Replace This with the Actual Title Using All Caps

HABITAT LOSS, FRAGMENTATION AND CLIMATIC DRIVERS OF AVIAN DIVERSITY ACROSS THE ANDES AND IMPLICATIONS FOR CLIMATE CHANGE

A Dissertation Presented to the Faculty of the Graduate School of Cornell University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

by Cristian Steven Sevillano-Ríos May 2020

HABITAT LOSS, FRAGMENTATION AND CLIMATIC DRIVERS OF AVIAN DIVERSITY ACROSS THE ANDES AND THEIR IMPLICATIONS FOR CLIMATE CHANGE

Cristian Steven Sevillano-Ríos, Ph. D

Cornell University 2020

ABSTRACT

The diversity of life on our planet is under siege. Around the world habitat loss, degradation, and fragmentation are among the main threats to biodiversity, with negative outcomes exacerbated by climate change. Our poor understanding of the mechanisms governing ecological responses to global change, especially in tropical montane systems, constrains our ability to predict and mitigate these threats to biodiversity. This dissertation addresses the effects of habitat loss, fragmentation and climatic drivers on avian diversity across the Andes, with a special focus on the implications for climate change. Specifically, we examined (1) sensitivity of birds to forest cover and patch size within one of the most threatened ecosystems of the Andes,

Polylepis forests, (2) how factors related to climate, landscape configuration and human activities at regional scales shape patterns of avian species richness and endemism along elevational gradients in the western slope of the Andes, (3) the extent to which relationships between elevation and avian diversity at continental scales are influenced by complex interactions between temperature and precipitation, and (4) the degree to which avian communities in the Andes can persist under future climate scenarios. We surveyed 88 bird species across 59 Polylepis patches within five glacier

ii valleys (3,300 to 4,700 m a.s.l.) of Cordillera Blanca (Ancash-Peru). We also compiled elevational ranges of 212 species within the region and 2,384 across the continent, together with climatic and environmental covariables across 16 elevational gradients across the Andes of Colombia, Ecuador, Peru, Bolivia, Argentina and Chile. At landscape scales, avian communities at high elevations seemed adapted to naturally patchy Polylepis landscapes. Species richness was greatest within small fragments and areas with reduced forest cover. Polylepis specialists persisted within small patches, as long as landscapes retained more than 10% of forest (>400 ha) within glacial valleys. At regional scales along the western slope of the Central Andes

(Ancash, Peru), avian species richness and endemism peaked at mid-elevations (3,300

– 4,300 masl) where temperatures and precipitation were comparably moderate (i.e, intermediate levels). At continental scales, the influence of climatic factors on elevational diversity gradients differed between xeric and mesic slopes. Diversity gradients were best explained by the interaction between temperature and precipitation along the xeric slopes of the western central Andes of Peru and Chile, but primarily by temperature along mesic slopes, like the East slope to the Amazon. Based upon four climate scenarios for 2070, changes in temperature and precipitation in the western central Andes will be most severe above 2,000 masl, whereas human activities will intensify and expand below 500 and above 3,000 masl. Collectively, these results provide evidence for three overarching conclusions: 1) even fragmented or heterogeneous Polylepis landscapes have potential to adequately support avian communities in High Andean systems as well as provide habitat for numerous species of conservation concern; 2) elevational gradients in avian diversity across the Andes

iii seem to be shaped more than by climatic than landscape or social factors; and 3) future changes in climate may profoundly alter the current avian communities, and efforts to forecast the consequences of climate change using temperature alone can produce misleading results, especially in xeric regions.

BIOGRAPHICAL SKETCH

C. Steven Sevillano-Ríos was born in the city of Huaraz, a city within the Cordillera

Blanca of Peru, where, from an early age, he was in close contact with nature. Steven’s interest in nature and ecology was kindled by his parents, Cristian Sevillano and Nini Rios, who shared their sightseeing tours and traveled frequently with him and his two brothers,

Sebastian, and Adrian to different natural parts of Peru. During his last years at school,

Steven met Professor Nini Sanchez, a true pedagogue of yore, which few are now, who sparked his great interest in the natural sciences and reaffirmed his wish to study biology.

Steven studied biology at Cayetano Heredia University in Lima, where his advisor,

Dr. Armando Valdés-Velásques introduced him to the scientific world of ecology. Twice during his college career, Steven was forced to interrupt his studies in order to take care of family affairs - for rheumatic fever affecting his father in March of 2002 and his mother’s diagnosis with breast cancer in 2006. Thankfully, both recovered well and despite these interruptions, Steven completed his bachelor’s in biology in 2007.

Steven always hoped that he would one day apply his knowledge to his birthplace,

Cordillera Blanca and mountains in general. Thus, after finishing his BS degree, he conducted his undergrad thesis research in Huascarán National Park, working with the incredible Polylepis forests and its birds. Steven also developed several projects through the Mountain Institute in order to disseminate knowledge about the goods and services provided by Polylepis forests and their potential for sustainable use. After publishing the results of his research, Steven frequently gave talks at conferences and workshops about

Polylepis, and he helped to create a community conservation area in the Cordillera

Huayhuash.

Soon thereafter, Steven moved to southern Peru in order to explore new opportunities. Working on Manu National Park, Tambopata National Park, and Bahuaja

Sonene National Park, three of the most biodiverse areas of the world provided Steven with the incredible opportunity to learn from nature, especially from birds, and to realize that he could expand the ecological knowledge of poorly understood natural systems. At

Wayquecha Biological Station, he learned from Dr. Huw Lloyd and several graduate students that were conducting fieldwork. During his time with Wildlife Conservation

Society’s (WCS) in the Alto Tambopata Landscape Project, he worked with coffee farmers and local people to facilitate conservation of these natural areas. He also co-coordinated two rapid biological assessments in very remote Amazon areas. While there, Steven saw first-hand the massive destruction of biological richness by illegal mining, unrestricted burning and the use of land for cultivation of illegal crops.

Between 2014 - 2016, Steven, as Fulbright Fellow, received his Master degree in

Natural Resources at Cornell University under the guidance of Dr. Amanda Rodewald. In

2016, Steven transitioned to the doctoral program at Cornell, thanks to a fellowship from the Peruvian CONCYTEC. To date, he has published 14 research articles (7 as the first author), 2 theses, 2 book chapters, with more on the way (1 article in review, 1 in press and

4 in preparation). In September 2019, Steven started to work as Resident Lecturer in

Conservation Science at the School for Field Studies in the Center for Amazonian Studies in Peru, and in May 2020 he started a postdoc position in Peru where he looks forward to further personal and professional development.

I dedicate this work to my loving family,

Nini Rios de Sevillano, Cristian Sevillano,

Sebastian Sevillano, and Adrian Sevillano.

vii

Ilustration: Katterine Sandoval

ACKNOWLEDGMENTS

This dissertation was made possible by the effort, enthusiasm, generosity, and support of many institutions and people, including biologists, ecologists, conservationists, many friends, and my family. For the financial support, I am grateful to the FONDECYT-

CONCYTEC Scholarship from the Peruvian Government and to the Athena Grant, the

Bergstrom Award of the Association of Field Ornithologists, Fulbright Scholarship and the

American Alpine Club Research Grants 2017 without whose funding, this research and my studies would not have been possible. I also extend gratitude to members of my committee.

To Stephen Morreale for accepting this challenge and for the critical and intensive discussions that improved my research and always challenged my understanding and scientific knowledge. You were a great support during all these years, including your support at the Cornell Tropical Biology and Conservation Graduate Student Association.

To André Dhondt for his insightful questions, comments, and suggestions about the ecology and biology of mountain birds. To Viviana Ruíz-Gutierrez for encouraging me to think deeply about my ecological questions, the ways to test and analyzing them and the ways of how I can apply this knowledge across South America. Finally, to my advisor,

Amanda Rodewald, who from our first discussions showed a great interest in my study in

Peru and has been extraordinarily supportive during all these years at Cornell and as part of the Lab of Ornithology. Thank you for guiding me and facilitating many processes of my research, from strengthening my theoretical background to promoting critical thinking so helpful now and definitely in my future. Thank you!

I also thank the members of the Conservation Science and Bird Population Studies

Lab at Cornell Lab of Ornithology. To Ken Rosenberg, Peter Arcese, Wesley Hochachka,

Eliot Miller and Orin Robinson for their valuable comments and constructive criticism that

viii helped me to improve my whole research. To my friends Gerardo Soto, Ruth Bennett, DJ

MacNeil, Rose Swift, Bryant Dossman, Anna Lello-Smith, Maritere Reinoso and Sharon-

Rose Alonzo with whom we shared memorable and wonderful moments. To several members of the Department of Natural Resources, through our meetings in the Tropical

Biology and Conservation Association Group, gave me ideas about how to communicate my study results to different audiences.

Thank you very much to the core field crew of Jessica Pisconte, Nathaniel Young,

Rosaura Watanabe, Nicolas Mamani, James Purcell, Julio Salvador, and Celia Sierra, who spent enormous effort climbing mountains and time collecting data and transcribing the recordings. I owe you more meetings in the Tambo! Thanks to the staff of Huascaran

National Park, especially Ricardo Jesús Gómez, Selwyn Valverde, Oswaldo Gonzales,

Martin Salvador, and the rangers who from the beginning gave me the resources to develop the research. To the student interns at the HNP, Dennis Rojas, Ingrid Peña, Fiorella

Bernales, Gabriela Salazar. Thanks to Jack Valderrama (MAGSER Rent a Car) for the transportation services. To the Mountain Institute, under the direction of Dr. Jorge

Recharte. To Manuel Plenge for the firsthand literature and to Katterine Sandoval for the great illustrations of the Ash-breasted Tit-tyrant and the Giant Conebill.

I would like to thank my family, especially my mother Nini Ríos and my father

Cristian Sevillano, who opened the possibility of discovering “La Chacra”, the mountains and nature. Thanks for the great support of the whole crew at home! Thanks to my brothers

Sebastian and Adrian Sevillano who accompanied me to these mountains with unforgettable moments. Finally, I want to thank Laura V. Morales, for her love and support throughout this time. Thanks for your words and long hours of scientific discussion that helped improve not only manuscripts but our relationship.

TABLE OF CONTENTS

ABSTRACT………………………………………………………………………………...ii BIOGRAPHICAL SKETCH ...... ii ACKNOWLEDGMENTS ...... viiiiii TABLE OF CONTENTS ...... x LIST OF FIGURES ...... xii LIST OF TABLES ...... xvii

CHAPTER I: INTRODUCTION ...... 1 Study system ...... 2 Dissertation organization…..………………………………………………………………………….5 References ...... 6

CHAPTER II: RESPONSES OF POLYLEPIS BIRDS TO PATCH AND LANDSCAPE ATTRIBUTES IN THE HIGH ANDES ...... 11 Abstract……..………………………………………………………………………………………..11 Resumen……………………………………………………………………………………………...13 Introduction ...... 15 Methods ...... 17 Study area ...... 17 Bird surveys ...... 18 Data analysis ...... 20 Occupancy models and sensitivity to landscape attibutes ...... 20 Local species richness ...... 23 Cumulative species-area curve analysis ...... 24 Results ...... 25 Species richness and composition response……………………………………………...…….26 Community response…………………………………………………………………………...27 Sensitivity of individual species ...... 27 Cumulative species-area curves………………………………………………………………………..28 Discussion……………………………………………………………………………………………29 The importance of the total amount of Polylepis forest…………………………...…………...29 The importance of small Polylepis patches.……………………………………………………31 High-Andes species richness and composition………………………………..………….……32 Conservation implications…………………………...………………………………………...33 Acknowledgments……………………………………………………………………………………34 References……………………………………………………………………………………………36 Tables………………………………………………………………………………………………...51 Figures……………………………………………………………………………………………...... 53 Appendices……………….…………………………………………………………………………..64

CHAPTER III: DRIVERS OF AVIAN DIVERSITY ALONG THE WESTERN SLOPE OF THE CENTRAL ANDES AND IMPLICATIONS FOR CLIMATE CHANGE ...... 94 Abstract ...... 94 Resumen ...... 95 Introduction ...... 96 Methods ...... 99 Study area ...... 99 Estimating avian species richness ...... 100 Explanatory variables………………………………………………………………….………………….101 Climatic data……………………………………………………………………………………...………..101 Primary productivity………………………………………………………………………………………102

Land area……………………………………………………………………………………………………103 Mid domaineffect )MDE…………………………………………………………………………………..103 Human impact………………………………………………………………………………….………….. 103 Bird-enviromental relationships...…………………..………………………………………………… 103 Future climatic conditions and human impact………...……………………………………………….105 Results ...... 105 Changes along the elevation gradient ...... 105 Species-enviromental relatioships ...... 106 Vulnerability to land use and climate change………………………………………………………….107 Discussion ...... 108 Climatic, spatial and human-related drivers…………………………………………………………..109 Vulnerability to climate change and human activities…...…………………………………………...111 Acknowledgments…………………………………………………………………………………..112 References ...... 114 Tables ...... 130 Figures……………………………………………………………………………………...……….134 Appendices……………… …………………………………………………………………………143

CHAPTER IV: CLIMATIC DRIVERS OF AVIAN SPECIES RICHNESS ACROSS THE ANDES ...... 157 Abstract ...... 15757 Resumen ...... 159 Introduction ...... 161 Methods ...... 161 Study areas ...... 163 Estimated bird species richness ...... 164 Climated data ...... 165 Analysis ...... 165 Future climatic conditions ...... 166 Results ...... 166 Avian species richness along the Andess ...... 166 Temperature and precipitation patterns ...... 167 Climatic-species richness relationships …………………………………………………………..……167 Prediction under future climate conditions……………………………………………………….…...167 Discussion ...... 168 Implications for forecasting effects of climate change………………………………………………170 Acknowledgments…………………………………………………………………………………..171 References ...... 172 Tables ...... 180 Figures……………………………………………………………………………………...……….185 Appendices……………….…………………………………………………………………………193

CHAPTER II SUPPLEMENTARY INFORMATION A...... 199 CHAPTER II SUPPLEMENTARY INFORMATION B...... 199 CHAPTER III SUPPLEMENTARY INFORMATION C...... 199 CHAPTER IV SUPPLEMENTARY INFORMATION D…………………….…………199

LIST OF FIGURES

Figure 2.1. Expectations for species occupancy given different tolerances to habitat loss (amount of forest cover) and fragmentation (patch size). If Polylepis specialists are adapted to patchy configurations of Polylepis forest, they are expected to be unaffected by or less sensitive to patch size reduction (dashed line) (a-b), but either negatively affected by the reduction of amount of forest (solid line) (a) or unaffected by both (b). If not, Polylepis bird species are expected to be equally affected by patchiness and forest loss (c). In contrast, species associated with the Puna matrix would positively respond to reductions in amount of forest and patch size (d)...... 53

Figure 2.2. Map of the five valleys studied within Cordillera Blanca, Peru. A total of 59 patches were included, ranging from 0.01 ha to 200 ha...... 54

Figure 2.3. Examples of Polylepis forest from the glacial valleys of Ulta (A), Rajucolta (B), Llaca (C), Llanganuco (D) and Parón (E) within Cordillera Blanca. Matrix habitats are dominated by Puna grassland (A and E) or shrublands (F). Photos: Steven Sevillano-Ríos……………………………………………………………………...55

Figure 2.4. Contrasting predictions based on cumulative species-area curves, including (a) a positive association with patch size (i.e., samples in larger patches, while controlling for surveyed area, have more species), (b) negative association with patch size (i.e., more species in smaller patches), or (c) no association with patch size...... 56

Figure 2.5. Average (among three years) local species richness (alpha diversity) at each site (n=187) declines with increasing elevation (a), patch size (b) and amount of Polylepis forest within 200 m-radio (c). Black dots and vertical lines represent the average mean and 95% posterior intervals of the species richness estimated through MSOM for each site. The solid black and shade areas represent the mean ± 95% CI of lineal regressions...... 56

Figure 2.6. Avian community composition across five different categories of patch size and amount of forest within the landscape. The community was classified according to (1) habitat associations (a, b), (2) five foraging guilds (c, d), and (3) status of conservation priority (e, f). The total numbers of observations during the three years on each category are in parenthesis over each bar (this applies to the other two graphs below)...... 57

Figure 2.7. Mean community occupancy for birds in the High Andes was highest at mid- lower elevations (a), in large forest patches (b) and in local landscapes with

xii

relatively low forest cover (c) while accounting for (d) detectability. Shade areas represent the 95% of posterior intervals...... 58

Figure 2.8. Species- and community- level responses to elevation (a-b), patch size (c-d) and amount of Polylepis forest in the landscape (e-f). Open dots represent the species- specific mean response. Horizontal lines represent the 95% posterior intervals. Intervals that do not overlap zero (blue lines) indicate that estimated coefficients differed significantly from zero, indicating an effect of the parameter. Solid and dashed vertical lines (red) are the mean and 95% PI of the community response, respectively. Species ID correspond those in Table S2.1...... 59

Figure 2.9. Mean species occupancy probabilities in response to Polylepis patch size (left in a,b,c,d) and amount of forest (%) within 200-m radio (right in a,b,c,d). The solid lines and shade areas represent the mean ± 95% Posterior Intervals of the reduced MSOM. Cobin = Giant Conebill, Crbar = Baron's Spinetail, Atruf = Rufous-eared Brushfinch, Grand = Stripe-headed Antpitta, Lepil = Rusty-crowned tit-spinetail, Mepho = Black Metaltail, Mialt = Plain-tailed Warbling-Finch, Scaff = Ancash Tapaculo, Xepar = Tit-like Dacnis, Zastr = White-cheeked Cotinga, Ocruf = Rufous-breasted Chat-tyrant, Geser = Striated Earthcreeper. Individual plots of the 88 species are in Figure S2.7...... 61

Figure 2.10. Species richness estimation across both orders (large to small and vise-versa) of cumulative forest area of 59 Polylepis surveyed patches for all (a), forest (b), grassland Puna (c) and shrubland (d) species. Shade lines correspond to the 95% confidence interval...... 63

Figure 3.1. Location of the study area along the Cordillera Negra (western) and Cordillera Blanca (eastern) and areas currently protected by Huascaran Biosphere Reserve (red line). The western slope of the Central Andes (0-6,768 m) was surveyed using 1000 survey points located in a 220 x 150 km study area...... 134

Figure 3.2. Estimated species richness and endemism within raw (a-b) and grouped (c-d) elevational bands along the western slope of the central Andes (a). Species richness (212 species) was highly correlated with the number of endemic species (22 species) at both scales (b-d). Lines are the mean and the shaded polygon, the 95% CI of a smoother (a-c) and linear (b-d) fit...... 135

xiii

annual temperature range, (i) isothermality (g/h), (j) land area for 100-m elevational bands, (k) expected species richness under mid domain effect (DME), (l) Human Footprint Index (HFI) 2009...... 136

Figure 3.4. Raw (a-c) and grouped (d-e) species richness relationships with: annual mean temperature (a, d), annual mean precipitation (b, e) and the average primary productivity (NDVI) (c, f) along the elevation gradient of the western slope of the Peruvian Central Andes. Raw data correspond to 1,000 random points in our study site whereas grouped data correspond to the mean value of all points within each 200-m elevational bands. Black lines and shaded areas represent the best models fits for each parameter (mean, 95% CI)...... 1237

Figure 3.5. Modelled grouped species richness (low-high: grey-yellow-red) in response to the interaction of (a) temperature2*NDVI2, (b) precipitation2*NDVI2 and (c) temperature2*precipitation2. The interactive effects of temperature and precipitation on species richness are evident only at lower NDVIs values (a-b). Modelled primary productivity patterns in response to temperature and precipitation (d). Intermediate values of temperature and precipitation are related with greater species richness and primary productivity (c-d). Black dots are the environmental domain sampled along the elevation gradient...... 138

Figure 3.6. Avian species richness is most closely related to primary productivity, which at the same time is drive by the interaction of the annual mean temperature and precipitation...... 139

Figure 3.7. Relationship between species richness and the (a) mid domain effect (MDE), (b) topographic land area and (c) Human Footprint Index (HFI) along the elevation gradient were less supported by the models (Generalized R2 = 0.65; 0.59; 0.33 respectively)...... 140

Figure 3.8. Observed change in the Human Footprint Index between 1993-2009 across the elevation. Points are the difference between 1993 and 2009...... 141

Figure 3.9. Expected changes of the annual mean (a) temperature and (b) precipitation along the elevation by 2070; based on four (grey=optimist, black=pessimist) different scenarios of global change. In both cases, the strongest changes are expected to occur >2,500 m...... 142

xiv

the estimated species richness grouped for every 200-m elevational bands whereas predicted (red dots) data correspond to the outcome of the Temp2*Prec2 model (Table 2). Correspondingly, the RCP2.6 – RCP8.0 (dots and lines) are the outcomes of this model based on the four future climatic scenarios, where RCP8.0 correspond to the more severe scenario. The maximum amount of land area occurs at 4,200 masl (dashed line)...... 142

Figure 4.1. The interaction of temperature and precipitation is expected to define the number of species along the elevational gradient...... 185

Figure 4.2. A total of 16 elevational gradients were analyzed for species richness and climatic factors across the Andes. Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito area = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16...... 186

Figure 4.3. Patterns of species richness along elevational gradients in the Andes. Species richness (a) decreased constantly with elevation in all the Amazonian slopes and (b) presented a slight low-elevation plateau in the northern Central Andes of Colombia and Ecuador. (c) A low elevation peak was observed in the Cordillera Central and Occidental of Colombia as well as in the Central Andes of Peru (d); which occurred at the base of the sampled plot. (e) A mid-elevation peak around 4,000 m was observed on the western slopes of the central Andes whereas in the southern (f), species richness decreased again with elevation...... 187

Figure 4.4. Elevational patterns of annual mean temperature (a) and precipitation (b) along sixteen different elevational gradients along the Andes. Temperature always decreased with elevation and had similar values in all but the two southernmost gradients in south Chile (15 and 16), which were significantly colder given the latitudinal gradient. Overall, precipitation patterns decreased with elevation although they were more erratic along the elevations. Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito área = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16...... 188

Figure 4.5. Climatic (temperature=black and precipitation=blue) and avian species richness patterns (transitioning from beige to red with increasing diversity) over 16 Andean elevational gradients. Climatic values correspond to random points ~1,000 per site. and showed contrasting patterns along the elevation gradient. While temperature always decreased constantly with elevation and maintained similar values for most part of the Andes, precipitation was more erratic and no single pattern was

dominant; remaining constant (a, d, e), increasing (g, k, m), decreasing (b, l, o), decreasing drastically before to remain constant (c, h, n) or showing mixture patterns (f, I, p)...... 189

Figure 4.7. Difference (modelled – actual) on species richness in response to changes on temperature and precipitation predicted according to the four RCPs climatic models by 2070...... 192

xvi

LIST OF TABLES

Table 2.1. Community-level hyper-parameters for occupancy and detection in relation to elevation, patch size and amount of forest area. Each value represents the mean, standard deviation, and the 95% posterior interval of the effect of each parameter on the whole bird community. Rhat is the Brooks-Gelman-Rubin (BGR) measure of convergence of the three Markovian chains. Rhat <1.1 indicate convergence. These parameters correspond to our reduced model...... 51

Table 2.2. Proportion of forest cover with respect to the landscape extension of each of the five glacier valleys studied in Cordillera Blanca, Peru. The landscape extension was delimitated by the natural entrance of the valleys and the 5,000 m elevational line and calculated based on a Digital Elevation Model. The forest cover was calculated in the same way as the studied patches...... 52

Table 3.1. Model included in the analysis of the three tested hypotheses (climatic, spatial and human). Within each group, models are ordered by AIC...... 130

Table 3.2. Additive and interactive climatic model included in the analysis of grouped data. Within each group, models are ordered by AIC, which is highly correlated with the generalized R2 in a log-log scale (Spearman’s r correlation = 0.99, p<0.0001).. ... 132

Table 3.3. Main and quadratic effects (Std. Error) of the three climatic factors (mean annual temperature, mean annual precipitation and primary productivity) defining species richness pattern along the elevational gradient in the western slope of the central Andes...... 133

Table 4.1. Names, elevational ranges and country were the 16 elevational gradients under study are located. Previous bird species inventories along the elevation are given as references...... 181

Table 4.2 Main and interactive effects of temperature and precipitation defining species richness pattern along sixteen elevational gradients across the Andes…………...182

xvii

CHAPTER I

INTRODUCTION

The diversity of life on our planet is under siege. Habitat loss, degradation, and fragmentation are among the main threats to biodiversity around the world (Brooks et al., 2002;

Fahrig 2003; Cushman 2006; Willis and Bhagwat 2009; Bellard et al., 2012) and are implicated in the loss of 13 to 75% of the world’s species (Haddad et al., 2015). Climate change is expected to exacerbate these consequences (Parmesan 1996, Parmesan and Yohe 2003), particularly in tropical mountain systems (Şekercioğlu et al., 2008; Wormworth and Şekercioğlu 2011;

Şekercioğlu et al., 2012). Our limited understanding of the mechanisms governing ecological responses to changes in habitat and climate constrains our ability to predict and mitigate threats to biodiversity.

Fortunately, one of the defining attributes of mountains – elevational gradients – provides an excellent opportunity to study the responses of species to changes in habitat and climate and, by using comparative approaches, to determine their separate and combined influences on biodiversity (Tingley et al., 2009; Tingley et al., 2012; Jankowski et al., 2010; Elsen et al., 2017).

Unlike latitudinal gradients, elevational gradients, especially in the tropics, are represented by multiple life zones, including tropical, subtropical, temperate, boreal-artic and snowline, that occur within small areas, often less than 6 km of elevational range and/or 10 km of distance

(Körner 2000). The close proximity of distinct life zones and ecosystems with shared geological and evolutionary history (Antonelli et al., 2018) means that each gradient replicates, to some extent, the response of systems to habitat and climatic changes (Elsen et al., 2017). Elevational gradients also permit us to distinguish long-term from seasonal shifts (Körner 1999) as well as to correct by land area effects (Elsen and Tingley 2015), helping to identify drivers of local and

1 global patterns of biodiversity (Gaston 2000; Körner 2000; McCain 2009). This understanding is critical if we aim to predict the impacts of future changes in habitat and climatic conditions, identify species or areas at higher risk, and propose strategies to mitigate these impacts

(Parmesan and Yohe 2003; Araujo and Rahbek 2006; La Sorte and Jetz 2010).

STUDY SYSTEM

My dissertation focuses on tropical montane systems in The Andes Mountains of South

America. The Andes Mountains are one of the most biodiverse, and also threatened, mountain regions of the world (Myers et al., 2000; Borsdorf and Stadel 2015). Distributed from Venezuela to Patagonia, the Andes are the most latitudinally-extensive tropical mountain range in the world and are characterized by unusually high levels of endemism, species turnover (beta diversity), ecosystem diversity along elevational gradients exceeding 6,000 meters (Borsdorf and Stadel

2015). The complex topography shapes interactions among abiotic and biotic components of ecosystems, promotes high rates of speciation and provides habitat for many species (Terborgh and Weske 1975; Vuilleumier and Simberloff 1980).

Yet the Andes, like many montane systems in the world, are facing enormous pressures from climate, human activities, and land-use change. The magnitude of climatic changes, for example, is well-illustrated by mountain glaciers studies across the Andes (Vuille et al., 2008;

Silverio et al., 2005; Mark et al., 2010) ), where a warming trend of 0.39 °C per decade (1951-

1999) (Mark and Seltzer, 2005) resulted in >30% loss of glacial areas during the last century

(Vuille et al, 2008). Worrisomely, strong human pressures within several Andean regions continue to alter – by way of habitat loss, degradation, or fragmentation – the integrity of ecosystems and landscapes, with effects of climate change becoming increasingly evident

(Herzog et al. 2011). The combined effect of climate and habitat changes are altering the

2 distribution and health of many ecosystems, including lakes, wetlands, forests, paramos, and

Puna (Anderson et al., 2017) along with the manner in which human populations interact with them (Kronik and Verner 2010). Uncertainty about how such changes influence biodiversity constrains our ability to develop adequate conservation strategies, especially for current globally endangered and vulnerable ecosystems (IUCN 2019).

Birds probably represent one of the best-studied and biodiverse groups across the Andes

(Fjeldså and Krabbe 1990; Ridgely et al. 2001; McMullan et al., 2010; Schulenberg et al. 2010;

Herzog et al. 2016; Ascanio et al., 2017), and as such, are useful tools to help us understand how global changes may impact biodiversity (Herzog et al., 2011). Birds are recognized as one of the best indicators of habitat disturbance and “sentinels for climate change effects” (Wikelski and

Tertitski 2016) as many species track changes in habitat (Devictor et al., 2008; Tingley et al.,

2009; Şekercioĝlu et al., 2012;), landscape composition and configuration (Rodewald and

Yahner 2001), or climate (Charmantier et al., 2008; Devictor et al., 2008; Tingley et al., 2009; La

Sorte and Jetz 2010). Andean birds may be particularly sensitive to these changes given their limited dispersal ability (Moore et al., 2008; La Sorte and Jetz 2010; Lloyd and Marsden 2011), high levels of eco-physiological specialization over narrow elevational ranges (Jankowski et al.,

2013), and small population sizes (Crick 2004) restricted to specific mountains ranges within the

Andes (Fjeldså and Krabbe 1990). Not surprisingly then, several regions in the Andes have been identified as centers of endemism and diversity hotspots that support large numbers of globally threatened or endangered species (Sevillano-Rios et al., 2018). Studying how Andean birds respond to habitat and climatic changes can, therefore, help us to predict and guide conservation strategies that can mitigate future threats to biodiversity.

One globally endangered and vulnerable ecosystem that is being further compromised by human pressures and climate change is the Polylepis forest, which represents one of the highest elevation forests in the world (Purcell et al., 2004; Hoch and Körner, 2005). Although palynological studies indicate that Polylepis ecosystems were naturally patchy in their configuration and did not occur as large areas of continuous forest (Gosling et al., 2009;

Valencia et al., 2018), we also know that Polylepis ecosystems have long been destroyed, fragmented, and degraded by a wide range of anthropogenic threats (Fjeldså et al., 1996; Kessler

2006). Included among main threats are uncontrolled use of fire along Puna grasslands (Renison et al., 2002), firewood collection (Fjeldså, 1993), browsing by livestock (Teich et al, 2005;

Cierjacks et al., 2008, Kessler, 2002), road construction, and mining development (Purcell and

Brelsford 2004). The extent to which these pressures will result in population declines, extirpations, or extinctions of Polylepis-associated birds is unknown, but there is the possibility that many species that have persisted for at least 20,000 years in a patchy ecosystem are pre- adapted to survive – and perhaps even thrive – in fragmented landscapes.

Within this context, and using the avian bird community across the Andes, this research aims to elucidate 1) how habitat loss and fragmentation at landscape scales affects avian species and biodiversity gradients within Polylepis ecosystems, 2) how factors related to climate, landscape configuration, and human activities at regional scales shape patterns of avian species richness and endemism along elevational gradients in the western slope of the Andes, 3) the extent to which relationships between elevation and avian diversity at continental scales are influenced by complex interactions between temperature and precipitation along latitudinal gradients, and 4) the degree to which these systems can persist under future climate scenarios.

DISSERTATION ORGANIZATION

This thesis consists of three main chapters that will be submitted for publication independently. Chapter II evaluates the effect of habitat loss and fragmentation in the avian community associated with Polylepis forest ecosystems, which are patchily distributed within landscapes of Puna grasslands and shrublands in the High Andes of Peru (3,300 – 4,700 m).

Results support the hypothesis that Polylepis bird specialists are adapted to this naturally patchy landscape so long as sufficient forest (>10% of the glacier valley or >400 ha forest) remains; highlighting the importance of a heterogeneous matrix and small patches for maintaining key habitats for numerous species of conservation concern. Chapter III evaluates the extent to which relationships between elevation and avian diversity were explained by factors related to climate

(e.g. temperature, precipitation, primary productivity, climatic stability), landscape configuration

(e.g. land area and spatial constraints) and human activities along the western slope of the central

Andes. Results suggest that the mid-elevation peak in avian species richness and endemism most likely results from high primary productivity and, indirectly, by intermediate levels of warm temperatures and precipitation at mid-elevations (3,300 – 4,300 masl). Future climatic conditions and human activities are expected to continue to intensify and expand at elevations below 500 and above 3,000 masl expecting to produce lowland attrition of species, upslope range shifts, mountain top extinctions and possibly downslope shifts for a few species closely tied to precipitation. In Chapter IV, we expand this understanding to a continental scale, evaluating the interaction of temperature and precipitation shaping avian species richness along an altitudinal and latitudinal gradient across the Andes. Our results indicate that both climatic factors interact differently across the Andes, where precipitation alone was a critical driver of diversity along the xeric western slopes, and temperature was most important in the rest of the Andes.

REFERENCES

Anderson, E. P., Marengo, J., Villalba, R., Halloy, S., Young, B., Cordero, D., ... & Martinez, R. (2017). Consequences of climate change for ecosystems and ecosystem services in the tropical Andes. Antonelli, A., Kissling, W. D., Flantua, S. G., Bermúdez, M. A., Mulch, A., Muellner-Riehl, A. N., ... & Fritz, S. A. (2018). Geological and climatic influences on mountain biodiversity. Nature Geoscience, 11(10), 718. Araújo, M. B., & Rahbek, C. (2006). How does climate change affect biodiversity?. Science, 313(5792), 1396-1397. Ascanio, D., Rodriguez, G. and Restall, R., (2017). Birds of Venezuela. Bloomsbury Publishing. Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., & Courchamp, F. (2012). Impacts of climate change on the future of biodiversity. Ecology letters, 15(4), 365-377. Borsdorf, A. and Stadel, C. (2015). The Andes: A geographical portrait. Switzerland: Springer. Brooks, T. M., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A., Rylands, A. B., Konstant, W. R., ... & Hilton‐Taylor, C. (2002). Habitat loss and extinction in the hotspots of biodiversity. Conservation biology, 16(4), 909-923. Charmantier, A., McCleery, R. H., Cole, L. R., Perrins, C., Kruuk, L. E., & Sheldon, B. C. (2008). Adaptive phenotypic plasticity in response to climate change in a wild bird population. science, 320(5877), 800-803. Cierjacks, A., Rühr, N. K., Wesche, K., & Hensen, I. (2008). Effects of altitude and livestock on the regeneration of two tree line forming Polylepis species in Ecuador. Plant Ecology, 194(2), 207-221. Crick, H. Q. (2004). The impact of climate change on birds. Ibis, 146, 48-56. Cushman, S. A. (2006). Effects of habitat loss and fragmentation on amphibians: a review and prospectus. Biological Conservation, 128(2), 231-240. Devictor, V., Julliard, R., Couvet, D., & Jiguet, F. (2008). Birds are tracking climate warming, but not fast enough. Proceedings of the Royal Society B: Biological Sciences, 275(1652), 2743-2748. Elsen, P. R., & Tingley, M. W. (2015). Global mountain topography and the fate of montane species under climate change. Nature Climate Change, 5(8), 772.

Elsen, P. R., Tingley, M. W., Kalyanaraman, R., Ramesh, K., & Wilcove, D. S. (2017). The role of competition, ecotones, and temperature in the elevational distribution of Himalayan birds. Ecology, 98(2), 337-348. Fahrig, L. (2003). Effects of habitat fragmentation on biodiversity. Annual review of ecology, evolution, and systematics, 34(1), 487-515. Fjeldså, J. (1993). The avifauna of the Polylepis woodlands of the Andean highlands: the efficiency of basing conservation priorities on patterns of endemism. Bird Conservation International, 3(01), 37-55. Fjeldså, J., & Krabbe, N. (1990). Birds of the high Andes. Fjeldså, J., Kessler, M., Engblom, G., & Driesch, P. (1996). Conserving the biological diversity of Polylepis woodlands of the highland of Peru and Bolivia: a contribution to sustainable natural resource management in the Andes. Copenhagen: Nordeco. Freeman, B. G., Scholer, M. N., Ruiz-Gutierrez, V., & Fitzpatrick, J. W. (2018). Climate change causes upslope shifts and mountaintop extirpations in a tropical bird community. Proceedings of the National Academy of Sciences, 201804224. Gaston, K. J. (2000). Global patterns in biodiversity. Nature, 405(6783), 220. Gosling, W. D., Hanselman, J. A., Knox, C., Valencia, B. G., & Bush, M. B. (2009). Long‐term drivers of change in Polylepis woodland distribution in the central Andes. Journal of Vegetation Science, 20(6), 1041-1052. Haddad, N. M., Brudvig, L. A., Clobert, J., Davies, K. F., Gonzalez, A., Holt, R. D., ... & Cook, W. M. (2015). Habitat fragmentation and its lasting impact on Earth’s ecosystems. Science Advances, 1(2), e1500052. Herzog S.K., Terrill RS, Jahn AE, Remsen, JV, Maillard OZ, García-Solíz VH, MacLeod R, MacCormick A, Vidoz JQ, Tofte CC, Slongo H., Tintaya O, Kessler M, Fjeldså J. (2016). Birds of Bolivia: Field Guide. Asociación Armonía. Herzog, S. K., Martínez, R., Jørgensen, P. M., & Tiessen, H. (2011). Climate change and biodiversity in the tropical Andes. Inter-American Institute for Global Change Research (IAI) and Scientific Committee on Problems of the Environment (SCOPE), 348. Hoch, G., & Körner, C. (2005). Growth, demography and carbon relations of Polylepis trees at the world's highest treeline. Functional Ecology, 19(6), 941-951.

IUCN 2019. The IUCN Red List of Threatened Species. Version 2019-2. https://www.iucnredlist.org Jankowski, J. E., Londoño, G. A., Robinson, S. K., & Chappell, M. A. (2013). Exploring the role of physiology and biotic interactions in determining elevational ranges of tropical animals. Ecography, 36(1), 1-12. Jankowski, J. E., Robinson, S. K., & Levey, D. J. (2010). Squeezed at the top: interspecific aggression may constrain elevational ranges in tropical birds. Ecology, 91(7), 1877- 1884. Kessler, M. (2002). The “Polylepis problem”: where do we stand. Ecotropica,8(2), 97-110. Kessler, M. (2006). Bosques de Polylepis. Botánica económica de los Andes Centrales, 11. Körner, C. (2000). Why are there global gradients in species richness? Mountains might hold the answer. Trends Ecol Evol, 15, 513-514. Körner, Ch. (1999) Alpine Plant Life, Springer- Verlag Kronik, J., & Verner, D. (2010). Indigenous peoples and climate change in Latin America and the Caribbean. The World Bank. La Sorte, F. A., & Jetz, W. (2010). Projected range contractions of montane biodiversity under global warming. Proceedings of the Royal Society B: Biological Sciences, 277(1699), 3401-3410. Lloyd, H., & Marsden, S. J. (2011). Between‐patch bird movements within a high‐Andean Polylepis woodland/matrix landscape: Implications for habitat restoration. Restoration Ecology, 19(1), 74-82. Mark BG, Bury J, McKenzie JM, French A, Baraer M (2010) Climate change and tropical Andean glacier recession: Evaluating hydrologic changes and livelihood vulnerability in the Cordillera Blanca Peru. Ann Assoc Am Geogr. 100: 794-805 Mark, B. G., & Seltzer, G. O. (2005). Evaluation of recent glacier recession in the Cordillera Blanca, Peru (AD 1962–1999): spatial distribution of mass loss and climatic forcing. Quaternary Science Reviews, 24(20), 2265-2280. McCain, C. M. (2009). Global analysis of bird elevational diversity. Global Ecology and Biogeography, 18(3), 346-360. McMullan, M., Donegan, T.M. and Quevedo, A., (2010). Field guide to the birds of Colombia. Fundación ProAves.

Moore, R. P., Robinson, W. D., Lovette, I. J., & Robinson, T. R. (2008). Experimental evidence for extreme dispersal limitation in tropical forest birds. Ecology letters, 11(9), 960- 968. Myers, N., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A., & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403(6772), 853. Parmesan, C. (2006). Ecological and evolutionary responses to recent climate change. Annu. Rev. Ecol. Evol. Syst., 37, 637-669. Parmesan, C., & Yohe, G. (2003). A globally coherent fingerprint of climate change impacts across natural systems. Nature, 421(6918), 37. Primack, R. B. (1993). Essentials of conservation biology (Vol. 23). Sunderland: Sinauer Associates. Purcell, J., & Brelsford, A. (2004). Reassessing the causes of decline of Polylepis, a tropical subalpine forest. Ecotropica, 10, 155-158. Renison, D., Cingolani, A. M., & Suarez, R. (2002). Efectos del fuego sobre un bosquecillo de Polylepis australis (Rosaceae) en las montañas de Córdoba, Argentina. Revista chilena de historia natural, 75(4), 719-727. Ridgely, R.S. and Greenfield, P.J. (2001). The birds of Ecuador, vol. 1. Field guide. Cornell. Rodewald, A. D., Yahner, R. H., & Brawn, J. (2001). Avian nesting success in forested landscapes: influence of landscape composition, stand and nest-patch microhabitat, and biotic interactions. The Auk, 118(4), 1018-1028. Schulenberg, T.S., Stotz, D.F., Lane, D.F., O'Neill, J.P. and Parker III, T.A., 2010. Birds of Peru: revised and updated edition. Princeton University Press. Şekercioğlu, Ç. H., Primack, R. B., & Wormworth, J. (2012). The effects of climate change on tropical birds. Biological conservation, 148(1), 1-18. Şekercioğlu, C. H., Schneider, S. H., Fay, J. P., & Loarie, S. R. (2008). Climate change, elevational range shifts, and bird extinctions. Conservation biology, 22(1), 140-150. Sevillano-Ríos, C. S., Rodewald, A. D., & Morales, L. V. (2018). Ecología y conservación de las aves asociadas con Polylepis:¿ qué sabemos de esta comunidad cada vez más vulnerable?. Ecología Austral, 28(1-bis), 216-228. Silverio W, Jaquet JM (2005) Glacial cover mapping 1987–1996 of the Cordillera Blanca Peru using satellite imagery. Remote Sens Environ. 953:342-350

Teich, I., Cingolani, A. M., Renison, D., Hensen, I., & Giorgis, M. A. (2005). Do domestic herbivores retard Polylepis australis Bitt. woodland recovery in the mountains of Córdoba, Argentina?. Forest Ecology and Management,219(2), 229-241. Terborgh, J. and Weske, J.S. (1975). The role of competition in the distribution of Andean birds. ECOLOGY, 56(3), pp.562-576. Tingley, M. W., Koo, M. S., Moritz, C., Rush, A. C., & Beissinger, S. R. (2012). The push and pull of climate change causes heterogeneous shifts in avian elevational ranges. Global Change Biology, 18(11), 3279-3290. Tingley, M. W., Monahan, W. B., Beissinger, S. R., & Moritz, C. (2009). Birds track their Grinnellian niche through a century of climate change. Proceedings of the National Academy of Sciences, 106(Supplement 2), 19637-19643. Valencia, B. G., Bush, M. B., Coe, A. L., Orren, E., & Gosling, W. D. (2018). Polylepis woodland dynamics during the last 20,000 years. Journal of Biogeography. Vuille, M., Francou, B., Wagnon, P., Juen, I., Kaser, G., Mark, B. G., & Bradley, R. S. (2008). Climate change and tropical Andean glaciers: Past, present and future. Earth-science reviews, 89(3-4), 79-96. Vuilleumier, F. and Simberloff, D. (1980). Ecology versus history as determinants of patchy and insular distributions in high Andean birds. Plenum. in Hecht, M. K, W. C. Steere, & B. Wallace (eds.). Evolutionary biology, Vol. 12. Plenum, New York. Wikelski, M., & Tertitski, G. (2016). Living sentinels for climate change effects. Science, 352(6287), 775-776. Willis, K. J., & Bhagwat, S. A. (2009). Biodiversity and climate change. Science, 326(5954), 806-807. Wormworth, J., & S̜ ekercioğlu, C. (2011). Winged sentinels: birds and climate change. Cambridge University Press

CHAPTER II

RESPONSES OF POLYLEPIS BIRDS TO PATCH AND LANDSCAPE ATTRIBUTES IN

THE HIGH ANDES

C. Steven Sevillano-Ríos and Amanda D. Rodewald

ABSTRACT

Habitat loss and fragmentation can devastate biodiversity, especially at regional and global scales. However, generalizing to individual species is challenging given the wide variety of intrinsic and extrinsic factors that shape species-specific responses – particularly among species that are specialists, generalists, or adapted to naturally patchy landscapes. In this study, we examined how patch and landscape attributes affected bird communities within Polylepis forest ecosystems, which are patchily distributed within landscapes of Puna grasslands and shrublands in the High Andes of Peru (3,300 – 4,700 m). We surveyed birds in 59 Polylepis patches and 47 sites in the Puna matrix, resulting in 13,210 observations of 88 bird species, including 15 species of conservation concern specialized on Polylepis. Data were analyzed using

Multi-Species Occupancy-Models (MSOM) and cumulative species-area curves. Species richness was generally greatest at mid-to-low elevations, within small fragments, and in landscapes with comparatively little forest cover; this was especially true for birds associated with the Puna matrix. Consistent with the hypothesis that Polylepis specialists are adapted to naturally patchy landscapes, we found no evidence that Polylepis specialists were sensitive to patch size, though two of nine species were positively related to forest cover within 200m. Our work shows that small patches of Polylepis have high ecological value and that conservation of

11 species of concern may depend more on retaining at least 10% forest cover within landscapes than on the presence of large patches of Polylepis.

Key-words: Andes, habitat fragmentation, habitat loss, hierarchical modeling, landscape,

MSOM, mountains, naturally fragmented, spatial configuration.

RESUMEN

La pérdida y fragmentación de hábitat puede devastar la biodiversidad, especialmente a escala regional y global. Sin embargo, generalizarlo a especies individuales puede no ser apropiado dada la variabilidad de factores intrínsecos y extrínsecos que determina respuestas especificas – particularmente entre especies que son especialistas, generalistas or adatadas a paisajes naturalmente fragmentados. En este estudio, examinamos cómo la extensión del tamaño del parche y la cantidad de bosque en el paisaje afecta la comunidad de aves dentro de los ecosistemas de bosques de Polylepis, que se distribuyen de manera irregular dentro de paisajes de pastizales y matorrales de Puna en los Altos Andes del Perú (3,300 – 4,700 m). Examinamos las aves en 59 parches de Polylepis y 47 sitios en la matriz de Puna, lo que resultó en 13,210 observaciones de aves de 88 especies, incluyendo 15 especies de interés para la conservación.

Los datos se analizaron utilizando modelos de ocupación de múltiples especies (MSOM) y curvas acumulativas de especies-área. En general, la riqueza de especies fue mayor a elevaciones medias-a-bajas, en fragmentos pequeños y en paisajes con una cubierta forestal comparativamente baja; esto fue especialmente cierto para las aves asociadas a la matriz de

Puna. Consistente con la hipótesis que las aves especilitas a Polylepis están adaptadas a un paisaje naturalmente fragmentado, no encontramos evidencia que los especialistas a Polylepis fueran sensibles al tamaño de parche, aunque dos de nueve especies fueron asociadas positivamente a la cobertura forestal dentro de 200m de radio. Nuestros hallazgos resaltan la contribución ecológica que cumplen los parches pequeños de Polylepis en los paisajes altoandinos y que la conservación de especies amenazadas depende más en retener al menos un

10% de cobertura forestal (> 400 ha) de estos valles galaciares que la sola presencia de grandes bosques de Polylepis.

Palabras clave: Andes, fragmentación de hábitat, Perdida de hábitat, modelamiento jerárquico, paisaje, MSOM, montañas, paisajes naturalmente fragmentados, configuración espacial.

INTRODUCTION

Anthropogenic changes in landscape composition and configuration can drive biodiversity loss at local, regional and global scales (Fahrig 1997, 2002; Lindenmayer and Franklin 2002) and are already implicated in the loss of 13 to 75% of the world’s species (Haddad et al., 2015). Despite strong evidence of devastating ecological consequences of habitat loss and fragmentation (Fahrig

2003; Wiegand et al., 2005; Liu et al., 2018), relatively few studies have parsed out their differential effects of these two factors because the two are often confounded within study designs (Fahrig 2003, 2013, 2017). Habitat loss can occur in the absence of fragmentation, which breaks apart continuous habitat (McGarigal and Cushman 2002; Fahrig 2003, 2017) in ways that reduce habitat patch size and increase isolation (Andrén 1994; Rodewald and Yahner 2001;

Fahrig 2003; Yaacobi et al., 2007; Fahrig 2017). Distinguishing between the effects of habitat loss and fragmentation on species of conservation concern is essential to identifying specific drivers of population decline, as well as effective conservation strategies (Bissonette and Storch

2002; Collinge 2009).

Habitat loss and fragmentation affect species in different ways, depending upon sensitivity to landscape context, degree of habitat specialization, and other intrinsic traits related to dispersal ability, morphology, life history and behavior (Kareiva 1987; Lomolino 2005; Ferraz et al., 2007; Duckworth and Kruuk 2009; Cote et al., 2010). For example, while habitat loss negatively impacts a wide range of species by limiting access to resources or reducing population size, fragmentation disproportionately impacts species that avoid edges, are area-sensitive, or have limited mobility (Radford et al., 2005; Haddad et al., 2015; Fletcher et al., 2018). Small populations within fragmented habitats may be particularly vulnerable to extinction from demographic and environmental stochasticity, loss of genetic variability and inbreeding

15 depression (Sih et al. 2000; Gaggiotti and Hanski 2004). At the same time, fragmentation can sometimes improve stability and resilience of naturally patchy systems and can promote persistence of edge-associated or generalist species (Huffaker 1958; Hastings 1977; Grez et al.,

2004; Ethier and Fahrig 2011). Indeed, a recent review by Fahrig (2017) reported that 76% of

118 studies detected positive outcomes of fragmentation, including improved functional connectivity (Holzschuh et al., 2010; Saura et al., 2014), higher habitat diversity or heterogeneity

(Quinn and Harrison 1988; Tscharntke et al., 2002), or positive edge effects (Klingbeil and

Willig 2009; Henden et al., 2011). This body of evidence collectively suggests that the consequences of habitat loss and fragmentation may be context-dependent and may differ among species adapted to continuous versus naturally patchy ecosystems, such as Serpentine soil and plant communities, rocky coastal ecosystems, coral reefs, African woodlands and savannas, and

Amazonian white-sand forests (Proctor and Woodell 1975; Nyström and Folke 2001; Fine et al.,

2010; Pennington et al., 2018).

In our study, we examined the extent to which forest patch size and the amount of forest in the landscape affect occurrence of High Andean birds across an elevation gradient (3,300 –

4,700 masl) dominated by the world’s highest forests – Polylepis forests – and a Puna matrix of grasslands and shrublands (Simpson 1979; Gareca 2010). Though naturally patchy ecosystems,

Polylepis forests are hotspots of avian diversity throughout the High Andes (Fjeldså 1992, 2002;

Kessler 2006; Lloyd 2008 a, b, c; Sevillano-Ríos et al., 2018) that support a group of highly specialized and threatened birds (Kessler 2002; Fjeldså 2002; Gosling et al., 2009, IUCN 2018).

Biogeographical and historical studies point to dramatic Polylepis forest expansions and contractions during interglacial periods (van der Hammen 1974; Pirie et al., 2006; Hughes and

Eastwood 2006; Schmidt-Lebuhn et al., 2010; Hanselman et al., 2011; Valencia et al., 2018) and

16 in response to fire, climate, and topographic/substrate changes across the last 370 kyr BP

(Chepstow-Lusty et al., 2005; Weng et al., 2006; Williams et al., 2011; Rodríguez and Behling

2012). Consequently, Polylepis was never configured as continuous forest and, rather, was a dynamic, naturally patchy ecosystem (Valencia et al., 2018). Unfortunately, loss and fragmentation of Polylepis forests have accelerated since the Andes were settled by humans

(~11-12 kyr BP to the present; Valencia et al., 2018) due to human-caused fires, cattle grazing, and timber harvesting (Renison et al., 2002; Purcell and Brelsford 2004; Cingolani et al., 2013).

These activities also caused striking changes to the composition of the Puna grassland and shrubland plant community, which is inhabited by many other High-Andean bird species

(Fjeldså and Krabbe 1990; Sylvester et al., 2014; Sylvester et al., 2017). Given the historical and current accounts of naturally-patchy and heterogeneous landscapes in the High Andes, we examined the evidence for four different relationships between avian occupancy and, either fragmentation per se (patch size), or the amount of habitat (forest cover within a given area)

(Figure 2.1). We expected that numbers of Polylepis specialists would increase with forest cover but show limited sensitivity to patch size, as they would presumably be adapted to patchily- distributed habitat. In contrast, we predicted that matrix-associated species, including generalists or species that use shrubland or Puna habitats, would be positively associated with small patches and landscapes with comparatively less forest cover.

METHODS

Study area

We studied bird communities in five glacial valleys of Cordillera Blanca, within Huascaran

National Park and Biosphere Reserve, Ancash, Peru (9°06′19″S, 77°36′21″W) (see Sevillano-

Ríos (2016) and Sevillano-Ríos and Rodewald (2017) for a more detailed description). This area

17 is recognized as a center of avian endemism and diversity of the High Andes (Fjeldså at al.,

1996; Fjeldså 2002; Servat 2006; Sevillano-Ríos et al., 2011; Sevillano-Ríos and Rodewald

2017) and contains some of the largest remaining areas and patches of Polylepis forest in the world (Zutta et al., 2012; Sevillano-Ríos et al., 2018). The five glacial valleys (Parón,

Llanganuco, Ulta, Llaca and Rajucolta) are located on the western slope of Cordillera Blanca and discharge into the Santa River and the Pacific Ocean (Figure 2.2).

We selected 59 Polylepis patches ranging in size from 0.01 ha to 199 ha (mean =13.9, SD

= 31.7), totaling 822 ha and distributed along the entire elevational gradient of each of our five glacial valleys (3,300 ̶ 4,700 masl) (Figure 2.2 and Figure 2.3). A patch of Polylepis forest was defined as a continuous woodland >30 m from any other patch (Lloyd 2008 a, b). Because

Polylepis forests are easily distinguished from other plant communities in satellite images

(Sevillano-Ríos et al., 2018), we manually delimited the boundaries of each patch on images of

0.5 m-resolution from Google Earth (Google Earth Pro, DigitalGlobe, © 2017). The topographic complexity of the area prevented us from surveying patches located on cliffs or otherwise inaccessible areas. Patches were dominated by Polylepis sericea at lower elevations (3,300 ̶

4,000 masl) forest or by P. weberbaueri at upper elevations (4,000 ̶ 4,700 masl), but a variety of trees and shrubs occurred within patches (see Sevillano-Ríos et al., 2017).

Bird surveys

A robust sampling design for multiple species (Kendall et al., 1997; Kendall 2001; MacKenzie and Royle 2005) was used to survey birds at 187 points located within both forest (n = 140) and the matrix, which was comprised of Puna grassland or shrubland (n = 47). At each point, we used a GPS (GARMIN GPSMAP64) to record UTM coordinates and elevation (±10 m), the latter of which were highly correlated (r = 0.995) with the Advanced Spaceborne Thermal

Emission and Reflection Radiometer (ASTER) – Global Digital Elevation Model v2 (GDEM2)

(NASA and METI 2011, 30-m resolution). Small patches contained only one point placed roughly in the center of the patch, but large patches contained up to 10 points distributed throughout the patch and separated by >150 m (mean ± SD = 190.00 ± 74.15; min – max = 150 –

480 m). Points in the matrix were separated by > 200 m (245 ± 85.80; 200 – 650 m) and placed along the whole elevation gradient.

Surveys were conducted during the dry seasons (May to August) of 2014 and 2016 and the wet season (January to April) of 2015 by four trained observers. Each point was surveyed 3-

13 times (mean ± SD = 7.57 ± 2.53) between sunrise (~0500 ̶ 0600 h) and 1200 h, totaling 1416 visits during which an observer recorded all birds seen or heard within 50 m over a 10-min period. Survey order and observers were changed across visits to avoid bias related to bird activity, time of day, or observer experience (Lloyd 2008 a, b; Sevillano-Ríos and Rodewald

2017). We excluded large raptors, aquatic birds, and birds identified only to genus.

Based upon published literature (Fjeldså and Krabbe 1990; Fjeldså 1992; Servat 2006;

Cahill et al., 2008; Lloyd 2008 a, b, c; Sevillano-Ríos et al., 2018), each species was assigned to one of three habitat guilds: (1) Polylepis forest (species primarily dependent or strongly associated with Polylepis forest); (2) grassland Puna (species primarily associated with open areas dominated by grasslands such as Stipa or Calamagrostis spp.); or (3) shrubland (species primarily associated with shrublands composed of Baccharis spp., Berberis spp., Gynoxys spp.,

Weinmannia spp., Senecio spp., Lupinus spp. Brachyotum spp., among others). Birds of conservation concern (15 species) included specialists of Polylepis forest (11 species), endemics to Peru (10 species), and species included in The International Union for Conservation of

Nature’s Red List of Threatened Species (IUCN 2018) (6 species) (Table S2.1.).

We manually calculated the amount of Polylepis forest within a 200 m radius around each survey point (12.5 ha in area) using the “Measure Area” tool of QGIS -v3.0-Girona (QGIS

Development Team 2018) on the same Google Earth Pro images (mean ± SD = 4.46 ha ± 3.88; range = 0 ̶ 12.5 ha). Although our landscapes were smaller than often used in other studies, the size was consistent with the relatively small scale of movements described (Lloyd and Marsden

2008). Our landscape size also avoided problems with spatial autocorrelation of the amount of habitat among points (sill = -10.3, range = 104,698 m, nugget = 12.1) (Figure S2.1.).

Data analysis a) Occupancy models and sensitivity to landscape attributes

We applied a multispecies occupancy model (MSOM) to our survey data (Dorazio and Royle

2005; Zipkin et al., 2009; Kéry and Royle 2016). This approach uses species-specific models

of occurrence within a hierarchical (i.e., multilevel) framework, while accounting for

heterogeneity in species detectability and habitat relationships (Zipkin et al., 2009; Iknayan et

al., 2014; Wright et al., 2020). MSOMs make inferences about the aggregated response of all

modeled species in a community by specifying a common mean and variance

hyperparameters for the occupancy and detection parameters for each species (Kéry and

Royle 2016). The method can incorporate the responses of rare species (Zipkin et al., 2009;

Tingley and Beissinger 2013, Wright et al., 2020) and estimate species richness while

accounting for species not detected during surveys (Dorazio et al., 2006; Kéry and Royle

2008; Kéry and Royle 2016). By stacking the detection histories by year, we allowed the

probabilities of occupancy and detectability to be modeled independently for each year while

using a “single-season” occupancy model approach. This parameterization is an alternative to

multiseason models where the parameters of colonization and extinction are unable to

20 converge due to a limited number of observations (e.g. endemic and rare species)

(MacKenzie et al., 2003; Broms et al., 2016).

Species observations ( yi,j,k, of species i = 1,2,…,88 at site j = 1,2,…,561 (187 sites stacked in 3 years) during each survey k = 1,..,5) were modeled as imperfect observations of true occurrence states, zi,j, given a certain detection probability, pi,j,k. Detection was modeled as yi,j,k│zi,j ~ Bernoulli (zi,j pi,j,k), where zi,j = 1 if species i was present at site j, or zi,j = 0 if not.

As elevation can modulate the effects of site covariates (McCain 2007, 2009), we simultaneously estimated species-specific occupancy probabilities in function of linear, quadratic, and interactive relationships of elevation with patch size, and amount of Polylepis forest within a 200m-radio. The year effect was defined by two dummy variables, y2 and y3 that correspond to the second (2015) and third (2016) survey year. An initial full model (Eq.

1) was defined as:

2 2 logit(ψi,j) = β0i + β1i (elevj)+ β2i (elevj ) + β3i (p_sizej) + β4i (p_sizej ) (Eq. 1)

2 + β5i (a_forj) +β6i (a_forj ) + β7i (elevj * p_sizej) + β8i (elevj * a_forj)

+ β9i (p_sizej * a_forj) + β10i (elevj + p_sizej * a_forj)

+ β11i (y2j) + β12i (y3j)

where β0i was the species-specific intercept and first year (2014) effect, β1i...β12i where the species-specific occupancy model coefficients treated as random effects with a normal

2 2 distribution of their hyperparameters μ and σ , such as β0i,...β12i ~ Normal(μβ0,…, μβ12, σ β0,…,

2 σ β12). Coefficients β1, β3 and β5 represented the linear main effects of elevation, forest patch size and amount of forest for species i while coefficients β2, β4 and β6 were the main effects of the quadratic components respectively. The coefficients β7, β8, and β9 represented the interactive effect of the three covatiates whereas β10 represent their full interaction.

Parameters β11 and β12 are the second- and third-year effect respectively for each j site.

2 2 Likewise, μβ0,…, μβ12 and σ β0,…, σ β12 were the mean and variance community responses

2 (across species) to each covariate (e.g. μβ3 , σ β3 for forest patch size) (Kéry and Royle 2008).

Following the recommendations of Kéry and Royle (2016) and Broms et al., 2016, we used uninformative priors for the mean (μ) and variance (σ2) of the hyper-parameters (e.g. normal distribution with mean zero and variance of 0.1 for the mean community response (μ), and uniform distribution with mean zero and variance of 5 for the community standard deviation

(σ)). This initial model was later contrasted to a reduced model (Eq. 2) that did not include the interactive terms with elevation:

2 2 logit(ψi,j) = β0i + β1i (elevj) + β2i (elevj ) + β3i (p_sizej) + β4i (p_sizej ) (Eq. 2)

2 + β5i (a_forj) +β6i (a_forj ) + β7i (p_sizej * a_forj)

+ β8i (y2j) + β9i (y3j)

As our ability to detect the species could vary along the day and across open and forested sites; detectability was modeled for each species i using forest cover, survey time (e.g., 06:30 h = 6.5) and survey year defined by the same two dummy variables, y2 and y3:

logit(pi,j,k) = α0i + α1i * forcovj + α2i * timej,k + α3i * y2j + α4i * y3j (Eq. 3)

where α0i ,…, α4i were the species-specific detectability model coefficients treated as

2 random effects with a normal distribution of their hyperparameters μ and σ , that follow α0i

2 … α4i ~ Normal(μα0- α4, σ α0- α4).

Models were run using JAGS (Plummer 2003) via R (v 3.3.1; R Development Core Team

2019) and the “jagsUI” package (v 1.4.4; Kellner 2016). We standardized all the continuos

covariates (elev, p_size, a_for, forcov, time) around a mean of zero. For each analysis, we

ran three parallel Markovian chains of 20,000 iterations, applying burn-in to the first 5,000

and thinning rate of 15, which left us with 3,000 samples to build the posterior distribution of

the different parameters. Though a visual inspection of traceplots and using the Brooks-

Gelman-Rubin (BGR, Brooks and Gelman 1998) convergence diagnostic value of Rhat <1.1

(Gelman et al., 2004), we assessed the convergence of the results. Both, the full and reduced

model were evaluated by comparing their respective DIC (Deviance Information Criterion;

Spiegelhalter et al., 2002) values and by direct observations of their interactive effects. The

model structure and model script are provided in the Supplementary Information A and B. b) Local species richness

The number of species expected to occur at each survey site was a derived community metric

from the posterior draws of Markov Chain Monte Carlo (MCMC) runs. This, allowed the

incorporation of uncertainty of the parameters into the local species richness estimation that

followed:

푁j = ∑ 푧 i,j (Eq. 4) 푖=1

where Nj is the total number of species occurring at site j from the true occurrence state zi,j.

Relationships between species richness and elevation, patch size and forest cover were then

evaluated by simple ordinary least squares (OLS) regressions (Tingley and Beissinger 2013).

23 c) Cumulative species-area curve analysis

We discerned effects of patch size from amount of forest by estimating cumulative species richness and comparing slopes of species–area relationships among a set of samples ordered by ascending versus descending patch size while controlling for the cumulative amount of habitat surveyed (sensu Quinn and Harrison (1988) and Fahrig (2017)). “Cumulative forest area” referred to the consecutive sum of different Polylepis patches ordered from small to large and vice versa. The cumulative species–area curves associated with both ordering schemes were then constructed from the estimated species richness of each cumulative forest area (Fahrig 2020).

The following relationships were possible: 1) species richness increases with patch size, such that a single large patch supports more species than several smaller patches of equal cumulative area (Figure 2.4a); 2) species richness declines with increasing patch size, where smaller patches contain more species than a large patch whose area is equal to the sum of the smaller patches

(Figure 2.4b); or 3) no association with patch size; instead species richness increases with total area of habitat, irrespective of the number of patches and therefore, both curves overlaps (Figure

2.4c). A positive response to patch size was indicated by a steeper cumulative species-area curve when patches were ordered from large to small versus when ordered from small to large (Figure

2.4a), whereas the opposite pattern indicated a negative response. Similar slopes were interpreted to indicate lack of evidence for a meaningful response to patch size (Fahrig 2013) (Figure 2.4c).

Although our MSOM facilitates estimating local species richness at each site, it does not allow to estimate cumulative species–area curves because species identities were not retained.

Therefore, we estimated species richness using the R package “iNEXT” (iNterpolation and

EXTrapolation), a new unbiased species richness estimator based on Hill numbers and sample completeness that retains species identity (Chao and Chiu 2014; Hsieh et al., 2016; Hsieh et al.,

2018). The calculation was based upon an abundance matrix, Y ≡ {yij, i, j} where i represents each of the observed (Sobs) and unobserved (SDA) species during the study on j = 1,…, J cumulative forest area (Figure S2.2). Given that multiple (x = 1,…10) points were surveyed k = 3,... 5 times on each patch, yij is the consecutive sum of different species observations after k visits to x surveys points. Then, yij ≥1 when species i was detected in any of the k visits at survey point x within the j cumulative forest area and yij = 0 when it was not detected at all (Hsieh et al., 2016).

RESULTS

In total, 13,780 observations of 100 bird species were registered, although only 13,210 observations of 88 bird species were included in the analysis after removing raptors, aquatic birds, and birds identified only to genus (Table S2.2.). Over the three seasons of sampling, 13 of the 15 species of conservation concern had at least 30 detections: Xenodacnis parina (Tit-like

Dacnis; 900 observation at 164 different sites), Scytalopus affinis (Ancash Tapaculo; 404, 146),

Cranioleuca baroni (Baron’s Spinetail; 386, 134), Metallura phoebe (Black Metaltail; 294, 114),

Grallaria andicolus (Stripe-headed Antpitta, 261, 114), Leptasthenura pileata (Rusty-crowned tit-spinetail, 230, 103), Atlapetes rufigenis (Rufous-eared Brushfinch, 228, 100), Geocerthia serrana (Striated Earthcreeper, 181, 95), Conirostrum binghami (Giant Conebill, 158, 80),

Leptasthenura yanacensis (Tawny Tit-Spinetail, 105, 41), Zaratornis stresemanni (104, 41),

Microspingus alticola (Plain-tailed Warbling-Finch, 67, 49), Anairetes alpinus (32, 15).

Conclusions are based on our reduced MSOM model, which received the strongest support (Full

Model: pD = 46894.6 and DIC = 76089.87; Reduced Model: pD = 18757.6 and DIC = 47643.52) and lacked significant interactive effects among elevation, patch size and amount of forest

(Figure S2.3, Figure S2.6).

In general, the High-Andean bird community of Cordillera Blanca was characterized by many rare, but only a few common species (e.g. Tit-like Dacnis). Avian occurrence and detectability, though variable across species, were usually low each year (Occurrence: mean ±

SD = 0.24 ± 0.05, range = 0.01 ̶ 0.89; detectability: mean ± SD = 0.06 ± 0.001, range=0.1 ̶

0.64) (Figure S2.4.). Some species were nearly ubiquitous and easily detected across years (e.g.

Xenodacnis parina, psi=0.80, p= 0.63); others were rare and difficult to detect when present

(e.g. Anairetes alpinus, psi=0.09, p=0.31) despite being consistently observed in the same few sites (outside of the point count period) among multiple years.

Species richness and composition response

The estimated number of species occurring at each site slightly differed among years (Figure

S2.5.), yet only year 3 had a significant effect with respect to the first year (Table 2.1.; μα4 (95%

PI) = -0.32 (-0.56, -0.1)). Local species richness slightly peaked at mid-elevation (~3,800 m) and declined with forest patch size and amount of Polylepis forest in the landscape independently of the year (Figure 2.5). Estimated number of species at a point was 19 species (range = 13 ̶ 28) in year one, 22 (14 ̶ 35) in year two, and 23 (13 ̶ 34) in year three. Some of the highest estimates corresponded to points at ~4,000 m, within fragments under 50 ha or in landscapes with less than

50% of forest cover.

Avian composition within Polylepis forests contrasted sharply with the comparatively more homogeneous composition of Puna and shrubland landscape matrix (Figure 2.6). Polylepis forests were dominated by forest-associated species (~70%, especially aerial (55%) and terrestrial (26.4%) insectivores) and species of conservation concern (~60%), irrespective of forest patch size (Figure 2.6a, c, e, f). In contrast, most species detected within the matrix were species associated to Puna (29.4% of species detected) and shrublands (28%) although species

26 associated with forest (42.6%) were surprisingly common (Figure 2.6a). Foraging guilds in the matrix were dominated by aerial insectivores (37.6%), terrestrial insectivores (29.4%), and granivores (22%) (Figure 2.6c, d).

Community response

Mean occupancy for the entire High-Andean bird community peaked around 3,800 m (Figure

2.7a) (linear component: μβ1 (95% PI) = -0.51 (-0.84, -0.17); quadratic component: μβ2 (95% PI)

= -0.22 (-0.36, -0.09)). Mean community occupancy increased with patch size (Figure 2.7b)

(linear component: μβ3 (95% PI) = 0.05 (-0.22, 0.36); quadratic component: μβ4 (95% PI) = 0.03

(-0.15, 0.2) but declined with the amount of Polylepis forest in the landscape (Figure 2.7c)

(linear component: μβ5 (95% PI) = -0.3 (-0.54, -0.06)) (Table 2.1). Overall, our estimates of bird community occupancy and detectability were comparable across years (Figure S2.3), though there was some variability related to effect sizes among years. Notably, patch size was significant in year two (μβ8 (95% PI) = 0.41 (0.13, 0.67) but not year three (μβ9 (95% PI) = 0.39

(-0.04, 0.84). Detection probability declined slightly with forest cover (Figure 2.7d) (μα1 (95%

PI) = -0.22 ( -0.33, -0.11), over the course of a day (μα2 (95% PI) = -0.09 (-0.14, -0.03)) (Figure

2.7e), and in year three (2016) (Figure 2.7f; μα4 (95% PI) = -0.32 (-0.56, -0.1)).

Sensitivity of individual species

Occupancy probability for 63 of 88 bird species was significantly associated with elevation (46 linear and 17 quadratic relationships) (Figure 2.8a, b; Table S2.1., Figure S2.7). Size of Polylepis patches was not significantly related to occupancy of any bird species (Figure 2.8c – d). Forest cover within 200 m was negatively related to 18 species and positively to three species (Figure

2.8e; Figure S2.7), two of which were species of conservation concern that primarily forage in

Polylepis trunks (the near-threatened Giant Conebill and endemic Baron's Spinetail). Sixteen

27 species that declined with increasing amount of forest were generalists or species that use shrubland or Puna habitats (Table S2.2.; Supplementary Information B). No species responded quadratically to the amount of forest or the interaction of patch size and amount of forest (Figure

2.8f; Figure S2.7).

Our four predictions of relationships with patch size and forest cover captured all species- specific occurrence profiles (Figure 2.1; Figure 2.9). Occupancy of Giant Conebill and Baron's

Spinetail were negatively related to forest cover (Figures 2.9a). In contrast, neither patch size nor forest cover were related to seven other Polylepis-associated species of conservation concern

(Rufous-eared Brushfinch, Stripe-headed Antpitta, Rusty-crowned tit-spinetail, Black Metaltail,

Plain-tailed Warbling-Finch, Ancash Tapaculo, Tit-like Dacnis) (Figure 2.9b). Occurrence of only one species of conservation concern (White-cheeked Cotinga) along with one widespred and common Andean bird (Rufous-breasted Chat-tyrant) increased with patch size (Figure 2.9c).

Most Puna species, including one species of conservation concern (Striated Earthcreeper), were more likely to occur in smaller patches and in areas with less forest cover (Figure 2.9d). Finally, no threshold effects (a faster reduction) were observed for any bird species.

Cumulative species-area curves

Cumulative species-area curves suggested a negative association between species richness and patch size overall, and for forest and shrubland birds in particular (Figure 2.10), such that greater numbers of species were detected in smaller than larger patches. No clear pattern was detected for species associated with Puna grasslands (Figure 2.10c). Notably, however, the two curves

(LS and SL) overlapped in all cases with ~400 - 600 ha of cumulative forest cover, suggesting that birds were less sensitive to patch size in landscapes with comparably more forest.

DISCUSSION

We examined the extent to which patch size and amount of Polylepis forest within 200 m were related to occurrence of High-Andean birds across an elevation gradient (3,300 – 4,700 m).

Overall, the patterns we detected at both community and species levels were consistent with the hypothesis that many High Andean birds, especially Polylepis specialists, are adapted to naturally patchy landscapes. Specifically, we found that most species responded more strongly to habitat loss than fragmentation per se, as measured by forest cover and patch size or showed no response to patch and landscape characteristics. These results contrast with decades of research from heavily forested systems in temperate and lowland tropical regions, in which researchers have documented harmful consequences of fragmentation to species that are habitat specialists, range-restricted, sensitive to disturbance, have low vagility, and/or occur within small populations (Saunders et al., 1991; Turner 1996; Devictor et al., 2008). Yet despite possessing many traits that are typically associated with vulnerability to fragmentation, most Polylepis-Puna species seemed relatively insensitive to changes in landscape configuration, like forest patch size, and were able to exploit small or remnant patches as long as sufficient forest remained within the landscape.

The importance of the total amount of Polylepis forest

Species occurrence was better explained by the total amount of Polylepis forest within 200 m than patch size, though we recognize that species did not uniformly show this pattern (Figures

2.8 and 2.9). For example, we estimate occupancy probability of the Giant Conebill should decline by roughly half (0.54, with a range from 0.83 to 0.29) given a loss of 90% of forest within the landscape, but only by 0.34 (from 0.90 to 0.56) when patch size is reduced from 200 to 1 ha (Figure 2.9a, b). Likewise, occupancy by Baron's Spinetail is expected to decline by half

29 to similar losses offorest cover but show little response to changes in patch size. Despite some variation among species, our results suggest that habitat loss would have more severe consequences for these species than habitat fragmentation, as indicated by patch size. In addition to our hypothesis that High-Andean birds are adapted to patchy landscapes, we suggest two additional contributors to the patterns we describe. First, the specific habitat features required by many Polylepis specialists are likely to be a function of the amount of forest rather than patch size (Lloyd 2008a; Lloyd and Marsden 2008; Sevillano-Ríos and Rodewald 2017). Second, many Polylepis specialists readily use matrix habitats and the resources they provide (Tinoco et al., 2013).

Our failure to detect strong responses to patch size is unlikely to be an artifact of working within intact or heavily forested landscapes. Previous research shows that the effects of patch size and isolation usually manifest when less than ~30% of habitat remains in a landscape

(Andrén 1994). Throughout most of the Andes, Polylepis woodlands comprise less than 20% of their original cover (although these estimates are debated, see Valencia et al., 2018), with 8-23% cover in our study region in Cordillera Blanca (Table 2.2). Another source of bias could be paucity of detections for certain species along the range of patch size and amount of forest.

However, our results are more precise for small patches as uncertainy increase with patch size and forest cover (Figure 2.9). In this way, our findings are consistent with the Habitat Amount

Hypothesis (HAH), which posits that habitat availability is more important than spatial configuration (Fahrig 2013), and the idea that many High Andean birds ̶ specialized to

Polylepis forest ̶ are adapted to this patchy landscape configurations but not to habitat loss.

The importance of small Polylepis patches

Most conservation and restoration efforts place high value on large Polylepis forests, sometimes to the exclusion of small fragments (Lloyd 2008a, b; Lloyd and Marsden 2008;

Sevillano-Ríos and Rodewald 2017), but our results challenge this assumption. Rather, we provide a striking illustration of the ability of many Polylepis specialist to use very small

Polylepis fragments (<1 ha). In fact, small patches of Polylepis forest in the Cordillera Blanca were as likely to be used as large patches by seven forest specialists of conservation concern, including the endangered and endemic Plain-tailed Warbling-Finch (Figure 2.9).

The weak signal for patch size may reflect the fact that small patches often contain habitat features required by certain specialists, or in other cases, that some species prefer or regularly use habitat edges. The Tawny Tit-Spinetail, for which 56% of records were in fragments smaller than 6 ha (mean patch size = 1.71 ha, SD = 2.07), provides an example of the former, as the species requires undisturbed forest with tall or mature Polylepis trees, rich in mosses cover and steep rocky terrain at high elevations (>4000 m) (Fjeldså and Krabbe 1990;

Lloyd 2008b; Sevillano-Ríos et al., 2011; Sevillano-Ríos and Rodewald., 2017). On the other hand, a preference for habitat edges may explain why Ash-breasted Tit-tyrant occurred either within small patches (mean patch size = 2.34 ha, SD = 2.3, range = 0.046 ̶ 5.6 ha) or along the irregular edge of a large forest patch (mean patch size = 47.7 ha, SD = 14.76, range = 17.6 ̶

53.77 ha; Figure 2.3a-b).

Small Polylepis fragments also can be a used as secondary habitats, as suggested by our observations that some Polylepis specialists patrol (individually, in flocks, or in multispecies flocks) several small patches for specific resources (e.g insects, flowers and fruits), probably as part of their home ranges. We observed patrolling behavior for the Giant Conebill, the Rufous-

31 eared Brushfinch, the Baron's Spinetail, Rusty-crowned Tit-Spinetail, Plain-tailed Warbling-

Finch and the White-cheeked Cotinga (Sevillano-Ríos 2016; Sevillano-Ríos and Rodewald

2017). Certainly, the home ranges of some Polylepis specialist are composed of multiple patches, including small ones, that are visited routinely everyday (De Coster et al., 2009; Lloyd and

Marsden 2011). Our finding that small Polylepis patches are important habitats for multiple

High-Andes species of conservation concern signals another potential risk given that small patches may be most vulnerable to loss or degradation (Wintle et al. 2019; Lindenmayer 2019).

High-Andes species richness and composition

Avian species richness within our patchy system was greatest within small fragments and in landscapes with comparably low amount of forest cover (and therefore more isolation, Fahrig

2003) (Figure 2.5). This finding is consistent with reports from Cuzco (Peru) (Lloyd 2008a;

Lloyd and Marsden 2008), Cochabamba (Bolivia) (Hjarsen 1998; Matthysen et al., 2008), and

Jujuy, Tucumán and Córdoba (Argentina) (Bellis et al., 2009). Although inconsistent with predictions from Theory of Island Biogeography, where the number of species in a fragment is defined in terms of patch size and isolation (MacArthur and Wilson 1967), our results underscore the importance of matrix attributes (Rodewald and Yahner 2001, Rodewald 2003), patch quality

(Mortelliti et al., 2010), and amount of habitat (Andrén 1994; Fahrig 2013) in terrestrial systems where the matrix is not considered to be “inhospitable” but instead can strongly shape community organization and structure (Ricketts 2001; Rodewald 2003; Prugh et al., 2008).

In the High Andes, Puna and shrubland habitats were used by many High-Andean birds that also use Polylepis forests to some extent (Lloyd 2008a; Tinoco et a., 2013; Sevillano-Ríos and Rodewald 2017). Notable, Gynoxys plants seems to play an important role in the Polylepis landscapes of Ecuador (Tinoco et al., 2013; Astudillo et al., 2019). In this way, matrix attributes

32 should modulate the degree of connectivity among fragments, influence the nature of the edge effects, and be the source of not only many opportunistic and generalist species, but also forest associated species (e.g. Tit-like Dacnis). Further studies should consider that the influence of matrix may change with latitude based on climatic and floristic conditions (Rundel et al., 1994), modulating the effect of patch size and isolation across the Andes (Bender & Fahrig 2005;

Franklin and Lidenmayer 2009; Prug et al., 2008; Watling et al 2011).

Conservation implications

Although many others have studied the relative importance of landscape configuration versus composition (Trzcinski et al., 1999; De Camargo et al., 2018; Shoffner et al., 2018; Valente and

Betts 2018), we believe ours is the first to explicitly test these differences in a tropical mountain system. We recognize that fragmentation negatively impacts biodiversity and drives species loss in many systems (Andrén 1994; Hanski, 2015; Haddad et al. 2015; Pfeifer et al., 2017; Martin

2018; Fletcher et al., 2018), but we show here that patchy landscapes and small patches can contribute to biodiversity conservation (Ethier and Fahrig 2011, Fahrig 2017; Fahrig et al., 2019.

Our work has three central implications for conservation.

First, conservation efforts should focus most heavily on avoiding loss of Polylepis forest rather than on avoiding fragmentation per se. Although landscape configuration is important in some contexts (Andrén 1994; Thompson et al., 2002) and for some species (Bender et al., 1998;

Villard et al., 1999; Lee et al., 2002), birds in our system responded more strongly to forest cover than patch size. That said, we caution that fragmentation might promote species richness in small fragments by increasing the abundance of generalist and matrix-associated species that might conceivably displace other species of conservation concern or have other negative outcomes

(Debinski and Holt 2000; Haddad et al., 2015).

Second, small patches of Polylepis seem to contribute as much to conservation as large fragments, given the same amount of forest within the landscape. Even the loss of small fragments can have potentially serious consequences, especially if they contain special habitat features absent in larger fragments (Lloyd 2008 a, b; Lloyd and Marsden 2008; Tinoco et al.,

2013; Sevillano-Ríos and Rodewald 2017). As first noted by Fjeldså (1993), the persistence of many Polylepis specialists in small and isolated Polylepis forest is no less than remarkable. Our study reinforces this observation and further provides evidence that many Polylepis species are adapted to patchiness.

Third, attributes of the landscape matrix play important roles in structuring Andean bird communities (Lloyd 2008a; Tinoco et al., 2013; Sevillano-Ríos and Rodewald., 2017; Astudillo et al., 2019). Specifically, a more heterogeneous matrix (e.g shrublands) is known to enhance connectivity and bird movements (Tinoco et al., 2013; Astudillo et al., 2019). Plant species like

Gynoxys, Baccharis, Buddleia, Lupinus among others, should be included in matrix restoration projects as many bird species seems to depend of the diversity of resources that they provide

(e.g. nectar, insects, fruits).

ACKNOWLEDGMENTS

This research was funded by FONDECYT-CONCYTEC (Nº 237-2015-FONDECYT). We are also grateful to the Fulbright Scholarship and support from Cornell University that supported

CSSR during his graduate studies. We also deeply appreciate financial support provided by

Cornell Lab of Ornithology, the Athena Grant, the Bergstrom Award of the Association of Field

Ornithologists as well as the American Alpine Club Research Grants 2017. Thanks to Jessica

Pisconte, Nathaniel Young, Rosaura Watanabe, Nicolas Mamani, Julio Salvador, Celia Sierra and James Purcell for being the core field crew for data gathering and data entry; to the

Huascaran National Park staffing, Ing. Jesús Gómez, Selwyn Valverde, Martín Salvador, and

Oswaldo Gonzales for facilitating permits; as well as to many park rangers that granted access and helped in different ways. We thank to André Dhondt, Stephen Morreale, Viviana Ruíz-

Gutiérrez, Wesley M. Hochachka, Lynn Marie Johnson and Laura Morales for comments, recommendations and guidance that helped to improve this manuscript.

REFERENCES

Andrén, H. (1994). Effects of habitat fragmentation on birds and mammals in landscapes with

different proportions of suitable habitat: a review. Oikos, 355-366.

Astudillo, P. X., Schabo, D. G., Siddons, D. C., & Farwig, N. (2019). Patch-matrix movements

of birds in the páramo landscape of the southern Andes of Ecuador. Emu-Austral

Ornithology, 119(1), 53-60.

Bellis, L. M., Rivera, L., Politi, N., Martin, E., Perasso, M. L., Cornell, F., & Renison, D. (2009).

Latitudinal patterns of bird richness, diversity and abundance in Polylepis australis

mountain forest of Argentina. Bird Conservation International, 19(3), 265-276.

Bender, D. J., & Fahrig, L. (2005). Matrix structure obscures the relationship between interpatch

movement and patch size and isolation. Ecology, 86(4), 1023-1033.

Bissonette, J. A., & Storch, I. (Eds.). (2002). Landscape ecology and resource management:

linking theory with practice. Island Press.

Brooks, S. P., & Gelman, A. (1998). General methods for monitoring convergence of iterative

simulations. Journal of computational and graphical statistics, 7(4), 434-455.

Broms, K. M., Hooten, M. B., & Fitzpatrick, R. M. (2016). Model selection and assessment for

multi‐species occupancy models. Ecology, 97(7), 1759-1770.

Cahill, J. R. A., Matthysen, E., & Huanca, N. E. (2008). Nesting biology of the giant conebill

(Oreomanes fraseri) in the High Andes of Bolivia. The Wilson Journal of Ornithology,

120(3), 545-549.

Chao, A., & Chiu, C. H. (2014). Species richness: estimation and comparison. Wiley StatsRef:

Statistics Reference Online, 1-26.

Chepstow-Lusty, A., Bush, M. B., Frogley, M. R., Baker, P. A., Fritz, S. C., & Aronson, J.

(2005). Vegetation and climate change on the Bolivian Altiplano between 108,000 and

18,000 yr ago. Quaternary Research, 63(1), 90-98.

Cingolani, AM; M. V Vaieretti; MA Giorgis; N La Torre; JI Whitworth-Hulse & D Renison.

2013. Can livestock and fres convert the sub-tropical mountain rangelands of central

Argentina into a rocky desert? The Rangeland Journal, 35:285-297.

Collinge, S. K. (2009). Ecology of fragmented landscapes. JHU Press.

Cote, J., Clobert, J., Brodin, T., Fogarty, S., & Sih, A. (2010). Personality-dependent dispersal:

characterization, ontogeny and consequences for spatially structured

populations. Philosophical Transactions of the Royal Society B: Biological

Sciences, 365(1560), 4065-4076.

De Camargo, R. X., Boucher‐Lalonde, V., & Currie, D. J. (2018). At the landscape level, birds

respond strongly to habitat amount but weakly to fragmentation. Diversity and

Distributions, 24(5), 629-639.

De Coster, G., Matthysen, E., Cahill, J. R., & Lens, L. (2009). Home range characteristics of the

Near Threatened Giant Conebill Oreomanes fraseri in fragmented Polylepis forest. Bird

Conservation International, 19(3), 215-223.

Debinski, D. M., & Holt, R. D. (2000). A survey and overview of habitat fragmentation

experiments. Conservation Biology, 14(2), 342-355.

Devictor, V., Julliard, R., & Jiguet, F. (2008). Distribution of specialist and generalist species

along spatial gradients of habitat disturbance and fragmentation. Oikos, 117(4), 507-514.

Dorazio, R. M., & Royle, J. A. (2005). Estimating size and composition of biological

communities by modeling the occurrence of species. Journal of the American Statistical

Association, 100(470), 389-398.

Dorazio, R. M., Royle, J. A., Söderström, B., & Glimskär, A. (2006). Estimating species richness

and accumulation by modeling species occurrence and detectability. Ecology, 87(4), 842-

854.

Duckworth, R. A., & Kruuk, L. E. (2009). Evolution of genetic integration between dispersal and

colonization ability in a bird. Evolution, 63(4), 968-977.

Ethier, K., & Fahrig, L. (2011). Positive effects of forest fragmentation, independent of forest

amount, on bat abundance in eastern Ontario, Canada. Landscape Ecology, 26(6), 865-876.

Fahrig, L. (1997). Relative effects of habitat loss and fragmentation on population

extinction. The Journal of Wildlife Management, 603-610.

Fahrig, L. (2002). Effect of habitat fragmentation on the extinction threshold: a

synthesis. Ecological applications, 12(2), 346-353.

Fahrig, L. (2003). Effects of habitat fragmentation on biodiversity. Annual review of ecology,

evolution, and systematics, 34(1), 487-515.

Fahrig, L. (2013). Rethinking patch size and isolation effects: the habitat amount

hypothesis. Journal of Biogeography, 40(9), 1649-1663.

Fahrig, L. (2017). Ecological responses to habitat fragmentation per se. Annual Review of

Ecology, Evolution, and Systematics, 48, 1-23.

Fahrig, L., Arroyo-Rodríguez, V., Bennett, J. R., Boucher-Lalonde, V., Cazetta, E., Currie, D. J.,

... & Koper, N. (2019). Is habitat fragmentation bad for biodiversity?. Biological

conservation, 230, 179-186.

Fahrig, L. (2020). Why do several small patches hold more species than few large

patches?. Global Ecology and Biogeography, 29(4), 615-628.

Ferraz, G., Nichols, J. D., Hines, J. E., Stouffer, P. C., Bierregaard, R. O., & Lovejoy, T. E.

(2007). A large-scale deforestation experiment: effects of patch area and isolation on

Amazon birds. science, 315(5809), 238-241.

Fine, P. V., García-Villacorta, R., Pitman, N. C., Mesones, I., & Kembel, S. W. (2010). A

floristic study of the white-sand forests of Peru. Annals of the Missouri Botanical Garden,

97(3), 283-305.

Fjeldså, J. (1992). Biogeographic patterns and evolution of the avifauna of relict high-attitude

woodlands of the Andes. Steenstrupia, (18), 9-62.

Fjeldså, J. (1993). The avifauna of the Polylepis woodlands of the Andean highlands: the

efficiency of basing conservation priorities on patterns of endemism. Bird Conservation

International, 3(01), 37-55.

Fjeldså, J. (2002). Polylepis forests - Vestiges of a vanishing ecosystem in the Andes.

Ecotropica, 8, 111–123.

Fjeldså, J., & Krabbe, N. (1990). Birds of the High Andes. Zoological Museum, University of

Copenhagen.

Fjeldså, J., Kessler, M., Engblom, G., & Driesch, P. (1996). Conserving the biological diversity

of Polylepis woodlands of the highland of Peru and Bolivia: a contribution to sustainable

natural resource management in the Andes (p. 250). Copenhagen: Nordeco.

Fletcher Jr, R. J., Didham, R. K., Banks-Leite, C., Barlow, J., Ewers, R. M., Rosindell, J., ... &

Melo, F. P. (2018). Is habitat fragmentation good for biodiversity?. Biological

conservation, 226, 9-15.

Franklin, Jerry F.; Lindenmayer, David B. (2009) Importance of matrix habitats in maintaining

biological diversity. Proceedings of the National Academy of Sciences, 2009, vol. 106, no

2, p. 349-350.

Gaggiotti, O. E., & Hanski, I. (2004). Mechanisms of population extinction. In Ecology, genetics

and evolution of metapopulations (pp. 337-366).

Gareca, E. E., Hermy, M., Fjeldså, J., & Honnay, O. (2010). Polylepis woodland remnants as

biodiversity islands in the Bolivian high Andes. Biodiversity and conservation, 19(12),

3327-3346.

Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2004). Bayesian data analysis 2nd edn

Chapman & Hall. CRC, Boca Raton FL.

Gosling, W. D., Hanselman, J. A., Knox, C., Valencia, B. G., & Bush, M. B. (2009). Long‐term

drivers of change in Polylepis woodland distribution in the central Andes. Journal of

Vegetation Science, 20(6), 1041-1052.

Grez, A., Zaviezo, T., Tischendorf, L., & Fahrig, L. (2004). A transient, positive effect of habitat

fragmentation on insect population densities. Oecologia, 141(3), 444-451.

Haddad, N. M., Brudvig, L. A., Clobert, J., Davies, K. F., Gonzalez, A., Holt, R. D., ... & Cook,

W. M. (2015). Habitat fragmentation and its lasting impact on Earth’s ecosystems. Science

Advances, 1(2), e1500052.

Hanselman, J. A., Bush, M. B., Gosling, W. D., Collins, A., Knox, C., Baker, P. A., & Fritz, S.

C. (2011). A 370,000-year record of vegetation and fire history around Lake Titicaca

(Bolivia/Peru). Palaeogeography, Palaeoclimatology, Palaeoecology, 305(1-4), 201-214.

Hanski, I. (2015). Habitat fragmentation and species richness. Journal of Biogeography, 42(5),

989-993.

Hastings, A. (1977). Spatial heterogeneity and the stability of predator-prey systems. Theoretical

population biology, 12(1), 37-48.

Henden, J. A., Ims, R. A., Yoccoz, N. G., & Killengreen, S. T. (2011). Declining willow

ptarmigan populations: the role of habitat structure and community dynamics. Basic and

Applied Ecology, 12(5), 413-422.

Hjarsen, T. K. S. (1999). Biological diversity in high altitude woodlands and plantations in the

Bolivian Andes: Implications for development of sustainable land-use. In III Simposio

Internacional De Desarrollo Sustentable De Montanas, Quito, Ecuador (pp. 145-149).

Holzschuh, A., Steffan‐Dewenter, I., & Tscharntke, T. (2010). How do landscape composition

and configuration, organic farming and fallow strips affect the diversity of bees, wasps and

their parasitoids?. Journal of Animal Ecology, 79(2), 491-500.

Hsieh, T. C., Ma, K. H., & Chao, A. (2016). iNEXT: an R package for rarefaction and

extrapolation of species diversity (H ill numbers). Methods in Ecology and Evolution,

7(12), 1451-1456.

Hsieh, T. C., K. H. Ma & Anne Chao. (2018) iNEXT: iNterpolation and EXTrapolation for

species diversity. R package version 2.0.18 URL:

http://chao.stat.nthu.edu.tw/blog/software-download/.

Huffaker, C. (1958). Experimental studies on predation: dispersion factors and predator-prey

oscillations. Hilgardia, 27(14), 343-383.

Hughes, C., & Eastwood, R. (2006). Island radiation on a continental scale: exceptional rates of

plant diversification after uplift of the Andes. Proceedings of the National Academy of

Sciences, 103(27), 10334-10339.

Iknayan, K. J., Tingley, M. W., Furnas, B. J., & Beissinger, S. R. (2014). Detecting diversity:

emerging methods to estimate species diversity. Trends in Ecology & Evolution, 29(2), 97-

106.

IUCN 2018. The IUCN Red List of Threatened Species. Version 2018-2.

http://www.iucnredlist.org

Kareiva, P. (1987). Habitat fragmentation and the stability of predator–prey interactions. Nature,

326(6111), 388.

Kellner, K. (2016). jagsUI: A Wrapper Around 'rjags' to Streamline 'JAGS' Analyses. R package

version 1.4.4. https://CRAN.R-project.org/package=jagsUI

Kendall, W. L., Nichols, J. D., & Hines, J. E. (1997). Estimating temporary emigration using

capture–recapture data with Pollock’s robust design. Ecology, 78(2), 563-578.

Kendall, W. L., & Bjorkland, R. (2001). Using open robust design models to estimate temporary

emigration from capture—recapture data. Biometrics, 57(4), 1113-1122.

Kéry, M., & Royle, J. A. (2008). Hierarchical Bayes estimation of species richness and

occupancy in spatially replicated surveys. Journal of Applied Ecology, 45(2), 589-598.

Kéry, M., Royle, J. A., Plattner, M., & Dorazio, R. M. (2009). Species richness and occupancy

estimation in communities subject to temporary emigration. Ecology, 90(5), 1279-1290.

Kéry, M., & Royle, J. A. (2016). Applied hierarchical modeling in ecology. Volumen 1.

Academic, Waltham.

Kessler, M. (2002). The “Polylepis problem”: where do we stand. Ecotropica,8(2), 97-110.

Kessler, M. (2006). Bosques de Polylepis. Botánica económica de los Andes centrales, 11.

Klingbeil, B. T., & Willig, M. R. (2009). Guild‐specific responses of bats to landscape

composition and configuration in fragmented Amazonian rainforest. Journal of applied

ecology, 46(1), 203-213.

Lee, M., Fahrig, L., Freemark, K., & Currie, D. J. (2002). Importance of patch scale vs landscape

scale on selected forest birds. Oikos, 96(1), 110-118.

Lindenmayer, D. (2019). Small patches make critical contributions to biodiversity conservation.

Proceedings of the National Academy of Sciences, 116(3), 717-719.

Lindenmayer, D. B., & Franklin, J. F. (2002). Conserving forest biodiversity: a comprehensive

multiscaled approach. Island press.

Liu, J., Wilson, M., Hu, G., Liu, J., Wu, J., & Yu, M. (2018). How does habitat fragmentation

affect the biodiversity and ecosystem functioning relationship?. Landscape Ecology, 33(3),

341-352.

Lloyd, H. (2008) a. Abundance and patterns of rarity of Polylepis birds in the Cordillera

Vilcanota, southern Perú: implications for habitat management strategies. Bird

Conservation International, 18(02), 164-180.

Lloyd, H. (2008) b. Influence of within‐patch habitat quality on high‐Andean Polylepis bird

abundance. Ibis, 150(4), 735-745.

Lloyd, H. (2008) c. Foraging ecology of high Andean insectivorous birds in remnant Polylepis

forest patches. The Wilson Journal of Ornithology, 120(3), 531-544.

Lloyd, H., & Marsden, S. J. (2008). Bird community variation across Polylepis woodland

fragments and matrix habitats: implications for biodiversity conservation within a high

Andean landscape. Biodiversity and Conservation, 17(11), 2645-2660.

Lomolino, M. V. (2005). Body size evolution in insular vertebrates: generality of the island rule.

Journal of Biogeography, 32(10), 1683-1699. https://doi.org/10.1111/j.1365-

2699.2005.01314.x

MacArthur, R. H. and Wilson, E. O., (1967). The theory of island biogeography. Princeton, NJ.

Princeton University press.

MacKenzie, D. I., & Royle, J. A. (2005). Designing occupancy studies: general advice and

allocating survey effort. Journal of applied Ecology, 42(6), 1105-1114.

MacKenzie, D. I., Nichols, J. D., Hines, J. E., Knutson, M. G., & Franklin, A. B. (2003).

Estimating site occupancy, colonization, and local extinction when a species is detected

imperfectly. Ecology, 84(8), 2200-2207.

Martin, C. A. (2018). An early synthesis of the habitat amount hypothesis. Landscape ecology,

33(11), 1831-1835.

Matthysen, E., Collet, F., & Cahill, J. (2008). Mixed flock composition and foraging behavior of

insectivorous birds in undisturbed and disturbed fragments of high-Andean Polylepis

woodland. Ornitologia Neotropical, 19, 403-416.

McGarigal, K., & Cushman, S. A. (2002). Comparative evaluation of experimental approaches to

the study of habitat fragmentation effects. Ecological applications, 12(2), 335-345.

Mortelliti, A., Amori, G., & Boitani, L. (2010). The role of habitat quality in fragmented

landscapes: a conceptual overview and prospectus for future research. Oecologia, 163(2),

535-547.

Nyström, M., & Folke, C. (2001). Spatial resilience of coral reefs. Ecosystems, 4(5), 406-417.

Pennington, R. T., Lehmann, C. E., & Rowland, L. M. (2018). Tropical savannas and dry forests.

Current Biology, 28(9), R541-R545.

Pfeifer, M., Lefebvre, V., Peres, C. A., Banks-Leite, C., Wearn, O. R., Marsh, C. J., ... &

Cisneros, L. (2017). Creation of forest edges has a global impact on forest vertebrates.

Nature, 551(7679), 187.

Pirie, M. D., Chatrou, L. W., Mols, J. B., Erkens, R. H., & Oosterhof, J. (2006). ‘Andean‐

centred’genera in the short‐branch clade of Annonaceae: testing biogeographical hypotheses

using phylogeny reconstruction and molecular dating. Journal of Biogeography, 33(1), 31-

46.

Plummer, M. (2003). JAGS: A program for analysis of Bayesian graphical models using Gibbs

sampling. In Proceedings of the 3rd international workshop on distributed statistical

computing (Vol. 124, No. 125, p. 10).

Proctor, J., & Woodell, S. R. (1975). The ecology of serpentine soils. In Advances in ecological

research (Vol. 9, pp. 255-366). Academic Press.

Prugh, L. R., Hodges, K. E., Sinclair, A. R., & Brashares, J. S. (2008). Effect of habitat area and

isolation on fragmented animal populations. Proceedings of the National Academy of

Sciences, 105(52), 20770-20775.

Purcell, J., & Brelsford, A. (2004). Reassessing the causes of decline of Polylepis, a tropical

subalpine forest. Ecotropica, 10, 155-158.

QGIS Development Team (2018). QGIS Geographic Information System. Open Source

Geospatial Foundation Project. http://qgis.osgeo.org”

Quinn, J. F., & Harrison, S. P. (1988). Effects of habitat fragmentation and isolation on species

richness: evidence from biogeographic patterns. Oecologia, 75(1), 132-140.

R Development Core Team, R. F. F. S. C. (2019). R: A language and environment for statistical

computing.

Radford, J. Q., Bennett, A. F., & Cheers, G. J. (2005). Landscape-level thresholds of habitat

cover for woodland-dependent birds. Biological conservation, 124(3), 317-337.

Renison, D; AM Cingolani & R Suarez. 2002. Efectos del fuego sobre un bosquecillo de

Polylepis australis (Rosaceae) en las montañas de Córdoba, Argentina. Revista Chilena de

Historia Natural, 75: 719-727.

Ricketts, T. H. (2001). The matrix matters: effective isolation in fragmented landscapes. The

American Naturalist, 158(1), 87-99.

Rodewald, A. D. 2003. The importance of land uses within the landscape matrix. Wildlife

Society Bulletin 31:586-592.

Rodewald, A. D., & Yahner, R. H. (2001). Influence of landscape composition on avian

community structure and associated mechanisms. Ecology, 82(12), 3493-3504.

Rodríguez, F., & Behling, H. (2012). Late Quaternary vegetation, climate and fire dynamics, and

evidence of early to mid-Holocene Polylepis forests in the Jimbura region of the

southernmost Ecuadorian Andes. Palaeogeography, Palaeoclimatology, Palaeoecology, 350,

247-257.

Rundel, P. W., Smith, A. P., & Meinzer, F. C. (Eds.). (1994). Tropical alpine environments:

plant form and function. Cambridge University Press.

Saunders, D. A., Hobbs, R. J., & Margules, C. R. (1991). Biological consequences of ecosystem

fragmentation: a review. Conservation biology, 5(1), 18-32.

Saura, S., Bodin, Ö., & Fortin, M. J. (2014). EDITOR'S CHOICE: Stepping stones are crucial for

species' long‐distance dispersal and range expansion through habitat networks. Journal of

Applied Ecology, 51(1), 171-182.

Schmidt‐Lebuhn, A. N., Fuchs, J., Hertel, D., Hirsch, H., Toivonen, J., & Kessler, M. (2010). An

Andean radiation: polyploidy in the tree genus Polylepis (Rosaceae, Sanguisorbeae). Plant

Biology, 12(6), 917-926.

Servat, G. P. (2006). The role of local and regional factors in the foraging ecology of birds

associated with Polylepis woodlands (Vol. 70, No. 11).

Sevillano- Ríos, C. S. (2016). Diversity, Ecology, and Conservation of Bird Communities of

Polylepis Woodlands in The Northern Andes of Peru. Master Thesis, Cornell University,

New York.

Sevillano- Ríos, C. S. (2017). Breve Historia de la Ornitología en los Altos Andes del Norte del

Perú y Su Importancia para la Conservación. Revista de Glaciares y Ecosistemas de

Montaña, 2, 87-102.

Sevillano- Ríos, C. S., Lloyd, H., & Valdés-Velásquez, A. (2011). Bird species richness,

diversity and abundance in Polylepis woodlands, Huascaran biosphere reserve, Peru.

Studies on neotropical fauna and environment, 46(1), 69-76.

Sevillano-Ríos, C. S., & Rodewald, A. D. (2017). Avian community structure and habitat use of

Polylepis forests along an elevation gradient. PeerJ, 5, e3220.

Sevillano-Ríos, C. S., Rodewald, A., & Morales, L. (2018). Ecología y conservación de las aves

asociadas con Polylepis: qué sabemos de esta comunidad cada vez más vulnerable. Ecología

Austral, 28, 216-228.

Shoffner, A., Wilson, A. M., Tang, W., & Gagné, S. A. (2018). The relative effects of forest

amount, forest configuration, and urban matrix quality on forest breeding birds. Scientific

reports, 8(1), 17140.

Sih, A., Jonsson, B. G., & Luikart, G. (2000). Habitat loss: ecological, evolutionary and genetic

consequences. Trends in Ecology & Evolution, 15(4), 132-134.

Simpson, B. B. (1979). A Revision of the Genus Polylepis (Rosaceae: Sanguisorbeae).

Smithsonian Contributions to Botany, (43), 1–62. http://doi.org/10.5479/si.0081024X.43.1

Spiegelhalter, D. J., Best, N. G., Carlin, B. P., & Van Der Linde, A. (2002). Bayesian measures

of model complexity and fit. Journal of the royal statistical society: Series b (statistical

methodology), 64(4), 583-639.

Sylvester, S. P., Sylvester, M. D., & Kessler, M. (2014). Inaccessible ledges as refuges for the

natural vegetation of the high Andes. Journal of vegetation Science, 25(5), 1225-1234.

Sylvester, S. P., Heitkamp, F., Sylvester, M. D., Jungkunst, H. F., Sipman, H. J., Toivonen, J.

M., ... & Kessler, M. (2017). Relict high-Andean ecosystems challenge our concepts of

naturalness and human impact. Scientific reports, 7(1), 3334.

Thompson, F. R., Donovan, T. M., DeGraff, R. M., Faaborg, J., & Robinson, S. K. (2002). A

multi-scale perspective of the effects of forest fragmentation on birds in eastern forests. In:

George, T. Luke; Dobkin, David S., eds. Effects of Habitat Fragmentation on Birds in

Western Landscapes: Contrasts with Paradigms from the Eastern United States. Studies in

Avian Biology. 25:8-19

Tingley, M. W., & Beissinger, S. R. (2013). Cryptic loss of montane avian richness and high

community turnover over 100 years. Ecology, 94(3), 598-609.

Tinoco, B. A., Astudillo, P. X., Latta, S. C., Strubbe, D., & Graham, C. H. (2013). Influence of

patch factors and connectivity on the avifauna of fragmented Polylepis forest in the

Ecuadorian Andes. Biotropica, 45(5), 602-611.

Trzcinski, M. K., Fahrig, L., & Merriam, G. (1999). Independent effects of forest cover and

fragmentation on the distribution of forest breeding birds. Ecological Applications, 9(2),

586-593.

Tscharntke, T., Steffan-Dewenter, I., Kruess, A., & Thies, C. (2002). Characteristics of insect

populations on habitat fragments: a mini review. Ecological research, 17(2), 229-239.

Turner, I. M. (1996). Species loss in fragments of tropical rain forest: a review of the evidence.

Journal of applied Ecology, 200-209.

Valencia, B. G., Bush, M. B., Coe, A. L., Orren, E., & Gosling, W. D. (2018). Polylepis

woodland dynamics during the last 20,000 years. Journal of Biogeography.

Valente, J. J., & Betts, M. G. (2019). Response to fragmentation by avian communities is

mediated by species traits. Diversity and Distributions, 25(1), 48-60. van der Hammen, T. (1974). The Pleistocene changes of vegetation and climate in tropical South

America. Journal of Biogeography, 3-26.

Villard, M. A., Trzcinski, M. K., & Merriam, G. (1999). Fragmentation effects on forest birds:

relative influence of woodland cover and configuration on landscape occupancy.

Conservation Biology, 13(4), 774-783.

Watling, J. I., Nowakowski, A. J., Donnelly, M. A., & Orrock, J. L. (2011). Meta‐analysis

reveals the importance of matrix composition for animals in fragmented habitat. Global

Ecology and Biogeography, 20(2), 209-217.

Weng, C., Bush, M. B., Curtis, J. H., Kolata, A. L., Dillehay, T. D., & Binford, M. W. (2006).

Deglaciation and Holocene climate change in the western Peruvian Andes. Quaternary

Research, 66(1), 87-96.

Wiegand, T., Revilla, E., & Moloney, K. A. (2005). Effects of habitat loss and fragmentation on

population dynamics. Conservation Biology, 19(1), 108-121.

Williams, J. J., Gosling, W. D., Brooks, S. J., Coe, A. L., & Xu, S. (2011). Vegetation, climate

and fire in the eastern Andes (Bolivia) during the last 18,000 years. Palaeogeography,

Palaeoclimatology, Palaeoecology, 312(1-2), 115-126.

Wintle, B. A., Kujala, H., Whitehead, A., Cameron, A., Veloz, S., Kukkala, A., ... & Bekessy, S.

A. (2019). Global synthesis of conservation studies reveals the importance of small habitat

patches for biodiversity. Proceedings of the National Academy of Sciences, 116(3), 909-

914.

Wright, A. D., Grant, E. H. C., & Zipkin, E. F. (2020). A hierarchical analysis of habitat area,

connectivity, and quality on amphibian diversity across spatial scales. Landscape

Ecology, 35(2), 529-544.

Yaacobi, G., Ziv, Y., & Rosenzweig, M. L. (2007). Habitat fragmentation may not matter to

species diversity. Proceedings of the Royal Society of London B: Biological Sciences,

274(1624), 2409-2412.

Zipkin, E. F., DeWan, A., & Andrew Royle, J. (2009). Impacts of forest fragmentation on

species richness: a hierarchical approach to community modelling. Journal of Applied

Ecology, 46(4), 815-822.

Zutta, B. R., Rundel, P. W., Saatchi, S., Casana, J. D., Gau-, P., Soto, A., … Buermann, W.

(2012). Prediciendo la distribución de Polylepis: bosques Andinos vulnerables y cada vez

más importantes Predicting Polylepis distribution: vulnerable and increasingly important

Andean woodlands, 19(August), 1–8.

TABLES

Standard 95% Posterior Community-level hyper-parameter Mean Rhat deviation Intervals

μβ0 Mean intercept -2.08 0.29 -2.65 - -1.51 1.01 σβ0 SD of intercept 2.32 0.23 1.91 - 2.79 1.00 μβ1 Elevation (lineal term) -0.51 0.18 -0.84 - -0.17 1.00 σβ1 Elevation (lineal term) 1.46 0.15 1.19 - 1.78 1.01 μβ2 Elevation (quadratic term) -0.22 0.07 -0.36 - -0.09 1.00 σβ2 Elevation (quadratic term) 0.43 0.07 0.31 - 0.57 1.00 μβ3 Patch Size (lineal term) 0.05 0.15 -0.22 - 0.36 1.10 σβ3 Patch Size (lineal term) 0.26 0.14 0.03 - 0.54 1.07 μβ4 Patch Size (quadratic term) 0.03 0.09 -0.15 - 0.2 1.35 σβ4 Patch Size (quadratic term) 0.06 0.04 0 - 0.15 1.01 μβ5 Amount of Area (lineal term) -0.3 0.12 -0.54 - -0.06 1.13 σβ5 Amount of Area (lineal term) 0.72 0.09 0.56 - 0.91 1.00 μβ6 Amount of Area (quadratic term) -0.02 0.09 -0.18 - 0.15 1.34 σβ6 Amount of Area (quadratic term) 0.16 0.08 0.02 - 0.3 1.04 μβ7 Patch Size * Amount of Area -0.17 0.22 -0.55 - 0.22 1.45 σβ7 Patch Size * Amount of Area 0.08 0.06 0 - 0.22 1.01 μβ8 Year 2 0.41 0.14 0.13 - 0.67 1.00 σβ8 Year 2 0.7 0.14 0.43 - 0.99 1.02 μβ9 Year 3 0.39 0.23 -0.04 - 0.84 1.00 σβ9 Year 3 1.46 0.2 1.09 - 1.89 1.00 μα0 Mean intercept -1.88 0.17 -2.22 - -1.56 1.03 σα0 SD of intercept 1.07 0.13 0.83 - 1.36 1.02 μα1 Forest cover -0.22 0.05 -0.33 - -0.11 1.00 σα1 Forest cover 0.38 0.05 0.3 - 0.49 1.00 μα2 Survey time -0.09 0.03 -0.14 - -0.03 1.00 σα2 Survey time 0.15 0.03 0.1 - 0.21 1.00 μα3 Year 2 -0.03 0.11 -0.24 - 0.17 1.00 σα3 Year 2 0.58 0.1 0.4 - 0.8 1.03 μα4 Year 3 -0.32 0.12 -0.56 - -0.1 1.00 σα4 Year 3 0.53 0.12 0.31 - 0.79 1.02

Glacier Landscape Forest Cover Proportion Valleys (ha) (ha) Paron 3468.32 581.22 16.76 Llanganuco 5177.21 555.17 10.72 Ulta 4509.15 926.4 20.54 Llaca 625.34 140.15 22.41 Rajucolta 2578.67 228.47 8.86

FIGURES

Figure 2.1. Expectations for species occupancy given different tolerances to fragmentation (patch size) and habitat loss (amount of forest cover). If Polylepis specialists are adapted to patchy configurations of Polylepis forest, they are expected to be unrelated or less sensitive to patch size reduction (dashed line) (a-b), but either negatively affected by the reduction of amount of forest (solid line) (a) or unaffected by both (b). If not, Polylepis bird species are expected to be equally affected by patch size and forest loss (c). In contrast, species associated with the Puna matrix would positively respond to reductions in patch size and amount of forest cover (d).

Figure 2.2. Map of the five valleys studied within Cordillera Blanca, Peru. A total of 59 patches were included, ranging from 0.01 ha to 200 ha.

Figure 2.3. Examples of Polylepis forest from the glacial valleys of Ulta (a), Rajucolta (b), Llaca (c), Llanganuco (d) and Parón (e) within Cordillera Blanca. Matrix habitats are dominated by Puna grassland (a and e) or shrublands (f). Photos: Steven Sevillano-Ríos

Figure 2.6. Avian community composition across five different categories of patch size and amount of forest cover within the landscape. The community was classified according to (1) habitat associations (a, b), (2) five foraging guilds (c, d), and (3) status of conservation priority (e, f). The total numbers of observations during the three years on each category are in parenthesis over each bar (this applies to the other two graphs below).

Figure 2.7. Mean community occupancy for birds in the High Andes was highest at mid-to-low elevations (a), in large forest patches (b) and in local landscapes with relatively low forest cover (c) while accounting for detectability per year (d,e,f). Shade areas represent the 95% of posterior intervals. Parameters y2 and y3 correspond to second- and third-year effect with respect to the initial year (α0) (f).

Figure 2.8. Species- and community- level responses to elevation (a-b), patch size (c-d) and amount of Polylepis forest cover in the landscape (e-f). Open dots represent the species-specific mean response. Horizontal lines represent the 95% posterior intervals. Intervals that do not overlap zero (blue lines) indicate that estimated coefficients differed significantly from zero, indicating an effect of the parameter. Solid and dashed vertical lines (red) are the mean and 95% PI of the community response, respectively. Species ID correspond those in Table S2.1.

Figure 2.9. Mean species occupancy probabilities in response to Polylepis patch size (left in a,b,c,d) and amount of forest cover (%) within 200-m radio (right in a,b,c,d). The solid lines and shade areas represent the mean ± 95% Posterior Intervals of the reduced MSOM. Cobin = Giant Conebill, Crbar = Baron's Spinetail, Atruf = Rufous-eared Brushfinch, Grand = Stripe-headed Antpitta, Lepil = Rusty-crowned tit-spinetail, Mepho = Black Metaltail, Mialt = Plain-tailed Warbling-Finch, Scaff = Ancash Tapaculo, Xepar = Tit-like Dacnis, Zastr = White-cheeked Cotinga, Ocruf = Rufous-breasted Chat-tyrant, Geser = Striated Earthcreeper. Individual plots of the 88 species are in Figure S2.7.

APPENDICES

Table S2.1. Species-specific occupancy probability response of the 88 bird species evaluated with respect changes on elevation, patch size and amount of forest. Values represent the difference across each gradient (e.g. ψ at lower limit – ψ at upper limit for species i). A positive or negative value indicated the magnitude of increase or decrease on the occupancy probability on the upper limit of the gradient. Species ID and codes correspond to those used in Figures 7,8, supplementary Figure S6. E= Endemic, EN=Endangered, VU=Vulnerable, NT=Near threatened, ζ = moderately specialist, ζ ζ = highly specialists.

IUCN Amount Species Primary Polylepis Patch ID Scientific Name Common Name E Red Elevation of forest Code habitat specialist size List cover 1 Aeand Aeronautes andecolus Andean Swift Forest -0.62 0 0 2 Agcup Aglaeactis cupripennis Shining Sunbeam Forest -0.48 0.33 -0.04 3 Agmon Agriornis montanus Black Billed Shrike Tyrant Shrubs 0.44 -0.01 -0.56 4 Amrub Ampelion rubrocristatus Red Crested Cotinga Forest -0.54 0.02 0 5 Analp Anairetes alpinus Ash-Breasted Tit-Tyrant Forest EN ζζ 0.93 0.02 0.01 6 Anfla Anairetes flavirostris Yellow-Billed Tit-Tyrant Shrubs -0.55 0.01 0 7 Annig Anairetes nigrocristatus Maranon Tit Tyrant Shrubs -0.8 0.05 -0.38 8 Anreg Anairetes reguloides Black-Crested Tit-Tyrant Forest -0.18 -0.22 -0.05 9 Asfla Asthenes flammulata Many-Striped Canastero Forest 0.82 -0.32 -0.46 10 Ashum Asthenes humilis Streak-Throated Canastero Grassland 0.46 -0.21 -0.53 11 Asmod Asthenes modesta Cordilleran Canastero Grassland -0.29 -0.09 0.15 12 Aspud Asthenes pudibunda Canyon Canastero Grassland -0.77 0.21 0.06 13 Atruf Atlapetes rufigenis Rufous-Eared Brush-Finch Forest E NT ζ 0.3 0.14 0.05 14 Boorb Bolborhynchus orbygnesius Andean Parakeet Forest -0.04 0.05 0 15 Caana Catamenia analis Band-Tailed Seedeater Shrubs -0.59 -0.06 -0.1 16 Caino Catamenia inornata Plain-Coloured Seedeater Shrubs -0.26 -0.01 -0.37 17 Choli Chalcostigma olivaceum Olivaceous Thornbill Grassland E 0.36 0.33 -0.08 18 Chsta Chalcostigma stanleyi Blue-Mantled Thornbill Grassland 0.21 -0.31 0.15 19 Cialb Cinclodes albidiventris Chestnut-Winged Cinclodes Grassland -0.04 0.13 -0.53 20 Ciata Cinclodes atacamensis White-Winged Cinclodes Grassland 0.01 -0.14 -0.42 21 Cileu Cinclus leucocephalus White-Capped Dipper Forest -0.03 0.01 -0.06 22 Cobin Conirostrum binghami Giant Conebill Forest NT ζζ 0.83 0.34 0.54

23 Cocin Conirostrum cinereum Cinereous Conebill Shrubs -0.98 -0.3 -0.68 24 Cocor Colibri coruscans Sparkling Violet-Ear Shrubs -0.82 -0.01 -0.01 25 Coiri Coeligena iris Rainbow Starfrontlet Forest -0.5 0.01 0.01 26 Corup Colaptes rupicola Andean Flicker Forest 0.16 -0.07 -0.28 27 Crbar Cranioleuca baroni Baron's Spinetail Forest E ζ 0.22 0.01 0.53 28 Dibru Diglossa brunneiventris Black-Throated Flowerpiercer Shrubs -0.39 -0.13 -0.26 29 Disit Diglossa sittoides Rusty Flowerpiercer Shrubs -0.64 -0.02 0.01 30 Dispe Diuca speculifera White-Winged Diuca-Finch Shrubs 0.09 -0.09 -0.22 31 Elalb Elaenia albiceps White-Crested Elaenia Shrubs -0.35 0 0 32 Geple Geospizopsis plebejus Ash-Breasted Sierra-Finch Grassland -0.68 -0.34 -0.64 33 Geser Geocerthia serrana Striated Earthcreeper Grassland E 0.46 -0.2 -0.32 34 Geten Geositta tenuirostris Slender-Billed Miner Grassland -0.01 -0.01 -0.01 35 Geuni Geospizopsis unicolor Plumbeous Sierra-Finch Grassland 0.51 -0.33 -0.72 36 Grand Grallaria andicolus Stripe-Headed Antpitta Forest ζ 0.92 0.08 -0.14 37 Inper Incaspiza personata Rufous-Backed Inca-Finch Shrubs E -0.29 0 -0.01 38 Leand Leptasthenura andicola Andean Tit-Spinetail Shrubs 0.38 0.45 0.14 39 Lefum Leuconotopicus fumigatus Smoky Brown Woodpecker Forest -0.01 -0.01 -0.02 40 Lenun Lesbia nuna Green-Tailed Trainbearer Shrubs -0.13 -0.15 -0.39 41 Lepil Leptasthenura pileata Rusty-Crowned Tit-Spinetail Shrubs E ζ -0.96 0.13 -0.07 42 Lestr Leptasthenura striata Streaked Tit-Spinetail Shrubs -0.96 -0.03 -0.09 43 Levic Lesbia victoriae Black-Tailed Trainbearer Shrubs -0.09 -0.15 -0.25 44 Mekoe Megascops koepckeae Koepcke's Screech Owl Forest 0.01 -0.01 -0.02 45 Meleu Mecocerculus leucophrys White-Throated Tyrannulet Forest 0.54 -0.3 0.38 46 Memel Metriopelia melanoptera Black-Winged Ground-Dove Shrubs -0.4 0 0 47 Mepho Metallura phoebe Black Metaltail Shrubs E -0.96 -0.09 0.02 48 Metyr Metallura tyrianthina Tyrian Metaltail Shrubs -0.84 0.03 0.27 49 Mialt Microspingus alticola Plain-Tailed Warbling-Finch Forest E EN ζ -0.85 0.19 -0.22 50 Mualb Muscisaxicola albilora White-browed ground tyrant Grassland -0.01 -0.01 -0.02 51 Mucin Muscisaxicola cinereus Cinereous Ground Tyrant Grassland 0.88 -0.15 -0.53 52 Mugri Muscisaxicola griseus Taczanowski's Ground Tyrant Grassland 0.22 -0.1 -0.06 53 Mujun Muscisaxicola juninensis Puna Ground-Tyrant Grassland -0.21 0 -0.13

54 Mumac Muscisaxicola maculirostris Spot-Billed Ground-Tyrant Grassland 0.03 -0.02 -0.13 55 Muruf Muscisaxicola rufivertex Rufous-Naped Ground-Tyrant Shrubs -0.16 -0.05 -0.16 56 Mynig Myiothlypis nigrocristata Black-Crested Warbler Forest -0.99 0.01 0.04 57 Mystr Myiotheretes striaticollis Streak-Throated Bush-Tyrant Shrubs -0.98 -0.01 -0.01 58 Ocleu Ochthoeca leucophrys White-Browed Chat-Tyrant Shrubs -0.17 -0.12 -0.16 59 Ocoen Ochthoeca oenanthoides D'Orbigny's Chat-Tyrant Shrubs 0.2 0.24 -0.51 60 Ocruf Ochthoeca rufipectoralis Rufous-Breasted Chat-Tyrant Forest 0.7 0.36 0.25 61 Orand Orochelidon andecola Andean Swallow Forest 0.01 -0.02 -0.03 62 Orest Oreotrochilus estella Andean Hillstar Shrubs 0.08 0.1 -0.33 63 Ormur Orochelidon murina Brown-Bellied Swallow Forest -0.19 -0.05 -0.2 64 Pafas Patagioenas fasciata Band-Tailed Pigeon Shrubs -0.98 0 0 65 Pagig Patagona gigas Giant Hummingbird Shrubs -0.07 -0.12 -0.04 66 Phpun Phrygilus punensis Peruvian Sierra Finch Shrubs -0.86 -0.04 -0.51 67 Pibon Pipraeidea bonariensis Blue-And-Yellow Tanager Shrubs -0.94 0 0.05 68 Poruf Polioxolmis rufipennis Rufous-Webbed Tyrant Shrubs -0.15 -0.11 -0.31 69 Psaur Psilopsiagon aurifrons Mountain Parakeet Shrubs 0 -0.01 0.04 70 Pycya Pygochelidon cyanoleuca Blue-And-White Swallow Shrubs 0.01 -0.01 -0.01 71 Saaur Saltator aurantiirostris Golden-Billed Saltator Shrubs -0.9 -0.01 0.01 72 Scaff Scytalopus affinis Ancash Tapaculo Forest E ζ 0.8 0.04 0.16 73 Sijel Silvicultrix jelskii Jelski's Chat Tyrant Grassland -0.01 0.06 0.01 74 Sioli Sicalis olivascens Greenish Yellow-Finch Grassland 0 -0.12 -0.28 75 Siuro Sicalis uropygialis Bright-Rumped Yellow-Finch Grassland 0.01 0.01 -0.41 76 Spatr Spinus atratus Black Siskin Shrubs 0.33 -0.05 -0.07 77 Spcra Spinus crassirostris Thick-Billed Siskin Forest 0.68 -0.08 0.06 78 Spmeg Spinus magellanicus Hooded Siskin Shrubs -0.84 -0.03 -0.67 79 Sylon Systellura longirostris Band-Winged Nightjar Shrubs 0.01 -0.01 -0.03 80 Leyan Leptasthenura yanacensis Tawny Tit-Spinetail Forest NT ζ 0.93 -0.01 0.04 81 Thorn Thlypopsis ornata Rufous-Chested Tanager Forest -0.77 0.06 0 82 Traed Troglodytes aedon House Wren Shrubs -0.04 0 -0.07 83 Tuchi Turdus chiguanco Chiguanco Thrush Shrubs/forest -0.47 -0.11 -0.14 84 Tufus Turdus fuscater Great Thrush Shrubs/forest -0.91 -0.17 -0.07

85 Upval Upucerthia validirostris Plain-Breasted Earthcreeper Grassland 0.1 -0.27 -0.47 86 Xepar Xenodacnis parina Tit-Like Dacnis Forest ζ 0.89 0.04 0.11 87 Zastr Zaratornis stresemanni White-Cheeked Cotinga Forest E VU ζζ 0.21 0.2 -0.13 88 Zocap Zonotrichia capensis Rufous-Collared Sparrow Grassland -0.28 -0.14 -0.48

Figure S2.1. Variogram where no evidence of spatial auto-correlation of the amount of forest in the landscape is observed. Variance between closer locations are similar to locations far apart. Survey stations located closer of each other (points at the left) have almost the same amount of variance of point apart (points at the right).

Figure S2.2. Diagram of the species richness estimation for each cumulative species-area curve.

Figure S2.3. High Andean bird community occupancy and detectability in response to elevation (a), Polylepis patch size (b), amount of forest (c), forest cover (d) and survey time (e) across the three years.

Figure S2.4. High Andean bird community distribution of average species occupancy (a) 2 and detection (b) probabilities based on the species intercept model (μβ0, σ β0, μα0, σα0). The red and blue vertical lines are the median and the mean respectively.

Density Density

0 0

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Species occupancy probability Species detection probability

Figure S2.5. Local species richness (alpha diversity) at each site (n=187) during the three years declines with increasing elevation (a), patch size (b) and amount of Polylepis forest within 200 m-radio (c). Open symbols and vertical lines represent the mean and 95% posterior intervals of the species richness estimated through MSOM of each site and year.

Figure S2.6. Species- and community- level responses to the interaction of elevation and patch size (upper left), elevation and amount of Polylepis forest (AF, upper right), and patch size and amount of forest in the landscape (bottom).

Figure S2.7. Species-specific occupancy response of 88 bird species to elevation (a), Polylepis forest patch size (b), amount of forest within 200 m-radio (c) while accounting for species detection probability across forest cover (d) and survey time (e). Species that: decrease = red, increase = blue, peak in between (a) or do not change but have an occupancy > 0.3 (d to e) = black, do not change and have occupancy < 0.3 = grey.

Figure S2.8. Species-specific mean (95% CI in grey lines) occupancy probabilities along elevation (left), patch size (center) and amount of forest (right) for each of the 88 bird species evaluated. Points are sites with at least one detection/non-detection. The occupancy of 32 species decreased monotonically with elevation, including three species of conservation concern: Rusty-crowned tit-spinetail Leptasthenura pileata (endemic and Polylepis asociated), Black Metaltail Metallura phoebe (endemic) and Plain-tailed Warbling-Finch Microspingus alticola (endemic, EN, and Polylepis asociated) whereas other 15 peacked at middle elevation (Figure S7). The occupancy of sixteen species increased either monotonically (n= 14) or quadratically (n=2) with elevation, including 11 of conservation concern: Ash-breasted Tit-Tyrant Anairetes alpinus (EN, Polylepis specialist), Rufous-eared Brushfinch Atlapetes rufigenis (Endemic, NT, Polylepis asociated), Olivaceous Thornbill Chalcostigma olivaceum (endemic), Giant Conebill Conirostrum binghami (NT, Polylepis specialist), Striated Earthcreeper Geocerthia serrana (endemic), Stripe-headed Antpitta Grallaria andicolus (Polylepis asociated), Ancash Tapaculo Scytalopus affinis (endemic, Polylepis asociated), Tawny Tit-Spinetail Leptasthenura yanacensis (NT, Polylepis specialist), Tit-like Dacnis Xenodacnis parina (Polylepis asociated) and the White-cheeked Cotinga Zaratornis stresemanni (endemic, VU, Polylepis specialist) (Table 2; Suplementary Figure S6). Only two species, the Baron's Spinetail Cranioleuca baroni (Endemic, Polylepis asociated), and the House Wren Troglodytes aedon, remain highly constant across the whole elevation. The main species responses to patch size (center) and amount of forest (right) are in the main text.

CHAPTER III

DRIVERS OF AVIAN DIVERSITY ALONG THE WESTERN SLOPE OF THE

CENTRAL ANDES AND IMPLICATIONS FOR CLIMATE CHANGE

C. Steven Sevillano-Ríos and Amanda D. Rodewald

ABSTRACT

Climate change is expected to alter species distributions and, thus, patterns of biodiversity around the world; yet the underlying mechanisms – especially in tropical montane systems – are poorly understood. We evaluated the extent to which avian diversity along elevational gradients was explained by factors related to climate, primary productivity, landscape configuration, and human activities along a section of the western slope of the Central Andes (department of Ancash, Peru), which represents one of the largest elevational gradients in the world (0-6768 masl). Avian species richness (212 species) and endemism (22 species) were highly correlated (r = 0.91) and peaked at mid- upper elevations (3,500 – 4,000 masl). This mid-elevation peak in numbers of species and endemics was most closely associated with high primary productivity and intermediate temperatures and precipitation amounts at mid-elevations (3,300 – 4,300 masl). Land area, the mid-domain effect and human activities received less support. In our statistical model based upon four climate scenarios for 2070, changes in temperature and precipitation will be most severe above 2,000 masl, whereas human activities will intensify and expand below 500 and above 3,000 masl. Our work provides evidence that elevational gradients in species richness along the western slope of the Andes will likely shift due to complex and changing interactions among temperature, precipitation, primary productivity, and human activities.

RESUMEN

Se espera que el cambio climático altere la distribución de especies y, por lo tanto, los patrones de biodiversidad en todo el mundo; sin embargo, los mecanismos subyacentes, especialmente en sistemas tropicales de montaña, son poco conocidos. Evaluamos la medida en que la diversidad aviar a lo largo de los gradientes de elevación es explicada por factores relacionados con el clima, la productividad primaria, la configuración del paisaje y las actividades humanas a lo largo de una sección occidental de los Andes centrales

(departamento de Ancash, Perú), que representa uno de los mayores gradientes de elevación del mundo (0-6768 msnm). La riqueza de especies (212 especies) y el endemismo de aves (22 especies) estuvieron altamente correlacionados (r = 0.91) y alcanzaron su punto máximo en elevaciones medias-altas (3,500 - 4,000 msnm). Este pico del número de especies y endemismo a elevación media se asoció más estrechamente con una alta productividad primaria, temperaturas y cantidades de precipitación intermedias entre los 3,300 - 4,300 msnm. La cantidad de área topográfica, el efecto de dominio intermedio (MDE) y las actividades humanas recibieron menos soporte. En nuestro análisis estadístico, según cuatro escenarios climáticos para el 2070, los cambios en la temperatura y la precipitación serán más severos por encima de los 2.000 msnm, mientras que las actividades humanas se intensificarán y expandirán por debajo de los 500 y por encima de los 3.000 msnm. Nuestro trabajo proporciona evidencia de que los gradientes de elevación en la riqueza de especies a lo largo de la ladera occidental de los Andes probablemente cambiarán debido a las complejas y cambiantes interacciones entre temperatura, precipitación, productividad primaria y actividades humanas.

INTRODUCTION

Climate change is among the most serious threats facing biodiversity around the world, with montane species particularly vulnerable (Şekercioğlu et al., 2008; Şekercioğlu et al.,

2012; Bellard et al., 2012; Freeman et al., 2018). Climate change is expected to alter distributional patterns of biodiversity via changes in population dynamics, behavior, and species interactions (Şekercioğlu et al., 2008; Herzog et al., 2011; Wormworth and

Şekercioğlu 2011), but the underlying mechanisms – especially in montane systems – are poorly understood (Perrigo et al., 2019). Consequently, our ability to predict and mitigate the effects of climate change on biodiversity along elevational gradients remains weak

(Huber et al., 2006; Mawdsley et al., 2009).

Understanding factors that shape species distributions is particularly important in biodiversity hotspots like the Tropical Andes (Myers et al., 2000), where mammals

(Patterson et al., 1996; Patterson et al., 1998; Voss 2003), birds (Terborgh 1977; Kessler et al., 2001; Kattan and Franco 2004; Herzog et al., 2005), plants (Kessler et al., 2001; Küper et al., 2004; Krӧmer et al., 2005; Salazar et al., 2015) and other taxa (Hutter et al., 2013;

Meza-Joya and Torres 2016) regularly exceed records of alpha, beta and gamma diversity

(Terborgh 1971, 1985; Terborgh and Weske 1975; Remsen 1985; Jankowski et al., 2013).

Yet even where diversity is unusually high, the manner in which it changes along elevational gradients varies widely among regions and taxa (McCain and Grytnes 2010).

Species richness may decline with elevation (Terborgh 1977; Patterson et al., 1998), peak mid-slope (Kattan and Franco 2004; Küper et al., 2004; Krӧmer et al., 2005), remain constant, plateau at low elevations before declining at the highest elevations (McCain

2009), or show unexpected patterns (Herzog et al., 2005). As such, no single factor (e.g.

96 climatic, spatial, biotic or historical) satisfactorily explains the spatial patterning of diversity in the Tropical Andes (Grytnes and McCain 2007; Rahbek et al., 2019).

In this study, we examine avian diversity and endemism in a poorly known section of the western slope of the Central Andes (department of Ancash, Peru), which represents one of the largest elevational gradients in the world (0-6768 masl). The region has been occupied by humans for over 5,000 years (e.g. Caral and Chavín) and is currently home to over 1,200,000 adults (INEI 2017). In contrast to most previous research on elevational gradients and biodiversity that focused on the eastern slope of the Andes (Terborgh 1971,

1977, 1985; Terborgh and Weske 1975; Remsen 1985; Patterson et al., 1996; Patterson et al., 1998; Kessler et al., 2001; Kattan and Franco 2004; Küper et al., 2004; Krӧmer et al.,

2005; Herzog et al., 2005; Jankowski et al., 2013; Hutter et al., 2013; Salazar et al., 2015;

Meza-Loja and Torres 2016), ours is the first study, to our knowledge, to focus specifically on the western slope of the Central Andes.

We evaluated the extent to which relationships between elevation and avian diversity were explained by factors related to climate (e.g. temperature, precipitation, climatic stability), primary productivity, landscape configuration (e.g. land area and spatial constraints) and human activities. The following hypotheses were explicitly tested:

Climatic hypotheses: (1) The temperature hypothesis proposes that species richness declines with elevation due to increasing energetic needs as temperatures decrease monotonically with increasing elevation (McCain and Grytnes 2010). This is based partly upon the metabolic theory of ecology (Brown et al., 2004), the metabolic theory of biodiversity (Allen and Gillooly 2006), and the species-energy (thermal) theory (Clarke and

Gaston 2006) where mutation and speciation rates increase with temperature (Allen et al.,

2006). 97

(2) The temperature and water hypothesis proposes that species richness will be greatest where temperature and precipitation best accommodate physiological tolerances of species and provide adequate resources (Hawkins et al., 2003; Evans et al., 2005; McCain

2007; McCain and Grytnes 2010). Although temperature and water availability can interact in unexpected ways (Hawkins et al., 2003; Evans et al., 2005; McCain 2007), the elevational climate model (ECM) predicts that warm and wet conditions (local climate optima) promote species richness, though the geographic and topographic correlates of these conditions vary regionally and among mountain systems (McCain 2007; McCain

2009).

(3) The climatic stability hypothesis posits that stable, or predictable, climates generate and maintain higher diversity (Connell and Orias 1964; Janzen 1967) by promoting niche partitioning, specialization, and speciation (Hawkins et al., 2003; Evans et al., 2005). In this case, species richness should peak at lower elevations because stability usually declines with elevation, as it does with latitude (Körner 2000).

Primary productivity hypothesis proposes that the net amount of energy in a system that is stored in the biomass of primary producers can limit species richness by constraining resources available to higher trophic levels (Connell and Orias 1964; Waide et al., 1999; Mittelback et al., 2001; Hawkins et al., 2003; Evans et al., 2005). As resources become more available, increased niche specialization and reduced mobility (Connell and

Orias 1964) can promote species richness and endemism. Although primary productivity generally declines with increasing elevation (Rahbek 1997; Nogués-Bravo et al., 2008), patterns are highly variable among mountains (McCain and Grytnes 2010; Toszogyova and

Storch 2019).

Landscape hypotheses. (1) The land area hypothesis, based on the species-area relationship, proposes that species richness increases with land area due to associated changes in speciation and extinction rates or habitat heterogeneity (Terborgh 1973;

Rosenzweig 1992, 1995). As such, the amount of land area present at different elevations might drive patterns in richness (Elsen and Tingley 2015).

(2) The spatial constraints hypothesis (e.g. mid-domain-effect) states that spatial boundaries, including bases and peaks of mountains, constrain species ranges in ways that can promote overlap of species in the middle of the domain (e.g. elevational gradient) with comparatively fewer species at the extremes (Colwell and Lees 2000; Colwell et al., 2004).

Because the mid-domain-effect (MDE) requires no biological forces, many authors evoke it as a null model of elevational biodiversity (Colwell et al., 2004) and it has been supported in many altitudinal (Kessler 2001; Sanders 2002; McCain 2004; Cardelús et al., 2006) and latitudinal studies (Jetz and Rahbek 2001, 2002; McCain 2003; Connolly et al., 2003).

Anthropogenic impacts hypothesis: This hypothesis proposes that humans have modified the ‘natural’ biodiversity patterns along elevations such that diversity is generally depressed in areas with long history of human occupation or intensive activities (Nogués-

Bravo et al., 2008).

METHODS

Study area

In order to include the full elevational gradient of the western Andes (0 – 6,768 masl), we selected a 220 x 150 km section of the Cordillera Occidental located parallel to the Pacific coast in the department of Ancash, Peru (-9.34°N, -77.39°E) (Figure 3.1). This area is dominated by eight ecosystems including coastal desert, riparian dry forest, xerophytic

99 coastal shrublands, Puna grassland, High Andean forests, shrublands, wetlands and glaciers in the Andes (<2,000 m) (Ministerio del Ambiente del Perú 2015). The Andes in this region is comprised of two sub-mountain ranges running parallel to the coast for about 180 km.

Cordillera Negra is an ice-free mountain range dominated by a mixture of intermontane shrubby forests and Puna grasslands, whereas Cordillera Blanca is a snowcapped range with glaciers extending from 5,000-6,768 masl (Ames et al., 1989) and some of the highest concentrations of Polylepis forest (Zutta 2012; Sevillano-Ríos and Rodewald 2017).

Currently, Cordillera Blanca is protected by Huascaran National Park (HNP) and the

Huascaran Biosphere Reserve (HBR), which is recognized as a world heritage site by

UNESCO (PROFONANPE - HNP Master Plan 2010). Both cordilleras include a mix of human-modified areas, such as cities, towns, crop fields and exotic forest plantations

(Ministerio del Ambiente 2015).

Estimating avian species richness

Similar to previous studies, we used bird elevation ranges to estimate the number of species expected to occur at discreate elevational bands (gamma diversity) along the elevation gradient (Kattan and Franco 2008; McCain 2009; Antonelli et al., 2018; Quintero and Jetz,

2018). We compiled elevation ranges of 239 bird species reported to occur within our (220 x 150 km) study area according to Birds of Peru (Schulenberg et al., 2010) and elsewhere

(Fjeldså 1987; Frimer and Nielsen 1989; Fjeldså and Krabbe 1990; Barrio 2002; Servat

2006; Sevillano-Ríos et al. 2011; Sevillano-Ríos 2016; Sevillano-Ríos and Rodewald 2017;

Sevillano-Ríos 2017). For each species, we recorded elevation range (min-max in meters), migratory behavior, and endemic status in Peru. In cases where information was unclear or missing, we referred to the Neotropical Birds web page

100

(https://neotropical.birds.cornell.edu/), the IUCN Red List of Threatened Species web page

(https://www.iucnredlist.org/), eBird (https://ebird.org/) records, or the original sources of information. Finally, we complemented this information with 13,780 direct observations of

100 bird species occurring between 3,300 ̶ 4,700 masl. After excluding introduced, vagrant, non-resident, pelagic (e.g. petrels), or widespread boreal/austral migratory species, we estimated species richness within 200-m elevation bands from a list of 212 species, including 22 endemics (Table S3.1), using the “RangeModel v. 5.0” software (Colwell,

2006; 2008) (See DME section).

Explanatory variables

Within our 220 x 150-km study area, we randomly placed 1,000 points along the elevational gradient in rough proportion to the relative amount of land area (Figure 3.1).

For each point, we recorded latitude, longitude, and mean elevation (m) within a 1-km radius plot from the Space Shuttle Radar Topography Mission - Digital Elevation Model

(SRTM, 30-m resolution), downloaded from the U. S. Geological Survey (USGS). These values were then averaged within each 200-m elevation band.

Climatic data

Seven of 19 classic bioclimatic variables were sourced from the most recent version of

WorldClim 2 (Fick and Hijmans 2017), which were derived from spatially-interpolated monthly climate data at high resolution (~1 km2 at the equator) recorded between 1970 to

2000 (Hijmans et al., 2005; Fick and Hijmans 2017). Annual mean temperature and precipitation (BIO1 and BIO12) were used as proxies of the total amount of energy and water annually available (O’Donnell and Ignizio 2012). We also considered temperature and precipitation seasonality (BIO4 and BIO15), which was expressed as the standard

101 deviation of the 12 mean monthly values multiplied by 100 (Hijmans et al., 2005).

Variability in mean diurnal and annual temperature ranges (BIO2 and BIO7) and isothermality (BIO3 = BIO2/BIO7*100), interpreted as the degree of oscillation of daily temperatures relative to the annual temperature range (O’Donnell and Ignizio 2012) were also included. For each sample point, we extracted the mean, maximum and minimum values of the seven climatic variables within 1-km radius and averaged within each 200-m elevation band.

Primary productivity

We used the Normalized Difference Vegetation Index (NDVI) as a proxy of the total amount of energy available in an ecosystem, usually related to vegetative cover and biomass (Rouse et al., 1974; Waide et al., 1999; Bailey et al., 2004; Pettorelli et al., 2005;

We et al., 2017; Lizaga et al., 2019). Because NDVI changes seasonally, we generated measures for the dry and wet seasons of 2015 and 2016 respectively by using the formula:

(퐵푎푛푑 5−퐵푎푛푑 4) 푁퐷푉퐼 = (1) (퐵푎푛푑 5+퐵푎푛푑 4) where Band 5 and Band 4 correspond to each of the four Landsat8 (L8 OLI/TIRS) extents of 30 m of resolution (Table S3.2), downloaded and preprocessed through the Semi-

Automatic Classification Plugin (Congedo 2016) in QGIS (QGIS Development Team,

2018). Although the amount of cloud cover varied between 8 to 42% for each extent, our area of interest was clean of clouds and no extra process was required. We extracted nine

NDVI variables, including the main, maximum and minimum NDVI values for the dry and wet seasons as well as an average of both seasons within a 1-km radius. We then averaged these values for each 200-m elevation band.

102

Land Area

We calculated the amount of land area within 100-m elevational bands based on SRTM –

DEM (U. S. Geological Survey) in QGIS (Version 3.0-Girona).

Mid domain effect (MDE)

We modeled the number of species expected to occur along each elevation band under geometric constraints (e.g., the Pacific Ocean and the glacier base line at 0 and 5,000 m respectively) using the “RangeModel v. 5.0” (Colwell 2006). This software uses the Jetz and Rahbek Spreading Dye simulation algorithm that accommodates overlapping species distributions along boundaries (Jetz and Rahbek 2001; Colwell 2006). Elevational ranges of birds (Table S3.1) were randomized 5,000 times (Monte Carlo simulations without replacement) and results (mean, 95% CI) for each 200-m elevational band exported for further analysis.

Human impact

We used the Human Footprint Index (Venter et al., 2016), a high-resolution (~1 km2) measure of the overall impact of human activities on the environment between 1993 and

2009. This global index is comprised of eight variables (built environments, population density, artificial night lighting, crop lands, pasture lands, roads, railways and navigable waters). For each point, we extracted the mean, maximum and minimum value within a 1- km radius and then averaged them for each 200-m elevation band.

Bird-environmental relationships

Data were analyzed in two ways. First, we conducted a raw analysis where we used a negative binomial distribution to model the relationship of avian diversity and the raw

103 climatic-related, spatial and human data extracted from the 1,000 random points. Models using such negative binomial distributions incorporate an extra parameter (theta) to adjust the variance of over-dispersed count data (Ver Hoef and Boveng, 2007). The minimum and maximum values of each of the seven climatic variables were excluded from the analysis as they were highly correlated with the mean (Spearman’s rho > 0.98, p <0.0001). All nine metrics of primary productivity (mean, minimum, maximum for the dry/wet and average seasons) were included as significant differences were observed among them (Kluskal-

Wallis test: chi2=5096, p<0.0001). Together with the land area, MDE and the Human

Footprint Index, a total of 19 explanatory covariates were included in analyses (Figure

S3.1).

Because our study design was subject to pseudo-replication (i.e., multiple points occurred within each 200-m elevational band), we performed a second analysis in which point values were grouped by calculating the mean covariate values for each 200-m elevation band using Standard Least Squares Models. We tested main and quadratic effects for each covariate, as well as additive and interactive effects of the factors identified as most relevant in our individual covariate analysis. Model performance was assessed through the

Akaike Information Criteria (AIC) and the Generalized R squared (likelihood-ratio-based

R2, Maddala 1983, Cox and Snell 1989, Magee 1990). We used the “MASS” (v 1.1)

(Venables and Ripley 2002), “MuMIn” (v 1.15.6) (Barton 2016), “rsq” (v 1.1) (Zhang

2018), and “corrplot” (v 0.84) (Taiyun and Viliam 2017) packages of R (Version 3.5.3) to perform all the analysis.

104

Future climatic conditions and human impact

We forecast future changes in climate along the elevational gradient based upon climatic data from the four Representative Concentration Pathways (RCPs) (RCP2.6, RCP4.5,

RCP6.0, RCP8.5) adopted by the Intergovernmental Panel on Climate Change (IPCC) in their Fifth Assessment Report (Allen et al., 2014). Each RCP represents different scenarios of greenhouse gas concentration, with RCP8.5 considered the most severe (Riahi et al.,

2011). For each survey point, we extracted the same 7 climatic variables (BIO1, BIO2,

BIO3, BIO4, BIO07, BIO12, BIO15) from the Hadley Global Environment Model 2 for

Earth System (HadGEM2-ES, ~1-km resolution) for the year 2070. We used LOESS models to characterize each climatic covariate along the elevation and used the difference between current (WorldClim 2) and future (HadGEM2-ES) models to estimate the expected degree of change. We used the best model coefficients obtained in the previous analysis to predict future species richness responses under the four climate scenarios.

Changes in species richness along elevation were also assessed through LOESS models.

Finally, as future human impact models are not available for our study area, we used the degree of change of Human Footprint Index (Venter et al., 2016) between 1993-2009 as proxy of the current trend in human impact along the elevational gradient.

RESULTS

Changes along the elevational gradient

Species richness peaked at middle-upper elevations (3,500 – 4,000 masl) and was highly correlated (r = 0.91) with the number of endemic species (Figure 3.2). Species richness remained relatively stable at ~60 species between 0-1,500 m, increasing to ~120 species at

105

~3,500 m and then decreasing to fewer than 10 species above 5,000 m. The proportion of endemic species, which was roughly 1:10, was similar throughout the elevation gradient.

Climatic variables showed inconsistent and sometimes contrasting relationships with elevation. Whereas annual mean temperature decreased linearly with elevation (BIO1

= -8E-07*Elev2 - 0.0003*Elev + 20.332, R2=0.98, p<0.001), the annual mean precipitation increased more variably, especially at mid-elevations (BIO12 = 1E-05*Elev2 + 0.1238*Elev

- 15.936, R² = 0.8649, p<0.001) (Figures 3.3a-b). Likewise, elevations between ~2,000 –

4,000 m had more stable temperatures but greater seasonality in precipitation (Figures 3.3d- e). Across our system, daily temperature fluctuations within a single day (Figure 3.3g) were typically as wide as annual fluctuations (Figure 3.3h), resulting in high (~ 80) and stable isothermality in areas above 2,000 masl (Figure 3.3i). Primary productivity (NDVI) peaked between 3,500 – 4,000 masl in each season (Figure 3.3c) and tended to be greater during wet than dry seasons (wet season mean = 0.30, mean difference = 0.083, 95% CI

0.08 – 0.09, t-test=-22.637, p<0.00001) (Figure 3.3f). Low elevations along the coast tended to have the greatest land area, with another peak in area occurring at 4,300 m

(Figure 3.3j). The species richness pattern under the mid domain effect (MDE) model peaked between 1,500 to 3,500 masl (Figure 3.3k). Finally, the Human Footprint Index

(HFI) of 2009 also peaked along the coast (e.g. Chimbote) and again around 3,000 masl where most of the towns of Callejon de Huaylas are located (Figure 3.3l).

Species-environmental relationships

Climate and primary productivity best explained variation in species richness along the western slope in both raw and grouped analyses (Table 3.1; Figure 3.4; Table S3.3, Figure

S3.3). Species richness was greatest at intermediate levels of annual temperatures (~ 8 oC;

106

Figure 3.4a, d) and precipitation (~ 680 mm; Figure 3.4b, e) and was positively related to primary productivity (Figure 3.4c, f). In particular, we found strong support for two interactive models containing primary productivity and temperature ([Temp2 * NDVI2] and

[Temp2 * Prec2 * NDVI2]) (ΔAIC= 0.26; Tables 3.2 and 3.3; Table S3.4) such that the importance of temperature and precipitation rose as NDVI declined (Figure 3.5a, b, c). One challenge with our analyses was collinearity between temperature and primary productivity

(VIF: Temp2 * NDVI2=5.08; Temp2 * Prec2 * NDVI2 = 398.54), though not between

2 2 2 temperature and precipitation (Temp *Prec , R = 0.94; F = 89.885,22, p-value <0.00001,

VIF=2.43). Hence, these results suggest that avian species richness may primarily respond to primary productivity, which is driven by the interaction of temperature and precipitation

(Figure 3.6).

Like in previous studies (McCain 2007; 2009), the Area and MDE hypotheses received less support in our analysis than other models, despite significant main effects

(Generalized R2 = 0.59; 0.65 respectively; p<0.0001, Figure 3.7). The Human Footprint

Index (HFI, R2=0.33, p=0.0009) received relatively little support in our model selection approach (Figure 3.7, Table 3.1; Table S3.3).

Vulnerability to land use and climate change

Both land use and climate are expected to change markedly in our system over the next 50 years (Figure 3.8, Figure 3.9). Human activities, as measured by the Human Footprint

Index, are expected to intensify and expand, especially at elevations below 500 and above

3,000 masl (Figure 3.8). Based upon four climate scenarios for 2070, changes in temperature and precipitation will be greatest above 2,000 masl, with the RCP8.5 forecasting the most severe increases in temperature (3.55 (SD= 0.32) oC) and precipitation

107

(107 (SD=44.48) mm), whereas conditions will likely become drier and warmer below

2,000 masl (Figure 3.9). Based on future temperature and precipitation changes, our quadratic model (Temp2 * Prec2) suggested that as climates become more severe, species richness should decline at elevations below 3,000 masl, increasing richness at mid-upper elevations (3,000 – 4,200 masl) but also causing mountaintop extinction at highest elevation (>4,200 masl) (Figure 3.10; Figure S3.4).

DISCUSSION

We evaluated the extent to which elevation gradients in avian diversity and/or endemism were be explained by climatic-related factors (e.g. temperature, precipitation, primary productivity, climatic stability), landscape configuration (e.g. land area and spatial constraints) and human activities in the western slope of the central Andes. Avian species richness and endemism peaked at mid-elevations (3,300 – 4,300 masl, Figure 3.2), where approximately 60% of the species in the entire study area occurred. Similar “hump-shaped” patterns in diversity have been reported for xeric mountains (McCain 2007; 2009) and across taxa including mammals (McCain 2004, 2007; Hu et al., 2014), birds (Kessler et al.,

2001; Pan et al., 2016; Araneda et al., 2018; Quintero and Jetz 2018), reptiles (McCain

2010), and plants (Kessler 2000; Kessler et al., 2001; Kessler 2011; Godoy-Bürki et al.,

2017). The mid-elevation peak in avian species richness along the western slope of the

Central Andes most likely results from high primary productivity and, indirectly, warm temperatures and high precipitation at mid-elevations (3,300 – 4,300 masl), which are known to promote high diversity (McCain 2007; McCain 2009). More proximately, the mid-elevation peak in richness might reflect overlapping distributions of endemic, globally threatened, or highly specialized species with small ranges at middle-upper elevations (e.g.

108

Polylepis forest; Sevillano-Ríos and Rodewald 2017), in contrast to the common or widely distributed species that tended to dominate lower elevations. The confluence of high diversity, endemism, and species of conservation concern underscores the importance of mid-elevations for avian biodiversity along the western slope of the central Andes.

Climatic, spatial and human-related drivers

Temperature, primary productivity, and precipitation explained 90%, 88%, and 78%, respectively, of the variation in species richness along the elevational gradient we studied

(Tables 3.1, Table 3.2). This finding is consistent with previous research demonstrating that annual mean temperature and precipitation are important drivers of primary productivity in temperate (Wang et al., 2003), tropical (Funk 2006; Gillman et al., 2015), marine

(Behrenfeld et al., 2006) and global systems (Wu et al., 2011; Del Grosso et al., 2008; Chu et al., 2016). Likewise, previous studies from the Himalayas (Acharya et al., 2011; Pan et al., 2016) and Alps (Dainese and Poldini 2012) concluded that mid-elevation peaks in avian diversity were related to vegetation productivity and diversity as well as intermediate climatic conditions. In other parts of the dry Andes, such as in northern Argentina and

Chile, avian richness was related to intermediate levels of temperature and water availability (Godoy-Bürki et al., 2017) and higher levels of vegetation heterogeneity

(Araneda et al., 2018). The consistency among regions suggests that the interaction of temperature and precipitation indirectly drive avian species richness in dry mountain systems by determining plant productivity and, hence, the energy available to consumers

(Toszogyova and Storch 2019).

Consistent with other studies, we found that locally warmer and wetter elevations were generally most productive and supported the greatest number of species and endemics

109

(Connell and Orias 1964; Janzen 1967; Currie 1991; Waide et al., 1999; McCain 2007,

2009). Species richness peaked at intermediate temperatures (i.e., ~7 to 13 oC degrees;

Figure 3.4), contradicting the negative linear relationship predicted by the metabolic theory of ecology (Brown et al., 2004). Likewise, species richness also peaked at intermediate levels of precipitation, between ~250 – 750 mm. We suggest that mid-elevations (3,300 –

4,300 masl), may provide the most suitable combination of climatic conditions, with adequate temperatures and substantial water from surface or groundwater sources (e.g., lakes, peatbogs, and glaciers) (Chevallier et al., 2011). Hence, when sufficient water is available in locales with fairly moderate temperatures, one should expect high primary productivity and, therefore, avian diversity and endemism (Encalada et al., 2019; Rahbek et al., 2019).

We found limited evidence that landscape attributes or human activities drove diversity patterns. Though species richness peaked at elevations with the greatest amount of land area (~4,300 masl; Figure 3.7b), we had weaker statistical support than either land area or spatial constraints were responsible for changes in diversity. The MDE, in particular, has been widely criticized for failing to recognize that species ranges and biological diversity cannot exist in absence of environmental gradients (Hawkins and Diniz-Filho 2002; Pimm and Brown 2004; Hawkins et al., 2005). However, the degree of correspondence between species richness and land area (Figures 3.3j; Figure 3.7) support recent studies (Rahbek et al., 2019; Rahbek et al., 2019) and suggests that further analysis is required to fully understand species-area relationships in this part of the Andes.

110

Vulnerability to climate change and human activities

Threats from climate and human activities are expected to rise under future scenarios and may eventually drive some species to extinction (Şekercioğlu et al., 2008, Møller et al.,

2010; Bellard et al., 2012). Species restricted to mountain ecosystems should be most vulnerable (Wormworth and Şekercioğlu 2011; Şekercioğlu et al., 2012). In our system, species that inhabit over 2,000 masl will be facing the greatest changes in mean annual temperature and precipitation by 2070, but lowland species may also be vulnerable to warmer and dryer conditions (Khaliq et al., 2014; Brodie et al., 2017).

Like species elsewhere in the world, Andean birds can theoretically respond to climate change by migrating or dispersing to suitable habitat (Parmesan and Yohe 2003; Parmesan

2006), adapting in situ, or going extinct (Hannah 2015; Habary et al., 2017). However, their life history traits suggest limited capacity to migrate or adapt in response to climate change.

In particular, many Andean birds are adapted to live within very narrow climatic ranges, have low dispersal capabilities, and occur in small populations (Polato et al., 2018). These attributes may make movement or in situ adaptation of such species slow, increasing their risk of extinction (Şekercioğlu et al., 2008; Lawler et al., 2009; Şekercioğlu et al., 2012).

As such, we expect to see lowland attrition of species, upslope range shifts, mountain top extinctions (Figure 3.10), and possibly downslope shifts for a few species closely tied to precipitation (Tingley et al., 2012). These predictions are consistent with reports from the eastern Andean, where increasing temperatures already have prompted shifts in species distributions (Forero-Medina et al., 2011; Tingley et al., 2012, Freeman et al. 2018).

While our models suggest that climate and productivity will facilitate the persistence of more species at upper than lower elevations, ecological disruptions or mismatches have the potential to hasten or exacerbate extinctions (Colwell et al., 2008; Rehm and Freeley 2016). 111

Differences in dispersal ability or mobility between plant and animal species might lead to situations where habitats shift more slowly than species and, thereby, constrain their ability to track climate changes (Rehm and Feeley 2016) and decouple ecological relationships in ways that drive extinctions (Visser and Both 2005). Distributional shifts of species also have the potential to alter species interactions, including mutualism, competition, and predation, as has been reported in Cerro Pantiacolla, Peru (Freeman et al., 2018). For instance, upslope dispersal of a competitive, generalist bird might displace subordinate species that are currently well-adapted to higher elevations. Moreover, certain interactions, like competition, may become especially intense if species are concentrated into sites above

4,200 m, which collectively have less land area (Figure 3.10).

In addition to climate change, the western slope of the Central Andes is likely to face intensifying human pressures, especially at mid-elevations where the Human Footprint

Index increased between 1993-2009 (Figure 3.8). If the road construction, mining development and population growth that we observed over the course of our study

(especially in Callejon de Huaylas and Conchucos; 3,300 – 4,300 masl) continues, many species stand to be negatively impacted. Human activities can interact with climatic drivers in ways that dramatically shape patterns of species diversity and ecosystem function along elevational gradients (Oliver and Morecroft 2014). Overall, our research shows that elevational gradients in species richness are likely to be affected by complex and changing interactions among temperature, precipitation, primary productivity, and human activities.

ACKNOWLEDGMENTS

This research was funded by FONDECYT-CONCYTEC (Nº 237-2015-FONDECYT).

We are also grateful to the Fulbright Scholarship and support from Cornell University that

112 supported CSSR during his graduate studies. We also deeply appreciate financial support provided by Cornell Lab of Ornithology, the Athena Grant, the Bergstrom Award of the

Association of Field Ornithologists as well as the American Alpine Club Research Grants

2017. We thank to André Dhondt, Stephen Morreale, Viviana Ruíz-Gutiérrez, Wesley M.

Hochachka, Lynn Marie Johnson, Sharon-Rose Alonzo and Laura Morales for comments, recommendations and guidance that helped to improve this manuscript.

113

REFERENCES

Acharya, B. K., Chettri, B., & Vijayan, L. (2011). Distribution pattern of trees along an

elevation gradient of Eastern Himalaya, India. Acta Oecologica, 37(4), 329-336.

Allen, A. P., & Gillooly, J. F. (2006). Assessing latitudinal gradients in speciation rates and

biodiversity at the global scale. Ecology letters, 9(8), 947-954.

Allen, A. P., Gillooly, J. F., Savage, V. M., & Brown, J. H. (2006). Kinetic effects of

temperature on rates of genetic divergence and speciation. Proceedings of the

National Academy of Sciences, 103(24), 9130-9135.

Allen, M. R., Barros, V. R., Broome, J., Cramer, W., Christ, R., Church, J. A., ... &

Edenhofer, O. (2014). IPCC fifth assessment synthesis report-climate change 2014

synthesis report.

Antonelli, A., Kissling, W. D., Flantua, S. G. A., Bermúdez, M. A., Mulch, A., Muellner-

riehl, A. N., … & Hoorn, C. (2018). Nature Geoscience.

https://doi.org/10.1038/s41561-018-0236-z

Ames A S., Muñoz, G., Verastegui J., Vigil R., Zamora M., and Zapata M. (1989) Glacier

inventory of Peru. Consejo Nacional de Ciencia y Tecnología, Perú.

Araneda, P., Sielfeld, W., Bonacic, C., & Ibarra, J. T. (2018). Bird diversity along

elevational gradients in the Dry Tropical Andes of northern Chile: The potential

role of Aymara indigenous traditional agriculture. PloS one, 13(12), e0207544.

Bailey, S. A., Horner‐Devine, M. C., Luck, G., Moore, L. A., Carney, K. M., Anderson, S.,

... & Fleishman, E. (2004). Primary productivity and species richness: relationships

among functional guilds, residency groups and vagility classes at multiple spatial

scales. Ecography, 27(2), 207-217.

114

Barrio J (2002) Evaluación de la avifauna de los bosques de Polylepis del Parque Nacional

Huascarán. Informe al PNH.

Barry G. (2008) Mountain weather and climate. Cambridge, UK: Cambridge University

Press.

Barton K. (2016). MuMIn: Multi-Model Inference. R package version 1.15.6.

https://CRAN.R-project.org/package=MuMIn

Behrenfeld, M. J., O’Malley, R. T., Siegel, D. A., McClain, C. R., Sarmiento, J. L.,

Feldman, G. C., ... & Boss, E. S. (2006). Climate-driven trends in contemporary

ocean productivity. Nature, 444(7120), 752.

Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., & Courchamp, F. (2012). Impacts

of climate change on the future of biodiversity. Ecology letters, 15(4), 365-377.

Brodie, J. F., Strimas-Mackey, M., Mohd-Azlan, J., Granados, A., Bernard, H., Giordano,

A. J., & Helmy, O. E. (2017). Lowland biotic attrition revisited: body size and

variation among climate change ‘winners’ and ‘losers’. Proceedings of the Royal

Society B: Biological Sciences, 284(1847), 20162335.

Brown J. H., Gillooly J. J., Allen A. P., et al. (2004) Toward a metabolic theory of ecology.

Ecology 85:1771-1789.

Cardelús, C. L., Colwell, R. K., & Watkins Jr, J. E. (2006). Vascular epiphyte distribution

patterns: explaining the mid‐elevation richness peak. Journal of Ecology, 94(1),

144-156.

Chen, I. C., Hill, J. K., Ohlemüller, R., Roy, D. B., & Thomas, C. D. (2011). Rapid range

shifts of species associated with high levels of climate warming. Science,

333(6045), 1024-1026.

115

Chu, C., Bartlett, M., Wang, Y., He, F., Weiner, J., Chave, J., & Sack, L. (2016). Does

climate directly influence NPP globally?. Global Change Biology, 22(1), 12-24.

Clarke, A., & Gaston, K. J. (2006). Climate, energy and diversity. Proceedings of the Royal

Society B: Biological Sciences, 273(1599), 2257-2266.

Colwell R. K. Rahbek C., Nicholas J. G. (2004) The Mid‐Domain Effect and Species

Richness Patterns: What Have We Learned So Far?. The American Naturalist.

1633: E1-E23.

Colwell, R. K. (2006). RangeModel: a Monte Carlo simulation tool for assessing geometric

constraints on species richness. Version, 5, 10.

Colwell, R. K. (2008). RangeModel: tools for exploring and assessing geometric

constraints on species richness (the mid‐domain effect) along transects. Ecography,

31(1), 4-7.

Colwell, R. K., & Lees, D. C. (2000). The mid-domain effect: geometric constraints on the

geography of species richness. Trends in ecology & evolution, 15(2), 70-76.

Colwell, R. K., Brehm, G., Cardelús, C. L., Gilman, A. C., & Longino, J. T. (2008). Global

warming, elevational range shifts, and lowland biotic attrition in the wet tropics.

Science, 322(5899), 258-261.

Congedo L. (2016). Semi-Automatic Classification Plugin Documentation. DOI:

http://dx.doi.org/10.13140/RG.2.2.29474.02242/1

Connell, J. H., & Orias, E. (1964). The ecological regulation of species diversity. The

American Naturalist, 98(903), 399-414.

Connolly, S. R., Bellwood, D. R., & Hughes, T. P. (2003). Indo‐Pacific biodiversity of

coral reefs: deviations from a mid‐domain model. Ecology, 84(8), 2178-2190.

116

Cox, D. R. and Snell, E. J. (1989) The Analysis of Binary Data, 2nd ed. London: Chapman

and Hall.

Currie, D. J. (1991). Energy and large-scale patterns of animal-and plant-species richness.

The American Naturalist, 137(1), 27-49.

Del Grosso, S., Parton, W., Stohlgren, T., Zheng, D., Bachelet, D., Prince, S., ... & Olson,

R. (2008). Global potential net primary production predicted from vegetation class,

precipitation, and temperature. Ecology, 89(8), 2117-2126.

Encalada, A. C., Flecker, A. S., Poff, N. L., Suárez, E., Herrera-R, G. A., Ríos-Touma, B.,

... & Anderson, E. P. (2019). A global perspective on tropical montane

rivers. Science, 365(6458), 1124-1129.

Elsen, P. R., & Tingley, M. W. (2015). Global mountain topography and the fate of

montane species under climate change. Nature Climate Change, 5(8), 772.

Evans, K. L., Warren P. H., and Gaston, K. J. (2005). Species-energy relationship at the

macroecological scale: a review of the mechanisms. Biological Reviews, 80:1-25.

Feeley, K. J., Silman, M. R., Bush, M. B., Farfan, W., Cabrera, K. G., Malhi, Y., ... &

Saatchi, S. (2011). Upslope migration of Andean trees. Journal of Biogeography,

38(4), 783-791.

Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: new 1‐km spatial resolution climate

surfaces for global land areas. International journal of climatology, 37(12), 4302-

4315.

Fjeldså J (1987) Birds of Relict Forests in the High Andes of Peru and Bolivia: Technical

Report from the Polylepis - Forest Expedition of the Zoological Museum 1987 with

Some Preliminary Suggestions for Habitat Preservation. Zoological Museum

University of Copenhagen 117

Fjeldså J and Krabbe NK (1990) Birds of the high Andes. Zoological Museum University

of Copenhagen and Apollo Books. Copenhagen Denmark

Forero-Medina, G., Terborgh, J., Socolar, S. J., & Pimm, S. L. (2011). Elevational ranges

of birds on a tropical montane gradient lag behind warming temperatures. PLoS

One, 6(12), e28535.

Freeman, B. G., Scholer, M. N., Ruíz-Gutierrez, V., & Fitzpatrick, J. W. (2018). Climate

change causes upslope shifts and mountaintop extirpations in a tropical bird

community. Proceedings of the National Academy of Sciences, 115(47), 11982-

11987.

Frimer O, Nielsen SM (1989) The status of Polylepis forests and their avifauna in

Cordillera Blanca Peru: technical report from an inventory in 1988 with suggestions

for conservation management. Zoological Museum University of Copenhagen

Funk, C. C., & Brown, M. E. (2006). Intra-seasonal NDVI change projections in semi-arid

Africa. Remote Sensing of Environment, 101(2), 249-256.

Gillman, L. N., Wright, S. D., Cusens, J., McBride, P. D., Malhi, Y., & Whittaker, R. J.

(2015). Latitude, productivity and species richness. Global Ecology and

Biogeography, 24(1), 107-117.

Godoy-Bürki, A. C., Biganzoli, F., Sajama, J. M., Ortega-Baes, P., & Aagesen, L. (2017).

Tropical high Andean drylands: species diversity and its environmental

determinants in the Central Andes. Biodiversity and Conservation, 26(6), 1257-

1273.

Grytnes, J. A., & McCain, C. M. (2007). Elevational trends in biodiversity. Encyclopedia of

biodiversity, 5, 1-8.

118

Habary, A., Johansen, J. L., Nay, T. J., Steffensen, J. F., & Rummer, J. L. (2017). Adapt,

move or die–how will tropical coral reef fishes cope with ocean warming?. Global

change biology, 23(2), 566-577.

Hannah, L. (2014). Climate change biology. Academic Press.

Hawkins, B. A., & Diniz‐Filho, J. A. F. (2002). The mid‐domain effect cannot explain the

diversity gradient of Nearctic birds. Global Ecology and Biogeography, 11(5), 419-

426.

Hawkins, B. A., Field, R., Cornell, H. V., Currie, D. J., Guégan, J. F., Kaufman, D. M., ...

& Porter, E. E. (2003). Energy, water, and broad‐scale geographic patterns of

species richness. Ecology, 84(12), 3105-3117.

Hawkins, B. A., Diniz-Filho, J. A. F., & Weis, A. E. (2005). The mid-domain effect and

diversity gradients: is there anything to learn?. The American Naturalist, 166(5),

E140-E143.

Herzog S. K, Kessler M, Bach K (2005) The elevational gradient in Andean bird species

richness at the local scale: a foothill peak and a high‐elevation plateau. Ecography

28: 209-222

Herzog SK, Martínez R, Jørgensen PM, Tiessen H (2011) Climate change and biodiversity

in the tropical Andes (p. 348) São José dos Campos: Inter-American Institute for

Global Change Research

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

of Climatology: A Journal of the Royal Meteorological Society, 25(15), 1965-1978.

https://CRAN.R-project.org/package=rsq

119

Huber, U. M., Bugmann, H. K., & Reasoner, M. A. (Eds.). (2006). Global change and

mountain regions: an overview of current knowledge (Vol. 23). Springer Science &

Business Media.

Hutter, C. R., Guayasamin, J. M., & Wiens, J. J. (2013). Explaining Andean megadiversity:

the evolutionary and ecological causes of glassfrog elevational richness patterns.

Ecology letters, 16(9), 1135-1144.

INEI (2017). Instituto Nacional de Estadistica e Informatica. Censo Nacional 2017.

https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib083

8/Libro16/cap01.pdf

Jankowski, J. E., Robinson, S. K., & Levey, D. J. (2010). Squeezed at the top: interspecific

aggression may constrain elevational ranges in tropical birds. Ecology, 91(7), 1877-

1884.

Jankowski J. E, Merkord CL, Ríos W. F, Cabrera K. G, Revilla N. S, Silman M. R (2013)

The relationship of tropical bird communities to tree species composition and

vegetation structure along an Andean elevational gradient. Journal of

Biogeography, 40: 950-962.

Janzen, D. H. (1967). Why mountain passes are higher in the tropics. The American

Naturalist, 101(919), 233-249.

Jetz, W., & Rahbek, C. (2001). Geometric constraints explain much of the species richness

pattern in African birds. Proceedings of the National Academy of Sciences, 98(10),

5661-5666.

Jetz, W., & Rahbek, C. (2002). Geographic range size and determinants of avian species

richness. Science, 297(5586), 1548-1551.

120

Kattan G. H, & Franco P. (2004) Bird diversity along elevational gradients in the Andes of

Colombia: area and mass effects. Global Ecology and Biogeography, 135: 451-458.

doi:10.1111/j.1466-822X.2004.00117.x

Kessler M, Herzog SK, Fjeldså J, Bach K (2001) Species richness and endemism of plant

and bird communities along two gradients of elevation humidity and land use in the

Bolivian Andes. Diversity and distributions, 7: 61-77

Kessler, M. (2000). Elevational gradients in species richness and endemism of selected

plant groups in the central Bolivian Andes. Plant ecology, 149(2), 181-193.

Kessler, M. (2001). Patterns of diversity and range size of selected plant groups along an

elevational transect in the Bolivian Andes. Biodiversity & Conservation, 10(11),

1897-1921.

Kessler, M., Kluge, J., Hemp, A., & Ohlemüller, R. (2011). A global comparative analysis

of elevational species richness patterns of ferns. Global ecology and biogeography,

20(6), 868-880.

Khaliq, I., Hof, C., Prinzinger, R., Böhning-Gaese, K., & Pfenninger, M. (2014). Global

variation in thermal tolerances and vulnerability of endotherms to climate change.

Proceedings of the Royal Society B: Biological Sciences, 281(1789), 20141097.

Körner, C. (2000). Why are there global gradients in species richness? Mountains might

hold the answer. Trends in Ecology & Evolution, 15(12), 513-514.

Krömer, T. , Kessler, M. , Robbert Gradstein, S. and Acebey, A. (2005), Diversity patterns

of vascular epiphytes along an elevational gradient in the Andes. Journal of

Biogeography, 32: 1799-1809. doi:10.1111/j.1365-2699.2005.01318.x

121

Küper, W. , Kreft, H. , Nieder, J. , Köster, N. and Barthlott, W. (2004), Large‐scale

diversity patterns of vascular epiphytes in Neotropical montane rain forests. Journal

of Biogeography, 31: 1477-1487. doi:10.1111/j.1365-2699.2004.01093.x

Lawler JJ, Shafer SL, White D, Kareiva P, Maurer EP, Blaustein AR, Bartlein PJ (2009)

Projected climate-induced faunal change in the Western Hemisphere. Ecology 90:

588-597

Lizaga, I., Gaspar, L., Quijano, L., Dercon, G., & Navas, A. (2019). NDVI, 137Cs and

nutrients for tracking soil and vegetation development on glacial landforms in the

Lake Parón Catchment (Cordillera Blanca, Perú). Science of The Total

Environment, 651, 250-260.

Maddala, G. S. (1983) Limited-Dependent and Qualitative Variables in Econometrics.

Cambridge University.

Magee, L. (1990) R^2 measures based on Wald and likelihood ratio joint significance tests.

The American Statistician, 44: 250-253.

Martín-López, B., Leister, I., Cruz, P. L., Palomo, I., Grêt-Regamey, A., Harrison, P. A., ...

& Walz, A. (2019). Nature’s contributions to people in mountains: A review. PloS

one, 14(6), e0217847.

Mawdsley, J. R., O’malley, R., & Ojima, D. S. (2009). A review of climate‐change

adaptation strategies for wildlife management and biodiversity conservation.

Conservation Biology, 23(5), 1080-1089.

McCain C. M, & Grytnes J. A. (2010) Elevational gradients in species richness eLS

McCain, C. M. (2003). North American desert rodents: a test of the mid-domain effect in

species richness. Journal of Mammalogy, 84(3), 967-980.

122

McCain C. M. (2004) The mid‐domain effect applied to elevational gradients: species

richness of small mammals in Costa Rica. Journal of Biogeography, 311: 19-31

McCain, C. M. (2007) Could temperature and water availability drive elevational species

richness patterns? A global case study for bats." Global Ecology and biogeography,

16.1 1-13.

McCain C. M. (2009) Global analysis of bird elevational diversity. Global Ecology and

Biogeography. 183: 346-360

McCain, C. M. (2010). Global analysis of reptile elevational diversity. Global Ecology and

Biogeography, 19(4), 541-553.

Meza‐Joya, F. L., & Torres, M. (2016). Spatial diversity patterns of Pristimantis frogs in

the Tropical Andes. Ecology and evolution, 6(7), 1901-1913.

Ministerio del Ambiente del Perú. (2015). Mapa nacional de cobertura vegetal: memoria

descriptiva / Ministerio del Ambiente, Dirección General de Evaluación, Valoración

y Financiamiento del Patrimonio Natural. -- Lima: MINAM, 2015. 100 : il. col.,

maps., tbls. Depósito Legal en la Biblioteca Nacional del Perú n.° 2015-13226

Mittelbach, G. G., Steiner, C. F., Scheiner, S. M., Gross, K. L., Reynolds, H. L., Waide, R.

B., ... & Gough, L. (2001). What is the observed relationship between species

richness and productivity?. Ecology, 82(9), 2381-2396.

Møller, A. P., Fiedler, W., & Berthold, P. (Eds.). (2010). Effects of climate change on

birds. Oxford University Press.

Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GA, Kent J (2000) Biodiversity

hotspots for conservation priorities. Nature, 403: 853-858

123

Nogués-Bravo, D., Araújo, M. B., Romdal, T., & Rahbek, C. (2008). Scale effects and

human impact on the elevational species richness gradients. Nature, 453(7192),

216.

O’Donnell, M. S., & Ignizio, D. A. (2012). Bioclimatic predictors for supporting ecological

applications in the conterminous United States. US Geological Survey Data Series,

691(10).

Oliver, T. H., & Morecroft, M. D. (2014). Interactions between climate change and land

use change on biodiversity: attribution problems, risks, and opportunities. Wiley

Interdisciplinary Reviews: Climate Change, 5(3), 317-335.

Pan, X., Ding, Z., Hu, Y., Liang, J., Wu, Y., Si, X., ... & Jin, K. (2016). Elevational pattern

of bird species richness and its causes along a central Himalaya gradient, China.

PeerJ, 4, e2636.

Parmesan, C. (2006). Ecological and evolutionary responses to recent climate change.

Annu. Rev. Ecol. Evol. Syst., 37, 637-669.

Parmesan, C., & Yohe, G. (2003). A globally coherent fingerprint of climate change

impacts across natural systems. Nature, 421(6918), 37.

Patterson, B. D., Pacheco, V., & Solari, S. (1996). Distribution of bats along an elevational

gradient in the Andes of south‐eastern Peru. Journal of Zoology, 240(4), 637-658.

Patterson B. D, Stotz DF, Solari S, Fitzpatrick J. W, Pacheco V (1998) Contrasting patterns

of elevational zonation for birds and mammals I n the Andes of southeastern Peru.

Journal of Biogeography, 25(3) 593-607

Perrigo, A., Hoorn, C., & Antonelli, A. (2019). Why mountains matter for

biodiversity. Journal of Biogeography.

124

Pettorelli, N., Vik, J. O., Mysterud, A., Gaillard, J. M., Tucker, C. J., & Stenseth, N. C.

(2005). Using the satellite-derived NDVI to assess ecological responses to

environmental change. Trends in ecology & evolution, 20(9), 503-510.

Pimm, S. L., & Brown, J. H. (2004). Domains of diversity. Science, 304(5672), 831-833.

Polato, N. R., Gill, B. A., Shah, A. A., Gray, M. M., Casner, K. L., Barthelet, A., ... &

Kondratieff, B. C. (2018). Narrow thermal tolerance and low dispersal drive higher

speciation in tropical mountains. Proceedings of the National Academy of Sciences,

201809326.

PROFONANPE (2010) Plan Maestro del Parque Nacional Huascarán 2010-2015.

Resolución Presidencial Nº 241-2010-SERNANP

QGIS Development Team, 2018. QGIS Geographic Information System. Open Source

Geospatial Foundation. URL http://qgis.osgeo.org

Quintero, I., & Jetz, W. (2018). Global elevational diversity and diversification of birds.

Nature, 555(7695), 246.

Rahbek, C. (1997). The relationship among area, elevation, and regional species richness in

neotropical birds. The American Naturalist, 149(5), 875-902.

Rahbek, C., Borregaard, M. K., Antonelli, A., Colwell, R. K., Holt, B. G., Nogues-Bravo,

D., ... & Fjeldså, J. (2019). Building mountain biodiversity: Geological and

evolutionary processes. Science, 365(6458), 1114-1119.

Rahbek, C., Borregaard, M. K., Colwell, R. K., Dalsgaard, B., Holt, B. G., Morueta-Holme,

N., ... & Fjeldså, J. (2019). Humboldt’s enigma: What causes global patterns of

mountain biodiversity?. Science, 365(6458), 1108-1113.

125

Rehm, E. M., & Feeley, K. J. (2016). Seedling transplants reveal species-specific responses

of high-elevation tropical treeline trees to climate change. Oecologia, 181(4), 1233-

1242.

Remsen Jr, J. V. (1985). Community organization and ecology of birds of high elevation

humid forest of the Bolivian Andes. Ornithological Monographs, 733-756.

Riahi, K., Rao, S., Krey, V., Cho, C., Chirkov, V., Fischer, G., ... & Rafaj, P. (2011). RCP

8.5—A scenario of comparatively high greenhouse gas emissions. Climatic Change,

109(1-2), 33.

Rosenzweig, M. L. (1992). Species diversity gradients: we know more and less than we

thought. Journal of mammalogy, 73(4), 715-730.

Rosenzweig, M. L. (1995). Species diversity in space and time. Cambridge University

Press.

Rouse, J. W., Haas, R. H., Deering, D. W. & Sehell, J. A., (1974): Monitoring the vernal

advancement and retrogradation (Green wave effect) of natural vegetation. Final

Rep. RSC 1978-4, Remote Sensing Center, Texas A&M Univ., College Station.

Salazar, L., Homeier, J., Kessler, M., Abrahamczyk, S., Lehnert, M., Krömer, T., & Kluge,

J. (2015). Diversity patterns of ferns along elevational gradients in Andean tropical

forests. Plant Ecology & Diversity, 8(1), 13-24.

Sanders, N. J. (2002). Elevational gradients in ant species richness: area, geometry, and

Rapoport's rule. Ecography, 25(1), 25-32.

Schulenberg, T. S., Stotz, D. F., Lane, D. F., O'Neill, J. P., & Parker III, T. A. (2010). Birds

of Peru: revised and updated edition (Vol. 63). Princeton University Press.

Şekercioğlu, C. H., Schneider, S. H., Fay, J. P., & Loarie, S. R. (2008). Climate change,

elevational range shifts, and bird extinctions. Conservation biology, 22(1), 140-150. 126

Şekercioğlu, Ç. H., Primack, R. B., & Wormworth, J. (2012). The effects of climate change

on tropical birds. Biological Conservation, 148(1), 1-18.

Servat GP (2006) The role of local and regional factors in the foraging ecology of birds

associated with Polylepis woodlands. Doctoral Thesis University of Missouri–St.

Louis Vol 70 No 11

Sevillano-Ríos CS (2016) Diversity ecology and conservation of bird communities of

Polylepis woodlands in the northern Andes of Peru. Master Thesis Cornell

University

Sevillano-Ríos CS, Lloyd H, Valdés-Velásquez A (2011) Bird species richness diversity

and abundance in Polylepis woodlands Huascaran biosphere reserve Peru. Studies

on Neotropical Fauna and Environment, 46: 69-76

Sevillano-Ríos, C.S. and Rodewald, A.D., (2017). Avian community structure and habitat

use of Polylepis forests along an elevation gradient. PeerJ, 5, p.e3220.

Sevillano-Ríos, C.S., (2017). Breve Historia de la Ornitología en los Altos Andes del Norte

del Perú y Su Importancia para la Conservación. Revista De Glaciares y

Ecosistemas De Montaña, 2, pp.87-102.

Taiyun, W., & Viliam, S. (2017). R package corrplot: visualization of a correlation matrix

(Version 0.84).

Terborgh, J. (1971). Distribution on environmental gradients: theory and a preliminary

interpretation of distributional patterns in the avifauna of the Cordillera

Vilcabamba, Peru. Ecology, 52(1), pp.23-40.

Terborgh J (1977) Bird species diversity on an Andean elevational gradient. Ecology 1007-

1019

127

Terborgh, J. (1973). On the notion of favorableness in plant ecology. The American

Naturalist, 107(956), 481-501.

Terborgh, J. (1985). The role of ecotones in the distribution of Andean birds. Ecology,

66(4), pp.1237-1246.

Terborgh, J. and Weske, J.S. (1975). The role of competition in the distribution of Andean

birds. Ecology, 56(3), pp.562-576.

Tingley, M. W., Koo, M. S., Moritz, C., Rush, A. C., & Beissinger, S. R. (2012). The push

and pull of climate change causes heterogeneous shifts in avian elevational ranges.

Global Change Biology, 18(11), 3279-3290.

Toszogyova, A., & Storch, D. (2019). Global diversity patterns are modulated by temporal

fluctuations in primary productivity. Global Ecology and Biogeography.

Venables, W. N. & Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth Edition.

Springer, New York. ISBN 0-387-95457-0

Venter, O., Sanderson, E. W., Magrach, A., Allan, J. R., Beher, J., Jones, K. R., ... & Levy,

M. A. (2016). Sixteen years of change in the global terrestrial human footprint and

implications for biodiversity conservation. Nature Communications, 7, 12558.

Ver Hoef, J. M., & Boveng, P. L. (2007). Quasi‐Poisson vs. negative binomial regression:

how should we model overdispersed count data?. Ecology, 88(11), 2766-2772.

Visser, M. E., & Both, C. (2005). Shifts in phenology due to global climate change: the

need for a yardstick. Proceedings of the Royal Society of London B: Biological

Sciences, 272(1581), 2561-2569.

Voss, R. S. (2003). A new species of Thomasomys (Rodentia: Muridae) from eastern

Ecuador, with remarks on mammalian diversity and biogeography in the Cordillera

Oriental. American Museum Novitates, 1-47. 128

Waide, R. B., Willig, M. R., Steiner, C. F., Mittelbach, G., Gough, L., Dodson, S. I., ... &

Parmenter, R. (1999). The relationship between productivity and species richness.

Annual review of Ecology and Systematics, 30(1), 257-300.

Wang, J., Rich, P. M., & Price, K. P. (2003). Temporal responses of NDVI to precipitation

and temperature in the central Great Plains, USA. International journal of remote

sensing, 24(11), 2345-2364.

Wormworth J, & Şekercioğlu C. H. (2011). Winged sentinels: birds and climate change.

Cambridge University Press

Wu, C., Peng, D., Soudani, K., Siebicke, L., Gough, C. M., Arain, M. A., ... & Xu, S.

(2017). Land surface phenology derived from normalized difference vegetation

index (NDVI) at global FLUXNET sites. Agricultural and forest meteorology, 233,

171-182.

Wu, Y., Yang, Q., Wen, Z., Xia, L., Zhang, Q., & Zhou, H. (2013). What drives the species

richness patterns of non‐volant small mammals along a subtropical elevational

gradient?. Ecography, 36(2), 185-196.

Wu, Z., Dijkstra, P., Koch, G. W., Peñuelas, J., & Hungate, B. A. (2011). Responses of

terrestrial ecosystems to temperature and precipitation change: a meta‐analysis of

experimental manipulation. Global Change Biology, 17(2), 927-942.

Zhang D. (2018). rsq: R-Squared and Related Measures. R package version 1.1.

Zutta BR, Rundel PW, Saatchi S, Casana JD, Gauthier P, Soto A, Buermann W (2012)

Prediciendo la distribución de Polylepis: bosques Andinos vulnerables y cada vez

más importantes. Revista Peruana de Biología, 19: 205-212

129

TABLES

Table 3.1. Model included in the analysis of the three tested hypotheses (climatic, spatial and human). Within each group, models are ordered by AIC.

Models AIC ΔAIC AIC weights Generalized R2 Intercept 283.21 102.40 0.00 -- Individual climatic factors Temperature2 223.58 42.77 0.00 0.897 Precipitation2 244.85 64.04 0.00 0.780 Temperature seasonality 2 269.92 89.11 0.00 0.461 Temperature seasonality 276.50 95.69 0.00 0.267 Isothermality2 279.93 99.12 0.00 0.229 Isothermality 281.69 100.88 0.00 0.118 Temperature 282.63 101.82 0.00 0.088 Temperature seasonality2 282.96 102.15 0.00 0.141 Mean diurnal temperature ranges 283.50 102.70 0.00 0.059 Temperature seasonality 283.99 103.18 0.00 0.043 Precipitation 284.37 103.57 0.00 0.029 Mean annual temperature ranges 285.15 104.34 0.00 0.002 Mean diurnal temperature ranges2 285.50 104.69 0.00 0.059 Mean annual temperature ranges2 286.68 105.88 0.00 0.019 Primary productivity NDVI 226.96 46.15 0.00 0.875 NDVI Wet season 227.42 46.61 0.00 0.873 NDVI2 228.94 48.13 0.00 0.875 NDVI Dry season 229.51 48.70 0.00 0.863 NDVI Wet Min 248.24 67.43 0.00 0.733 NDVI Min 250.00 69.19 0.00 0.716 NDVI Dry Min 256.25 75.44 0.00 0.645 NDVI Wet Max 263.40 82.59 0.00 0.541 NDVI Max 267.96 87.15 0.00 0.460 NDVI Dry Max 274.21 93.40 0.00 0.325 Landscape-spatial hypotheses MDE 255.49 74.685 0.000 0.654 Elevevation2 256.24 75.436 0.000 0.669 Land Area2 262.21 81.400 0.000 0.591 Land Area 271.73 90.926 0.000 0.382 Elevation 284.59 103.784 0.000 0.022 Human hypothesis Human Footprint Index 273.859 93.051 0.000 0.333 Elevation (m) Annual mean temperature (C) = BIO1

130

Annual mean precipitation (mm) = BIO12 NDVI=Normalized Difference Vegetation Index Temperature seasonality = BIO4 (SD*100) Precipitation seasonality = BIO15 (SD*100) Isothermality: BIO3= BIO2/BIO7*100 Mean diurnal temperature ranges = BIO2 Mean annual temperature ranges = BIO7

131

AIC Generalized Models AIC ΔAIC weights R2 Intercept 283.21 102.4 0.00 -- Additive and interactive climatic models Temp2*Prec2*NDVI2 180.81 0.00 0.50 0.987 Temp2 * NDVI2 181.07 0.26 0.44 0.982 Prec2 * NDVI2 185.05 4.24 0.06 0.979 Prec2 + NDVI2 191.29 10.48 0.00 0.972 Temp2 + NDVI2 193.74 12.94 0.00 0.969 Temp + NDVI 211.15 30.34 0.00 0.934 Temp * NDVI 211.92 31.11 0.00 0.937 Temp + Prec + NDVI 212.57 31.76 0.00 0.935 Prec + NDVI 212.79 31.98 0.00 0.930 Prec * NDVI 213.28 32.48 0.00 0.934 Temp2 + Prec2 220.21 39.41 0.00 0.921 Temp2 * Prec2 222.20 41.39 0.00 0.921 Temp * Prec 237.59 56.78 0.00 0.842 Temp + Prec 269.82 89.01 0.00 0.463 Seasonal Temp2 * Seasonal Prec2 270.27 89.46 0.00 0.559 Seasonal Temp2 + Seasonal Prec2 272.98 92.17 0.00 0.479 Seasonal Temp + Seasonal Prec 278.40 97.60 0.00 0.270 Seasonal Temp * Seasonal Prec 280.20 99.39 0.00 0.275 Temp: Annual mean temperature (C) = BIO1 Prec: Annual mean precipitation (mm) = BIO12 NDVI: Normalized Difference Vegetation Index Seasonal Temp: Temperature seasonality = BIO4 (SD*100) Seasonal Prec: Precipitation seasonality = BIO15 (SD*100)

132

Temp2*Prec2*NDVI2 Temp2 * NDVI2 Prec2 * NDVI2 Temp2 * Prec2 (Intercept) -76.29 (±393) 4.97 (±2.66) . 59.04 (±4.96) *** -4.78 (±635.7) Temperature -0.39 (±32.39) 2.08 (±2.74) 28.56 (±52.27) Temperature2 0.34 (±0.67) 0.03 (±0.13) -1.27 (±1.06) Precipitation 0.05 (±0.81) -0.03 (±0.04) -0.31 (±1.32) Precipitation2 0 (±0) 0 (±0) 0 (±0) NDVI 869.5 (±406.9) * 60.92 (±40.19) -66.19 (±63.55) NDVI2 761.1 (±274.5) * 537.99 (±73.14) *** 560.9 (±86.17) *** Temperature* Precipitation 0.01 (±0.03) 0 (±0.05) Temperature*NDVI -51.11 (±20.83) * -5.17 (±1.33) *** Precipitation*NDVI -0.96 (±0.44) * 0.1 (±0.04) * Temperature* Precipitation*NDVI 0 (±0.02) Signif. codes: 0 ‘***’; 0.001 ‘**’; 0.01 ‘*’; 0.05 ‘.’; 0.1 ‘ ’

133

FIGURES

134

135

Figure 3.3. Elevational patterns in raw (from 1,000 points) climatic (a-i), spatial (j-k) and human impact (l) covariates used as explanatory variables of bird species richness along the elevation. (a) Mean annual temperature, (b) mean annual precipitation, (c) average NDVI, (d) temperature seasonality, (e) precipitation seasonality, (f) seasonal NDVI (dry=black, wet=grey), (g) mean temperature diurnal range, (h) annual temperature range, (i) isothermality (g/h), (j) land area for 100-m elevational bands, (k) expected species richness under mid domain effect (DME), (l) Human Footprint Index (HFI) 2009.

136

137

138

Figure 3.6. Avian species richness is most closely related to primary productivity, which at the same time is drive by the interaction of the annual mean temperature and precipitation.

139

140

Figure 3.8. Observed change in the Human Footprint Index between 1993-2009 across the elevation. Points are the difference between 1993 and 2009.

141

Figure 3.10. Modeled species richness in response to changes on temperature and precipitation predicted according to the four RCPs climatic models (predicted estimates=dots and LOESS model = lines). Observed (blue dots) data correspond to the estimated species richness grouped for every 200-m elevational bands whereas predicted (red dots) data correspond to the outcome of the Temp2*Prec2 model (Table 3.2). Correspondingly, the RCP2.6 – RCP8.0 (dots and lines) are the outcomes of this model based on the four future climatic scenarios, where RCP8.0 correspond to the more severe scenario. The maximum amount of land area occurs at 4,200 masl (dashed line).

142

APPENDICES

Table S3.1. List of the 212 bird species included in the empirical species richness estimation.

Elevational range ID Family English name Scientific name Status (Min-Max) 1 Tinamidae Ornate Tinamou Nothoprocta ornata 3300-4400 R 2 Tinamidae Andean Tinamou Nothoprocta pentlandii 2000-3600 R 3 Tinamidae Puna Tinamou Tinamotis pentlandii 3900-5000 R* 4 Anatidae Andean Goose Chloephaga melanoptera 3700-4600 R 5 Anatidae Torrent Duck Merganetta armata 900-3500 R 6 Anatidae Crested Duck Lophonetta specularioides 3500-4800 R 7 Anatidae Yellow-billed Teal Anas flavirostris 2800-4800 R 8 Anatidae Yellow-billed Pintail Anas georgica 3200-4400 R 9 Anatidae White-cheeked Pintail Anas bahamensis 0-500 R 10 Anatidae Puna Teal Anas puna 3000-4600 R 11 Anatidae Cinnamon Teal Anas cyanoptera 0-4400 R 12 Anatidae Ruddy Duck Oxyura jamaicensis 2800-4500 R 13 Phoenicopteridae Chilean Flamingo Phoenicopterus chilensis 3200-4600 R 14 Podicipedidae White-tufted Grebe Rollandia rolland 3200-5000 R 15 Podicipedidae Silvery Grebe Podiceps occipitalis 3200-4700 R 16 Podicipedidae Pied-billed grebe Podilymbus podiceps 0-100 R* 17 Columbidae Band-tailed Pigeon Patagioenas fasciata 1000-3600 R 18 Columbidae White-tipped Dove Leptotila verreauxi 600-2700 R 19 Columbidae West Peruvian Dove Zenaida meloda 0-1000 R 20 Columbidae Eared Dove Zenaida auriculata 0-4000 R 21 Columbidae Croaking Ground Dove Columbina cruziana 0-2800 R 22 Columbidae Bare-faced Ground Dove Metriopelia ceciliae 1700-4100 R 23 Columbidae Black-winged Ground Dove Metriopelia melanoptera 2600-4500 R 24 Caprimulgidae Lesser Nighthawk Chordeiles acutipennis 0-950 R* 25 Caprimulgidae Band-winged Nightjar Systellura longirostris 1950-4400 R 26 Caprimulgidae Tschudi's Nightjar Systellura decussata 0-1300 R* 27 Apodidae White-collared Swift Streptoprocne zonaris 0-4300 R 28 Apodidae Andean Swift Aeronautes andecolus 0-4500 R 29 Trochilidae Sparkling Violetear Colibri coruscans 400-4500 R 30 Trochilidae Bronze-tailed Comet Polyonymus caroli 2100-3400 R(e) 31 Trochilidae Gray-bellied Comet Taphrolesbia griseiventris 2750-3200 R(e)EN 32 Trochilidae Andean Hillstar Oreotrochilus estella 3400-4600 R 33 Trochilidae Black-tailed Trainbearer Lesbia victoriae 2700-4100 R 34 Trochilidae Green-tailed Trainbearer Lesbia nuna 1700-3800 R 35 Trochilidae Olivaceous Thornbill Chalcostigma olivaceum 3500-4700 R 36 Trochilidae Blue-mantled Thornbill Chalcostigma stanleyi 2800-4500 R 37 Trochilidae Tyrian Metaltail Metallura tyrianthina 2400-4200 R

143

Elevational range ID Family English name Scientific name Status (Min-Max) 38 Trochilidae Black Metaltail Metallura phoebe 2700-4300 R(e) 39 Trochilidae Shining Sunbeam Aglaeactis cupripennis 2500-4600 R 40 Trochilidae Rainbow Starfrontlet Coeligena iris 1500-3800 R 41 Trochilidae Giant Hummingbird Patagona gigas 2000-3400 R 42 Trochilidae Purple-collared Woodstar Myrtis fanny 700-3200 R 43 Trochilidae Oasis Hummingbird Rhodopis vesper 0-3800 R 44 Trochilidae Peruvian Sheartail Thaumastura cora 0-2800 R 45 Trochilidae Spot-throated Hummingbird Leucippus taczanowskii 350-2800 R(e) 46 Rallidae Plumbeous Rail Pardirallus sanguinolentus 0-4400 R 47 Rallidae Common Gallinule Gallinula galeata 2200-4400 R 48 Rallidae Giant Coot Fulica gigantea 3900-4600 R 49 Rallidae Slate-colored Coot Fulica ardesiaca 2500-4600 R 50 Charadriidae Andean Lapwing Vanellus resplendens 3000-4600 R 51 Charadriidae Killdeer Charadrius vociferus 0-200 R 52 Charadriidae Diademed Sandpiper-Plover Phegornis mitchellii 4100-5000 R 53 Charadriidae Puna Plover Charadrius alticola 3800-4500 R* 54 Scolopacidae Whimbrel Numenius phaeopus 0-200 R* 55 Scolopacidae Jameson's Snipe Gallinago jamesoni 2800-3600 R 56 Scolopacidae Puna Snipe Gallinago andina 3000-4600 R 57 Thinocoridae Rufous-bellied Seedsnipe Attagis gayi 4400-5000 R 58 Thinocoridae Gray-breasted Seedsnipe Thinocorus orbignyianus 3000-4600 R 59 Laridae Andean Gull Chroicocephalus serranus 3000-4400 R 60 Laridae Gray-hooded Gull Chroicocephalus cirrocephalus 0-200 R 61 Ardeidae Black-crowned Night-Heron Nycticorax nycticorax 3100-4700 R 62 Ardeidae Cattle Egret Bubulcus ibis 0-2000 R 63 Ardeidae Great Egret Ardea alba 0-500 R 64 Ardeidae Snowy Egret Egretta thula 0-500 R 65 Ardeidae Little Blue Heron Egretta caerulea 0-500 R 66 Ardeidae Striated Heron Butorides striata 0-500 R* 67 Threskiornithidae Puna Ibis Plegadis ridgwayi 3200-4500 R 68 Threskiornithidae Andean Ibis Theristicus branickii 3700-4600 R 69 Cathartidae Turkey Vulture Cathartes aura 0-2200 R 70 Cathartidae Black Vulture Coragyps atratus 0-1200 R 71 Cathartidae Andean Condor Vultur gryphus 3000-5000 R NT 72 Accipitridae Cinereous Harrier Circus cinereus 2700-4200 R 73 Accipitridae Variable Hawk Geranoaetus polyosoma 0-4600 R 74 Accipitridae Black-chested Buzzard-Eagle Geranoaetus melanoleucus 1600-4600 R 75 Accipitridae White-throated Hawk Buteo albigula 1500-3700 R 76 Falconidae Mountain Caracara Phalcoboenus megalopterus 3200-4700 R 77 Falconidae American Kestrel Falco sparverius 0-4700 R 78 Falconidae Aplomado Falcon Falco femoralis 2400-4300 R 79 Falconidae Peregrine Falcon Falco peregrinus 1800-4300 R 144

Elevational range ID Family English name Scientific name Status (Min-Max) 80 Cuculidae Groove-billed Ani Crotophaga sulcirostris 0-2750 R 81 Strigidae Koepcke's Screech-Owl Megascops koepckeae 2200-4000 R(e) 82 Strigidae Great Horned Owl Bubo virginianus 2600-4400 R 83 Strigidae Peruvian Pygmy-Owl Glaucidium peruanum 900-2100 R 84 Strigidae Burrowing Owl Athene cunicularia 0-4600 R 85 Picidae Smoky-brown Woodpecker Picoides fumigatus 1200-2900 R 86 Picidae Black-necked Woodpecker Colaptes atricollis 600-2800 R(e) 87 Picidae Andean Flicker Colaptes rupicola 2700-4500 R 88 Psittacidae Mountain Parakeet Psilopsiagon aurifrons 500-3100 R 89 Psittacidae Andean Parakeet Bolborhynchus orbygnesius 2400-3900 R 90 Psittacidae Scarlet-fronted Parakeet Psittacara wagleri 2400-3000 R 91 Grallariidae Stripe-headed Antpitta Grallaria andicolus 3500-4600 R 92 Rhinocryptidae Ancash Tapaculo Scytalopus affinis 3000-4600 R(e) 93 Furnaridae Dark-winged Miner Geossita saxicolina 0-700 R(e) 94 Furnariidae Slender-billed Miner Geositta tenuirostris 2650-4600 R 95 Furnariidae Common Miner Geositta cunicularia 3100-4800 R 96 Furnariidae Striated Earthcreeper Geocerthia serrana 3000-4600 R(e) 97 Furnariidae Buff-breasted Earthcreeper Upucerthia validirostris 3400-4800 R 98 Furnariidae Chestnut-winged Cinclodes Cinclodes albidiventris 2750-4800 R 99 Furnariidae White-bellied Cinclodes Cinclodes palliatus 4400-5000 R* EN 100 Furnariidae White-winged Cinclodes Cinclodes atacamensis 2800-4600 R 101 Furnariidae Tawny Tit-Spinetail Leptasthenura yanacensis 3950-4600 R NT 102 Furnariidae Rusty-crowned Tit-Spinetail Leptasthenura pileata 2900-4200 R(e) 103 Furnariidae Streaked Tit-Spinetail Leptasthenura striata 2000-4200 R 104 Furnariidae Andean Tit-Spinetail Leptasthenura andicola 3500-4200 R 105 Furnariidae Pale-tailed Canastero Asthenes dorbignyi 2000-2500 R** 106 Furnariidae Many-striped Canastero Asthenes flammulata 2900-4400 R 107 Furnariidae Streak-backed Canastero Asthenes wyatti 3500-4600 R 108 Furnariidae Streak-throated Canastero Asthenes humilis 3700-4800 R 109 Furnariidae Cordilleran Canastero Asthenes modesta 3600-4600 R 110 Furnariidae Canyon Canastero Asthenes pudibunda 2450-4000 R 111 Furnariidae Baron's Spinetail Cranioleuca baroni 2000-4400 R(e)NT 112 Furnariidae Great Spinetail Synallaxis hypochondriaca 2150-2800 R 113 Furnariidae Russet-bellied Spinetail Synallaxis zimmeri 2100-2900 R(e)EN 114 Tyrannidae White-crested Elaenia Elaenia albiceps 800-3500 R 115 Tyrannidae Southern Beardless-Tyrannulet Camptostoma obsoletum 0-2600 R 116 Tyrannidae White-throated Tyrannulet Mecocerculus leucophrys 1800-4600 R 117 Tyrannidae Black-crested Tit-Tyrant Anairetes nigrocristatus 2100-4000 R 118 Tyrannidae Pied-crested Tit-Tyrant Anairetes reguloides 0-2800 R 119 Tyrannidae Ash-breasted Tit-Tyrant Anairetes alpinus 3700-4600 R EN 120 Tyrannidae Yellow-billed Tit-Tyrant Anairetes flavirostris 1900-4100 R 121 Tyrannidae Tufted Tit-Tyrant Anairetes parulus 1450-4400 R* 145

Elevational range ID Family English name Scientific name Status (Min-Max) 122 Tyrannidae Torrent Tyrannulet Serpophaga cinerea 800-2800 R 123 Tyrannidae Tawny-crowned Pygmy-Tyrant Euscarthmus meloryphus 0-2500 R 124 Tyrannidae Gray-and-white Tyrannulet Pseudelaenia leucospodia 0-800 R* 125 Tyrannidae Bran-colored Flycatcher Myiophobus fasciatus 0-1800 R 126 Tyrannidae Tropical Pewee Contopus cinereus 1000-2800 R 127 Tyrannidae Black Phoebe Sayornis nigricans 2500-3500 R 128 Tyrannidae Vermilion Flycatcher Pyrocephalus rubinus 0-2500 R 129 Tyrannidae Andean Negrito Lessonia oreas 3100-4600 R 130 Tyrannidae White-winged Black-Tyrant Knipolegus aterrimus 1800-3600 R 131 Tyrannidae Spot-billed Ground-Tyrant Muscisaxicola maculirostris 2000-4500 R 132 Tyrannidae Taczanowski's Ground-Tyrant Muscisaxicola griseus 3200-4800 R 133 Tyrannidae Puna Ground-Tyrant Muscisaxicola juninensis 3800-4800 R 134 Tyrannidae White-fronted Ground-Tyrant Muscisaxicola albifrons 3700-4900 R 135 Tyrannidae Rufous-naped Ground-Tyrant Muscisaxicola rufivertex 2700-4200 R 136 Tyrannidae Short-tailed Field Tyrant Muscigralla brevicauda 0-1200 R 137 Tyrannidae Black-billed Shrike-Tyrant Agriornis montanus 3000-4500 R 138 Tyrannidae White-tailed Shrike-Tyrant Agriornis albicauda 2400-4300 R* 139 Tyrannidae Streak-throated Bush-Tyrant Myiotheretes striaticollis 1700-3700 R 140 Tyrannidae Rufous-webbed Bush-Tyrant Polioxolmis rufipennis 3100-4600 R 141 Tyrannidae Jelski's Chat-Tyrant Ochthoeca jelskii 2300-3200 R 142 Tyrannidae Rufous-breasted Chat-Tyrant Ochthoeca rufipectoralis 2300-4100 R 143 Tyrannidae d'Orbigny's Chat-Tyrant Ochthoeca oenanthoides 3400-4600 R 144 Tyrannidae Piura Chat-Tyrant Ochthoeca piurae 0-1000 R(e) 145 Tyrannidae White-browed Chat-Tyrant Ochthoeca leucophrys 2400-4200 R 146 Tyrannidae Tropical Kingbird Tyrannus melancholicus 0-2100 R 147 Tyrannidae Dusky-capped Flycatcher Myiarchus tuberculifer 1000-3250 R 148 Cotingidae White-cheeked Cotinga Zaratornis stresemanni 3800-4400 R(e)VU 149 Cotingidae Red-crested Cotinga Ampelion rubrocristatus 2400-3700 R 150 Hirundinidae Blue-and-white Swallow Pygochelidon cyanoleuca 0-4300 R 151 Hirundinidae Brown-bellied Swallow Orochelidon murina 2200-4300 R 152 Hirundinidae Andean Swallow Orochelidon andecola 3500-4600 R 153 Troglodytidae House Wren Troglodytes aedon 0-4600 R 154 Cinclidae White-capped Dipper Cinclus leucocephalus 1500-3100 R 155 Turdidae Great Thrush Turdus fuscater 2400-4200 R 156 Turdidae Chiguanco Thrush Turdus chiguanco 2400-4300 R 157 Mimidae Long-tailed Mockingbird Mimus longicaudatus 0-2600 R 158 Motacillidae Short-billed Pipit Anthus furcatus 3500-4100 R 159 Motacillidae Correndera Pipit Anthus correndera 3800-4600 R 160 Motacillidae Paramo Pipit Anthus bogotensis 2950-4400 R 161 Thraupidae Rufous-chested Tanager Thlypopsis ornata 1600-3300 R 162 Thraupidae Fawn-breasted Tanager Pipraeidea melanonota 1100-2900 R 163 Thraupidae Blue-and-yellow Tanager Pipraeidea bonariensis 2000-4200 R 146

Elevational range ID Family English name Scientific name Status (Min-Max) 164 Thraupidae Cinereous Conebill Conirostrum cinereum 0-4200 R 165 Thraupidae Giant Conebill Oreomanes fraseri 3500-4600 R NT 166 Thraupidae Tit-like Dacnis Xenodacnis parina 3200-4600 R 167 Thraupidae Black-throated Flowerpiercer Diglossa brunneiventris 2400-4300 R 168 Thraupidae Rusty Flowerpiercer Diglossa sittoides 1200-3500 R 169 Thraupidae Peruvian Sierra-Finch Phrygilus punensis 2800-4700 R 170 Thraupidae Mourning Sierra-Finch Phrygilus fruticeti 2300-4200 R 171 Thraupidae Plumbeous Sierra-Finch Phrygilus unicolor 3000-4700 R 172 Thraupidae Ash-breasted Sierra-Finch Phrygilus plebejus 2400-4700 R 173 Thraupidae Band-tailed Sierra-Finch Phrygilus alaudinus 2000-3100 R 174 Thraupidae White-winged Diuca-Finch Diuca speculifera 3950-4800 R 175 Thraupidae Great Inca-Finch Incaspiza pulchra 1000-2700 R(e) 176 Thraupidae Rufous-backed Inca-Finch Incaspiza personata 2300-4000 R(e) 177 Thraupidae Buff-bridled Inca-Finch Incaspiza laeta 1000-2750 R(e) 178 Thraupidae Plain-tailed Warbling-Finch Poospiza alticola 3100-4600 R(e)EN 179 Thraupidae Rufous-breasted Warbling-Finch Poospiza rubecula 2500-3800 R(e)EN 180 Thraupidae Collared Warbling-Finch Poospiza hispaniolensis 0-2900 R 181 Thraupidae Bright-rumped Yellow-Finch Sicalis uropygialis 3300-4700 R 182 Thraupidae Greenish Yellow-Finch Sicalis olivascens 1650-4200 R 183 Thraupidae Saffron Finch Sicalis flaveola 0-3100 R 184 Thraupidae Raimondi's Yellow-Finch Sicalis raimondii 200-2500 R(e) 185 Thraupidae Grassland Yellow-Finch Sicalis luteola 1700-3900 R* 186 Thraupidae Streaked Saltator Saltator striatipectus 0-1200 R* 187 Thraupidae Golden-billed Saltator Saltator aurantiirostris 2100-4000 R 188 Thraupidae Blue-black Grassquit Volatinia jacarina 0-2400 R 189 Thraupidae Parrot-billed Seedeater Sporophila peruviana 0-600 R* 190 Thraupidae Chestnut-throated Seedeater Sporophila telasco 0-700 R* 191 Thraupidae Drab Seedeater Sporophila simplex 500-2500 R 192 Thraupidae Black-and-white Seedeater Sporophila luctuosa 1400-3200 R 193 Thraupidae Band-tailed Seedeater Catamenia analis 0-4000 R 194 Thraupidae Plain-colored Seedeater Catamenia inornata 2600-4400 R 195 Thraupidae Bananaquit Coereba flaveola 0-1600 R 196 Thraupidae Dull-colored Grassquit Tiaris obscurus 350-2100 R* 197 Emberizidae Rufous-collared Sparrow Zonotrichia capensis 0-4500 R 198 Emberizidae Rufous-eared Brushfinch Atlapetes rufigenis 3200-4600 R(e)NT 199 Emberizidae Yellow-breasted Brushfinch Atlapetes latinuchus 1850-3225 R 200 Emberizidae Bay-crowned Brushfinch Atlapetes seebohmi 1150-2800 R* 201 Emberizidae Rusty-bellied Brushfinch Atlapetes nationi 2100-4000 R(e) 202 Cardinalidae Hepatic Tanager Piranga flava 0-2700 R 203 Cardinalidae Golden Grosbeak Pheucticus chrysogaster 0-3500 R 204 Parulidae Masked Yellowthroat Geothlypis aequinoctialis 0-2700 R 205 Parulidae Black-crested Warbler Myiothlypis nigrocristata 1800-3300 R 147

Elevational range ID Family English name Scientific name Status (Min-Max) 206 Icteridae Scrub Blackbird Dives warczewiczi 0-2400 R 207 Icteridae Shiny Cowbird Molothrus bonariensis 0-700 R 208 Icteridae Peruvian Meadowlark Sturnella bellicosa 0-2800 R 209 Fringillidae Thick-billed Siskin Spinus crassirostris 3600-4600 R 210 Fringillidae Hooded Siskin Spinus magellanicus 0-4200 R 211 Fringillidae Black Siskin Spinus atratus 3500-4700 R 212 Fringillidae Yellow-rumped Siskin Spinus uropygialis 3200-4200 R

R: resident species *: resident with few records (e): endemic species NT: near threatened VU: vulnerable EN: endangered **: sub nw Spp.

148

Table S3.2. Landsat8 images used to generate the NDVI raster layers for the dry and wet seasons.

Season ProductID Cloud cover LC08_L1TP_009066_20150722_20170406_01_T1 40 LC08_L1TP_008067_20150731_20170406_01_T1 25 Dry LC08_L1TP_008066_20150731_20170406_01_T1 42 LC08_L1TP_007067_20150724_20170406_01_T1 22 LC08_L1TP_009066_20160419_20170326_01_T1 15 LC08_L1TP_008067_20160428_20170326_01_T1 8 Wet LC08_L1TP_008066_20160428_20170326_01_T1 34 LC08_L1TP_007067_20160421_20170326_01_T1 25

149

Table S3.3. Models for raw data used to test three climatic, spatial, and human hypotheses of species richness. Within each group, models are ordered by their Goodness of Fit (GOF = null deviance – deviance), measured as the degree of improvement after incorporating the explanatory covariates (High GOF values correspond to a high improvement), which is highly correlated with the generalized R2 in a log-log scale (Spearman’s rs correlation = 0.99, p<0.0001). d.f = degrees of freedom. Theta = the negative binomial dispersal parameter.

Models Theta null.deviance deviance GOF d.f. Generalized R2 Intercept 4.91 1081.48 1081.48 0.00 999 0.00 Individual climatic factors Temp2 19.95 3363.76 1251.03 2112.73 997 0.65 Prec2 7.57 1572.92 1115.59 457.33 997 0.31 Seasonal Temp2 7.00 1472.13 1107.55 364.58 997 0.26 Seasonal Temp 6.89 1452.05 1106.87 345.18 998 0.25 Mean diurnal temperature ranges2 6.38 1359.05 1098.50 260.55 997 0.21 Mean diurnal temperature ranges 6.36 1355.64 1098.41 257.23 998 0.20 Isothermality2 6.26 1337.89 1096.78 241.10 997 0.19 Isothermality 6.20 1327.02 1096.37 230.65 998 0.19 Mean annual temperature ranges2 5.52 1199.35 1088.21 111.14 997 0.10 Mean annual temperature ranges 5.52 1199.34 1088.22 111.13 998 0.10 Prec 5.29 1155.55 1084.76 70.79 998 0.07 Temp 5.07 1112.59 1082.62 29.96 998 0.03 Seasonal Prec2 4.91 1081.53 1081.48 0.05 997 0.00 Seasonal Prec 4.91 1081.51 1081.48 0.03 998 0.00 Primary productivity NDVI2 11.40 2202.29 1167.53 1034.76 997 0.49 NDVI 10.06 1991.51 1150.29 841.22 998 0.45 NDVI Dry season 8.82 1788.00 1132.82 655.18 998 0.39 NDVI Wet season 8.62 1754.01 1130.26 623.75 998 0.38 NDVI_ Max 6.74 1425.93 1104.09 321.85 998 0.24 NDVI_Wet_ Max 6.47 1375.48 1100.14 275.34 998 0.21 NDVI_Dry_ Max 6.43 1367.83 1099.49 268.34 998 0.21 NDVI_ Min 6.31 1347.09 1097.78 249.31 998 0.20 NDVI_ Wet_Min 6.05 1298.90 1096.88 202.02 998 0.17 NDVI_ Dry_Min 5.97 1283.08 1092.15 190.93 998 0.16 Landscape-spatial hipotheses MDE 10.29 2028.43 1147.972 880.46 998 0.46 Elev2 9.32 1870.856 1139.017 731.839 997 0.41 Land Area2 8.18 1679.205 1119.799 559.406 997 0.35 Land Area 6.36 1356.474 1097.496 258.978 998 0.20

150

Elev 5.42 1178.748 1085.592 93.156 998 0.13 Roughness terrain 4.98 1095.073 1082.146 12.93 998 0.01 Human hypothesis Human Footprint Index 4.97 1093.188 1082.009 11.18 998 0.01

Elev: Elevation (m) Temp: BIO1, Annual mean temperature (C0) Prec: BIO12, Annual mean precipitation (mm) NDVI: Normalized Difference Vegetation Index Seasonal Temp: BIO4, Temperature seasonality (SD*100) Seasonal Prec: BIO15, Precipitation seasonality (SD*100) Isothermality: BIO3= BIO2/BIO7*100 BIO2, mean diurnal temperature ranges BIO7, mean annual temperature ranges

151

Table S3.4. Additive and interactive climatic model included in the analysis. Models are ordered by their Goodness of Fit (GOF = null deviance – deviance), measured as the degree of improvement after incorporating the explanatory covariates (High GOF values correspond to a high improvement), which is highly correlated with the generalized R2 in a log-log scale (Spearman’s rs correlation = 0.99, p<0.0001). d.f = degrees of freedom. Theta = the negative binomial dispersal parameter.

Models Theta null.deviance deviance GOF d.f. Generalized R2 Intercept 4.91 1081.48 1081.48 0.00 999 0.00 Additive and interactive climatic models Temp2 * Prec2 * NDVI2 320.91 11201.60 1329.07 9872.53 989 0.90 Temp2 * NDVI2 215.90 10348.53 1303.99 9044.54 994 0.90 Temp2+ NDVI2 29.04 4340.43 1265.60 3074.82 995 0.74 Prec2 * NDVI2 22.15 3619.81 1144.03 2475.77 994 0.71 Temp2 * Prec2 20.57 3437.55 1252.11 2185.44 994 0.66 Temp2 + Prec2 20.34 3410.08 1250.19 2159.89 995 0.66 Temp * NDVI 13.41 2502.76 1122.46 1380.31 996 0.58 Prec2 + NDVI2 13.07 2453.12 1152.25 1300.88 995 0.56 Temp * Prec 13.22 2475.05 1172.98 1302.07 996 0.55 Prec * NDVI 11.85 2270.92 1133.24 1137.68 996 0.53 Temp+Prec+NDVI 11.48 2214.32 1144.45 1069.87 996 0.51 Temp + NDVI 11.42 2205.84 1142.96 1062.88 997 0.51 Prec + NDVI 10.74 2100.61 1144.65 955.96 997 0.48 Seasonal Temp2 * Seasonal Prec2 7.05 1480.45 1108.52 371.93 994 0.27 Seasonal Temp2 + Seasonal Prec2 7.04 1479.37 1108.42 370.95 995 0.27 Seasonal Temp * Seasonal Prec 6.95 1463.85 1107.60 356.26 996 0.26 Seasonal Temp + Seasonal Prec 6.94 1460.59 1107.48 353.11 997 0.26 Temp + Prec 5.37 1169.87 1086.08 83.79 997 0.08

Temp: Annual mean temperature (C) = BIO1 Prec: Annual mean precipitation (mm) = BIO12 NDVI: Normalized Difference Vegetation Index Seasonal Temp: Temperature seasonality = BIO4 (SD*100) Seasonal Prec: Precipitation seasonality = BIO15 (SD*100)

152

Figure S3.1. Spearman’s rank correlation among the 19 covariables used as explanatory variables in the analysis. The maximum and minimum values of climatic covariables were highly correlated (Spearman’s r > 0.98, p <0.0001) with the mean and therefore, excluded from the analysis. Positive (blue) and negative (red) slopes correspond to positive or negative relationships, whereas a more circular (wider and whitish) oval correspond to no correlation.

153

Figure S3.2. Modelled raw species richness pattern (grey-yellow-red) in response to interactions between (a) temperature2*NDVI2, (c) temperature2*precipitation2 and (e) NDVI2 and precipitation2. Correspondently, each pair of climatic variables are plotted along the elevation (b, c, d) with empirical species richness shown in red (red open circles). Annual mean temperature =black dots; NDVI = green crosses; annual mean precipitation =blue crosses. The models were based on the raw data (1000 points) gathered along the elevation gradient.

154

Figure S3.3. Species richness across elevation and in relation to interactions between (a) temperature and primary productivity, (b) temperature and precipitation and (c) precipitation and primary productivity. The 3D surface represents the 95% Kernel contour shell around the points; shading of surfaces are artefacts of the illustration and have no meaning.

155

Figure S3.4. Modeled relationships between species richness and annual mean temperature or precipitation in current and future climatic scenarios.

156

CHAPTER IV

CLIMATIC DRIVERS OF AVIAN SPECIES RICHNESS ACROSS THE ANDES

C. Steven Sevillano-Ríos and Amanda D. Rodewald

ABSTRACT

As the Earth’s climate continues to change, scientist and decision-makers are challenged to predict and mitigate risks to biodiversity. Whereas rising temperatures are expected to shift species distributions poleward and uphill, much less is understood about the effects of changes in precipitation patterns. We studied 16 elevational gradients across the

Andes of Colombia, Ecuador, Peru, Bolivia, Argentina and Chile to (1) evaluate the extent to which current relationships between elevation and avian diversity were explained by interactions between temperature and precipitation, and (2) predict responses to future climatic conditions. Two key patterns emerged, even amid regional variation. In mesic regions, like the Amazonian slopes of the Andes, avian species richness generally declined with elevation and latitude and was best explained by temperature. On the other hand, along xeric slopes (e.g., central-western slopes of the Andes) where precipitation is more limiting, avian richness peaked at mid-elevations and was most closely associated with an interaction between temperature and precipitation. Specifically, the warmer and wetter conditions at mid-elevations seemed to promote species richness in contrast to the warm-but-dry lowlands and wet-but-cold highlands. Based upon four climatic scenarios, mesic slopes are predicted to gain species by 2070, whereas xeric slopes, especially at low elevations that are expected to be both warmer and drier, are expected to lose species. Our findings highlight that efforts to forecast the consequences of climate change using temperature alone can produce

157

misleading results, especially in xeric regions that also seem to be the most vulnerable.

Hence, increasing the number of meteorological stations that quantify precipitation is an important step to improving predictions about climate-induced changes in avian communities.

Key words: Andes, climate change, temperature, precipitation, avian biodiversity, mountain systems.

158

RESUMEN

A medida que el clima de la Tierra continúa cambiando, los científicos y los encargados de tomar decisiones tienen el desafío de predecir y mitigar los riesgos para la biodiversidad.

Mientras que se espera que el aumento de las temperaturas cambie la distribución de las especies hacia los polos y a mayor elevación, mucho menos es entendido sobre los efectos de los cambios en los patrones de precipitación. Estudiamos 16 gradientes de elevación en los Andes de Colombia, Ecuador, Perú, Bolivia, Argentina y Chile para (1) evaluar en qué medida las relaciones actuales entre elevación y diversidad aviar se explicaron por las interacciones entre temperatura y precipitación y (2) predecir respuestas a las condiciones climáticas futuras. Surgieron dos patrones clave, dentro de la variación regional general. En las regiones húmedas como las laderas amazónicas de los Andes, la riqueza de las especies de aves generalmente disminuyó con la elevación y la latitud y se explica mejor por la temperatura. Por otro lado, a lo largo de las laderas xéricas o áridas (por ejemplo, laderas oeste de los Andes Centrales) donde la precipitación es más limitante, la riqueza aviar alcanzó su punto máximo en las elevaciones intermedias y se asoció más estrechamente con una interacción entre la temperatura y la precipitación. Específicamente, con las condiciones más cálidas y húmedas a elevaciones intermedias que contrastan con las tierras bajas cálidas pero secas y las tierras altas húmedas pero frías. Con base en cuatro escenarios climáticos, se pronostica que las laderas húmedas ganarán especies para 2070 en contraste con las pérdidas de especies a lo largo de las laderas xéricas, especialmente en elevaciones bajas donde se espera condiciones más cálidas y secas. Nuestros hallazgos resaltan que los esfuerzos para pronosticar las consecuencias del cambio climático utilizando solo la temperatura pueden producir resultados sesgados, especialmente en regiones xéricas las cuales parecen ser las

159

más vulnerables. Por lo tanto, incrementar el número de estaciones meteorológicas que cuantifiquen la precipitación es clave para comprender y mejorar nuestras predicciones sobre los impactos climáticos en conjunto con los cambios en temperatura.

160

INTRODUCTION

Global patterns in temperature and precipitation are changing in ways that are expected to profoundly affect ecological communities and the ecosystem processes that sustain them (Lovejoy and Hannah 2005; Clarke and Gaston 2006; Worm and Tittensor

2018). Though increasing temperatures are a relatively consistent and widespread phenomenon (Brohan et al., 2006; Trenberth et al., 2007; Post 2013), changes in precipitation are more variable. For example, though precipitation has increased in the northern hemisphere and decreased in tropical/subtropical zones since 1901 (Dore 2005, Gu and Adler 2015), future scenarios suggest that most mesic areas will become wetter and xeric areas dryer (Trenberth 2011).

Shifts in temperature and precipitation can strongly influence biodiversity, by way of mediating primary productivity and, hence, resource availability for consumers (Gaston

2000; Hawkins et al., 2003; Evans et al., 2005; Clarke and Gaston 2006; McCain 2009).

Until recently, most studies of species responses to climate change focused on temperature, often reporting poleward and upward shifts in species distributions (Grabherr et al., 1994,

Parmesan 1996, Pounds et al., 1999, Parmesan and Yohe 2003; Araujo and Rahbek 2006,

Nogués-Bravo et al., 2007, Lenoir et al., 2008, La Sorte and Jetz 2010a, b; Freeman et al.,

2018). However, our understanding of the manner in which changes in precipitation can affect biodiversity remains poor (McCain and Colwell 2011, Tingley et al., 2012).

Mountain systems, which often are hotspots of biodiversity and endemism, are particularly vulnerable to climate change. Fortunately, one of their defining attributes – elevational gradients – provides an excellent opportunity to study the responses of species to changes in climate and, by using comparative approaches, to determine the separate and

161

combined influence of temperature and precipitation on biodiversity (Tingley et al., 2009;

Tingley et al., 2012; Jankowski et al., 2010; Elsen et al., 2017). Elevation gradients are defined, in part, by the close proximity of multiple life zones, including tropical, subtropical, temperate, boreal-artic and snowline, that occur within short distances, often less than 6 km of elevation range and/or 10-km of distance (Körner 2000). The close proximity of these life zones coupled with their shared geological and evolutionary history (Antonelli et al., 2018) provides special insight into ecological responses to climatic changes often without confounding factors (Elsen et al., 2017). Elevational gradients also permit us to distinguish long-term from seasonal shifts (Körner 1999), as well as to correct by land area effects

(Elsen and Tingley 2015), helping to identify drivers of local and global patterns of biodiversity (Gaston 2000; Körner 2000; McCain 2009; Quintero and Jetz 2018).

Research conducted in montane systems to date shows that patterns of diversity are determined neither solely by temperature, nor elevation alone (Clarke and Gaston 2006;

Worm and Tittensor 2018). Instead, several elevational patterns have been described, including peaks at mid-elevations and/or plateaus at low elevations (McCain 2007).

Collectively these patterns have spurred the development of alternative hypotheses, including primary productivity (Wright 1983), environmental stability (Sanders 1968), habitat complexity (Pianka 1966); land area (Connor and McCoy 1979), seasonality (Haffer

1969), mid-domain effects (Colwell and Lees 2000), and the elevational climate model (i.e., interaction between temperature and precipitation; McCain 2007; McCain 2009).

In this study, we evaluated the extent to which relationships between elevation and avian diversity were explained by the interaction of temperature and precipitation along a latitudinal gradient along the Andes and applied this understanding to predict responses to

162

future climate changes. Given that McCain (2009) hypothesized that avian species richness would be positively related to warmer and wetter conditions across the elevation, we predicted the following possible patterns: (1) species richness declines with elevation due to concomitant declines in temperature and precipitation (Figure 4.1a), (2) species richness peaks at low elevations before declining with precipitation (Figure 4.1b), (3) species richness remains relatively constant if precipitation is invariable, (Figure 4.1c), or (4) species richness peaks in areas receiving the greatest precipitation (Figures 4.1d-e), or richness peaks at the point where precipitation and temperature intersect – in other words, peaking at locations with the warmest and wettest conditions across the elevational gradient (Figure

4.1f).

METHODS

Study areas

We studied 16 elevational gradients across the Andes, which included 12 gradients for which we had access to previous surveys of bird communities and 4 gradients that had not been studied previously (Table 4.1; Figure 4.2). In order to include the full elevational gradient within each study area, we established a 220 x 150 km plot with 1,000 random points separated by > 5 km and in rough proportion to the relative amount of land area. For each point, we recorded latitude, longitude and elevation within a 1-km radius plot from

Space Shuttle Radar Topography Mission - Digital Elevation Model (SRTM; 30-m resolution, U. S. Geological Survey) in QGIS (Version 3.0-Girona).

In Colombia, where the Andes are divided in three ranges separated by the valleys of

Cauca and Magdalena rivers, we selected five elevational gradients studied by Salaman et al.

(1999) and Kattan and Franco (2004). Three were located over the Western, Central and 163

Eastern Cordilleras; one over Upper Magdalena Valley; and a last one over the Serrania de la Macarena, Churumbelos and Puerto Bello, which included part of the Amazonian lowlands (Figure 4.2a). In Ecuador, where the Andes form a single Cordillera, we lacked a study to compare. Yet, here we selected a western elevational gradient that represent a transition zone between the northern and central western Andes. In Peru, northern Chile, and

Bolivia, where the western and eastern slopes of the Andes are separated by several small but high inter-valley hills, six elevational gradients were located based on Terborgh 1977;

Patterson et al., 1998; Kessler et al., 2001; Herzog et al., 2005; Forero-Medina et al., 2011;

Araneada et al., 2018; Quispe et al., 2019; Sevillano-Ríos and Rodewald (in Prep).

Arequipa, a site without a comparative study, was included to fill the gap of the western

Andes of southern Peru. Therefore, three sites were placed over each Andean slope, including Ancash, Arequipa and Tarapacá over the western slope and Cerros del Sira, Cuzco and Cochabamba in Amazonian slope, and one in the Central Andes of Junín where the elevation gradient is truncated at 1,250 masl (Figure 4.2b). In Argentina, we selected an elevation gradient in the eastern slope of the Andes, in the province of Salta (Figure 4.2c).

Finally, two last elevational gradients were located in the western slopes of the southern

Andes of Chile, over the regions of De Los Lagos (close to Puerto Montt) and Magallanes y la Antarctica Chilena (close to Puerto Natales) (Figure 4.2d)

Estimated bird species richness

Avian species richness was estimated through three steps. First, we compiled a list of potential birds at a site by extracting distribution maps for 2,384 species from the IUCN red list (Birdlife International 2019, Suplementary Information D), which follows taxonomy used in the Handbook of Birds of the World (HBW) and Birdlife International Illustrated

164

Checklist of the Birds of the World (Birdlife International 2018). We then excluded introduced, vagrant, non-resident, pelagic (e.g. petrels) or widespread boreal/austral migratory species and restricted our analysis to a list of 2,191 species (Appendix A). Next, we compiled the elevational ranges (min-max in meters) of each species based mainly on

Quintero and Jetz (2018) and the IUCN red list for 410 species not included in their work.

Finally, we used the elevational midpoints and ranges to estimate species richness within

100-m elevational bands using the RangeModel software (version 5.0; Colwell 2006; 2008).

We used the R package “rredlist” (Chamberlain 2017) to extract all IUCN data.

Climatic data

We used the mean annual temperatures and precipitation (BIO1 and BIO12) data from the recent version of WorldClim 2 (Fick and Hijmans 2017), which is derived from spatially-interpolated monthly climatic data at high resolution (~1 km2 in the equator) between 1970 to 2000 (Hijmans 2005: Fick and Hijmans 2017). We extracted both covariates within 1-km radius of each of the 1,000 sampling points placed over the 16 sites.

Points randomly placed in the ocean or lakes were excluded.

Analysis

Avian species richness and climatic data were grouped for each 200-m elevation band by calculating the mean values that were then centered and analyzed through a

Generalized Linear Model with normal distribution in program JMP Pro v14 (SAS Institute

Inc., Cary, NC). We estimated the main and interactive effects for temperature and precipitation for each of the 16 sites independently and used the FDR LogWorth (-log10(p- value)) value as a standardized measure of the effect size of each component.

165

Future climatic conditions

We used the climatic data from the four Representative Concentration Pathways

(RCPs) (RCP2.6, RCP4.5, RCP6.0, RCP8.5) adopted by the Intergovernmental Panel on

Climate Change (IPCC) in their Fifth Assessment Report (Allen et al., 2014) to assess future climatic change along the elevation. Each RCP defines a specific emissions trajectory and subsequent radiative forcing defined as a measure of influence on energy fluxes (watts/m2) in the Earth’s atmosphere by 2100 (Moss et.al. 2010). Hence, each RCP represents different scenarios of greenhouse gas concentration, with RCP8.5 (8.5 w/m2) considered the most severe (Riahi et al., 2011). For each survey point, we extracted the predicted main annual temperature and precipitation (BIO1 and BIO12) from the Hadley Global Environment

Model 2 for Earth System (HadGEM2-ES, ~1-km resolution) for the year 2070 (Jones et al.,

2011). We used LOESS models to characterize each climatic covariate along the elevation gradient and used the difference between current (WorldClim 2) and future (HadGEM2-ES) models to estimate the expected degree of change. We used the coefficients of the model obtained in the previous analysis to predict future species richness responses under the four scenarios over the 16 sites. LOESS models were used to assess the degree of change in species richness along elevation.

RESULTS

Avian species richness along the Andes

Patterns of avian species richness along the elevational gradients varied among regions, though richness most commonly decreased with increasing elevation and latitude. In particular, species richness declined with elevation along eastern slopes (Figure 4.3a), the central Andes of Colombia and Ecuador (Figure 4.3b) and southern Andes (Figure 4.3f). In 166

other regions, species richness peaked at mid-elevations. Although richness initially plateaued at low elevations along the western slopes of Colombia (Figure 4.3c) and central

Peru (Figure 4.3d), this mid-elevation patter was greater in the central-western Andes of

Peru and Chile (Figure 4.3e).

Temperature and precipitation patterns

Temperature and precipitation showed dissimilar relationships with elevation; temperature declined with elevation whereas precipitation gradients were highly variable

(Figures 4.4 and 4.5).

Climatic- species richness relationship

Avian species richness was related to both precipitation and temperature, but relationships varied across the Andes. Along most of the mesic slopes of the Andes temperature was the most important driver (Figure 4.6; Table 4.2), with little signal from precipitation which remained relatively uniform or high across the gradient (Figure 4.5). In contrast, along the xeric western slopes, where richness peaked at mid-elevation, the effect of temperature was modulated by precipitation in a way that was maximized at moderated levels of precipitation (Figure 4.6 in orange; Table 4.2).

Predictions under future climate conditions

Striking changes in both temperature and precipitation are forecast for the entire

Andean mountain system (Figures S4.1a). Temperatures should rise evenly across an elevation gradient in most locations, but with notable exceptions in the central Andes of

Junín (Locality=9; greater change at low elevations), and along dry western slopes

(Localities= 7,11, 13, 16; greater change at mid to high elevations). Precipitation is expected 167

to increase in most places, especially in the northern Andes of Colombia and Ecuador

(Figures S4.1b; Localities= 1:6). Declines in precipitation, on the other hand, are projected for the lowlands of Cusco and the areas close to Puerto Montt (Localities= 10, 15). Our models of species richness based on future conditions suggest that mesic Andean slopes should gain species, unlike the western dry Andean slopes along which richness is likely to decline at lower elevations and increase at mid-upper elevations (> 3,000 m) (Figure 4.7).

DISCUSSION

We evaluated the extent to which relationships between elevation and avian diversity were explained by the interaction of temperature and precipitation along a latitudinal gradient in the Andes and applied this understanding to predict responses to future climate changes. Though temperature alone was sufficient to explain patterns in species richness along many slopes in our system, we also detected complex interactions between temperature and precipitation, especially in xeric regions. In most regions, richness peaked at elevations where conditions were warmest and wettest (Figure 4.5), thereby supporting the elevational climate model (McCain 2007; McCain 2009).

Previous studies described a general reduction in avian richness with elevation across the

Andes (Terborgh 1971, 1977, 1985; Terborgh and Weske 1975; Patterson et al., 1996;

Patterson et al., 1998; Kattan and Franco 2004; Remsen 1985; Kessler et al., 2001; Küper et al., 2004; Krӧmer et al., 2005; Herzog et al., 2005; Jankowski et al., 2013; Quintero and Jetz

2018). However, the majority of these studies occurred within mesic regions such as the

Amazonian slope and, as such, do not broadly represent the Andes. To our knowledge, only two other studies (Araneada et al., 2018 in Tarapacá and Sevillano-Ríos and Rodewald (in

168

prep) in Ancash), examined these patterns in the western Andes and both similarly detected mid-elevation patterns of avian richness. If we integrate findings from across the Andes, species richness most commonly declines with elevation (Quintero and Jetz 2018) except for xeric regions of the central-western Andes of Peru and Chile, where richness peaks at mid- elevations. One caveat is that previous studies have overwhelmingly focused on more pristine and less-impacted regions (Terborgh 1971, 1977, 1985; Terborgh and Weske 1975;

Remsen 1985; Patterson et al., 1996; Patterson et al., 1998), as has been proposed that human activities may already modify these patterns in most parts of the world (Nogués-

Bravo et al., 2008). Yet, given the fact that the central-western Andes has been occupied by humans for thousands of years, it seems like climate regimes are the central drivers of avian patterns across the Andes and, while the influence of temperature was relatively consistent, precipitation was much more important in xeric than mesic regions.

Temperature is known to drive macroevolutionary and macroecological patterns of biodiversity (Currie et al., 1991; Erwin 2009; Mayhew et al., 2012), including those along elevational gradients (Janzen 1967; Grytnes and McCain 2007; McCain 2007; Freeman et al., 2018). In general, abundance and richness of species in montane areas is positively related to temperature through a variety of mechanisms (e.g. increased mutation rate; Brown et al., 2004; Evans and Gaston 2005; Clarke and Gaston 2006). However, warm temperatures alone do not necessarily promote biodiversity (e.g. hot deserts are normally species poor), as water can become limiting. Consistent with this idea, we found that so long as sufficient water was available in xeric regions, species richness was greatest where temperatures were warm. We suggest that our failure to detect a signal of precipitation in mesic regions stemmed from the high rates of precipitation along those slopes, which may

169

indicate that water was not limiting. Indeed, in mesic regions, such as in the Amazonian slope of the Andes, the annual mean precipitation is generally high (~800 - 6000 mm; Figure

4.5) and unlikely to limit the distribution of many species. In xeric regions however, precipitation ranges between ~0 – 1,000 mm and does not necessarily decline with elevation, leading to warm but dry lowlands and wet but cold highlands. The same effects of precipitation have been demonstrated experimentally (see Wu et al., 2011), with total net primary productivity and plant growth responding positively to the combination of wet and warm conditions (Wu et al., 2011).

Implications for forecasting effects of climate change

Regional variation in current and future temperature and precipitation regimes precludes broad generalizations about future impacts of climate change on species richness in the

Andes. Our analyses suggest that future climatic conditions in mesic slopes could favor species richness (Figure 4.7 and Figure S4.1). At higher elevations, such as in the Magdalena

Valley in Colombia, species richness is expected to rise sharply in response to increasing temperature and precipitation (Figure 4.7), and possibly facilitate colonization by lowland species (Tingley et al., 2009). However, this can produce the local extinction of mountaintop species in response of “the escalator effect” (Marris 2007) and the reduction of land with elevation (Elsen and Tingley 2015). Along the xeric slopes of the central-western Andes, species richness is projected to decline at low elevations, where conditions are expected to become warmer and dryer, but increase at mid-elevations (Figure 4.7). Although more research is required to fully understand future changes, species-habitat mismatches or mountain-top extinctions may occur as if generalist species outcompete or otherwise negatively affect specialists at high elevations (Sevillano-Rios et al., 2019). 170

A large number of studies have attempted to understand how biodiversity will respond to the global increase of temperature without considering changes in precipitation (e.g.

Parmesan 1996; Pounds et al., 2006; Lenoir et al., 2008; La Sorte and Jetz 2010a, b; La Sorte and Jetz 2012). Fortunately, a growing body of research has begun to explicitly consider the joint effects of temperature and precipitation (Wu et al., 2011; McCain and Colwell 2011;

Bellard et al., 2012; Tingley et al., 2012) along with land use changes (Mantyka‐Pringle et al., 2012; Oliver and Morecroft 2014; Segan et al., 2016; Peters et al., 2019). Our findings underscore the importance of simultaneously considering these two climatic drivers, especially in xeric regions where precipitation is likely to be a limiting factor in the next 50 years.

ACKNOWLEDGMENTS

This research was funded by FONDECYT-CONCYTEC (Nº 237-2015-FONDECYT).

We are also grateful to the Fulbright Scholarship and support from Cornell University that supported CSSR during his graduate studies. We also deeply appreciate financial support provided by Cornell Lab of Ornithology, the Athena Grant, the Bergstrom Award of the

Association of Field Ornithologists as well as the American Alpine Club Research Grants

2017. We thank to André Dhondt, Stephen Morreale, Viviana Ruíz-Gutiérrez, Wesley M.

Hochachka, Lynn Marie Johnson and Laura Morales for comments, recommendations and guidance that helped to improve this manuscript.

171

REFERENCES

Allen, M. R., Barros, V. R., Broome, J., Cramer, W., Christ, R., Church, J. A., ... & Edenhofer, O.

(2014). IPCC fifth assessment synthesis report-climate change 2014 synthesis

report. Intergovernmental Panel on Climate Change, Geneva, Switzerland.

Antonelli, A., Kissling, W. D., Flantua, S. G., Bermúdez, M. A., Mulch, A., Muellner-Riehl, A. N.,

... & Fritz, S. A. (2018). Geological and climatic influences on mountain biodiversity. Nature

Geoscience, 11(10), 718.

Araneda, P., Sielfeld, W., Bonacic, C., & Ibarra, J. T. (2018). Bird diversity along elevational

gradients in the Dry Tropical Andes of northern Chile: The potential role of Aymara

indigenous traditional agriculture. PloS one, 13(12), e0207544.

Araújo, M. B., & Rahbek, C. (2006). How does climate change affect biodiversity?.

Science, 313(5792), 1396-1397.

Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., & Courchamp, F. (2012). Impacts of climate

change on the future of biodiversity. Ecology letters, 15(4), 365-377.

Brohan, P., Kennedy, J. J., Harris, I., Tett, S. F., & Jones, P. D. (2006). Uncertainty estimates in

regional and global observed temperature changes: A new data set from 1850. Journal of

Geophysical Research: Atmospheres, 111(D12).

Brown J. H., Gillooly J. J., Allen A. P., et al. (2004) Toward a metabolic theory of ecology. Ecology

85:1771-1789.

Chamberlain, S. (2017). rredlist:'IUCN'Red List Client. R package version 0.4. 0.

Clarke, A., & Gaston, K. J. (2006). Climate, energy and diversity. Proceedings of the Royal Society

B: Biological Sciences, 273(1599), 2257-2266.

Clarke, A., & Gaston, K. J. (2006). Climate, energy and diversity. Proceedings of the Royal Society

B: Biological Sciences, 273(1599), 2257-2266.

172

Colwell, R. K. (2006). RangeModel: a Monte Carlo simulation tool for assessing geometric

constraints on species richness. Version, 5, 10.

Colwell, R. K. (2008). RangeModel: tools for exploring and assessing geometric constraints on

species richness (the mid‐domain effect) along transects. Ecography, 31(1), 4-7.

Colwell, R. K., & Lees, D. C. (2000). The mid-domain effect: geometric constraints on the

geography of species richness. Trends in ecology & evolution, 15(2), 70-76.

Connor, E. F., & McCoy, E. D. (1979). The statistics and biology of the species-area relationship.

The American Naturalist, 113(6), 791-833.

Currie, D. J. (1991). Energy and large-scale patterns of animal-and plant-species richness. The

American Naturalist, 137(1), 27-49.

Dore, M. H. (2005). Climate change and changes in global precipitation patterns: what do we

know?. Environment international, 31(8), 1167-1181.

Elsen, P. R., & Tingley, M. W. (2015). Global mountain topography and the fate of montane species

under climate change. Nature Climate Change, 5(8), 772.

Elsen, P. R., Tingley, M. W., Kalyanaraman, R., Ramesh, K., & Wilcove, D. S. (2017). The role of

competition, ecotones, and temperature in the elevational distribution of Himalayan

birds. Ecology, 98(2), 337-348.

Erwin, D. H. (2009). Climate as a driver of evolutionary change. Current Biology, 19(14), R575-

R583.

Evans, K. L., & Gaston, K. J. (2005). Can the evolutionary‐rates hypothesis explain species‐energy

relationships?. Functional Ecology, 19(6), 899-915.

Evans, K. L., Warren P. H., and Gaston, K. J. (2005). Species-energy relationship at the

macroecological scale: a review of the mechanisms. Biological Reviews 80:1-25.

173

Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: new 1‐km spatial resolution climate surfaces for

global land areas. International journal of climatology, 37(12), 4302-4315.

Forero-Medina, G., Terborgh, J., Socolar, S. J., & Pimm, S. L. (2011). Elevational ranges of birds on

a tropical montane gradient lag behind warming temperatures. PLoS One, 6(12), e28535.

Freeman, B. G., Scholer, M. N., Ruiz-Gutierrez, V., & Fitzpatrick, J. W. (2018). Climate change

causes upslope shifts and mountaintop extirpations in a tropical bird community.

Proceedings of the National Academy of Sciences, 115(47), 11982-11987.

Gaston, K. J. (2000). Global patterns in biodiversity. Nature, 405(6783), 220.

Grabherr, G., Gottfried, M., & Pauli, H. (1994). Climate effects on mountain

plants. Nature, 369(6480), 448.

Grytnes, J. A., & McCain, C. M. (2007). Elevational trends in biodiversity. Encyclopedia of

biodiversity, 2, 1-8.

Gu, G., & Adler, R. F. (2015). Spatial patterns of global precipitation change and variability during

1901–2010. Journal of Climate, 28(11), 4431-4453.

Haffer, J. (1990). Avian species richness in tropical South America. Studies on Neotropical Fauna

and Environment, 25(3), 157-183.

Hawkins, B. A., Field, R., Cornell, H. V., Currie, D. J., Guégan, J. F., Kaufman, D. M., ... & Porter,

E. E. (2003). Energy, water, and broad‐scale geographic patterns of species

richness. Ecology, 84(12), 3105-3117.

Herzog, S. K., Kessler, M., & Bach, K. (2005). The elevational gradient in Andean bird species

richness at the local scale: a foothill peak and a high‐elevation plateau. Ecography, 28(2),

209-222.

174

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 of Climatology: A

Journal of the Royal Meteorological Society, 25(15), 1965-1978.

Jankowski J. E, Merkord CL, Rios W. F, Cabrera K. G, Revilla N. S, Silman M. R (2013) The

relationship of tropical bird communities to tree species composition and vegetation structure

along an Andean elevational gradient. Journal of Biogeography 40: 950-962.

Jankowski, J. E., Robinson, S. K., & Levey, D. J. (2010). Squeezed at the top: interspecific

aggression may constrain elevational ranges in tropical birds. Ecology, 91(7), 1877-1884.

Janzen, D. H. (1967). Why mountain passes are higher in the tropics. The American

Naturalist, 101(919), 233-249.

Jones, C., Hughes, J. K., Bellouin, N., Hardiman, S. C., Jones, G. S., Knight, J., ... & Boo, K. O.

(2011). The HadGEM2-ES implementation of CMIP5 centennial simulations. Geoscientific

Model Development, 4(3), 543-570.

Kattan G. H, & Franco P. (2004) Bird diversity along elevational gradients in the Andes of

Colombia: area and mass effects. Global Ecology and Biogeography 135: 451-458.

doi:10.1111/j.1466-822X.2004.00117.x

Kessler M, Herzog SK, Fjeldså J, Bach K (2001) Species richness and endemism of plant and bird

communities along two gradients of elevation humidity and land use in the Bolivian Andes.

Diversity and distributions 7: 61-77

Körner, C. (2000). Why are there global gradients in species richness? Mountains might hold the

answer. Trends in Ecology & Evolution, 15(12), 513-514.

Krömer, T. , Kessler, M. , Robbert Gradstein, S. and Acebey, A. (2005), Diversity patterns of

vascular epiphytes along an elevational gradient in the Andes. Journal of Biogeography, 32:

1799-1809. doi:10.1111/j.1365-2699.2005.01318.x

175

Küper, W. , Kreft, H. , Nieder, J. , Köster, N. and Barthlott, W. (2004), Large‐scale diversity patterns

of vascular epiphytes in Neotropical montane rain forests. Journal of Biogeography, 31:

1477-1487. doi:10.1111/j.1365-2699.2004.01093.x

La Sorte, F. A., & Jetz, W. (2010a). Projected range contractions of montane biodiversity under

global warming. Proceedings of the Royal Society B: Biological Sciences, 277(1699), 3401-

3410.

La Sorte, F. A., & Jetz, W. (2010b). Avian distributions under climate change: towards improved

projections. Journal of Experimental Biology, 213(6), 862-869.

La Sorte, F. A., & Jetz, W. (2012). Tracking of climatic niche boundaries under recent climate

change. Journal of Animal Ecology, 81(4), 914-925.

Lenoir, J., Gégout, J. C., Marquet, P. A., De Ruffray, P., & Brisse, H. (2008). A significant upward

shift in plant species optimum elevation during the 20th century. science, 320(5884), 1768-

1771.

Lovejoy, T. E. and Hannah, L. J. (Eds.). (2005). Climate change and biodiversity. Yale university

press.

Marris, E. (2007). The escalator effect. Nature Reports Climate Change, 1, 94-96.

Mantyka‐pringle, C. S., Martin, T. G., & Rhodes, J. R. (2012). Interactions between climate and

habitat loss effects on biodiversity: a systematic review and meta‐analysis. Global Change

Biology, 18(4), 1239-1252.

Mayhew, P. J., Bell, M. A., Benton, T. G., & McGowan, A. J. (2012). Biodiversity tracks

temperature over time. Proceedings of the National Academy of Sciences, 109(38), 15141-

15145.

McCain, C. M. (2007). Could temperature and water availability drive elevational species richness

patterns? A global case study for bats. Global Ecology and biogeography, 16(1), 1-13.

176

McCain, C. M. (2009). Global analysis of bird elevational diversity. Global Ecology and

Biogeography, 18(3), 346-360.

McCain, C. M., & Colwell, R. K. (2011). Assessing the threat to montane biodiversity from

discordant shifts in temperature and precipitation in a changing climate. Ecology

letters, 14(12), 1236-1245.

Moss, R. H., Edmonds, J. A., Hibbard, K. A., Manning, M. R., Rose, S. K., Van Vuuren, D. P., ... &

Meehl, G. A. (2010). The next generation of scenarios for climate change research and

assessment. Nature, 463(7282), 747.

Nogués-Bravo, D., Araújo, M. B., Errea, M. P., & Martinez-Rica, J. P. (2007). Exposure of global

mountain systems to climate warming during the 21st Century. Global environmental

change, 17(3-4), 420-428.

Oliver, T. H., & Morecroft, M. D. (2014). Interactions between climate change and land use change

on biodiversity: attribution problems, risks, and opportunities. Wiley Interdisciplinary

Reviews: Climate Change, 5(3), 317-335.

Parmesan, C. (1996). Climate and species' range. Nature, 382(6594), 765.

Parmesan, C., & Yohe, G. (2003). A globally coherent fingerprint of climate change impacts across

natural systems. Nature, 421(6918), 37.

Patterson B. D, Stotz DF, Solari S, Fitzpatrick J. W, Pacheco V (1998) Contrasting patterns of

elevational zonation for birds and mammals I n the Andes of southeastern Peru Journal of

Biogeography 25(3) 593-607

Patterson B. D, Stotz DF, Solari S, Fitzpatrick J. W, Pacheco V (1998) Contrasting patterns of

elevational zonation for birds and mammals I n the Andes of southeastern Peru Journal of

Biogeography 25(3) 593-607

177

Patterson, B. D., Pacheco, V., & Solari, S. (1996). Distribution of bats along an elevational gradient

in the Andes of south‐eastern Peru. Journal of Zoology, 240(4), 637-658.

Peters, M. K., Hemp, A., Appelhans, T., Becker, J. N., Behler, C., Classen, A., ... & Gebert, F.

(2019). Climate–land-use interactions shape tropical mountain biodiversity and ecosystem

functions. Nature, 568(7750), 88.

Pianka, E. R. (1966). Convexity, desert lizards, and spatial heterogeneity. Ecology, 47(6), 1055-1059.

Post, E. (2013). Ecology of climate change: the importance of biotic interactions (Vol. 68). Princeton

University Press.

Pounds, J. A., Fogden, M. P., & Campbell, J. H. (1999). Biological response to climate change on a

tropical mountain. Nature, 398(6728), 611.

QGIS Development Team, 2018. QGIS Geographic Information System. Open Source Geospatial

Foundation. URL http://qgis.osgeo.org

Quintero, I., & Jetz, W. (2018). Global elevational diversity and diversification of

birds. Nature, 555(7695), 246.

Quispe-Melgar H, Sevillano-Rios CS, Navarro-Romo WC, Ames-Martinez FN, Camel-Paucar V,

Fjeldsa J and Kessler M (2019) The Central Andes of Peru: a key area for the conservation

of Polylepis forest biodiversity. Journal of Ornithology. https ://doi.org/10.1007/s1033 6-

019-01703 -5)

Remsen Jr, J. V. (1985). Community organization and ecology of birds of high elevation humid

forest of the Bolivian Andes. Ornithological Monographs, 733-756.

Riahi, K., Rao, S., Krey, V., Cho, C., Chirkov, V., Fischer, G., ... & Rafaj, P. (2011). RCP 8.5—A

scenario of comparatively high greenhouse gas emissions. Climatic Change, 109(1-2), 33.

Salaman, P. G., Donegan, M., & Cuervo, M. (1999). Ornithological surveys in Serrania de los

Churumbelos, southern Colombia.

178

Sanders, H. L. (1968). Marine benthic diversity: a comparative study. The American Naturalist,

102(925), 243-282.

Segan, D. B., Murray, K. A., & Watson, J. E. (2016). A global assessment of current and future

biodiversity vulnerability to habitat loss–climate change interactions. Global Ecology and

Conservation, 5, 12-21.

Sevillano-Rios CS, Rodewald A, & Morales LV. (2019). Alpine Birds of South America. Earth

Systems and Environmental Sciences. Encyclopedia of World's Biomes.

Terborgh J (1977) Bird species diversity on an Andean elevational gradient. Ecology 1007-1019

Terborgh, J. (1971). Distribution on environmental gradients: theory and a preliminary interpretation

of distributional patterns in the avifauna of the Cordillera Vilcabamba, Peru. ECOLOGY,

52(1), pp.23-40.

Terborgh, J. (1985). The role of ecotones in the distribution of Andean birds. ECOLOGY, 66(4),

pp.1237-1246.

Terborgh, J. and Weske, J.S. (1975). The role of competition in the distribution of Andean birds.

ECOLOGY, 56(3), pp.562-576.

Tingley, M. W., Koo, M. S., Moritz, C., Rush, A. C., & Beissinger, S. R. (2012). The push and pull

of climate change causes heterogeneous shifts in avian elevational ranges. Global Change

Biology, 18(11), 3279-3290.

Tingley, M. W., Monahan, W. B., Beissinger, S. R., & Moritz, C. (2009). Birds track their

Grinnellian niche through a century of climate change. Proceedings of the National Academy

of Sciences, 106(Supplement 2), 19637-19643.

Trenberth, K.E., P.D. Jones, P. Ambenje, R. Bojariu, D. Easterling, A. Klein Tank, D. Parker, F.

Rahimzadeh, J.A. Renwick, M. Rusticucci, B. Soden and P. Zhai, 2007: Observations:

Surface and Atmospheric Climate Change. In: Climate Change 2007: The Physical Science

Basis. Contribution of Working Group I to the Fourth Assessment Report of the

179

Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen,

M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press,

Cambridge, United Kingdom and New York, NY, USA.Trenberth, K. E. (2011). Changes in

precipitation with climate change. Climate Research, 47(1-2), 123-138.

Worm, B., & Tittensor, D. P. (2018). A Theory of Global Biodiversity (MPB-60) (Vol. 79). Princeton

University Press.

Wright, D. H. (1983). Species-energy theory: an extension of species-area theory. Oikos, 496-506.

Wu, Z., Dijkstra, P., Koch, G. W., Peñuelas, J., & Hungate, B. A. (2011). Responses of terrestrial

ecosystems to temperature and precipitation change: a meta‐analysis of experimental

manipulation. Global Change Biology, 17(2), 927-942.

180

TABLES

Table 4.1. Names, elevational ranges and country were the 16 elevational gradients under study are located. Previous bird species inventories along the elevation are given as references.

ID Name Country Elevation range (m) Reference 1 Central Cordillera Colombia 100 – 3,700 Kattan and Franco 2004 2 Eastern Cordillera Colombia 190 – 4,700 Kattan and Franco 2004 3 Western Cordillera Colombia 0 – 3,900 Kattan and Franco 2004 4 Magdalena Valley Colombia 200 – 5,200 Kattan and Franco 2004 5 Puerto Bello & Colombia 220 – 3,330 Salaman et al., 1999; Serranía de la Kattan and Franco Macarena 2004 6 Quito Ecuador 0 – 4,900 NA 7 Ancash Peru 0 – 6,768 Sevillano-Rios and Rodewald 2019 8 Cerros del Sira Peru 160 – 2,700 Terborgh 1969; Forero-Medina et al., 2011 9 Junín Peru 1,250 – 5,000 Quispe et al., 2019 10 Cuzco (Vilcabamba Peru 270 – 5,750 Terborgh 1969; and Manu) Patterson et al. 1998 11 Arequipa Peru 120 – 5,750 NA 12 Cochabamba Bolivia 200 – 4,800 Kessler et al., 2001; Herzog et al., 2005 13 Tarapaca Chile 650 – 5,300 Araneada et al., 2018 14 Salta Argentina 550 – 5,150 NA 15 Puerto Montt Chile 0 – 2,100 NA 16 Puerto Arenas Chile 0 – 2,800 Vuilleumier 1972

181

Table 4.2. Main and interactive effects of temperature and precipitation defining species richness pattern along sixteen elevational gradients across the Andes.

Locality Term Mean ± SD Wald X2 p-value Central Cordillera (1) Intercept -480.43 ± 64.6 55.31 <.0001 Temperature 8.47 ± 1.36 39.05 <.0001 Precipitation 0.27 ± 0.03 67.81 <.0001 (Temp-17.8011)*(Prec- 2382.47) 0.01 ± 0 7.55 0.006 Eastern Cordillera (2) Intercept -38.1 ± 19.68 3.75 0.0528 Temperature 18.28 ± 3.48 27.64 <.0001 Precipitation 0.04 ± 0.04 1.15 0.2826 (Temp-14.4344)*(Prec- 1810.04) -0.01 ± 0 12.47 0.0004 Western Cordillera (3) Intercept -113.26 ± 28.74 15.53 <.0001 Temperature 34.06 ± 2.57 175.28 <.0001 Precipitation -0.07 ± 0.03 4.62 0.0315 (Temp-16.752)*(Prec-2498.06) 0 ± 0 0.15 0.6945 Magdalena Valley (4) Intercept -362.03 ± 70.98 26.01 <.0001 Temperature 14.55 ± 1.74 69.97 <.0001 Precipitation 0.3 ± 0.06 26.05 <.0001 (Temp-14.4472)*(Prec-1573.3) -0.01 ± 0 4.34 0.0372 Puerto Bello & Serranía de la Macarena (5) Intercept -163.71 ± 77.83 4.42 0.0354 Temperature 35.47 ± 3.32 113.84 <.0001 Precipitation -0.07 ± 0.07 1.03 0.311 (Temp-17.523)*(Prec-1943.81) 0.01 ± 0.01 2.8 0.0941 Quito (6) Intercept -47.42 ± 7.56 39.32 <.0001 Temperature 15.02 ± 1.2 156.85 <.0001 Precipitation 0.07 ± 0.02 18.01 <.0001 (Temp-13.3185)*(Prec- 1309.68) 0 ± 0 9.21 0.0024 Ancash (7) Intercept 20.6 ± 61.24 0.11 0.7366 Temperature 4.13 ± 2.68 2.38 0.1231

182

Precipitation 0.08 ± 0.06 1.86 0.1721 (Temp-10.9919)*(Prec- 503.795) 0.01 ± 0 54.93 <.0001 Cerros del Sira (8) -224.95 ± Intercept 110.87 4.12 0.0425 Temperature 16.87 ± 7.24 5.42 0.0199 Precipitation 0.15 ± 0.03 19.72 <.0001 (Temp-21.8523)*(Prec- 1730.47) -0.01 ± 0.01 1.4 0.2373 Junín (9) Intercept 64.4 ± 36.44 3.12 0.0771 Temperature 14.75 ± 0.85 302.38 <.0001 Precipitation -0.06 ± 0.05 1.4 0.2371 (Temp-12.6853)*(Prec- 914.434) -0.01 ± 0.01 2.3 0.1294 Cuzco (Vilcabamba and Manu) (10) Intercept -63.38 ± 9.55 44.04 <.0001 Temperature 17.69 ± 1.53 133.5 <.0001 Precipitation 0.06 ± 0.02 6.38 0.0116 (Temp-12.7773)*(Prec- 1512.56) 0 ± 0 0.1 0.7571 Arequipa (11) Intercept 130.93 ± 40.27 10.57 0.0011 Temperature -1.15 ± 2.22 0.27 0.6065 Precipitation -0.02 ± 0.08 0.1 0.7537 (Temp-11.6072)*(Prec- 215.844) 0.02 ± 0 86.94 <.0001 Cochabamba (12) Intercept -105.51 ± 7.14 218.11 <.0001 Temperature 24.36 ± 0.87 787.45 <.0001 Precipitation 0.03 ± 0.01 13.63 0.0002 (Temp-15.9357)*(Prec- 1536.14) 0 ± 0 6.11 0.0135 Tarapaca (13) Intercept 0.43 ± 28.71 0 0.988 Temperature 5.27 ± 1.89 7.79 0.0052 Precipitation 0.49 ± 0.21 5.35 0.0207 (Temp-9.25196)*(Prec- 64.8062) 0.07 ± 0.01 66.98 <.0001 Salta (14) Intercept 4.04 ± 2.74 2.17 0.1403 Temperature 12.65 ± 0.39 1039.02 <.0001

183

Precipitation -0.02 ± 0.02 1.02 0.3135 (Temp-11.5052)*(Prec- 295.478) 0.01 ± 0 8.31 0.0039 Puerto Montt (15) Intercept 78.91 ± 2.43 1057.84 <.0001 Temperature 5.4 ± 0.89 37.13 <.0001 Precipitation 0 ± 0.01 0.61 0.4332 (Temp-7.00673)*(Prec- 1493.94) 0 ± 0 10.38 0.0013 Puerto Arenas (16) Intercept 73.52 ± 10.62 47.89 <.0001 Temperature 4.84 ± 0.26 350.92 <.0001 Precipitation 0.01 ± 0.01 1.54 0.2146 (Temp-0.5369)*(Prec-1532.29) 0 ± 0 4.6 0.0321

184

FIGURES

Figure 4.1. The interaction of temperature and precipitation is expected to define the number of species along the elevational gradient. Temperature=Solid lines, Precipitation= Dashed lines, Species richness = grey gradient (darker is greater species richness).

185

186

Figure 4.3. Patterns of species richness along elevational gradients in the Andes. Species richness (a) decreased constantly with elevation in all the Amazonian slopes and (b) presented a slight low-elevation plateau in the northern Central Andes of Colombia and Ecuador. (c) A low elevation peak was observed in the Cordillera Central and Occidental of Colombia as well as in the Central Andes of Peru (d); which occurred at the base of the sampled plot. (e) A mid- elevation peak around 4,000 m was observed on the western slopes of the central Andes whereas in the southern (f), species richness decreased again with elevation. The colors on top of the six figures (a-f) correspond to the group of study sites (plots, 1-16) with similar species richness patterns across the Andes (map in the right).

187

188

Figure 4.5. Climatic (temperature=black and precipitation=blue) and avian species richness patterns (black vertical lines and beige to red (more species) gradient) over 16 Andean elevational gradients. Climatic values correspond to random points ~1,000 per site and showed contrasting patterns along the elevation gradient. While temperature always decreased constantly with elevation and maintained similar values for most part of the Andes, precipitation was more erratic and no single pattern was dominant; remaining constant (a, d, e), increasing (g, k, m), decreasing (b, l, o), decreasing drastically before remaining constant (c, h, n), or showing mixture patterns (f, I, p). Elevations (x-axes) could differ across figures as they represent a unique elevation range (min - max) present on each elevation gradient (See Table 4.1).

189

190

Figure 4.6. Determinants of avian species richness along the Andes. Relative importance plots of LogWorth coefficients of predictor variables: TEMP=Temperature, PREC=Precipitation, TEMP*PREC= Temperature* Precipitation (the direction of effect is indicated as + or -). LogWorth above 2 (vertical dashed line) corresponds to a p-value below 0.01. The colors and this specific order correspond to the species richness pattern show in Figure 4.3. Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito área = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16.

191

Figure 4.7. Difference (modelled – actual) in species richness in response to changes in temperature and precipitation predicted by 2070 according to the four RCPs climatic models. Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito area = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16.

192

APPENDICES

Figure S4.1. Expected changes of the annual mean (a) temperature and (b) precipitation along the elevation by 2070. Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito área = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16. (a)

193

(b) Future precipitación patterns across Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito área = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16.

194

Figure S4.2. Modelled species richness in response to changes on temperature and precipitation predicted according to the four RCPs climatic models by 2070. Cordillera Central = 1, Eastern Cordillera = 2, Western Cordillera =3, Magdalena Valley = 4, Puerto Bello and Montes de la Macarena =5, Quito área = 6, Ancash =7, Cerros del Sira =8, Junín = 9, Cuzco (Vilcabamba and Manu) = 10, Arequipa = 11, Cochabamba = 12, Tarapacá = 13, Salta = 14, Puerto Montt = 15, Puerto Natales = 16.

195

Figure S4.3. Climatic (temperature (black) and precipitation (blue)) and avian species richness patterns over 16 elevational gradients across the Andes.

196

197

198

APPENDICES

CHAPTER II: SUPLEMENTARY INFORMATION A. 1.- R script code of the Dorazio/Royley Multispecies Occupancy Model structure used in the analysis. 10.6084/m9.figshare.9598517

SUPLEMENTARY INFORMATION B. Multispecies Occupancy Model (MSOM) structure, species-specific occupancy estimates and parameters values, and relationships against landscape composition covariates. 10.6084/m9.figshare.9598214

2.- Table S1. Dorazio/Royley Multispecies Occupancy Model structure used in the analysis.

3.- Table S2. Marginal species-specific occupancy estimates and parameters values (Mean, SD, 95% PI). E= Endemic, EN=Endangered, VU=Vulnerable, NT=Near threatened, ζ = moderately specialists, ζ ζ = Polylepis specialists.

4.- Table S3. Summary of the number of species observed (Sobs) and estimated (Sest) for each one of the 59 Polylepis patches survey during the study. The results are based on Hill number estimation evaluated through a measure of survey completeness (SC).

CHAPTER III SUPLEMENTARY INFORMATION C. Dataset used in the analysis. 10.6084/m9.figshare.9598694

CHAPTER IV SUPLEMENTARY INFORMATION D. Lower and upper elevational limits and Raw data base extracted from 16 elevational gradients. 10.6084/m9.figshare.9598709

5.- Table S1. Lower and upper elevational limits of 2191 species recorded among 16 plots across the Andes. Taxonomic names and ranges are primary based on IUCN redlist 2019 ranges and Quintero & Jetzs 2018.

6.- Table S2. Raw data base extracted from 16 elevational gradients (~1000 points/site) across the Andes.

199

Ash-breasted Tit-tyrant (Anairetes alpinus) Endangered (EN) Ilustrator: Katterine Sandoval

200