PF (2013) Angiosperm Phylogeny Website, version 13. Accessed on August 2nd, 2016. URL: http://www.mobot.org/MOBOT/research/APweb/

Stevenson PR, Aldana AM (2008) Potential Effects of Ateline Extinction and Forest Fragmentation on Plant Diversity and Composition in the Western Orinoco Basin, Colombia. Int J Primatol 29:365–377. doi: 10.1007/s10764-007-9177-x

Stevenson PR, Suescún M, Quiñones M (2004) Characterization of forest types at the CIEM, Tinigua Park, Colombia. F Stud Fauna Flora La Macarena Colomb 14:1–20

Swenson NG (2009) Phylogenetic resolution and quantifying the phylogenetic diversity and dispersion of communities. PLoS One 4:e4390. doi: 10.1371/journal.pone.0004390

Umaña MN, Norden N, Cano A, Stevenson PR (2012) Determinants of plant community assembly in a mosaic of landscape units in central Amazonia: ecological and phylogenetic perspectives. PLoS One 7:e45199. doi: 10.1371/journal.pone.0045199

Webb CO, Ackerly DD, Kembel SW (2008) Phylocom: software for the analysis of phylogenetic community structure and character evolution. Bioinformatics 24:2098–2100.

Webb CO, Ackerly DD, McPeek MA, Donoghue MJ (2002) Phylogenies and community ecology. Annu Rev Ecol Syst 33:475–505. doi: 10.1146/annurev.ecolsys.33.010802.150448

Wiens JJ, Graham CH (2005) Niche conservatism: integrating evolution, ecology, and conservation biology. Annu Rev Ecol Evol Syst 36:519–539. doi: 10.1146/annurev.ecolsys.36.102803.095431

Wittmann F, Householder E, Piedade MTF, et al (2013) Habitat specifity, endemism and the neotropical distribution of Amazonian white-water floodplain trees. Ecography (Cop) 36:690–707. doi: 10.1111/j.1600- 0587.2012.07723.x

32 Wittmann F, Schöngart J, Junk WJ (2011) Amazonian Floodplain Forests. Springer Netherlands, Dordrecht

33 Electronic supplementary material

ESM1: Detailed information for 32 1-ha vegetation plots used in this study. NRI values were calculated using a regional species pool with all the 32 plots (NRI Regional Pool) and with basin species pool with all plots belonging to each of the three basins (NRI Basin Pool).

NRI NRI Forest Number of Number of Plot Plot Code Locality Basin Longitude Latitude Basin Regional Type Trees Species Pool Pool 1 CAPA_PI_1 Caparu Flooded Amazon -69.51806 -1.08139 498 110 0.821 1.007 2 CAPA_PI_2 Caparu Flooded Amazon -69.51809 -1.0887 548 135 0.432 0.591 3 CAPA_TFC_1 Caparu Terra firme Amazon -69.51364 -1.07183 629 255 -0.517 -0.256 4 CAPA_TFC_2 Caparu Terra firme Amazon -69.51582 -1.07001 605 218 -1.907 -1.707 5 CAPA_TFT_1 Caparu Terra firme Amazon -69.51503 -1.07669 617 171 -2.053 -1.929 6 CAPA_TFT_2 Caparu Terra firme Amazon -69.50882 -1.07763 599 218 -0.211 -0.038 7 CASA_PII_2 Casanare Flooded Orinoco -70.1491 5.66733 453 31 0.888 0.894 8 CASA_PIV_1 Casanare Flooded Orinoco -70.11015 5.67092 433 26 0.344 0.366 9 QUIN_TF_1 Quinchas Terra firme Magdalena -74.26828 6.04681 568 171 -1.594 -1.733 10 QUIN_TF_3 Quinchas Terra firme Magdalena -74.1985 6.02275 606 187 -1.900 -2.167 11 QUIN_TF_4 Quinchas Terra firme Magdalena -74.20644 6.01911 446 159 0.063 -0.004 12 QUIN_TF_5 Quinchas Terra firme Magdalena -74.266 6.04717 545 144 -1.933 -2.034 13 SAJU_PI_1 SanJuan Flooded Magdalena -74.15702 6.68268 484 63 0.226 0.184 14 SAJU_PI_2 SanJuan Flooded Magdalena -74.13933 6.6858 344 58 0.625 0.621 15 SAJU_TF_3 SanJuan Terra firme Magdalena -74.11652 6.70198 474 96 0.250 0.245 16 SAJU_TF_4 SanJuan Terra firme Magdalena -74.11572 6.69518 425 78 -0.201 -0.161 17 SAJU_TF_5 SanJuan Terra firme Magdalena -74.12025 6.69678 525 90 -1.803 -1.916 18 SAMA_TF_1 SanMartin Terra firme Orinoco -73.63909 3.60844 524 60 -0.767 -0.840 19 SAMA_TF_2 SanMartin Terra firme Orinoco -73.63885 3.62834 566 60 -1.437 -1.402 20 SAMA_TF_3 SanMartin Terra firme Orinoco -73.454 3.581 466 98 -2.381 -2.330 21 TINI_PI_2 Tinigua Flooded Orinoco -74.03636 2.62456 399 31 0.503 0.537 22 TINI_PI_5 Tinigua Flooded Orinoco -74.03361 2.62456 449 77 0.313 0.314 23 TINI_TF_1 Tinigua Terra firme Orinoco -74.03859 2.62347 592 131 -0.320 -0.420 24 TINI_TF_3 Tinigua Terra firme Orinoco -74.04144 2.62092 683 165 -1.485 -1.616 25 TINI_TF_4 Tinigua Terra firme Orinoco -74.03785 2.61846 578 136 0.454 0.377 26 TINI_TF_6 Tinigua Terra firme Orinoco -74.04039 2.61601 529 120 0.117 0.113 27 TINI_TF_7 Tinigua Terra firme Orinoco -74.04308 2.61855 709 163 -1.355 -1.434 28 TOMO_PI_4 Tomo Flooded Orinoco -70.26383 4.82314 628 46 0.726 0.733 29 TOMO_PI_5 Tomo Flooded Orinoco -70.25089 4.83064 805 34 1.513 1.368 30 TOMO_TF_1 Tomo Terra firme Orinoco -70.27275 4.83925 433 80 -2.076 -2.190 31 TOMO_TF_2 Tomo Terra firme Orinoco -70.2757 4.83582 423 80 -2.328 -2.337 32 TOMO_TF_3 Tomo Terra firme Orinoco -70.25675 4.8418 451 63 -2.274 -2.206

34

ESM2: Phylogenetic tree containing 1,432 angiosperm tree species from Northern Southern

America. Green, eudicots; purple, other dicots; orange, monocots; blue, magnoliids.

35 ESM3: List of the 10 most important species in each of the 32 vegetation plots used in this study with information to the phylogenetic group that each family belongs to according to APG III.

Plot code & Frequency Density Dominance Importance Group Family Species Frequency Density Dominance Forest type % % % Value

Eudicots Calophyllaceae Caraipa densifolia 0.37 60 2470.16 8.08 11.98 8.87 28.93

Eudicots Combretaceae Buchenavia viridiflora 0.14 16 2038.65 3 3.19 7.32 13.52

Eudicots Fabaceae Dicorynia paraensis 0.15 17 1864.66 3.23 3.39 6.7 13.32

Eudicots Fabaceae Zygia cataractae 0.21 23 769.14 4.62 4.59 2.76 11.97

CAPA_PI_1 Magnoliids Myristicaceae Virola elongata 0.19 25 401.66 4.16 4.99 1.44 10.59 Seasonally- flooded Forest Eudicots Fabaceae Aldina heterophylla 0.16 16 1065.14 3.46 3.19 3.83 10.48 Eudicots Apocynaceae Aspidosperma excelsum 0.15 17 689.61 3.23 3.39 2.48 9.1

Eudicots Fabaceae Acosmium nitens 0.11 11 1268.16 2.31 2.2 4.55 9.06

Eudicots Lecythidaceae Eschweilera albiflora 0.08 8 1256.43 1.85 1.6 4.51 7.96

Eudicots Fabaceae Cynometra marginata 0.14 14 559.31 3 2.79 2.01 7.81

Eudicots Fabaceae Aldina heterophylla 0.38 45 1898.64 7.52 8.21 7.63 23.37

Eudicots Fabaceae Zygia cataractae 0.32 43 1217.19 6.34 7.85 4.89 19.07

Eudicots Lecythidaceae Eschweilera albiflora 0.2 22 2281.41 3.96 4.01 9.17 17.14

Eudicots Fabaceae Dicorynia paraensis 0.21 23 1806.01 4.16 4.2 7.26 15.61

CAPA_PI_2 Eudicots Phyllanthaceae Didymocistus chrysadenius 0.2 28 835.28 3.96 5.11 3.36 12.43 Seasonally- flooded Forest Eudicots Sapotaceae Manilkara bidentata 0.18 20 580.39 3.56 3.65 2.33 9.55

Eudicots Lecythidaceae Eschweilera integrifolia 0.2 20 464.07 3.96 3.65 1.86 9.47

Eudicots Chrysobalanaceae Licania egleri 0.18 21 373.84 3.56 3.83 1.5 8.9

Eudicots Goupiaceae Goupia glabra 0.05 5 1446.99 0.99 0.91 5.81 7.72

Magnoliids Myristicaceae Virola elongata 0.16 17 328.97 3.17 3.1 1.32 7.59

Eudicots Lecythidaceae Eschweilera coriacea 0.35 47 1873.85 5.72 7.47 6.6 19.79

CAPA_TFC_1 Eudicots Combretaceae Buchenavia congesta 0.01 1 1785.96 0.16 0.16 6.29 6.61 Terra firme Forest Eudicots Burseraceae Protium guianenese 0.16 16 378.93 2.61 2.54 1.34 6.49 Eudicots Sapotaceae Micropholis venulosa 0.1 10 517.03 1.63 1.59 1.82 5.05

36 Eudicots Fabaceae Clathrotropis macrocarpa 0.12 12 319.42 1.96 1.91 1.13 4.99

Eudicots Fabaceae Parkia oppositifolia 0.02 2 1207.34 0.33 0.32 4.25 4.9

Eudicots Fabaceae Andira macrothyrsa 0.05 5 919.95 0.82 0.79 3.24 4.85

Monocots Arecaceae Iriartea deltoidea 0.11 11 353.97 1.8 1.75 1.25 4.79

Eudicots Moraceae Brosimum lactescens 0.08 8 550.63 1.31 1.27 1.94 4.52

Eudicots Fabaceae Cynometra longicuspis 0.08 8 533.5 1.31 1.27 1.88 4.46

Eudicots Lecythidaceae Eschweilera coriacea 0.32 38 1448.63 5.55 6.28 3.69 15.52

Eudicots Sapotaceae Micropholis venulosa 0.02 2 5324.23 0.35 0.33 13.56 14.24

Eudicots Fabaceae Monopteryx uaucu 0.12 14 2420.01 2.08 2.31 6.16 10.56

Eudicots Fabaceae Cynometra longicuspis 0.11 11 1425.83 1.91 1.82 3.63 7.36

CAPA_TFC_2 Eudicots Fabaceae Enterolobium barnebianum 0.04 4 2312.98 0.69 0.66 5.89 7.25 Terra firme Forest Magnoliids Myristicaceae Iryanthera ulei 0.17 19 265.34 2.95 3.14 0.68 6.76 Magnoliids Myristicaceae Iryanthera laevis 0.13 13 848.78 2.25 2.15 2.16 6.56

Eudicots Violaceae Rinorea paniculata 0.16 19 235.55 2.77 3.14 0.6 6.51

Magnoliids Myristicaceae Virola carinata 0.14 14 561.89 2.43 2.31 1.43 6.17

Magnoliids Myristicaceae Osteophloeum platyspermum 0.07 8 1366.24 1.21 1.32 3.48 6.02

Monocots Arecaceae Oenocarpus bataua 0.67 114 2402.1 12.32 18.48 8.19 38.98

Eudicots Fabaceae Monopteryx uaucu 0.17 18 4378.6 3.13 2.92 14.93 20.97

Eudicots Euphorbiaceae Micrandra spruceana 0.27 32 3026.67 4.96 5.19 10.32 20.47

Monocots Arecaceae Euterpe precatoria 0.24 29 335.58 4.41 4.7 1.14 10.26

CAPA_TFT_1 Eudicots Moraceae Brosimum rubescens 0.15 15 1130.97 2.76 2.43 3.86 9.05 Terra firme Forest Eudicots Calophyllaceae Caraipa punctulata 0.15 16 630.21 2.76 2.59 2.15 7.5

Magnoliids Myristicaceae Virola pavonis 0.13 14 707.52 2.39 2.27 2.41 7.07

Eudicots Burseraceae Protium guianenese 0.14 16 501.53 2.57 2.59 1.71 6.88

Eudicots Moraceae Brosimum utile 0.11 11 856.55 2.02 1.78 2.92 6.73

Eudicots Euphorbiaceae Sandwithia heterocalyx 0.16 18 235.71 2.94 2.92 0.8 6.66

CAPA_TFT_2 Eudicots Euphorbiaceae Micrandra spruceana 0.5 64 5305.96 9.03 10.68 16.86 36.57 Terra firme Forest Eudicots Lecythidaceae Eschweilera coriacea 0.34 43 1118.08 6.14 7.18 3.55 16.87

37 Eudicots Fabaceae Monopteryx uaucu 0.1 10 2860.79 1.81 1.67 9.09 12.57

Eudicots Euphorbiaceae Pseudosenefeldera inclinata 0.22 26 483.54 3.97 4.34 1.54 9.85

Eudicots Lecythidaceae Cariniana micrantha 0.01 1 2415.82 0.18 0.17 7.68 8.03

Eudicots Fabaceae Clathrotropis macrocarpa 0.15 19 390.16 2.71 3.17 1.24 7.12

Eudicots Euphorbiaceae Hevea guianensis 0.14 14 643.24 2.53 2.34 2.04 6.91

Eudicots Fabaceae Cynometra longicuspis 0.11 11 627.21 1.99 1.84 1.99 5.82

Magnoliids Myristicaceae Osteophloeum platyspermum 0.07 7 788.12 1.26 1.17 2.51 4.94

Eudicots Olacaceae Minquartia guianensis 0.07 7 765.65 1.26 1.17 2.43 4.87

Eudicots Fabaceae Tachigali vaupesiana 0.85 257 16242.39 33.74 56.73 68.65 159.13

Eudicots Rubiaceae Duroia micrantha 0.3 39 776.11 11.93 8.61 3.28 23.82

Eudicots Vochysiaceae Vochysia obscura 0.21 24 2325.25 8.23 5.3 9.83 23.36

Eudicots Malpighiaceae Byrsonima japurensis 0.24 35 1050.78 9.47 7.73 4.44 21.63

CASA_PII_2 Eudicots Fabaceae Swartzia leptopetala 0.14 14 587.93 5.35 3.09 2.49 10.93 Seasonally- flooded Forest Eudicots Euphorbiaceae Mabea trianae 0.16 16 275.62 6.17 3.53 1.16 10.87 Eudicots Fabaceae Hydrochorea marginata 0.06 12 1088.06 2.47 2.65 4.6 9.72

Eudicots Calophyllaceae Caraipa llanorum 0.08 8 205.49 3.29 1.77 0.87 5.93

Eudicots Euphorbiaceae Alchornea discolor 0.06 6 100.16 2.47 1.32 0.42 4.22

Eudicots Chrysobalanaceae Licania heteromorpha 0.05 5 176.87 2.06 1.1 0.75 3.91

Eudicots Phyllanthaceae Phyllanthus elsiae 0.61 207 8473.21 25.75 47.81 37.57 111.12

Eudicots Boraginaceae Cordia tetrandra 0.46 74 6044.43 19.31 17.09 26.8 63.2

Eudicots Fabaceae Erythrina fusca 0.2 21 4253.65 8.58 4.85 18.86 32.29

Eudicots Bignoniaceae Crescentia amazonica 0.29 31 747.14 12.02 7.16 3.31 22.49

CASA_PIV_1 Eudicots Urticaceae Cecropia engleriana 0.22 32 1025.54 9.44 7.39 4.55 21.38 Seasonally- flooded Forest Eudicots Melastomataceae Mouriri guianensis 0.12 15 353.97 5.15 3.46 1.57 10.18

Eudicots Anacardiaceae Spondias mombin 0.07 13 936.68 3 3 4.15 10.16

Eudicots Fabaceae Lonchocarpus densiflorus 0.12 12 157.49 5.15 2.77 0.7 8.62

Eudicots Calophyllaceae Calophyllum brasiliense 0.05 5 93.98 2.15 1.15 0.42 3.72

Eudicots Euphorbiaceae Alchornea fluviatilis 0.03 3 84.76 1.29 0.69 0.38 2.36

38 Eudicots Malvaceae Cavanillesia indet 0.03 3 4211.01 0.58 0.53 13 14.1

Magnoliids Annonaceae Ephedranthus colombianus 0.16 18 1325.41 3.08 3.17 4.09 10.34

Eudicots Lecythidaceae Grias haughtii 0.17 24 571.34 3.27 4.23 1.76 9.26

Eudicots Rubiaceae Simira rubescens 0.14 21 918.95 2.69 3.7 2.84 9.23

QUIN_TF_1 Magnoliids Myristicaceae Virola flexuosa 0.16 17 871.84 3.08 2.99 2.69 8.76 Terra firme Forest Magnoliids Annonaceae Oxandra panamensis 0.16 18 565.83 3.08 3.17 1.75 7.99 Eudicots Malvaceae Apeiba tibourbou 0.12 15 596.7 2.31 2.64 1.84 6.79

Eudicots Lecythidaceae Eschweilera andina 0.14 14 493.33 2.69 2.46 1.52 6.68

Eudicots Urticaceae Cecropia insignis 0.13 17 289.3 2.5 2.99 0.89 6.39

Eudicots Burseraceae Tetragastris panamensis 0.12 13 458.01 2.31 2.29 1.41 6.01

Eudicots Malvaceae Huberodendron patinoi 0.04 4 5286.05 0.69 0.66 13.13 14.48

Eudicots Lecythidaceae Eschweilera andina 0.16 16 3369.04 2.76 2.64 8.37 13.77

Eudicots Fabaceae Andira chigorodensis 0.15 16 1674.84 2.59 2.64 4.16 9.39

Eudicots Euphorbiaceae Hura crepitans 0.03 3 3183.11 0.52 0.5 7.9 8.92

QUIN_TF_3 Magnoliids Myristicaceae Virola flexuosa 0.12 12 1575.35 2.07 1.98 3.91 7.96 Terra firme Forest Eudicots Fabaceae Clathrotropis brunnea 0.14 15 1066.4 2.41 2.48 2.65 7.54 Monocots Arecaceae Oenocarpus bataua 0.17 19 507.63 2.93 3.14 1.26 7.33

Eudicots Clusiaceae Garcinia madruno 0.17 18 527.45 2.93 2.97 1.31 7.21

Magnoliids Myristicaceae Virola sebifera 0.16 18 510.33 2.76 2.97 1.27 7

Eudicots Urticaceae Cecropia insignis 0.11 13 373.66 1.9 2.15 0.93 4.97

Eudicots Fabaceae Clathrotropis brunnea 0.21 21 2327.36 4.73 4.71 7.4 16.83

Eudicots Lecythidaceae Eschweilera andina 0.1 10 1270.6 2.36 2.24 4.04 8.64

Eudicots Fabaceae Indet indet 0.01 1 2344.85 0.24 0.22 7.45 7.91

QUIN_TF_4 Eudicots Fabaceae Hymenaea courbaril 0.07 7 1288.34 1.65 1.57 4.1 7.32 Terra firme Forest Eudicots Urticaceae Pourouma hirsutipetiolata 0.1 12 496.11 2.36 2.69 1.58 6.63

Eudicots Fabaceae Swartzia amabale 0.05 5 1151.76 1.18 1.12 3.66 5.96

Eudicots Bignoniaceae Jacaranda copaia 0.08 8 517.91 1.89 1.79 1.65 5.33

Eudicots Urticaceae Cecropia insignis 0.07 12 296.83 1.65 2.69 0.94 5.29

39 Magnoliids Myristicaceae Virola flexuosa 0.06 6 744.41 1.42 1.35 2.37 5.13

Magnoliids Annonaceae Unonopsis aviceps 0.09 10 124.49 2.13 2.24 0.4 4.77

Eudicots Urticaceae Pourouma melinonii 0.31 35 1820.63 6.01 6.42 7.23 19.67

Eudicots Malvaceae Trichospermum galeottii 0.18 25 845.46 3.41 4.59 3.36 11.35

Eudicots Salicaceae Laetia procera 0.15 16 834.93 3.01 2.94 3.32 9.26

Monocots Arecaceae Oenocarpus bataua 0.16 22 470.25 3.21 4.04 1.87 9.11

QUIN_TF_5 Eudicots Sapotaceae Chrysophyllum lucentifolium 0.15 17 678.9 3.01 3.12 2.7 8.82 Terra firme Forest Eudicots Malvaceae Catostemma digitata 0.04 4 1600.32 0.8 0.73 6.36 7.89 Monocots Arecaceae Socratea exorrhiza 0.15 17 246.7 3.01 3.12 0.98 7.11

Eudicots Malvaceae Apeiba tibourbou 0.12 14 524.24 2.4 2.57 2.08 7.06

Eudicots Moraceae Ficus americana 0.02 2 1550.31 0.4 0.37 6.16 6.93

Eudicots Moraceae Pseudolmedia rigida 0.13 15 330.08 2.61 2.75 1.31 6.67

Eudicots Boraginaceae Cordia collococca 0.62 102 4899.58 15.6 21.07 9.52 46.19

Eudicots Anacardiaceae Spondias mombin 0.42 63 11256.81 10.74 13.02 21.86 45.62

Eudicots Moraceae Ficus insipida 0.11 14 11380.71 2.81 2.89 22.11 27.81

Eudicots Rubiaceae Genipa americana 0.31 41 1399.89 7.93 8.47 2.72 19.12

SAJU_PI_1 Eudicots Euphorbiaceae Hura crepitans 0.15 15 3111.97 3.84 3.1 6.04 12.98 Seasonally- flooded Forest Eudicots Malvaceae Luehea seemannii 0.14 14 3276.34 3.58 2.89 6.36 12.84

Eudicots Malvaceae Guazuma ulmifolia 0.13 14 1808.3 3.32 2.89 3.51 9.73

Eudicots Sapotaceae Pouteria baehniana 0.16 19 352.54 4.09 3.93 0.68 8.7

Eudicots Chrysobalanaceae Licania platypus 0.02 3 3184.63 0.51 0.62 6.19 7.32

Magnoliids Annonaceae Pseudomalmea boyacana 0.13 15 384.49 3.32 3.1 0.75 7.17

Eudicots Malvaceae Luehea seemannii 0.35 41 10659.28 10.97 11.71 29.49 52.18

Eudicots Anacardiaceae Spondias mombin 0.26 28 9317.4 8.06 8 25.78 41.85

SAJU_PI_2 Eudicots Boraginaceae Cordia collococca 0.28 32 1524 8.71 9.14 4.22 22.07 Seasonally- flooded Forest Eudicots Fabaceae Zygia inaequalis 0.18 21 2324.2 5.48 6 6.43 17.92 Eudicots Rubiaceae Faramea capillipes 0.21 29 304.55 6.45 8.29 0.84 15.58

Eudicots Rubiaceae Genipa americana 0.15 16 702.54 4.52 4.57 1.94 11.03

40 Eudicots Combretaceae Terminalia oblonga 0.13 13 1162.26 3.87 3.71 3.22 10.8

Eudicots Moraceae Clarisia biflora 0.15 15 509.46 4.52 4.29 1.41 10.21

Eudicots Sapotaceae Pouteria baehniana 0.14 13 504.48 4.19 3.71 1.4 9.3

Eudicots Fabaceae Senegalia polyphylla 0.09 11 810.24 2.9 3.14 2.24 8.29

Eudicots Euphorbiaceae Alchornea triplinervia 0.19 25 1831.15 4.47 5.27 10.89 20.63

Eudicots Fabaceae Zygia ocumarensis 0.28 41 765.24 6.59 8.65 4.55 19.79

Eudicots Lecythidaceae Corythophora labriculata 0.28 32 886.18 6.59 6.75 5.27 18.61

Eudicots Fabaceae Inga indet 0.18 21 1274.65 4.24 4.43 7.58 16.25

SAJU_TF_3 Eudicots Bignoniaceae Jacaranda copaia 0.15 18 1370.81 3.53 3.8 8.15 15.48 Terra firme Forest Magnoliids Myristicaceae Virola peruviana 0.22 23 505.08 5.18 4.85 3 13.03 Eudicots Fabaceae Clathrotropis brunnea 0.15 15 662.96 3.53 3.16 3.94 10.64

Magnoliids Annonaceae Xylopia polyantha 0.15 15 405.45 3.53 3.16 2.41 9.11

Eudicots Burseraceae Protium sagotianum 0.1 11 628.44 2.35 2.32 3.74 8.41

Eudicots Fabaceae Inga acuminata 0.13 13 315.19 3.06 2.74 1.87 7.68

Eudicots Lecythidaceae Corythophora labriculata 0.61 91 4653.27 16.44 21.41 22.16 60.01

Eudicots Moraceae Pseudolmedia laevigata 0.31 36 1304.72 8.22 8.47 6.21 22.9

Eudicots Fabaceae Crudia indet 0.15 15 1631.24 4.11 3.53 7.77 15.41

Magnoliids Myristicaceae Virola peruviana 0.16 19 992.83 4.38 4.47 4.73 13.58

SAJU_TF_4 Eudicots Fabaceae Clathrotropis brunnea 0.15 15 829.48 4.11 3.53 3.95 11.59 Terra firme Forest Eudicots Moraceae Helianthostylis sprucei 0.14 20 400.39 3.84 4.71 1.91 10.45

Monocots Arecaceae Oenocarpus indet 0.15 17 376.18 4.11 4 1.79 9.9

Eudicots Fabaceae Swartzia oraria 0.11 11 792.04 3.01 2.59 3.77 9.37

Eudicots Moraceae Helicostylis tomentosa 0.14 15 349.77 3.84 3.53 1.67 9.03

Eudicots Cardiopteridaceae Dendrobangia boliviana 0.08 8 801.6 2.19 1.88 3.82 7.89

Eudicots Fabaceae Brownea stenantha 0.27 52 5516.51 6.09 9.9 18.4 34.39

SAJU_TF_5 Magnoliids Myristicaceae Virola sebifera 0.12 11 6674.14 2.58 2.1 22.26 26.93 Terra firme Forest Eudicots Lecythidaceae Corythophora labriculata 0.36 44 1938.68 7.96 8.38 6.47 22.81

Monocots Arecaceae Oenocarpus bataua 0.38 58 883.2 8.43 11.05 2.95 22.42

41 Magnoliids Myristicaceae Iryanthera ulei 0.2 25 648.71 4.45 4.76 2.16 11.38

Eudicots Lecythidaceae Eschweilera coriacea 0.14 18 936.7 3.04 3.43 3.12 9.6

Eudicots Rubiaceae Wittmackanthus stanleyanus 0.17 17 471.34 3.75 3.24 1.57 8.56

Eudicots Apocynaceae Aspidosperma spruceanum 0.04 5 1926.56 0.94 0.95 6.43 8.31

Magnoliids Lauraceae Indet indet 0.14 15 364.43 3.04 2.86 1.22 7.12

Magnoliids Annonaceae Xylopia polyantha 0.14 14 375.17 3.04 2.67 1.25 6.96

Eudicots Burseraceae Protium heptaphyllum 0.47 77 2408.19 11.99 14.69 9.99 36.68

Eudicots Euphorbiaceae Pera arborea 0.27 53 3916.13 6.89 10.11 16.25 33.26

Magnoliids Annonaceae Xylopia polyantha 0.33 44 1008.54 8.42 8.4 4.19 21

Monocots Arecaceae Oenocarpus bataua 0.23 33 1587.86 5.87 6.3 6.59 18.76

SAMA_TF_1 Eudicots Euphorbiaceae Alchornea triplinervia 0.14 19 2648.22 3.57 3.63 10.99 18.19 Terra firme Forest Magnoliids Myristicaceae Virola sebifera 0.22 28 1028.72 5.61 5.34 4.27 15.23 Eudicots Burseraceae Protium llanorum 0.15 31 1181.82 3.83 5.92 4.9 14.65

Eudicots Burseraceae Trattinnickia rhoifolia 0.1 10 2406.53 2.55 1.91 9.99 14.45

Eudicots Melastomataceae Miconia elata 0.2 31 452.1 5.1 5.92 1.88 12.89

Magnoliids Myristicaceae Iryanthera laevis 0.12 17 628.9 3.06 3.24 2.61 8.92

Eudicots Moraceae Pseudolmedia laevis 0.58 95 1671.83 12.58 16.78 8.39 37.76

Magnoliids Lauraceae Indet indet 0.38 46 2467.84 8.39 8.13 12.39 28.91

Eudicots Clusiaceae Garcinia madruno 0.37 58 1376.39 8.17 10.25 6.91 25.33

Eudicots Burseraceae Trattinnickia rhoifolia 0.29 34 1931.36 6.4 6.01 9.7 22.11

SAMA_TF_2 Eudicots Apocynaceae Himatanthus articulatus 0.3 33 1281.47 6.62 5.83 6.43 18.89 Terra firme Forest Magnoliids Annonaceae Xylopia polyantha 0.26 36 1113.02 5.74 6.36 5.59 17.69

Eudicots Ebenaceae Diospyros pseudoxylopia 0.16 20 1333.87 3.53 3.53 6.7 13.76

Eudicots Euphorbiaceae Pera arborea 0.17 20 858.54 3.75 3.53 4.31 11.6

Eudicots Burseraceae Protium heptaphyllum 0.19 22 671.76 4.19 3.89 3.37 11.45

Monocots Arecaceae Oenocarpus bataua 0.14 14 864.12 3.09 2.47 4.34 9.9

SAMA_TF_3 Magnoliids Myristicaceae Virola elongata 0.44 64 1206.53 10.46 13.73 6.94 31.14 Terra firme Forest Eudicots Bignoniaceae Jacaranda copaia 0.21 26 1863.55 5.11 5.58 10.73 21.42

42 Eudicots Sapotaceae Sarcaulus brasiliensis 0.27 35 1172.32 6.33 7.51 6.75 20.58

Eudicots Phyllanthaceae Hieronyma oblonga 0.17 20 810.13 4.14 4.29 4.66 13.09

Eudicots Fabaceae Dialium guianense 0.07 7 1197.95 1.7 1.5 6.9 10.1

Eudicots Chrysobalanaceae Licania kunthiana 0.1 10 879.2 2.43 2.15 5.06 9.64

Monocots Arecaceae Oenocarpus bataua 0.14 15 495.32 3.41 3.22 2.85 9.48

Magnoliids Annonaceae Duguetia odorata 0.13 14 407.55 3.16 3 2.35 8.51

Eudicots Achariaceae Lindackeria paludosa 0.15 15 221.22 3.65 3.22 1.27 8.14

Eudicots Apocynaceae Aspidosperma spruceanum 0.06 7 836.34 1.46 1.5 4.81 7.78

Eudicots Malvaceae Luehea tessmannii 0.67 126 11445.01 29.52 31.58 38.7 99.81

Eudicots Salicaceae Laetia corymbulosa 0.7 185 5243.19 30.48 46.37 17.73 94.57

Eudicots Moraceae Ficus insipida 0.02 2 3092.85 0.95 0.5 10.46 11.91

Eudicots Moraceae Ficus trigona 0.03 4 2491.77 1.43 1 8.43 10.86

TINI_PI_2 Eudicots Moraceae Brosimum lactescens 0.09 8 1358.45 3.81 2.01 4.59 10.41 Seasonally- flooded Forest Eudicots Sapotaceae Pouteria procera 0.1 11 637.39 4.29 2.76 2.16 9.2 Eudicots Moraceae Ficus maxima 0.05 5 1167.5 2.38 1.25 3.95 7.58

Eudicots Malvaceae Pseudobombax munguba 0.08 7 348.74 3.33 1.75 1.18 6.27

Eudicots Polygonaceae Ruprechtia indet 0.05 5 652.75 2.38 1.25 2.21 5.84

Eudicots Rubiaceae Calycophyllum spruceanum 0.05 5 340.03 2.38 1.25 1.15 4.78

Eudicots Meliaceae Guarea guidonia 0.37 70 11161.35 12.72 15.59 28.12 56.43

Eudicots Urticaceae Cecropia membranacea 0.18 89 4308.28 6.01 19.82 10.85 36.68

Eudicots Moraceae Ficus insipida 0.05 5 5386.75 1.77 1.11 13.57 16.45

Eudicots Salicaceae Laetia corymbulosa 0.16 23 688.91 5.65 5.12 1.74 12.51

TINI_PI_5 Monocots Arecaceae Socratea exorrhiza 0.16 25 382.4 5.65 5.57 0.96 12.19 Seasonally- flooded Forest Eudicots Moraceae Ficus andicola 0.01 1 4074.42 0.35 0.22 10.26 10.84

Eudicots Fabaceae Inga cylindrica 0.09 12 1876.35 3.18 2.67 4.73 10.58

Eudicots Euphorbiaceae Alchornea glandulosa 0.12 20 541.44 4.24 4.45 1.36 10.06

Eudicots Moraceae Ficus maxima 0.07 9 2200.85 2.47 2 5.54 10.02

Eudicots Fabaceae Inga marginata 0.11 14 386.18 3.89 3.12 0.97 7.98

43 Eudicots Moraceae Brosimum alicastrum 0.11 11 5864.19 2 1.86 19.25 23.11

Magnoliids Annonaceae Oxandra mediocris 0.34 42 1398.93 6.18 7.09 4.59 17.87

Eudicots Lecythidaceae Gustavia hexapetala 0.2 23 2239.86 3.64 3.89 7.35 14.87

Eudicots Anacardiaceae Spondias mombin 0.08 8 2225.48 1.45 1.35 7.31 10.11

TINI_TF_1 Monocots Arecaceae Astrocaryum chambira 0.18 19 775.08 3.27 3.21 2.54 9.03 Terra firme Forest Magnoliids Annonaceae Pseudomalmea diclina 0.14 14 728.52 2.55 2.36 2.39 7.3 Eudicots Rubiaceae Alibertia hadrantha 0.13 15 678.22 2.36 2.53 2.23 7.12

Eudicots Meliaceae Trichilia micrantha 0.15 16 477.36 2.73 2.7 1.57 7

Eudicots Burseraceae Protium sagotianum 0.14 17 474.81 2.55 2.87 1.56 6.98

Eudicots Araliaceae Dendropanax caucanus 0.15 15 368.16 2.73 2.53 1.21 6.47

Monocots Poaceae Guadua angustifolia 0.23 168 2122.53 4.6 24.6 10.83 40.03

Eudicots Malvaceae Theobroma glaucum 0.27 31 429.79 5.4 4.54 2.19 12.13

Eudicots Melastomataceae Henriettella fissanthera 0.11 18 689.08 2.2 2.64 3.52 8.35

Eudicots Malvaceae Apeiba aspera 0.15 15 535.92 3 2.2 2.73 7.93

TINI_TF_3 Eudicots Araliaceae Schefflera morototoni 0.07 8 932.12 1.4 1.17 4.76 7.33 Terra firme Forest Eudicots Fabaceae Dalbergia indet 0.16 18 224.25 3.2 2.64 1.14 6.98 Monocots Arecaceae Astrocaryum chambira 0.11 13 497.79 2.2 1.9 2.54 6.64

Eudicots Bignoniaceae Jacaranda copaia 0.08 8 607.7 1.6 1.17 3.1 5.87

Eudicots Malvaceae Quararibea wittii 0.11 12 193.61 2.2 1.76 0.99 4.94

Eudicots Fabaceae Inga stenoptera 0.08 9 396.81 1.6 1.32 2.02 4.94

Eudicots Moraceae Pseudolmedia laevigata 0.34 43 1481.26 6.65 7.44 5.84 19.93

Monocots Arecaceae Oenocarpus bataua 0.23 38 1380.25 4.5 6.57 5.44 16.52

Eudicots Moraceae Pseudolmedia laevis 0.26 32 712.25 5.09 5.54 2.81 13.43

TINI_TF_4 Eudicots Burseraceae Protium sagotianum 0.2 24 1152.69 3.91 4.15 4.54 12.61 Terra firme Forest Eudicots Malvaceae Theobroma glaucum 0.26 30 478.93 5.09 5.19 1.89 12.17

Eudicots Sapindaceae Talisia intermedia 0.17 21 775.59 3.33 3.63 3.06 10.02

Eudicots Fabaceae Dalbergia indet 0.17 21 386.83 3.33 3.63 1.53 8.49

Eudicots Urticaceae Pourouma bicolor 0.11 11 725.45 2.15 1.9 2.86 6.92

44 Eudicots Burseraceae Crepidospermum rhoifolium 0.15 16 197.43 2.94 2.77 0.78 6.48 Pseudopiptadenia Eudicots Fabaceae 0.05 5 1173.01 0.98 0.87 4.62 6.47 suaveolens Eudicots Burseraceae Protium sagotianum 0.48 64 2435.54 10.23 12.1 10.32 32.65

Monocots Arecaceae Oenocarpus bataua 0.43 60 2468.36 9.17 11.34 10.46 30.97

Eudicots Euphorbiaceae Mabea maynensis 0.27 32 631.23 5.76 6.05 2.68 14.48

Eudicots Lecythidaceae Gustavia hexapetala 0.15 17 1479.81 3.2 3.21 6.27 12.68

TINI_TF_6 Eudicots Moraceae Pseudolmedia obliqua 0.19 21 824.12 4.05 3.97 3.49 11.51 Terra firme Forest Monocots Arecaceae Euterpe precatoria 0.19 22 327.62 4.05 4.16 1.39 9.6 Eudicots Burseraceae Trattinnickia rhoifolia 0.09 10 1166.91 1.92 1.89 4.95 8.75

Eudicots Moraceae Pseudolmedia laevis 0.13 16 690.75 2.77 3.02 2.93 8.72

Eudicots Urticaceae Coussapoa orthoneura 0.04 4 1376.6 0.85 0.76 5.83 7.44

Eudicots Malvaceae Bombacopsis quinata 0.01 1 1640.23 0.21 0.19 6.95 7.35

Monocots Arecaceae Oenocarpus bataua 0.31 58 1885.85 5.07 8.17 6.48 19.71

Monocots Poaceae Guadua angustifolia 0.11 63 760.41 1.88 8.87 2.61 13.36

Eudicots Malvaceae Theobroma glaucum 0.27 31 432.58 4.5 4.37 1.49 10.36

Eudicots Anacardiaceae Spondias mombin 0.08 21 1236.35 1.31 2.96 4.25 8.52

TINI_TF_7 Eudicots Burseraceae Protium sagotianum 0.19 19 613.25 3.19 2.68 2.11 7.97 Terra firme Phenakospermum Monocots Strelitziaceae 0.1 32 301.98 1.69 4.51 1.04 7.23 Forest guyannense Eudicots Burseraceae Protium robustum 0.19 18 370.81 3.19 2.54 1.27 7

Monocots Arecaceae Socratea exorrhiza 0.19 18 342.08 3.19 2.54 1.18 6.9

Eudicots Moraceae Pseudolmedia laevigata 0.15 17 520.35 2.44 2.39 1.79 6.62

Eudicots Moraceae Castilla ulei 0.13 12 741.09 2.06 1.69 2.55 6.3

Eudicots Fabaceae Tachigali odoratissima 0.49 75 4731.95 9.03 11.94 19.17 40.14

Eudicots Fabaceae Tachigali vaupesiana 0.4 63 4068.58 7.41 10.03 16.48 33.92 TOMO_PI_4 Seasonally- Eudicots Euphorbiaceae Mabea trianae 0.56 81 1965.32 10.42 12.9 7.96 31.28 flooded Forest Magnoliids Annonaceae Guatteria brevicuspis 0.33 43 2734.93 6.02 6.85 11.08 23.94

Eudicots Calophyllaceae Caraipa llanorum 0.28 31 2565.25 5.09 4.94 10.39 20.42

45 Eudicots Lecythidaceae Eschweilera parvifolia 0.34 36 697.14 6.25 5.73 2.82 14.81

Eudicots Myrtaceae Calycorectes indet 0.24 31 670.06 4.4 4.94 2.71 12.05

Eudicots Salicaceae Laetia suaveolens 0.23 28 509.34 4.17 4.46 2.06 10.69

Eudicots Malpighiaceae Byrsonima japurensis 0.21 22 737.23 3.94 3.5 2.99 10.42

Eudicots Phyllanthaceae Discocarpus essequeboensis 0.26 24 390.2 4.86 3.82 1.58 10.26

Eudicots Euphorbiaceae Mabea trianae 0.73 160 4098.38 13.74 19.88 14.6 48.21

Eudicots Chrysobalanaceae Licania heteromorpha 0.52 77 3485.99 9.73 9.57 12.42 31.71

Eudicots Fabaceae Tachigali odoratissima 0.45 80 3490.11 8.59 9.94 12.43 30.96

Eudicots Rubiaceae Duroia micrantha 0.33 67 2903.49 6.3 8.32 10.34 24.96

TOMO_PI_5 Eudicots Lecythidaceae Eschweilera parvifolia 0.43 58 2069.32 8.21 7.2 7.37 22.78 Seasonally- flooded Forest Eudicots Chrysobalanaceae Licania apetala 0.37 50 1518.54 7.06 6.21 5.41 18.68 Eudicots Salicaceae Laetia suaveolens 0.3 47 1106.65 5.73 5.84 3.94 15.51

Eudicots Sapotaceae Pouteria elegans 0.3 36 1174.04 5.73 4.47 4.18 14.38

Eudicots Clusiaceae Tovomita spruceana 0.28 48 705.71 5.34 5.96 2.51 13.82

Eudicots Fabaceae Tachigali vaupesiana 0.21 24 1417.62 4.01 2.98 5.05 12.04

Eudicots Bignoniaceae Jacaranda copaia 0.41 66 6575.33 10.9 15.24 30.92 57.06

Monocots Arecaceae Attalea maripa 0.33 39 2455.48 8.72 9.01 11.55 29.27

Magnoliids Annonaceae Bocageopsis multiflora 0.33 41 627.88 8.72 9.47 2.95 21.14

Eudicots Goupiaceae Goupia glabra 0.18 19 1300.83 4.9 4.39 6.12 15.41

TOMO_TF_1 Eudicots Burseraceae Tetragastris panamensis 0.13 14 914.7 3.54 3.23 4.3 11.08 Terra firme Forest Magnoliids Annonaceae Guatteria foliosa 0.17 17 483.47 4.63 3.93 2.27 10.83

Eudicots Sapindaceae Cupania scrobiculata 0.16 22 282.95 4.36 5.08 1.33 10.77

Eudicots Moraceae Pseudolmedia laevis 0.14 18 322.43 3.81 4.16 1.52 9.49

Eudicots Sapotaceae Pouteria multiflora 0.09 10 640.9 2.45 2.31 3.01 7.78

Eudicots Rubiaceae Capirona decorticans 0.12 12 265.49 3.27 2.77 1.25 7.29

Monocots Arecaceae Attalea maripa 0.39 51 2858.51 10.54 12.06 16.53 39.12 TOMO_TF_2 Terra firme Eudicots Burseraceae Tetragastris panamensis 0.29 36 2197.87 7.84 8.51 12.71 29.06 Forest Magnoliids Annonaceae Xylopia polyantha 0.19 20 407.51 5.14 4.73 2.36 12.22

46 Eudicots Goupiaceae Goupia glabra 0.12 14 881.06 3.24 3.31 5.09 11.65

Eudicots Moraceae Pseudolmedia laevis 0.16 20 283.8 4.32 4.73 1.64 10.69

Magnoliids Annonaceae Bocageopsis multiflora 0.15 18 311.98 4.05 4.26 1.8 10.11

Eudicots Rubiaceae Macrocnemum indet 0.16 19 221.63 4.32 4.49 1.28 10.1

Eudicots Bignoniaceae Jacaranda copaia 0.06 6 858.58 1.62 1.42 4.96 8

Eudicots Sapotaceae Pouteria multiflora 0.09 9 422.74 2.43 2.13 2.44 7

Eudicots Euphorbiaceae Conceveiba tristigmata 0.11 11 221.38 2.97 2.6 1.28 6.85

Monocots Arecaceae Oenocarpus bataua 0.28 41 1537.13 6.96 9.09 10.37 26.42

Eudicots Urticaceae Pourouma aurea 0.26 34 915.52 6.44 7.54 6.18 20.16

Eudicots Moraceae Pseudolmedia laevis 0.28 36 437.32 6.96 7.98 2.95 17.89

Magnoliids Annonaceae Guatteria foliosa 0.19 26 906.93 4.9 5.76 6.12 16.78

TOMO_TF_3 Eudicots Euphorbiaceae Alchornea triplinervia 0.15 16 1232.3 3.87 3.55 8.31 15.73 Terra firme Forest Monocots Arecaceae Attalea maripa 0.15 18 1119.42 3.87 3.99 7.55 15.41 Eudicots Bignoniaceae Jacaranda copaia 0.09 11 1228.55 2.32 2.44 8.29 13.05

Magnoliids Annonaceae Bocageopsis multiflora 0.2 21 462.11 5.15 4.66 3.12 12.93

Monocots Arecaceae Socratea exorrhiza 0.16 24 276.31 4.12 5.32 1.86 11.31

Eudicots Fabaceae Stryphnodendron guianense 0.12 12 712.43 3.09 2.66 4.81 10.56

47

ESM4: Figure showing the correlation between NRI and species richness for flooded (squares) and terra firme (triangles) forest plots northern southamerica. Correlation= -0.36, pearsons t=-2.1, P=0.043

48 Appendix 1 The most important species, in terms of the Index of Importance, of flooded forests belong mainly to the Fabaceae and Malvaceae families. While the most important species of the terra firme forests belong to the Arecaceae, Fabaceae, Burseraceae and Moraceae families (ESM-Table 2).

The floristic composition of the várzea and igapó communities included in this study are very different at the family, genus and species levels (ESM-Table 2). Várzea forests of the Magdalena and Orinoco basins are dominated by species belonging to the Malvaceae and Salicaceae and the dominant species in the igapó forests of the Amazon and Orinoco basins belong mainly to the Fabaceae and Euphorbiaceae. These patterns coincide with what has been reported for terra firme and Várzea forests of the Amazon Basin (Wittman et a. 2010).

However, NRI calculations are done only considering species abundance, and not importance, and this fact confounds the trends, because at the plot level dominance varies greatly within localities and forest types. For instance, the most abundant species in the terra firme plots is the palm (monocot) Oenocarpus bataua, but it is not present in all terra firme plots (18/22), it is followed in abundance by the annonaceae (magnoliid) Xylopia polyantha, found in 17 of the 22 plots. Nonetheless, the abuncance of palms and species belonging to the myristicaceae and annonaceae could help explain high phylogenetic diversity in terra firme forests, when compared to flooded forests.

On the other hand, the most abundant species in the igapó forest is the legume (eudicot) Tachigali chrysophylla, which is not found in two of the five plots, it is followed in abundance by the euphorb (eudicot) Mabea trianae, which is found in all five Igapó plots. Dominance differs very much within plots in várzea forests, and the dominant species in one plot is usually not present in other plots; aditionally, these species belong to different families and orders (Phyllantaceae – Malpigiales; Boraginaceae – Boraginales; Malvaceae – Malvales; Malvaceae – Malvales; Meliaceae – Sapindales), but all belong to the eudicots.

Wittmann, F., Schöngart, J., Junk, W.J., 2010. Phytogeography, Species Diversity, Community Structure and Dynamics of Central Amazonian Floodplain Forests, in: Junk, W.J., Piedade, M.T.F., Wittmann, F., Schöngart, J., Parolin, P. (Eds.), Amazonian Floodplain Forests: Ecophysiology, Biodiversity and Sustainable Management, Ecological Studies. Springer Netherlands, Dordrecht.

49 Appendix 2 We calculated NRI for all plots following the process described in the Methods, with information of the abundance of species belonging only to the Eudicots clade. The results are similar: NRI values are higher in flooded forests than in terra firme forests (Figure A-2). This confirms the idea that seasonally flooded forests in Northern South America, have a community of species that are more related to each other than terra firme forests, suggesting that environmental filtering might have possiteviley selected for species of some orders of the eudicots.

Fig. A-2: Distribution of Net Relatedness Index (NRI) values for flooded and terra firme forests in Northern South America, calculated with information of species belonging to the Eudicots clade. NRI values of seasonally-flooded forests are higher (ANOVA, d.f=1, F= 4.21, p=0.048)

50

FLOODING AND DIFFERENCES IN SOIL COMPOSITION DETERMINE BETA- DIVERSITY OF LOWLAND FORESTS IN NORTHERN SOUTH AMERICA

Stevenson, P. R., Aldana, A. M., Cárdenas-Hoyos, S. & Negret, P. Flooding and differences in soil composition determine beta-diversity of lowland forests in Northern South America. Manuscript submitted to Journal of Biogeography on October 17th 2016.

51 Original Article

Title: Flooding and differences in soil composition determine beta-diversity of lowland forests in Northern South America

Pablo R. Stevenson1*, Ana M. Aldana1, Sasha Cárdenas1, Pablo José Negret1

1 Laboratorio de Ecología de Bosques Tropicales y Primatología, Departamento de Ciencias Biológicas, Universidad de Los Andes, Bogotá, Colombia.

* Corresponding author: Departamento de Ciencias Biológicas, Universidad de Los Andes, Carrera 1 No. 18ª -12, Bogotá, Colombia. Postcode: 111711 e-mail address: [email protected]

Running header: Beta-diversity of Northern South American Rainforests

Word count: 5,718

52 Abstract Aim

Beta diversity may be determined by dispersal limitation, environmental variables and phylogeographic history. Our objective is to advance in the understanding of plant species turnover in rainforests in Northern South America and determine which factors are affecting this species turnover.

Location

Northern South America: Amazon, Magdalena and Orinoco Basins

Methods

We evaluated the relative effect of environmental variables (i.e. soil, climate, fragmentation and flooding frequency) and dispersal limitation (i.e. geographical distance and resistance distance due mountain barriers) on tree beta diversity in 32 one-hectare vegetation plots lowland forests.

Results

We found that tree species turnover was better explained by environmental distance than by geographic distance. Although, soil traits and flooding regime were good predictors of tree species composition, the majority of the variance remained unexplained. In our study system the Eastern Andean ridge had no significant effect on plant beta diversity, probably because of its young age in relation to the phylogeny.

Main conclusions

Our results provide support for the hypotheses of niche theory and disregards a strong role of dispersal limitation. Therefore, we advise that conservation strategies of lowland trees should consider forest types peculiarities (e.g. seasonally flooded vs. terra firme, as well as piedmont vs. central Amazonian forests).

Key Words: b-diversity, Colombian forests, igapó, tree species turnover, várzea.

53 Introduction Beta diversity refers to the spatial change in the composition of species in ecological communities. Three major factors have been proposed to explain the spatial turnover of species composition: climatic variation (Condit et al., 2002, 2013; Engelbrecht et al., 2007), topographic and edaphic variation (Tuomisto et al., 1995, 2003; Higgins et al., 2011; Condit et al., 2013) and dispersal limitation (Rosindell et al., 2011; Hubbell 2001). In addition, the history of colonization and potential diversification of lineages in a region may also contribute to the patterns of diversity (Hickerson et al., 2010). Hubbell`s unified neutral theory (2001) proposes that dispersal limitation, speciation and ecological drift are the main factors shaping the turnover of species in nature. However, neutral theory is often not successful at explaining diversity patterns in communities with high climate and topographic variability (Condit et al., 2002, 2013; Umaña et al., 2012). For instance, across Panama, there is a large range of rainfall variation that results in larger beta-diversity than in lowland Amazonian Perú and Ecuador (Condit et al., 2002), where rainfall patterns are similar. In addition, barriers such as high mountains or large rivers may also obscure the relationship between beta diversity and linear geographical distance. However, few studies have compared the effect of such barriers on the composition of tree assemblages. One example is plant communities in the western and central Andean Cordilleras which showed important differences in species composition (Idárraga et al., 2016). A large impediment in testing if the Andean cordilleras are factors limiting dispersal of lowland plant species, is that the floras at different sides correspond to environmental dissimilar biotas and gamma diversity is known to affect the patterns of beta diversity (Quian & Song, 2012). However, in the Middle Magdalena Valley there are humid forest with similar rainfall patterns to that of the Amazonian forests, but patterns of beta-diversity in this region have not been explored before.

According to niche theory, it would be expected that species coexist in response to competition and specific adaptations regarding habitat conditions. Therefore, environmentally similar areas will be similar in terms of species composition (Hutchinson, 1959; Tilman & Pacala, 1993). Additionally, it has

54 also been observed that human disturbance (such as logging and habitat fragmentation), generates significant changes in abiotic conditions, altering floristic composition (Stevenson & Aldana, 2008; Laurance et al., 2002). Because of the large number of variables that may affect species turnover, it is important to generate studies that determine the relative role of each factor on the diversity in different regions and at different scales (Gentry, 1988; Barreto et al., 2010; Lopes & Duke, 2010).

Currently, there is no consensus on which is the most determinant process in species turnover (Condit et al., 2002; Rosindell et al., 2011). There are studies that have assessed species turnover and, support dispersal limitation (Bell, 2001; Volkov et al., 2003), or show strong environmental effects (Condit et al., 2013; Engelbrecht et al., 2007) and studies that support both (Duque et al., 2002, 2009). Dispersal limitation is often tested from the prediction that dissimilarity in species composition increases with geographical distance (Hubbell, 2001). However, positive correlations between dissimilarity and geographical distance may also result from the positive association between environmental factors and distance (i.e. because ecological factors tend to be similar nearby sites and increase with distance) (Gilbert & Lechowicz, 2004; Nekola & White, 1999). Therefore, to reveal the patterns of species turnover, it is necessary to assess the interaction between ecological factors and geographical distance (Tuomisto et al., 2003), and ideally, set a spatial sampling design in which the association between environment and distance is removed. This decoupling between environmental traits and geographical distance has been achieved by a posteriri choosing the communities under study to minimize the association (Gilbert & Lechowicz, 2004). In this study we take advantage that some environmental conditions related to soils and flooding regimes may vary at short geographic distance, and have established a sampling design where units in each site included some variation in environmental conditions at short distances.

The Northern South America region holds a great diversity of plants, and yet knowledge of the detailed distribution of many species is still lacking, as well as the turnover patterns across the region (Idárraga et al., 2016). The floristic

55 knowledge of lowland forests in this area has increased in recent decades, especially in the Amazon region where there have been several studies on composition and species turnover (Londoño & Alvarez, 1997; Duque et al., 2009; ter Steege et al., 2013). Some patterns found in these studies show that variation in floristic composition is higher when ecologically and geologically different forests are compared, and distance as a proxy of dispersal limitation can also play a role in turnover (Duque et al., 2009). However, these analyses are usually based on forest plots where detailed soil variables are not included and there was no attempt to uncouple geographical distance and ecological variables.

An interesting feature of lowland forests is that being at low sea level some may be located in flood plains. These flooded forests may be classified as: várzea floodplains, which are flooded by rivers originating in the Andes (white- water rivers) which are rich in nutrients; and igapó flood plains flooded by rivers with dissolved organic matter in the form of tannins, quite acidic waters and poor in nutrients (black-water rivers) (Wolfgang, 1997). The topographic and climatic variation associated to lowland forests makes them an interesting scenario to test the effects of dispersal limitation and environmental variability on the turnover in species composition (Umaña et al., 2012).

The purpose of this study is to advance in the understanding of plant species turnover in lowland forests in Northern South America and determine which factors are affecting this species turnover. For this, we consider four main questions: 1). Can we decouple the association between environmental and geographical distance by choosing dissimilar environments in each sampling location? Accordingly, we expected to find a positive association between ecological and geographical distance for terra firme forests, and lower relationships when including both flooded and terra firme forests. 2) Does floristic similarity depend on flooding regimes? In this regard, we expect to find significant clustering in the ordination of forest plots (according to species composition) belonging to similar forest types (e.g. terra firme, várzea and igapó). 3) Does the uplift of the of the Eastern Andean Ridge have an effect on tree species composition? Given that this is the youngest Andean Ridge in

56 Colombia, we do not expect to find significant differences when comparing the association between floristic similarity and geographical distance by using linear distance and least-cost distance (determined by the mountain barrier). 4) When reducing the association between ecological and geographic distance, is the variation in the floristic composition mainly explained by differences in environmental conditions (rainfall patterns, soil type and temperature)? Under this scenario, we expect to find that sites with higher environmental differences will have more dissimilar floristic composition, regardless of the distance that separates them and environmental variables would be good predictors of floristic composition.

Materials and methods Study system We used information from 32 (1-ha) vegetation plots that were established between the years 2000 and 2012 by researchers of the LEBTYP (Laboratorio de Ecología de Bosques Tropicales y Primatología) (Stevenson et al., 2004; Aldana et al., 2008; Stevenson & Aldana, 2008, Cano & Stevenson, 2008; Correa-Gómez & Stevenson, 2010; Cárdenas & Stevenson, 2012; Umaña et al., 2012; Casas-Caro & Stevenson, 2013) from 7 localities in the Amazon, Magdalena and Orinoco Basins (Fig. 1). Plots were classified into three geomorphological units, terra firme forests, várzea and igapó floodplains.

Plots were classified as terra firme (22) if they were never flooded during a 20-year period. Seasonally flooded forests plots (10) varied in the intensity and frequency of the flood. If a plot was classified as flooded, we calculated the number of months when the topsoil was submerged in a year. The researchers building plots provided an estimate of flooding intensity based on their experience in the field (0.4 - 4 yr.), and this metric was averaged with flooding frequency based on a topographic analysis using radar sensors (Quiñones-Fernández et al., 2015). We also classified flooded forests as either igapó or várzea depending on the type of river that inundated them (Prance, 1979).

57

Fig. 1: Fig. showing the approximate location of the 32 (1ha) vegetation plots used in this study. Squares represent plots (3 -4 per square) established in terra firme forests. Upward facing triangles represent plots established in igapó forests (1-2 per triangle). Downward facing triangles represent plots established in várzea forests (1-2 per triangle). Three major river basins in the region are delimited in white: Amazon, Magdalena and Orinoco.

In each plot, we sampled all woody stems with diameter at breast height (DBH) > 10 cm and identified them to species in the field when possible. When field determinations were not possible, we collected voucher specimens to compare them with reference herbarium collections (ANDES, COL, COAH). In cases were species determination was not possible we assigned them to morphospecies. From a total database of 17,451 stems contained in the plots, we identified 1,428 taxonomic units: 1,109 species and 319 morphospecies.

Geographical and Environmental Variables

To analyze the effect of the environment on the floristic composition we used climatic and edaphic variables and a deforestation index for each plot. For most plots (25) we took 25 samples (in 20 x 20 m subplots) of topsoil and analyzed it to obtain soil texture (% of clay, sand and silt) and pH. Soil analyses were performed at the Laboratory of Soil and Water at the Universidad Nacional of Colombia and Laboratory of Soil La Sabana in Villavicencio. Information on density of Nitrogen (g/m ²) and Carbon (kg/m ²) in the soil was extracted from global layers with a resolution of 5x5 arc-min (ca.

58 10 km x 10 km) from the "Oak Ridge National Laboratory Distributed Active Archive Center" (ORNL DAAC) NASA (Global Soil Data Group, 2000). The information on temperature (Temp), minimum temperature of the coldest month (Min_temp) and precipitation (Preci) were taken from Bioclim (Hijmans et al., 2005). An index of deforestation was generated from the definition of a circular 4-kilometer radius area around each plot. In the circular area the proportion of continuous forest canopy in relation to the total area of the circle was determined, based on the Google Earth program (Google Inc., 2001).

Data Analyses

To assess if the inclusion of different forest types in each site decouples the relationship between geographic and environmental distance, first we made a geographical distance matrix of plots. Then we compared it with a matrix of environmental similarity (using six soil variables, two precipitation variables, two temperature variables, and the flooding variable) including all plots from sites with contrasting forest types and just for terra firme plots at the same sites (n = 22). We compared matrices using mantel tests, expecting to find weaker correlations for the comparison using all plots and stronger correlations when using only terra firme plots.

The floristic similarity between plots was evaluated by Chao index, which is based on the abundance of species and therefore considers rare species, which are important in diverse ecosystems (Chao et al., 2005). To illustrate the floristic similarity patterns between plots a non-metric multidimensional scaling (NMDS) was conducted (Kenkel & Orloci, 1986) and significance between forest types was obtained through a MANOVA.

To analyze the effect of Andean ridges on dispersal limitation, we compared the relationship between floristic composition and both linear geographic distance and the distance taking into account the mountain's barrier effect. This barrier distance was calculated as the minimum distance connecting two plots in areas of climatic niche availability for our lowland species (Fig. 2), using Dismo (Hijmans et al., 2013), Raster, and Gdistance packages in R (R Development Core Team, 2012) (Hijmans & van Etten, 2014). The model was

59 constructed using Maxent (Phillips et al., 2006) algorithm in R. The area suitable for lowland species was estimated from climatic niche modeling, using the majority of named species in our study plots (n = 1163). Data for this analysis were extracted from Global Biodiversity Information Facility (accessed in May 2016). 10

1e-04 5 8e-05

6e-05

0 4e-05 -5 -80 -75 -70 -65

Fig. 2: Climatic niche modelling for all named plant species in the dataset, showing the likely areas of dispersal (green tones) between the inter-Andean Valleys and eastern Colombian lowlands. Southern plots in Amazonia and the piedmont showed the shortest resistance path with plots in the Magdalena Valley though the southern portion of the Eastern ridge, while northern plots in Vichada and Casanare showed the shortest path throughout Venezuela, avoiding the low temperatures at high altitudes (white tones).

To evaluate the effect of environmental variables on the variation in species composition between plots a redundancy analysis (RDA) was performed (Legendre & Anderson, 1999). The matrix of species abundance per plot was transformed with the Hellinger distance, which gives equal weight to plots with rare species and with many individuals (Svenning et al., 2004). We used data of 25 plots for which soil analyses were available (5 igapó, 3 várzea and 17 terra firme).

Finally, to determine the effect of dispersal limitation and environmental variability restricting the effect on each other and to analyze their combined effect, variance partition analysis was conducted (Borcard et al., 1992). For this analysis, the geographical distance matrix was transformed by an

60 analysis of the Principal Coordinates Neighboring Matrices (PCNM) (Borcard & Legendre, 2002; Dray et al., 2006). Variance partition analysis between the PCNM obtained and the selected environmental variables by forward selection was performed. For this analysis the 25 plots with soil analysis were used. All analyzes were performed in the R software Vegan package (3.0.1 version) (Oksanen et al., 2011, R Development Core Team, 2011).

Results In general, we found positive associations between environmental and geographical distance. For instance, in the subset of plots where we measured soil traits and where different forest types were available, we found high association between environmental and geographical distance when measured only for the terra firme forests (r = 0.64, p= 0.002). Although, the inclusion of different forest types lowered the correlation between the variables (r = 0.46), the association remained significant (p = 0.04). Therefore, in our effort to decouple the influence of geographic and environmental distance, it helps to include different forest types in each locality, but the whole data set still includes an association.

The non-parametric ordination showed that terra firme forests tend to be more similar in species composition, regardless of their geographical location (river basins) (Fig. 3). Also a MANOVA made between plots in different forest types (terra firme, igapó and varzea) showed that the aggrupation of the plots in relation to the forest type was statistically significant (F = 4.00, d.f. = 2, p < 0.001). It also shows that species diversity (spp/stem) is higher in terra firme forests than in flooded forests. The first axis of the ordination separates forests by flooding regime: flooded forests are all on the left side and terra firme forests are all on the right side of the axis. We also found differences in floristic composition of plots when comparing the three different basins (Manova F = 4.35, d.f. = 2, p < 0.001), despite that there is continuous forest between the southern Orinoco and the Amazonian plots.

61

Fig. 3: Ordination of species composition of 32 (1-ha) vegetation plots based on non-metric multidimensional scaling (NMDS) (Stress = 0.17). Each forest type is represented by a symbol and each major basin is represented in a color. Sizes of the symbols are a representation of the species diversity of each plot estimated as species per stem: bigger symbols represent more diverse tree communities.

We did not find support for the effect of the Andean ridges on the patterns of beta-diversity. Actually, using the whole data set (n = 32) we found no correlation between floristic similarity and geographical distance (r = -0.059). Using the geographic distance derived from the least-cost model, did not improve the correlation (r = -0.018). In fact, terra firme forests in the Magdallena Valley were floristically similar to undisturbed terra firme forest at the other side of the Eastern Ridge (Fig. 3).

Using the climatic and soil variables available for 25 of our plots and the converted species composition matrix, we performed redundancy analysis (RDA). For this analysis we selected the variables that explained most of the variation of the data using a forward selection method. The variables selected were: Flooding, minimum precipitation, nitrogen, pH, Clay, mean precipitations, minimum temperature, and carbon (Fig. 4). With this analysis we got a similar result as with the NMDS: the first axis of the ordination

62 separates flooded from terra firme forests. In addition, flooded forest tend to be associated with high soil fertility (N and C components), in sharp contrast with terra firme forest in Orinoquian gallery forests. Terra firme forests of the amazon basin are more similar in environmental conditions to flooded forests, and are more fertile, than other terra firme forests in the Orinoco basin. However, this analysis did not include all Orinoquian plots in the piedmont, which should show relatively high soil fertility.

0.6 pH Temp Carbon

0.4 Nitrogen 0.2 Min_preci Flooding 0

0.0 Preci Igapo

RDA2 Varzea Terra Firme

-0.2 Clay

Magdalena

-0.4 Orinoco Amazonas -0.6 -0.8 -1.0 -0.5 0.0 0.5

RDA1

Fig. 4: Ordination of species composition and environmental variables of 25 (1-ha) vegetation plots based on Redundancy Analysis. Gray crosses represent single species, squares represent terra firme plots, upward facing triangles represent igapó plots, and downward facing triangles represent várzea plots. Black symbols are plots in the Amazon Basin, red symbols are plots in the Magdalena Basin and blue symbols are plots in the Orinoco Basin. In order to determine the effect of dispersal limitation and environmental similarity on species composition of these forests we used a variance partition analysis, which shows that most of the variance in our data is explained by environmental distance (29%) and the interaction between geographic and

63 environmental distance (13%) (Fig. ). Pure geographic distance explained a small part of the variation (3%).

Fig. 5: Venn diagram showing the variance partition between geographic and environmental distance between 25 (1-ha) vegetation plots in Northern South America. Geographical distance matrix was transformed by an analysis of the Principal Coordinates Neighboring Matrices (PCNM).

Discussion We found that environmental factors are important determinants of floristic similarity in the tree assemblages of north-western South America. In our analysis of variance partitioning 29% of the variation was explained by the selected environmental variables, much higher than the percentage explained by geographical distance alone (3%). In spite of the effort to decouple environmental and geographic distance, by choosing plots in different forest types in the same locality, the association between these factors remained significant. Accordingly, 13% of the variation in floristic composition was associated to both geographic and environmental distance. Overall, the residual variation in this model was large (55%), suggesting that the factors included in this study are unable to explain the majority of the variation. Recent studies have suggested that some soil variables such as phosphorus and calcium contents, which were not included in our study, are strong predictors of floristic composition in Neotropical forests (John et al., 2007; Idárrraga et al., 2016). Therefore, it is possible that these and other soil

64 elements (Gentry, 1988) would increase the predictive power of floristic affinities in our study system. In addition, other environmental variables related to plant withering may also play important roles (Engelbrecht et al., 2007), which may not be taken into account by commonly used rainfall variables. Therefore, we suppose that the inclusion of additional and detailed environmental variables may increase the predictive power of models aiming to explain floristic affinities.

In any case, we found that the forest classifications based on flooding regime and origin of river sediments are strong predictors of the patterns of beta- diversity. Similar findings have emerged in many floristic comparisons involving terra firme and seasonal flooded forests (e.g. Assis et al., 2014; Fortunel et al., 2014; Haugaasen & Peres, 2006; Umaña et al., 2012; Wittmann et al., 2013). Much of the difference in floristic composition may also be attributed to differences in soil types, as well as flooding regimes (Fortunel et al., 2014; Wittmann et al., 2013). For instance, two seasonal flooded plots in our dataset located in várzea and igapó forests and less than 5 km apart, showed a strong floristic differentiation. Interestingly, soil traits were not as dissimilar, suggesting that other factors may be involved in floristic differentiation. Perhaps, the similarity in soil conditions may be caused by sporadic high water river pulses allowing the large Meta river to flood the small black water Picapico creek. However, the explication of the strong turnover in such short distance remains unexplained.

We did not find any improvement at predicting floristic affinities, when using the least-cost path geographic distance caused by the Andean ridges. This result was expected, since the Eastern ridge is believed to have reached its current height only 5-2 million years ago (Gregory-Wodzicki, 2000). Therefore, it is likely that many lowland tree species were present at both sides of the mountain prior to the final uplift (Hoorn et al., 2010), as part of the plant lineages associated to Andean sediments in the piedmont (Quesada et al., 2012).

An additional obstacle to finding highly predictive models of beta-diversity may be related to the still elusive knowledge of plant taxonomy in the rich Northern

65 Andes. There are many species to be described in the humid lowland forest of South America (ter Steege et al., 2013) and it is possible that cryptic species occur (Cavers et al., 2013), even at different sides of the mountains. Therefore, better understanding of the species identity is required to get more resolved estimates of diversity patterns.

In conclusion, we found that the patterns of beta-diversity in the northern South American lowlands are better determined by environmental than by geographic distance, providing support for the hypotheses of niche theory and disregarding a strong role of dispersal limitation. In spite that our analysis of variance partitioning showed that much of the variation remains unexplained, it is clear that soil traits and flooding regime are a strong predictor of tree species composition. In the case of the Eastern Andean ridge we did not find a significant role at affecting plant beta-diversity, probably because of its young age. Additionally, we suggest that better understanding of species identity and the inclusion of other environmental variables should increase the predictive power of the variation in floristic composition, improving the understanding of the patterns of beta-diversity in this region. It is clear that there are large differences in tree species composition between forest types defined by flooding regimes, as well as differences between distinct biogeographic regions. Therefore, the conservation of tree species should take into account these distinctions and accordingly guarantee the protection of each type, including the highly endangered flooded forests (Junk & Piedade, 2005). As fertile soils are facing more anthropogenic pressures, the conservation of plant and animal species restricted to this kind of ecosystems most be prioritized. For instance, the area of national parks in Colombia is quite biased to central Amazonian regions, while the fertile piedmont has less and quite deforested parks (Clark & Aide, 2012), and there are no national parks protecting the lowland forests in the Magdalena Valley.

Acknowledgements We thank the researchers who established vegetation plots: Ángela Cano, Luisa Fernanda Casas, Diego Felipe Correa, Mabel Suescún, María Natalia Umaña, Boris Villanueva; and Luis Francisco Henao for his help carrying the ecological modelling analysis and for additional comments on the manuscript.

66 Dr. Alvaro Duque provided comments on an early version of the manuscript and Sebastian González-Caro helped us suggesting data analyses. We also thank the funding institutions: Fundación para la Promoción de la Investigación y la Tecnología (Banco de la República), Margot Marsh Fund, Primate Conservation Inc., Lincoln Park Zoo, Ecopetrol, Colciencias and Universidad de Los Andes.

67 References Aldana, A.M., Beltrán, M., Torres-Neira, J. & Stevenson, P.R. (2008) Habitat characterization and population density of brown spider monkeys (Ateles hybridus) in Magdalena Valley, Colombia. Neotropical Primates, 15, 46–50.

Assis, R., Wittmann, F., Piedade, M. F. & Haugaasen, T. (2014) Effects of hydroperiod and substrate properties on tree alpha diversity and composition in Amazonian floodplain forests. Plant Ecology, 216 (1), 1-14.

Barreto, S., Duque, A., Cárdenas, D. & Moreno, F. (2010) Floristic variation of canopy tree species at a local scale on tierra firme forest in colombian Amazonia. Acta Amazonica, 40, 179–188.

Bell, G. (2001) Neutral Macroecology. Science, 293, 2413–2418.

Borcard, D. & Legendre, P. (2002) All-scale spatial analysis of ecological data by means of principal coordinates of neighbour matrices. Ecological Modelling, 153, 51–68.

Borcard, D., Legendre, P. & Drapeau, P. (1992) Partialling out the spatial component of ecological variation. Ecology, 73, 1045–1055.

Cano, A. & Stevenson, P.R. (2008) Diversidad y composición florística de tres tipos de bosque en la Estación Biológica Caparú, Vaupés. Colombia Forestal, 12, 63–80.

Cárdenas, S. & Stevenson, P.R. (2012) Patrones florísticos de los planos de inundación y bosques de tierra firme : efectos de filtros ambientales y azar. BSc. Thesis, Departamento de Ciencias Biólogicas. Universidad de los Andes. Bogotá.

Casas-Caro, L.F. & Stevenson, P.R. (2013) Variación de biomasa aérea y densidad de madera en bosques de tierras bajas con planos de inundación de aguas negras y aguas blancas. MSc Thesis, Departamento de Ciencias Biológicas. Universidad de Los Andes.

Cavers, S., Telford, A., Arenal Cruz, F., Pérez Castañeda, A. J., Valencia, R., Navarro, C., Buonamici, A., Lowe, A.J. & Vendramin, G. G. (2013) Cryptic species and phylogeographical structure in the tree Cedrela odorata L. throughout the Neotropics. Journal of Biogeography, 40(4), 732-746.

Chao, A., Chazdon, R. L., Colwell, R. & Shen, T. J. (2005) A new statistical approach for assessing similarity of species composition with incidence and abundance data. Ecology Letters, 8, 148–159.

Clark, M. L., Aide, T. M. & Riner, G. (2012) Land change for all municipalities in Latin America and the Caribbean assessed from 250-m MODIS imagery (2001–2010). Remote Sensing of Environment, 126, 84-103.

Condit, R., Engelbrecht, M. B. J., Pino, D., Pérez, R. & Turner, B. L. (2013) Species distributions in response to individual soil nutrients and seasonal drought across a community of tropical trees. Proceedings of the National Academy of Sciences USA, 110(13), 5064-5068.

Condit, R., Pitman, N., Leigh, E. G., Chave, J., Terborgh, J., Foster, R.B., Núñez, P., Aguilar, S., Valencia, R., Villa, G., Muller-Landau, H. C., Losos, E. & Hubbell, S. P. (2002) Beta- diversity in tropical forest trees. Science, 295 (5555): 666–669.

Correa-Gómez, D.F. & Stevenson, P.R. (2010) Estructura y diversidad de bosques de los llanos orientales colombianos (Reserva Tomo Grande, Vichada). Orinoquia, 14, 31–48.

Dray, S., Legendre, P. & Peres-Neto, P. R. (2006) Spatial modelling: a comprehensive framework for principal coordinate analysis of neighbour matrices (PCNM). Ecological Modelling, 196, 483–493.

Duque, Á., Phillips, J. F., von Hildebrand, P., Posada, C. A., Prieto, C., Rudas, A., Suescún,

68 M. & Stevenson, P. R. (2009) Distance decay of tree species similarity in protected areas on terra firme forests in Colombian Amazonia. Biotropica, 41, 599–607.

Duque, A., Sánchez, M., Cavelier, J. & Duivenvoorden, J. F. (2002) Different floristic patterns of woody understorey and canopy plants in Colombian Amazonia. Journal of Tropical Ecology, 18 (4), 499-525.

Engelbrecht, B. M., Comita, L. S., Condit, R., Kursar, T. A., Tyree, M. T., Turner, B. L., & Hubbell, S. P. (2007) Drought sensitivity shapes species distribution patterns in tropical forests. Nature, 447 (7140), 80-82.

Fortunel, C., Paine, C.E.T., Fine, P.V. a., Kraft, N.J.B. & Baraloto, C. (2014) Environmental factors predict community functional composition in Amazonian forests. Journal of Ecology, 102, 145–155.

Gentry, A. 1988. Changes in Plant Composition on Enviromental and Geographical Gradients. Annals of the Missouri Botanical Garden, 75, 1–34.

Gilbert, B. & Lechowicz, M. J. (2004) Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Sciences USA, 101, 7651–7656.

Global Soil Data Group. (2000) Global gridded surfaces of selected soil characteristics (IGBP- DIS). (International Geosphere-biosph. Programme - Data Information System. Tennessee, U.S.A.

Google Inc, 2001. Google Earth Pro.

Gregory-Wodzicki, K. M. (2000) Uplift history of the Central and Northern Andes: a review. Geological Society of America Bulletin, 112 (7), 1091-1105.

Haugaasen, T. & Peres, C. A. (2006) Floristic, edaphic and structural characteristics of flooded and unflooded forests in the lower Rio Purús region of central Amazonia, Brazil. Acta Amazonica, 36 (1), 25-35.

Hickerson, M. J., Carstens, B. C., Cavender-Bares, J., Crandall, K. A., Graham, C. H., Johnson, J. B., Rissler, L., Victoriano, P. F. & Yoder, A. D. (2010) Phylogeography’s past, present, and future: 10 years after. Molecular Phylogenetics and Evolution, 54 (1), 291-301.

Higgins, M., Ruokolainen, K., Tuomisto, H., Llerena, N. Cardenas, G., Phillips, O. L., Vásquez, R. & Räsänen, M. (2011) Geological control of floristic composition in Amazonian forests. Journal of Biogeography, 38, 2136–2149.

Hijmans, R.J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A. (2005) Very high resolution interpolated climate surfaces for global land areas. International Journal Climatology, 25, 1965–1978

Hijmans, R. J. & van Etten, J. (2014) Raster: Geographic data analysis and modeling. R package version, 2, 15.

Hijmans, R. J., Phillips, S., Leathwick, J. & Elith, J. (2013) Dismo: Species distribution modeling. R package version 0.8-17.

Hoorn, C., Wesselingh, F. P., Ter Steege, H., Bermudez, M. A., Mora, A., Sevink, J. & Jaramillo, C. (2010) Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science, 330 (6006), 927-931.

Hubbell, S.P. (2001) The Unified Neutral Theory of Biodiversity and Biogeography, United States, New Jersey: Princeton University Press.

Hutchinson, G. (1959) Why are There so Many Kinds of Animals? American Naturalist, 93, 145–159.

69 Kenkel, N. C. & Orlóci, L. (1986) Applying metric and nonmetric multidimensional scaling to ecological studies: some new results. Ecology, 67 (4), 919-928.

Idárraga, Á., Duque-Montoya, Á. J. & Feeley, K. (2016) Divergent drivers of tree community composition in lowland and highland forests of the northern tropical Andes, Colombia. Actual Biol, 38 (105), 145-156.

John, R., Dalling, J. W., Harms, K. E., Yavitt, J. B., Stallard, R. F., Mirabello, M., Hubbel, S. P., Valencia, R., Navarrete, H., Vallejo, M. & Foster, R. B. (2007) Soil nutrients influence spatial distributions of tropical tree species. Proceedings of the National Academy of Sciences, 104 (3), 864-869.

Junk W. J. & Piedade M. T. F. (2005) Amazonian wetlands. In: Fraser LH, Keddy PA, editors. Large wetlands: their ecology and conservation. pp. 63-117. Cambridge (UK): Cambridge University Press. Laurance, W.F., Lovejoy, T.E., Vasconcelos, H. L., Bruna, E. M., Didham, R. K. Stouffer, P. C. Gascon, C., Bierregaard, R. O., Laurance, S. G. & Sampaio, E. (2002) Ecosystem decay of amazonian forest fragments : a 22-Year Investigation. Conservation Biolology, 16, 605–618.

Legendre, P. & Anderson, M. J. (1999) Distance-based redundancy analysis: testing multispecies responses in multifactorial ecological experiments. Ecological Monographs, 69 (1), 1-24.

Londoño, A.C. & Alvarez, E. (1997) Composicion floristica de dos bosques (tierra firme y varzea) en la region de Araracuara, Amazonia colombiana. Caldasia, 19, 431–463.

Lópes, W. & Duque, A. J. (2010) Beta diversity in neotropical mountain forests. Caldasia, 32, 175–189.

Nekola, J. C. & White, P. S. (1999) The distance decay of similarity in biogeography and ecology. Journal of Biogeography, 26 (4), 867-878.

Oksanen, J., Blanchet, R., Guillaume Kindt, F., Legendre, P., Minchin, P. R., O’Hara, R. B., Simpson, G. L., Solymos, P., Henry, M., Wagner, S. & Wagner, H. (2011) vegan: Community Ecology Package. R package version 2.0-2.

Phillips, S. J., Anderson, R. P. & Schapire, R. E. (2006) Maximum entropy modeling of species geographic distributions. Ecological Modelling 190, 231–259.

Prance, G. T. (1979) Notes on the vegetation of Amazonia III. The terminology of Amazonian forest types subject to inundation. Brittonia, 31, 26-38.

Quesada, C. A., Phillips, O., Schwarz, M., Czimczik, C.I., Baker, T.R., Patiño, S., Fyllas, N.M., Hodnett, M.G., Herrera, R., Almeida, S., Alvarez Dávila, E., Arneth, A., Arroyo, L., Chao, K.J., Dezzeo, N., Erwin, T., di Fiore, A., Higuchi, N., Honorio Coronado, E., Jimenez, E.M., Killeen, T., Lezama, a. T., Lloyd, G., López-González, G., Luizão, F.J., Malhi, Y., Monteagudo, A., Neill, D. a., Núñez Vargas, P., Paiva, R., Peacock, J., Peñuela, M.C., Peña Cruz, A., Pitman, N., Priante Filho, N., Prieto, A., Ramírez, H., Rudas, A., Salomão, R., Santos, a. J.B., Schmerler, J., Silva, N., Silveira, M., Vásquez, R., Vieira, I., Terborgh, J. & Lloyd, J. (2012) Basin-wide variations in Amazon forest structure and function are mediated by both soils and climate. Biogeosciences, 9, 2203–2246.

Qian, H. & Song, J. S. (2012) Latitudinal gradients of associations between beta and gamma diversity of trees in forest communities in the New World. Journal of Plant Ecology. 1-7. doi: 10.1093/jpe/rts040.

Quiñones-Fernandez, M., Florez-Ayala, C. & Estipiñan-Suarez, L. (2015) Historias contadas por el agua. Frecuencias de inundación., in: Jaramillo-Villa, U., Cortés-Duque, J., Florez-Ayala, C. (Eds.), Colombia Anfibia. Un País de Humedales. pp. 80–81. Instituto de Investigación de recursos biológicos Alexander von Humboldt, Bogotá, D. C.

70 R Development Core Team. (2012) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Rosindell, J., Hubbell, S. P. & Etienne, R. S. (2011) The unified neutral theory of biodiversity and biogeography at age ten. Trends in Ecology and Evolution, 26, 340–8. ter Steege, H., Pitman, N. C., Sabatier, D., Baraloto, C., Salomão, R. P., Guevara, J. E., et al. (2013) Hyperdominance in the Amazonian tree flora Science, 342 (6156), 1243092.

Stevenson, P.R. & Aldana, A.M. (2008) Potential effects of ateline extinction and forest fragmentation on plant diversity and composition in the Western Orinoco Basin, Colombia. International Journal of Primatology, 29, 365–377.

Stevenson, P.R., Suescún, M. & Quiñones, M. (2004) Characterization of forest types at the CIEM, Tinigua Park, Colombia. Field Studies of Fauna and Flora La Macarena Colombia, 14, 1–20.

Svenning, J. C., Kinner, D., Stallard, R. F., Engelbrecht, B. M. J. & Wright, S. J. (2004) Ecological Determinism in plant community structure across a tropical forest landscape. Ecology, 85, 2526–2538.

Tilman, D. & Pacala, S. (1993) Species diversity in ecological communities: The Maintenance of species diversity in plant communities. Chicago Press. Chicago, Unite States.

Tuomisto, H., Ruokolainen, K., Kalliola, R., Linna, A., Danjoy, W. & Rodriguez, Z. (1995) Dissecting Amazonian Biodiversity. Science, 269 (5220), 63–66.

Tuomisto, H., Ruokolainen, K. & Yli-Halla, M. (2003) Dispersal, environment, and floristic variation of western Amazonian forests. Science, 299, 241–244.

Terborgh, J. & Andresen, E. (1998) The composition of Amazonian forests: patterns at local and regional scales. Journal of Tropical Ecology, 14 (05), 645-664.

Umaña, M.N., Norden, N., Cano, A., Stevenson, P.R. 2012. Determinants of plant community assembly in a mosaic of landscape units in central Amazonia: ecological and phylogenetic perspectives. PloS One 7, e45199.

Vieira, S., Alves, L., Aidar, M. & Araújo, L. (2008) Estimation of biomass and carbon stocks: the case of the Atlantic Forest. Biota, 8, 21–29.

Volkov, I., Banavar, J. R., Hubbell, S. P. & Maritan, A. (2003) Neutral theory and relative species abundance in ecology. Nature, 424, 1035–1037.

Wittmann, F., Householder, E., Piedade, M. T., de Assis, R. L., Schöngart, J., Parolin, P. & Junk, W. J. (2013) Habitat specifity, endemism and the neotropical distribution of Amazonian white-water floodplain trees. Ecography, 36 (6), 690-707.

Wolfgang, J. (1997) The Central Amazon Floodplain: Ecology of a Pulsin System. J. J. Wolfgang (Ed.), Germany, Berlin.

71

CAPÍTULO II – DETERMINANTES DE LAS RESERVAS DE CARBONO DE LOS BOSQUES EN EL NORTE DE SURAMÉRICA

72 Resumen

Con el auge de las estrategias de deforestación evitada para la conservación de bosques tropicales, que se basan en el servicio ecosistémico de acumulación de carbono, se hace necesario conocer los factores que controlan las reservas de carbono en éstos ecosistemas. Igualmente, ante futuros escenarios de cambio climático, conocer éstos factores permitirá predecir el efecto de los cambios del clima en las reservas de carbono para un mejor manejo de los bosques. En el estudio que presentamos a continuación se pretendió, primero, cuantificar las reservas de carbono, acumuladas en la biomasa aérea, de diferentes tipos de bosque del norte de Suramérica, y posteriormente, establecer el efecto del clima y otras variables ambientales sobre éstas reservas. Para esto se utilizó a la información de estructura y composición de treintaidos parcelas permanentes de vegetación distribuidas en bosques estacionalmente inundables y de tierra firme. Confirmamos el hecho de que la cantidad de árboles de gran tamaño, es la variable estructural que principalmente determina las reservas de carbono, en la biomasa aérea de los bosques. Otras variables como la inundación y la riqueza de especies también son importantes al momento de estimar las reservas de carbono de los bosques. La evidencia presentada en éste capítulo representa un insumo importante para optimizar el planteamiento de estrategias REDD+ en Colombia.

73

Abstract

With the increasing importance of the strategies of avoided deforeatation for the conservation of tropical forests, that are based on the ecosystem service of carbon storage, it becomes necesary to recognize which factors control the carbon stocks in these ecosystems. Furthermore, amidst future scenarios of climate change, knowing these factors will allow desition makers to predict the effect that these changes will have on the carbon stocks for better management of the forests. In the study we present in this chapter aimed to, first, cuantify the carbon stocks, accumulated in the aboveground biomass, of different forest types in Northern South America, and second, to determine the effect of the climate and other environmental variables on these stocks. For this, we used information on the structure and composition of thirty-two permanent forest plots distributed along seasonally flooded and terra firme forests. We confirmed the fact that the quantity of large trees is the most important structural variable determining carbon stocks in the aboveground biomass of these forests. Other variables, such as flooding and species richness are also important when estimating carbon stocks. The evidence we present in this chapter is a great input for the optimization of the REDD+ strategies in Colombia.

74

DRIVERS OF BIOMASS STOCKS IN NORTHWESTERN SOUTH AMERICAN FORESTS: CONTRIBUTING NEW INFORMATION ON THE NEOTROPICS

Aldana, A. M., Villanueva, B., Cano, Á., Correa, D. F., Umaña, M. N., Casas-Caro, L., Cárdenas Hoyos, S., Henao-Diaz, L. F. & Stevenson, P. R. 2017. Drivers of biomass stocks in Northwestern South American forests: contributing new information on the Neotropics. Forest Ecology and Management 389: 86-95

75

Forest Ecology and Management 389 (2017) 86–95

Contents lists available at ScienceDirect

Forest Ecology and Management

journal homepage: www.elsevier.com/locate/foreco

Drivers of biomass stocks in Northwestern South American forests: Contributing new information on the Neotropics

1 2 3 4 Ana M. Aldana ⇑, Boris Villanueva , Ángela Cano , Diego F. Correa , María Natalia Umaña , Luisa Fernanda Casas 5, Sasha Cárdenas, Luis Francisco Henao-Diaz, Pablo R. Stevenson

Centro de Investigaciones Ecológicas La Macarena – CIEM, Departamento de Ciencias Biológicas, Universidad de Los Andes, Carrera 1 # 18ª – 10, Bogotá, DC., Colombia article info abstract

Article history: The conservation of tropical forests has become an important mechanism for the mitigation of the neg- Received 17 August 2016 ative effects of climate change. Countries located in the Neotropical region, Central and South America, Received in revised form 6 December 2016 are aiming to understand the drivers of carbon stocks of their forests to build a better capacity for forest Accepted 20 December 2016 management. In this study, we calculated Above Ground Biomass – AGB stocks for 32 (1 ha) vegetation plots of forests classified as terra firme and seasonally flooded, and evaluated the effect of basin location, structural and environmental variables on the magnitude of AGB stocks. We report that variation among Keywords: river basins results from the effects of fragmentation and soil fertility. The most important variable, Várzea Igapó determining the magnitude of AGB stocks in lowland forests in the region is the number of large trees Colombia per hectare. Seasonally flooded forests should be studied and managed separately from terra firme forests Orinoco as these behave differently in the relationship between tree species diversity and AGB stocks. We found Magdalena that the proportions of endozoochoric and synzoochoric tree species are important variables for the mag- Amazon nitude of AGB stocks in these forests. We present the first account on the drivers of AGB stocks in Northwestern South America and show that environmental characteristics of forests, such as flooding and fragmentation should be taken into account to determine the variation on AGB stocks among forests in this region. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction account for 23% of the carbon accumulated in forests around the world, thus these are considered as carbon sinks (Meister et al., Avoided deforestation programs for carbon trade markets, such 2012). Within tropical forests, the highest estimates of AGB stocks as the United Nations’ Reduced Emissions from Degradation and are found in the Neotropics (Baccini et al., 2012; Nogueira et al., Deforestation – REDD, are key in the conservation of tropical for- 2015; Saatchi et al., 2011), here defined based on the revision by ests, as it has been proposed for the conservation of the Amazon Morrone (2002): tropical South America, Central America, south- forest (Gullison et al., 2007; Nepstad et al., 2009; Soares et al., central Mexico, the West Indies, and southern Florida. However, 2010). For this reason, it is important for tropical countries to biomass stocks can be highly variable and accurate estimates of investigate and understand the drivers of the carbon cycle in their AGB stocks. ABG estimations based on field measurements are still forests. Aboveground biomass – AGB stocks in tropical forests missing, especially in Northwestern South America, where studies have been overall scarce (Phillips et al., 2016). Sources of variation in biomass stocks in the Neotropics Corresponding author. ⇑ include: forest structural characteristics, local species diversity E-mail address: [email protected] (A.M. Aldana). and environmental conditions. For example, Slik et al. (2013) 1 Present address: Laboratorio de Dendrología, Universidad del Tolima, Ibagué, Colombia. revealed that biomass of moist tropical forests is determined 2 Present address: Plant Systematics and Biodiversity Laboratory, Conservatoire et mostly by the density of large trees (>70 cm DBH), leading to the Jardin botaniques de la Ville de Genève & University of Geneva, 1, Chemin de assumption that forests under some kind of disturbance (logged, l’Impératrice, 1292 Chambésy, Switzerland. flooded or fragmented) would hold less biomass than undisturbed 3 Present address: School of Agriculture and Food Sciences, Centre of Excellence for forests. AGB in undisturbed mature forests across a latitudinal gra- Environmental Decisions CEED, The University of Queensland, Brisbane, Australia. 4 Present address: Department of Biology, University of Maryland, College Park, MD dient in the Neotropics shows a strong positive correlation with 20742, USA. species diversity (Poorter et al., 2015); nevertheless, in the former 5 Present address: Fundación Natura Colombia, Bogotá, Colombia. http://dx.doi.org/10.1016/j.foreco.2016.12.023 0378-1127/Ó 2016 Elsevier B.V. All rights reserved. A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95 87 study, the lack of field measurements from Northwest South Amer- the political boundaries of Colombia, most of which remains unal- ica is evident. tered by human impacts (Gutiérrez et al., 2004). The Magdalena In the Amazon basin AGB stocks are partly determined by river basin extends for an area of 257,000 km2, all of which belongs edaphic and climatic conditions (Hawes et al., 2012; Quesada to Colombia (Rodríguez and Armenteras, 2005), this region has et al., 2012). In this region, environmental filtering in seasonally been the most populated, and thus, the most deforested region in flooded forests shapes the community composition acting in favor the country. During the last four decades large extension of contin- of some functional traits, such as high wood density which has an uous forest have been replaced by vast pastures for cattle ranching effect on AGB (Fortunel et al., 2014). Additionally, recent studies on and small forest patches (Etter et al., 2006a,b). The total extension the effect of defaunation in tropical forests have found, at least for of the Orinoco river basin is 991,587 km2, of which 35% extends the Neotropics, a strong correlation between animal dispersal sys- through Colombia and the remaining 65% through Venezuela tems, based on morphological characteristics of the seeds, and AGB (Romero Ruíz et al., 2004). The western Orinoco river basin is com- stocks: tree species dispersed by animals, which are abundant in posed by a matrix of diverse ecosystems (Romero Ruíz et al., 2004) these forests, tend to hold more biomass than tree species with which include forests fragments of the foothills of the Andes, gal- other dispersal systems, because their wood is usually denser lery forests in a savanna matrix and seasonally flooded forests, (Bello et al., 2015; Osuri et al., 2016; Peres et al., 2016). This is all of the former included in our study system. explained by a trade-off for old-growth shade-tolerant species These plots were established during the period of 2000–2012 with high wood density to have larger seeds and thus relay on ani- for different research purposes (Aldana et al., 2008; Cano and mal dispersal (Rüger et al., 2012; Ter Steege et al., 2006). Stevenson, 2008; Correa-Gómez and Stevenson, 2010; Stevenson Although previous research in the Neotropics on AGB usually et al., 2004; Stevenson and Aldana, 2008; Umaña et al., 2012). lacks information from Northwestern South America, some For this research, plots were classified as terra firme (22) if they researchers have reported estimations of carbon stocks in some were never flooded during a 20-year period. Seasonally flooded of these, particularly for natural forests (Alvarez et al., 2012; forests plots (10) varied in the intensity and frequency of the flood. Phillips et al., 2016, 2011; Sierra et al., 2007). Nevertheless, to date If a plot was classified as flooded, we calculated a flooding index as only few studies have tried to explore which factors explain the the average between flooding intensity (number of months sub- differences in AGB stocks; Alvarez et al. (2016) report that water merged in a year) and flooding frequency (number of times it availability plays a major role in shaping these differences. was detected as flooded in a 20-year period). Each researcher in Armenteras et al., 2015, report that places with high animal diver- the field estimated flooding intensity; flooding frequency was sity do not always hold the highest AGB stocks. established using information provided by Quiñones-Fernandez Through this study we aim to explore the patterns in AGB stocks et al. (2015). We also classified flooded forests as either igapó or for lowland forests of Northern South America. Specifically, we várzea depending on the type of river that inundated them want to evaluate if there is variation associated to biogeographical (Prance, 1989). regions (Magdalena, Orinoco, Amazon basins) and disturbances such as flooding. Regarding geographic regions, we hypothesize that forests of the Amazon basin, which are the least affected by 2.2. Field data collection human presence, should have: a greater proportion of large trees, higher species diversity, higher proportion of animal dispersed We sampled all woody stems with DBH > 10 cm and identified seeds, and thus higher AGB stocks. On the other hand, regarding them to species in the field when possible. When determinations flooding, we hypothesize that seasonally flooded forests should were not possible, we collected voucher specimens to compare have: less species diversity, smaller proportion of large trees due them with reference herbarium collections (ANDES, COL, COAH). to pressures that limit tree size (hydric stress), and a higher pro- In cases were species determination was not possible we assigned portion of species dispersed by water (flooding), and thus AGB them to morphospecies. From a total database of 17,451 stems stocks in these forests should be low compared to continuous terra contained in the plots, we identified 1428 taxonomic units: 1109 firme forests. species and 319 morphospecies. All species were determined by To test our hypothesis, we used information from 32 (1-ha) veg- PRS, to ensure a high level of taxonomic homogeneity within plots. etation plots distributed along Northwestern South America. The In the field, tree height was measured using a laser clinometer dataset includes plots in terra firme and seasonally-flooded forests; following the RAINFOR protocol (Phillips and Baker, 2015), mea- some of the forests in this study have been subject to anthro- suring at least 10 individuals of each of eight DBH categories in pogenic fragmentation and some are naturally fragmented gallery 16 plots, of the Orinoco and Magdalena Basin. This resulted in forests in the Orinoco savannas (as far back as can be established; direct height measurements for 36% stems of the complete dataset; Etter, 2013; Etter et al., 2006a,b). these measurements were used to calculate average stem height Our findings are of great importance to forest managers as they for all the stems in the dataset (32 plots) using the same eight give more insight on the factors that can affect the amount of car- DBH categories. bon accumulated in different forest types of the Neotropics. In some of the sites (within 16 plots and their surrounding for- ests), of terra firme and seasonally flooded forests of the Orinoco and Magdalena basins, we collected wood cores, from 366 tree spe- 2. Materials and methods cies, which were usually the most abundant within each of the plots, to estimate wood density (Casas-Caro et al., 2016) following 2.1. Study system the protocol recommended for CTFS sites (Chave, 2005), wood den- sity was calculated as wood specific gravity, a measurement which A total of thirty-two 1 ha plots were established in lowland for- does not have units, following the recommendations from ests in seven different localities, in the three major river basins in Williamson and Wiemann (2010). Colombia: Amazon, Magdalena and Orinoco (Fig. 1). The Amazon In twenty-five of the plots (17 terra firme and 8 seasonally basin extends for 7,989,004 km2 in South America, it covers territo- flooded), located in the three basins, we collected 25 samples of ries of Bolivia, Brazil, Colombia, Ecuador, French Guyana, Guyana, topsoil (in 20 20 m subplots) and analyzed it to obtain soil tex- Â Peru, Suriname and Venezuela; 6.6% of this area is located within ture (% of clay, sand and silt) and pH. 88 A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95

Fig. 1. Map showing the location of the 32 (1-ha) vegetation plots established in Colombia. Black squares represent sites were terra firme plots were established, 3–4 plots per square; downward facing black triangles represent sites were várzea plots were established, 1–2 plots per triangle; upward facing black triangles represent sites were igapó plots were established, 1–2 plots per triangle. Three major river basins in the region are delimited (black) as well as the political borders of Colombia (gray). Different forest types are shown in green colors; seasonally flooded forests are shown in light blue, the land cover distribution and conventions were adapted from the map of vegetation structures in Colombia (Quiñones et al., 2015).

2.3. Indirect measurements species based on fruit morphology (in the field, through literature search, or from herbarium specimens). The assignment of seed dis- To have an estimate of the degree of fragmentation (human or persal systems was based on a previous study (Correa-Gómez et al., natural), for each plot we estimated forest cover percentage in a 2013; Correa et al., 2015) per basic morphological traits (Table 1). 5 km radius around the plot using satellite images and a measure- Very few individuals in our database were classified as Myrme- ment tool provided by Google Earth Pro. If forest cover was less chocorous (15) and Unassisted (5), thus these two categories were than 40% forest plots were categorized as fragmented. Stems with excluded from the analysis of variance and the resulting plots. DBH > 70 cm were classified as large trees and counted for each plot (Slik et al., 2013). 2.4. Biomass estimations Dispersal modes or systems have been classified in terms of fruit and seed morphology and represent valuable approximations We calculated AGB for each individual tree (15,778 stems; 1312 to the major mechanism of seed dispersal and forest dynamics. taxonomic units) using the type I allometric equation for moist for- Dispersal systems can be classified as animal (endozoochory, syn- ests proposed by Alvarez et al. (2012), this equation uses informa- zoochory, myrmecochory), abiotic (wind and water dispersal), tion on wood density, DBH and tree height. For wood density, we autochory (explosive dehiscence), and unassisted (Correa et al., used the data gathered from our field measurements (Casas-Caro 2015). Although this classification represents a gross approxima- et al., 2016) and extrapolated it to 36% of the tree species in our tion to the main dispersal agents (i.e. do not take into account dataset. Wood density values for tree species which we did not when a fleshy fruited species is predominantly dispersed by birds, measure was retrieved from the information on Neotropical Taxa bats or primates), it avoids differences in dispersal outcomes that data base (Zanne et al., 2009). Unidentified morphospecies were could be caused by the abundance of frugivores in the environment assigned values as genus and family averages calculated from the (Stevenson et al., 2015) and determinations can be made for each Neotropical Taxa data base (Zanne et al., 2009). A total of 62% of A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95 89

Table 1 Main dispersal systems and their main morphological traits (from Correa et al., 2015).

1. Animal dispersal 1.1. Endozoochory Diaspores that can be swallowed by frugivores, usually associated with fruits with fleshy structures (but including also mimetic seeds) 1.2. Synzoochory Non-fleshy or fleshy fruits with a seed width larger than 2 cm, which are not commonly swallowed by large frugivores in the Neotropics, but can be transported externally by scatter-hoarding rodents (Jansen et al., 2012) 1.3. Myrmecochory Diaspores with seed width smaller than 3 mm and associated fleshy structures (i.e. elaiosomes) 2. Abiotic dispersal 2.1. Anemochory Diaspores without fleshy structures and with structures that allow wind dispersal (e.g. expanded wings, kapok or tufts). Very small seeds ( < 1 mm) released from dehiscent capsules, such as dust seeds, are also included here 2.2. Hydrochory Diaspores without fleshy structures, without the characteristics that allow dispersal by wind and with floating abilities 3. Other types 3.1 Explosive dehiscence Non-fleshy fruits that release seeds by explosive dehiscence 3.2 Unassisted Diaspores that did not show any of the previous combinations of characters or any reported dispersal mode

Table 2 ing) would be redundant and could possibly lead to multicollinear- Detail of the equations and the sources of AGB estimation equations used in this ity of the model. study, discriminated by life form. The combined effect of environmental and structural parame- Woody Percentage of Source of the Equation used ters on biomass stocks was established using an equation model life form individuals within our AGB equation with the Structural Equation Modeling-SEM package for R lavaan database used (Rosseel, 2012). For the equation, we established a model in which Trees 90% Alvarez et al. ln(AGB) = 2.261 AGB in each plot is defined by: the number of large trees, the num- À 2 (2012) + 0.933 ln (D H q) ber of stems, species richness and wood specific gravity calculated Palms 7% Goodman et al. 8 genus level Eq. (1) (2013) family level as the community weighted mean (Lavorel et al., 2008). These vari- equation ables, in our model, are defined by the environmental variables: Lianas 1.5% Schnitzer et al. AGB = exp[ 1.484 À flooding index, forest cover percentage, soil physical characteristics (2006) + 2.657 ln(D)] (% clay and pH) and dispersal systems (% Endozoochoric and Syn- Guadua 1.5% Quiroga et al. AGB = 2.66 D0,98 zoochoric stems). This analysis was performed with information (2013) from the 25 of the 32 forest plots for which we had soil samples (8 seasonally flooded and 17 terra firme). our tree wood density estimates were gathered from the literature. Taxonomic units (2%) with no family identification were assigned 3. Results wood density values as the plot average. Biomass estimations for woody life forms that are not trees, were calculated using different We found that the magnitude of AGB stocks differs between allometric equations, looking to minimize errors (Table 2). To esti- regions (river basins) and forest types within our study system. mate total AGB for each plot we summed all estimations from the We report a summary of the main results and variables for all plots individuals and converted the total to Mg per hectare. in Table 3. When examining the differences in AGB stocks among river basins we find that forests in the Orinoco river basin tend to have 2.5. Data analysis lower biomass stocks than the Magdalena and Amazon forests (ANOVA, F = 8.11, p < 0.05). This maybe the result of the large num- Differences in biomass stocks between forest types and river ber of plots categorized as fragmented in this region (Table 3). Bio- basins, as well as differences in wood density between dispersal mass stocks also vary greatly within forests types in Northwestern systems, and differences in average stem diameter and dispersal South America: the forest type with the highest AGB stocks was the systems were calculated using analyses of variance – ANOVA, in várzea forest; this was also the forest type with the greatest varia- the freely available software R (R Development Core Team, tion within. Forest cover around the plots seems to have a negative 2008). Estimations of groups with different means were done with effect on AGB stocks among terra firme forests (Fig. 2). We did not a HSD post hoc test using the package agricolae for R (De evaluate the two effects with a two-way ANOVA because, as can be Mendiburu, 2014). Simple correlations between variables were seen in Table 3 the samples are not balanced within regions and evaluated using a Pearson correlation test. between forest types. We adjusted a linear model in which the AGB of each plot is explained in terms of the percentages of each dispersal system, to evaluate the degree in which the proportion of individuals 3.1. Biomass variation and structural variables belonging to species of different dispersal systems influences ABG stocks. In this analysis, we excluded individuals from species Following the idea proposed by Slik et al. (2013), we explored for which the dispersal system was unknown (11% of our data- the relationship between the number of large trees in each plot base). Precisely the model fit (in R) was: and the calculated ABG, this resulted in a strong positive correla- tion (rp = 0.88, t = 10.13, p < 0.05). We also found that most plots AGB % Anemochry % Endozoochory % Hydrochory with the highest proportion of large trees (DBH > 70 cm) are várzea  þ þ forests. There is no clear trend for the number of large trees and the % Explosive dehiscence % Synzoochory þ þ region that forests belong to or the species richness in each plot % Myrmechochory % Unassisted 1 (Fig. 3). þ þ À We tested the correlation between species richness and AGB with the 1 accounting for the fact that the percentages add to 100, (Poorter et al., 2015) in our database. When all plots (32) are À so adding a constant term (which R always does for linear model- included in the analysis the correlation is weak and not significant 90 A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95

Table 3 Summary of the results of AGB stocks, species richness, number of large trees, forest cover and flooding average per hectare for the regions and forest types included in this study, based on information from thirty-two 1 ha vegetation plots. All variables are presented as average; standard deviation is shown in parenthesis.

Basin Bosque Number of Average AGB Average Number of Average Number of Large Average Forest Average Flooding plots (Mg/ha) Species trees Cover index Amazon Igapo 2 261.71 (23.36) 122 (16.97) 16 (2.82) 1 8.25 (0.35) Terra firme 4 292.52 (53.66) 214.25 (33.68) 17.25 (6.65) 1 0 Magdalena Terra firme 4 269.9 (57.52) 165.5 (18.08) 18.5 (6.60) 0.64 (0.25) 0 Terra firme 3 182.69 (46.94) 88.33 (9.60) 5.33 (1.15) 0.35 (0.02) 0 fragmented Varzea 2 405.84 (83.16) 61.5 (3.53) 39.5 (0.70) 0.47 (0.01) 2.75 (2.47) Orinoco Igapo 3 198.52 (14.19) 37.3 (8.50) 2.66 (1.15) 0.34 (0.10) 8.66 (1.04) Terra firme 6 189.63 (49.03) 135.66 (25.88) 10 (5.58) 0.91 (0.21) 0 Terra firme 5 145.66 (27.15) 68.6 (10.47) 3.4 (2.60) 0.24 (0.06) 0 fragmented Varzea 3 208.54 (69.90) 44 (27.87) 19.33 (11.59) 0.8 (0.33) 2.66 (3.40)

Fig. 2. Scatterplot showing the variation in AGB between and within forest types in Northern South America. Variation was estimated from 32 (1-ha) vegetation plots (ANOVA, F = 3.86, p < 0.05). The thick line represents the median of the group and the thin lines represent the quartiles. Letters above the points represent the groups estimated with a post hoc HSD test. Red dots in the igapó category represent forest plots where forest cover estimates around the plot were lower than 0.4 and thus should also be considered as ‘‘fragmented”, however, these are located in a landscape of the Orinoco basin were forests are naturally fragmented (gallery forests).

Fig. 3. Relationship between above ground biomass stocks and the number of trees with DBH greater than 70 cm, for each of the 32 (1-ha) vegetation plots. Symbol size represents species richness (number of species/stem ratio) for each plot. The black line represents the fit of the data which was estimated as 0.879. A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95 91

Fig. 4. Plot of the correlation between species richness and AGB. This figure shows the relationship for all the 32 plots (rp = 0.34, t = 1.94, p > 0.05).

(Fig. 4); but if seasonally flooded forests (10) are excluded from the (t = 3.02, p < 0.01), anemochory (t = 2.52, p < 0.01), and synzoo- analysis the correlation becomes stronger and significant (rp = 0.74, chory (t = 2.14, p < 0.05). Detailed information on the proportions t = 4.91, p < 0.05). of dispersal systems for each plot can be found in Table A1. Given this result, we decided to explore the relationship of dispersal sys- 2.2. Biomass and dispersal systems tems to AGB variables such as wood density and DBH (Fig. 5). When evaluating the differences in wood specific gravity for dis- We tested the effect of dispersal systems on variation in AGB in persal systems we found that synzoochorous plants showed the an independent analysis (see methods), including all plots. We highest wood density (wood specific gravity) values than other dis- found that the proportion of individuals of endozoochoric species persal systems (ANOVA, F = 9.501, p < 0.001). When examining the is very important in explaining the variation in AGB (t = 4.83, distribution of DBH sizes of trees related to dispersal systems we p < 0.001), followed by the proportion of explosive dehiscence found that anemochorous tree species tend to be larger than trees

Fig. 5. Scatterplots showing the distribution of wood specific gravity values (left) and the distribution of tree DBH (right) for different dispersal systems of Neotropical woody species. Wood density values and DBH measurements were extracted from trees of which we knew the species, and we could assign them to the different dispersal systems in our database of 32 (1-ha) vegetation plots in lowland forests of Northern South America. The thick lines represent the median of the groups and the thin lines represent the quartiles. Letters above the dots represent the groups estimated with post hoc HSD tests. 92 A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95

Fig. 6. Path analysis showing the relationship between AGB and structural and ecological variables for 25 (1-ha) vegetation plots in seasonally flooded (8) and terra firme (17) forests in Colombia. Arrow size is proportional to the magnitude of the relationship, which is expressed to the right of each arrow, significance of the regressions and covariance are: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’. Single head arrows represent regressions and double headed arrows represent covariance. Abbreviations of variables are: Flood (estimated value of flood index based on flood intensity and frequency, see methods), % End (% of endozoochoric individuals in each plot), % Syn (% of Synzoochoric individuals in each plot), Stem70 (number of large trees in each plot), Stems (total number of stems in each plot), spp (species richness for each plot), WD (Community Weighted Mean of Wood Specific Gravity for each plot), AGB (estimated total Above Ground Biomass stock for each plot in Mg per hectare). classified as other dispersal systems (ANOVA, F = 6.705, p < 0.001). 4. Discussion The scatterplot allows for the observation that most of the stems in these forests belong to endozoochous species and that most belong In general, our hypotheses are not supported by the analyses: to small DBH classes. forests of the Amazon basin do not have the highest AGB stocks, regardless of high species diversity and a higher proportion of ani- mal dispersed species. The number of large trees is the most 2.3. Biomass variation with structural and environmental variables important variable explaining AGB stocks in our data set. Season- ally flooded forests (várzea) of the Magdalena basin have the high- When we tested our model for AGB stocks including 25 of the est numbers of large trees, and thus the highest AGB stocks. This plots for which we had soil analysis, we found that the number result is likely explained by high nutrient content in the soils, as of large trees is still the most important variable explaining the shown by the path analysis, which included soil variables linked variation in AGB stocks (Fig. 6). In turn, pH, soil texture (% clay) with soil fertility (pH and % clay). The Magdalena river has great and forest cover are important variables explaining the number sediment loads from the highly fertile soils of the Andean moun- of large trees in a plot. Stem density and wood specific gravity tains and these are accumulated in the flood plains (Junk and are also correlated to AGB stocks on a lesser extent. Furch, 1993), a condition that favors plant growth (Baribault Contrary to what we found with the previous analysis, in this et al., 2012). model species richness, does not seem to affect AGB. However, for- Our hypothesis concerning flooding was only partially sup- est cover, as expected, is strongly correlated to species richness, ported by seasonally flooded forests classified as igapó of the Ori- another variable affecting species richness is the flooding index. noco basin. Igapo forests have the lowest AGB stocks across all Given the importance of dispersal systems in the previous anal- the sites, likely due to a smaller number of large trees. This result ysis, we included the proportion of endozoochoric and synzoo- is not surprising, as most of the forests in the region are riparian or choric species within this model, expecting those variables to gallery forests of short stature (Correa-Gómez and Stevenson, influence directly on AGB and wood density. Nevertheless, the 2010; Gónzalez et al., 2016). This result could also be explained effect of the dispersal systems directly on AGB stocks was diluted. by low nutrient content on the soils, as the soils of the Orinoco We did find that the proportion of trees of synzoochoric species has basin tend to be nutrient poor (Junk and Furch, 1993). Due to the some effect on the community weighted mean for wood density, environmental stress from the flood, we also expected flooded for- confirming our results from the previous analysis on dispersal sys- ests to be species poor and to have a higher proportion of seeds dis- tems and wood density (Fig. 5). persed by water, these predictions were met; however, those The path analysis shows that the proportion of endozoochorous variables do not seem to be related with AGB stocks. stems has a negative covariation with flooding index and forest Our results on AGB stocks seem to be higher than the estima- cover. Covariance with the flooding index (i.e. higher proportion tions calculated previously in the framework of the assessment of endozoochorous stems in forest with shorter flooding period) of AGB stocks for natural forests in Colombia (Phillips et al., can be explained by the higher importance of other dispersal sys- 2016, 2011). We believe this can be attributed to the fact that we tems within seasonally flooded forests such as hydrochory, which considered palms and lianas as such, for allometries, while previ- is higher in these forests than in terra firme forests (ANOVA, ous studies have used only tree allometries for all growth forms. F = 3.17, p = 0.039). The negative covariance with the percentage Discriminating between forests types, as we have done, evi- of forest cover can be explained as an effect of the inclusion of nat- dences great differences in the amount of AGB stored in Neotropi- urally fragmented forests as the gallery forests of the Orinoco cal forests. In fact, even within forest types there are still great basin. differences. Nonetheless, there is a tendency for seasonally flooded A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95 93 forests to behave differently between them: igapó forests hold less AGB is calculated using equations, that implicitly have a an associ- AGB stocks and show less variation than váreza forests; frag- ation environmental characteristics (controlling for forest type, e.g. mented and riparian forests tend to hold the smaller amounts of Alvarez et al., 2012; Chave et al., 2005), then, a direct relationship AGB stocks. This result may be coupled with the differences with rainfall and other environmental factors should be expected. between geographical regions (see results), as many of the plots Our model analysis gives a stronger role to soil fertility, here studied in the Orinoco are found in fragmented (natural and measured indirectly from pH and % clay, forest plots with higher human) landscapes. fertility harbor higher number of large trees (DBH > 70 cm) than Species richness is an important factor in AGB stocks, but only low-nutrient plots. Then, soil fertility tends to be the most deter- for terra firme forests. Forests holding high species diversity tend minant factor in AGB content of different tropical forest types in to have high AGB stocks. However, seasonally flooded forests tend the Neotropics. to have lower species diversity but high AGB stocks. This result In our model, in contrast to what was reported by Poorter et al. stresses the need for differentiated management of flooded and (2015), we found that species diversity does not play an important terra firme, as they are structurally and ecologically different role in determining AGB stocks. The importance of species diver- (Gónzalez et al., 2016; Hawes et al., 2012). sity is dissolved when environmental variables, including season- ally flooded forests, are considered. However, this does not mean 4.1. Biomass and dispersal systems that species diversity is not important, as we showed in our regres- sion including only terra firme forests, species richness is very well Endozoochoric and synzoochoric tree species are the most correlated to AGB stocks. important dispersal systems in the maintenance of AGB stocks in We understand the limitations of our path analysis, which was our study system. The proportion of endozoochoric species influ- carried out with information from 25 plots, rather than the 32 of ences the magnitude of AGB stocks: the greater proportion of our database, however, this analysis highlighted the importance endozoochorus species, the greater the AGB stocks in a forest. This of environmental variables such as forest cover and soil fertility. result shows the importance of the conservation of large mammals We also included variables such as flooding and dispersal systems, and birds for the maintenance of the populations of these tree spe- which do not seem to influence AGB stocks but appear to have cies (Bufalo et al., 2016; Peres and van Roosmalen, 2002; Stevenson some importance in the ecology of these forests. et al., 2005). On the other hand, syzoochoric species tend to have denser woods than species with other dispersal systems. This fact 4. Conclusions influences the carbon stocks of Neotropical forests giving impor- tance to the conservation of medium-sized rodents and other We found that AGB stocks in lowland forests of Northwestern mammals such as large bats. Along with the conservation of forest South America depend mainly on the number of large trees present cover, these vectors can aid on the maintenance of the populations in the forest; this variable is correlated to soil fertility. Seasonally of the species they disperse. Although the role of other agents such flooded forests in this region behave very different than terra firme as birds, primates, tapirs and bats is well recognized (Howe, 2016), forests; AGB stocks of the later are negatively affected by fragmen- our data emphasize on the conservation of vectors that move seeds tation and loss of species richness. Zoochorous tree species are externally (such as medium sized rodents). To our knowledge, this important in the maintenance of AGB stocks in this system; thus, is the first time that the relevance of these two zoochorous disper- the populations of seed dispersers should be protected. sal systems for carbon stocks is evaluated at the same time. Our Our study complements previous findings, as we are publishing results agree with previous studies which highlight the need to information from 32 forest plots established Colombia, which has have conservation efforts to preserve the fauna in the forests were almost always been a missing point in Neotropical studies. carbon has become the main conservation tool, because defau- nated forests may, in the long future, reduce the magnitude of their carbon storage capacity (Bello et al., 2015; Peres et al., 2016). Acknowledgments

4.2. Biomass variation with structural and environmental variables We would like to thank all researchers and friends involved in the establishment of the plots. Pablo Negret calculated plot forest Poorter et al. (2015) reported that AGB in a system of Neotrop- cover. Adolfo Quiroz and Alf Onshuus helped in the development ical forests, is driven by rainfall, average stem diameter and species of the model to correlate AGB and dispersal systems. Two anony- richness. We agree that rainfall may have a direct effect on plant mous reviewers helped in the improvement of a previous version growth, as well as a positive indirect effect on forest cover and a of the manuscript. negative one on soil fertility. However, evaluating a relationship between AGB and rainfall might be misleading, as AGB is currently Appendix A calculated using rainfall estimates (Chave et al., 2014). And, when

Table A1 Detailed information on forest type, region and proportion of stems belonging to different dispersal systems in 32 (1-ha) lowland forests of Northern South America.

Plot_Code Forest Type Region % % % Explosive % % % % Anemochory Endozoochory Dehiscence Hydrochory Myrmechochory Synzoochory Unassisted CAPA_PI_1 Igapo Amazon 22.7 32.6 6.9 18.5 0.0 19.1 0.2 CAPA_PI_2 Igapo Amazon 7.6 36.7 8.9 15.3 0.0 31.6 0.0 CASA_PII_2 Igapo Orinoco 2.8 69.7 7.0 2.3 0.7 17.5 0.0 TOMO_PI_4 Igapo Orinoco 5.5 73.3 7.5 2.2 0.2 11.4 0.0 TOMO_PI_5 Igapo Orinoco 6.0 66.8 12.0 2.5 3.5 9.3 0.0 CAPA_TFC_1 TerraFirme Amazon 3.9 54.6 21.1 2.2 2.6 15.6 0.0 CAPA_TFC_2 TerraFirme Amazon 17.5 62.1 7.8 9.7 0.0 2.9 0.0 CAPA_TFT_1 TerraFirme Amazon 0.2 38.9 55.6 4.4 0.0 0.9 0.0

(continued on next page) 94 A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95

Table A1 (continued)

Plot_Code Forest Type Region % % % Explosive % % % % Anemochory Endozoochory Dehiscence Hydrochory Myrmechochory Synzoochory Unassisted CAPA_TFT_2 TerraFirme Amazon 13.0 75.1 1.3 0.0 0.0 10.6 0.0 QUIN_TF_1 TerraFirme Magdalena 9.3 74.0 1.6 0.5 0.7 13.3 0.5 QUIN_TF_3 TerraFirme Magdalena 8.6 76.2 1.7 1.0 0.0 12.5 0.0 QUIN_TF_4 TerraFirme Magdalena 11.6 77.1 1.5 0.0 0.2 9.7 0.0 QUIN_TF_5 TerraFirme Magdalena 13.4 78.5 3.9 1.3 0.0 2.9 0.0 SAMA_TF_3 TerraFirme Orinoco 23.9 67.3 0.0 0.9 0.0 8.0 0.0 TINI_TF_1 TerraFirme Orinoco 7.7 72.6 5.8 9.1 0.0 4.8 0.0 TINI_TF_3 TerraFirme Orinoco 2.5 63.2 2.0 26.9 0.0 5.5 0.0 TINI_TF_4 TerraFirme Orinoco 7.6 58.4 16.6 10.5 0.2 6.6 0.0 TINI_TF_6 TerraFirme Orinoco 5.1 94.5 0.0 0.0 0.0 0.4 0.0 TINI_TF_7 TerraFirme Orinoco 8.2 91.7 0.0 0.0 0.0 0.2 0.0 SAJU_TF_3 TerraFirme_Frag Magdalena 9.7 87.9 0.2 0.0 0.0 2.2 0.0 SAJU_TF_4 TerraFirme_Frag Magdalena 36.0 61.0 0.0 2.8 0.0 0.3 0.0 SAJU_TF_5 TerraFirme_Frag Magdalena 6.7 84.1 0.7 0.4 0.0 6.3 1.8 SAMA_TF_1 TerraFirme_Frag Orinoco 3.3 91.7 0.3 0.0 0.0 4.7 0.0 SAMA_TF_2 TerraFirme_Frag Orinoco 8.4 61.1 0.5 0.3 0.0 4.1 25.6 TOMO_TF_1 TerraFirme_Frag Orinoco 5.9 87.8 2.8 0.0 0.0 3.5 0.0 TOMO_TF_2 TerraFirme_Frag Orinoco 4.6 80.1 10.6 1.0 0.0 3.7 0.0 TOMO_TF_3 TerraFirme_Frag Orinoco 3.9 78.5 1.9 0.4 0.3 5.7 9.3 CASA_PIV_1 Varzea Orinoco 22.1 52.2 18.3 1.2 0.0 6.2 0.0 SAJU_PI_1 Varzea Magdalena 12.9 56.8 21.6 1.3 0.0 7.5 0.0 SAJU_PI_2 Varzea Magdalena 25.2 62.7 0.2 0.0 1.2 10.6 0.0 TINI_PI_2 Varzea Orinoco 11.3 69.5 1.5 0.0 2.7 15.0 0.0 TINI_PI_5 Varzea Orinoco 7.1 77.3 2.4 0.0 3.3 9.9 0.0

References Correa-Gómez, D., Stevenson, P., Älvarez, E., Aldana, A., Umaña, M., Cano, A., Adarve, J., Benítez, D., Castaño, A., Cogollo, Á., Devia, W., Fernández, F., García, L., Melo, O., Peñuela, M., Restrepo, Z., Serna, M., Velásquez, O., Velásquez, C., Von Aldana, A.M., Beltrán, M., Torres-Neira, J., Stevenson, P.R., 2008. Habitat Hildbrand, P., 2013. Patrones De Frecuencia Y Abundancia De Sistemas De characterization and population density of brown spider monkeys (Ateles Dispersión De Plantas En Bosques Colombianos Y Su Relación Con Las Regiones hybridus) in Magdalena Valley, Colombia. Neotropical Primates 15, 46–50. Geográficas Del País. Colomb. Forestal 16, 33–51. http://dx.doi.org/10.1896/044.015.0203. Correa-Gómez, D.F., Stevenson, P.R., 2010. Estructura y diversidad de bosques de los Alvarez, E., Cayuela, L., Gonzalez-Caro, S., Aldana, A.M., Stevenson, P.R., Phillips, O.L., llanos orientales colombianos (Reserva Tomo Grande, Vichada). Orinoquia 14, Von Hildebrand, P., Jiménez, E., Melo, O., Mendoza, I., Restrepo, Z., Velásquez, O., 31–48. Rey-Benayas, J.M., 2016. Forest Biomass Density across Large Climate Gradients Correa, D.F., Álvarez, E., Stevenson, P.R., 2015. Plant dispersal systems in Neotropical in Northern South America is related to Water Availability but not with forests: availability of dispersal agents or availability of resources for Temperature. Manuscript submitted for publication. constructing zoochorous fruits? Glob. Ecol. Biogeogr. 24, 203–214. http://dx. Alvarez, E., Duque, A., Saldarriaga, J., Cabrera, K., de las Salas, G., del Valle, I., Lema, doi.org/10.1111/geb.12248. A., Moreno, F., Orrego, S., Rodríguez, L., 2012. Tree above-ground biomass De Mendiburu, F., 2014. Agricolae: statistical procedures for agricultural research. R allometries for carbon stocks estimation in the natural forests of Colombia. For. package version 1, 1–6. Ecol. Manage. 267, 297–308. http://dx.doi.org/10.1016/j.foreco.2011.12.013. Etter, A., 2013. La transformación del uso de la tierra y los ecosistemas durante el Armenteras, D., Rodríguez, N., Retana, J., 2015. National and regional relationships periodo colonial en Colombia (1500–1800). La Economía Colonial de La Nueva of carbon storage and tropical biodiversity. Biol. Conserv. 192, 378–386. http:// Granada, 1–44. dx.doi.org/10.1016/j.biocon.2015.10.014. Etter, A., McAlpine, C., Pullar, D., Possingham, H., 2006a. Modelling the conversion of Baccini, A., Goetz, S.J., Walker, W.S., Laporte, N.T., Sun, M., Sulla-Menashe, D., Colombian lowland ecosystems since 1940: drivers, patterns and rates. J. Hackler, J., Beck, P.S.A., Dubayah, R., Friedl, M.A., Samanta, S., Houghton, R.A., Environ. Manage. 79, 74–87. http://dx.doi.org/10.1016/j.jenvman.2005.05.017. 2012. Estimated carbon dioxide emissions from tropical deforestation improved Etter, A., McAlpine, C., Wilson, K., Phinn, S., Possingham, H., 2006b. Regional by carbon-density maps. Nat. Clim. Change 2, 182–185. http://dx.doi.org/ patterns of agricultural land use and deforestation in Colombia. Agric. Ecosyst. 10.1038/nclimate1354. Environ. 114, 369–386. http://dx.doi.org/10.1016/j.agee.2005.11.013. Baribault, T.W., Kobe, R.K., Finley, A.O., 2012. Tropical tree growth is correlated with Fortunel, C., Paine, C.E.T., Fine, P.V.A., Kraft, N.J.B, Baraloto, C., 2014. Environmental soil phosphorus, potassium, and calcium, though not for legumes. Ecol. Monogr. factors predict community functional composition in Amazonian forests. J. Ecol. 82, 189–203. http://dx.doi.org/10.1890/11-1013.1. 102, 145–155. http://dx.doi.org/10.1111/1365-2745.12160. Bello, C., Galetti, M., Pizo, M.A., Magnago, L.F.S., Rocha, M.F., Lima, R.A.F., Peres, C.A., Gónzalez, J.S., Aldana, A.M., Correa, D., Casas, L., Stevenson, P.R., 2016. Dinámica, Ovaskainen, O., Jordano, P., 2015. Defaunation affects carbon storage in tropical Estructura Y Diversidad De Los Bosques De Galería En La Región De Los Llanos. forests. Sci. Adv. 1, 1–11. http://dx.doi.org/10.1126/sciadv.1501105. Universidad de Los Andes, Colombia. Bufalo, F.S., Galetti, M., Culot, L., 2016. Seed dispersal by primates and implications Goodman, R.C., Phillips, O.L., del Castillo Torres, D., Freitas, L., Cortese, S.T., for the conservation of a biodiversity hotspot, the Atlantic forest of South Monteagudo, A., Baker, T.R., 2013. Amazon palm biomass and allometry. For. America. Int. J. Primatol. 37, 333–349. http://dx.doi.org/10.1007/s10764-016- Ecol. Manage. 310, 994–1004. http://dx.doi.org/10.1016/j.foreco.2013.09.045. 9903-3. Gullison, R.E., Frumhoff, P.C., Canadell, J.G., Field, C.B., Nepstad, D.C., Hayhoe, K., Cano, A., Stevenson, P.R., 2008. Diversidad y composición florística de tres tipos de Avissar, R., Curran, L.M., Friedlingstein, P., Jones, C.D., Nobre, C., 2007. Tropical bosque en la Estación Biológica Caparú,Vaupés. Colomb. Forestal 12, 63. http:// forests and climate policy. Science 316, 1136163–1136986. http://dx.doi.org/ dx.doi.org/10.14483/udistrital.jour.colomb.for.2009.1.a06. 10.1126/science.1136163. Casas-Caro, L., Aldana, A.M., Henao-Diaz, F., Villanueva, B., Stevenson, P.R., 2016. Gutiérrez, F., Acosta, L.E., Salazar, C.A., 2004. Perfiles Urbanos en la Amazonia Specific gravity of woody tissue from lowland Neotropical plants: variance colombiana: Un enfoque para el desarrollo sostenible. among forest types. Data paper submitted for publication. Hawes, J.E., Peres, C.A, Riley, L.B., Hess, L.L., 2012. Landscape-scale variation in Chave, J., 2005. Measuring wood density for tropical forest trees A field manual for structure and biomass of Amazonian seasonally flooded and unflooded forests. the CTFS sites. For. Ecol. Manage. 281, 163–176. http://dx.doi.org/10.1016/ Chave, J., Andalo, C., Brown, S., Cairns, M.A., Chambers, J.Q., Eamus, D., Fölster, H., j.foreco.2012.06.023. Fromard, F., Higuchi, N., Kira, T., Lescure, J.-P., Nelson, B.W., Ogawa, H., Puig, H., Howe, H.F., 2016. Making dispersal syndromes and networks useful in tropical Riéra, B., Yamakura, T., 2005. Tree allometry and improved estimation of carbon conservation and restoration. Glob. Ecol. Conserv. 6, 152–178. http://dx.doi.org/ stocks and balance in tropical forests. Oecologia 145, 87–99. http://dx.doi.org/ 10.1016/j.gecco.2016.03.002. 10.1007/s00442-005-0100-x. Jansen, P.A., Hirsch, B.T., Emsens, W.J., Zamora-Gutierrez, V., Wikelski, M., Kays, R., Chave, J., Réjou-Méchain, M., Búrquez, A., Chidumayo, E., Colgan, M.S., Delitti, W.B. 2012. Thieving rodents as substitute dispersers of megafaunal seeds. Proc. Nat. C., Duque, A., Eid, T., Fearnside, P.M., Goodman, R.C., Henry, M., Martínez-Yrízar, Acad. Sci. 109 (31), 12610–12615. http://dx.doi.org/10.1073/pnas.1205184109. A., Mugasha, W.A., Muller-Landau, H.C., Mencuccini, M., Nelson, B.W., Junk, W.J., Furch, K., 1993. A general review of tropical South American floodplains. Ngomanda, A., Nogueira, E.M., Ortiz-Malavassi, E., Pélissier, R., Ploton, P., Wetlands Ecol. Manage. 2, 231–238. http://dx.doi.org/10.1007/bf00188157. Ryan, C.M., Saldarriaga, J.G., Vieilledent, G., 2014. Improved allometric models Lavorel, S., Grigulis, K., McIntyre, S., Williams, N.S.G., Garden, D., Dorrough, J., to estimate the aboveground biomass of tropical trees. Glob. Change Biol. 3177– Berman, S., Quétier, F., Thébault, A., Bonis, A., 2008. Assessing functional 3190. http://dx.doi.org/10.1111/gcb.12629. A.M. Aldana et al. / Forest Ecology and Management 389 (2017) 86–95 95

diversity in the field – methodology matters! Funct. Ecol. 22, 134–147. http:// R Development Core Team, R.F.F.S.C., 2008. R: A Language and Environment for dx.doi.org/10.1111/j.1365-2435.2007.01339.x. Statistical Computing. Vienna Austria R Foundation for Statistical Computing. Meister, K., Ashton, M.S., Craven, D., Griscom, H., 2012. Carbon dynamics of tropical Rodríguez, N., Armenteras, D., 2005. Ecosistemas naturales de la cuenca del rio forests. In: Ashton, M.S., Tyrrell, M.L., Spalding, D., Gentry, B. (Eds.), Managing Magdalena. Los sedimentos del río Magdalena: reflejo de la crisis ambiental, Forest Carbon in a Changing Climate. Springer, Netherlands, Dordrecht, pp. 51– 79–98. 75. http://dx.doi.org/10.1007/978-94-007-2232-3_4. Romero Ruíz, M., Galindo, G., Otero, J., Armenteras, D., 2004. Ecosistemas De La Morrone, J.J., 2002. Biogeographical regions under track and cladistic scrutiny. J. Cuenca Del Orinoco Colombiano. Instituto de Investigación de Recursos Biogeogr. 29, 149–152. http://dx.doi.org/10.1046/j.1365-2699.2002.00662.x. Biológicos Alexander von Humboldt, Bogotá, D. C.. Nepstad, D., Soares-Filho, B.S., Merry, F., Lima, A., Moutinho, P., Carter, J., Bowman, Rosseel, Y., 2012. Lavaan: an R package for structural equation modeling. J. Stat. M., Cattaneo, A., Rodrigues, H., Schwartzman, S., McGrath, D.G., Stickler, C.M., Softw. 48, 1–36. http://dx.doi.org/10.18637/jss.v048.i02. Lubowski, R., Piris-Cabezas, P., Rivero, S., Alencar, A., Almeida, O., Stella, O., Rüger, N., Wirth, C., Wright, S.J., Condit, R., 2012. Functional traits explain light and 2009. The end of deforestation in the Brazilian Amazon. Science 326, 1350– size response of growth rates in tropical tree species. Ecology 93, 2626–2636. 1351. http://dx.doi.org/10.1126/science.1182108. http://dx.doi.org/10.1890/12-0622.1. Nogueira, E.M., Yanai, A.M., Fonseca, F.O.R., Fearnside, P.M., 2015. Carbon stock loss Saatchi, S.S., Harris, N.L., Brown, S., Lefsky, M., Mitchard, E.T.A., Salas, W., Zutta, B.R., from deforestation through 2013 in Brazilian Amazonia. Glob. Change Biol. 21, Buermann, W., Lewis, S.L., Hagen, S., Petrova, S., White, L., Silman, M., Morel, A., 1271–1292. http://dx.doi.org/10.1111/gcb.12798. 2011. Benchmark map of forest carbon stocks in tropical regions across three Osuri, A.M., Ratnam, J., Varma, V., Alvarez-Loayza, P., Hurtado Astaiza, J., Bradford, continents. Proc. Natl. Acad. Sci. U.S.A. 108, 9899–9904. http://dx.doi.org/ M., Fletcher, C., Ndoundou-Hockemba, M., Jansen, P.A., Kenfack, D., Marshall, A. 10.1073/pnas.1019576108. R., Ramesh, B.R., Rovero, F., Sankaran, M., 2016. Contrasting effects of Schnitzer, S.A., DeWalt, S.J., Chave, J., 2006. Censusing and measuring lianas: a defaunation on aboveground carbon storage across the global tropics. Nat. quantitative comparison of the common methods. Biotropica 38, 581–591. Commun. 7, 11351. http://dx.doi.org/10.1038/ncomms11351. http://dx.doi.org/10.1111/j.1744-7429.2006.00187.x. Peres, C.A., Emilio, T., Schietti, J., Desmoulière, S.J.M., Levi, T., 2016. Dispersal Sierra, C.A., del Valle, J.I., Orrego, S.A., Moreno, F.H., Harmon, M.E., Zapata, M., limitation induces long-term biomass collapse in overhunted Amazonian Colorado, G.J., Herrera, M.A., Lara, W., Restrepo, D.E., Berrouet, L.M., Loaiza, L.M., forests. Proc. Natl. Acad. Sci. 113, 892–897. http://dx.doi.org/10.1073/ Benjumea, J.F., 2007. Total carbon stocks in a tropical forest landscape of the pnas.1516525113. Porce region, Colombia. For. Ecol. Manage. 243, 299–309. http://dx.doi.org/ Phillips, J.F., Duque, A., Cabrera, K., Navarrete, D.A., Garcia, M.C., Alvarez, E., Cabrera, 10.1016/j.foreco.2007.03.026. E., Cárdenas, D., Galindo, G., Ordóñez, M.F., Rodríguez, M.L., Vargas, D.M., 2011. Slik, J.W.F., Paoli, G., Mcguire, K., Amaral, I., Barroso, J., Bastian, M., Blanc, L., Bongers, Estimación de las Reservas Potenciales de Carbono Almacenadas en la Biomasa F., Boundja, P., Clark, C., Collins, M., Dauby, G., Ding, Y., Doucet, J.L., Eler, E., Aérea en Bosques Naturales de Colombia. Instituto de Hidrología, Meteorología, Ferreira, L., Forshed, O., Fredriksson, G., Gillet, J.F., Harris, D., Leal, M., Laumonier, y Estudios Ambientales-IDEAM, Bogotá. Y., Malhi, Y., Mansor, A., Martin, E., Miyamoto, K., Araujo-Murakami, A., Phillips, J.F., Duque, Á., Scott, C., Wayson, C., Galindo, G., Cabrera, E., Chave, J., Peña, Nagamasu, H., Nilus, R., Nurtjahya, E., Oliveira, Á., Onrizal, O., Parada- M., Álvarez, E., Cárdenas, D., Duivenvoorden, J., Hildebrand, P., Stevenson, P., Gutierrez, A., Permana, A., Poorter, L., Poulsen, J., Ramirez-Angulo, H., Reitsma, Ramírez, S., Yepes, A., 2016. Live aboveground carbon stocks in natural forests of J., Rovero, F., Rozak, A., Sheil, D., Silva-Espejo, J., Silveira, M., Spironelo, W., ter Colombia. For. Ecol. Manage. 374, 119–128. http://dx.doi.org/10.1016/ Steege, H., Stevart, T., Navarro-Aguilar, G.E., Sunderland, T., Suzuki, E., Tang, J., j.foreco.2016.05.009. Theilade, I., van der Heijden, G., van Valkenburg, J., Van Do, T., Vilanova, E., Vos, Phillips, O., Baker, T., 2015. Field Manual for Plot Establishment and V., Wich, S., Wöll, H., Yoneda, T., Zang, R., Zhang, M.G., Zweifel, N., 2013. Large Remeasurement – RAINFOR. trees drive forest aboveground biomass variation in moist lowland forests Poorter, L., van der sande, T., Thompson, J., Arets, E.J.M.M., Alarcón, A., Álvarez- across the tropics. Glob. Ecol. Biogeogr. 22, 1261–1271. http://dx.doi.org/ Sánchez, J., Ascarrunz, N., Balvanera, P., Barajas-Guzmán, G., Boit, A., Bongers, F., 10.1111/geb.12092. Carvalho, F.A., Casanoves, F., Cornejo-Tenorio, G., Costa, F.R.C., de Castilho, C.V., Soares, B., Moutinho, P., Nepstad, D., Anderson, A., Rodrigues, H., Garcia, R., Dietzsch, Duivenvoorden, J.F., Dutrieux, L.P., Enquist, B.J., Fernández-Méndez, F., Finegan, L., Merry, F., Bowman, M., Hissa, L., Silvestrini, R., Maretti, C., 2010. Role of B., Gormley, L.H.L., Healey, J.R., Hoosbeek, M.R., Ibarra-Marínquez, G., Junqueira, Brazilian Amazon protected areas in climate change mitigation. Proc. Natl. Acad. A.B., Levis, C., Licona, J.C., Lisboa, L.S., Magnusson, W.E., Martínez-Ramos, M., Sci. U.S.A. 107, 10821–10826. http://dx.doi.org/10.1073/pnas.0913048107. Martínez-Yrizar, A., Martorano, L.G., Masskell, L.C., Mazzei, L., Meave, J.A., Mora, Stevenson, P.R., Aldana, A.M., 2008. Potential effects of ateline extinction and forest F., Muñoz, R., Nytch, C., Pansonato, M.P., Parr, T.W., Paz, H., Simoes Penello, M., fragmentation on plant diversity and composition in the Western Orinoco Pérez-Garcia, E.A., Rentería, L.Y., Rodríguez-Velazquez, J., Rosendaal, D.M.A., Basin, Colombia. Int. J. Primatol. 29, 365–377. http://dx.doi.org/10.1007/ Ruschel, A.R., Sakschewski, B., Salgado Negret, B., Schietti, J., Sinclair, F.L., Souza, s10764-007-9177-x. P.F., Souza, F.C., Stropp, J., ter Steege, H., Swenson, N.G., Thonicke, K., Toledo, M., Stevenson, P.R., Link, A., González-Caro, S., Torres-Jiménez, M.F., 2015. Frugivory in Uriarte, M., van der Hout, P., Walker, P., Zamora, N., Peña-Claros, M., 2015. canopy plants in a western amazonian forest: dispersal systems, phylogenetic Diversity enhances carbon storage in tropical forests. Glob. Ecol. Biogeogr. 24 ensembles and keystone plants. PLoS ONE 10, e0140751. http://dx.doi.org/ (11), 1314–1328. http://dx.doi.org/10.1111/geb.12364. 10.1371/journal.pone.0140751. Prance, G., 1989. American tropical forests. Tropical rain forest ecosystems- Stevenson, P.R., Pineda, M., Samper, T., 2005. Influence of seed size on dispersal biogeographical and ecological studies 99–132. http://dx.doi.org/10.1016/ patterns of woolly monkeys (Lagothrix lagothricha) at Tinigua Park, Colombia. B978-0-444-42755-7.50012-2. Oikos 110, 435–440. Quesada, C.A., Phillips, O., Schwarz, M., Czimczik, C.I., Baker, T.R., Patiño, S., Fyllas, N. Stevenson, P.R., Suescún, M., Quiñones, M., 2004. Characterization of forest types at M., Hodnett, M.G., Herrera, R., Almeida, S., Alvarez Dávila, E., Arneth, A., Arroyo, the CIEM, Tinigua Park, Colombia. Field Studies of Fauna and Flora La Macarena L., Chao, K.J., Dezzeo, N., Erwin, T., di Fiore, A., Higuchi, N., Honorio Coronado, E., Colombia 14, 1–20. Jimenez, E.M., Killeen, T., Lezama, A.T., Lloyd, G., López-González, G., Luizão, F.J., Ter Steege, H., Pitman, N.C.A., Phillips, O., Chave, J., Sabatier, D., Duque, A., Molino, J.- Malhi, Y., Monteagudo, A., Neill, D.A., Núñez Vargas, P., Paiva, R., Peacock, J., F., Prévost, M.-F., Spichiger, R., Castellanos, H., von Hildebrand, P., Vásquez, R., Peñuela, M.C., Peña Cruz, A., Pitman, N., Priante Filho, N., Prieto, A., Ramírez, H., 2006. Continental-scale patterns of canopy tree composition and function Rudas, A., Salomão, R., Santos, A.J.B., Schmerler, J., Silva, N., Silveira, M., Vásquez, across Amazonia. Nature 443, 444–447. http://dx.doi.org/10.1038/nature05134. R., Vieira, I., Terborgh, J., Lloyd, J., 2012. Basin-wide variations in Amazon forest Umaña, M.N., Norden, N., Cano, A., Stevenson, P.R., 2012. Determinants of plant structure and function are mediated by both soils and climate. Biogeosciences 9, community assembly in a mosaic of landscape units in central Amazonia: 2203–2246. http://dx.doi.org/10.5194/bg-9-2203-2012. ecological and phylogenetic perspectives. PLoS ONE 7, e45199. http://dx.doi. Quiñones-Fernandez, M., Florez-Ayala, C., Estipiñan-Suarez, L., 2015. Historias org/10.1371/journal.pone.0045199. contadas por el agua. Frecuencias de inundación. In: Jaramillo-Villa, U., Cortés- Williamson, G.B., Wiemann, M.C., 2010. Measuring wood specific gravity Duque, J., Florez-Ayala, C. (Eds.), Colombia Anfibia. Un País de Humedales. Ellipsiscorrectly. Am. J. Bot. 97, 519–524. http://dx.doi.org/10.3732/ Instituto de Investigación de recursos biológicos Alexander von Humboldt, ajb.0900243. Bogotá, D.C., pp. 80–81. Zanne, A.E., Lopez-Gonzalez, G., Coomes, D.A.A., Ilic, J., Jansen, S., Lewis, S.L.S.L., Quiñones, M., Vissers, M., Pacheco, A., Huertas, C., 2015. Vegetation Structural Map Miller, R.B.B., Swenson, N.G.G., Wiemann, M.C.C., Chave, J., 2009. Global wood for Continental Colombia. density database. Dryad Digital Repository. http://dx.doi.org/10.5061/ Quiroga, R.R., Li, T., Lora, G., Andersen, L., 2013. A Measurement of the Carbon dryad.234. Sequestration Potential of Guadua Angustifolia in the Carrasco National Park. ECONSTOR, Bolivia, p. 04. Appendices

Table A.1. Detailed information on forest type, region and proportion of stems belonging to different dispersal systems in 32 (1-ha) lowland forests of Northern South America.

% % % % % % % Plot_Code Forest Type Region Explosive Anemochory Endozoochory Hydrochory Myrmechochory Synzoochory Unassisted Dehiscence

CAPA_PI_1 Igapo Amazon 22.7 32.6 6.9 18.5 0.0 19.1 0.2

CAPA_PI_2 Igapo Amazon 7.6 36.7 8.9 15.3 0.0 31.6 0.0

CASA_PII_2 Igapo Orinoco 2.8 69.7 7.0 2.3 0.7 17.5 0.0

TOMO_PI_4 Igapo Orinoco 5.5 73.3 7.5 2.2 0.2 11.4 0.0

TOMO_PI_5 Igapo Orinoco 6.0 66.8 12.0 2.5 3.5 9.3 0.0

CAPA_TFC_1 TerraFirme Amazon 3.9 54.6 21.1 2.2 2.6 15.6 0.0

CAPA_TFC_2 TerraFirme Amazon 17.5 62.1 7.8 9.7 0.0 2.9 0.0

CAPA_TFT_1 TerraFirme Amazon 0.2 38.9 55.6 4.4 0.0 0.9 0.0

CAPA_TFT_2 TerraFirme Amazon 13.0 75.1 1.3 0.0 0.0 10.6 0.0

QUIN_TF_1 TerraFirme Magdalen 9.3 74.0 1.6 0.5 0.7 13.3 0.5 a

86

QUIN_TF_3 TerraFirme Magdalen 8.6 76.2 1.7 1.0 0.0 12.5 0.0 a

QUIN_TF_4 TerraFirme Magdalen 11.6 77.1 1.5 0.0 0.2 9.7 0.0 a

QUIN_TF_5 TerraFirme Magdalen 13.4 78.5 3.9 1.3 0.0 2.9 0.0 a

SAMA_TF_3 TerraFirme Orinoco 23.9 67.3 0.0 0.9 0.0 8.0 0.0

TINI_TF_1 TerraFirme Orinoco 7.7 72.6 5.8 9.1 0.0 4.8 0.0

TINI_TF_3 TerraFirme Orinoco 2.5 63.2 2.0 26.9 0.0 5.5 0.0

TINI_TF_4 TerraFirme Orinoco 7.6 58.4 16.6 10.5 0.2 6.6 0.0

TINI_TF_6 TerraFirme Orinoco 5.1 94.5 0.0 0.0 0.0 0.4 0.0

TINI_TF_7 TerraFirme Orinoco 8.2 91.7 0.0 0.0 0.0 0.2 0.0

SAJU_TF_3 TerraFirme_Fra Magdalen 9.7 87.9 0.2 0.0 0.0 2.2 0.0 g a

SAJU_TF_4 TerraFirme_Fra Magdalen 36.0 61.0 0.0 2.8 0.0 0.3 0.0 g a

SAJU_TF_5 TerraFirme_Fra Magdalen 6.7 84.1 0.7 0.4 0.0 6.3 1.8 g a

SAMA_TF_1 TerraFirme_Fra Orinoco 3.3 91.7 0.3 0.0 0.0 4.7 0.0 g

87

SAMA_TF_2 TerraFirme_Fra Orinoco 8.4 61.1 0.5 0.3 0.0 4.1 25.6 g

TOMO_TF_1 TerraFirme_Fra Orinoco 5.9 87.8 2.8 0.0 0.0 3.5 0.0 g

TOMO_TF_2 TerraFirme_Fra Orinoco 4.6 80.1 10.6 1.0 0.0 3.7 0.0 g

TOMO_TF_3 TerraFirme_Fra Orinoco 3.9 78.5 1.9 0.4 0.3 5.7 9.3 g

CASA_PIV_1 Varzea Orinoco 22.1 52.2 18.3 1.2 0.0 6.2 0.0

SAJU_PI_1 Varzea Magdalen 12.9 56.8 21.6 1.3 0.0 7.5 0.0 a

SAJU_PI_2 Varzea Magdalen 25.2 62.7 0.2 0.0 1.2 10.6 0.0 a

TINI_PI_2 Varzea Orinoco 11.3 69.5 1.5 0.0 2.7 15.0 0.0

TINI_PI_5 Varzea Orinoco 7.1 77.3 2.4 0.0 3.3 9.9 0.0

88

CAPÍTULO III – DINAMICA DE COMUNIDADES Y DE CARBONO DE LOS BOSQUES DE TIERRAS BAJAS EN EL NORTE DE SUR AMÉRICA

89

TREE TURNOVER AND CARBON DYNAMICS OF SEASONALLY-FLOODED AND TERRA FIRME FORESTS OF COLOMBIA

Aldana, A. M., Villanueva, B., & Stevenson, P. R. Tree turnover and carbon dynamics of seasonally-flooded and terra firme forests of Colombia. Manuscript in preparation to be submitted to the Journal of Tropical Ecology.

90

Tree turnover and carbon dynamics of seasonally-flooded and terra firme forests of Colombia

Ana M. Aldana1*

Boris Villanueva2

Pablo R. Stevenson1

1. Departamento de Ciencias Biológicas, Universidad de Los Andes, Bogotá, Colombia 2. Laboratorio de Dendrología, Uiversidad del Tolima, Ibagué, Colombia

* Corresponding author: [email protected]

91

Abstract Tropical forest conservation is a mechanism to mitigate the negative effects of climate change. Long term monitoring efforts have produced a growing line of evidence that extreme climatic events result in a decline of the tropical forest carbon sink. Our main objective was to describe forest dynamics of seasonally-flooded and terra firme forests of the Magdalena and Orinoco Basins over a short period of time, and to evaluate the relationship with aboveground biomass – AGB dynamics. Additionally, we evaluated the effect of environmental variables: extreme climatic events, soil fertility and fragmentation on forest and AGB dynamics. During the period of 2011 and 2015 we re-sampled sixteen 1ha vegetation plots, previously established by us or in collaboration with other researchers during the period of 2005 and 2010, eight plots are located in the Magdalena Valley and eight plots in the western Orinoco river basin. The most fertile forests, várzea forests of the Magdalena, had the highest mortality rates and, consequently, lost AGB during the sampling period. There was no strong correlation between mortality rates and climatic variables; however, we presume that high mortality was caused by the intensity of the Magdalena river flood during a La Niña event late 2011-mid 2012.

Keywords: Recruitment, Relative growth rate, Soil fertility, Igapó

92

Resumen La conservación de los bosques tropicales es un mecanismo para mitigar los efectos del cambio climático. Esfuerzos de monitoreo de largo plazo han producido una creciente línea de evidencia que los eventos climáticos extremos resultan en la disminución del sumidero de carbono de los bosques tropicales. Nuestro principal objetivo era describir la dinámica de los bosques estacionalmente inundables y los de tierra firme de las cuencas de los ríos Magdalena y Orinoco en un periodo corto de tiempo, y evaluar la relación con la dinámica de la biomasa aérea. Adicionalmente, evaluamos los efectos de variables ambientales: eventos climáticos extremos, fertilidad del suelo y la fragmentación sobre la dinámica del bosque y de la biomasa aérea. Durante el periodo entre 2011 y 2015 re-muestreamos dieciséis parcelas permanentes de vegetación de 1ha, establecidas previamente por nosotros o en colaboración con otros investigadores durante el periodo de 2005 y 2010. Ocho parcelas están en el Magdalena medio y ocho parcelas en la Orinoquía. Los bosques más fértiles, las várzeas del Magdalena, tuvieron las mayores tasas de mortalidad y consecuentemente perdieron biomasa (carbono), durante el periodo de muestreo. No hubo una correlación fuerte entre las tasas de mortalidad y las variables climáticas; sin embargo, suponemos que la alta mortalidad fue causada por la intensidad de la inundación del río Magdalena durante el evento de La Niña en 2011-2012.

93

Introduction The protection of tropical forests has become imminent as it serves as a mechanism to mitigate the negative effects of climate change, given the fact that forests act as carbon sinks, accumulating carbon, a major greenhouse gas in the soil and the biomass of trees (Pan et al. 2011, 2013). The accumulation of carbon in forests has been studied primarily by the estimation of above ground biomass – AGB: the amount of biomass accumulated in the trunks, twigs and leaves of live trees, and its equivalent in carbon, which is usually estimated as half of the biomass (Chave et al. 2005, 2014, Feldpausch et al. 2012, Baraloto et al. 2013). AGB stocks in Neotropical forests have been proven to be controlled by structural components of the forests such as the density of large trees (Slik et al. 2013), tree species composition (Fauset et al. 2015), and tree species diversity (Poorter et al. 2015); and by environmental variables such as soil fertility (Quesada et al. 2012) and climate (Alvarez et al. 2016- Anexo 2).

In terms of AGB dynamics, it has been established that forest dynamics (mortality, recruitment and stem turnover) are a determinant factor of the carbon balance in tropical forests. In forests where tree mortality rates are high there is a tendency to lose biomass and thus, the carbon accumulated in the biomass is liberated back to the atmosphere. This has been observed by the monitoring, over four decades, of forests in the Amazon basin (Brienen et al. 2015, Johnson et al. 2016). These long term monitoring efforts have produced a growing line of evidence that extreme environmental events such as prolonged droughts can cause high tree mortality rates, resulting in a decline of the tropical forest carbon sink (Phillips et al. 2009, Feldpausch et al. 2016).

Lately, in light of the implementation of a Reducing Emissions from Deforestation and Forest Degradation – REDD+ strategy in Colombia, some studies have reported AGB stocks for different forest types (Sierra et al. 2007, Alvarez et al. 2016-Anexo 2, Phillips et al. 2016) and even a research group lead by Alvarez (et al 2012) produced AGB allometric equations for Colombian forests. However, to date, less than a handful of studies have reported forest or AGB dynamics for Colombian forests (Aldana & Stevenson 2016 - Capítulo 3, Peña & Duque 2013, Restrepo et al. 2016 - Anexo 1). These studies suggest that the history of land use and its effects on forest dynamics are the most determinant factors for AGB dynamics.

94

In Colombia, there are marked effects on climate and river hydraulics during the years when there are extreme phases of El Niño Southern Oscilation – ENSO events (Poveda & Mesa 1996, Poveda 2004). During an El Niño event the western region of the country experiences a prolonged drought, while during a La Niña event there is the opposite pattern, and the western region receives excess rainfall. Due to the recent effects of climate change, these ENSO events have been occurring in shorter time intervals (IDEAM 2016). These extreme climatic events could have a negative effect on forest and AGB dynamics in Colombian forests, particularly in the western part of the country (Magdalena river Valley).

Our main objective is to describe the forest dynamics of seasonally-flooded and terra firme forests of the Magdalena and Orinoco Basins over a short period (2005 – 2015), and to evaluate the relationship of these dynamics with AGB dynamics. Additionally, we want to evaluate the effect of environmental variables, such as extreme climatic events, soil nutrient content and fragmentation on forest and AGB dynamics. In general, we expect that the ENSO events that occurred recently had a negative effect on forest and AGB dynamics, enhancing tree mortality and biomass loses in forests of the Magdalena basin. Because the ENSO is not so strong in the east part of Colombia, we do not expect negative effects in the Orinoco basin. Considering soil nutrient content, we expect rich soils to hold more dynamic forests that should be at AGB equilibrium. Regarding fragmentation, we expect to find higher mortality in forests with the highest degrees of fragmentation due to higher edge effects; these forests would be favored by the opening of gaps, creating an optimal environment for fast growing pioneer plants, so the highest recruitment should be found in these forests as well. Consequently, AGB changes in fragmented forests should be negative due to the replacement of old-growth trees with low wood density pioneer species.

We present the first integrative effort to understand forest and AGB dynamics for Colombian forests. Our findings show that non-amazonic rainforests of Northern South America are affected by different climatic effects than forests of the Amazon basin, and reinforce the need to continue with long term monitoring of these forests, in order to plan better conservation strategies on the light of future climate change.

95

Methods Field study

The inventoried forest plots are located in the Magdalena Valley and the western Orinoco river basin in Colombia (Figure 1). The Magdalena Valley is located in the Andean region which has been the most populated, and thus, the most deforested region in the country. During the last four decades large extensions of continuous forest have been replaced by vast pastures for cattle ranching and small forest patches (Etter et al. 2006). Seasonally-flooded forests in the Magdalena Valley are inundated by Andean rivers with high nutrient contents and were classified by Prance (1989) as seasonal vázea. Forests in this region are disappearing due to increased human pressure for crop production (Rodríguez Eraso et al. 2013). Climate in the region is bi-modal with two rainy seasons (April-May; October- November), mean annual precipitation of 2,900mm, and mean annual temperature of 28ºC (data from this study).

Figure 1: Map of the approximate location of the 16 1ha vegetation plots inventoried in this study. The two main regions studied are the Magdalena and Orinoco basin, delimited with gray dotted lines. Major rivers are shown in light blue. Country limits are shown white. Triangles represent places where seasonally flooded forests were sampled, two plots per triangle. Squares represent places were terra firme forests were sampled, 2 to 4 plots per triangle.

96

The western Orinoco river basin is composed by a matrix of diverse ecosystems (Romero Ruíz et al. 2004) which include forests fragments of the foothills of the Andes, gallery forests in a savanna matrix and seasonally flooded forests, all of the former included in our study system. The floodplain forests included in this study are inundated by the Tomo River, a nutrient poor black river, and are thus classified as seasonal igapó (Prance 1989). The natural and human-created savannas in the region are exploited for cattle ranching (Etter et al. 2010). Climate in the region is monomodal with one dry season (December-February) and one rainy season (April- September), mean annual precipitation of 2,500mm and mean annual temperature of 26ºC (data from this study).

During the period of 2011 and 2015 we re-sampled sixteen 1ha vegetation plots that had been previously established by us or in collaboration with other researchers during the period of 2005 and 2010 (Aldana et al. 2008, Stevenson & Aldana 2008, Correa-Gómez & Stevenson 2010). Average time between the establishment and re- sampling of the plots was 5.4 years, with a maximum time of 7 years and a minimum of 4 years. Of these sixteen plots, eight are located in the Magdalena Valley and eight are located in the western Orinoco river basin. In each of the basins two of the plots are found in seasonally flooded forests. The premises where the plots are located are either cattle ranches or private nature reserves.

Following the RAINFOR protocol, for plot establishment and re-sampling, we measured Diameter at Breast Height – DBH increase of trees greater than 10 cm; we also recorded the number of new recruits and dead individuals; when possible, we documented the cause of death. When plots were established the Point of Measurement – POM was not marked, but DBH was almost always measured at 1,30 meters from the ground. During re-samplig, we tried to maintain this stantard for the POM except when we encountered deformations, in these cases the POM was moved 30cm, either up or down from the original POM. Trees with buttresses where measured above buttresses and when these where too high we used photographies with a scale to calculate DBH. We did not delete from our database tres that had excesive growth rates as there was a trend within some pioneer species, for example Trema and Apeiba species. In the few cases where we found “ghost” recruits these were included in the original database and asummed to have had no DBH increases. 97

During plot re-sampling, we additionally measured height for 68% of the individuals using a laser clinometer.

To determine nutrient availability in the soil, while re-sampling, we collected 25 samples of topsoil (in the center of 20m x 20m subplots) in each of the plots. Each soil sample was analyzed for macronutrient content (P, Al, Mg, Ca, K, Na), effective cation exchange capacity – ECEC, pH, and texture either at the Laboratorio de Aguas y Suelos, Universidad Nacional de Colombia Sede Bogotá or at the Laboratorio de Suelos La Sabana in Villavicencio (Meta).

To determine a level of fragmentation in each of the plots, we measured forest cover percentage in a circle of 5km around the plot, using Google Earth Plus images and measurement tools.

Climatic variables such as total annual precipitation, average monthly precipitation, duration and total precipitation of the dry season (number of consecutive months precipitation < 50mm), duration and total precipitation of the wet season (number of consecutive months with precipitation > 250mm), were calculated using the data extracted of the sampled period, from the nearest meteorological station to the locality where the plots are located. This information was provided by the Instituto de Hidrología, Meteorología, y Estudios Ambientales – IDEAM via a request for research purposes.

Data analysis Environmental and climatic variable selection Soil fertility is very well represented by ECEC, because this variable is calculated adding the concentration of Ca, Mg, Na and K cations, so we used this variable as a summary for soil fertility and soil data. All the climatic variables we extracted from the meteorological data were highly correlated between them, so we chose the average duration of the dry season to represent these climatic variables, because it had the highest correlation values with all the other variables.

Community Dynamics – tree turnover rates Comparing the information from the establishment and the re-sampling of the plots, we calculated mortality, recruitment, population change and relative growth (RGR) rates, using the equations proposed by Sherman et al.(2012) for all plots.

98

Here, tree turnover rate is defined as the difference between recruitment and mortality rates. This follows the equation for population change proposed by Sherman et al.(2012), but it was calculated at the community level (plot level) and not at the population level (each species within plots). Relative growth rate is calculated as the percentage of DBH increase per year within the sampling period.

RGR = (ln(dbht2)−ln(dbht1))/(t2−t1),

AGB stocks and AGB change For each individual stem in our dataset, we calculated AGB for the year of establishment and the year of sampling and added all stems in each plot to get results of AGB stocks per plot, using a combination of allometric equations, to minimize errors in the estimations. For trees we used the allometric equation for Tropical Moist Forest (Type I.3), which takes into account DBH, height and wood density (Alvarez et al. 2012); for palm species we calculated the biomass using the equations proposed by Goodman et al. (2013), and for lianas we calculated the biomass using the equation proposed by Schnitzer et al. (2006).

Height measurements from the time of the re-sampling were used to calculate height average for the time of establishment and the time of re-sampling for all stems in the database, subdivided in to eight DBH categories.

Wood density estimations for tree species were taken from a wood density database of tropical species, constructed by the combination of published and unpublished data (Zanne et al. 2009, Casas-Caro et al. 2016 - Anexo 4). When tree species were not reported in this database, or when morphospecies were only identified to genus, we calculated a genus wood density average assuming trait conservatism (Chave et al. 2006), as well as a family average for morphospecies only identified to family. For unidentified morphospecies we calculated a plot wood density average.

The net change in AGB was calculated as the difference between the AGB stock in the year of establishment of the plots and the AGB stock in the time of the re- sampling, divided by the number of years between each sampling moment.

99

Results Community Dynamics – tree turnover rates Estimated dynamic rates were highly variable among forest types and regions (Table 1), seasonally flooded forests in both regions had negative community changes (mortality higher than recruitment), and terra firme forests in both regions had higher recruitment rates than mortality. RGR was lowest in the igapó forest (Orinoco).

100

Table 1: Summary of the biomass per hecatare (Mg ha), biomass change rate (Mg ha year), rates of mortality, recruitment, community dynamics and relative growth –RGR estimated, ECEC, duration of the dry season (Dry), change in the number of large trees (DBH>70cm; Large); percentage forests cover around the plot (Frag); change in species richness (Spp) for each of the 16 1ha plots, values are show as averages for each forest type in the two studied regions.

Population AGB AGB Mortality Recruitment RGR Basin Forest type change ECEC Dry Large Frag Spp Mg ha Mg ha year % % % %

Terrafirme 252.76 1.50 2.13 2.41 0.28 1.32 6.56 0.63 -1.67 0.55 5.33

Magdalena Flooded 371.29 -17.28 7.40 3.29 -4.11 1.75 21.26 0.63 -3.50 0.48 0.00

Average 282.39 -3.19 3.45 2.63 -0.82 1.43 10.23 0.63 -2.13 0.54 4.00

Terrafirme 147.53 0.73 3.08 3.54 0.46 1.52 4.56 1.94 1.50 0.28 1.50

Orinoco Flooded 207.26 1.51 2.17 1.72 -0.45 0.61 5.19 2.30 2.00 0.29 1.50

Average 162.46 0.93 2.85 3.08 0.23 1.30 4.71 2.03 1.63 0.28 1.50

101

AGB stocks and AGB change We found that forests of the Orinoco tend to have smaller (100-200 Mg per ha) AGB stocks than forests of the Magdalena (200-400 Mg per ha). In both regions seasonally flooded forests tend to have higher AGB stocks than terra firme forests. Regarding AGB change, both regions behaved differently: forests in the Orinoco accumulated biomass at a rate of 0.93 Mg ha-1 y-1 during the sampling period, and forests in the Magdalena Valley lost biomass at a rate of 3.2 Mg ha-1 y-1 during the sampling period (Figure 2), this trend is mainly controlled by the várzea forests that lost, in average, 17 Mg of biomass per year.

Figure 2: Above ground biomass change in terra firme and seasonally flooded forests. Left panel forests of the western Orinoco river basin, right panel forests of the Magdalena Valley. Forest dynamics and AGB change Exploring the relationship between tree turnover dynamics and AGB change we found that there was a strong positive correlation between these two variables. Forest plots with negative turnover rates lost AGB during the sampling period (Figure 3).

102

Figure 3: Correlation between the annual turnover rate and the annual above ground biomass AGB change for 16 1ha vegetation plots of terra firme and seasonally flooded forests surveyed in the Magdalena Valley and the western Orinoco river basin (pearson´s correlation= 0.88; p<0.001).

This result is derived from the high mortality rates experimented by some the seasonally flooded forests (Table 1), which was higher than the recruitment and thus results in a negative tree turnover rate.

Mortality was positively correlated with soil fertility, expressed as ECEC (pearson´s correlation= 0.80; p<0.001). This result can be a coincidence of high mortality in fertile flooded forests of the Magdalena Valley, as it can be observed in Figures 3 and 4, plots with the highest values of mortality are the várzea forests of the Magdalena and these are also the forests with the highest values of ECEC.

103

Figure 4: Boxplot showing the variation in soil fertility, represented by effective cation exchange capacity - ECEC for 16 1ha vegetation plots of seasonally flooded and terra firme forests in the Magdalena Valley (red) and Orinoco river basin (black). Differences were estimated with a ANOVA (F=39.93, p<0.001), varzea forests have the highest ECEC, confirmed with a post hoc test. Taking into account that AGB stocks in these forests are mainly driven by the number of large trees (DBH> 70cm) (Slik et al. 2013), we evaluated if the change measured as mortality or recruitment within that DBH category, was correlated with the change in AGB. We found that this correlation is weak and not significant (Table 2). However, we established that the change in the number of large trees is positively correlated with the total precipitation during the dry season (Table 2).

We also wanted to evaluate the effect of losing or gaining species in the AGB accumulation change, because previous research has shown that species richness is a very important variable for AGB stocks (Poorter et al. 2015), so we calculated the change in species richness and we evaluated the correlation of this variable with all other measured variables (Table 2). We found that this variable is marginally correlated with AGB change (0.44) but this correlation is not significant.

104

Table 2: Summary table of the correlations evaluated between climatic, dynamic and environmental variables of 16 1ha vegetation plots in terra firme and seasonally flooded forests in the Magdalena and Orinoco regions in Northern South America. Correlations in bold font are significant (**p- value<0.001; *p-value<0.01). Names of variables were shortened for ease of presentation: Dry, average precipitation during the dry season; AGB, annual rate of AGB change; M, annual mortality rate; R, annual recruitment rate; L, change in the number of large trees (DBH>70cm); F, percentage forests cover around the plot; S, change in species richness; T, annual tree turnover rate.

Dry AGB M R L F S ECEC RGR T

Dry 1

AGB -0.26 1

M 0.07 -0.79** 1

R -0.35 0.01 0.45 1

L -0.72* 0.38 -0.31 0.01 1

F 0.64 -0.15 -0.06 -0.02 -0.59 1

S 0.17 0.44 -0.33 0.19 -0.13 0.35 1

ECEC 0.46 -0.64 0.80** 0.17 -0.50 0.22 -0.06 1

RGR 0.01 -0.14 0.45 0.48 -0.12 -0.13 0.10 0.37 1

T -0.32 0.88** -0.79** 0.19 0.35 0.06 0.49 -0.77** -0.16 1

Discussion Community Dynamics – tree turnover rates In general, the flooded forest of the Magdalena (várzea) was more dynamic than the other forest studied; mortality, recruitment and RGR were higher in this forest when compared to other forest types (Table 1). This maybe a result of high soil fertility (Figure 4), as it has been reported before for várzea forests of the Amazon (Schöngart et al. 2010). Opposed to this result, flooded forests of the Orinoco (igapó) were less dynamic. Specially, RGR was very low when compared to other forest types. This result maybe a consequence of the physiological constrains of the flood, and low soil fertility. When examining the growth patterns of Macrolobium acaciifolium in igapó and várzea forests in the Amazon basin (Schöngart et al. 2005) that had similar flooding periods, found that wood growth was much slower in the Igapó, showing that soil nutrient content is a major limiting factor for tree growth in the floodplains.

105

Forest dynamics and AGB change In many plots, mortality was higher than recruitment, which resulted in negative stem turnover, and this, in turn, had an effect on AGB change. In fact, tree turnover was the variable with the highest effect on AGB change than the other variables studied (Table 2). In the Amazon basin, where most of our knowledge from Neotropical forests comes from, this has been also the case. Mortality rates are determinant for the AGB change rates (Phillips et al. 2009, Brienen et al. 2015). Forest managers should be able to understand the drivers of tree mortality in order to be able to plan strategies that will help diminish mortality rates and the maintenance of the forest carbon sink.

AGB stocks and AGB change Forests in the Magdalena region have higher AGB stocks than in the Orinoco, due to the fact that these forests are taller, have superior numbers of large trees (DBH>70cm) and are more diverse (Aldana et al. 2016 -Capítulo 2, Restrepo et al. 2016 - Anexo 1, González et al. 2016 - Anexo 3). This fact may also have an effect on the magnitude of the AGB change rates, which are usually steeper in the Magdalena (Figure 2). This result has a direct impact on the definition of REDD+ strategies and “payment for ecosystem services”, which, in Colombia should differentiate between regions, not only on the opportunity costs (Rivera & Harb 2014) but also on forest function.

Effects of environmental variables The most fertile forests, várzea forests of the Magdalena, were the forests with the highest mortality rates and, consequently, these forests lost biomass during the sampling period. Although there was no strong correlation between mortality rates and climatic variables in our study system, from our field observations during plot re- sampling, we evidenced that in the plots with the highest mortality rates, the cause of death was the uprooting of the trees. The majority of the trees that died in this period were small trees (DBH < 20cm), and many of them (21%) were identified as Cordia collococca a dominant species in this system (Chapter 1).

We presume that this mortality was caused by the intensity of the flood from the Magdalena river during the La Niña event, which occurred a couple of months before the re-sampling of the plots, late 2011-mid 2012 (IDEAM), this La Niña event was 106 very strong, making the Magdalena river reach up to 10mts of it´s usuall level, very similar to what happended in La Niña event in 1999. High fertility is associated to high percentage of clay in the soil (Damasco et al. 2013), which has been correlated to tree mortality of trees, specially tree uprooting (dos Santos et al. 2015, Ribeiro et al. 2016).

Another important factor that can be causing the differences between forest plots are the differences in management and history of land use. As it was evidenced by (Aldana & Stevenson 2016) forests that are located inside cattle ranches tend to have lower recruitment rates due to the negative effects of cattle that uses the forest for water supply and shadow. Most plots in our study system are located in cattle ranches where the cattle are free to range inside the forest. Some studies have also shown that the history of land use, such as selective logging and crop production have a long term effect on AGB dynamics (Peña & Duque 2013, Restrepo et al. 2016 - Anexo 1).

We expected forest fragmentation to have a negative effect on forest and AGB dynamics, due to enhanced tree mortality caused by edge effects. This prediction was not fulfilled: there was no significant correlation between percentage forest cover and forest dynamics and AGB change. The effects of forest fragmentation could be coupled with the effects of forest structure (see above), because plots with the lowest percentage of forest cover are also located in the Orinoco basin (Appendix 3).

We also expected that, because the number of large trees is a major driver of AGB (Slik et al. 2013, Aldana et al. 2016 - Capitulo 2), the change in the number of large trees should have an effect on AGB change. We did not find this relationship to be strong. However, this variable was correlated with the duration of the dry season, implying that climate is an important variable controlling tree longevity (Schöngart et al. 2005); thus, under future climate change scenarios where the periodicity of extreme droughts is predicted to increase these forests will decrease their AGB stocks. Previous studies on forest carbon dynamics highlighted the importance of species richness. We found that the change in species richness had no correlation with the change on AGB. Although, it important to note that the magnitude of the change was very small. 107

We acknowledge the fact that this data set is small and may not be representative of all forests of the study regions (Clark & Clark 2011); however, this is, to date, the first attempt to describe forest and AGB dynamics of these forests that have been long understudied and are at high risk. We also understand that this is a short period of time and it was a very atypical in terms of the frequency of extreme climatic events; nonetheless, in the future, due to climate change effects and enhanced greenhouse gasses emissions, the periodicity of this events is expected to increase further studies should focus on larger temporal periods.

Acknowledgements This research was funded by the Facultad de Ciencias, Universidad de los Andes, the Colombia L´Oréal – Unesco For Women in Science grant awarded to AMA, and the Universidad del Tolima supporting BVT. We would like to thank all the volunteers who helped us during the re-sampling of the plots. Fundación ProAves, Familia Sánchez-Rey, Familia Enciso, and Familia Lalinde were very kind to let us work in their premises and supported us with logistics.

108

Cited literature

Aldana, A. M., Beltrán, M., Torres-Neira, J. & Stevenson, P. R. 2008. Habitat Characterization And Population Density Of Brown Spider Monkeys (Ateles Hybridus) In Magdalena Valley, Colombia. Neotropical Primates 15:46–50.

Aldana, A. M. & Stevenson, P. R. (N.D.). Forest Fragments Of The Andean Piedmont As Carbon Sinks: Short- Term Gain Of Above Ground Biomass In Fragments Used By Cattle Ranches. Tropical Conservation Science.

Aldana, A. M., Villanueva, B., Cano, Á., Correa, D. F., Umaña, M. N., Casas-Caro, L., Cárdenas Hoyos, S., Henao-Diaz, L. F. & Stevenson, P. R. 2016. Drivers Of Biomass Stocks In Northwestern South American Forests: Complementing Information For The Neotropics. Manuscript Submitted For Publication.

Alvarez, E., Cayuela, L., Gonzalez-Caro, S., Aldana, A. M., Stevenson, P. R., Phillips, O. L., Von Hildebrand, P., Jiménez, E., Melo, O., Mendoza, I., Restrepo, Z., Velásquez, O. & Rey-Benayas, J. M. 2016. Forest Biomass Density Across Large Climate Gradients In Northern South America Is Related To Water Availability But Not With Temperature. Manuscript Submitted For Publication.

Alvarez, E., Duque, A., Saldarriaga, J., Cabrera, K., De Las Salas, G., Del Valle, I., Lema, A., Moreno, F., Orrego, S. & Rodríguez, L. 2012. Tree Above-Ground Biomass Allometries For Carbon Stocks Estimation In The Natural Forests Of Colombia. Forest Ecology And Management 267:297–308. Jour, .

Baraloto, C., Molto, Q., Rabaud, S., Hérault, B., Valencia, R., Blanc, L., Fine, P. V. A. & Thompson, J. 2013. Rapid Simultaneous Estimation Of Aboveground Biomass And Tree Diversity Across Neotropical Forests: A Comparison Of Field Inventory Methods. Biotropica 45:288–298. Jour, .

Brienen, R. J. W., Phillips, O. L., Feldpausch, T. R. & Et Al. 2015. Long-Term Decline Of The Amazon Carbon Sink. Nature 519:344–348. Nature Publishing Group.

Casas-Caro, L., Aldana, A. M., Henao-Diaz, F., Villanueva, B. & Stevenson, P. R. 2016. Specific Gravity Of Woody Tissue From Lowland Neotropical Plants: Variance Among Forest Types. Data Paper Submitted For Publication.

Chave, J., Andalo, C., Brown, S., Cairns, M. A., Chambers, J. Q., Eamus, D., Fölster, H., Fromard, F., Higuchi, N., Kira, T., Lescure, J.-P., Nelson, B. W., Ogawa, H., Puig, H., Riéra, B. & Yamakura, T. 2005. Tree Allometry And Improved Estimation Of Carbon Stocks And Balance In Tropical Forests. Oecologia 145:87– 99. Jour, .

Chave, J. J., Muller-Landau, H. C., Baker, T. R., Easdale, T. A., Steege, H. Ter, Webb, C. O., Hans Steege, T. E. R. & Webb, C. O. 2006. Regional And Phylogenetic Variation Of Wood Density Across 2456 Neotropical Tree Species. Ecological Applications 16:2356–2367. Jour, .

Chave, J., Réjou-Méchain, M., Búrquez, A., Chidumayo, E., Colgan, M. S., Delitti, W. B. C., Duque, A., Eid, T., Fearnside, P. M., Goodman, R. C., Henry, M., Martínez-Yrízar, A., Mugasha, W. A., Muller-Landau, H. C., Mencuccini, M., Nelson, B. W., Ngomanda, A., Nogueira, E. M., Ortiz-Malavassi, E., Pélissier, R., Ploton, P., Ryan, C. M., Saldarriaga, J. G. & Vieilledent, G. 2014. Improved Allometric Models To Estimate The Aboveground Biomass Of Tropical Trees. Global Change Biology:3177–3190.

Clark, D. A. & Clark, D. B. 2011. Assessing Tropical Forests’ Climatic Sensitivities With Long-Term Data. Biotropica 43:31–40.

Correa-Gómez, D. F. & Stevenson, P. R. 2010. Estructura Y Diversidad De Bosques De Los Llanos Orientales Colombianos (Reserva Tomo Grande, Vichada). Orinoquia 14:31–48.

Damasco, G., Vicentini, A., Castilho, C. V., Pimentel, T. P. & Nascimento, H. E. M. 2013. Disentangling The Role Of Edaphic Variability, Flooding Regime And Topography Of Amazonian White-Sand Vegetation. Journal Of Vegetation Science 24:384–394.

Etter, A., Mcalpine, C., Pullar, D. & Possingham, H. 2006. Modelling The Conversion Of Colombian Lowland Ecosystems Since 1940: Drivers, Patterns And Rates. Journal Of Environmental Management 79:74–87.

Etter, A., Sarmiento, A. & Romero, M. 2010. Land Use Changes (1970–2020) And Carbon Emissions In The Colombian Llanos. Pp. 383–402in Hill, M. & Hanna, N. . (Eds.).Ecosystem Function In Savannas. Crc Press.

Fauset, S., Johnson, M. O., Gloor, M., Baker, T. R., Monteagudo M., A., Brienen, R. J. W., Feldpausch, T. R.,

109

Lopez-Gonzalez, G., Malhi, Y., Ter Steege, H., Pitman, N. C. A., Baraloto, C., Engel, J., Pétronelli, P., Andrade, A., Camargo, J. L. C., Laurance, S. G. W., Laurance, W. F., Chave, J., Allie, E., Vargas, P. N., Terborgh, J. W., Ruokolainen, K., Silveira, M., Aymard C., G. A., Arroyo, L., Bonal, D., Ramirez-Angulo, H., Araujo-Murakami, A., Neill, D., Hérault, B., Dourdain, A., Torres-Lezama, A., Marimon, B. S., Salomão, R. P., Comiskey, J. A., Réjou-Méchain, M., Toledo, M., Licona, J. C., Alarcón, A., Prieto, A., Rudas, A., Van Der Meer, P. J., Killeen, T. J., Marimon Junior, B.-H., Poorter, L., Boot, R. G. A., Stergios, B., Torre, E. V., Costa, F. R. C., Levis, C., Schietti, J., Souza, P., Groot, N., Arets, E., Moscoso, V. C., Castro, W., Coronado, E. N. H., Peña-Claros, M., Stahl, C., Barroso, J., Talbot, J., Vieira, I. C. G., Van Der Heijden, G., Thomas, R., Vos, V. A., Almeida, E. C., Davila, E. Á., Aragão, L. E. O. C., Erwin, T. L., Morandi, P. S., De Oliveira, E. A., Valadão, M. B. X., Zagt, R. J., Van Der Hout, P., Loayza, P. A., Pipoly, J. J., Wang, O., Alexiades, M., Cerón, C. E., Huamantupa-Chuquimaco, I., Di Fiore, A., Peacock, J., Camacho, N. C. P., Umetsu, R. K., De Camargo, P. B., Burnham, R. J., Herrera, R., Quesada, C. A., Stropp, J., Vieira, S. A., Steininger, M., Rodríguez, C. R., Restrepo, Z., Muelbert, A. E., Lewis, S. L., Pickavance, G. C. & Phillips, O. L. 2015. Hyperdominance In Amazonian Forest Carbon Cycling. Nature Communications 6:6857.

Feldpausch, T. R., Lloyd, J., Lewis, S. L., Brienen, R. J. W., Gloor, M., Monteagudo Mendoza, A., Lopez- Gonzalez, G., Banin, L., Abu Salim, K., Affum-Baffoe, K., Alexiades, M., Almeida, S., Amaral, I., Andrade, A., Aragão, L. E. O. C., Araujo Murakami, A., Arets, E. J. M. M., Arroyo, L., Aymard C., G. A., Baker, T. R., Bánki, O. S., Berry, N. J., Cardozo, N., Chave, J., Comiskey, J. A., Alvarez, E., De Oliveira, A., Di Fiore, A., Djagbletey, G., Domingues, T. F., Erwin, T. L., Fearnside, P. M., França, M. B., Freitas, M. A., Higuchi, N., Iida, Y., Jiménez, E., Kassim, A. R., Killeen, T. J., Laurance, W. F., Lovett, J. C., Malhi, Y., Marimon, B. S., Marimon-Junior, B. H., Lenza, E., Marshall, A. R., Mendoza, C., Metcalfe, D. J., Mitchard, E. T. A., Neill, D. A., Nelson, B. W., Nilus, R., Nogueira, E. M., Parada, A., Peh, K. S.-H., Pena Cruz, A., Peñuela, M. C., Pitman, N. C. A., Prieto, A., Quesada, C. A., Ramírez, F., Ramírez-Angulo, H., Reitsma, J. M., Rudas, A., Saiz, G., Salomão, R. P., Schwarz, M., Silva, N., Silva-Espejo, J. E., Silveira, M., Sonké, B., Stropp, J., Taedoumg, H. E., Tan, S., Ter Steege, H., Terborgh, J., Torello-Raventos, M., Van Der Heijden, G. M. F., Vásquez, R., Vilanova, E., Vos, V. A., White, L., Willcock, S., Woell, H. & Phillips, O. 2012. Tree Height Integrated Into Pantropical Forest Biomass Estimates. Biogeosciences 9:3381–3403.

Feldpausch, T. R., Phillips, O. L., Brienen, R. J. W., Gloor, E., Lloyd, J., Lopez-Gonzalez, G., Monteagudo- Mendoza, A., Malhi, Y., Alarcón, A., Álvarez Dávila, E., Alvarez-Loayza, P., Andrade, A., Aragao, L. E. O. C., Arroyo, L., Aymard C., G. A., Baker, T. R., Baraloto, C., Barroso, J., Bonal, D., Castro, W., Chama, V., Chave, J., Domingues, T. F., Fauset, S., Groot, N., Honorio C., E., Laurance, S., Laurance, W. F., Lewis, S. L., Licona, J. C., Marimon, B. S., Marimon-Junior, B. H., Mendoza Bautista, C., Neill, D. A., Oliveira, E. A., Oliveira Dos Santos, C., Pallqui Camacho, N. C., Pardo-Molina, G., Prieto, A., Quesada, C. A., Ramírez, F., Ramírez-Angulo, H., Réjou-Méchain, M., Rudas, A., Saiz, G., Salomão, R. P., Silva-Espejo, J. E., Silveira, M., Ter Steege, H., Stropp, J., Terborgh, J., Thomas-Caesar, R., Van Der Heijden, G. M. F., Vásquez Martinez, R., Vilanova, E. & Vos, V. A. 2016. Amazon Forest Response To Repeated Droughts. Global Biogeochemical Cycles.

Goodman, R. C., Phillips, O. L., Del Castillo Torres, D., Freitas, L., Cortese, S. T., Monteagudo, A. & Baker, T. R. 2013. Amazon Palm Biomass And Allometry. Forest Ecology And Management 310:994–1004. Elsevier B.V.

Ideam. 2016. Instituto De Hidrología, Meteorología Y Estudios Ambientales - Colombia.

Johnson, M. O., Galbraith, D., Gloor, E., De Deurwaerder, H., Guimberteau, M., Rammig, A., Thonicke, K., Verbeeck, H., Von Randow, C., Monteagudo, A., Phillips, O. L., Brienen, R. J. W., Feldpausch, T. R., Lopez Gonzalez, G., Fauset, S., Quesada, C. A, Christoffersen, B., Ciais, P., Gilvan, S., Kruijt, B., Meir, P., Moorcroft, P., Zhang, K., Alvarez, E. A, Alves De Oliveira, A., Amaral, I., Andrade, A., Aragao, L. E. O. C., Araujo-Murakami, A., Arets, E. J. M. M., Arroyo, L., Aymard, G. A, Baraloto, C., Barroso, J., Bonal, D., Boot, R., Camargo, J., Chave, J., Cogollo, A., Cornejo, F. V., Costa, L. Da, Di Fiore, A., Ferreira, L., Higuchi, N., Honorio, E., Killeen, T. J., Laurance, S. G., Laurance, W. F., Licona, J., Lovejoy, T., Malhi, Y., Marimon, B., Marimon, B. H. J., Matos, D. C. L., Mendoza, C., Neill, D. A, Pardo, G., Peña-Claros, M., Pitman, N. C. A, Poorter, L., Prieto, A., Ramirez-Angulo, H., Roopsind, A., Rudas, A., Salomao, R. P., Silveira, M., Stropp, J., Ter Steege, H., Terborgh, J., Thomas, R., Toledo, M., Torres-Lezama, A., Van Der Heijden, G. M. F., Vasquez, R., Vieira, I., Vilanova, E., Vos, V. A & Baker, T. R. 2016. Variation In Stem Mortality Rates Determines Patterns Of Aboveground Biomass In Amazonian Forests: Implications For Dynamic Global Vegetation Models. Global Change Biology.

Pan, Y., Birdsey, R. A., Fang, J., Houghton, R., Kauppi, P. E., Kurz, W. A., Phillips, O. L., Shvidenko, A., Lewis, S. L., Canadell, J. G., Ciais, P., Jackson, R. B., Pacala, S. W., Mcguire, A. D., Piao, S., Rautiainen, A., Sitch, S. & Hayes, D. 2011. A Large And Persistent Carbon Sink In The World’s Forests. Science 333:988– 993. Jour, .

Pan, Y., Birdsey, R. A., Phillips, O. L. & Jackson, R. B. 2013. The Structure, Distribution, And Biomass Of The World’s Forests. Annual Review Of Ecology, Evolution, And Systematics 44:593–622.

110

Peña, M. A. & Duque, A. 2013. Patterns Of Stocks Of Aboveground Tree Biomass, Dynamics, And Their Determinants In Secondary Andean Forests. Forest Ecology And Management 302:54–61.

Phillips, J. F., Duque, Á., Scott, C., Wayson, C., Galindo, G., Cabrera, E., Chave, J., Peña, M., Álvarez, E., Cárdenas, D., Duivenvoorden, J., Hildebrand, P., Stevenson, P., Ramírez, S. & Yepes, A. 2016. Live Aboveground Carbon Stocks In Natural Forests Of Colombia. Forest Ecology And Management 374:119– 128. Elsevier B.V.

Phillips, O. L., Aragao, L. E. O. C., Lewis, S. L., Fisher, J. B., Lloyd, J., Lopez-Gonzalez, G., Malhi, Y., Monteagudo, A., Peacock, J., Quesada, C. A, Van Der Heijden, G., Almeida, S., Amaral, I., Arroyo, L., Aymard, G., Baker, T. R., Banki, O., Blanc, L., Bonal, D., Brando, P., Chave, J., De Oliveira, A. C. A., Cardozo, N. D., Czimczik, C. I., Feldpausch, T. R., Freitas, M. A., Gloor, E., Higuchi, N., Jimenez, E., Lloyd, G., Meir, P., Mendoza, C., Morel, A., Neill, D. A, Nepstad, D., Patino, S., Penuela, M. C., Prieto, A., Ramirez, F., Schwarz, M., Silva, J., Silveira, M., Thomas, A. S., Steege, H. Ter, Stropp, J., Vasquez, R., Zelazowski, P., Davila, E. A., Andelman, S., Andrade, A., Chao, K.-J., Erwin, T., Di Fiore, A., C., E. H., Keeling, H., Killeen, T. J., Laurance, W. F., Cruz, A. P., Pitman, N. C. A, Vargas, P. N., Ramirez-Angulo, H., Rudas, A., Salamao, R., Silva, N., Terborgh, J. & Torres-Lezama, A. 2009. Drought Sensitivity Of The Amazon Rainforest. Science 323:1344–1347.

Poorter, L., Van Der Sande, T., Thompson, J., Arets, E. J. M. M., Alarcón, A., Álvarez-Sánchez, J., Ascarrunz, N., Balvanera, P., Barajas-Guzmán, G., Boit, A., Bongers, F., Carvalho, F. A., Casanoves, F., Cornejo- Tenorio, G., Costa, F. R. C., De Castilho, C. V, Duivenvoorden, J. F., Dutrieux, L. P., Enquist, B. J., Fernández-Méndez, F., Finegan, B., Gormley, L. H. L., Healey, J. R., Hoosbeek, M. R., Ibarra-Marínquez, G., Junqueira, A. B., Levis, C., Licona, J. C., Lisboa, L. S., Magnusson, W. E., Martínez-Ramos, M., Martínez-Yrizar, A., Martorano, L. G., Masskell, L. C., Mazzei, L., Meave, J. A., Mora, F., Muñoz, R., Nytch, C., Pansonato, M. P., Parr, T. W., Paz, H., Simoes Penello, M., Pérez-Garcia, E. A., Rentería, L. Y., Rodríguez-Velazquez, J., Rosendaal, D. M. A., Ruschel, A. R., Sakschewski, B., Salgado Negret, B., Schietti, J., Sinclair, F. L., Souza, P. F., Souza, F. C., Stropp, J., Ter Steege, H., Swenson, N. G., Thonicke, K., Toledo, M., Uriarte, M., Van Der Hout, P., Walker, P., Zamora, N. & Peña-Claros, M. 2015. Diversity Enhances Carbon Storage In Tropical Forests. Global Ecology And Biogeography Accepted.

Poveda, G. 2004. La Hidroclimatología De Colombia: Una Síntesis Desde La Escala Inter-Decadal Hasta La Escala Diurna. Revista De La Academia Colombiana De Ciencias 28:201–222.

Poveda, G. & Mesa, O. J. 1996. Las Fases Extremas Del Fenómeno Enso (El Niño Y La Niña) Y Su Influencia Sobre La Hidrología De Colombia. Ingeniería Hidráulica En México Xi:31–37.

Prance, G. 1989. American Tropical Forests. Tropical Rain Forest Ecosystems-Biogeographical And Ecological Studies:99–132. Elsevier B.V.

Quesada, C. A., Phillips, O., Schwarz, M., Czimczik, C. I., Baker, T. R., Patiño, S., Fyllas, N. M., Hodnett, M. G., Herrera, R., Almeida, S., Alvarez Dávila, E., Arneth, A., Arroyo, L., Chao, K. J., Dezzeo, N., Erwin, T., Di Fiore, A., Higuchi, N., Honorio Coronado, E., Jimenez, E. M., Killeen, T., Lezama, A. T., Lloyd, G., López- González, G., Luizão, F. J., Malhi, Y., Monteagudo, A., Neill, D. A., Núñez Vargas, P., Paiva, R., Peacock, J., Peñuela, M. C., Peña Cruz, A., Pitman, N., Priante Filho, N., Prieto, A., Ramírez, H., Rudas, A., Salomão, R., Santos, A. J. B., Schmerler, J., Silva, N., Silveira, M., Vásquez, R., Vieira, I., Terborgh, J. & Lloyd, J. 2012. Basin-Wide Variations In Amazon Forest Structure And Function Are Mediated By Both Soils And Climate. Biogeosciences 9:2203–2246.

Restrepo, I. ., Aldana, A. M. & Stevenson, P. R. 2016. Dinámica De Bosques En Diferentes Escenarios De Tala Selectiva En El Magdalena Medio (Colombia). Colombia Forestal 19:71–83.

Ribeiro, G., Chambers, J., Peterson, C., Trumbore, S., Magnabosco Marra, D., Wirth, C., Cannon, J., Négron- Juárez, R. I., Lima, A., De Paula, E. V. C. M., Santos, J. & Higuchi, N. 2016. Mechanical Vulnerability And Resistance To Snapping And Uprooting For Central Amazon Tree Species. Forest Ecology And Management 380:1–10. Elsevier B.V.

Rivera, S. V. & Harb, A. 2014. Análisis De Costos De Oportunidad De La Iniciativa De Implementación Temprana Redd En El Sector Güejar-Cafre.

Rodríguez Eraso, N., Armenteras-Pascual, D. & Alumbreros, J. R. 2013. Land Use And Land Cover Change In The Colombian Andes: Dynamics And Future Scenarios. Journal Of Land Use Science 8:154–174. Jour, .

Romero Ruíz, M., García, G., García, G. O., I Pascual, J. A. & Ruíz, D. M. R. 2004. Ecosistemas De La Cuenca Del Orinoco Colombiano. Book, .

Dos Santos, L. T., Magnabosco Marra, D., Trumbore, S., Camargo, P. B., Chambers, J. Q., Negrón-Juárez, R. I.,

111

Lima, A. J. N., Ribeiro, G. H. P. M., Dos Santos, J. & Higuchi, N. 2015. Windthrows Increase Soil Carbon Stocks In A Central Amazon Forest. Biogeosciences Discussions 12:19351–19372.

Schnitzer, S. A., Dewalt, S. J. & Chave, J. 2006. Censusing And Measuring Lianas: A Quantitative Comparison Of The Common Methods. Biotropica 38:581–591.

Schöngart, J., Piedade, M. T. F., Wittmann, F., Junk, W. J. & Worbes, M. 2005. Wood Growth Patterns Of Macrolobium Acaciifolium (Benth.) Benth. (Fabaceae) In Amazonian Black-Water And White-Water Floodplain Forests. Oecologia 145:454–61.

Schöngart, J., Wittmann, F. & Worbes, M. 2010. Biomass And Net Primary Production Of Central Amazonian Floodplain Forests. Pp. 347–388in Junk, W. J., Piedade, M. T. ., Wittmann, F., Schöngart, J. & Parolin, P. (Eds.).Amazonian Floodplain Forests: Ecophysiology, Biodiversity And Sustainable Management. Cambridge University Press, Cambridge.

Sherman, R. E., Fahey, T. J., Martin, P. H. & Battles, J. J. 2012. Patterns Of Growth, Recruitment, Mortality And Biomass Across An Altitudinal Gradient In A Neotropical Montane Forest, Dominican Republic. Journal Of Tropical Ecology 28:483–495. Jour, .

Sierra, C. A., Del Valle, J. I., Orrego, S. A., Moreno, F. H., Harmon, M. E., Zapata, M., Colorado, G. J., Herrera, M. A., Lara, W., Restrepo, D. E., Berrouet, L. M., Loaiza, L. M. & Benjumea, J. F. 2007. Total Carbon Stocks In A Tropical Forest Landscape Of The Porce Region, Colombia. Forest Ecology And Management 243:299–309. Jour, Elsevier B.V.

Slik, J. W. F., Paoli, G., Mcguire, K., Amaral, I., Barroso, J., Bastian, M., Blanc, L., Bongers, F., Boundja, P., Clark, C., Collins, M., Dauby, G., Ding, Y., Doucet, J. L., Eler, E., Ferreira, L., Forshed, O., Fredriksson, G., Gillet, J. F., Harris, D., Leal, M., Laumonier, Y., Malhi, Y., Mansor, A., Martin, E., Miyamoto, K., Araujo- Murakami, A., Nagamasu, H., Nilus, R., Nurtjahya, E., Oliveira, Á., Onrizal, O., Parada-Gutierrez, A., Permana, A., Poorter, L., Poulsen, J., Ramirez-Angulo, H., Reitsma, J., Rovero, F., Rozak, A., Sheil, D., Silva-Espejo, J., Silveira, M., Spironelo, W., Ter Steege, H., Stevart, T., Navarro-Aguilar, G. E., Sunderland, T., Suzuki, E., Tang, J., Theilade, I., Van Der Heijden, G., Van Valkenburg, J., Van Do, T., Vilanova, E., Vos, V., Wich, S., Wöll, H., Yoneda, T., Zang, R., Zhang, M. G. & Zweifel, N. 2013. Large Trees Drive Forest Aboveground Biomass Variation In Moist Lowland Forests Across The Tropics. Global Ecology And Biogeography 22:1261–1271.

Stevenson, P. R. & Aldana, A. M. 2008. Potential Effects Of Ateline Extinction And Forest Fragmentation On Plant Diversity And Composition In The Western Orinoco Basin, Colombia. International Journal Of Primatology 29:365–377. Jour, .

Zanne, A. E., Lopez-Gonzalez, G., Coomes, D. A. A., Ilic, J., Jansen, S., Lewis, S. L. S. L., Miller, R. B. B., Swenson, N. G. G., Wiemann, M. C. C. & Chave, J. 2009. Global Wood Density Database. P. Dryad Digital Repository. . 33 Pp.

112

Appendices Appendix 3: Table with information on the sixteen forest plots used in this study.

Tree Delta Delta Forest AGB Mortality Recruitment Clay Plot code Region RGR turnover ECEC Large Fragmentation Spp type change rate rate % rate Trees Richness

QUIN_TF_1 8.27 2.08 3.60 1.20 1.51 9.36 8.94 -2 0.42 12

QUIN_TF_3 -6.42 1.79 2.14 0.98 0.35 8.40 2.93 -5 0.87 2

QUIN_TF_4 4.72 2.22 3.19 1.13 0.96 18.32 10.92 -1 0.86 7 Terra firme QUIN_TF_5 Magdalena 6.43 2.36 3.03 2.02 0.67 5.68 6.77 0 0.43 4

SAJU_TF_4 Valley 4.17 1.76 1.59 1.35 -0.18 20.56 5.10 -4 0.37 9

SAJU_TF_5 -8.16 2.55 0.91 1.21 -1.64 14.84 4.69 2 0.38 -2

SAJU_PI_1 -20.84 8.41 4.21 1.39 -4.20 47.16 19.45 -4 0.49 -5 Várzea SAJU_PI_2 -13.72 6.39 2.37 2.12 -4.03 46.42 23.07 -3 0.46 5

SAMA_TF_1 0.73 4.67 4.58 1.71 -0.09 21.88 5.94 1 0.17 -6

SAMA_TF_2 -0.77 3.93 3.44 1.77 -0.49 31.89 6.19 1 0.18 -4 Orinoco SAMA_TF_3 Terra firme 1.77 3.30 5.22 1.76 1.92 14.56 3.72 2 0.46 18 Basin TOMO_TF_1 1.04 1.87 2.10 1.13 0.23 30.64 3.92 2 0.27 -2

TOMO_TF_2 -0.91 3.03 3.26 1.38 0.24 22.73 4.13 2 0.29 1

113

TOMO_TF_3 2.54 1.69 2.63 1.40 0.94 18.33 3.43 1 0.30 2

TOMO_PI_4 0.07 2.45 1.92 0.49 -0.53 44.27 6.07 2 0.30 1 Igapó TOMO_PI_5 2.95 1.88 1.51 0.74 -0.37 25.32 4.31 2 0.29 2

114

FOREST FRAGMENTS OF THE ANDEAN PIEDMONT AS CARBON SINKS: SHORT-TERM GAIN OF ABOVE GROUND BIOMASS IN FRAGMENTS USED BY CATTLE RANCHES

Aldana, A. M., & Stevenson, P. R. 2016. Forest fragments of the Andean piedmont as carbon sinks: Short-term gain of above ground biomass in fragments used by cattle ranches. Tropical Conservation Science. Vol 9 (4): 1-9

115

Research Article

Tropical Conservation Science October-December 2016: 1–9 Forest fragments of the Andean piedmont ! The Author(s) 2016 Reprints and permissions: as carbon sinks: Short-term gain sagepub.com/journalsPermissions.nav DOI: 10.1177/1940082916667339 of above ground biomass in fragments trc.sagepub.com used by cattle ranches

Ana M. Aldana1 and Pablo R. Stevenson1

Abstract Some tropical countries, such as Colombia, are venturing into the international carbon market by means of avoided defor- estation programs, i.e., they seek carbon credits by not cutting down forests. Consequently, basic information about carbon storage in different forest types is urgently needed. We describe the tree community dynamics of forest fragments in cattle ranches of the western Orinoco basin, and their relationship to the forests’ ability to accumulate carbon. We re-sampled three 1-ha vegetation plots after six and seven years of establishment. We found that these forests have different population dynamics; smaller fragments have negative net population change, but the dynamics allow for an annual carbon accumulation 1 rate of around 0.57 Mg haÀ . This rate is similar to the rate reported for some mature Amazon forests and may be attributed to high soil fertility.

Keywords Forest dynamics, above ground biomass accumulation rate, wood specific gravity

Introduction These forests, if conserved and restored, could be useful for carbon trade (Saatchi et al., 2011). Research on land-use Carbon storage estimation in different ecosystems has changes and ecosystem services is still scarce (Etter, become a central subject in conservation science, because McAlpine, Wilson, Phinn, & Possingham, 2006; Rodrı´guez natural ecosystems can mitigate climate change effects by Eraso, Armenteras-Pascual, & Retana Alumbreros, 2012; acting as carbon sinks, limiting the availability of this Sa´ nchez-Cuervo, Aide, Clark, & Etter, 2012) and is needed greenhouse gas in the atmosphere (Ashton, Tyrrell, for planning conservation strategies in geographic areas Spalding, & Gentry, 2012; Pan, Birdsey, Phillips, & dedicated to agriculture and livestock. Jackson, 2013). In tropical forests, which account for Estimations of carbon accumulated in tropical forests 55% of the carbon stored in the world’s forests, 65% of depend on climate, soil type (since in cold climates and the carbon is stored as biomass in the roots, trunks, and anoxic conditions carbon usually accumulates in soils), leaves of woody plants (Pan et al., 2011). In Colombia, a and the structure of plant communities and their product- tropical country, avoided deforestation initiatives have ivity (Carvalhais et al., 2014; La¨ hteenoja, Ruokolainen, emerged during the last decade from a variety of stake- Schulman, & Oinonen, 2009). The amount of carbon in holders such as the agriculture and livestock industries, plants is divided in two components, above and below which hope to mitigate the damage of their own product- ive activities while offering the service of carbon storage to other productive sectors (Fedegan, 2006; Fedepalma, 1Departamento de Ciencias Biolo´gicas, Universidad de Los Andes, Bogota´, 2013). However, the impact of these initiatives is still Colombia unknown, because the carbon stocks in Colombia have Received 26 May 2016; Revised 27 June 2016; Accepted 12 July 2016 been estimated only for primary forests (Alvarez et al., Corresponding Author: 2012; Phillips et al., 2016; Sierra et al., 2007). Ana M. Aldana, Departamento de Ciencias Biolo´gicas, Universidad de Los High deforestation rates in Colombia have left vast areas Andes, Cra 1 No 18A-12 Bogota´ 111711, Colombia. of fragmented and secondary forests (Phillips et al., 2016). Email: [email protected]

Creative Commons CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 3.0 License (http://www.creativecommons.org/licenses/by-nc/3.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/ open-access-at-sage). 2 Tropical Conservation Science ground, which are estimated based on allometric equations cattle ranching in large areas. The original landscape (Gibbs, Brown, Niles, & Foley, 2007). However, these was mainly continuous forest from the foothills of the equations are more common for aboveground biomass Andes, with some natural savannas to the southeast. (AGB), using information on the diameter of trees at Land-use change estimations from the Colonia (Etter, breast height (DBH) and wood density (Alvarez et al., 2015; Etter, Sarmiento, & Romero, 2010) show that 2012; Chave et al., 2005). In some cases, equations can fragmentation has been occurring since the late 1600s, also include tree height (Feldpausch et al., 2012; resulting in a landscape of riparian forest fragments Goodman et al., 2013). The ecosystem service related to (Appendix 1). carbon storage from the atmosphere may also depend on The remaining forest in the area has been categorized as forest dynamics, because a forest in which the recruitment tropical moist forest (Romero Ruı´z, Garcı´a, Garcı´a, i of new trees is higher than mortality would show high Pascual, & Ruı´z, 2004). The mean temperature is 27.5C, accumulation rates (if variables such as wood density ranging from 19 to 35C (Stevenson, 2011). Annual rain- and tree height are kept constant) (Johnson et al., 2016). fall in the region is usually between 2,600 and 3,000 mm Cattle ranching is a productive activity involving and is seasonal, with two dry months with less than deforestation, the use of forest resources, fragmentation, 100 mm (usually January and February) (Stevenson, and edge effects associated with high mortality of large Castellanos, Corte´ s, & Link, 2008). During the last three trees (Laurance, Delamo´ nica, Laurance, Vasconcelos, & decades cattle ranches have been increasingly replaced Lovejoy, 2000), which are the main drivers of AGB with oil palm plantations, making the Meta Department stocks (Slik et al., 2013). Fragments, especially small or the most extensive producer in Colombia (Wagner, Castro, narrow forest fragments, have a strong relationship & Stevenson, 2007). It has been observed that when grass- between perimeter and area, and thus are susceptible to lands are free from human intervention, a secondary forest large edge effects (Nascimento & Laurance, 2004). Forest is established (Stevenson, 2011). fragments show drastic changes in plant species compos- We re-sampled three forest plots (1 ha) in forest frag- ition, because there is high recruitment of pioneer species ments in three cattle ranches in San Martı´ n (Appendix 1). near edges (Laurance et al., 2006). As pioneer trees are We established the plots during the first semester of 2005 characterized by low wood density and usually short stat- (Stevenson & Aldana, 2008) and the second semester of ure (Muller-Landau, 2004), it is expected that forest frag- 2009. Fragment 1 has an approximate area of 28 ha, and ments should have greater abundance of low wood the 1-ha plot (336030.4000N, 7338020.7200W) was re- density species. In addition, the mortality of large trees sampled six years after establishment (second semester caused by more wind exposure and changes in abiotic of 2011); Fragment 2 has an approximate area of 23 ha, conditions should exacerbate the pattern (Laurance and the 1-ha plot (337042.0200N, 7338019.8600W) was re- et al., 2011). sampled seven years after establishment (second semester Our main objective is to describe the tree community of 2012); Fragment 3 is the biggest remaining fragment in dynamics of forest fragments in cattle ranches of the the municipality, 1,000 ha in area; the 1-ha plot Orinoco Basin, located in the Andean piedmont, and to (334054.900N, 7327014.700W) was re-sampled six years determine their relationship to carbon accumulation. after establishment (second semester of 2015). Specifically, we want to: 1. Determine growth, mortality, During the re-sampling of the plots we measured the and recruitment rates of woody plant species in these increased DBH of trees greater than 10 cm, recording the forest fragments; 2. Quantify the variation in wood dens- number of dead individuals and new recruits. We mea- ity of tree species in these fragments; and 3. Estimate the sured height of 76% of the individuals in our database, carbon stocks and carbon accumulation rates of the using a laser clinometer. We collected wood cores to forest fragments. Considering the predictable changes in determine wood density, here measured as Wood the tree community, we expected to find a low estimate of Specific Gravity (WSG sensu, Williamson & Wiemann, carbon storage in the forest fragments and a negative 2010), of at least one individual from the most abundant balance of carbon accumulation over time. species, following the protocol proposed by Chave (2005). Green volume was calculated using the dimensional method, and wood cores were oven dried at 103C for Methods 72 hours. Field study Data analysis The municipality of San Martı´ n (Meta, Colombia) is one of the oldest cities in the region, founded in 1585. We compared the census data from 2005 (Stevenson The surrounding region has since been used for extense & Aldana, 2008) and 2009 with census data from Aldana and Stevenson 3

2011 – 2015 to determine annual growth, mortality, and fragments 1 and 2 was lower than fragment 3, resulting recruitment rates for each plot using the formulae pre- in a negative change in stem densities. Only fragment 3 sented by Sherman and collaborators (Sherman, Fahey, showed a positive change in stem density, although rela- Martin, & Battles, 2012). tive tree growth rate was lower in fragment 3 than frag- We calculated above ground biomass (AGB) for each ments 1 and 2. sampling year, using the allometric equations for Tropical Moist Forest of the type I, which take into account DBH, Wood Specific Gravity height and wood density. We used the best predictive model for neotropical moist forests (Type I.3) (Alvarez The distribution and mean values of Wood Specific et al., 2012). Biomass calculations for other woody life Gravity for woody plant species differ among forest forms were done using different allometric equations to fragments (Figure 1). The analysis showed differences minimize errors; for palms we calculated the biomass among fragments (ANOVA, F 0.26, P < 0.01) and using the equations proposed by Goodman et al. (2013), a Tukey test showed that mean WSG¼ is lower in frag- and for lianas we calculated the biomass using the equa- ment 1. tion proposed by Schnitzer, DeWalt, and Chave (2006). For cases where the WSG was not measured by us Above Ground Biomass Change (33% of species), we used information on species mea- sured by our research group, following the same proto- The estimations of the rate of biomass change were posi- cols (Casas et al., 2016), and values reported in the Dryad tive for fragments 1 and 3 and negative for fragment 2 global database (Chave et al., 2009; Zanne et al., 2009). (Figure 2). These forests fragments, on average, are accu- For species with no reported WSG values in either of the mulating 0.57 Mg of AGB per hectare per year. Sierra databases, we used the mean for the genus or family et al. (2007) calculated that carbon biomass content in (using a combined database with the three sources men- moist tropical forests is around 45%, from which we esti- tioned previously), assuming that this trait is phylogenet- mate that, on average, each hectare of these forest frag- ically conserved (Chave et al., 2006). To evaluate whether ments is accumulating 0.26 Mg of carbon every year. mean WSG was different among plots we performed an ANOVA and a Tukey test, using the information from Growth rate and Wood Specific Gravity the year of re-sampling. The net change in AGB was calculated as the differ- Given the surprising result of positive AGB change and ence between the AGB in 2011 – 2015 and the AGB in negative stem density change in fragments 1, we explored 2005–2009, divided by the number of years between each the possibility of the mortality rate being compensated by sampling. We compared the results among plots and the rapid growth of some tree species. A negative rela- between sampling years with a t-test. tionship between growth rates and wood density is expected in contexts of r and K strategies. Although we did not find this relationship, we found high WSG values Results (WSG > 0.6) for some species with high relative growth Population Dynamics rate (RGR) (RGR > 2%). These results are summarized in Figure 3, which was drawn with species that had more The three forest fragments have different mortality than two individuals with observations of RGR in our and recruitment rates (Table 1). Recruitment rate for database.

Table 1. Community dynamics of forest fragments: Recruitment, Discussion mortality, net population change and average relative growth rates of the tree communities of three forest plots (1ha) in fragments Population Dynamics located in cattle ranches of the Andean piedmont in Meta, Colombia. We found that the smaller fragments have a negative change in stem densities, due to a mortality rate higher Mortality Recruitment Net Change Average than the recruitment rate. A higher mortality rate can be 1 1 1 Fragment (% yÀ ) (% yÀ ) (% yÀ ) RGR attributed to selective logging that has been occurring in Fragment 1 4.67 4.49 –0.18 2.56 these fragments in past years. Similarly, a lower recruitment Fragment 2 3.93 3.39 –0.54 2.36 rate may be due to intensive use by the cattle that go daily into the forest to drink water from the streams within these Fragment 3 3.13 5.19 2.07 1.80 fragments. Although we do not have measurements of the 4 Tropical Conservation Science

Figure 1. Histogram of the distribution of wood specific gravity values for tree species in three forest fragments of the Andean pied- mont in Meta, Colombia. Values were assigned to each individual stem from a species average calculated based on the measurements made by us in the present study and complemented with information from unpublished data (Casas et al., 2016). For species missing in our database (10%) we used information from Global Data Bases (Zanne et al., 2009). Dotted lines represent the mean values for each fragment. effect of cattle in these forests, cattle presence in the This idea is supported by a positive change in stem forest may cause high seedling mortality and soil compac- densities for fragment 3, which is isolated from cattle tion (Herna´ ndez-Vargas, Sa´ nchez-Vela´ squez, Carmona- presence. Also, the negative effect of selective logging Valdovinos, & Cuevas-Guzma´ n, 2000). may be attenuated by the size of the fragment. Aldana and Stevenson 5

Figure 2. Above Ground Biomass change in three forest plots (1-ha) in fragments located in cattle ranches of the Andean piedmont in Meta, Colombia, after 6 and 7 years of establishment.

Figure 3. Relationship between Relative Growth Rate and Wood Specific Gravity for 63 tree species (species in our data set with more than 2 observations), found in forest fragments located in cattle ranches of the Andean piedmont in Meta, Colombia. Species colored in red may be of interest for native wood production and ecosystem restoration (WSG>0.6, RGR>2%). Species names have been shortened for easier reading, species names from top to bottom are: Eschweilera cabrerana (cabo de hacha), Hieronyma oblonga (candelo), Hirtella americana (moradito), Garcinia madruno (madron˜o), Eugenia sp., Alchornea discolor, Licania subarachnophylla.

Wood Specific Gravity shared, which allows for a higher frequency of species with higher WSG values (Figure 1). Fragment 3 is the We found that fragment 1 has a lower WSG mean value, most diverse in species (127) and it is also the fragment which is possibly an effect of selective logging and low with the greatest frequency of species with high WSG species richness (67 species per hectare). Fragment 2 is values. This result is consistent with the reduction in frag- not richer in species (69), but some species are not ment size. However, in our study system selective logging 6 Tropical Conservation Science may have eliminated individuals of some old-growth spe- subarachnophylla could be considered promising for res- cies from the fragments, specially fragment 1 (Stevenson toration and reforestation programs in the Andean pied- & Aldana, 2008). mont, as these are native species that grow fast and could enhance the ecosystem services of the forest by accumu- Biomass Stocks and Above Ground Biomass Change lating more carbon in their trunks (given their wood density). Comparing our results with previous estimations, based on extrapolations (Phillips et al., 2011), we found that Implications for conservation these forests have lower biomass stocks than previously estimated. Although AGB stocks in these forest fragments are lower We found that fragment 2 had a negative change in than in Amazonian primary forests (Baraloto et al., 2013), AGB which maybe a consequence of a negative change in we found that these forests are capable of accumulating at stem density; opposite to this pattern, fragment 1 had a least 0.26 Mg of carbon every year, very similar to many negative stem density change, but a positive rate of AGB primary forests in the Neotropics (Phillips, Lewis, Baker, accumulation, which may be due to a high RGR (see Chao, & Higuchi, 2008). We believe this accumulation rate below). Although fragment 3 is more conserved than is due to fertile soils and could be enhanced by reducing the other two fragments, it has not reached a neutral selective logging and cattle use of the forest. If well man- carbon balance as have primary forests of the Amazon aged and protected, these fragments could be used by cattle (Phillips et al., 1998) and it accumulated more AGB per ranchers, oil palm producers, and other productive sectors year than the other fragments. In the tropics, the Amazon of the region either to compensate for the impacts of their region has been subject of many studies of AGB accumu- activities or to receive payment for ecosystem services that lation, which show that Amazon primary forests tend to enhance forest cover and connectivity between forest frag- be on an equilibrium state in terms of carbon storage ments. Connectivity among these forest fragments would (Herna´ ndez-Vargas et al., 2000). Therefore, this result not only benefit tree species populations, but also popula- can be explained by past removal of large trees for tions of big mammals like Panthera onca and other native domestic wood use, which allows for growth and recruit- fauna. We found some native tree species that have high ment of competing trees. Fragments 1 and 2 also have a values of WSG and are faster growing than other native higher RGR than fragment 3, which means that some species; these species should be studied for their potential tree species in these fragments may have high growth for timber production, and if successful, may be used to rates and high WSG values (see below). replace Eucalyptus pellita (eucalipto – Red Mahogany) Because these three forest plots may not be represen- and Pinus caribea (pino - Caribbean Pine). Plantations of tative of all the forest fragments in the Orinoco Basin, the the latter species are replacing natural forest and savanna results of this study should be compared with other stu- ecosystems in the region. dies from the same region, in order to reach sound con- clusions about the carbon dynamics of forests fragments Appendix 1 of the region. Detail of the study area in the municipality of San Martı´n Growth rate and Wood Specific Gravity: Promising (Meta, Colombia) in the western Orinoco Basin. Yellow Tree Species squares show the location of the three 1-ha vegetation plots in the forest fragments. Areas shaded in orange The tree species we highlight in Figure 3: Eschweilera are oil palm plantations, which may be confused with cabrerana (cabo de hacha), Hieronyma oblonga (can- forest fragments. Light green areas are pastures for cattle delo), Hirtella americana (moradito), Garcinia madruno ranching. Images were created using GoogleEarthPro in (madron˜o), Eugenia sp., Alchornea discolor, Licania June 2016. Aldana and Stevenson 7 8 Tropical Conservation Science

Acknowledgements Chave, J., Coomes, D., Jansen, S., Lewis, S. L., Swenson, N. G., & The Sa´nchez-Rey family and the Enciso family were very kind to Zanne, A. E. (2009). Towards a worldwide wood economics let us work on their premises; they also helped with fieldwork spectrum. Ecology Letters, 12, 351–366. logistics. Sasha Ca´rdenas, Marcela Co´rdoba, Indira Leo´n, Camila Chave, J., Muller-Landau, H. C., Baker, T. R., Easdale, T. A., Hans Monje, A´ ngela Perilla, Erika Rodrı´guez, Vanessa Rubio, Luis Steege, T. E. R., & Webb, C. O. (2006). Regional and phylo- Francisco Henao, Eduardo Pinel, Diana Acosta, Edna Beltra´n, genetic variation of wood density across 2456 neotropical tree Efraı´n Rinco´n, Camilo Quiroga, Maria Juliana Pardo, Felipe species. Ecological Applications, 16, 2356–2367. Aramburo and A´ ngela Sa´nchez helped with the re-sampling of Etter, A. (2015). La transformacio´n del uso de la tierra y los the plots and wood specific gravity measurements. We thank an ecosistemas durante el periodo colonial en Colombia anonymous reviewer who helped us improve the quality of the (1500–1800). La Economı´a Colonial de la Nueva Granada, manuscript. (pp. 1–44). Bogota´, Colombia: Fondo de Cultura Econo´mica; Banco de la Repu´blica. Etter, A., McAlpine, C., Wilson, K., Phinn, S., & Possingham, H. Declaration of Conflicting Interests (2006). Regional patterns of agricultural land use and deforest- ation in Colombia. Agriculture, Ecosystems & Environment, The author(s) declared no potential conflicts of interest with respect 114, 369–386. to the research, authorship, and/or publication of this article. Etter, A., Sarmiento, A., & Romero, M. (2010). Land use changes (1970–2020) and carbon emissions in the Colombian Llanos. In: M. Hill, & N. Hanna (Eds.). Ecosystem function in Funding Savannas Boca Raton, FL: CRC Press, pp. 383–402. The author(s) disclosed receipt of the following financial support Fedegan Plan Estrate´gico De La Ganaderı´a Colombiana 2019 for the research, authorship, and/or publication of this article: The Bogota´, Colombia: Author, pp. 1–296. authors received finantial support from Fondo de Investigaciones of Fedepalma Conservacio´n de la Biodiversidad en las Zonas de the Facultad de Ciencias at Universidad de Los Andes. AMA Cultivos de Palma. Bogota´, Colombia: Author. received support from the Lo´re´al – Unesco grant For Women in Feldpausch, T. R., Lloyd, J., Lewis, S. L., Brienen, R. J. W., Gloor, Science. M., Monteagudo Mendoza, A., ...; Phillips, O. (2012). Tree height integrated into pantropical forest biomass estimates. Biogeosciences, 9, 3381–3403. References Gibbs, H. K., Brown, S., Niles, J. O., & Foley, J. (2007). Alvarez, E., Duque, A., Saldarriaga, J., Cabrera, K., de las Salas, Monitoring and estimating tropical forest carbon stocks: G., del Valle, I., ...; Rodrı´guez, L. (2012). Tree above-ground Making REDD a reality. Environmental Research Letters, 2, biomass allometries for carbon stocks estimation in the natural 045023. forests of Colombia. Forest Ecology and Management, 267, Goodman, R. C., Phillips, O. L., del Castillo Torres, D., Freitas, L., 297–308. Cortese, S. T., Monteagudo, A., ...; Baker, T. R. (2013). Ashton, M., Tyrrell, M., Spalding, D., & Gentry, B. (2012). Amazon palm biomass and allometry. Forest Ecology and Managing forest carbon in a changing climate (M. S. Ashton, Management, 310, 994–1004. M. L. Tyrrell, D. Spalding, & B. Gentry Eds.). Dordrecht, Herna´ndez-Vargas, G., Sa´nchez-Vela´squez, L., Carmona- Netherlands: Springer Netherlands. Valdovinos, T., & Cuevas-Guzma´n, R. (2000). Efecto de la Baraloto, C., Molto, Q., Rabaud, S., He´rault, B., Valencia, R., ganaderı´a extensiva sobre la regeneracio´n arbo´rea de los bos- Blanc, L., ...; Thompson, J. (2013). Rapid simultaneous ques de la Sierra de Manantla´n. Madera y Bosques, 6, 13–28. estimation of aboveground biomass and tree diversity across Johnson, M. O., Galbraith, D., Gloor, E., De Deurwaerder, H., neotropical forests: A comparison of field inventory methods. Guimberteau, M., Rammig, A., ..., Baker, T. R. (2016). Biotropica, 45, 288–298. Variation in stem mortality rates determines patterns of above- Carvalhais, N., Forkel, M., Khomik, M., Bellarby, J., Jung, M., ground biomass in Amazonian forests: Implications for dynamic Migliavacca, M., ...; Reichstein, M. (2014). Global covariation global vegetation models. Global Change Biology. Advance of carbon turnover times with climate in terrestrial ecosystems. online publication. doi:10.1111/gcb.13315. Nature, 514, 213–217. La¨hteenoja, O., Ruokolainen, K., Schulman, L., & Oinonen, M. Casas, L., Aldana, A.M., Henao-Diaz, F., Villanueva, B., & (2009). Amazonian peatlands: An ignored C sink and potential Stevenson, P. R. (2016). Specific gravity of woody tissue from source. Global Change Biology, 15, 2311–2320. lowland Neotropical plants: Variance among forest types. Laurance, W. F., Camargo, J. L. C., Luiza˜o, R. C. C., Laurance, S. Manuscript submitted for publication. G., Pimm, S. L., Bruna, E. M., ...; Vasconcelos, H. L. (2011). Chave, J. (2005). Measuring wood density for tropical forest trees: The fate of Amazonian forest fragments: A 32-year investiga- A field manual for the CTFS sites. Pan-Amazonia. Retrieved tion. Biological Conservation, 144, 56–67. from http://www.rainfor.net/upload/ManualsEnglish/wood_ Laurance, W. F., Delamo´nica, P., Laurance, S. G., Vasconcelos, H. density_english[1].pdf. L., & Lovejoy, T. E. (2000). Rainforest fragmentation. Nature, Chave, J., Andalo, C., Brown, S., Cairns, M. A., Chambers, J. Q., 404, 836. Eamus, D., ...; Yamakura, T. (2005). Tree allometry and Laurance, W. F., Nascimento, H. E. M., Laurance, S. G., Andrade, improved estimation of carbon stocks and balance in tropical A., Ribeiro, J. E. L., Giraldo, J. P., ...; Angelo, S (2006). Rapid forests. Oecologia, 145, 87–99. decay of tree-community composition in Amazonian forest Aldana and Stevenson 9

fragments. Proceedings of the National Academy of Sciences, carbon stocks in tropical regions across three continents. 103, 19010–19014. Proceedings of the National Academy of Sciences of the Muller-Landau, H. C. (2004). Interspecific and inter-site variation United States of America, 108, 9899–9904. in wood specific gravity of tropical trees. Biotropica, 36, 20–32. Sa´nchez-Cuervo, A. M., Aide, T. M., Clark, M. L., & Etter, A. Nascimento, H. E. M., & Laurance, W. F. (2004). Biomass dynam- (2012). Land cover change in Colombia: Surprising for- ics in Amazonian forest fragments. Ecological Applications, 14, est recovery trends between 2001 and 2010. PLoS ONE, 7, 127–38. e43943. Pan, Y., Birdsey, R. A., Fang, J., Houghton, R., Kauppi, P. E., Schnitzer, S. A., DeWalt, S. J., & Chave, J. (2006). Censusing and Kurz, W. A., ...; Hayes, D. (2011). A large and persistent measuring lianas: A quantitative comparison of the common carbon sink in the world’s forests. Science, 333, 988–993. methods. Biotropica, 38, 581–591. Pan, Y., Birdsey, R. A., Phillips, O. L., & Jackson, R. B. (2013). Sherman, R. E., Fahey, T. J., Martin, P. H., & Battles, J. J. (2012). The structure, distribution, and biomass of the World’s forests. Patterns of growth, recruitment, mortality and biomass across an Annual Review of Ecology, Evolution, and Systematics, 44, altitudinal gradient in a Neotropical Montane forest, Dominican 593–622. Republic. Journal of Tropical Ecology, 28, 483–495. Phillips, J. F., Duque, A., Cabrera, K., Navarrete, D. A., Garcia, M. Sierra, C. A., del Valle, J. I., Orrego, S. A., Moreno, F. H., Harmon, C., Alvarez, E., ...; Vargas, D. M. (2011). Estimacio´n de las M. E., Zapata, M., ...; Benjumea, J. F. (2007). Total car- Reservas Potenciales de Carbono Almacenadas en la Biomasa bon stocks in a tropical forest landscape of the Porce region, Ae´rea en Bosques Naturales de Colombia. Bogota´, Colombia: Colombia. Forest Ecology and Management, 243, 299–309. Instituto de Hidrologı´a, Meteorologı´a, y Estudios Ambientales- Slik, J. W. F., Paoli, G., Mcguire, K., Amaral, I., Barroso, J., IDEAM. Bastian, M., ...; Zweifel, N. (2013). Large trees drive forest Phillips, J. F., Duque, A´ ., Scott, C., Wayson, C., Galindo, G., aboveground biomass variation in moist lowland forests across Cabrera, E., ...; Yepes, A. (2016). Live aboveground carbon the tropics. Global Ecology and Biogeography, 22, 1261–1271. stocks in natural forests of Colombia. Forest Ecology and Stevenson, P. R. (2011). The abundance of large ateline monkeys is Management, 374, 119–128. positively associated with the diversity of plants regenerating in Phillips, O., Lewis, S. L., Baker, T. R., Chao, K.-J., & Higuchi, N. neotropical forests. Biotropica, 43, 512–519. (2008). The changing Amazon forest. Philosophical Stevenson, P. R., & Aldana, A. M. (2008). Potential effects of Transactions of the Royal Society B: Biological Sciences, 363, ateline extinction and forest fragmentation on plant diversity 1819–1827. and composition in the Western Orinoco Basin, Colombia. Phillips, O., Malhi, Y., Higuchi, N., Laurance, W. F., Vargas, P. N., International Journal of Primatology, 29, 365–377. Va´squez, R., ...; Grace, J. (1998). Changes in the carbon bal- Stevenson, P. R., Castellanos, M. C., Corte´s, A. I., & Link, A. ance of tropical forests: Evidence from long-term plots. Science, (2008). Flowering patterns in a seasonal tropical lowland 282, 439–442. forest in Western Amazonia. Biotropica, 40, 559–567. Rodrı´guez Eraso, N., Armenteras-Pascual, D., & Retana Wagner, M., Castro, F., & Stevenson, P. R. (2007). Habitat char- Alumbreros, J. (2012). Land use and land cover change in the acterization and population status of the Dusky Titi (C allicebus Colombian Andes: Dynamics and future scenarios. Journal of ornatus) in fragmented forests, Meta, Colombia. Neotropical Land Use Science, 8, 154–174. Primates, 16, 621–627. Romero Ruı´z, M., Garcı´a, G., Garcı´a, G. O., i Pascual, J. A., & Williamson, G. B., & Wiemann, M. C. (2010). Measuring wood Ruı´z, D. M. R. (2004). Ecosistemas de la cuenca del Orinoco specific gravity correctly. American Journal of Botany, 97, colombiano (p. 187). Bogota´, Colombia: Instituto de 519–524. Investigacio´n de Recursos Biolo´gicos Alexander von Zanne, A. E., Lopez-Gonzalez, G., Coomes, D. A. A., Ilic, J., Humboldt (IAVH); Instituto Geogra´fico Agustı´n Codazzi. Jansen, S., Lewis, S. L., ..., Chave, J. (2009). Global wood Saatchi, S. S., Harris, N. L., Brown, S., Lefsky, M., Mitchard, E. T. density database. Dryad Digital Repository. Retrieved from A., Salas, W., ...; Morel, A. (2011). Benchmark map of forest http://dx.doi.org/10.5061/dryad.234.

CONCLUSIONES GENERALES Debido a que los bosques húmedos de Colombia se distribuyen en terrenos heterogéneos que tienen diversas historias geológicas y presiones ambientales, éstos se diferencian en cuanto a la composición florística, diversidad filogenética, reservas de biomasa y dinámicas del recambio de árboles.

Los bosques húmedos de tierra firme que están geográficamente cercanos tienden a tener las mismas condiciones ambientales, lo que hace que en términos de riqueza florística y filogenética éstos sean bastante similares entre sí; sin embargo, la historia biogeográfica de las regiones donde se distribuyen éstos bosques tiene un efecto sobre la composición florística y la estructura que a su vez ejerce un efecto sobre las reservas de biomasa y la dinámica de la comunidad. De esta manera, los bosques húmedos de tierra firme de las cuencas del Amazonas, del Magdalena y del Orinoco, en Colombia, se diferencian en la magnitud del servicio ecosistémico de acumulación de carbono que prestan a la humanidad. Igualmente, se evidencian los efectos negativos de las actividades humanas sobre éstos bosques y el servicio ambiental, particularmente la fragmentación, la tala selectiva y ganadería.

Los efectos de la inundación estacional son muy fuertes sobre la riqueza y la diversidad filogenética; los bosques estacionalmente inundables son menos diversos que los bosques de tierra firme. Adicionalmente, se evidencia el efecto de las características del suelo, que son afectadas por el tipo de rio que inunda los bosques; bosques que son estacionalmente inundados por ríos de origen andino (várzeas), con altas cargas sedimentarias, tienen suelos más ricos que facilitan el acelerado crecimiento de los árboles, que éstos alcancen mayor estatura y tamaño, y mayor dinámica de los individuos; lo que, a su vez, posibilita que los bosques de várzea tengan mayores reservas de carbono. Los bosques estacionalmente inundados por ríos originados en los bosques de tierras bajas de la Amazonía y Orinoquía (igapó), con bajos contenidos de nutrientes, tienen suelos pobres que limitan las tasas de crecimiento de los árboles, su estatura y tamaño. Por esta razón, estos

125

bosques, a pesar de tener especies que tienen densidades de madera altas tienen menores reservas de carbono que otros bosques en la misma region.

Los eventos climáticos extremos como las fases intensas del ENSO, tienen efectos negativos, principalmente sobre los bosques de la cuenca del Magdalena. En esta región, los aumentos en la precipitación causados por los eventos de La Niña generan aumento en los niveles del río Magdalena y sus afluentes, lo que promueve en la mortalidad de árboles en bosques de las zonas bajas. Se evidencia la importancia del monitoreo a largo plazo de los bosques, con el fin de prever los efectos que tendrá en cambio climático global, que tiende a aumentar la frecuencia de los eventos climáticos extremos.

Los resultados presentados en éste trabajo pueden no ser representativos de todos los bosques húmedos de Colombia, dado que la información proviene de unas pocas parcelas permanentes. Sin embargo, se encontraron diferencias marcadas entre tipos de bosque y entre regiones, lo que debería ser un indicativo de la necesidad de establecer y monitorear más parcelas permanentes de vegetación en otras regiones del país.

En cuanto a la beta-diversidad filigenética y de especies, las treintaydos parcelas que se estudiaron no estaban distribuidas uniformemente en las regiones, por lo que no sería sorprendente encontrar que, por ejemplo, dentro de la cuenca del Orinoco, los bosques ubicados en la transición Orinoquía- Amazonía sean más similares a los bosques de la cuenca del Amazonas que a los de la cuenca de Orinoco en cuanto a diversidad florística y filogenética.

Estudiando las reservas de biomasa con una buena cantidad de parcelas permanentes (206) distribuidas a lo largo de una gradianete climático, se hizo evidente la importancia del clima. Este hallazgo puede ser extrapolado a todo el país, y debería tenerse en cuenta en la planeación de futuros estudios y estrategias de conservación relacionadas con la provisión de servicios ecosistémicos.

Considerando las limitaciones de un monitoreo de largo plazo en pocas parcelas (16), no es posible hacer conclusiones robustas sobre las variables que mayor control tienen sobre las dynamicas del bosque y la biomasa aérea

126

en la región. No obstante, este es un punto de inicio y una referencia para futuros programas de monitoreo.

127

GENERAL CONCLUSIONS Because the rainforests of Colombia are distributed along heterogeneous topographies, with diverse geologic histories and environmental pressures, these are different in terms of floristic composition, phylogenetic diversity, biomass stocks and tree turnover dynamics.

Terra firme wet forests that are geographically close, have similar environmental conditions, which makes them also very similar in floristic and phylogenetic richness; however, the biogeographic history of the regions where the forests are located at, has an effect on the floristic composition and structure, that in turn affects the forests´ biomass stocks and the tree turnover dynamics. This way, the terra firme rainforests of the Amazon, Magdalena and Orinoco basins differ in the magnitude of the carbon accumulation ecosystem service they provide to humanity. Similarly, the negative effects of the human activities on the forests and the environmental service are evidenced, particularly fragmentation, selective logging and cattle ranching.

The seasonal flood is a very strong force controlling tree species richness and phylogenetic diversity; seasonally flooded forests are less diverse than terra firme forests. Additionally, there is a strong effect of the soil characteristics, which in turn are affected by the type of river that inundates the forests. Seasonally flooded forests that are inundated by rivers of Andean origin (várzea), with high sediment loads, have more fertile soils that allow for accelerated tree growth, high stature and great size, with more turnover dynamics; thus, these várzeas have the highest carbon stocks. Seasonally flooded forests inundated by rivers originated within the lowlands of the Amazonian and Orinoco basins (igapó), with low nutrient content, have poor soils, which limit tree growth, stature and size. This is why, igapó forests, regardless of having tree species with denser woods, have the lowest carbon stocks when compared to terra firme forests of the same region.

Extreme climatic events, such as the ENSO phases, have negative effects, mainly in the forests of the Magdalena basin. In this region, the increase in precipitation caused by La Niña events generated increases in the water levels of the Magdalena river and it´s affluents which promoted tree mortality. There is and imperative need to continue with long term forest monitoring, that 128

can help forecast the effects of the global climate change, which has augmented the frequency of extreme climatic events.

The results summarized here may not be representative of all rainforests of Colombia, as the information comes from a few permanent vegetation plots. However, we found marked differences between forest types and basins, which should be an indicative for the need to establish and monitor more permanent vegetation plots in other regions.

Regarding the phylogenetic and species beta-diversity, the thirty-two plots studied are not evenly distributed across the basins, thus, it would not be surprising to see that, for example, within the Orinoco basin, forests of the Orinoco-Amazon transition may be more similar to forests of the Amazon than to the Orinoco in terms of floristic and phylogenetic diversity.

When examining above ground biomass stocks with a fair number of plots (206) distributed along a climatic gradient, there is an evident effect of the environment. This finding can be extrapolated to the whole country, and it is important to take it into account when planning future studies and conservation strategies involving the provision of ecosystem services.

Considering the limitations of a short-term monitoring of a few (16) plots, it is not possible to make more conclusive remarks on the most important variables controlling tree turnover and biomass dynamics in the region. However, this is an important starting point and a reference for future monitoring programs.

129

LITERATURA CITADA Ackerly, D. D., S. R. Loarie, W. K. Cornwell, S. B. Weiss, H. Hamilton, R. Branciforte, and N. J. B. Kraft. 2010. The geography of climate change: implications for conservation biogeography. Diversity and Distributions 16:476–487.

Alizadeh, K., M. Cohen, and H. Behling. 2015. Origin and dynamics of the northern South American coastal savanna belt during the Holocene – the role of climate, sea-level, fire and humans. Quaternary Science Reviews 122:51–62.

Alvarez, E., A. Duque, J. Saldarriaga, K. Cabrera, G. de las Salas, I. del Valle, A. Lema, F. Moreno, S. Orrego, and L. Rodríguez. 2012. Tree above-ground biomass allometries for carbon stocks estimation in the natural forests of Colombia. Forest Ecology and Management 267:297–308.

Antonelli, A., J. A. A. Nylander, C. Persson, and I. Sanmartín. 2009. Tracing the impact of the Andean uplift on Neotropical plant evolution. Proceedings of the National Academy of Sciences of the United States of America 106:9749–54.

Baker, T. R., R. T. Pennington, S. Magallon, E. Gloor, W. F. Laurance, M. Alexiades, E. Alvarez, A. Araujo, E. J. M. M. Arets, G. Aymard, A. A. de Oliveira, I. Amaral, L. Arroyo, D. Bonal, R. J. W. Brienen, J. Chave, K. G. Dexter, A. Di Fiore, E. Eler, T. R. Feldpausch, L. Ferreira, G. Lopez-Gonzalez, G. van der Heijden, N. Higuchi, E. Honorio, I. Huamantupa, T. J. Killeen, S. Laurance, C. Leaño, S. L. Lewis, Y. Malhi, B. S. Marimon, B. H. Marimon Junior, A. Monteagudo Mendoza, D. Neill, M. C. Peñuela-Mora, N. Pitman, A. Prieto, C. a. Quesada, F. Ramírez, H. Ramírez Angulo, A. Rudas, A. R. Ruschel, R. P. Salomão, A. S. de Andrade, J. N. M. Silva, M. Silveira, M. F. Simon, W. Spironello, H. ter Steege, J. Terborgh, M. Toledo, A. Torres-Lezama, R. Vasquez, I. C. G. Vieira, E. Vilanova, V. a. Vos, and O. L. Phillips. 2014. Fast demographic traits promote high diversification rates of Amazonian trees. Ecology letters 17:527–36.

Brienen, R. J. W., O. L. Phillips, T. R. Feldpausch, and Et Al. 2015. Long-term decline of the Amazon carbon sink. Nature 519:344–348.

Bunker, D. E., F. Declerck, J. C. Bradford, R. K. Colwell, I. Perfecto, O. L. Phillips, M. Sankaran, and S. Naeem. 2005. Species loss and aboveground carbon storage in a tropical forest. Science (New York, N.Y.) 310:1029–31.

Cavanaugh, K. C., J. S. Gosnell, S. L. Davis, J. Ahumada, P. Boundja, D. B. Clark, B. Mugerwa, P. a. Jansen, T. G. O’Brien, F. Rovero, D. Sheil, R. Vasquez, and S. Andelman. 2014. Carbon storage in tropical forests correlates with taxonomic diversity and functional dominance on a global scale. Global Ecology and Biogeography 23:563– 573.

Etter, A. 2013. La transformación del uso de la tierra y los ecosistemas durante el periodo colonial en Colombia (1500-1800). Pages 1–44La Economía colonial de la nueva granada.

Etter, A., C. McAlpine, K. Wilson, S. Phinn, and H. Possingham. 2006. Regional patterns of agricultural land use and deforestation in Colombia. Agriculture, Ecosystems & Environment 114:369–386.

FAO, and JRC. 2012. Global forest land-use change 1990-2005. Page FAO forestry paper. Rome.

Fauset, S., M. O. Johnson, M. Gloor, T. R. Baker, A. Monteagudo M., R. J. W. Brienen, T. R. Feldpausch, G. Lopez-Gonzalez, Y. Malhi, H. ter Steege, N. C. a. Pitman, C. Baraloto, J. Engel, P. Pétronelli, A. Andrade, J. L. C. Camargo, S. G. W. Laurance, W. F. Laurance, J. Chave, E. Allie, P. N. Vargas, J. W. Terborgh, K. Ruokolainen, M. Silveira, G. a. Aymard C., L. Arroyo, D. Bonal, H. Ramirez-Angulo, A. Araujo-Murakami, D. Neill, B. Hérault, A. Dourdain, A. Torres-Lezama, B. S. Marimon, R. P. Salomão, J. a. Comiskey, M. Réjou-Méchain, M. Toledo, J. C. Licona, A. Alarcón, A. Prieto, A. Rudas, 130

P. J. van der Meer, T. J. Killeen, B.-H. Marimon Junior, L. Poorter, R. G. a. Boot, B. Stergios, E. V. Torre, F. R. C. Costa, C. Levis, J. Schietti, P. Souza, N. Groot, E. Arets, V. C. Moscoso, W. Castro, E. N. H. Coronado, M. Peña-Claros, C. Stahl, J. Barroso, J. Talbot, I. C. G. Vieira, G. van der Heijden, R. Thomas, V. a. Vos, E. C. Almeida, E. Á. Davila, L. E. O. C. Aragão, T. L. Erwin, P. S. Morandi, E. A. de Oliveira, M. B. X. Valadão, R. J. Zagt, P. van der Hout, P. A. Loayza, J. J. Pipoly, O. Wang, M. Alexiades, C. E. Cerón, I. Huamantupa-Chuquimaco, A. Di Fiore, J. Peacock, N. C. P. Camacho, R. K. Umetsu, P. B. de Camargo, R. J. Burnham, R. Herrera, C. a. Quesada, J. Stropp, S. a. Vieira, M. Steininger, C. R. Rodríguez, Z. Restrepo, A. E. Muelbert, S. L. Lewis, G. C. Pickavance, and O. L. Phillips. 2015. Hyperdominance in Amazonian forest carbon cycling. Nature Communications 6:6857.

Feldpausch, T. R., O. L. Phillips, R. J. W. Brienen, E. Gloor, J. Lloyd, G. Lopez-Gonzalez, A. Monteagudo-Mendoza, Y. Malhi, A. Alarcón, E. Álvarez Dávila, P. Alvarez-Loayza, A. Andrade, L. E. O. C. Aragao, L. Arroyo, G. A. Aymard C., T. R. Baker, C. Baraloto, J. Barroso, D. Bonal, W. Castro, V. Chama, J. Chave, T. F. Domingues, S. Fauset, N. Groot, E. Honorio Coronado, S. Laurance, W. F. Laurance, S. L. Lewis, J. C. Licona, B. S. Marimon, B. H. Marimon-Junior, C. Mendoza Bautista, D. A. Neill, E. A. Oliveira, C. Oliveira dos Santos, N. C. Pallqui Camacho, G. Pardo-Molina, A. Prieto, C. A. Quesada, F. Ramírez, H. Ramírez-Angulo, M. Réjou-Méchain, A. Rudas, G. Saiz, R. P. Salomão, J. E. Silva-Espejo, M. Silveira, H. ter Steege, J. Stropp, J. Terborgh, R. Thomas-Caesar, G. M. F. van der Heijden, R. Vásquez Martinez, E. Vilanova, and V. A. Vos. 2016. Amazon forest response to repeated droughts. Global Biogeochemical Cycles 30:964– 982.

Gentry, A. 1982. Neotropical floristic diversity: phytogeographical connections between Central and South America, Pleistocene climatic fluctuations, or an accident of the Andean orogeny. Annals of the Missouri Botanical Garden 69:557–593.

Honorio Coronado, E. N., K. G. Dexter, R. T. Pennington, J. Chave, S. L. Lewis, M. N. Alexiades, E. Alvarez, A. Alves de Oliveira, I. L. Amaral, A. Araujo-Murakami, E. J. M. M. Arets, G. a. Aymard, C. Baraloto, D. Bonal, R. Brienen, C. Cerón, F. Cornejo Valverde, A. Di Fiore, W. Farfan-Rios, T. R. Feldpausch, N. Higuchi, I. Huamantupa-Chuquimaco, S. G. Laurance, W. F. Laurance, G. López-Gonzalez, B. S. Marimon, B. H. Marimon- Junior, A. Monteagudo Mendoza, D. Neill, W. Palacios Cuenca, M. C. Peñuela Mora, N. C. a. Pitman, A. Prieto, C. a. Quesada, H. Ramirez Angulo, A. Rudas, A. R. Ruschel, N. Salinas Revilla, R. P. Salomão, A. Segalin de Andrade, M. R. Silman, W. Spironello, H. ter Steege, J. Terborgh, M. Toledo, L. Valenzuela Gamarra, I. C. G. Vieira, E. Vilanova Torre, V. Vos, and O. L. Phillips. 2015. Phylogenetic diversity of Amazonian tree communities. Diversity and Distributions 21:1295–1307.

Lewis, W. M. J., S. K. Hamilton, and J. F. I. Saunders. 1995. Rivers of Northern South America. Pages 219–256in C. Cushing and K. Cummis, editors.Ecosystems of the World. Elsevier, New York, NY.

Ortega-P, S. C., A. Garcia-Guerrero, C. A. Ruíz, J. Sabogal, and J. D. Vargas, editors. 2010. Deforestación Evitada Una Guía REDD + Colombia. BOOK, Ministerio de Ambiente, Vivienda y Desarrollo Territorial; Conservación Internacional Colombia; Fondo Mundial para la Naturaleza (WWF); The Nature Conservancy; Corporación Ecoversa; Fundación Natura; Agencia de Cooperación Americana (USAID); Patrimonio Na, Bogotá.

Pan, Y., R. a. Birdsey, O. L. Phillips, and R. B. Jackson. 2013. The Structure, Distribution, and Biomass of the World’s Forests. Annual Review of Ecology, Evolution, and Systematics 44:593–622.

Peña, M. a., and A. Duque. 2013. Patterns of stocks of aboveground tree biomass, dynamics, and their determinants in secondary Andean forests. Forest Ecology and Management 302:54–61.

Pennington, R. T., and C. W. Dick. 2004. The role of immigrants in the assembly of the South American rainforest tree flora. Philosophical transactions of the Royal Society of

131

London. Series B, Biological sciences 359:1611–22.

Phillips, J. F., A. Duque, K. Cabrera, D. A. Navarrete, M. C. Garcia, E. Alvarez, E. Cabrera, D. Cárdenas, G. Galindo, M. F. Ordóñez, M. L. Rodríguez, and D. M. Vargas. 2011. Estimación de las Reservas Potenciales de Carbono Almacenadas en la Biomasa Aérea en Bosques Naturales de Colombia. BOOK, Instituto de Hidrología, Meteorología, y Estudios Ambientales-IDEAM, Bogotá.

Phillips, J. F., Á. Duque, C. Scott, C. Wayson, G. Galindo, E. Cabrera, J. Chave, M. Peña, E. Álvarez, D. Cárdenas, J. Duivenvoorden, P. Hildebrand, P. Stevenson, S. Ramírez, and A. Yepes. 2016. Live aboveground carbon stocks in natural forests of Colombia. Forest Ecology and Management 374:119–128.

Phillips, O., S. L. Lewis, T. R. Baker, K.-J. Chao, and N. Higuchi. 2008. The changing Amazon forest. Philosophical Transactions of the Royal Society B: Biological Sciences 363:1819–1827.

Prance, G. 1989. American tropical forests. Tropical rain forest ecosystems-biogeographical and ecological studies:99–132.

Quesada, C. a., O. Phillips, M. Schwarz, C. I. Czimczik, T. R. Baker, S. Patiño, N. M. Fyllas, M. G. Hodnett, R. Herrera, S. Almeida, E. Alvarez Dávila, A. Arneth, L. Arroyo, K. J. Chao, N. Dezzeo, T. Erwin, A. di Fiore, N. Higuchi, E. Honorio Coronado, E. M. Jimenez, T. Killeen, a. T. Lezama, G. Lloyd, G. López-González, F. J. Luizão, Y. Malhi, A. Monteagudo, D. a. Neill, P. Núñez Vargas, R. Paiva, J. Peacock, M. C. Peñuela, A. Peña Cruz, N. Pitman, N. Priante Filho, A. Prieto, H. Ramírez, A. Rudas, R. Salomão, a. J. B. Santos, J. Schmerler, N. Silva, M. Silveira, R. Vásquez, I. Vieira, J. Terborgh, and J. Lloyd. 2012. Basin-wide variations in Amazon forest structure and function are mediated by both soils and climate. Biogeosciences 9:2203–2246.

Sierra, C. A., J. I. del Valle, S. A. Orrego, F. H. Moreno, M. E. Harmon, M. Zapata, G. J. Colorado, M. A. Herrera, W. Lara, D. E. Restrepo, L. M. Berrouet, L. M. Loaiza, and J. F. Benjumea. 2007. Total carbon stocks in a tropical forest landscape of the Porce region, Colombia. Forest Ecology and Management 243:299–309.

Ter Steege, H., N. C. a Pitman, D. Sabatier, C. Baraloto, R. P. Salomão, J. E. Guevara, O. Phillips, C. V Castilho, W. E. Magnusson, J.-F. Molino, A. Monteagudo, P. Núñez Vargas, J. C. Montero, T. R. Feldpausch, E. N. H. Coronado, T. J. Killeen, B. Mostacedo, R. Vasquez, R. L. Assis, J. Terborgh, F. Wittmann, A. Andrade, W. F. Laurance, S. G. W. Laurance, B. S. Marimon, B.-H. Marimon, I. C. Guimarães Vieira, I. L. Amaral, R. Brienen, H. Castellanos, D. Cárdenas López, J. F. Duivenvoorden, H. F. Mogollón, F. D. D. A. Matos, N. Dávila, R. García-Villacorta, P. R. Stevenson Diaz, F. Costa, T. Emilio, C. Levis, J. Schietti, P. Souza, A. Alonso, F. Dallmeier, A. J. D. Montoya, M. T. Fernandez Piedade, A. Araujo-Murakami, L. Arroyo, R. Gribel, P. V. a Fine, C. a Peres, M. Toledo, G. a Aymard C, T. R. Baker, C. Cerón, J. Engel, T. W. Henkel, P. Maas, P. Petronelli, J. Stropp, C. E. Zartman, D. Daly, D. Neill, M. Silveira, M. R. Paredes, J. Chave, D. D. A. Lima Filho, P. M. Jørgensen, A. Fuentes, J. Schöngart, F. Cornejo Valverde, A. Di Fiore, E. M. Jimenez, M. C. Peñuela Mora, J. F. Phillips, G. Rivas, T. R. van Andel, P. von Hildebrand, B. Hoffman, E. L. Zent, Y. Malhi, A. Prieto, A. Rudas, A. R. Ruschell, N. Silva, V. Vos, S. Zent, A. a Oliveira, A. C. Schutz, T. Gonzales, M. Trindade Nascimento, H. Ramirez-Angulo, R. Sierra, M. Tirado, M. N. Umaña Medina, G. van der Heijden, C. I. a Vela, E. Vilanova Torre, C. Vriesendorp, O. Wang, K. R. Young, C. Baider, H. Balslev, C. Ferreira, I. Mesones, A. Torres-Lezama, L. E. Urrego Giraldo, R. Zagt, M. N. Alexiades, L. Hernandez, I. Huamantupa-Chuquimaco, W. Milliken, W. Palacios Cuenca, D. Pauletto, E. Valderrama Sandoval, L. Valenzuela Gamarra, K. G. Dexter, K. Feeley, G. Lopez-Gonzalez, and M. R. Silman. 2013. Hyperdominance in the Amazonian tree flora. Science (New York, N.Y.) 342:1243092.

Stevenson, P. R., M. Suescún, A. M. Aldana, Á. Cano, M. N. Umaña, D. F. Correa-Gómez, L. F. Casas Caro, and B. Villanueva. 2011. Diversidad arbórea en bosques de tierras bajas en Colombia : efectos del ambiente , las perturbaciones y la geografía. Hipotesis:29–35.

132

Werner, G. D. a., W. K. Cornwell, J. I. Sprent, J. Kattge, and E. T. Kiers. 2014. A single evolutionary innovation drives the deep evolution of symbiotic N2-fixation in angiosperms. Nature Communications 5:1–9.

Whitfeld, T. J. S., V. Novotny, S. E. Miller, J. Hrcek, P. Klimes, and G. D. Weiblen. 2012. Predicting tropical insect herbivore abundance from host plant traits and phylogeny. Ecology 93:S211–S222.

133

ANEXOS

134

ANEXO 1 – Dinámica de bosques en diferentes escenarios de tala selectiva en el magdalena medio (colombia).

Restrepo, I.C., Aldana, A. M. & Stevenson, P. R. 2016. Dinámica de bosques en diferentes escenarios de tala selectiva en el magdalena medio (colombia). Colombia Forestal Vol. 19 (2):71–83.

135

Publicación de la Facultad del Medio Ambiente y Recursos Naturales - Proyecto Curricular de Ingeniería Forestal UNIVERSIDAD DISTRITAL FRANCISCO JOSÉ DE CALDAS revistas.udistrital.edu.co/ojs/index.php/colfor/index

ARTÍCULO DE INVESTIGACIÓN DINÁMICA DE BOSQUES EN DIFERENTES ESCENARIOS DE TALA SELECTIVA EN EL MAGDALENA MEDIO (COLOMBIA)

Forest dynamics under different levels of selective logging in the middle Magdalena River (Colombia)

Isabel C. Restrepo1, Ana M. Aldana2 & Pablo R. Stevenson3

Restrepo, I.C, Aldana, A.M. & Stevenson, P.R. (2016). Dinámica de bosques en diferentes escenarios de tala se- lectiva en el Magdalena medio (Colombia). Colombia Forestal, 19(2), 71-83

Recepción: 31 de agosto de 2015 Aprobación: 8 de febrero de 2016

Resumen pendientes de la zona provocan una elevada mor- La tala selectiva es una actividad de uso forestal uti- talidad y promueven la formación de claros tanto lizada con frecuencia, la cual ha demostrado tener en bosques intervenidos como en los que han sido un menor impacto sobre la biodiversidad que la tala poco alterados. generalizada. Sin embargo, tanto la magnitud como Palabras clave: Biomasa aérea, bosque primario, la dirección del cambio ecológico después de la composición florística, demografía, serranía de las tala dependen de su intensidad y de la subsecuente Quinchas. dinámica del bosque. Por esto, es importante reali- zar estudios que permitan comprender el funciona- Abstract miento de diferentes ecosistemas tras practicar tala Selective logging is a growing forestry activity with selectiva. En este estudio se analizó la dinámica del less impact on biodiversity than complete defores- bosque de la reserva El Paujil (Magdalena medio, tation. Despite this, both the magnitude and the di- Colombia) en términos demográficos, de regenera- rection of ecological change after logging, strongly ción, dinámica de claros, acumulación de bioma- depend on the intensity, and subsequent temporal sa y composición florística al comparar dos parcelas forest dynamics. For this reason, it is important to de una hectárea en un fragmento del bosque poco conduct studies to understand the dynamics of diffe- perturbado (primario y dos parcelas de una hectárea rent ecosystems after selective logging. We analyzed en un fragmento del bosque que fue objeto de tala forest dynamics at El Paujil Reserve (middle Magda- selectiva en el pasado. Como se esperaba, la estruc- lena River, Colombia), in terms of demography, gap tura del bosque y la acumulación de biomasa se ven dynamics, changes in floristic composition, and abi- alteradas por efecto de la tala selectiva, sin embar- lity to gather biomass, comparing two 1 ha plots in go, no generó un impacto significativo en los demás a fragment of undisturbed “primary” forest, and two aspectos mencionados ya que, al parecer, las fuertes 1 ha plots in a fragment of forest that was subject

1 Laboratorio de ecología de bosques tropicales y primatología (LEBTYP), Universidad de Los Andes, Bogotá, Colombia. [email protected]. Autor para correspondencia. 2 Laboratorio de ecología de bosques tropicales y primatología (LEBTYP), Universidad de Los Andes, Bogotá, Colombia. [email protected]. 3 Laboratorio de ecología de bosques tropicales y primatología (LEBTYP), Universidad de Los Andes, Bogotá, Colombia. [email protected]. http://dx.doi.org/10.14483/udistrital.jour.colomb.for.2016.2.a05

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 71 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

to selective logging in the past. As expected, forest generate high mortality and gap formation in both structure and biomass accumulation were altered as forest types. a result of selective logging; however, it did not ge- Keywords: Aboveground biomass, primary forest, nerate a significant impact on the other aspects eva- floristic composition, demographics, Serranía de las luated. We suggest that perhaps local steep slopes Quinchas.

INTRODUCCIÓN el bosque. Otro efecto importante de la tala selecti- va sobre los bosques es la formación de claros, que La tala selectiva en bosques tropicales es una activi- permiten la entrada de una mayor cantidad de luz dad creciente que ha demostrado producir menos al sotobosque, lo cual beneficia a las lianas y espe- daños en el suelo y el dosel frente a la deforesta- cies pioneras, pues son plantas que en su mayoría ción generalizada (Asner et al., 2004). Además, es crecen rápidamente cuando hay una alta disponi- la actividad de uso forestal con menor impacto so- bilidad de luz (Philipson et al., 2014). La dinámi- bre la diversidad, pues los bosques talados selec- ca subsecuente es variable, los árboles pioneros tivamente presentan una alta riqueza de especies pueden alcanzar la altura del dosel y establecerse de bosque maduro (Gibson et al., 2011; Norden et o morir por falta de luz cuando el dosel se forme al., 2009). Sin embargo, tanto la magnitud como nuevamente sobre ellos (Philipson et al., 2014). Las la dirección del cambio ecológico después de la lianas, si logran crecer hasta el dosel y establecerse tala dependen fuertemente de su intensidad y de sobre árboles de bosque maduro, pueden llegar a la subsecuente dinámica espacial y temporal del derribarlos con su peso, fomentando la formación bosque afectado (Asner et al., 2004). de nuevos claros (Schnitzer & Bongers, 2002). De Entre las consecuencias más importantes que este modo, Schnitzer & Bongers (2002) demuestran tiene la tala selectiva en los bosques está la pérdi- que hay una relación negativa en la densidad de da de biomasa, no solo durante el periodo de ex- lianas y la densidad de árboles de bosque maduro, tracción, sino también en el tiempo después de la mientras que la densidad de lianas se relaciona po- actividad. En estudios realizados en el Amazonas sitivamente con la densidad de especies pioneras. brasilero, Figueira et al. (2008) encontraron que En países neotropicales como Colombia, que durante los cuatro años siguientes a las actividades presenta diversidad de clima y relieves, se vuel- de extracción el bosque perdió biomasa. La mor- ve aún más difícil predecir el rumbo que tomará talidad fue mayor a la predicha para este tipo de la recuperación de los bosques objeto de tala se- bosque, posiblemente por los efectos de daño me- lectiva. A pesar de tener un 52.6% (60 millones cánico, dado el aumento en la exposición al viento de hectáreas) del territorio nacional ocupado por en los árboles adyacentes a los sitios de tala (Fi- bosques (IDEAM, 2014), Colombia presenta altas gueira et al., 2008). Sin embargo, estudios a largo tasa de deforestación, con pérdidas de 5.4 millo- plazo como el de Gourlet-Fleury et al. (2013) han nes de hectáreas de bosque en los últimos 20 años encontrado que la tala selectiva favorece el aumen- (García-Romero, 2013). La cuenca del Magdalena to en la biomasa, el crecimiento y la supervivencia es la más deforestada de Sudamérica y la décima de árboles en todas las clases de diámetro inferior del mundo (García-Romero, 2013), pero presen- a 70 cm, así como el reclutamiento de especies de ta regiones como la serranía de las Quinchas, en árboles de crecimiento rápido (pioneros), lo que el Magdalena medio, donde se conservan algunas aumenta las tasas de ganancia de biomasa aérea en áreas de bosque continuo. Por ejemplo, la reserva

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 72 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

de aves El Paujil incluye 3419 hectáreas protegidas noviembre el último. La humedad relativa oscila bajo la figura de Reserva privada de la sociedad ci- entre 85% y 89% (Balcázar-Vargas et al., 2000 ci- vil (ProAves, s.f.) y presenta al menos cuatro tipos tado en Aldana et al., 2008). de bosque identificados por Aldana et al. (2008): La reserva fue creada en noviembre de 2003 bosque con un nivel moderado de tala selectiva para conservar El Paujil de pico azul (Crax alberti) (bosques talados), bosques en llanuras aluviales, (ProAves, s.f.) y su hábitat, el bosque húmedo tro- bosque secundario joven (bosque secundario) y pical. Antes del establecimiento de la reserva, la bosque con tala de subsistencia (bosque primario). reforestadora Bosques del Futuro practicó tala se- El objetivo de este estudio es analizar el efecto lectiva para obtener madera durante cinco años en de la tala selectiva en la dinámica de la comuni- una parte del bosque. Silva-Herrera (1999) reportó dad de árboles en el bosque de la reserva El Paujil que la reforestadora planeaba explotar 50 m³ de en el Magdalena medio colombiano. Esta dinámica madera en pie en el bosque por hectárea al año, evaluada en términos de capacidad de acumular distribuidos así: 15 m3 de maderas finas y 35 m3 de biomasa, dinámica de claros, regeneración, demo- maderas ordinarias. De acuerdo con esto, la refo- grafía y composición florística. Dados los efectos de restadora taló aproximadamente cinco árboles por aumento en disponibilidad de luz y exposición al hectárea, lo cual pudo ser verificado por los inves- viento después de la tala se espera que los bosques tigadores al momento de establecer las parcelas en talados presenten una mayor tasa de reclutamiento 2006. Este bosque está ubicado en el departamen- y mortalidad, una tasa de cambio poblacional ne- to de Boyacá, en la vereda Puerto Pinzón del mu- gativa y menor capacidad de acumular biomasa. nicipio de Puerto Boyacá. El bosque primario, fue Así mismo, por efectos directos de la tala se espe- explotado por los dueños del predio para la sub- ra encontrar una menor densidad de individuos de sistencia, que corresponde al menos de un árbol tallas grandes y una menor cobertura del dosel, lo por ha por año, como pudo ser evidenciado por que permitirá que haya además una mayor densi- los investigadores durante el establecimiento de dad de plántulas y juveniles en el bosque talado. las parcelas. Este bosque se encuentra en el depar- Adicionalmente, dado que la tala selectiva no ha tamento de Santander, en el municipio de Bolívar. sido considerada de gran impacto en la diversidad La distancia entre los dos tipos de bosques (los lu- de los bosques, se espera encontrar similitudes en gares de muestreo) es de aproximadamente 8 km. composición y recambio de especies vegetales en- Los bosques muestreados presentan una topografía tre los dos tipos de bosque, pero mayor abundancia con pendientes de hasta 40 °, y una altitud geográ- de plantas pioneras en el bosque talado. fica desde los 194 msnm hasta los 471 msnm.

Toma de datos MATERIALES Y MÉTODOS Para cada tipo de bosque se muestrearon dos par- Área de estudio celas de una hectárea, establecidas en el año 2006 por Aldana et al. (2008). Ellos realizaron medicio- Este estudio se realizó en la reserva de aves El Pau- nes de diámetro a la altura del pecho (DAP) para jil, en los departamentos de Santander y Boyacá todos los individuos con DAP mayor a 5 cm y los (74° 11´ W, 5° 56’ N) con altitud geográfica des- identificaron hasta especie (o morfoespecie de no de los 150 m hasta los 1200 m y temperatura pro- ser posible). La decisión de incluir individuos a par- medio anual de 27.8°C (Aldana et al., 2008). En tir de los 5 cm de DAP y no desde los 10 cm como el año se presentan dos periodos de lluvias, en- generalmente suele hacerse, obedece a la presen- tre abril y mayo el primero y entre septiembre y cia de una gran cantidad de especies de árboles

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 73 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

que no alcanzan esta talla (anexo 1), y por ende bosque. Utilizando el programa estadístico R ver- se quedarían por fuera del muestreo, lo que lleva- sión 3.0.1 (R Core Team, 2013) se hicieron pruebas t ría a una subestimación en la diversidad y bioma- de una cola para dos muestras, con la aproximación sa acumulada del lugar (Baraloto, 2013). En 2013 de Welch cuando las varianzas no eran similares. se censaron nuevamente, midiendo el DAP para Para el análisis de claros se utilizaron las fotos los individuos muestreados en 2006, reportando tomadas en el punto central de cada una de las los individuos muertos o desaparecidos y tomando subparcelas de 20x20 m, escogiendo una por sub- nota de las causas de mortalidad en los casos que parcela para determinar el valor de gris por pixel fuera posible, siguiendo el protocolo de Phillips et (donde 0 significa negro y 255 significa blanco) al. (2009). Se incluyeron como nuevos reclutas to- con el programa Image J (Image J, 2007). Este valor dos los individuos que entraron a la categoría de representa la intensidad lumínica en el sotobosque tamaño mayor a 5 cm de DAP. Se colectaron mues- y es comparado con una prueba t de una cola para tras de madera para determinar su densidad, usan- dos muestras entre los dos tipos de bosque utili- do un barrenador y tomando muestras de mínimo zando el programa estadístico R versión 3.0.1, li- cinco individuos de las especies más abundantes brería Vegan (R Core Team, 2013). Adicionalmente, de los dos tipos de bosque. como indicadores del proceso de regeneración de Para los estimativos de regeneración se compa- los bosques después del disturbio, se comparó la ró la cantidad de plántulas y juveniles presentes cantidad de plántulas y juveniles presentes en los en los dos tipos de bosque tomando datos de 100 dos tipos de bosque con pruebas t de una cola para parcelas de 2x2 m para plántulas y 100 de 5x5 m dos muestras, con la aproximación de Welch cuan- para juveniles (Stevenson, 2011). Este muestreo se do las varianzas no eran similares en el programa realizó de manera sistemática, dentro de las parce- estadístico R versión 3.0.1 (R Core Team, 2013). las de 1 ha. Para cuantificar la entrada de luz al so- La densidad de las muestras de madera obteni- tobosque se trabajó con 50 subparcelas de 20x20 das en campo se determinó como la gravedad es- m por tipo de bosque, ubicadas dentro de las par- pecífica (peso seco / volumen verde) siguiendo el celas de 1 ha. Se tomaron dos fotos del dosel en el protocolo de densidad de madera de Chavé (2005). punto central de cada una de las subparcelas con Para las especies de las que no se tomó muestra de la cámara paralela al suelo a 1 m de altura. Se uti- madera se utilizaron los datos de densidad madera lizó un lente ojo de pez y se programó la cámara de Casas et al. (datos sin publicar) y los de Zanne para tomar las fotos a blanco y negro, con 22 cm et al. (2009). Posteriormente, se estimó la altura de de apertura constante y velocidad variable. cada árbol utilizando la ecuación alométrica deri- vada de la función de Weibull adaptada para Sur Análisis de datos América presentada por Feldpausch et al. (2012), y se calculó la biomasa acumulada de cada árbol Se comparó la información obtenida en el año 2006 utilizando la ecuación I.3 de Alvarez et al. (2012) con la de 2013 para establecer las tasas anuales de para bosque húmedo tropical en Colombia, que crecimiento, mortalidad y reclutamiento de la co- tiene en cuenta el diámetro a la altura del pecho, munidad de árboles para cada parcela utilizando la altura del árbol y la densidad de la madera y, las formulas presentadas por Sherman et al. (2012). según Alvarez et al. (2012), es la mejor ecuación Luego, para poder comparar estadísticamente estos para estimar carbono en Colombia, dada la baja comportamientos y que sean equiparables con la incertidumbre y variabilidad con respecto a las de- información obtenida de dinámica de claros y rege- más ecuaciones generadas por ellos. Con los va- neración, se realizaron comparaciones a una escala lores de biomasa inicial y final se determinó el de 20x20 m, con 50 subparcelas para cada tipo de cambio de biomasa anual en Mg ha-1 año-1.

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 74 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

Finalmente, se identificaron las cuatro espe- a 2 % (tabla 1). El reclutamiento fue mayor para el cies más importantes en cada parcela con el índi- bosque primario (t = 2.91, gl = 98, p < 0.01; figura ce de valor de importancia que se determina con 1b), y, al igual que en el caso de la mortalidad a es- la densidad, frecuencia y área basal relativa de cala de 1 ha se corrobora el resultado (tabla 1). El cada especie. Se buscó si estas especies son ex- cambio poblacional fue cero para bosque primario plotadas por su madera y la calidad de la misma y negativo para el talado, a escala de 20 x 20 m se (Fern, 2014) para relacionar algún posible cambio encuentran diferencias significativas entre tipos de en las especies dominantes con la tala. Se buscó bosque (t = 2.16, gl = 98, p = 0.02; figura 1c). además el estado de conservación en Colombia de estas especies (Bernal et al., 2014) para indagar la Tabla 1. Tasa anual de mortalidad (m), reclutamiento vulnerabilidad en la que se encuentran los árboles (r) y cambio poblacional (λ) en 2 parcelas de una maderables presentes en este bosque. Adicional- hectárea establecidas en bosques primarios (P3 y P4) mente, se realizó una comparación de las espe- y 2 parcelas de una hectárea en bosques talados (P1 y cies presentes y su abundancia en las parcelas con P5) en la reserva el Paujil (Colombia). un cluster representado en un dendrograma. Dado que las especies pioneras juegan un papel impor- Tipo Bosque Parcela m r λ tante como indicadores de disturbios (tala o caída P3 2.0 1.8 -0.2 Primario de árboles), se categorizaron los individuos mues- P4 2.0 2.2 0.2 treados en tres grupos dependiendo de la densidad P1 1.7 0.6 -1.1 de su madera, por ser uno de los rasgos funcionales Talado P5 2.3 2.2 -0.1 más importantes en la determinación de especies de bosque primario y especies pioneras (Philipson et al., 2014). Se consideraron especies de árboles con densidad de madera baja aquellas con valores entre 0.10 g/cm³ y 0.39 g/cm³, media con madera entre 0.40 g/cm³ y 0.69 g/cm³ y alta las de madera entre 0.70 g/cm³ y 0.90 g/cm³. Posteriormente se comparó la proporción de árboles de cada catego- ría entre tipos de bosque con una prueba G en R versión 3.0.1 (R Core Team, 2013).

RESULTADOS Figura 1. Comparación de las tasas demográficas Demografía anuales en 50 subparcelas de 20x20 m para bosques primarios y 50 subparcelas de 20x20 m para bosques Los bosques talado y primario en la reserva El Pau- talados de los árboles en los bosques de la reserva el jil no presentaron diferencias significativas en la Paujil (Colombia). (a) mortalidad, (b) reclutamiento tasa anual de mortalidad (t = 0.13, gl = 98, p = y (c) cambio poblacional. La mortalidad no presentó 0.45; figura 1a), a pesar de la variación encontrada diferencias significativas, el reclutamiento fue mayor al hacer el análisis por subparcelas de 20x20 m, se para bosque primario *(t = 2.91, gl = 98, p < 0.01) y puede ver que al comparar entre parcelas de 1 ha, el cambio poblacional fue cero para bosque primario el promedio es igual para los dos tipos de bosque, y negativo para el talado, presentan diferencias con una tasa anual de mortalidad promedio igual significativas bn *(t = 2.16, gl = 98, p = 0.02).

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 75 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

Análisis de claros y regeneración.

La cantidad de luz que llega al sotobosque en los bosques de la reserva El Paujil es similar en am- bos tipos de bosque (t = 0.32, gl = 98, p = 0.63; figura 2a). Así mismo, tanto la cantidad de plántu- las como la de juveniles no presentan diferencias significativas entre tipos de bosque (t de welch = -2.18, gl = 197, p = 0.98; t de welch = -4.09, gl = 155, p = 1 respectivamente; figuras 2b y 2c).

Figura 3. Comparación de aspectos estructurales entre 50 subparcelas de 20x20 m para bosques primarios y 50 subparcelas de 20x20 m para bosques talados en la reserva El Paujil (Colombia). (a) Porcentaje de árboles grandes (DAP > 50 cm), donde el primario presenta un valor mayor * (t = 1.40, gl = 98, p = 0.04). (b) Densidad de madera de los árboles grandes para bosque primario y talado, donde el primario presenta un valor mayor *(t = 2.17, gl= 107, p=0.016).

Composición Figura 2. Análisis de la cantidad de luz en el sotobosque y la densidad de plantas regenerando 50 Las especies más importantes de las parcelas de subparcelas de 20x20 m para bosques primarios y 50 bosque primario difieren de las del bosque talado, subparcelas de 20x20 m para bosques talados en la además, son en su mayoría árboles con madera de reserva El Paujil (Colombia). (a) Intensidad lumínica. buena calidad (Fern, 2014), objetivo de las refores- (b) Plántulas. (c) Juveniles. tadoras (tabla 2). El componente que más influyó en la determinación del índice de importancia fue Cambio de biomasa el área basal relativa (anexo 2). El estado de con- servación de estas especies en su mayoría no ha El bosque primario ganó en promedio 4 Mg de sido evaluado, no obstante se conoce que Clathro- biomasa ha-1 año-1, pasando de tener en promedio tropis brunnea está en peligro, Hymenaea courba- 459.5 Mg ha-1 en el 2006 a tener 487.6 Mg ha-1 en ril se encuentra en preocupación menor y Grias el 2013, mientras que el talado ganó menos bio- haughtii es vulnerable (tabla 2). masa, con un promedio de 0.9 Mg ha-1 año-1, pa- Se encontró una alta afinidad florística en las sando de tener 440.6 Mg ha-1 en el 2006 a tener parcelas del bosque primario, mientras que el bos- 447.1 Mg ha-1 en el 2013. Encontramos que hay que talado es más heterogéneo, con una parcela menos árboles con un DAP mayor a 50 cm en el más similar al grupo de bosque primario que la de bosque talado (t = 1.40, gl = 98, p = 0.04; figura su mismo tipo de bosque (figura 4). 3a), y, que la densidad de madera de estos árboles Como resultado de la clasificación por densi- (<50 cm DAP) es menor que en el bosque primario dad de madera, se encontró que la proporción de (t = 2.17, gl = 107, p = 0.016; figura 3b). árboles en cada categoría de densidad de madera

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 76 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

Tabla 2. Especies más importantes para cada parcela de una hectárea (dos por tipo de bosque) establecidas en la reserva El Paujil, con su valor de índice de importancia, uso de madera, calidad de madera y estado de conservación en Colombia.

Tipo de Índice de Uso de Calidad de Estado de bosque Parcela Especie importancia madera madera conservación P3 Eschweilera andina 11.87 x buena NE Andira chigorodensis 7.62 x muy buena NE Garcinia madruno 7.11 mala NE Clathrotropis brunnea 6.21 x mala EP Primario P4 Clathrotropis brunnea 14.66 x mala EP Pseudolmedia rigida 7.95 x mala NE Hymenaea courbaril 6.58 x media CA Eschweilera andina 6.2 x buena NE Cavanillesia platanifolia 13.86 mala NE Simira rubescens 7.95 x media NE P1 Grias haughtii 7.02 x media PM Ephedranthus colombianus 6.79 mala NE Talado Pourouma melinonii 13.53 N/A NE Laetia procera 8.15 x N/A NE P5 Chrysophyllum lucentifolium 7.63 x buena NE Trichospermum galeottii 7.11 mala NE

Convenciones: NE=No evaluado, EP= En peligro, CA= Casi amenazada, N/A= Información no encontrada, PM= Preocupación menor. depende del tipo de bosque (g = 44.266, gl = 2, p < 0.01), encontrando una mayor proporción de es- pecies con densidad de madera baja en el bosque talado (figura 5).

Figura 5. Proporción de individuos muestreados en 2 parcelas de 1 ha para bosque primario y 2 parcelas de Figura 4. Dendrograma comparando la composición 1 ha para bosque talado de la en la reserva el Paujil florística de 2 parcelas de una hectárea establecidas (Colombia) agrupados en tres categorías de densidad en bosques primarios (P3 y P4) y 2 parcelas de una de madera, que corresponden a: baja entre 0.10 g/cm³ hectárea en bosques talados (P1 y P5), en la serranía y 0.39 g/cm³, media entre 0.40 g/cm³ y 0.69 g/cm³ y de las Quinchas (Colombia). alta entre 0.70 g/cm³ y 0.90 g/cm³.

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 77 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

DISCUSIÓN 0.02). Sin embargo, en comparación con otras par- celas de la misma extensión establecidas por inves- Demografía tigadores del Laboratorio de ecología de bosques tropicales y primatología (LEBTYP) en Colombia, se Los componentes demográficos analizados para pudo ver que el porcentaje de lianas en el bosque los bosques de la reserva El Paujil no se compor- primario de la reserva El Paujil tiene un valor alto tan como era esperado. La mortalidad fue igual para (2.98% en promedio), pero no significativamente los dos tipos de bosque (tabla 1 y figura 1a), a pe- mayor en comparación con las demás parcelas de sar que se esperaba mayor en el bosque talado por tierra firme (1.46% en promedio N=20) (t de Welch daño mecánico a causa de una mayor exposición al = 2.92, gl = 2 p = 0.93). Por lo tanto, aún hay in- viento de árboles contiguos a zonas de tala (Figueira certidumbre sobre el efecto de la abundancia de et al., 2008). El reclutamiento, que se esperaba ma- lianas sobre la dinámica de estos bosques. yor en bosques talados por la mayor cantidad de luz disponible en consecuencia de la tala, resultó mayor Análisis de claros y regeneración en bosque primario (tabla 1 y figura 1b) y de hecho no hubo diferencias en la cantidad de luz que entra La magnitud de los claros en los bosques de la re- al sotobosque en los dos tipos de bosque (figura 2a). serva El Paujil, similar en ambos tipos de bosque La población del bosque primario no presentó cam- (figura 3) al igual que los anteriores resultados, lle- bios, mientras que la del bosque talado disminuyó va a pensar en una dinámica moldeada principal- (tabla 1 y figura 1c), presentando diferencias signifi- mente por el efecto de la pendiente del terreno, cativas entre los dos tipos de bosque. causando una alta mortalidad por caída de árboles Como se mencionó anteriormente, los sitios de semejante o en mayor magnitud que la ocasionada bosque muestreados tienen una altitud geográfica por el efecto del disturbio (viento) tras actividades que oscila desde 194 m hasta 471 m, en zonas con de tala. Los vientos se han determinado como un pendientes de hasta 40o. Se ha demostrado que la factor clave en la mortalidad y las dinámicas post pendiente del suelo es un factor ecológico prima- disturbio principalmente en zonas relativamente rio, que controla la tasa de mortalidad por caída de planas, como por ejemplo en el Amazonas, donde árboles y, por tanto, induce un fuerte gradiente de la influencia del viento puede ser mayor (Etter & luz en el sotobosque que favorece el crecimiento Botero, 1990; Laurance & Curran, 2008). y reclutamiento de especies pioneras (Ferry et al., A pesar de existir una proporción de claros y 2010). En el bosque de la reserva El Paujil, donde una cantidad de plántulas y juveniles similar en los la alta mortalidad es casi igual en áreas con tala se- dos tipos de bosque, el mayor reclutamiento ob- lectiva y de bosque primario, se puede decir que el servado en el bosque primario podría ser resultado efecto de la inclinación del terreno sobre la mor- del efecto del daño mecánico causado a plántu- talidad de los árboles puede llegar a enmascarar el las y juveniles en el bosque talado, por la circula- efecto de la tala como lo sugiere Ferry (2010). ción constante de personas (por ejemplo: turistas e Por otro lado, una alta abundancia de lianas en investigadores), que por su fácil acceso tiene una el bosque primario contribuiría a la formación de afluencia mucho mayor a la del bosque primario. claros, pues afecta negativamente a los árboles de bosque maduro (Schnitzer, 2002). Para evaluar esta Cambio de biomasa explicación se determinó y comparó la abundan- cia relativa de lianas en ambos tipos de bosque, lo Los estimativos de reservas de biomasa por hectá- cual corroboró lo predicho, pues resultó mayor en rea para los dos tipos de bosque son relativamen- el bosque primario (t de Welch = 2.09, gl = 85, p = te altos cuando se comparan con estimaciones

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 78 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R. realizada para estos bosques en Colombia (Phillips Composición et al., 2011), sin embargo, esto se explica con el hecho de que en este estudio se incluyeron árbo- Asumiendo que los bosques tenían una compo- les desde 5 cm de DAP. Esta no es una práctica sición similar antes de la tala, las diferencias al común en estudios de estimación de biomasa, sin comparar la composición y dominancia entre ti- embargo, es importante incluir los individuos de pos de bosque (tabla 2 y figura 4) se pueden re- menor talla, dado que no solo aumentan los cál- lacionar con esta actividad. Ya que, por ejemplo, culos de reservas biomasa por ha (Baraloto et al., las especies de árboles de grandes tallas (mayor 2013), por efecto del aumento del número de in- área basal) y madera con densidad alta (conside- dividuos censados, que puede ser de casi el doble, rado un factor importante de calidad) son menos sino también los cálculos de diversidad de espe- importantes en el bosque talado (tabla 2, figura 5). cies (anexo 1). Nuevamente, se pone en evidencia que el efecto Como era de esperarse el bosque primario tuvo más fuerte de la tala selectiva sobre el bosque fue una ganancia neta de biomasa por hectárea mayor generar cambios en la composición de especies, que el bosque talado. Esto concuerda con lo que al disminuir las poblaciones de especies de árbo- reportado en estudios de dinámica de biomasa en les maderables. Este factor debe ser considerado bosques primarios tropicales, donde se han repor- en planes de manejo de las industrias madereras, tado incrementos anuales de biomasa por hectá- pues algunas de estas especies se encuentran en rea del orden de 3 a 20 toneladas (Meister et al., estado de vulnerabilidad (tabla 2) o se descono- 2012). Sin embargo, es notable que los menores ce su estatus de conservación actual (Bernal et al., valores, reportados en el presente estudio para el 2014). Adicionalmente, puede tener implicaciones bosque talado, se asemejan a los valores reporta- en las interacciones tróficas del ecosistema, don- dos por otros estudios de dinámica de biomasa de habitan dispersores de gran importancia como en bosques fragmentados (Nascimento & Lauran- el mono araña (Ateles hybridus hybridus) crítica- ce, 2004). Adicionalmente, la mayor cantidad de mente amenazados y con poblaciones decrecien- árboles grandes observada en el bosque primario tes (Aldana et al., 2008). (figura 2b), sumada que estos tienen árboles con Otros estudios han reportado cambios en la maderas de mayor densidad (figura 2c), pone en composición de especies de bosques de la región, evidencia un efecto de la tala selectiva que no como una respuesta a los cambios en el clima (Du- puede ser comparado tan fácilmente con factores que et al., 2015), que sumados a los efectos de intrínsecos de la dinámica del bosque y que tarda la tala selectiva que evidenciamos en este estudio, largos periodos de tiempo en volver a la normali- podrían causar fuertes disminuciones en las pobla- dad. Es clara la disminución en las poblaciones de ciones de especies vegetales de estos bosques. ciertas especies de árboles, que son objeto de ex- plotación forestal por la calidad de su madera. A la luz de estos resultados, es importante resaltar que, CONCLUSIONES de continuar la fragmentación y la tala selectiva en esta región, los efectos sobre las dinámicas del car- La dinámica del bosque de la reserva El Paujil no bono podrán ser extremadamente negativos hasta muestra una gran diferenciación a causa de la tala el punto que estos bosques pueden dejar de ser selectiva practicada en el pasado. Es importante re- reservorios de carbono para convertirse en fuen- saltar el efecto de la topografía del terreno sobre tes de emisiones de CO2, como se ha previsto para la dinámica del bosque y se recomiendan estudios bosques fragmentados de la amazonia (Laurance sobre el efecto de las lianas y el tránsito de perso- et al., 2011). nas. La estructura del bosque cambia y las reservas

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 79 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

de biomasa disminuyen a causa de la tala selectiva, Balcázar, M. P., Rangel, J. O., & Linares, E. L. (2000). pues hay una importante disminución en las pobla- Diversidad florística de la serranía de Las Quin- ciones de árboles grandes de especies con madera chas, Magdalena medio (Colombia). Caldasia, de buena calidad en el bosque talado, lo que puede 22(2), 191-224. repercutir en las interacciones tróficas del bosque y Baraloto, C., Molto, Q., Rabaud, S., Hérault, B., Va- en su capacidad de ser sumideros de carbono. Este lencia, R., Blanc, L., Fine, P. V. A. & Thompson, tipo de especies deben ser manejadas para evitar la J. (2013). Rapid simultaneous estimation of abo- disminución irreversible de sus poblaciones. veground biomass and tree diversity across Neo- tropical forests: a comparison of field inventory methods. Biotropica, 45(3), 288-298. AGRADECIMIENTOS Bernal, R., Robbert Grandstein, S., & Celis, M. (2014). Catálogo de plantas y líquenes de Colombia. Recu- Por los fondos proporcionados para la toma de da- perado de: http://catalogoplantasdecolombia.unal. tos agradecemos al Fondo de apoyo doctoral de la edu.co/. Facultad de Ciencias de la Universidad de los An- Chave, J. (2005). Measuring wood density for tropical des. A la fundación ProAves, que gracias los con- forest trees. A field manual for the CTFS sites. Tou- venios de investigación en la Reserva Natural de luse, France. 1-7. las Aves El Paujil permitió el establecimiento de las Duque, A., Stevenson, P. R., & Feeley, K. J. (2015). Ther- parcelas y posterior remuestreo. Al equipo de tra- mophilization of adult and juvenile tree commu- bajo de Aldana et al. (2008) por el establecimiento nities in the northern tropical Andes. Proceedings de las parcelas y a Guillermo Rivas, Angela Perilla, of the National Academy of Sciences of the United Pablo Negret, Sebastian Gonzales, Alejandra Jime- States of America, 112(34), 10744-9. nez, Diana Pizano y Sasha Cárdenas por la ayuda Etter, A., & Botero, P. J. (1990). Efectos de los procesos en el remuestreo de las parcelas. climáticos y geomorfológicos en la dinámica del Bosque Húmedo Tropical de la Amazonía Colom- biana. Colombia Amazonica, 4(2), 7-21. REFERENCIAS BIBLIOGRÁFICAS Feldpausch, T. R., Lloyd, J., Lewis, S. L., Brienen, R. J., Gloor, M., Monteagudo Mendoza, A., Lopez-Gon- Aldana, A. M., Beltrán, M., Torres-Neira, J., & Steven- zalez, G., Banin, L., Abu Salim, K., Affum-Baffoe, son, P. R. (2008). Habitat characterization and po- K., Alexiades, M., Almeida, S., Amaral, I., Andra- pulation density of brown spider monkeys (Ateles de, A., Arag˜ao, L. E. O. C., Araujo Murakami, A., hybridus) in Magdalena Valley, Colombia. Neotro- Arets, E. J. M. M., Arroyo, L., Aymard C., G. A.,- pical Primates, 15(agosto), 46-50. Baker, T. R., B´anki, O. S., Berry, N. J., Cardozo, Álvarez, E., Duque, A., Saldarriaga, J., Cabrera, K., N., Chave, J., Comiskey, J. A., Alvarez, E., de Oli- de las Salas, G., del Valle, I., Lema, A., Moreno, veira, A., Di Fiore, A., Djagbletey, G., Domingues, F., Orrego, S., Rodríguez, L. (2012). Tree abo- T. F., Erwin, T. L., Fearnside, P. M., Franca, M. B., ve-ground biomass allometries for carbon stocks Freitas, M. A., Higuchi, N., Honorio C., E., Iida, estimation in the natural forests of Colombia. Forest Y., Jiménez, E., Kassim, A. R., Killeen, T. J., Lauran- Ecology and Management, 267, 297-308. ce, W. F., Lovett, J. C.,Malhi, Y., Marimon, B. S., Asner, G. P., Keller, M., & Silva, J. N. M. (2004). Spa- Marimon-Junior, B. H., Lenza, E., Marshall, A. R., tial and temporal dynamics of forest canopy gaps Mendoza, C., Metcalfe, D. J., Mitchard, E. T. A., following selective logging in the eastern Amazon. Neill, D. A., Nelson, B.W., Nilus, R., Nogueira, E. Global Change Biology, 10(5), 765-783. M., Parada, A., Peh, K. S.-H., Pena Cruz, A., Peñue- la, M. C., Pitman, N. C. A., Prieto, A., Quesada,

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 80 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

C. A., Ramírez, F., Ramírez-Angulo, H., Reitsma, ImageJ. Rasband, W. (2007). ImageJ: Image Processing J. M., Rudas, A., Saiz, G., Salomao, R. P., Schwarz, and Analysis in Java. Bethesda, Maryland: US Na- M., Silva, N., Silva-Espejo, J. E., Silveira, M. Sonké, tional Institutes of Health. B., Stropp, J., Taedoumg, H. E., Tan, S., Ter Stee- Laurance, W. F., Camargo, J. L. C., Luizão, R. C. C., Lau- ge, H., Terborgh, J., Torello-Raventos, M., van der rance, S. G., Pimm, S. L., Bruna, E. M., Stouffer, Heijden, G. M. F., Vásquez, R., Vilanova, E., Vos, V. P. C., Bruce Williamson, G., Benítez-Malvido, J., A., White, L., Willcock, S., Woell, H. & Phillips, O. Vasconcelos, H. L. (2011). The fate of Amazonian L. (2012). Tree height integrated into pan-tropical forest fragments: A 32-year investigation. Biological forest biomass estimates. Biogeosciences Discus- Conservation, 144(1), 56-67. sions, 9(3), 2567-2622. Laurance, W. F., & Curran, T. J. (2008). Impacts of wind Fern, K. (2014). Useful Tropical Plant Database. Recu- disturbance on fragmented tropical forests: A review perado de: http://tropical.theferns.info/. and synthesis. Austral Ecology, 33(4), 399-408. Ferry, B., Morneau, F., Bontemps, J.-D., Blanc, L., & Meister, K., Ashton, M. S., Craven, D., & Griscom, H. Freycon, V. (2010). Higher treefall rates on slopes (2012). Managing Forest Carbon in a Changing Cli- and waterlogged soils result in lower stand biomass mate, 51-75. and productivity in a tropical rain forest. Journal of Nascimento, H. E. M., & Laurance, W. F. (2004). Bio- Ecology, 98(1), 106–116. mass dynamics in amazonian forest. Fragments, Figueira, A. M. E. S., Miller, S. D., de Sousa, C. A. D., 14(4), s127-s138. Menton, M. C., Maia, A. R., da Rocha, H. R., & Norden, N., Chazdon, R. L., Chao, A., Jiang, Y.-H., & Goulden, M. L. (2008). Effects of selective logging Vílchez-Alvarado, B. (2009). Resilience of tropical on tropical forest tree growth. Journal of Geophysi- rain forests: tree community reassembly in secon- cal Research, 113(G00B05), 1–11. dary forests. Ecology Letters, 12(5), 385-94. García Romero, H. G. (2013). Deforestación en Colom- Phillips J.F., Duque A.J., Yepes A.P., Cabrera K.R., Na- bia : Retos y perspectivas. En: F. Dane (Ed.), El De- varrete D.A., Álvarez E., Cárdenas D. (2011). Es- safío del Desarrollo Sustentable en América Latina, timación de las reservas potenciales de carbono 123-142. Rio de Janerio: Konrad Adenauer Stiftung. almacenadas en la biomasa aérea en bosques na- Gibson, L., Lee, T. M., Koh, L. P., Brook, B. W., Gard- turales de Colombia. Informe Final. Instituto de ner, T. A, Barlow, J., Peres, C. A., Bradshaw, C. J. A., Hidrología, Meteorología, y Estudios Ambienta- Laurance, W. F., Lovejoy, T. E., Sodhi, N. S. (2011). les-IDEAM-. Bogotá D.C., Colombia. 68 pp. Primary forests are irreplaceable for sustaining tro- Phillips, O., Baker, T., Feldpausch, T., Brienen, R. pical biodiversity. Nature, 478(7369), 378-81. (2009). Manual de campo para la remedición y es- Gourlet-Fleury, S., Mortier, F., Fayolle, A., Baya, F., tablecimiento de parcelas. En: RAINFOR (Ed., 2nd Ouédraogo, D., Bénédet, F., & Picard, N. (2013). ed.), Proyecto PAN-AMAZONIA. Tropical forest recovery from logging: a 24 year sil- Philipson, C. D., Dent, D. H., O’Brien, M. J., Chamag- vicultural experiment from Central Africa. Philoso- ne, J., Dzulkifli, D., Nilus, R., Philips, S. Reynolds, phical Transactions of the Royal Society of London G., Saner, P., Hector, A. (2014). A trait-based tra- B: Biological Sciences, 368(1625), 20120302. de-off between growth and mortality: evidence IDEAM (Instituto de Hidrología, Meteorología y Es- from 15 tropical tree species using size-specific re- tudios Ambientales), Subdirección de Ecosiste- lative growth rates. Ecology and Evolution, 4(18), mas e Información Ambiental, Grupo de Bosques. 3675-3688. (2014). Proyecto Sistema de Monitoreo de Bosques ProAves (s.f.). Reserva Natural de las Aves El Paujil. Re- y Carbono. Bogotá, D. C., Colombia. cuperado de: http://www.proaves.org/rna-el-paujil/.

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 81 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

R Core Team. (2013). R: A language and environment ubicada en el Magdalena medio. Bogotá: Universi- for statistical computing. Vienna, Austria: R Foun- dad de La Salle. 1-130. dation for Statistical Computing. Stevenson, P. R. (2011). The Abundance of Large Ateline Schnitzer, S. A., & Bongers, F. (2002). The ecology of Monkeys is Positively Associated with the Diversity lianas and their role in forests. Trends in Ecology of Plants Regenerating in Neotropical Forests. Bio- and Evolution, 17(5), 223-230. tropica, 43(4), 512–519. Sherman, R. E., Fahey, T. J., Martin, P. H., & Battles, J. Zanne, A. E., Lopez-Gonzalez, G., D.A., C., Ilic, J., Jan- J. (2012). Patterns of growth, recruitment, mortality sen, S., Lewis, S. L. S. L., Miller, R.B. B.,Swenson, and biomass across an altitudinal gradient in a neo- N.G. G., Wiemann, M.C. C., Chave, J. (2009). Glo- tropical montane forest, Dominican Republic. Jour- bal Wood Density Database. Dryad Digital Reposi- nal of Tropical Ecology, 28(05), 483-495. tory. Recuperado de: http://hdl.handle.net/10255/ Silva Herrera, L. J. (1999). Plan de factibilidad y estra- dryad.235. tegico de la reforestadora Bosques del Futuro S.A.

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 82 ] Dinámica de bosques en diferentes escenarios de tala selectiva en el Magdalena medio (Colombia

RESTREPO, I.C, ALDANA, A.M. & STEVENSON, P.R.

ANEXOS

Anexo 1. Número de individuos y de especies para cada parcela de una hectárea incluyendo árboles de más de 5 cm de DAP y de más de 10 cm de DAP, tomado de Aldana et al. (2008).

DAP > 5 cm DAP > 10 cm Tipo bosque Parcela N.° Especies N.° Individuos N.° Especies N.° Individuos P3 246 1048 185 606 Primario P4 226 924 158 446 P1 234 1070 181 721 Talado P5 201 1000 142 545

Anexo 2. Especies más importantes para cada parcela de una hectárea (dos por tipo de bosque) establecidas en la reserva El Paujil con los valores de densidad, frecuencia, área basal relativa e índice de importancia.

Tipo de Densidad Área basal Frecuencia Índice de bosque Parcela Especie relativa relativa relativa importancia Eschweilera andina 2.19 9.57 0.11 11.87 Andira chigorodensis 1.90 3.81 1.90 7.62 P3 Garcinia madruno 3.24 1.97 1.90 7.11 Clathrotropis brunnea 1.81 3.17 1.23 6.21 Primario Clathrotropis brunnea 3.87 8.06 2.73 14.66 Pseudolmedia rigida 4.62 1.03 2.30 7.95 P4 Hymenaea courbaril 1.00 4.72 0.86 6.58 Eschweilera andina 1.87 4.33 0.04 6.24 Cavanillesia sp01 0.26 13.25 0.35 13.86 Simira rubescens 3.55 2.88 1.52 7.95 P1 Grias haughtii 3.29 1.73 1.99 7.02 Ephedranthus colombianus 1.90 3.37 1.52 6.79 Talado Pourouma melinonii 4.62 5.91 3.00 13.53 Laetia procera 2.49 3.80 1.86 8.15 P 5 Chrysophyllum lucentifolium 2.72 2.91 2.00 7.63 Trichospermum galeottii 2.72 2.81 1.57 7.11

Colombia Forestal • ISSN 0120-0739 • e-ISSN 2256-201X • Bogotá-Colombia • Vol. 19 No. 2 • pp. 71-83 [ 83 ]

ANEXO 2 – Forest biomass density across large climate gradients in northern south america is related to water availability but not with temperature

Alvarez, E., Cayuela, L., Gonzalez-Caro, S., Aldana, A. M., Stevenson, P. R., Phillips, O. L., Von Hildebrand, P., Jiménez, E., Melo, O., Mendoza, I., Restrepo, Z., Velásquez, O. & Rey-Benayas, J. M. 2016. Forest Biomass Density across Large Climate Gradients in Northern South America is related to Water Availability but not with Temperature. Manuscript submitted for publication to PLOS ONE.

148

Running title: Water availability and aboveground biomass of Colombian forests

Title: Forest Biomass Density across Large Climate Gradients in Northern South America is related to Water Availability but not with Temperature.

Álvarez-Dávila E.1,2,3*, Cayuela L.4, González-Caro S.5, Aldana A.M.6, Stevenson P.R.6, Phillips O.7, Cogollo A.5, Peñuela M.C.8, von Hildebrand P.9, Jiménez E.10, Melo O.11, Londoño-Vega AC.12, Mendoza I.2, Velasquez O.13, Fernández F.11, Serna M.14, Velázquez-Rua C2, Benitez, D.5, Rey-Benayas J.M.1

*Corresponding Author: [email protected]

1- Departamento de Ciencias de la vida, Universidad de Alcalá, Alcalá, España.

2- Grupo de Servicios Ecosistemicos y Cambio Climático, Fundación Convida, Medellín, Colombia.

3- Grupo de Agroforestería y Biodiversidad Tropical, Universidad Nacional Abierta y a Distancia, Medellín, Colombia.

4- Área de Biodiversidad y Conservación, Escuela Superior de Ciencias Experimentales y Tecnológicas, Universidad Rey Juan Carlos, Departamental I, Madrid, Spain.

5- Herbario JAUM, Jardín Botánico de Medellín, Colombia

6- Departamento de Ciencias Biológicas, Universidad de Los Andes, Bogotá, Colombia.

7- School of Geography, University of Leeds, Leeds, UK

8- Universidad Regional Amazónica, Napo, Ecuador

9- Fundación Puerto Rastrojo

149

10- Grupo de Ecología de Ecosistemas Terrestres Tropicales, Universidad Nacional de Colombia, Bogotá, Colombia.

11- Facultad de Ingeniería Forestal, Universidad del Tolima, Ibagué, Colombia

12- Investigador independiente, Medellín, Colombia.

13- Facultad de Ciencias Agronómicas, Universidad Nacional de Colombia, Medellín, Colombia

14- Institución Universitaria Tecnológico de Antioquia, Facultad de Ingeniería. Medellín, Colombia.

Key words: Water availability, Elevational gradient, Basal area, Temperature, Carbon stocks, Climate change

150

Abstract

Understanding and predicting the likely response of ecosystems to climate change are crucial challenges for ecology and for conservation biology. Nowhere is this challenge greater than in the tropics as these forests store more than half the total atmospheric carbon stock in their biomass. Biomass is determined by the balance between biomass inputs (i.e., growth) and outputs (mortality). We can expect therefore that conditions that favor high growth rates, such as abundant water supply, warmth, and nutrient-rich soils will tend to correlate with high biomass stocks. Our main objective is to describe the patterns of above ground biomass (AGB) stocks across major tropical forests across climatic gradients in Northwestern South America. We gathered data from 206 plots across the region, at elevations ranging between 0 to 3400 m. We estimated AGB based on allometric equations and values for stem density, basal area, and wood density weighted by basal area at the plot-level. We used two groups of climatic variables, namely mean annual temperature and actual evapotranspiration as surrogates of environmental energy, and annual precipitation, precipitation seasonality, and water availability as surrogates of water availability. We found that AGB is more closely related to water availability variables than to energy variables. In northwest South America, water availability influences carbon stocks principally by determining stand structure, i.e. basal area. When water deficits increase in tropical forests we can expect negative impact on biomass and hence carbon storage.

151

Introduction

Understanding and predicting the likely response of ecosystems to climate change are crucial challenges for ecology and for conservation biology [1]. A key ecosystem service and one of the most studied ecosystem characteristics in forests is the storage of carbon in trees. Knowledge of the distribution of above-ground biomass (AGB) is an essential basis for forest conservation strategies and programs, including the reduction of emissions by degradation and deforestation (REDD) [2,3], to be successful.

The challenge of understanding forest biomass is particularly important in tropical forests, where there are about 460 billion tons of carbon in their biomass and soil, equivalent to more than half the total atmospheric stock [4]. In addition, in terms of carbon fluxes tropical forests process 40 billion tonnes of carbon annually [5]. Deforestation and other anthropic processes significantly impact both stocks and fluxes [2], and recent droughts and climate trends are already impacting tropical biomass in Amazonia and elsewhere (e.g. [6]). There are several different approaches to understanding the distribution of tropical forest biomass. One is based on comparison of remote sensing data with stand variables (wood density, basal area and stem density), allowing AGB estimations [7]. This kind of correlative studies provide spatially-explicit and verifiable estimates of AGB and may allow for extensive assessments of carbon stocks [3]. This approach has great value for mapping carbon and evaluating risks from land-use change and potential benefits from policy interventions (e.g. [8]). Alternatively, environmentally based models can be developed to test and quantify potential ecological mechanisms controlling AGB and so have predictive properties, although the prediction accuracy for any given locality is likely to be low because of the multiple factors involved.

Biomass is the emergent outcome of many complex processes, acting at different temporal and spatial scales. From a reductionist point of view, however, it is determined by the balance between biomass inputs (i.e., growth) and outputs (mortality). We can expect therefore that conditions that favor high growth rates, such as abundant water supply, warmth, and nutrient-

152

rich soils will tend to correlate with high biomass. Environments with high mortality risks – whether via wind extremes, extremes of temperature, or drought, will tend to support lower biomass. Water supply and temperature have multiple impacts on both growth and mortality processes, and so are likely to exert major control on AGB. This expectation is reinforced by the global pattern of covariation of ecosystem carbon turnover times with both precipitation and climate [9].Within the tropics, different climatic variables have been found to covary with AGB in diverse regions around the world. For example, precipitation in the drier quarter is positively correlated with AGB in Amazonian forests [10]. In contrast, AGB is weakly related to climate across a latitudinal gradient in the Neotropics [11]. Models developed for broad climatic and geographic scales are not easily applied to finer scales and vice versa. Nonetheless, Stegen et al. [11] concluded that water availability does have an important effect on aboveground biomass, mainly via limiting the size of larger trees. Also, Baraloto et al. [12] found that AGB is locally more related to stand variables than to climate in Amazon forests. In turn, stand variables are related to the variation in soil water availability, which is determined by topography [13]. This suggests that climate effects on AGB in the tropics may vary regionally, and may be scale-dependent.

Other stand properties that are related to climatic factors could help explain climate impacts on biomass [14]. Notably, recent studies have shown the importance of large trees and their sensitivity to water availability as drivers of AGB. Stegen et al. [11] showed that the very largest tree in a stand are limited by water availability across forests in the Americas because the maximum potential stand AGB is climatically-controlled. Such a relationship appears to have a mechanistic basis in terms of water relations. Thus, for example, Phillips et al. [15] found that drought influences tree mortality across the world’s tropical forests and that the impact of water deficits on mortality rates is greatest for large trees. In addition, Nepstad et al. [16] based on induced drought experiments in Amazonian rainforests, showed that the mortality rates of large trees (> 30 cm of DBH) increased by 4.5 times on dry conditions, and are higher than mortality rates for small trees..

153

Recent studies have shown the risk of embolism increased in large trees, with hydraulic failure during dry periods linked to mortality (e.g., [17]). This mechanism should exert climatic control on stand AGB by impacting especially the physiology of larger trees and thus acting as a filter on tree size. The demonstrated global convergence in forest vulnerability to drought [18] is consistent with this, with safety margins against hydraulic failure being largely independent of precipitation, and mediated in part by changes in species composition and size (hence, AGB) along precipitation gradients.

Further, temperature is suggested to influence biomass because photosynthetic activity is temperature dependent. Clark et al. [19] showed that tree growth rates decreased in warmth years between 1984 and 2000, possibly due to photosynthetic reductions and respiration increments driven by the effect of maximum daily temperature. Also, Doughty & Goulden [20] showed a negative relationship between CO2 exchange and temperature, with a 3ºC rise in environmental temperature causing a 35% decrease of gas exchange at the forest level. Other experiments found that the relationship between temperature and photosynthetic activity have a wider range than expected across plants species [21]. Thus, we can expect a hump-shaped relationship between biomass and temperature based on the temperature- dependence photosynthetic activity or a low plateau relationship, if high temperature does not reduce photosynthetic activity at the community level through wide species tolerances.

Important variation in tropical ecosystem properties at large spatial scales is associated with tens and hundreds of millions of years of evolutionary history, independent of climate. For example, the forests of the Pre-Cambrian Guyana Shield have up to 50% greater biomass density than forests growing in similar climate conditions on Neogene sediments in south- western Amazonia [22]. The lowland rainforests of Borneo are half as productive as climatically and edaphically matched equivalent forests in north- western Amazonia, a difference apparently driven by the preeminence of a single hyper-successful family, Dipterocarpaceae, in Southeast Asian forests [23]. Thus, analysis at cross-continental or multi-continental scales [11,15] might obscure true climatic impacts on tropical AGB due to geological and/or 154

deep phylogenetic controls. To better identify and isolate the climate factors controlling biomass, it is therefore important to analyze tropical biomass variation at smaller scales, where such differences are proportionately less important. One approach is to analyze elevation transects along large mountain ranges (e.g. [24]). An alternative approach, taken here, is to analyze ecological variation within a few parts of the tropics where extremely long dry to wet and cold to hot climate gradients occur within relatively short distances, affording a rich set of contrasts and replication. Probably the single most climatically complex region of the world exists in northwestern South America. For example, three mountain ranges run from south to northeast, dividing lowlands and reaching an altitude of 6000 m (some peaks have permanent snow). The inter-Andean Magdalena and Cauca valleys, where dry forests are the predominant natural habitat, dissect these mountain ranges. In addition, two very wet forests (Chocó Biogeographic region and the Amazon basin) limit the Andean mountains.

Our main objectives are to i) quantify and (ii) describe the patterns of forest AGB stocks across major tropical climatic gradients and (iii) to understand the effects of water availability and environmental energy on stand variables, and indirectly on AGB, in Northwestern South America. In particular, we want to answer the following questions: i) Are stand variables (wood density, basal area and stem density) related to AGB along broad climatic gradients of temperature and moisture? ii) What is the shape of the relationship between stand variables and climate? iii) Is climatic variation related to AGB? Such answers will help improving our knowledge on tropical forest stocks and contribute to understand the likely effects of climate change on ecosystem functioning.

Materials and Methods

Plot sampling and aboveground biomass estimation

The sampled region (northwest South America) includes some of the wettest, driest, hottest, and coldest tropical forests on Earth. Within the region

155

there is a large variety of forest types according to climatic and biogeographic regions: Amazon forests (moist forests), Andean montane forests (> 1000 masl), Chocó forests (pluvial forests), Caribbean forests (dry forests), Orinoquia forest (continuous and riparian forests) and, finally, inter Andean valley forests (a dry to moist lowland forest gradient along Magdalena and Cauca rivers). We gathered data from 156 plots in old-growth forest across the region at elevations ranging between 0 to 3400 m (Fig 1) in the period 1989-2013. The vast majority of the plots were established by the authors, using standardized methods [25], some of the plot data has been already published in recent work from the coauthors [22,26]. The vegetation plots are located in private-owned land and some in National Parks, for which we obtained permits. In some cases, an agreement was signed between the researchers and the landowners. In any case, this research did not focus on endangered species. These plots represent very large climatic gradients, notably from < 10oC to almost 30oC mean annual temperature, and from <1,000 mm to almost 10,000 mm annual rainfall (Fig 2). Plot area ranges from 0.25 to 25 ha, although most have an area of 1 ha. We converted all data to 1 ha equivalent units prior to analysis. Diameter at breast height (DBH) was measured for all trees in each plot. We used the values of wood density for a total of ~55% individual species reported in the Wood Density Global Database [27]. For 45% additional species not found in this database, acknowledging the strong phylogenetic signal for this trait [28], we used the genus or family mean, depending on data availability. For the relatively few cases when neither family nor genus were reported in our database (14%), we used wood density averaged per plot [29].

For our initial, exploratory analyses two sets of allometric equations were used to calculate AGB for each individual tree. One set is widely used for tropical analyses Chave’s AGB equation [30]), which has been reported to overestimate biomass stocks in some localities; the other set is based only on Colombian forest tree allometries (Alvarez’s GB equation [31]). We summed AGB values of all individuals in each plot to calculate AGB per plot (Mg ha-1). Both sets of equations in fact produced similar stand-level results (R2=0.97; F= 1165; p < 0.001) (S1 Fig.). We will show here the results based on Alvarez

156

et al. [31] equations, because these are regionally adjusted and allow controlling for allometric variation along the elevational gradient. As for AGB, -1 we also derived plot-level values for stem density (Nind; N ha ), basal area (BA; m2 ha-1), and wood density average (WD; g cm-3). In addition, we gathered 34 1ha plot from published studies which report Chave’s plot AGB and plot stand variables [22]; we transformed these Chave’s AGB values to Alvarez’s plot AGB using a previously fitted linear model. Finally, we included 10 additional plots from Colombia estimating the biomass indirectly from information on Nind and BA. For this purpose, we use a model calibrated with the 156 plots for which we have the tree-by-tree data: AGB Mg ha-1 = exp (1.048 * ln (BA) + 0.299 * ln (Nin)); R2 -adjusted = 0.92, F-ratio = 833.0, P < 0.0001). Overall, we used 200 plots in further analyses. Data from the plots included in this study is available in the Supporting Information (S1 File).

157

Figure 1. Distribution of plots used to estimate aboveground biomass in the studied Northwest South American region (Colombia, Brazil, Peru and Ecuador). Color of the symbols represent the forest types in the region: Blue for forest plots in Amazonia; yellow for forest plots in the Andean uplands; red for plots in the inter- andean valleys; Orange for the Caribbean plots; Green for the Orinoco region and the green triangles for the forest plot in the Choco region. The grey scale is displayed to denote altitude (m. a.s.l)

Climatic data

Climatic variables such as annual mean temperature (AMT; °C), annual precipitation (AP; mm) and within year precipitation variability (PV %; Variation coefficient of the monthly precipitation along the year) were downloaded from the WorldClim Global Climate Data [32]. Actual evapotranspiration (AET; mm)

158

and potential evapotranspiration (PET; mm) were extracted from the Geospatial Database CGIAR Consortium for Spatial Information [33]. To distinguish the two main climate effects hypothesized to lead to spatial variation in AGB, we split climatic predictors onto two groups of variables representing environmental energy and water availability. We chose the following variables: AMT and AET, as surrogates of environmental energy; and AP, PV and water availability (WA mm; AP minus PET), as surrogates of water availability. Correlations between pairs of variables was moderate to low (maximum r = 0.68).

Figure 2. Climatic space represented for each vegetation plot used in this analysis. The climatic space is shown as principal components analysis to reduce climatic variables used. The first axis represents temperature variability and second axis represents precipitation variability. Gray points represent the climatic space availability across Northwest South America. Blue points represent actual climatic conditions of each of the vegetation plots sampled.

159

Statistical analyses

To answer our questions, we followed five analytical steps: (1) We correlated pairs of explanatory variables, including climatic and stand variables, to examine collinearity. (2) We standardized explanatory variables by fitting a mean equal to 0 and variance to 1 to directly compare the effects of all explanatory variables. (3) We used generalized linear squares (GLS) models to relate each of these variables to AGB (response variable), and used generalized nonlinear squares (GNLS) models to test possible non- linear relationships. The parameters of GNLS were assigned using the brute force method, which consists of a heuristic approach to estimate model parameters. (4) We controlled the effect of spatial data distribution on model results by including a spatial correlation matrix on the models. This procedure allows detecting differences in parameter estimation when spatial variation is considered. (5) We used the Akaike Information Criterion (AIC) to choose between resulting models, and a pseudo-R2 (calculated as: (Null deviance - Residual deviance)/ Null deviance) to show explained deviance [34]. These analyses were conducted using packages pls [35], nlme [36], proto [37], MASS [38] and raster [39] from the R environment for statistical computing [40].

We applied the above five steps to evaluate the relationship between: i) stand variables (wood density, basal area and stem density) and AGB, ii) stand variables and climate, and iii) AGB and climate.

Results

Amount of aboveground biomass and structural parameters

Northwest South American forests showed huge variation in their AGB and structure (Table 1). AGB ranged between 7.7 and 386.9 Mg ha-1 (mean = 194.4, SD = 87.4), basal area between 1.6 and 46.8 m2 ha-1 (mean = 23.1; SD = 8.2), stem density between 61 and 1,388 ha-1 (mean = 583; SD = 214.7), and basal-area-weighted wood density between 0.37 to 0.75 g cm-3 (mean = 0.59; SD = 0.05). The highest and lowest values of AGB and BA were found

160

in the Amazon forests and the Dry Inter Andean Valley forests, respectively (Table 1).

Correlations & explanatory models

AGB was more strongly related with BA (R2 = 0.85, p < 0.001, Fig 3a) than with stem density (R2 = 0.38, p < 0.001, Fig 3b), whereas it was unrelated with wood density (R2 = 0.00; Fig 3c)

Figure 3. Plots of forest structure parameters and aboveground biomass. The Coefficient of determination is showed in each plot. *P<0.01; **P<0.001; ***P<0.000; ns: non-significant.

161

Table 1. Mean and standard deviation of aboveground biomass (AGB) across geographic regions of Northwest South America. N= plot number per region; AGB Aboveground Biomass; BA= Basal area; WDBA= Wood density weighted by basal area.

-1 2 -1 Stem density 3 Region N AGB (Mg ha ) BA (m ha ) WDBA (g cm ) (N ha-1)

Mean SD Mean SD Mean SD Mean SD

Amazonia 52 259.7 51.8 26.9 4.5 658.5 135.9 0.63 0.04

Andean total 63 211.9 70.8 26.0 7.3 689.6 191.0 0.57 0.02

Andean (Quercus 19 229.9 85.8 28.3 8.9 798.6 258.2 0.59 0.03 present)

Andean (Not 44 204.1 62.8 25.0 6.3 642.6 131.1 0.57 0.02 Quercus present)

Inter Andean valleys 16 44.5 20.0 8.9 4.5 297.3 151.9 0.60 0.06 (Dry)

Inter Andean valleys 10 156.6 41.6 20.8 4.8 618.5 225.0 0.57 0.05 (Moist)

Caribbean 19 75.4 52.7 14.4 8.6 340.8 195.0 0.59 0.07

Choco 35 217.5 46.6 24.8 5.0 554.3 155.7 0.55 0.05

Orinoquia 5 138.7 43.3 16.9 3.3 402.0 64.9 0.53 0.03

200 194.4 87.4 23.1 8.2 582.6 214.7 0.59 0.05

162

AGB was significantly related to climate and stand structure variables and that it was more closely related to water availability variables (Fig 4a; Fig 4b) than to energy variables (Fig 4c; Fig 4d). Water availability (WA; annual precipitation minus potential evapotranspiration) was the single best determinant of AGB (Table 2). WA was better adjusted as a quadratic form, clearly suggesting a biomass maximum at intermediate levels of water availability; and lower biomass in sites with extreme precipitation. (Fig 4a). Additionally, precipitation variability (PV, i.e. the evenness of water supply through the year) was negatively related to AGB, suggesting that low rain variation promotes large biomass stocks in northwest South American forests (Fig 4b).

Figure 4. Relationship between aboveground biomass (AGB) and (a) water availability (WA), (b) precipitation variability (PV), actual evapotranspiration (AET) and (d) annual mean temperature (AMT). Bioregions are shown with different colors. Solid lines represent the trend of relationships, based on the original data (without transformation), according to the best models (highest AIC scores) presented in Table 2; pR2 is a partial regression coefficient for each of the relationships.

163

AGB was weakly related to actual evapotranspiration (AET) and annual mean temperature (AMT; Table 2). As these variables were highly correlated with elevation, this indicates little change in AGB across Andean mountains in Northwest South America. Finally, our assessment of spatial autocorrelation on AGB variation along climate gradients revealed that the model parameters were weakly affected by spatial variation (Table 2).

164

Table 2. Explanatory models of aboveground biomass. WA: Water availability; AP: Annual precipitation; PV: Precipitation variability; AMT:

Annual mean temperature; AET: Actual evapotranspiration; BA= Basal area; Nind: Individual density; WDAB= Wood density weighted by basal area. AIC= Akaike information criteria. RSE= Residual sum error. The best model based on AIC is bolded.

MODELS GLS (Only environment) GLS (environment + space)

a b c α β AIC RSE a b c α β AIC RSE

Water availability

α (WA)β - - - 1.920 0.307 527.8 0.896 - - - 1.883 0.324 513.2 0.914

2 a (WA) + b(WA) -0.380 2.172 - - - 482.2 0.800 -0.370 2.137 - - - 473.0 0.818

- b(WA) + c - 0.312 1.634 - - 559.3 0.953 - 0.340 - - 542.3 0.971 0.032

165

- b(AP) + c - 0.270 2.140 - - 564.5 0.965 - 0.296 - - 547.6 0.985 0.033

- - b(PV) + c - -0.589 2.140 - - 495.1 0.810 - - - 485.0 0.829 0.594 0.016

Environmental energy a (AMT)2 + b(AMT) 1.285 1.023 - - - 791.6 1.733 1.201 1.052 - - - 731.0 1.668

- - b(AMT) + c - 570.7 0.981 - - - 553.8 1.000 -0.208 2.140 - - 0.233 0.019

2 a (AET ) + b(AET) 0.545 0.997 - - - 866.7 2.092 0.542 1.019 - - - 793.0 1.982

- b(AET) + c - 0.360 2.140 - - 552.0 0.935 - 0.359 - - 538.7 0.958 0.024

Stand variables

- b(BA) + c - 0.923 2.140 - - 200.3 0.385 - 0.955 - - 165.1 0.404 0.018

166

- b(WDBA) + c - 0.058 2.140 - - 578.8 1.001 - 0.072 - - 562.4 1.024 0.030

- b(Nind) + c - 0.623 2.140 - - 482.1 0.784 - 0.614 - - 466.2 0.794 0.021

167

Discussion

Aboveground biomass variation across the study region

AGB values for the set of Amazon forests were within the range of previously reported values elsewhere in the Amazonian Basin (200-360 Mg ha-1; [10,41,42]).

Similarly, our basal area estimates were within the range reported (25-35 m2 ha-1) in two large forest inventories [10,41] . In contrast, Choco and inter-Andean lowland tropical humid forest had lower values than expected (Table 1). Also, the studied

Andean forests had higher values than previously reported at the pan-tropical scale

[3], supporting the finding that mountain carbon stocks are still little known [43]. In addition, several Andean plots in our dataset have larger AGB than the maximum

AGB value expected for Neotropical forests (350 Mg ha-1; [44]).

AGB pantropical maps [2,3], are widely used by governments when presenting proposals to reduce deforestation through economic compensations [45]; however, these maps can contain significant regional biases. Recently, Mitchard et al. [22] reported great differences between the AGB stocks reported in pantropical maps and the field data in the Amazon. In our study, in which we include a great number of field data, we also found great differences with the AGB estimations reported in pantropical maps (S2 Fig). First, AGB estimations extracted from these maps poorly predicts the observed AGB in the field, with an explained variation of only 24.0% for Saatchi et al [3] and 39.0% for Baccini et al [2]; Second, mean deviation, or error, of the AGB values predicted from the pantropical maps for each of our plots: Error =100 * (AGBpredict - GBmeasured) / AGBmeasured), is very high for both maps, with -45.8 ± 52.1 % and 23.0 ± 109.5% for [3] and [2] respectively, showing low precision for these pantropical maps. This comparison confirms the findings by

168

Mitchard et al. [22] and suggests that pantropical AGB maps should be revised to include the spatial variation in wood density [41] and the allometric relationship between tree diameter and height [46] to have more precise maps of carbon stocks in tropical forests.

Water availability and forest structure

We found that water availability variables were highly related to AGB, rather than environmental energy variables (Table 2). This result is consistent with previous studies in the Neotropics [47] and African forests [48]. In northwest South America, water availability influences carbon stocks principally by determining stand structure, i.e. basal area. Also, this relationship is humped, which forests between 2500-3500 mm have higher values of AGB. For example, Choco and Amazon region have divergent values of AGB, although these may be considered structurally similar.

Moreover, the influence of water supply on forest structure may be a pan-tropical phenomenon at different scales, for example Chave et al. [49] improved allometric equations including a similar water availability coefficient than ours. This suggests that it may be possible to improve the spatial accuracy of remote-sensing estimates of carbon stocks [22] by accounting for local water supply across the mapped surface. This is confirmed by Detto et al. [13] who demonstrated, in a mesic forest in

Panama, that canopy height was highest in areas of positive convexity (valleys, depressions) close to draining channels, which seems like a response to greater water availability in the soil.

Water deficits in the shape of occasional or regular droughts are well-known to drive mortality, particularly of larger trees [11,15,16], and these mortality impacts

169

may be limiting AGB in our forests too. The close dependence of AGB on BA (and hence on size of larger trees) is consistent with this mechanism. However, until sufficient long-term monitoring data are available it is unlikely to distinguish this from an alternative mechanism whereby extended dry seasons limit AGB simply by suppressing tree growth. Regardless of the exact mechanism, our results from a wide climatic gradient in a region with relatively little spatial extent underline the importance of water availability on AGB. If and when water deficits increase in tropical forests, we can expect negative impact on biomass and hence carbon storage [16]. However, future scenarios of precipitation remain rather poorly understood [50] and in recent years some Neotropical environments have become wetter (e.g. [51]). The future importance of water-limitation on AGB stocks in tropical forests thus remains unclear.

Temperature and forest structure

We found a weak relationship between AGB and environmental energy variables, particularly with temperature. Several studies have shown that global temperature is an important determinant of the spatial distribution of biomass [52], due to its effect on the ecophysiological processes that control the net primary productivity rate (mainly photosynthesis and respiration). However, at the local or regional level, the tropical forests biomass is influenced by a large number of variables besides temperature and usually presents a high spatial variability. In the present study, the stronger effect of water availability than that of temperature hints that biomass output drivers could be more important than biomass input drivers shaping biomass stocks. Furthermore, temperature is correlated with elevation. This

170

imply simultaneous changes in temperature, humidity, solar radiation, soils, species composition and historical factors that could affect biomass in concert [38,39,40]

[53–55]. For example, Culmsee et al. [54] found higher AGB values in Southeast

Asian montane forest plots dominated by Fagaceae species with high wood that contributed significantly to AGB. Similar biogeographic patterns may be occurring in

Northern Andean forests that have landscapes dominated by temperate immigrants such as Fagaceae species (Table 1).

The ratio of height to diameter in tropical forests tends to decline with elevation and temperature (e.g. [31]), and so excluding such allometric variation by simply applying ‘universal’ allometric equations would lead to overestimating the biomass of Andean forests. Here, AGB was calculated using allometric equations specifically developed for Colombian forest types that implicitly include the known variation of tree height with elevation [31]. Additionally, we found that stem density was negatively related to temperature (Rp = -0.40; p < 0.001; i.e. positively related to elevation), similar to previous studies on tropical elevational transects [24,56]. Thus, other stand variables such as wood density could be important in driving this observation. However, wood density data of mountain species is scarce and requires more sampling effort [28], pointing to the challenge of AGB quantification on mountain ranges and the need of field work on these regions.

Database and methodological limitations

Our study evidences that climatic variables, related to water availability are the most determinant of the spatial distribution patterns of primary forests of

Northeastern South America. Nonetheless, there are some limitations, both in the

171

data and the methods that could have affected our results. Regarding the data, the forest inventories do not include all the forest types and there are some missing points of the climatic gradient. For instance, the forests of the eastern Colombia

(llanos), swamps, mangroves and highland forests (>3.000 m a.s.l.) are poorly represented; this fact may explain the high level of uncertainty of the predictive models for these forest types. Additionally, the distribution of the inventories is not random, mainly due to the intrinsic bias of ease of access in all tropical forest inventories [57], as well as the limited availability of forest remnants, such as the dry forests of the Caribbean. Regardless of these limitation, our data set includes forest plots of various sizes (0.25-1 ha) which is considered representative for the study of the structure of tropical forests [22].

Regarding the methods, the equations we used do not include height as a predictive variable of AGB, which could produce a bias in the estimations, given the fact that low precipitation and altitude can influence the positive allometry for tree diameter and height. However, the equations we used are calibrated for each forest type in the region and the specific allometry for diameter and height is intrinsically included in these biomass equations. Further similar analysis should include tree height as a variable to reduce uncertainty in the AGB estimations [46]. Another possible limitation of our methods is the resolution of the climatic data (1 km2) and the resolution of the forest plot data (1 ha or less). This lack of correspondence between climatic and forest data maybe ignoring the fine scale effects of the climatic heterogeneity on AGB [22,58]. Finally, some studies in the Amazon Basin have established that edaphic conditions maybe more important variables than climate determining forest structure and AGB stocks [14]. Future studies should include local edaphic variables.

172

Conclusions

Water availability has an important role shaping spatial patterns of carbon stocks in northwestern South America across a huge climate gradient representative of the whole tropical forest zone, consistently with water deficits enhancing tree mortality and shifting size distribution of stands to favor small trees. AGB did not vary systematically with temperature, suggesting that temperature-mediated processes such as autotrophic respiration do not have a major impact on forest biomass in our study region. Increasing the understanding of how forest stand variables respond to climatic variability on spatial gradients could inform us about likely tropical biomass responses to climate change. By incorporating forest/climate relationships like those identified here it should be possible to improve calibration and accuracy of remote- sensing based maps of tropical forest carbon stocks.

173

Acknowledgments

We thank Z. Restrepo, W. Lopez, J.C. Rodriguez and Y. Alvarez for field assistances on different plots used in this study. We also thank P.C. Zalamea for early comments and ideas on this manuscript. We acknowledge Interconexión Eléctrica

S.A (ISA), Instituto Alexander von Humboldt (Colombia), INCIVA and National Parks, for access to some plot data. We are grateful to the field researchers for forest plot data.

174

References

1. Pan Y, Birdsey RA, Phillips OL, Jackson RB. The Structure, Distribution, and Biomass of the World’s Forests. Annu Rev Ecol Evol Syst. 2013;44: 593–622. doi:10.1146/annurev-ecolsys-110512-135914 2. Baccini A, Goetz SJ, Walker WS, Laporte NT, Sun M, Sulla-Menashe D, et al. Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nat Clim Chang. 2012;2: 182–185. doi:10.1038/nclimate1354 3. Saatchi SS, Harris NL, Brown S, Lefsky M, Mitchard ET, Salas W, et al. Benchmark map of forest carbon stocks in tropical regions across three continents. Proceedings of the National Academy of Sciences of the United States of America. 2011. pp. 9899–904. doi:10.1073/pnas.1019576108 4. Pan Y, Birdsey RA, Fang J, Houghton R, Kauppi PE, Kurz WA, et al. A large and persistent carbon sink in the world’s forests. Science. 2011;333: 988–993. doi:10.1126/science.1201609 5. Beer C, Reichstein M, Tomelleri E, Ciais P, Jung M, Carvalhais N, et al. Terrestrial gross carbon dioxide uptake: global distribution and covariation with climate. Science. 2010;329: 834–838. doi:10.1126/science.1184984 6. Phillips OL, Aragão LEOC, Lewis SL, Fisher JB, Lloyd J, López-González G, et al. Drought sensitivity of the Amazon rainforest. Science. 2009;323: 1344– 1347. doi:10.1126/science.1164033 7. Baraloto C, Rabaud S, Molto Q, Blanc L, Fortunel C, Hérault B, et al. Disentangling stand and environmental correlates of aboveground biomass in Amazonian forests. Glob Chang Biol. 2011;17: 2677–2688. doi:10.1111/j.1365-2486.2011.02432.x 8. Asner GP, Mascaro J. Mapping tropical forest carbon: Calibrating plot estimates to a simple LiDAR metric. Remote Sens Environ. Elsevier Inc.; 2014;140: 614–624. doi:10.1016/j.rse.2013.09.023 9. Carvalhais N, Forkel M, Khomik M, Bellarby J, Jung M, Migliavacca M, et al. Global covariation of carbon turnover times with climate in terrestrial ecosystems. Nature. Nature Publishing Group; 2014;514: 213–217. doi:10.1038/nature13731 10. Malhi Y, Wood D, Baker T, Wright J, Phillips O, Cochrane T, et al. The regional variation of aboveground live biomass in old-growth Amazonian forests. Glob Chang Biol. 2006;12: 1107–1138. doi:10.1111/j.1365-2486.2006.01120.x 11. Stegen JC, Swenson NG, Enquist BJ, White EP, Phillips OL, Jørgensen PM, et al. Variation in above-ground forest biomass across broad climatic gradients. Glob Ecol Biogeogr. 2011;20: 744–754. doi:10.1111/j.1466- 8238.2010.00645.x 12. Baraloto C, Rabaud S, Molto Q, Blanc L, Fortunel C, Hérault B, et al.

175

Disentangling stand and environmental correlates of aboveground biomass in Amazonian forests. Glob Chang Biol. 2011;17: 2677–2688. doi:10.1111/j.1365-2486.2011.02432.x 13. Detto M, Muller-Landau HC, Mascaro J, Asner GP. Hydrological Networks and Associated Topographic Variation as Templates for the Spatial Organization of Tropical Forest Vegetation. PLoS One. 2013;8. doi:10.1371/journal.pone.0076296 14. Quesada CA., Phillips OL, Schwarz M, Czimczik CI, Baker TR, Patiño S, et al. Basin-wide variations in Amazon forest structure and function are mediated by both soils and climate. Biogeosciences. 2012;9: 2203–2246. doi:10.5194/bg-9- 2203-2012 15. Phillips OL, van der Heijden G, Lewis SL, López-González G, Aragão LEOC, Lloyd J, et al. Drought-mortality relationships for tropical forests. New Phytol. 2010;187: 631–646. doi:10.1111/j.1469-8137.2010.03359.x 16. Nepstad DC, Tohver IM, Ray D, Moutinho P, Cardinot G. Mortality of Large Trees and Lianas following Experimental Drought in an Amazon Forest. Ecology. 2015;88: 2259–2269. 17. Anderegg WRL, Berry JA., Smith DD, Sperry JS, Anderegg LDL, Field CB. From the Cover: The roles of hydraulic and carbon stress in a widespread climate-induced forest die-off. Proc Natl Acad Sci. 2012;109: 233–237. doi:10.1073/pnas.1107891109 18. Choat B, Jansen S, Brodribb TJ, Cochard H, Delzon S, Bhaskar R, et al. Global convergence in the vulnerability of forests to drought. Nature. 2012;491: 752–5. doi:10.1038/nature11688 19. Clark DA, Piper SC, Keeling CD, Clark DB. Tropical rain forest tree growth and atmospheric carbon dynamics linked to interannual temperature variation during 1984-2000. Proc Natl Acad Sci. 2003;100: 5852–5857. doi:10.1073/pnas.0935903100 20. Doughty CE, Goulden ML. Are tropical forests near a high temperature threshold? J Geophys Res Biogeosciences. 2009;114: 1–12. doi:10.1029/2007JG000632 21. Way DA, Oren R. Differential responses to changes in growth temperature between trees from different functional groups and biomes: a review and synthesis of data. Tree Physiol. 2010;30: 669–688. doi:10.1093/treephys/tpq015 22. Mitchard ET, Feldpausch TR, Brienen RJW, Lopez-Gonzalez G, Monteagudo A, Baker TR, et al. Markedly divergent estimates of Amazon forest carbon density from ground plots and satellites. Glob Ecol Biogeogr. 2014;23: 935– 946. doi:10.1111/geb.12168 23. Banin L, Lewis SL, Lopez-Gonzalez G, Baker TR, Quesada CA, Chao KJ, et al. Tropical forest wood production: A cross-continental comparison. J Ecol. 2014;102: 1025–1037. doi:10.1111/1365-2745.12263

176

24. Girardin CAJ, Malhi Y, Aragão LEOC, Mamani M, Huaraca Huasco W, Durand L, et al. Net primary productivity allocation and cycling of carbon along a tropical forest elevational transect in the Peruvian Andes. Glob Chang Biol. 2010;16: 3176–3192. doi:10.1111/j.1365-2486.2010.02235.x 25. Vallejo-Joyas MI, Londoño-Vega AC, López-Camacho R, Galeano G, Álvarez- Dávila E, Devia-Álvarez W. Establecimiento de parcelas permanentes. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt. Bogotá D. C., Colombia. 310 p. (Serie: Métodos para estudios ecológicos a largo plazo; No. 1). 2005. 26. Aldana AM, Villanueva B, Cano Á, Correa DF, Umaña MN, Casas-Caro L, et al. Drivers of biomass stocks in Northwestern South American forests: complementing information for the Neotropics. For Ecol Manage. Elsevier B.V.; 2016;389: 86–95. doi:10.1016/j.foreco.2016.12.023 27. Zanne AE, Lopez-Gonzalez G, Coomes DA, Ilic J, Jansen S, Lewis SL, et al. Data from: Towards a worldwide wood economics spectrum. Dryad Digital Repository. 2009; Available: http://datadryad.org/handle/10255/dryad.235 28. Chave J, Coomes D, Jansen S, Lewis SL, Swenson NG, Zanne AE. Towards a worldwide wood economics spectrum. Ecol Lett. 2009;12: 351–66. doi:10.1111/j.1461-0248.2009.01285.x 29. Baker T, Phillips O, Malhi Y. Variation in wood density determines spatial patterns inAmazonian forest biomass. Glob Chang …. 2004; 545–562. doi:10.1111/j.1529-8817.2003.00751.x 30. Chave J, Andalo C, Brown S, Cairns MA., Chambers JQ, Eamus D, et al. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia. 2005;145: 87–99. doi:10.1007/s00442-005-0100-x 31. Alvarez E, Duque A, Saldarriaga J, Cabrera K, de las Salas G, del Valle I, et al. Tree above-ground biomass allometries for carbon stocks estimation in the natural forests of Colombia. For Ecol Manage. 2012;267: 297–308. doi:10.1016/j.foreco.2011.12.013 32. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A. Very high resolution interpolated climate surfaces for global land areas. Int J Climatol. 2005;25: 1965–1978. doi:10.1002/joc.1276 33. Zomer RJ, Trabucco A, Bossio DA, Verchot L V. Climate change mitigation: A spatial analysis of global land suitability for clean development mechanism afforestation and reforestation. Agric Ecosyst Environ. 2008;126: 67–80. doi:10.1016/j.agee.2008.01.014 34. Chambers JM, Hastie TJ. Statistical Models in S. Chapman and Hall, London.; 1991. 35. Mevik BH. Partial Least Squares and Principal Component Regression [Internet]. 2015. Available: http://cran.r-project.org/package=pls 36. Pinheiro J, Bates D, DebRoy S, Sarkar D. NLME: Linear and nonlinear mixed

177

effects models. [Internet]. 2013. pp. 1–336. Available: https://cran.r- project.org/web/packages/nlme/index.html 37. Kates L, Petzoldt T. proto: Prototype object-based programming [Internet]. 2012. pp. 1–8. Available: http://cran.r-project.org/package=proto 38. Venables WN, Ripley BD. Modern Applied Statistics With S [Internet]. Fourth. Technometrics. New York: Springer; 2003. pp. 111–111. doi:10.1198/tech.2003.s33 39. Hijmans R, van Etten J, Cheng J, Mattiuzzi M, Sumner M, Greenberg JA, et al. Package “ raster ”. CRAN -R.2.5-8 [Internet]. 2016. Available: http://cran.r- project.org/package=raster 40. R Core Team. R: A Language and Environment for Statistical Computing [Internet]. 2016. Available: http://www.r-project.org/ 41. Baker T, Phillips O, Malhi Y, Almeida S, Arroyo L, Di Fiore A, et al. Variation in wood density determines spatial patterns in Amazonian forest biomass. Glob Chang Biol. 2004;10: 545–562. doi:10.1111/j.1529-8817.2003.00751.x 42. DeWalt SJ, Chave J. Structure and Biomass of Four Lowland Neotropical Forest 1.-. Biotropica. 2004;36: 7–19. doi:10.1111/j.1744-7429.2004.tb00291.x 43. Selmants PC, Litton CM, Giardina CP, Asner GP. Ecosystem carbon storage does not vary with mean annual temperature in Hawaiian tropical montane wet forests. Glob Chang Biol. 2014; doi:10.1111/gcb.12636 44. Keeling HC, Phillips OL. The global relationship between forest productivity and biomass. Glob Ecol Biogeogr. 2007;16: 618–631. doi:10.1111/j.1466- 8238.2007.00314.x 45. Diaz D, Hamilton K, Johnson E. State of the Forest Carbon Markets 2011: From Canopy to Currency. For Trends Ecosyst Marketpl. 2011; 93. Available: http://www.forest-trends.org/documents/files/doc_2963.pdf 46. Feldpausch TR, Lloyd J, Lewis SL, Brienen RJW, Gloor M, Monteagudo Mendoza A, et al. Tree height integrated into pantropical forest biomass estimates. Biogeosciences. 2012;9: 3381–3403. doi:10.5194/bg-9-3381-2012 47. Saatchi S, Houghton RA, Dos Santos Alvalá RC, Soares J V., Yu Y. Distribution of aboveground live biomass in the Amazon basin. Glob Chang Biol. 2007;13: 816–837. doi:10.1111/j.1365-2486.2007.01323.x 48. Lewis SL, Sonke B, Sunderland T, Begne SK, Lopez-Gonzalez G, van der Heijden GMF, et al. Above-ground biomass and structure of 260 African tropical forests. Philos Trans R Soc B Biol Sci. 2013;368: 20120295– 20120295. doi:10.1098/rstb.2012.0295 49. Chave J, Réjou-Méchain M, Búrquez A, Chidumayo E, Colgan MS, Delitti WBC, et al. Improved allometric models to estimate the aboveground biomass of tropical trees. Glob Chang Biol. 2014;20: 3177–3190. doi:10.1111/gcb.12629

178

50. Mccain CM, Colwell RK. Assessing the threat to montane biodiversity from discordant shifts in temperature and precipitation in a changing climate. Ecol Lett. 2011;14: 1236–1245. doi:10.1111/j.1461-0248.2011.01695.x 51. Gloor M, Brienen RJW, Galbraith D, Feldpausch TR, Schöngart J, Guyot JL, et al. Intensification of the Amazon hydrological cycle over the last two decades. Geophys Res Lett. 2013;40: 1729–1733. doi:10.1002/grl.50377 52. Reich PB, Luo Y, Bradford JB, Poorter H, Perry CH, Oleksyn J. Temperature drives global patterns in forest biomass distribution in leaves, stems, and roots. Pnas. 2014;111: 13721–13726. doi:10.1073/pnas.1216053111 53. Alves LF, Vieira SA., Scaranello MA., Camargo PB, Santos FAM, Joly CA., et al. Forest structure and live aboveground biomass variation along an elevational gradient of tropical Atlantic moist forest (Brazil). For Ecol Manage. 2010;260: 679–691. doi:10.1016/j.foreco.2010.05.023 54. Culmsee H, Leuschner C, Moser G, Pitopang R. Forest aboveground biomass along an elevational transect in Sulawesi, Indonesia, and the role of Fagaceae in tropical montane rain forests. J Biogeogr. 2010;37: 960–974. doi:10.1111/j.1365-2699.2009.02269.x 55. Unger M, Homeier J, Leuschner C. Effects of soil chemistry on tropical forest biomass and productivity at different elevations in the equatorial Andes. Oecologia. 2012;170: 263–274. doi:10.1007/s00442-012-2295-y 56. Leuschner C, Moser G, Bertsch C, Röderstein M, Hertel D. Large altitudinal increase in tree root/shoot ratio in tropical mountain forests of Ecuador. Basic Appl Ecol. 2007;8: 219–230. doi:10.1016/j.baae.2006.02.004 57. Marvin DC, Asner GP, Knapp DE, Anderson CB, Martin RE, Sinca F, et al. Amazonian landscapes and the bias in field studies of forest structure and biomass. 2014; doi:10.1073/pnas.1412999111 58. Slik JWF, Aiba SI, Brearley FQ, Cannon CH, Forshed O, Kitayama K, et al. Environmental correlates of tree biomass, basal area, wood specific gravity and stem density gradients in Borneo’s tropical forests. Glob Ecol Biogeogr. 2010;19: 50–60. doi:10.1111/j.1466-8238.2009.00489.x

179

ANEXO 3 – Dinámica, estructura y diversidad de los bosques de galería en la región de los llanos, colombia

Gonzalez, J.S., Aldana, A.M., Correa, D., Casas, L.F & Stevenson P.R. Dinámica, estructura y diversidad de los bosques de galería en la región de los llanos, colombia. Manuscrito en preparación para ser sometido a Revista de Biología Tropical.

180

DINÁMICA, ESTRUCTURA Y DIVERSIDAD DE LOS BOSQUES DE GALERÍA EN LA REGIÓN DE LOS LLANOS, COLOMBIA

Dynamic, structure and diversity of gallery forest in the Colombian Llanos

Juan Sebastián González2, Ana María Aldana3, Diego Correa4, Luisa Casas5, Pablo R. Stevenson6

Resumen

Estudios de dinámica y acumulación de biomasa son indispensables para entender el funcionamiento y productividad de bosques. Además, son valiosos para la planeación de estrategias de conservación y uso sostenible. Este estudio se realizó (llevo a cabo) con información de 5 hectáreas de bosque de galería (3 en tierra firme y 2 en planos de inundación de Igapó), ubicadas en la reserva de Tomo-Grande, Vichada. Se evaluaron y cuantificaron las diferencias entre ambos tipos de bosque en cuanto a diversidad, composición florística, biomasa aérea, dinámica de bosque y se determinó la influencia de la composición química del suelo. El bosque más diverso fue el de tierra firme. No se encontraron diferencias en las tasas de mortalidad y cambio poblacional, pero sí en las de reclutamiento, que fueron mayores en tierra firme. Se encontró que la tasa de crecimiento relativa es mayor en tierra firme pero el cambio en la biomasa es mayor en planos de inundación. También, se distingue una marcada diferencia en la textura y el contenido de nutrientes en el suelo. Este es el primer estudio que destaca las diferencias en la dinámica y diversidad de bosques de galería inundables y de tierra firme en la región de los Llanos. Palabras clave: Cambio Biomasa aérea, planos de inundación, tierra firme, suelo, Vichada.

2 Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, Colombia, [email protected]. Autor para correspondencia. 3 Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, Colombia, [email protected] 4 Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, Colombia, [email protected] 5 Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, Colombia, [email protected] 6 Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, Colombia, [email protected] 181

Abstract

Dynamic and biomass accumulation studies are essential to understand the forest functioning and productivity. Also, they are useful in conservation planning and sustainable use of these ecosystems. This study was undertaken with information from 5 gallery forest hectares (3 in terra firme and 2 in Igapó flood plains), located in the Tomo-Grande Reserve, Vichada. The plots were resampled 5 years after their establishment. We evaluated and quantified the differences in diversity, floristic composition, aerial biomass, forest dynamics and the soil chemistry composition between the two types of forests. We found that the terra firme forest was the most diverse and the floristic composition was considerably different from Igapó. There were no differences in the mortality rates and population change, but terra firme forest had a highest recruitment rate. The relative growth rate was highest in terra firme although the biomass change was highest in flood plain. We found clear differences in the soil texture and nutrient content between forests. This is the first study that recognizes the dynamic and diversity differences between terra firme and flood plain gallery forest in the Llanos region.

Key words: Aerial biomass change, flood plain, terra firme, soil, Vichada.

182

Introducción

La comprensión de los patrones de riqueza de especies y abundancia de individuos en una comunidad y los mecanismos que las determinan han sido desde hace mucho tiempo de gran interés para los ecólogos (Hubbell, 2001; Berazategui, 2012). Así pues, se han propuesto dos enfoques para entender los procesos involucrados en el ensamblaje y mantenimiento de la diversidad de las especies. De una parte la Teoría Neutral propone un ensamblaje por efectos estocásticos donde la dispersión juega un papel fundamental, ya que todos los individuos de todas las especies pueden establecerse en cualquier lugar, independiente de las características ecológicas (Hubbell, 2001). Por el contrario el ensamblaje según la teoría de nicho ecológico es el resultado de los rasgos ecológicos de las especies y de la competencia entre ellas. Así, las especies mejor adaptadas a los filtros impuestos por el ambiente son las que finalmente se pueden establecer (Berazategui, 2012). Estos filtros ambientales pueden estar relacionados con ambientes estresantes para las planta, como puede ser el caso del constante estrés hídrico en los desiertos o los periodos anóxicos en bosques inundables, en donde solo un subconjunto de especies pueden llegar a establecerse (Umaña et al. 2012; Casas y Stevenson, 2013).

Se sabe que las condiciones de los bosques pueden variar en el tiempo debido a patrones externos como pueden ser las inundaciones, sequias, fuegos o procesos antropogénicos (Connel, 1978; Phillips, 2004). Según las teorías de nicho, podría haber cambios temporales en la composición de especies, dependiendo de este tipo de cambios (Enquist y Nicklas, 2001). Para entender estas variaciones se han realizado una gran cantidad de estudios enfocados a examinar las relaciones entre la estructura, dinámica y factores ambientales (e.g. Peacock et al., 2007). Algunos de los temas más examinados son los que se refieren a los patrones espaciales y temporales de biomasa, productividad de madera y tasas de mortalidad, reclutamiento y crecimiento. (Phillips, 2004; Phillips y Lewis, 2014). Estos últimos tres patrones se conocen como la dinámica de bosque, que es el objeto de evaluación en este estudio, en conjunto con los cambios en la biomasa y la influencia del suelo.

183

La dinámica de bosques se centra en el estudio de los cambios temporales en composición y reclutamiento (Veblen et al., 2004). Estos cambios pueden ser también producto de un proceso de sucesión ecológica, es decir de cambios direccionales en la composición de especies después de un disturbio. De otra parte, se ha sugerido que las características fisicoquímicas del suelo, que definen la fertilidad de éste, están directamente correlacionadas con las tasas de cambio y la producción de biomasa, a través de toda la Cuenca Amazónica (Phillips, 2004; Quesada et al., 2012).

A nivel del trópico se ha sugerido en muchos trabajos el amplio rango de diferencias tanto en la dinámica como en la variación de acumulación de biomasa en los bosques neotropicales a nivel espacial y nivel temporal (Phillips, 1994; Phillips, 2004; Sherman et al. 2012). Sin embargo, teniendo en cuenta que estos cambios pueden manifestar funcionamiento y ecología de los bosques neotropicales a gran escala, hay que recalcar que hace falta evidencia que permita entender mejor estos patrones en ecosistemas específicos del neotrópico (Phillips et al., 2014). Por ejemplo, los bosques de galería de la altillanura del Orinoco en Colombia, han sido un ecosistema muy poco estudiado de manera cuantitativa y no conocemos ningún estudio de dinámica de bosques en estos lugares.

El bosque de galería es un ecosistema que hace parte del paisaje de sabana tropical y que en su mayoría son estrechas franjas de bosque asociadas a caños y ríos; y aunque la fracción de territorio en el paisaje sea pequeña, su valor para la región es muy grande (Veneklaas et al., 2005). Estos ecosistemas ribereños contienen una gran riqueza de especies y recursos que no se encuentran en las zonas de sabana y adicionalmente cumplen con importantes funciones ecológicas (Casas y Stevenson, 2013). Los bosques de galería pueden llegar a ser de gran importancia en un futuro no muy lejano ya que gracias a que la mayoría de fragmentación en los bosques tropicales es tan reciente, ha sido difícil de evaluar en un largo periodo las consecuencia de este proceso; sin embargo, estos ecosistemas pueden llegar a ser buenos indicadores del estado estable de un fragmento de bosque tropical (Kellman et al., 1998).

En Colombia el conocimiento de los bosque de galería de la Orinoquía es muy limitado, según nuestro conocimiento solo existen 7 parcelas permanentes en

184

bosques de galería de sabanas estacionales; 2 en bosques de galería inundables en el Meta, y las otras 5 en el Vichada, de las cuales 3 son de tierra firme (Correa y Stevenson, 2010) y las otras dos ubicadas en bosque de Igapó (Casas y Stevenson, 2013). Estas últimas 5 parcelas fueron precisamente las que se escogieron para realizar este estudio.

En el presente estudio nos propusimos evaluar la diversidad, dinámica, acumulación de biomasa y la variación y efecto de los suelos de los bosques de galería de tierra firme e inundable, con planos de inundación de aguas negras (Igapó), en la Reserva de Tomo-Grande, Vichada. Teniendo en cuenta aspectos teóricos y algunos de los estudios citados anteriormente, hipotetizamos que a pesar de la cercanía geográfica van a existir diferencias tanto en la diversidad, estructura, suelos como también en las tasas demográficas, de crecimiento y acumulación de biomasa, entre ambos tipos de bosque, debido a los diferentes regímenes y condiciones a que se ven expuestos los árboles en ambos bosques.

Materiales y métodos

Zona de estudio

El estudio se realizó en la Reserva Tomo-Grande (Santa Rosalía, Vichada). La reserva de Tomo Grande se localiza en el área de desembocadura del rio Caño Grande en el rio Tomo en la vereda Nazareth. La zona se encuentra en la altillanura disectada de los llanos orientales y hace parte del pedobioma de sabana tropical estacional, caracterizado por sabanas extensas y vegetación boscosa restringida a bosques de galería, con una marcada temporada seca en el año y ocurrencia de quemas periódicas (Hernández & Sánchez, 1994; Romero et al., 2009; Correa & Stevenson, 2010; Casas y Stevenson, 2013). El promedio anual de temperatura oscila entre 24,5 y 27º C. La precipitación es estacional y monomodal, con un máximo en el mes de Junio. La precipitación registrada en la estación más cercana, Las Gaviotas, sobre el caño Urimica (Vichada), registra un promedio anual de 2.673 mm (Hurtado et al., 2005; Correa & Stevenson, 2010; Casas, 2013). La cobertura vegetal de esta área de estudio corresponde a grandes sabanas y la vegetación boscosa, principalmente bosques de galería, que podemos catalogar como tierra firme, ya que solo sufren inundaciones esporádicas en momentos de

185

exceso de lluvia. En la Reserva también hay bosques inundables, que se encuentran en las orillas del río Tomo (Casas y Stevenson, 2013), y corresponden a planos de inundación de ríos de aguas negras por lo que pueden considerarse bosques de Igapó (Prance, 1970). Geográficamente la reserva hace parte de la cuenca media del Río Tomo, donde aún hay un bajo porcentaje de cobertura boscosa (Romero et al., 2009) y un nivel bajo de intervención antrópica (e.g. fuegos frecuentes en sabanas y entresaca de árboles para cercas).

Figura 1. Imagen satelital la zona de estudio, donde se resaltan las 5 parcelas de vegetación que fueron re-muestreadas entre Enero y Mayo del 2014.

Fase de campo

Entre enero y mayo de 2014 se re-muestrearon 5 parcelas de vegetación de 1 hectárea, 3 de bosque de tierra firme y 2 de bosques inundables, localizadas en la reserva de Tomo-Grande (Figura 1). Las parcelas de tierra firme fueron establecidas durante el primer semestre del año 2009 (Correa & Stevenson, 2010) y las parcelas de bosque inundable fueron establecidas entre febrero y marzo del 2010 (Casas y Stevenson, 2013). Para el re-muestreo se realizaron mediciones de diámetro a la altura del pecho (DAP) y altura total (altura de la rama más alta de la copa) para todos los individuos. Se incluyeron como nuevos reclutas todos los individuos con DAP mayor a 10 cm. Para los individuos muertos se registró el tipo de muerte,

186

siguiendo el protocolo propuesto en el manual de campo de RAINFOR (Phillips et al., 2009).

Adicionalmente, durante el re-muestreo del 2010 (Casas y Stevenson, 2013) se tomaron un total 125 muestras de suelo. Las muestras se tomaron a 20 cm de profundidad, tomando 1 muestra de suelo por cada sub parcela de 20 x 20 en cada parcela de 1 hectárea, es decir, en total se tomaron 25 muestras por parcela. A todas las muestras se les hicieron análisis de textura, carbono orgánico (CO), nitrógeno (N), fósforo (P), pH, acidez de cambio (Al), capacidad de intercambio catiónico (CIC), capacidad de intercambio catiónico efectivo (CICE), y contenido de bases (Ca, Mg, K, y Na) en el Laboratorio de Suelos y Aguas de la Facultad de Agronomía de la Universidad Nacional sede Bogotá. Igualmente durante este período se tomaron muestras de madera de la mayoría de las especies presentes en las 5 parcelas, para ser analizadas en el laboratorio y se éste modo obtener su gravedad específica (Williamson & Wiemann, 2010).

Análisis de datos

Diversidad y estructura de bosques

Se estimó y se cuantificó la diversidad en las 5 parcelas y por tipo de bosque. Para estimar la diversidad se calcularon 4 aspectos: 1. Riqueza de especies. 2. Proporción de especies por tallo. 3. Índice de diversidad α de Fisher.4. Índice de diversidad de Shannon exponencial: Esta es una corrección del índice de Shannon el cual estima las especies precisamente por su frecuencia, sin favorecer de manera desproporcionada a las especies, ya sean raras o comunes (Jost, 2006).

Adicionalmente, para comparar la diversidad entre los dos tipos de bosque se construyeron dos curvas de rarefacción, una por cada tipo de bosque. Las curvas de rarefacción se construyeron con el software Estimates 9.1.0 (Colwel, 2013), a partir de una matriz que agrupa a los individuos presentes en muestras de sub- parcelas de 20 x 20 m. De igual manera, estas curvas pueden darnos información acerca del esfuerzo de muestreo y la diversidad en general.

Para determinar la relación entre el DAP y la altura total en cada bosque, se realizó una regresión lineal simple, tomando como variable respuesta la altura total. Se excluyeron las palmas del análisis pues en la mayoría de las especies, el

187

ensanchamiento del tallo ocurre antes de su elongación (Henderson, 2002). Adicionalmente, se cuantificó el número de individuos según el rango del DAP y la densidad de madera de los árboles presentes en ambos tipos de bosque. Por otra parte, siguiendo los resultados obtenidos de la clasificación por tipo de muerte propuesta por RAINFOR (Phillips et al., 2009), se obtuvo el porcentaje de cada tipo de muerte por bosque.

Composición florística

Se realizó un análisis de ordenación no paramétrico: escalamiento multidimensional no métrico (NMDS, por sus siglas en inglés) utilizando las abundancias de las especies y el método de disimilitud de Bray-Curtis, mediante el software estadístico R (v. 3.12; R Core Team Development 2014), comparando la composición florística entre las 5 parcelas. Adicionalmente se calculó el índice de importancia (IVI) para cada especie (Rangel-Ch y Velázquez, 1997). Dinámica de bosque

Se realizaron comparaciones con los datos obtenidos de 2009 y 2010 para establecer las tasas crecimiento relativo anual (TCR) , mortalidad (m), reclutamiento (r) y cambio poblacional (γ), por tipo de bosque, por parcela y por especie; las cuales fueron calculadas usando un modelo de crecimiento exponencial en un tiempo continuo propuesto por Lewis et al. (2004).

Adicionalmente, se utilizaron unas ecuaciones equivalentes para calcular la tasa de mortalidad (mBA) y reclutamiento (rBA) en términos de biomasa aérea. Sustituyendo la BA al inicio del muestreo por N0, la BA de los árboles muertos por Dt y la BA al final del censo por Nt.

Para el cálculo de la tasa de crecimiento relativo se utilizó la siguiente formula:

!"# = ( ln )*+,- − ln )*+,/ ) 1- − 1/ . 34(5)

Donde el DAP es el diámetro a la altura del pecho de un individuo en el tiempo t2 y en el tiempo t1 dividido por el número de años entre los censos. Adicionalmente se calculó el incremento del diámetro proporcional anual (% del DAP t1) para ambos tipos de bosque (Kellman et al., 1998).

188

Para realizar los análisis estadísticos se utilizó el programa estadístico R (v.3.1.2; R Core Team Development 2014). En primer lugar se realizó la prueba de Wilcoxson para analizar las diferencias de TCR y cambio poblacional (γ) entre los bosques. De otra parte se realizó una prueba ANOVA para evaluar las diferencias entre las parcelas en cuanto a mortalidad, reclutamiento, cambio poblacional y TCR; para este último se recurrió a una prueba Tukey HSD para especificar contrastes entre parcelas.

Análisis de suelos

Para analizar las diferencias entre los suelos, según sus propiedades, se realizó un análisis NMDS usando el método de disimilitud de Bray-Curtis con ayuda del software R. Igualmente, se hicieron pruebas t para cada una de las variables medidas en ambos bosques, para ver si existía diferencia estadística en estos parámetros.

Estimación de biomasa aérea y acumulación de carbono

Para la estimación de la biomasa aérea de los árboles, se utilizaron las ecuaciones desarrolladas por Álvarez et al. (2012) generadas para Colombia, las cuales se ajustan a las características ecológicas del área de estudio. Se utilizó el modelo tipo I el cual relaciona el diámetro (D), la altura total de los árboles (H) y la densidad de madera (ρ) de las distintas especies. La ecuación se asignó en función de la zonas de vida propuestas por Holdridge et al. (1971) y de acuerdo con Chave (2005). En particular se tomó la ecuación para un bosque húmedo tropical Ec (6).

6* = exp (−2,261 + 0.937 ∗ ln )-D E . 34(6)

La biomasa aérea de cada parcela se obtuvo del resultado de la sumatoria de BA de los individuos registrados en ella. Se probó si existían diferencias entre parcelas utilizando una prueba Wilcoxson por parejas y el método de ajuste de Bonferroni. Y para analizar diferencias entre bosques se utilizó una prueba de Wilcoxon. La acumulación de carbono se obtuvo teniendo en cuenta que para el trópico el 45% de la biomasa aérea es carbono (Pan et al., 2013).

189

Resultados

Diversidad y estructura de bosque

En el análisis de diversidad por se observó que el bosque de tierra firme tiene cerca del doble del número de especies que el bosque inundable (Tabla1). Sin embargo, para saber si esta diferencia es debida al número de individuos presentes, se calculó el número de especies dividido por el número de tallos con el que se comprobó es el bosque más diverso el de tierra firme. Adicionalmente, los índices de diversidad calculados, el índice de α Fisher y el de eH, también confirman lo anterior (Tabla 1).

Tabla1. Resultados generales de diversidad y estructura por tipo de bosque de 5 parcelas de vegetación en la Reserva Tomogrande.

No. Área No. sp Inclinados Bosque individuo basal total Sp/tallo α Fisher eH (riqueza) (%) s (m2)

Tierra 1460 94,50 120 0,08 30,40 51,93 1,92 firme

Inundable 1367 88,88 59 0,04 12,14 22,82 8,78

Las curvas de rarefacción corroboran una vez más que el bosque de tierra firme es más diverso que el bosque inundable (Figura 1 2) y también sugieren que el esfuerzo de muestreo en cada bosque abarca gran parte de las especies presentes.

190

140

120

100 de especiesde 80

60 acumulado

40 Inundable

Número 20 Tierra firme

0 0 10 20 30 40 50 60 70 80 Número acumulado de individuos Figura 2. Curva de rarefacción para comparar la riqueza del bosque inundable y el bosque de tierra firme en la Reserva Tomo-Grande, Vichada (Colombia). En la relación del DAP vs altura, los datos se ajustan mejor en el bosque de tierra firme (R2=0,50) que en el bosque inundable (R2=0,26). Además, se observa que esta relación es mayor en el bosque de tierra firme, es decir que si se tiene un árbol con el mismo DAP en ambos bosques va a tender a ser más alto en el bosque de tierra firme que en el bosque inundable (Figura 3).

Figura 3. Relación entre el diámetro a la altura del pecho (DAP) y altura total de los árboles en bosques inundables de Igapó y bosque de galería en tierra firme. Las líneas hacen referencia al ajuste de la regresión lineal.

191

La estructura de ambos tipos de bosque, a partir del número de individuos y de su DAP indica que en proporción los árboles jóvenes, es decir con un DAP más pequeño, son mayoría en el bosque inundable (Figura 4.a). En ambos bosques se encontraron árboles de más de 60 cm de DAP, sin embargo, en las parcelas de tierra firme fue donde encontraron los individuos de mayor DAP, como es el caso de un individuo de Erisma uncinatum Warm. (DAP = 74,6 cm) y uno de la especie Enterolobium schomburgkii (Benth.) Benth. (DAP= 77,3 cm).

De otra parte, se encontró una diferencia significativa en la densidad de la madera para ambos tipos de bosque (W = 4572, p < 0.005). Donde el bosque de aguas negras (Igapó) presentó un mayor promedio en la densidad de la madera (0,625 g/cm3 ±0,11), con respecto a el bosque de tierra firme (0,585 g/cm3 ± 0,11) (Figura 4.b). a) b)

Figura 4. a. Estructura de la vegetación de acuerdo a la frecuencia de individuos, en los diferentes rangos de diámetro a la altura del pecho (DAP), por hectárea. b. Densidad de madera; en los bosques inundables de Igapó y bosques de tierra firme de la Reserva de Tomo-Grande, Vichada.

Asimismo, siguiendo el protocolo de RAINFOR (Phillips et al., 2009) encontramos que existen 11 diferentes tipos de muerte en ambos bosques. En primer lugar, tenemos que más del 80% de los árboles mueren de tres formas principalmente:

192

Muerto en pie, caído desde la raíz o por el tronco roto. Sin embargo, al parecer existen patrones específicos de muerte para cada tipo de bosque. Donde, en el bosque inundable el tipo de muerte más común es la muerte en pie, con un poco más del 50%. Por otro lado, en tierra firme los patrones de muerte son más equilibrados, en donde el 35% de árboles mueren por tronco roto y el 28% y 22% caídos desde la raíz y muertos en pie, respectivamente.

Composición florística

En la ordenación se puede observar claramente la disimilitud que existe en la composición florística entre parcelas de tierra firme y de bosque inundable (Figura 5). Además, cabe resaltar que las dos parcelas de bosque inundable difieren notoriamente entre ellas.

Las figura 6 muestra los resultados del índice de importancia (IVI) de las 10 especies más importantes en cada bosque. Las especies con mayor IVI en el bosque de tierra firme correspondieron a Attalea maripa (Aubl.) Mart. (25,30) y Jacaranda copaia (Aubl.) D. Don. (23,96) debido a sus altos valores de densidad y dominancia. Cabe agregar que aunque J. copaia registró los valores más altos de densidad (11,04%), la especie más dominante del bosque fue A. maripa (8,12%) .

Para el bosque inundable, sobresale Mabea nitida (Spruce ex Benth). como la especie más dominante (16,44%) e importante del bosque (35,75). También sobresalen dos especies de la familia de las leguminosas, Tachigali odoratissima y Tachigali chrysophylla, segunda y tercera en el rango de importancia en el bosque.

193

Figura 5. Ordenación NMDS tomando la composición florística de plantas leñosas con diámetro mayor o igual a 10 cm presentes en 5 parcelas de una hectárea en la Reserva de Tomogrande, Vichada (Colombia).

De otra parte, es importante notar que las especies de bosque inundable arrojaron valores de dominancia mucho más altos que los que se observan en las especies de bosque de tierra firme, como se observa claramente en las especies más importantes de cada bosque, donde para bosque inundable M. nítida y T. odoratissima tiene valores de dominancia de (16,44%) y (10,14%) respectivamente y para bosque de tierra firme A. maripa y J. copaia con valores de (8,12%) y (5,71%) respectivamente (Anexo 1).

194

a) Tierra firme 0 5 10 15 20 25 30

Attalea maripa Jacaranda copaia Bocageopsis multiflora Tetragastris… Oenocarpus bataua Pseudolmedia laevis Couepia glabra Guatteria metensis Pourouma aurea Xylopia polyantha

b) Plano inundable

0 5 10 15 20 25 30 35 40

Mabea nitida Tachigali odoratissima Tachigali chrysophylla Licania heteromorpha Eschweilera parvifolia Guatteria brevicuspis Caraipa llanorum Laetia suaveolens Duroia micrantha Licania apetala

% Dom % DR % FR

Figura 6. Especies más importantes en los bosques de la Reserva Tomo-Grande (Vichada), según el índice IVI. Dominancia relativa (% Dom), Densidad relativa (% DR), Frecuencia relativa (% F).

195

Tasa de reclutamiento, mortalidad y cambio poblacional

Se encontró que la tasa de mortalidad anual promedio en el bosque inundable (1,43% anual) es menor a la tasa encontrada en el bosque de tierra firme (2,05% anual) (Tabla 4). Sin embargo, según el estadístico de Wilcoxon, esta diferencia no es significativa (W=1757, p=0,55) (Figura 7a). Por el contrario, sí se encontraron diferencias en la tasa de reclutamiento anual (W=970, p<0,01), siendo mayor el reclutamiento en tierra firme (Figura 7b).

Para el bosque de Igapó el porcentaje de muertos, en su mayoría, se da en el estadio juvenil de los árboles. El 68% de los individuos muertos tiene DAP ≤ 25 cm. En tierra firme ese porcentaje fue 50,5%. Asimismo, se observó que el porcentaje de individuos muertos en tierra firme es más alto en la mayoría de casos, siendo esta diferencia más notoria en el estadio longevo de los árboles. Sin embargo, hay que tener en cuenta que el número de individuos iniciales es muy diferente en por hectárea de cada tipo de bosque. No se encontraron diferencias en las tasas de cambio poblacional de las especies de ambos bosques (W=1504, p=0,06), siendo la media del cambio poblacional positivo en el bosque de tierra firme . El valor de la media de γ cercano a cero se puede explicar debido a que en ambos tipos de bosque la mayoría de las especies tuvieron un cambio poblacional de cero. a) b)

196

c) d)

Figura 7.a: Tasa de mortalidad anual (%). b: Tasa de reclutamiento anual (%). c. Tasa de cambio poblacional (%) anual. d. Tasa de crecimiento relativo anual (%). En el bosque inundable y bosque de tierra firme en la Reserva Tomo-Grande, Vichada (Colombia).

Tasa de crecimiento relativo anual

Las medias de la tasa de crecimiento relativo anual (TCRA) de las especies del bosque de Tierra firme son mucho más altas que las de las especies del bosque inundable (W = 1289, p < 0.01) (Figura 6.c). Adicionalmente, es pertinente resaltar a algunas especies abundantes e importantes con una media muy por encima del promedio de TCRA; como T. odoratissma (1,259 %), T. chrysophylla (1,02%) en bosque inundable y Bocageopsis multiflora (Mart.) R.E. Fr. (1,632%) en el bosque de tierra firme. Estimación de biomasa y acumulación de carbono.

En cuanto al cambio de BA los valores negativos indican una pérdida de biomasa que ocurre por la muerte de árboles en su mayoría. Vale la pena resaltar que en las parcelas de tierra firme, la varianza alcanza valores más negativos que las parcelas de bosque inundable, es decir que hubo grandes pérdidas de biomasa en dichas parcelas. Por otra parte, encontramos que en las parcelas de 1 hectárea la BA oscila entre 104 y 170 toneladas. Donde, las parcelas del bosque inundable, que en promedio

197

contienen 169,95 ± 0,79 Ton, albergan más biomasa por hectárea que las parcelas de tierra firme 128,73 ± 22,13 Ton. Asimismo, tomando las medias en el cambio de BA en sub parcelas de 20x20, se encontraron diferencias entre ambos tipos de bosque (W= 936157, p < 0,01), siendo mayor el cambio en el bosque inundable. De igual manera, se obtuvo que en el bosque inundable el cambio en la biomasa por hectárea al año es casi el doble al hallado en tierra firme, es decir que la acumulación de carbono es casi el doble en el bosque de Igapó (Tabla 2). Por otra parte, tenemos que tanto mBA como rBA son mayores en tierra firme (Tabla 2) siendo esta diferencia significativa en ambos casos, (W = 1468, p = 0.04) y (W=970, p < 0,01) respectivamente. Cabe anotar que los árboles grandes contribuyeron sustancialmente en las tasas de mortalidad en términos de biomasa.

180

160

140

120 Biomasa (Ton/Ha) Biomasa

100

80 2008 2009 2010 2011 2012 2013 2014 Año de muestreo

Figura 8. Cambio de biomasa aérea en las parcelas durante el periodo entre muestreos, donde tierra firme está representado en gris oscuro e inundables en gris claro.

Adicionalmente, en las parcelas de bosque inundable el cambio de la BA es positivo y mayor al encontrado en cualquier parcela de bosque de tierra firme y además ambas son muy similares en este aspecto. Por otra parte, vemos que sucede lo contrario en las parcelas de tierra firme, pues a pesar de estar a no más de 1.5 km una de la otra la dinámica de biomasa es diferente en cada una. Es más en el caso de la parcela 2,

198

el comportamiento fue diferente a las demás, pues su cambio de BA fue negativo (Figura 8). Una posible explicación para este comportamiento es que en esta parcela

se identificó que mBA (tasa de mortalidad en términos de BA) fue 3.04%, casi el doble al encontrado en las demás parcelas.

Tabla 2. Valores de biomasa aérea (BA) y acumulación de carbono, especificados por tipo de bosque.

Cambio en Acumulación Tasa de Tasa de Promedio BA Bosque BA (Ton. de carbono mortalidad en Reclutamiento total (Ton/ha) ha/año) (Ton. ha/año) BA (% año) en BA (% año) Tierra firme 0,73 0,328 128,739 2,039 2,613 Inundable 1,47 0,661 169,9495 1,320 2,201

Análisis de suelos

Figura 9. Ordenación NMDS de las características del suelo (.Arcilla (A), Limo (L), Arena (Ar), pH, CIC, CICE, CO, P, Ca, N, K, Mg) en bosques inundables (Igapó) y de tierra firme, en la Reserva de Tomogrande, Vichada.

199

En el análisis de ordenación NMDS observamos que los suelos de bosque inundables de Igapó son diferentes a los de tierra firme (Figura 9). Sin embargo, algunas muestras de tierra firme son bastante similares a las del bosque inundable. Las características que agruparon a los suelos fueron la textura, porcentaje de arena (A), limo (L) y arcilla (Ar) en la muestra, el contenido de aluminio (Al), potasio (K), fosforo (P) y parámetros como CICE, CIC y pH. En todos los casos se encontró que la diferencia era muy marcada (Anexo 2). Por otra parte, se observa que el patrón de las muestras de bosque inundables es más uniforme en comparación con tierra firme, es decir que la diferencia de las características entre las muestras es más pequeña.

Discusión

Diversidad y estructura

Con los resultados de este estudio, se confirmó la hipótesis en donde la diversidad del bosque de tierra firme es mayor a la del bosque inundable. La figura 1 y los índices de α de Fisher y eH (Tabla 1) no solo soportan lo dicho, sino que además dan una idea de lo diferente que son ambos bosques, a pesar de su proximidad espacial y de formar un dosel continuo y conectado. Este patrón es el mismo encontrado por ter Steege (2000) para toda la Amazonia y se explica debido a las condiciones de estrés a las que se ve enfrentado el bosque inundable, tales como los periodos prolongados de inundación, los cuales pueden limitar el número de especies que se establecen en el bosque de Igapó. Esto sugiere que solo las especies adaptadas a estas condiciones pueden sobrevivir (Duivenvoorden, 1996; Umaña et al. 2012).

Aunque la diversidad a nivel local, por cada parcela de 1 hectárea, es comparativamente similar a la de un bosque amazónico, a nivel regional los bosque de galería en esta zona son menos diversos que cualquier bosque amazónico (ter Steege, 2000; Correa y Stevenson, 2010). Lo anterior es congruente con la hipótesis que explica que a menor historia y área de cobertura del bosque, menor es la diversidad (Ter Steege et al., 2000), además, el efecto de borde en estos bosques angostos puede afectar negativamente a muchas especies (Laurence et al., 2002).

200

De la relación entre el diámetro y la altura de los árboles encontramos que es mayor en el bosque de tierra firme. Una posible explicación a este resultado es que en las parcelas de bosque inundable el porcentaje de árboles inclinados (8,72%) fue mucho mayor al encontrado en las parcelas de tierra firme (1,91) % (Tabla 1). Cabe resaltar que, Parolin (2009) afirma que no existe un efecto mecánico significativo del agua sobre los árboles. Sin embrago, nuestros resultados muestran que sí existe una leve diferencia, cuando se compara con tierra firme. Una explicación a este hecho, es la que propone Londoño (2011), ella señala que la periodicidad de las inundaciones y el poco drenaje del suelo pueden generar raíces superficiales y volver más susceptibles a los árboles a inclinarse o morir caídos desde la raíz.

De otra parte, encontramos que el bosque inundable posee individuos con densidades de madera más altas a las encontradas en tierra firme. Una explicación de este patrón es el requerimiento que tienen los árboles en los planos de inundación de poseer fuerte estructura estable, es decir producción de madera fuerte, debido a que se ven enfrentados a un alto estrés mecánico (Parolin, 2002).

Uno de los factores que puede estar influenciando la dinámica de estos bosques está relacionado con el tipo de muerte de los árboles. En este caso, se pudo observar que en el bosque inundable la proporción de los individuos muertos en pie es mucho mayor a la de tierra firme, en donde el tronco roto y la caída desde la raíz fueron muertes muy comunes. Este resultado puede estar relacionado con efectos climáticos, de la topografía o efectos biológicos como la herbívora o enfermedades naturales (Chao et al., 2008). En el caso del bosque inundable, los árboles, en especial los jóvenes, tienden a morir en pie debido a que en la mayoría de casos pueden sufrir anoxia, en cambio en tierra firme las muertes posiblemente fueron causadas en su mayoría por el mal anclaje, muriendo caídos desde la raíz por enfermedades, la senescencia (Quesada 2012). Composición florística

Se confirmó la hipótesis esperada donde la composición florística de los bosques de tierra firme e inundables difiere fuertemente (Figura 5), debido a las adaptaciones morfológicas y anatómicas de las especies de los planos inundables (Parolin, 2004). Adicionalmente se puede apreciar que los bosque inundables al ser menos diversos, están compuestos de unas pocas especies dominantes como es el caso de M. nítida

201

con un índice de dominancia del 16% y T. odoratissima con un 10%. En otras palabras, estas dos especies componen el 26% de los bosques inundables de galería de esta zona en términos de importancia ecológica.

Por otra parte, según Correa y Stevenson (2010), estos bosques están compuestos por individuos de las familias más abundantes en la Amazonia y bosques aledaños de los llanos orientales venezolanos (Arecaceae, Annonaceae, Rubiaceae, Moraceae, Burseraceae, Fabaceae, Chrysobalanaceae, Euphorbiaceae). Asimismo, Behling y Hooghiemstra (1999) y Veneklass y colaboradores (2005) sugieren que la flora de estos bosques no es única, pues según registros palinológicos, estos bosques eran más extensos y dominados por estas especies en periodos más húmedos del Holoceno; lo que presumiblemente influyó en la diversidad del presente. Tasa de reclutamiento, mortalidad y cambio poblacional

Las tasas anuales de reclutamiento y mortalidad observadas en los bosques de galería inundable y de tierra firme están dentro del rango reportado previamente para bosques neotropicales (Nebel et al., 2001; Phillips et al., 2004). Comparando con los resultados obtenidos por Phillips et al. (2004) tenemos que el recambio encontrado en nuestro estudio coincide con el rango reportando para los bosques de la amazonia central y del este, la cual registra tasas de recambio más bajas a las del resto de la amazonia. Cabe anotar que se encontró un patrón en la mortalidad de los árboles, en donde los arboles más jóvenes mueren en mayor proporción en el bosque inundable. Esto según Quesada et al. (2012) se debe a que en los bosques de planos de inundación los individuos en estados juveniles no sobreviven fácilmente las épocas de inundación y mueren por anoxia a pesar de sus rasgos funcionales especializados. De otra parte, se esperaría altas tasas de mortalidad y reclutamiento especialmente en los bosques inundables al estar enfrentados a altos niveles de estrés ambiental como, el fuego y el flujo de los ríos (Felfili, 1997; Kellman et al., 1998). Esto no se encontró, posiblemente porque varios de estos disturbios ocurren tanto en bosques inundables como de tierra firme. Adicionalmente, cabe anotar que se ha reportado que bajos niveles de mortalidad esta correlacionado con bajos contenidos de nutrientes en el suelo (Phillips, 2004). Otra explicación a esto, es que las especies en los bosques de galería soportan efectos de borde, que implicaría un fuerte filtro ambiental. Sin embargo, no se 202

descarta que estas especies se localicen principalmente en los bordes y actúen como barreras contra el viento y el fuego, permitiendo un funcionamiento normal al interior del bosque (Kellman, 1998). Otra posible explicación a esto es que en ambos casos hubo varias especies con pocos individuos cuyas tasas de mortalidad y reclutamiento fueron cero. Lo anterior posiblemente puede deberse a que el tiempo entre censos no fue muy largo (ca. 5 años).

De igual manera, es importante resaltar que en el bosque de tierra firme el cambio poblacional fue positivo 0,601%, contrario al bosque inundable donde fue - 0,201%, este resultado aunque nos da un indicio de la dinámica del bosque en los últimos 5 años. Sin embargo, no es un resultado muy contundente que se pueda extrapolar, pues se requieren muchos más censos a largo plazo para llegar a estar cerca a comprender como es la dinámica del bosque (Londoño, 2011).

Es importante tener en cuenta que es muy diferente la muerte de un individuo de gran envergadura, la cual provee luz, nutrientes y espacio, a la muerte de un individuo delgado que no tiene tanto impacto en el bosque (Lewis et al., 2004). Por esto se calcularon las tasas de mortalidad y reclutamiento en términos de biomasa, como lo veremos más adelante.

Tasa de crecimiento

Los valores de incremento de diámetro para los bosques de galería de tierra firme e inundable 1,50% y 0,69 % respectivamente, están dentro del rango reportado en otros bosques tropicales (0,3- 5 %) (Condit et al., 1992; Kellman, 1998). Comparando los resultados con otro bosques de galería, Kellman en 1998 encontró para un bosque de galería, en Belize a 500 msnm, un incremento de diámetro del 1,21 %, sin embargo el incluyó las especies de borde, en donde registró tasas de crecimiento mayores a las de las especies de interior. Sin embargo, nuestras parcelas no estaban tan cercanas a los bordes (> de 20 metros en todos los casos). Igualmente, nuestro rango de crecimiento es comparable con el encontrado por Sherman et al. (2012) para un bosque tropical montano, donde el valor de TCR a alturas bajas (1300 msnm) es similar al resultado de TCR para tierra firme y el TCR de alturas elevadas (1700 msnm) es más similar a nuestro resultado de bosque inundable (Tabla 2). De esta manera podemos resaltar que en promedio los bosques de galería de la reserva de Tomo-Grande tienen un promedio relativamente

203

bajo respecto a otros bosques amazónicos, sin embargo es comparable con bosques montanos y otros bosques de galería. Los resultados mostraron que efectivamente si existen diferencias en cuanto al TCR entre ambos tipos de bosque. Algunos factores que pueden estar influenciando esta diferencia, son las largas temporadas de inundación, la fertilidad y composición del suelo (Londoño, 2011). Específicamente se encontró lo mismo que Baker (2003) para los bosques húmedos tropicales; variabilidad en el crecimiento de especies y entre individuos de la misma especie. Sin embargo, esta variación es mucho más notoria en el bosque de tierra firme, en donde el rango de crecimiento es mayor, posiblemente debido a la mayor cantidad de especies presentes. Por otra parte, en el plano de inundación el nivel del agua está directamente relacionado con el grado de inundación, donde los suelos con el agua superficial limitan la aeración de las raíces y reducen el crecimiento de los árboles (Martins et al., 2015). Así mismo se sabe que en los bosques inundables existen periodos de crecimiento, es decir que los árboles entran en un estado de dormancia para evitar el estrés de las inundaciones (Parolin, 2009). De nuevo sobresalen T. chrysophilla y T. odoratissima con TCR´s de 1,02% y 1,59% respectivamente. Dos especies que pertenecen a un género muy común en los planos de inundación, muy tolerante a este ecosistema, lo cual podría explicar su habilidad de crecimiento. Además, se sabe que muchos rasgos fisiológicos y estructurales relacionados con el crecimiento y la mortalidad tienen una fuerte relación filogenética (Baker et al., 2004).

Por último, tenemos que algunas de las especies que sobresalen por ser muy comunes en estos bosques y además contar con tasas altas de recambio poblacional y de crecimiento, son además usadas o con un potencial uso para la región. Este es el caso de los individuos del genero Tachigali, los cuales son usados en la región por su valor maderable y otras especies como B.multiflora y Byrsonima japurensis que no son muy usadas en la zona, pero que existen fuentes que las catalogan como especies de un valor maderable alto (Landinez y Linares, 2000). Igualmente existen varios estudios que le dan un gran valor medicinal a B. japurensis por su capacidad de antiinflamatorio, antioxidante y antihiperalgésico (Guilhon-Simplicio et al., 2012; Guilhon-Simplicio et al. 2013). Las anteriores menciones podrían llegar a ser de beneficio para la comunidad de la región si se

204

manejan correctamente las especies anteriormente mencionadas, teniendo en cuenta algunos resultados de este estudio.

Estimación de biomasa aérea

El mayor aporte de biomasa se dio por parte del bosque inundable de aguas negras, Igapó (170 Ton/ha). Este resultado básicamente se puede explicar gracias a que en las parcelas de Igapó, el número de individuos es mayor que en las parcelas de tierra firme (Tabla 1). Igualmente, otro factor que puede estar afectando positivamente la biomasa de los bosques inundables es la densidad de madera (Casas et al.,2016 – Anexo 4), pues bajo condiciones de estrés, como inundaciones periódicas o suelos pobres en nutrientes, la densidad de madera tiende a ser mayor y el crecimiento relativo menor (Parolin, 2002, Chave, 2006).

Basándonos en el estudio de Saatchi et al. (2005), donde a partir de 544 parcelas en la Amazonia se hizo un análisis de biomasa, tenemos qué nuestros resultados de biomasa aérea encajan en el rango (100-200 Ton/ha) reportado por ellos para los bosques transicionales y estacionales, del margen sur y nororiental de la cuenca amazónica. Asimismo, en este trabajo reportan un promedio de 161 ton/ha para los bosques con planos de inundación, el cual es muy similar al resultado nuestro para este tipo de bosque (169,9 ton/ha). Igualmente, el promedio de BA en ambos tipos de bosque al parecer es mayor al encontrado en un bosque secundario de la amazonia, pero menores al promedio de BA por hectárea en un bosque maduro Amazónico de tierra firme o en bosque andino (Saatchi et al., 2005, Clark et al., 2001, Yepes et al., 2014).

Un punto que vale la pena resaltar, es la diferencia que existe en tomar las tasas de mortalidad y reclutamiento por tallo y por biomasa. En la figura 6, tenemos que el bosque inundable tuvo un cambio en la población negativo debido a que la tasa de reclutamiento fue más baja que la tasa de mortalidad. Sin embargo, al observar la tabla 2, vemos que en términos de biomasa sucede lo contrario, es decir que la tasa de reclutamiento es mucho mayor a la de mortalidad, lo que hace que el cambio en la BA sea mucho mayor a el cambio en tierra firme. De esta manera, las tasas de cambio en términos de biomasa pueden ilustrar mejor lo que sucede en el bosque, como por ejemplo la muerte de un individuo de gran porte el cual puede liberar una

205

gran cantidad de recursos (espacio, luz y nutrientes), comparado con la muerte de un individuo pequeño (Londoño, 2011).

El promedio en el incremento de la biomasa en ambos tipos de bosque (tabla 2), está por debajo del promedio reportado para bosques tropicales de tierras bajas y de montaña (Clark et al., 2001; Sherman et al., 2012). Sin embargo, en este aspecto es comparable con bosques de Hawaii, Mexico y Paragominas en el Brasil. El cambio en la BA en el bosque de tierra firme fue muy variable entre parcelas (0,726 ± 1,41 Ton/ha. año). Esto al parecer está relacionado con las tasas de mortalidad y reclutamiento,donde mBA y rBA estuvieron muy por encima del promedio en las parcelas 2 y 3 respectivamente. Por otro lado, en el bosque inundable el cambio en la BA no fue tan variable (1,43±0,09 Ton/ha. año), debido a que mBA y rBA fueron muy similares entre parcelas. Lo anterior claramente se reflejó en la acumulación de carbono anual, en donde según nuestros resultados el bosque de Igapó acumula 0,646 ± 0,04 Ton/ha anual, casi el doble del bosque de tierra firme (Tabla 2).

Análisis de suelo

En las parcelas muestreadas encontramos que los bosques de tierra firme son de suelo arcilloso mientras que los suelos del bosque inundable tienen un mayor contenido de arena y limo. Adicionalmente se encontró que en promedio los bosques de tierra firme son más ácidos que los del bosque inundable, lo cual está relacionado con una mayor concentración de nutrientes (Mg, Al, P, K, N, CO) y una elevada capacidad de intercambio catiónico efectivo (CICE) en el bosque de Igapó, ya que un elevado CICE es un indicativo de alto contenido de arcillas y materia orgánica. Lo anterior es relevante ya que la textura del suelo y la concentración de nutrientes afectan fuertemente el crecimiento de los árboles tropicales, en diferentes formas. En donde la concentración de arcilla entre el 30 y 60%, y altos niveles de N, P y K tiende a favorecer el crecimiento radial y la biomasa de las plantas (Clark et al., 1998; Castilho et al., 2006). Adicionalmente, se ha encontrado que la CICE está muy correlacionada con la biomasa aérea en los bosques tropicales (Quesada et al., 2012).

Basados en nuestros resultados podemos inferir que el mayor número de individuos, y por tanto mayor biomasa aérea, en el bosque inundable es producto tanto de la

206

capacidad y características que tiene este suelo a nivel fisicoquímico como de la alta densidad de madera encontrada en los individuos de este bosque. En ambientes estresantes, tales como planos de inundación, las plantas invierten más en adaptaciones que mejoren su tolerancia, en cambio en ambientes no estresantes las plantas invierten más en la competencia (Grime, 1977). Así pues, la abundancia en las parcelas de bosque inundable se podía explicar debido a un balance en la reproducción, es decir el tamaño de la población, en las especies tolerantes (Martins et al., 2015). Asimismo, estas adaptaciones afectan la distribución y estructura de las comunidades (Martins et al., 2015), como es el caso de la correlación entre la relación arena:arcilla y la composición de las especies, donde a mayor concentración de arcilla hay más especies presentes. Es otras palabras, este resultado apoyaría la gran diferencia en la composición florística que se encontró en este estudio.

Conclusiones

Se confirma el patrón encontrado a lo largo de los bosques del neotrópico en donde los bosques de tierra firme tienen una mayor riqueza de especies y son más diversos que los bosques con planos de inundación, sin embargo, la abundancia promedio por hectárea es mayor en este bosque. En este bosque encontramos un mayor número de árboles inclinados, lo que puede generar más luz llegando al sotobosque y más posibilidades de reclutamiento. En términos de estructura identificamos que la relación del DAP vs la altura fue mayor en tierra firme debido a que en el bosque de Igapó los arboles tenían un promedio más bajo de altura y además un alto porcentaje de individuos inclinados. Asimismo, debido a las adaptaciones anatómicas y fisiológicas de los árboles del bosque inundable, la composición de ambos bosques fue totalmente diferente. En cuanto a la dinámica del bosque tenemos que sí existen diferencias en las tasas de reclutamiento y crecimiento relativo entre tipos de bosque. Por lo contrario, no existen diferencias en la tasa de mortalidad y cambio poblacional por individuo, pero sí hubo disimilitudes el tasa de mortalidad por biomasa debido a que la muerte de árboles de gran tamaño en el bosque de tierra firme marcan la diferencia. Cabe anotar que el cambio poblacional en el bosque de tierra firme fue positivo y en el plano de inundación negativo, lo cual nos hace inferir sobre lo que sucedió durante

207

el periodo entre muestreos. Asimismo, tanto la biomasa promedio total como el cambio en la biomasa por hectárea fueron mayores en el bosque con plano de inundación debido a la abundancia y homogeneidad en las tasas demográficas. A partir de lo anterior, obtuvimos que la acumulación de carbono resulto en 0,66 y 0,32 (Ton/Ha. Año) en el bosque de Igapó y de tierra firme respectivamente. Por último, según los parámetros fisicoquímicos que mostró el suelo, el bosque más fértil fue el bosque inundable, sin embargo, este resultado no se vio reflejado en las tasas demográficas, pero al parecer sí está correlacionado con la abundancia y biomasa total en este bosque.

Aunque los resultados obtenidos nos dan una idea sobre la dinámica de los bosques de galería de la Reserva de Tomogrande, es importante continuar haciendo censos periódicamente y a largo plazo incluyendo datos sobre topografía, variación climática y de los pulsos de inundación para de este modo llegar a entender mejor el funcionamiento y productividad de estos bosques. Adicionalmente es relevante incluir las plántulas y las especies de borde, pues pueden estar jugando un papel importante.

Agradecimientos

A la Facultad de Ciencias por el apoyo económico. A Alejandra Jiménez, Luz Dary Rivas, John Fredy por su colaboración en la fase de campo.

208

Referencias bibliográficas

Alvarez, E., Duque, A., Saldarriaga, J., Cabrera, K., de las Salas, G., del Valle, I., … Rodríguez, L. (2012). Tree above-ground biomass allometries for carbon stocks estimation in the natural forests of Colombia. Forest Ecology and Management, 267, 297–308. doi:10.1016/j.foreco.2011.12.013

Behling, H., & Hooghiemstra, H. (1999). Environmental history of the Colombian savannas of the Llanos Orientales since the Last Glacial Maximum from lake records El Pinal and Carimagua *. Journal of Paleolimnology, 21(4), 461–476.

Berazategui, M. (2012). Evaluación de las teorías Neutral y de Nichos en comunidades temporales. Tesis de maestría, Universidad de la Republica, Uruguay.

Calderón García, A. (2010). Revisión literaria de los bosques de galería (pp. 6– 20). Tomado de: http://medcontent.metapress.com/index/A65RM03P4874243N.pdf

Casas Caro, L. F, & Stevenson, P. R (2013). Variación de biomasa aérea y densidad de madera en bosques de tierras bajas con planos de inundación de aguas negras y aguas blancas. Tesis de maestría, Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, Colombia.

De Castilho, C. V., Magnusson, W. E., de Araújo, R. N. O., Luizão, R. C. C., Luizão, F. J., Lima, A. P., & Higuchi, N. (2006). Variation in aboveground tree live biomass in a central Amazonian Forest: Effects of soil and topography. Forest Ecology and Management, 234(1-3), 85–96. doi:10.1016/j.foreco.2006.06.024.

Chao, K.-J., Phillips, O. L., Gloor, E., Monteagudo, A., Torres-Lezama, A., & Martínez, R. V. (2008). Growth and wood density predict tree mortality in Amazon forests. Journal of Ecology, 96(2), 281–292. doi:10.1111/j.1365-2745.2007.01343.x

Chave, J., Andalo, C., Brown, S., Cairns, M. a, Chambers, J. Q., Eamus, D., Yamakura, T. (2005). Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia, 145(1), 87–99. doi:10.1007/s00442-005-0100-x.

Chave, J., Muller-Landau, H., Baker, T. R., Easdale, T. A., Ter Steege, H., & Webb, C. (2006). Refional and phylogenetic variation of wood density across neotropical tree species, 16(6), 2356–2367.

Clark, D. B., Clark, D. a., & Read, J. M. (1998). Edaphic variation and the mesoscale distribution of tree species in a neotropical rain forest. Journal of Ecology, 86(1), 101–112. doi:10.1046/j.1365- 2745.1998.00238.x.

Clark, D. A., Brown, S., Kicklighter, D. W., Chambers, J. Q., Tomlinson, J. R., Ni, J., & Holland, E. A. (2001). Net primary production in tropical forests : an evaluation and synthesis of existing field data. Ecology Applications, 11(2), 371–384.

Colwell, R. K. 2013. Estimate S: Statistical estimation of species richness and shared species from samples. Versión 9.0.1

Condit, R., Hubbell, S. P., & Foster, R. B. (1992). Short-Term dynamics of a neotropical forest. BioSciencie, 42(11), 822–828.

Condit, R. (1998). Tropical Forest Census Plots (pp. 1–15). New York: Springer - Verlag Berlin Heidelber and R.G Company Georgetown.

Connel, J. H. (1978). Diversity in Tropical Rain Forests.pdf. Science, Vol. 199, 1302-1309.

209

Correa-Gómez, D. F., & Stevenson, P. R. (2010). Estructura y diversidad de bosques de los llanos orientales colombianos (Reserva Tomo-Grande, Vichada), Revista Orinoquia, 31–45.

Duivenvoorden, J. (1996). Patterns of tree species richness in rain forest of the middle Caqueta area, Colombia, NW Amazonia. Biotropica, 28, 142–158.

Ettema, C. H., & Wardle, D. A. (2002). Spatial soil ecology. Trends in Ecology and Evolution, 17(4), 177–183.

Felfili, J. M. (1997). Diameter and height distributions in a gallery forest tree community and some of its main species in central Brazil over a six-year period ( 1985-1991 ), 20(2), 155–16

Ferry, B., Morneau, F., Bontemps, J.-D., Blanc, L., & Freycon, V. (2010). Higher treefall rates on slopes and waterlogged soils result in lower stand biomass and productivity in a tropical rain forest. Journal of Ecology, 98(1), 106–116. doi:10.1111/j.1365-2745.2009.01604.x.

Fisher AA, Corbet AS, Williams CB. The relation between the number of species and the number of individuals in a random sample of an animal population. J Anim Ecol 1943; 12: 42-58.

Furch, K. (1997). Chemistry of V arzea and Igapo Soils and Nutrient Inventory of Their Floodplain Forests. Ecological Studies, 126(2).

Grime, J. . (1977). Evidence for the existence of three primary strategies in plants and its relevance to ecological and evlutionary theory. The American Naturalist, 111(982), 1169–1194.

Guilhon-Simplicio, F., Pinheiro, C. C. D. S., Conrado, G. G., Barbosa, G. D. S., Santos, P. A. Dos, Pereira, M. D. M., & Lima, E. S. (2012). Anti-inflammatory, anti-hyperalgesic, antiplatelet and antiulcer activities of Byrsonima japurensis A. Juss. (Malpighiaceae). Journal of Ethnopharmacology, 140(2), 282–6. doi:10.1016/j.jep.2012.01.018

Guilhon-Simplicio, F., Souza, T. P. De, Alonso, A. A., Almeida, D. O. De, Alexandre, P., Ohana, D. T., … Pereira, M. D. M. (2013). Antioxidant activity of a standardized extract of Byrsonima japurensis A . Juss . ( Malpighiaceae ) stem bark. Medicinal Plants Research, 7(26), 1926–1930. doi:10.5897/JMPR12.113.

Henderson A. Evolution and ecology of palms. Bronx, Nueva York, The New York Botanical Garden Press, 2002.

Holdridge, L.R., Grenke, W.C., Hatheway, W.H., Liang, T., Tosi, J.A., (1971). Forest Environments in Tropical Life Zones. Pergamon Press, Oxford.

Hubbell, S.P. (2001) The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton, NJ.

Hurtado G, González OC, Montaña JA. Tercera Parte: Aspectos departamentales. En Henríquez M (Editor), Atlas climatológico de Colombia. Bogotá, Instituto de Hidrología, Meteorología y Estudios Ambientales de Colombia IDEAM, 2005.

Jost, L. (2006). Entropy and diversity. OIKOS, 363–369.

Kellman, M., Tackaberry, R., & Rigg, L. (1998). Structure and function in two tropical gallery forest communities: implications for forest conservation in fragmented systems. Journal of Applied Ecology, 35(2), 195–206. doi:10.1046/j.1365-2664.1998.00300.x

Kozlowski, T. T. (1984). Flooding and Plant Growth . In FLOODING AND PLANT GROWTH (pp. 129– 163). Elsevier. doi:10.1016/B978-0-12-424120-6.50009-2

210

Landinez, A., & Linares, E. (2006). Plantas dendroenergéticas utilizadas por una comunidad indígena Piapoco en Guanía, Colombia, (2000).

Laurance, W. F., Lovejoy, T. E., Vasconcelos, H. L., Bruna, E. M., Didham, R. K., Stouffer, P. C., Sampaio, E. (2002). Ecosystem Decay of Amazonian Forest Fragments : a 22-Year Investigation. Conservation Biology, 16(3), 605–618.

Lewis, S. L., Phillips, O. L., Sheil, D., Vinceti, B., Baker, T. R., Brown, Graham, Andrew W.H., Niro, H., David W., Laurance, W. F., Lejoly, J.,Malhi, Y.,Monteagud, A.,Nuñez Vargas, P.,Sonke, B. (2004). Tropical forest tree mortality , recruitment and turnover rates : calculation , interpretation and comparison when census intervals vary. Journal of Ecology, 92, 929–944.

Londoño, A. C. (2011). Flora and dynamics of an upland and a floodplain forest in Peña Roja , Colombian Amazonia. Medellin.

Martins, K. G., Marques, M. C. M., dos Santos, E., & Marques, R. (2015). Effects of soil conditions on the diversity of tropical forests across a successional gradient. Forest Ecology and Management. doi:10.1016/j.foreco.2015.04.018.

Nebel, G., Kvist, L. P., & Vanclay, J. K. (2001). Forest dynamics in flood plain forests in the Peruvian Amazon : effects of disturbance and implications for management. Forest Ecology and Management, 150(1), 79–92.

Niklas, K. J., & Enquist, B. J. (2001). Invariant scaling relationships for interspecific plant biomass production rates and body size. Proceedings of the National Academy of Sciences of the United States of America, 98(5), 2922–7. doi:10.1073/pnas.041590298.

Parolin, P. (2002). Radial gradients in Wood specific gravity in trees of central amazonian floodplains, 23(4), 449–457.

Parolin, P., De Simone, O., Haase, K., Waldhoff, D., Rottenberger, S., Kuhn, E., … Pledade, T. F. (2004). Central Amazonian floodplain forest: Tree adaptations in a pulsing system. The Botanical Review, 70(3), 357–380.

Parolin, P. (2009). Submerged in darkness: adaptations to prolonged submergence by woody species of the Amazonian floodplains. Annals of Botany, 103(2), 359–76. doi:10.1093/aob/mcn216

Peacock, J., Baker, T.R., Lewis, S.L., Lopez-Gonzalez, G. & Phillips, O.L. 2007. The RAINFOR database: monitoring forest biomass and dynamics. Journal of Vegetation Science 18: 535–542.

Phillips, O. L., Hall, P., Gentry, A. H., Sawyer, S. A., & Vasquez, R. (1994). Dynamics and species richness of tropical rain forests. Ecology, 91(March), 2805–2809.

Phillips, O. L., Baker, T. R., Arroyo, L., Higuchi, N., Killeen, T. J., Laurance, W. F, Vinceti, B. (2004). Pattern and process in Amazon tree turnover, 1976-2001. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 359(1443), 381–407. doi:10.1098/rstb.2003.1438

Phillips, O., Baker, T., Feldpausch, T., & Brienen, R. (2009). RAINFOR manual de campo para la remedición y establecimiento de parcelas.

Phillips, O. L., & Lewis, S. L. (2014). Recent changes in tropical forest biomass and dynamics. Forest and Global Change, (Cambridge University Press), Chapter 4.

Prance, G. T. (1979). Notes on the vegetation of Amazonia III. The terminology of Amazonian forest types subject to inundation. Brittonia, 31(1), 26–38.

211

Quesada, C. a., Phillips, O. L., Schwarz, M., Czimczik, C. I., Baker, T. R., Patiño, S., … Lloyd, J. (2012). Basin-wide variations in Amazon forest structure and function are mediated by both soils and climate. Biogeosciences, 9(6), 2203–2246. doi:10.5194/bg-9-2203-2012.

R development core team. 2014. R: A language&enviroment for statistical computing. Disponible: http://www.R-project.org

Rangel-Ch JO, Lowy-C PD, y Aguilar-P M , Métodos de estudio de la vegetación. Colombia Diversidad Biótica II: Tipos de Vegetación en Colombia. Bogotá, Universidad Nacional de Colombia, Instituto de Ciencias Naturales, 1997. Pp. 59-87.

Redford, K. H., Fonseca, G. A. B., & Redford, K. H. (2014). The Role of Gallery Forests in th Zoogeography of the Cerrado ’ s Non-volant Mammalian Fauna The Role of Gallery Forests in the Zoogeography of the Cerrado ’ s Non-volant Mammalian Fauna1, 18(2), 126–135.

Romero M.N., Maldonado-Ocampo J.A., Bogotá-Gregory J.D., Usma J.S., Umaña-Villaveces A.M., Murillo J.I.,Restrepo-Calle S., Álvarez M.,Palacios-Lozano M.T., Valbuena M.S., Mejía S.L., Aldana.Domínguez J. y Payán E. Informe sobre el estado de la biodiversidad en Colombia 2007- 2008: Piedemonte orinoquense, sabanas y bosques asociados al norte del río Guaviare. Instituto de Investigaciones de recursos Biológicos Alexander von Humboldt. Bogotá D.C.,Colombia. 133p.

Sherman, R. E., Fahey, T. J., Martin, P. H., & Battles, J. J. (2012). Patterns of growth, recruitment, mortality and biomass across an altitudinal gradient in a neotropical montane forest, Dominican Republic. Journal of Tropical Ecology, 28(05), 483–495. doi:10.1017/S0266467412000478

Ter Steege, H., Sabatier, D., Castellanos, H., Van Andel, T., Duivenvoorden, J., Adalardo de Oliveira, A.,Renske, E., Lilwah,R.,Maas, P., Mori, S. (2000). An analysis of the floristic composition and diversity of Amazonian forests including those of the Guiana Shield. Journal of Tropical Ecology, 16, 801–828.

Timothy R. Baker, David F. R. P. Burslem and Michael D. Swaine (2003). Associations between tree growth, soil fertility and water availability at local and regional scales in Ghanaian tropical rain forest. Journal of Tropical Ecology, 19, pp 109-125. doi:10.1017/S0266467403003146.

Umaña, M. N., Norden, N., Cano, A., & Stevenson, P. R. (2012). Determinants of plant community assembly in a mosaic of landscape units in central Amazonia: ecological and phylogenetic perspectives. PloS One, 7(9), e45199. doi:10.1371/journal.pone.0045199

Veblen, T. T., Kitzberger, T., & Villalba, R. (2004). Nuevos paradigmas en ecologia y su influencia sobre el conocimiento de la dinamica de los bosques del sur de argentina y chile. In M. . Arturi, J. . Frangi, & J. . Goya (Eds.), .

Veneklaas, E. J., Fajardo, A., Obregon, S., & Lozano, J. (2005). Gallery forest types and their environmental correlates in a Colombian savanna landscape. Ecography, 2(October 2004), 236– 252.

Williamson, G. B., & Wiemann, M. C. (2010). Measuring wood specific gravity...Correctly. American Journal of Botany, 97(3), 519–24. doi:10.3732/ajb.0900243

212

Anexos

Anexo 1. Listado de las 15 especies más importantes en los bosque de galería inundable y de tierra firme en la Reserva de Tomo-Grande, donde: D = Dominancia relativa, ρ = Densidad relativa, F = Frecuencia relativa, IVI = Índice de importancia.

Especie D ρ F IVI Mabea nitida 16,44 11,31 8,00 35,75 Tachigali odoratissima 10,14 15,80 6,50 32,44 Tachigali chrysophylla 6,58 11,19 6,00 23,77 Licania heteromorpha 6,51 7,49 5,33 19,33 Eschweilera parvifolia 6,23 4,84 6,67 17,74

Guatteria brevicuspis 4,25 7,28 4,67 16,19 Caraipa llanorum 3,01 7,68 4,00 14,70 Laetia suaveolens 5,21 3,07 5,00 13,28 Duroia micrantha 4,79 5,46 3,00 13,26

Bosque Inundable Bosque Licania apetala 3,90 3,12 4,50 11,52 Byrsonima japurensis 3,01 2,63 4,50 10,15 Pouteria elegans 2,88 2,65 4,00 9,53 Tovomita longifolia 4,11 1,60 3,17 8,88 Discocarpus essequeboensis 3,29 1,59 3,50 8,38 Calycorectes sp16 2,47 1,31 2,33 6,11 Attalea maripa 8,12 11,79 5,39 25,30 Jacaranda copaia 5,71 15,04 3,21 23,96 Bocageopsis multiflora 6,58 2,77 5,08 14,43 Tetragastris panamensis 3,88 5,81 3,11 12,80

Oenocarpus bataua 5,41 4,58 2,80 12,79 Pseudolmedia laevis 5,71 1,92 4,15 11,77

rrafirme Couepia glabra 3,22 3,88 3,21 10,31 Guatteria metensis 3,00 3,03 2,38 8,41 Pourouma aurea 3,44 2,47 2,07 7,98 Xylopia polyantha 2,93 1,56 3,11 7,60 Bosque de Tie de Bosque Cupania scrobiculata 3,07 0,95 2,80 6,82 Conceveiba tristigmata 2,78 1,02 2,90 6,70 Ocotea schomburgkiana 2,12 1,70 2,18 6,00 Himatanthus articulatus 1,68 2,09 2,18 5,95 Macrocnemum sp01 2,34 0,71 2,28 5,34

213

Anexo 2. Valores promedio de medidas de todas las variables muestreadas en el suelo de los bosques de tierra firme e inundable. Arcilla (A), Limo (L), Arena (Ar), Carbono orgánico (CO), Nitrógeno (N), Fosforo (P), pH, Acidez de cambio (Al), capacidad de intercambio catiónico (CIC), capacidad de intercambio catiónico efectivo (CICE), Calcio (Ca) Magnesio (Mg) Potasio (K) Sodio (Na). El valor p indica el valor de comparación estadística, obtenido de la prueba de Wilcoxon, de los parámetros físico-químicos del suelo, donde los valores en negrilla indican que existe una diferencia estadística.

Bosque Variable Inundable Tierra firme Valor p % A 15,00 54,59 < 0,01 % L 49,90 21,93 < 0,01 % Ar 34,70 23,20 < 0,01 % CO 3,80 3,12 0,05 % N 0,27 0,18 0,01 P(mg/Kg) 8,37 1,90 < 0,01 pH 4,56 3,68 < 0,01

Al (cmolc/Kg) 4,11 2,79 < 0,01 CIC 15,82 10,36 < 0,01 CICE 5,18 3,83 < 0,01

Ca (cmolc/Kg) 0,18 0,26 0,03

Mg (cmolc/Kg) 0,16 0,15 0,15

K (cmolc/Kg) 0,23 0,10 < 0,01

Na (cmolc/Kg) 0,50 0,53 0,05

214

ANEXO 4 – Specific gravity of woody tissue from lowland Neotropical plants: variance among forest types

Casas-Caro, L., Aldana, A. M., Henao-Diaz, F., Villanueva, B. & Stevenson, P. R. 2016. Specific gravity of woody tissue from lowland Neotropical plants: variance among forest types. Artículo de datos sometido para publicación en Ecology, el 26 de agosto de 2016, actualmente en revisión.

215

Specific gravity of woody tissue from lowland Neotropical plants: variance among forest types

Luisa Fernanda Casas-Caro1*, Ana María Aldana2, Francisco Henao-Diaz2, Boris Villanueva3, Pablo R. Stevenson2

1Laboratorio de Ecología de Bosques Tropicales y Primatología, Universidad de los Andes. 111711 Bogotá D.C., Colombia. Actual linked Fundación Natura Colombia.

2Laboratorio de Ecología de Bosques Tropicales y Primatología, Universidad de los Andes. 111711 Bogotá D.C., Colombia.

3Grupo de Investigación en Biodiversidad y Dinámica de Ecosistemas Tropicales - GIBDET, Universidad del Tolima, Colombia.

*Corresponding author: Luisa Fernanda Casas-Caro, phone +573164727050, email: [email protected]

216

INTRODUCTION

Working with above ground biomass estimations for lowland forests in Colombia, we identified an information gap for many functional traits of a high number of tree species in our dataset. In our research we are interested in evaluating if environmental differences between forests, such as flooding regime, are important factors determining differences in above ground biomass. Since wood specific gravity is an important factor for estimating biomass stocks as it is one of the three components included in biomass allometric equations (Chave et al. 2005, Alvarez et al. 2012). However, when there is no information for a particular species, many authors working usually extrapolate this trait value from the closest relatives from available information (Chave et al. 2006a, Zanne et al. 2009) assuming trait phylogenetic conservatism (Chave et al. 2006b). Williamson and Wiemann (2010) called for researchers to be cautious of the information published and to use appropriate measurement techniques when dealing with wood density. Under this scenario, we thought that for our research it was necessary to generate a new data set that included information from new species, that was measured following standard protocols, and that accounted for the variation between forest types. Our study system includes terra firme forests and flooded forests that we have classified as either igapó or várzea sensu Prance (1979).

217

METADATA

CLASS I. DATA SET DESCRIPTORS

A. Data set identity: Specific gravity of woody tissue from lowland Neotropical plants: variance among forest types

B. Data set identification code: Spec_Grav_Lowland_NeoTrop_Plants.txt

C. Data set description

Originators: Luisa Fernanda Casas-Caro1, Ana María Aldana2, Francisco Henao- Diaz2, Boris Villanueva3, Pablo R. Stevenson2

1Laboratorio de Ecología de Bosques Tropicales y Primatología, Universidad de los Andes. 111711 Bogotá D.C., Colombia. Actual linked Fundación Natura Colombia.

2Laboratorio de Ecología de Bosques Tropicales y Primatología, Universidad de los Andes. 111711 Bogotá D.C., Colombia.

3Grupo de Investigación en Biodiversidad y Dinámica de Ecosistemas Tropicales - GIBDET, Universidad del Tolima, Colombia.

Abstract: Wood density, or more precisely, wood specific gravity, is an important parameter when estimating above ground biomass, which has become a central tool for the management and conservation of forests around the world. When using biomass allometric equations for tropical forests, researchers are often required to assume phylogenetic trait conservatism and assign genus and family level mean wood specific gravity values to many woody species. The lack of information on this

218

trait for many Neotropical plant species, has led to an imprecise estimation of the biomass stored in Neotropical forests. The data presented here has information of woody tissue specific gravity from 2,708 individual stems for 395 species, including trees, palms, lianas and hemi-epiphytes of lowland tropical forests in Colombia. This dataset was produced by us collecting wood cores from woody species in five localities in the Orinoco and Magdalena Basins in Colombia. Our dataset complements other datasets previously published.

D. Key words: Wood specific gravity, wood density, above ground biomass, plant functional trait.

CLASS II. RESEARCH ORIGIN DESCRIPTORS

A. Overall project description

Identity: Specific gravity of woody tissue from lowland Neotropical plants: variance among forest types

Originators: L.F. Casas-Caro, A.M. Aldana, F. Henao-Díaz, B. Tamayo. P.R. Stevenson

Period of Study: The database was compiled between 2010 and 2015.

Objectives: To produce a wood specific gravity dataset of lowland Neotropical plants

Abstract: Same as above.

219

Sources of funding: This project was funded by Ecopetrol and Universidad de los Andes.

Site description: Samples were collected at five localities in the Magdalena and Orinoco Basins (<1000 m.a.s.l.), classified as tropical humid forest zone of life (Holdridge 1967). Most of samples were collected from 1-hectare permanent forest plots, the rest from identified species at field.

Methods

Woody tissue samples were collected with an increment borer using the protocol described in Chave (2005). To calculate wood specific gravity (WSG, sensu Williamson and Wiemann 2010), fresh volume and dry mass were measured for each sample. Fresh volume was obtained using two methods depending sample consistency. Cylindrical samples that maintained the increment borer diameter (D=5.1mm) were measured in length (L) with a vernier scale and then was calculate the volume using the formula π/4*D^2L. Irregular samples were processed using the volumetric procedure, which require sample saturation with water overnight and then submerge it in a test tube over a precision scale. Dry mass was obtained in a precision scale with dried samples at 103°C for 72h.

The data-file we are sharing contains information for each individual stem sampled: coordinates of the plot or site were the samples were collected; forest type, a classification of the plot based on the frequency and intensity of the flood, as well as the type of river that floods them; order and family classification of the species based on the APG III classification system; scientific name and author of the species; habit as observed by us in the field; specific gravity of the woody tissues, we emphasize in the terms “woody tissue” as palms and other woody growth-forms do not have real wood (xylem tissues).

220

CLASS III. DATA SET STATUS AND ACESSIBILITY

A. Status Latest update: 2016 Latest archive data: 2016 Metadata status: metadata is complete. Data verification: all data have been quality checked.

B. Accessibility Storage location and medium: Wiley Online Library

Contact person: Pablo R. Stevenson, Laboratorio de Ecología de Bosques Tropicales y Primatología, Universidad de los Andes, 111711 Bogotá D.C., Colombia, phone: +5713394949 ext. 3770

Copyright restrictions: None

Proprietary restrictions: None

Costs: None

CLASS IV. DATA STRUCTURAL DESCRIPTORS

A. Data set file

Identity: Spec_Grav_Lowland_NeoTrop_Plants.txt

Size: 2708 rows + header, 9 columns (244 KB)

221

Format and storage: Text, tab delimited file (.txt)

B. Variable information Each row of the dataset represents one plant individual and shows information about its wood specific gravity, taxonomic classification and location.

Variable Name Variable Definition Unit ID Consecutive number Coordinates Latitud-longitud Decimal degrees Forest_Type A classification of the plot based on the frequency and intensity of the flood, as well as the type of river: Terra firme: non flooded Varzea: flooded by river of white water Igapo: flooded by river of black water Order Classification based on the APG III classification system Family Classification based on the APG III classification system Scientific_name Genus_species Authority Species authors Habit Kind of habit: tree/liana/palm/hemi (hemi-epiphytes) Specific_gravity Mass/volume g/cm3 C. Data limitations We collected samples from palms, because other data sets have reported “wood density” values for species of this family, but we consider that data from this trait in

222

palms should be treated carefully, because there is great variation among individuals, specially due to differences in diameter. When possible, species and family allometric equations for palms should be used (Goodman et al. 2013), avoiding to use “wood density” values.

All species were determined by the authors of the dataset and flora experts of Magdalena and Orinoco Basins, but taxonomic classification was not reviewed by an expert of each problematic group, for example Myrtaceae or Rubiaceae, so the actual classification could have an uncertainty degree on some taxonomic groups. However the database is being permanently reviewed and updated with the constant revision, so the possible taxonomic errors are being solved.

CLASS V. SUPLEMENTAL DESDRIPTORS Specific gravity ranged from 0.093 in Astrocaryum chambira (Arecaceae) to 1.056 in Hymenaea courbaril (Fabaceae). Both mean and median were near to 0.6 (Fig. 1).

223

Figure1. Histogram of woody tissue specific gravity data from 2,708 individual stems for 395 Neotropical woody species.

When comparing between collection plots, we found differences between forest types: igapo, varzea and terra firme (ANOVA, F=12.95 p<0.001). Samples collected in varzea have lower specific gravity values than igapo and terra firme (Figure 2). We suggest that the differences found could be associated to structural and nutritional soil features, Wittmann et al. 2006), as well as floristic composition and stress given by flooding seasons (Parolin and Wittmann 2010, Wittmann et al. 2013). The difference is sustained after we remove the effect of low density tissue species such as palms (ANOVA, F= 14.63 p=4.83-07 results not shown )

224

Figure 2. Wood specific gravity difference between forest type in the Neotropics. Asterisks are drawn to represent the different group as determined by a Tukey test.

Acknowledgments: To Nathali Rodriguez, Sasha Cárdenas, Maria Fernanda Torres, Juan Camilo Muñoz, Vanessa Rubio, Erika Rodríguez, Marcela Córdoba, Indira León, Isabel Restrepo, Guillermo Rivas, Juan Sebastián González, Alejandra Jiménez, Camilo Quiroga, Eduardo Pinel, Edna Beltrán, Diana Acosta, Felipe Aramburo, Ángela Sánchez, who helped us gather and process data at Laboratorio de Ecología de Bosques Tropicales y Primatología. This project was funded by Facultad de Ciencias Universidad de Los Andes & Ecopetrol (Convenio DHS No. 135 de 2009).

225

Literature cited

Alvarez, E., A. Duque, J. Saldarriaga, K. Cabrera, G. de las Salas, I. del Valle, A. Lema, F. Moreno, S. Orrego, and L. Rodríguez. 2012. Tree above-ground biomass allometries for carbon stocks estimation in the natural forests of Colombia. Forest Ecology and Management 267:297–308.

Chave, J. 2005. Measuring wood density for tropical forest trees A field manual for the CTFS sites.

Chave, J., C. Andalo, S. Brown, M. A. Cairns, J. Q. Chambers, D. Eamus, H. Fölster, F. Fromard, N. Higuchi, T. Kira, J.-P. Lescure, B. W. Nelson, H. Ogawa, H. Puig, B. Riéra, and T. Yamakura. 2005. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia 145:87–99.

Chave, J., H. C. Muller-Landau, T. R. Baker, T. A. Easdale, T. E. R. Hans Steege, and C. O. Webb. 2006a. Regional and phylogenetic variation of wood density across 2456 neotropical tree species. Ecological Applications 16:2356–2367.

Chave, J., H. C. Muller-Landau, T. R. Baker, T. A. Easdale, H. ter Steege, and C. O. Webb. 2006b. Regional And Phylogenetic Variation Of Wood Density Across 2456 Neotropical Tree Species. Ecological Applications 16:2356–2367.

Goodman, R. C., O. Phillips, D. del Castillo Torres, L. Freitas, S. T. Cortese, A. Monteagudo, and T. R. Baker. 2013. Amazon palm biomass and allometry. Forest Ecology and Management.

Holdridge, L. R. 1967. Life zone ecology.:206.

Parolin, P., and L. Ferreira. 1998. Are there differences in specific wood gravities between trees in várzea and igapó (central amazonia)? Ecotropica 4:25–32.

Parolin, P., and F. Wittmann. 2010. Struggle in the flood: tree responses to flooding stress in four tropical floodplain systems. AoB Plants 2010:plq003–plq003.

Prance, G. 1979. Notes on the vegetation of Amazonia III. The terminology of Amazonian forest types subject to inundation. Brittonia 31:26–38.

Williamson, G. B., and M. C. Wiemann. 2010. Measuring wood specific gravity...correctly. American Journal of Botany 97:519–524.

Wittmann, F., Schöngart, J., Parolin, P., Worbes, M., Piedade, M. T., & Junk, W. J. (2006). Wood specific gravity of trees in Amazonian white-water forests in relation to flooding. IAWA Journal, 27(3), 255-268.

Wittmann, F., E. Householder, M. T. F. Piedade, R. L. de Assis, J. Schöngart, P. Parolin, and W. J.

226

Junk. 2013. Habitat specifity, endemism and the neotropical distribution of Amazonian white- water floodplain trees. Ecography 36:690–707.

Zanne, A. E., G. Lopez-Gonzalez, D. A. A. Coomes, J. Ilic, S. Jansen, S. L. S. L. Lewis, R. B. B. Miller, N. G. G. Swenson, M. C. C. Wiemann, and J. Chave. 2009. Global wood density database. Dryad Digital Repository.

227

Biofuels illustration by Tatiana Arocha

228