Protected areas (PAs) cover 12.9% of Earth’s land, while just 5.8% has strict protection for biodiversity (Earth’s variety of ecosystems, species, and genetic variation). Constraints of size and configuration, mismanagement, anthropogenic pressure, and climate change hamstring the capacity of PAs to conserve biodiversity. Increasingly, studies of biodiversity in human-modified landscapes provide an evidence base to support policies to make land outside of PAs as amenable as possible for biodiversity persistence. I reviewed research on biodiversity in sub-Saharan Africa’s human-modified landscapes within four ecosystem categorizations: rangelands, tropical forest, Cape Floristic Region, and urban and rural built environment. I found potential for humanmodified landscapes to contribute to conservation across ecosystems. Available research could guide policy-making; nonetheless, several issues require further investment, e.g. research deficiencies, implementation strategies, and conflict with biodiversity. I also conducted case studies that could support land-use planning in South Africa’s coastal forest, part of a biodiversity hotspot. By comparing herpetofaunal communities over a land-use gradient, I found old-growth forest harbored the highest richness and abundance. Richness was low in sugar cane cultivation and degraded forest but substantial in acacia woodland and eucalyptus plantation. Composition differed between natural and anthropogenic vegetation types. Functional group richness decreased monotonically along the gradient, driven by sensitivity of fossorial herpetofauna and vegetation-dwelling frogs. Environmental variables were good predictors of frog abundance, but less so for reptiles. Maintaining forest and preventing degradation is important for herpetofaunal conservation while restoration and plantations have more value than cultivation. Old-growth remnants and post-disturbance regenerating vegetation also provide habitat for birds. However, occurrence does not ensure persistence. I calculated population trends for 37 bird species and general trends in overall bird density in different vegetation types. Seventy-six percent of species assessed have declined, 57% significantly so at an average rate of 13.9% per year. Overall, bird density fell at 12.2% per year across vegetation types. Changes in rainfall, habitat area, and survey coverage may partly explain trends. However, species with larger range extents declined more sharply than others and may be responding to environmental changes on a broad scale. These results cast doubt on the future persistence of birds in this human-modified landscape and justify further study. Such studies can support sensible land-use management; however, biases in study topics should not lead to gaps in the evidence base. By reviewing the global literature, I demonstrated clear geographical bias among biomes and geopolitical regions and taxonomic bias among species groups. Furthermore, distribution of published papers did not generally reflect threats of low PA coverage, high land conversion, and high human population density. Forests were the subject of 87% of papers, and 75% focused on the Americas and Europe, while Africa and Asia were critically understudied. This thesis highlights that managing human-modified landscapes for biodiversity could contribute to conservation. However, responses to land uses are complex, locationand species-specific, and often poorly understood, hindering integration of information into policy recommendations. Further research is needed to elucidate what, where, and how biodiversity persists alongside humans to enhance conservation efficacy, especially in understudied regions. / Thesis (PhD)--University of Pretoria, 2014. / gm2014 / Zoology and Entomology / unrestricted

O conhecimento da ecologia das espécies em ecossistemas em constante transformação se faz necessário para entender como essas persistem, como utilizam os novos habitats criados pelas atividades humanas, e se ainda desempenham funções ecológicas, informações essenciais para subsidiar a conservação dos ecossistemas. A presente tese teve como objetivos: 1) avaliar por meio da dieta da onça-parda (Puma concolor), se essa pode ocupar o nicho da onça-pintada (Panthera onca) onde essa está funcionalmente ausente, comparando a dieta de ambas as espécies; 2) analisar mudanças no uso dos recursos, habitat e estrutura trófica de assembleias de mamíferos entre paisagens conservadas e modificadas da Mata Atlântica por meio da análise de isótopos estáveis (SIA) de carbono e nitrogênio; 3) identificar alterações na efetividade funcional de assembleias de mamíferos em relação a estrutura da paisagem na Mata Atlântica. Para determinar a dieta da onça-parda, coletamos amostras fecais em duas áreas no maior contínuo florestal de Mata Atlântica, identificando 15 tipos de presas. Observamos uma preferência da onça-parda por grandes mamíferos, superior a observada em outras áreas do bioma. Comparada a dieta da onça-pintada, a porcentagem de grandes mamíferos também foi superior. Portanto, a onça-parda pode atuar como equivalente funcional da onça-pintada onde essa está funcionalmente ausente, e há disponibilidade de grandes presas. Para o estudo com a SIA, utilizamos os pelos dos mamíferos para análise, coletados em paisagens conservadas e modificadas. Para comparação das paisagens, classificamos os mamíferos em guildas tróficas e utilizamos fatores de fracionamento para corrigir os valores isotópicos. Observamos uma grande mudança no uso dos recursos pelos mamíferos entre as paisagens, com predominância de recursos C3 nas conservadas, e maior proporção de C4 nas modificadas. A estrutura trófica nas paisagens conservadas foi clara, com enriquecimento escalonado de 15N, enquanto desordenado nas modificadas, com enriquecimento flutuante. Destacamos que a matriz agrícola desempenha um papel importante como fonte de recursos e habitat para mamíferos resilientes, incluindo espécies ameaçadas, sendo seu manejo imprescindível para a conservação das espécies. Para o estudo da efetividade funcional, criamos dois bancos de dados, um de assembleias de mamíferos, e outro para determinar a contribuição das espécies nas funções. Selecionamos 10 funções tróficas para análise, classificadas em vulneráveis (espécies sensíveis) e persistentes (espécies resilientes). Para cada assembleia calculamos o nível de defaunação e cinco variáveis de paisagem, utilizando a Modelagem Hierárquica de Comunidades para análise dos dados, e extrapolando os resultados para toda Mata Atlântica. A riqueza de espécies, a massa corporal média e as funções vulneráveis apresentaram relação positiva com aumento na proporção de habitat, e negativa para os usos antrópicos do solo, sendo o inverso para as funções persistentes, similar ao padrão dos níveis de defaunação. As funções vulneráveis ficaram restritas aos grandes blocos florestais, áreas com elevada diversidade de espécies, destacando sua importância para manutenção das funções ecológicas. As paisagens modificadas podem desempenhar um importante papel na manutenção das funções, principalmente se conectadas a grandes blocos florestais. / The knowledge about species ecology in ever changing ecosystems is necessary to understand how they persist, how they used the new habitats molded by human activities, and if they are still performing ecological functions; this kind of information is essential to subsidize the conservation of ecosystem. The objectives of this thesis are: 1) to determine if the puma (Puma concolor) can occupy the niche of the jaguar (Panthera once) where it is functionally absent, by assessing the puma\' feeding habits and comparing it to the jaguar; 2) analyze changes in resource and habitat use, and trophic structure of mammal assemblages in preserved and modified landscapes of the Atlantic Forest, Brazil, by using the analysis of carbon and nitrogen stable isotopes; 3) identify changes in the functional effectiveness of mammal assemblages in response to changes in landscape structure within the Atlantic Forest. To determine the diet of the puma, we collected fecal samples in two areas of the largest Atlantic Forest continuum and identified 15 prey taxa. We observed that pumas consumed preferentially large prey, a proportion superior to other areas in the biome. Compared to the jaguar\' diet, the proportion of large prey was also higher. Thus, the puma may at as a functional equivalent where the jaguar is functionally absent, and there is availability of large prey. In the study with isotopic ecology, we used mammal\' hair for analysis, which were collected in preserved and modified landscapes. To compare these areas, we classified mammals in trophic guilds and corrected isotopic values using species-specific fractionation factors. We observed a huge difference in mammals\' resource use, with predominant use of C3 resource in preserved landscapes, and a higher incorporation of C4 carbon in the modified ones. The trophic structure was clear in preserved landscapes, with an orderly 15N enrichment, while unordered in modified landscapes, with floating enrichment. We highlight that the agricultural matrix plays an important role as source of food items and as habitat for resilient mammals, including threatened species, and its management is essential for species conservation. To study the functional effectiveness of mammal species, we created two databases, one with mammal assemblages, and other to assess the contribution of species in ecological functions. We selected 10 trophic functions for analysis, which were classified in vulnerable (performed by sensitive species) and persistent (performed by resilient species). For each assemblage we calculated the defaunation level and five landscape variables. We analyze the data using the Hierarchical Modelling of Species Communities and extrapolate the results for the entire Atlantic Forest. Species richness, body mass and vulnerable functions showed a positive relationship with increasing habitat amount, while negative for anthropogenic land uses, similar to the defaunation pattern; the inverse was observed for persistent functions. Vulnerable functions were restricted to the large forest block, which have high species diversity, highlighting its importance for ecological functions maintenance. Human-modified landscapes may still perform an important role for functions persistence, especially if connected to the largest forest blocks.

Understanding the impacts of land-use change on biodiversity and the ecosystem services that it provides is of great importance given unprecedented growth of the human population. Past studies attempting to explore these impacts have described the overall structure of communities (i.e. species richness and trait diversity) across gradients of local scale degradation and fragmentation, and have sought to identify whether the loss of species following land-use change is non-random. Yet, despite a wealth of research we still lack a generalised understanding of how land-use change impacts on traits responsible for determining species sensitivity and their role within ecosystems, particularly for vertebrates. Moreover, despite the importance of niche-based processes in the assembly of communities, we have not yet elucidated whether these are important in mediating the collapse of communities in human-dominated landscapes. To fill these existing research gaps, I collated comprehensive avian species inventories from fragmented and degraded forests and compared their structure with communities existing in continuous forests. In Chapter 2, I tested whether sensitivity of species to forest fragmentation varies between the temperate zone and the tropics and whether there are key differences in the size of fragments required to maintain ecosystem processes in these regions. I found that sensitivity to fragmentation varies according to functional group and body mass, with the prevalence of insectivores and large frugivores declining in relation to fragment size, particularly in tropical fragments smaller than 100 ha. In Chapter 3, I tested whether functional diversity and the mean position of trait diversity of insectivores and frugivores, changed across a gradient of intensifying land-use change. I found a decline in the functional diversity of forest species and a shift in the mean community traits for both forest and non-forest species. In Chapter 4, I tested whether the structure of tropical bird communities are influenced by species interactions in a fragmented landscape. I found increasing over-dispersion in functional and phylogenetic trait relatedness among species with decreasing fragment size, suggesting that competitive interactions are important in the disassembly of avian communities. In Chapter 5, I modelled the impact of forest cover change on ecosystem function across the Brazilian Amazon, focusing on seed dispersal by birds. Furthermore, I tested whether ecosystem function declined linearly with decreased forest cover after accounting for differences in the underlying pools of species. I found the lowest levels of functional diversity along the southern arc of deforestation and that the dispersal of large seeds showed some resilience to declining forest cover. Taken together, my results suggest that the loss of species from communities in degraded and fragmented landscapes is strongly non-random. Insectivores and large frugivores are most sensitive to land-use change, with species located in the densest parts of trait space being most threatened by a decline in forest patch size, suggesting that species interactions regulate the collapse of avian diversity in human-modified forests. I conclude that land-use change has important implications for the provisioning of ecosystem services, including seed dispersal and the control of insect herbivores. The impact of future land-use change is likely to be mediated by the composition of the original pool of species and the amount of redundancy in the ecosystem services that they provide. I discuss the relevance of my findings to land-use management strategies and policy interventions, and in particular conclude that these should, where possible, maintain pristine forest patches above 1000 ha, improve connectivity among habitat patches, and ensure greater protection for logged and burnt forests. Future studies should focus on clarifying the link between shifts in vertebrate community structure and the functioning of forest ecosystems.

No description available.

A expansão das atividades humanas está associada a várias ameaças antrópicas que afetam a biodiversidade tropical, especialmente a perda e fragmentação de habitat, o continuado crescimento populacional humano e a expansão de infraestrutura, como estradas e rodovias. Embora seja esperado que as ameaças antrópicas interajam e criem mosaicos de manchas de habitat de diferentes qualidades, há ainda poucos estudos sobre os efeitos aditivos e interativos de diferentes ameaças sobre a biodiversidade ou sobre a capacidade de espécies nativas de usar remanescentes de diferentes qualidades. De fato, a maioria dos estudos focaram em apenas uma ou poucas ameaças isoladas, na escala do fragmento em vez da escala da paisagem e não consideraram as variações na qualidade dos remanescentes. Mamíferos de maior porte são um bom modelo de estudo porque possuem um conjunto de características que os tornam particularmente vulneráveis a várias ameaças antrópicas. Dada a dificuldade de obter dados sobre estas espécies elusivas em escalas espaciais adequadas, entrevistas com moradores locais vêm sendo cada vez mais usadas para acessar a distribuição de mamíferos de maior porte, e podem fornecer informações confiáveis. Através da estimativa da ocorrência de 15 espécies de mamíferos de maior porte por meio de entrevistas com o chefe de 12 unidades domésticas em cada uma de 20 paisagens em uma região de pós-fronteira de 1 milhão ha na Amazônia, investigamos: (i) que tipo de habitat e em qual escala espacial está associado com a ocorrência das espécies, e (ii) os efeitos, a importância relativa e as interações de quatro ameaças antrópicas - cobertura de habitat, fragmentação de habitat, densidade populacional humana e densidade de estrada - sobre a persistência das espécies. Dados de presença/ausência nos arredores de 227 unidades domésticas foram analisados usando modelos lineares generalizados mistos e seleção de modelos baseada no Critério de Informação de Akaike em duas etapas. A persistência das espécies maiores e ameaçadas não foi necessariamente afetada pela qualidade do habitat, mas elas responderam majoritariamente a cobertura de habitat em escalas maiores. A chance de persistência de todas as espécies - mesmo das duas menores, que não responderam à perda de habitat isoladamente - foi afetada por alguma combinação das quatro ameaças antrópicas. Efeitos aditivos entre as ameaças antrópicas foram mais importantes do que as interações entre elas na determinação da ocorrência de mamíferos de maior porte. Enquanto os efeitos da perda de habitat foram mais fortes do que os efeitos da fragmentação de habitat per se, a densidade de estradas foi tão importante quanto à perda de habitat para a ocorrência das espécies. Por último, não houve um padrão claro em termos do grau de ameaça e do tamanho corpóreo por trás da reposta dos mamíferos de maior porte a ameaças antrópicas combinadas. Nosso estudo sugere que áreas protegidas em terras públicas ou privadas na Amazônia devem ser grandes para assegurar a persistência de mamíferos de maior porte, mas podem incluir mosaicos de florestas primárias e secundárias. Os resultados também ressaltam a necessidade de considerar os impactos acumulados de múltiplas ameaças simultaneamente - em especial, as consequências da expansão da malha viária - em planejamentos e manejos, de maneira a evitar subestimar a chance de extinções. Evitar que paisagens na Amazônia se tornem muito desmatadas e alteradas é fundamental, dado que a persistência até mesmo das espécies mais comuns pode ser prejudicada, afetando um dos serviços ecossistêmicos mais importantes para moradores locais - a carne de caça - e potencialmente erodindo o valor que estes atribuem às florestas / The expansion of human activities is associated to a myriad of anthropogenic threats that affect tropical biodiversity, especially habitat loss and fragmentation, the continued human population growth and the expansion of infrastructure, as roads and highways. Although anthropogenic threats are expected to interact and to create mosaics of patches of varying quality, their additive and interaction effects on biodiversity, as well as the extent native species are able to use remnants of distinct quality, have yet been poorly studied. Indeed most studies focused on only one or few threats in isolation, on the patch rather than the landscape scale and did not considered the varying quality of remnants. Large mammals are a good study model as they have a set of traits that make them particularly vulnerable to several anthropogenic threats. Due to the difficulties in gathering data on these elusive species at adequate spatial scales, interviews with local residents have increasingly being used to access their distribution and can provide reliable information. By estimating the occurrence of 15 large mammal species through interviews with the head of 12 households within each of 20 landscapes across a 1-million ha post-frontier region in Amazonia, we investigated: (i) which habitat type at which spatial scale is associated with species occurrence, and (ii) the effects, relative importance and interactions of four anthropogenic threats - habitat cover, habitat fragmentation, human population density and road density - on species persistence. Presence/absence data across the surroundings of 227 households was analyzed using generalized linear mixed-effects models, and a two-step model selection based on Akaike’s Information Criteria. The persistence of larger and endangered species was not necessarily affected by habitat quality, but they responded to habitat cover mostly at larger spatial scales. The chance of persistence of all species - even the two smallest that were not affected by habitat loss alone - was disrupted by some combination of the four anthropogenic threats. Additive effects between anthropogenic threats were more important than their interaction in determining the occurrence of large mammals. While the effects of habitat loss were stronger than the effects of habitat fragmentation per se, road density was as important as habitat loss to species occurrence. Finally, there was no clear pattern in threat status or body size underlying the response of large mammals to combined anthropogenic threats. Our study suggest that protected areas in public or private lands in Amazonia should be large to secure the persistence of large mammals, but may include mosaics of primary and secondary forest. The findings also highlight the need to take into account the accumulated impacts of multiple threats simultaneously - particularly, the consequences of the expansion of the road network - in planning and management, as a way to avoid underestimating the chance of extinctions. Preventing Amazonian landscapes to become heavily deforested and altered is critical, as the persistence of even common species can be impaired, affecting one of the most important ecosystem services provided to local residents - bushmeat - and potentially eroding the value people attribute to forests

Fauna conservation outside protected areas can make an important complementary contribution to conservation within reserves. This thesis aimed to contribute new information and analytical frameworks to the science of fauna conservation in human-modified landscapes. Two approaches were used: (1) empirical data collection and analysis, and (2) the discussion and development of conceptual landscape models. ¶ Empirical work focused on lizard distribution patterns in two production landscapes in southeastern Australia. Lizards were targeted because ectotherms are frequently neglected by conservation biologists. The “Nanangroe grazing landscape” was used for sheep and cattle grazing. In this landscape, approximately 85% of pre-European woodland cover had been cleared, and understorey vegetation was sparse. Lizards were surveyed at 16 landscape units, which were stratified by aspect, topographic position and amount of tree cover. Each landscape unit contained three sites, and each site contained three plots. Regression modelling showed that different species responded differently to their environment. For example, the four-fingered skink (Carlia tetradactyla) and Boulenger’s skink (Morethia boulengeri) were more likely to occur at woodland sites with northerly aspects, whereas the striped skink (Ctenotus robustus) and olive legless lizard (Delma inornata) were more likely to inhabit sites with a simple microhabitat structure. Statistical analysis further showed that the habitat attributes that lizards were related to varied continuously through space, and over different spatial scales. For example, invertebrate abundance (a proxy for food availability) varied most strongly over tens of metres, whereas the amount of grass cover varied most strongly over hundreds to thousands of metres. Thus, work at Nanangroe revealed spatially complex patterns of lizard occurrence and habitat variables. ¶ The “Tumut plantation landscape” was a spatial mosaic of native eucalypt (Eucalyptus) forest patches embedded within a plantation of the introduced radiata pine (Pinus radiata). In this landscape, thirty sites were surveyed for lizards. Sites were stratified by forest type and patch size, and included eucalypt patches, pine sites, and extensive areas of eucalypt forest adjacent to the plantation. Regression modelling showed that lizard species responded to various habitat attributes, including elevation, the amount of eucalypt forest within 1 km of a site, invertebrate abundance and ground cover. Variables related to habitat fragmentation often were significant predictors of lizard occurrence. However, work at Tumut suggested that important additional insights into lizard distribution patterns could be obtained by considering variables related to food and shelter resources, and climatic conditions. ¶ The Nanangroe and Tumut landscapes were in close proximity, but together spanned an altitudinal gradient of 900 m. An investigation of changes in lizard community composition with altitude showed that (1) only one species was common to Nanangroe and Tumut, (2) different species had different altitudinal preferences, and (3) ecologically similar species replaced one another with increasing altitude. These results highlighted that even in highly modified landscapes, natural gradients (such as climate) can play an important role in shaping animal assemblage composition and species distribution patterns. ¶ Empirical work suggested that, in some landscapes, the frequently used “fragmentation model” is a relatively weak conceptual basis for the study of animal distribution patterns. The fragmentation model implicitly assumes that “habitat patches” can be defined unequivocally across many species, and that patches are located within a relatively inhospitable matrix. Where these assumptions are breached, conservation guidelines arising from the fragmentation model may be too simplified. In spatially complex production landscapes, it may be more appropriate to maintain habitat heterogeneity at multiple spatial scales than to focus solely on the management of large, pre-defined patches. ¶ Given the potential limitations of the fragmentation model, a new, more holistic landscape model was developed. The “continuum model” was derived from continuum theory as developed for plant ecology. The continuum model recognises (1) spatial continua of environmental variables, and (2) species’ individualistic responses to these variables. For animals, key environmental variables may be related to the availability of food, shelter, sufficient space, and suitable climatic conditions. Unlike the fragmentation model, the continuum model is inherently process-based and thus may help to link the perceived gap between patterns and processes in landscape ecology. ¶ Three general conclusions arise from this thesis: 1. Some heterogeneous production landscapes support many native species, and therefore represent important conservation opportunities. 2. In some modified landscapes, the fragmentation model does not capture the complexity of animal distribution patterns. In those landscapes, conservation recommendations derived from the fragmentation model may be overly simplistic. 3. The continuum model may be a useful extension of the fragmentation model. It provides a process-based conceptual basis for empirical work on animal distribution patterns.

A expansão das atividades humanas está associada a várias ameaças antrópicas que afetam a biodiversidade tropical, especialmente a perda e fragmentação de habitat, o continuado crescimento populacional humano e a expansão de infraestrutura, como estradas e rodovias. Embora seja esperado que as ameaças antrópicas interajam e criem mosaicos de manchas de habitat de diferentes qualidades, há ainda poucos estudos sobre os efeitos aditivos e interativos de diferentes ameaças sobre a biodiversidade ou sobre a capacidade de espécies nativas de usar remanescentes de diferentes qualidades. De fato, a maioria dos estudos focaram em apenas uma ou poucas ameaças isoladas, na escala do fragmento em vez da escala da paisagem e não consideraram as variações na qualidade dos remanescentes. Mamíferos de maior porte são um bom modelo de estudo porque possuem um conjunto de características que os tornam particularmente vulneráveis a várias ameaças antrópicas. Dada a dificuldade de obter dados sobre estas espécies elusivas em escalas espaciais adequadas, entrevistas com moradores locais vêm sendo cada vez mais usadas para acessar a distribuição de mamíferos de maior porte, e podem fornecer informações confiáveis. Através da estimativa da ocorrência de 15 espécies de mamíferos de maior porte por meio de entrevistas com o chefe de 12 unidades domésticas em cada uma de 20 paisagens em uma região de pós-fronteira de 1 milhão ha na Amazônia, investigamos: (i) que tipo de habitat e em qual escala espacial está associado com a ocorrência das espécies, e (ii) os efeitos, a importância relativa e as interações de quatro ameaças antrópicas - cobertura de habitat, fragmentação de habitat, densidade populacional humana e densidade de estrada - sobre a persistência das espécies. Dados de presença/ausência nos arredores de 227 unidades domésticas foram analisados usando modelos lineares generalizados mistos e seleção de modelos baseada no Critério de Informação de Akaike em duas etapas. A persistência das espécies maiores e ameaçadas não foi necessariamente afetada pela qualidade do habitat, mas elas responderam majoritariamente a cobertura de habitat em escalas maiores. A chance de persistência de todas as espécies - mesmo das duas menores, que não responderam à perda de habitat isoladamente - foi afetada por alguma combinação das quatro ameaças antrópicas. Efeitos aditivos entre as ameaças antrópicas foram mais importantes do que as interações entre elas na determinação da ocorrência de mamíferos de maior porte. Enquanto os efeitos da perda de habitat foram mais fortes do que os efeitos da fragmentação de habitat per se, a densidade de estradas foi tão importante quanto à perda de habitat para a ocorrência das espécies. Por último, não houve um padrão claro em termos do grau de ameaça e do tamanho corpóreo por trás da reposta dos mamíferos de maior porte a ameaças antrópicas combinadas. Nosso estudo sugere que áreas protegidas em terras públicas ou privadas na Amazônia devem ser grandes para assegurar a persistência de mamíferos de maior porte, mas podem incluir mosaicos de florestas primárias e secundárias. Os resultados também ressaltam a necessidade de considerar os impactos acumulados de múltiplas ameaças simultaneamente - em especial, as consequências da expansão da malha viária - em planejamentos e manejos, de maneira a evitar subestimar a chance de extinções. Evitar que paisagens na Amazônia se tornem muito desmatadas e alteradas é fundamental, dado que a persistência até mesmo das espécies mais comuns pode ser prejudicada, afetando um dos serviços ecossistêmicos mais importantes para moradores locais - a carne de caça - e potencialmente erodindo o valor que estes atribuem às florestas / The expansion of human activities is associated to a myriad of anthropogenic threats that affect tropical biodiversity, especially habitat loss and fragmentation, the continued human population growth and the expansion of infrastructure, as roads and highways. Although anthropogenic threats are expected to interact and to create mosaics of patches of varying quality, their additive and interaction effects on biodiversity, as well as the extent native species are able to use remnants of distinct quality, have yet been poorly studied. Indeed most studies focused on only one or few threats in isolation, on the patch rather than the landscape scale and did not considered the varying quality of remnants. Large mammals are a good study model as they have a set of traits that make them particularly vulnerable to several anthropogenic threats. Due to the difficulties in gathering data on these elusive species at adequate spatial scales, interviews with local residents have increasingly being used to access their distribution and can provide reliable information. By estimating the occurrence of 15 large mammal species through interviews with the head of 12 households within each of 20 landscapes across a 1-million ha post-frontier region in Amazonia, we investigated: (i) which habitat type at which spatial scale is associated with species occurrence, and (ii) the effects, relative importance and interactions of four anthropogenic threats - habitat cover, habitat fragmentation, human population density and road density - on species persistence. Presence/absence data across the surroundings of 227 households was analyzed using generalized linear mixed-effects models, and a two-step model selection based on Akaike’s Information Criteria. The persistence of larger and endangered species was not necessarily affected by habitat quality, but they responded to habitat cover mostly at larger spatial scales. The chance of persistence of all species - even the two smallest that were not affected by habitat loss alone - was disrupted by some combination of the four anthropogenic threats. Additive effects between anthropogenic threats were more important than their interaction in determining the occurrence of large mammals. While the effects of habitat loss were stronger than the effects of habitat fragmentation per se, road density was as important as habitat loss to species occurrence. Finally, there was no clear pattern in threat status or body size underlying the response of large mammals to combined anthropogenic threats. Our study suggest that protected areas in public or private lands in Amazonia should be large to secure the persistence of large mammals, but may include mosaics of primary and secondary forest. The findings also highlight the need to take into account the accumulated impacts of multiple threats simultaneously - particularly, the consequences of the expansion of the road network - in planning and management, as a way to avoid underestimating the chance of extinctions. Preventing Amazonian landscapes to become heavily deforested and altered is critical, as the persistence of even common species can be impaired, affecting one of the most important ecosystem services provided to local residents - bushmeat - and potentially eroding the value people attribute to forests

In spite tropical forests are the most important terrestrial global carbon sinks due to carbon storage in aboveground biomass, it is also the primary target of deforestation. The conversion of Tropical forests into anthropogenic areas might disrupt biological flux and also lead to severe microclimatic changes at forest edges. These combined effects can trigger profound changes in plant composition through both high mortality of fragmentation-sensitive species and proliferation of disturbed-adapted species which will ultimately impacts carbon storage. Thus, our main objective in this study was understand the role of human-induced disturbances in modulate the dimension of biomass loss at tropical forests. We applied a systematic literature review searching for empirical evidences that edge effects can drive biomass loss in tropical forests (Chapter 2). Our findings highlighted the gap of knowledge about the pattern and process related to biomass loss in tropical forests. To strengthen this understanding, we formulated a conceptual model linking landscape structure and patch-level attributes to severity of edge effects affecting aboveground biomass. Our model hypothesizes that habitat amount, isolation, time since edge creation, and the synergism between edge distance, patch size, and matrix type are the main drivers of biomass loss in anthropogenic tropical forests. We thus used a large plant dataset (18 503 trees ≥ 10 cm dbh) from 146 sites distributed across four Mexican and four Brazilian rainforest regions to test our conceptual model predictions, specifically the influence of forest cover, site isolation, edge distance, patch size and type of matrix on biomass (Chapter 3). We observed that carbon-rich sites presented species that are typical of old-growth forests (shade-tolerant, large-seeded, zoocoric) contrasting to carbon-poor sites composed by disturbed-adapted species (pioneer occupying the understory). Large shade-tolerant trees (≥ 40 cm dbh) were impacted severely by the combination of forest loss and edge effects. Edge distance, patch size, and the amount of open-matrix strongly influence small shade-tolerant trees (≤ 20 cm dbh). Although our results do not fully corroborate the initial predictions of the conceptual model, they support the idea that landscape composition interact with patch structure and ultimately impacts biomass stocks in fragmented tropical forests. Finally, we further investigated if the disturbance level of the region influences plant-structure responses to forest loss (Chapter 4). Biomass, but not plant density, was affected by forest loss in regions with intermediate disturbance levels, i.e. regions showing a combination of moderate deforestation (20-40% of remaining forest cover) disturbed during the past 30-60 years, high defaunation but harboring relictual populations of large-mammals, and areas mostly composed by heterogeneous matrices. In general, our findings highlight that both landscape composition and patch structure are the main drivers of biomass loss in Neotropical forests, and that the landscape context must be considered to obtain more reliable estimations of carbon emissions due to forest degradation. Landscape planning (e.g. restoration of forest cover) should be included in conservation strategies in order to sustain carbon storage. Moreover, we advocate that conservation initiatives will be less costly and more effective if implemented in areas under intermediate disturbance levels / Apesar das florestas tropicais serem a mais importante fonte mundial de carbono da porção terrestre do globo devido ao armazenamento de carbono na biomassa acima do solo, elas são também o alvo primário do desmatamento. A conversão das florestas Tropicais em áreas antropogênicas pode interromper o fluxo biológico e também levar a severas mudanças microclimáticas na borda dos fragmentos. A combinação desses efeitos pode engatilhar profundas mudanças na composição da vegetação através tanto da mortalidade de espécies sensíveis à fragmentação como também pela proliferação de espécies adaptadas distúrbios, com impactos finais nos estoques de carbono. Assim, o maior objetivo desse estudo foi compreender o papel dos distúrbios induzidos pelo homem na modulação da dimensão da perda de biomassa em florestas Tropicais. Nós aplicamos uma revisão sistemática da literatura procurando por evidências empíricas de que o efeito de borda pode levar a perda de biomassa em florestas tropicais (Capítulo 2). Nossos resultados destacam a lacuna de conhecimento entre padrões e processos relacionados à perda de biomassa em florestas Tropicais. Para fortalecer esse conhecimento, nós formulamos um modelo conceitual conectando estrutura da paisagem e atributos na escala do fragmento à severidade do efeito de borda, e assim afetando a biomassa acima do solo. Nosso modelo hipotetiza que a quantidade de hábitat, o isolamento, o tempo desde a formação da borda e o sinergismo entre tamanho do fragmento, distância da borda e tipo de matriz são os principais condutores de perda de biomassa em florestas Tropicais antropogênicas. Utilizando um grande banco de dados (18 503 árvores ≥ 10 cm dap) provenientes de 146 locais distribuídos em quatro regiões de floresta úmida no México e quatro no Brasil, nós então testamos as predições do nosso modelo conceitual. Especificamente, a influência da cobertura florestal, isolamento, distância da borda, tamanho do fragmento e tipo de matriz sobre a biomassa (Capítulo 3). Nós observamos que áreas com muito carbono apresentaram espécies típicas de florestas maduras (tolerantes ao sombreamento, zoocóricas, com sementes grandes) contrastando com áreas com pouco carbono compostas por espécies adaptadas à distúrbio (pioneiras ocupando o sub-bosque). Árvores grandes tolerantes ao sombreamento (≥ 40 cm dap) foram impactadas severamente pela combinação de perda de cobertura florestal e efeitos de borda. Distância da borda, tamanho do fragmento e a extensão da área de matriz aberta influenciaram fortemente as árvores pequenas tolerantes a sombreamento (≤ 20 cm dap). Apesar dos nossos resultados não corroborarem completamente as predições iniciais do nosso modelo conceitual, eles dão suporte à ideia de que a composição da paisagem interage com a estrutura do fragmento com impactos finais nos estoques de biomassa em florestas Neotropicais. Por fim, nós investigamos se o nível de distúrbio da região pode influenciar nas respostas da estrutura da vegetação à perda de cobertura florestal. Biomassa, mas não a densidade de indivíduos, foi afetada pela perda de cobertura florestal em regiões com nível intermediário de distúrbio, i.e. regiões apresentando uma combinação de níveis moderados de desmatamento (20-40% de cobertura florestal remanescente) em que a perturbação ocorreu ao longo dos últimos 30-60 anos, com alto grau de defaunação mas ainda abrigando populações relictuais de grandes mamíferos e, em sua maioria, compostos por uma matriz heterogênea. Em geral, nossos resultados destacaram que tanto a composição da paisagem como a estrutura do fragmento são os principais condutores de perda de biomassa em florestas Neotropicais e que o contexto da paisagem deve ser considerado para se obter estimativas mais confiáveis de emissão de carbono devido à degradação florestal. O planejamento da paisagem (e.g. restauração da cobertura florestal) deve ser incluído em estratégias de conservação em ordem de sustentar o armazenamento de carbono. Além disso, nós defendemos que iniciativas de conservação serão menos custosas e mais efetivas se implementadas em áreas sob níveis intermediários de distúrbio

In spite tropical forests are the most important terrestrial global carbon sinks due to carbon storage in aboveground biomass, it is also the primary target of deforestation. The conversion of Tropical forests into anthropogenic areas might disrupt biological flux and also lead to severe microclimatic changes at forest edges. These combined effects can trigger profound changes in plant composition through both high mortality of fragmentation-sensitive species and proliferation of disturbed-adapted species which will ultimately impacts carbon storage. Thus, our main objective in this study was understand the role of human-induced disturbances in modulate the dimension of biomass loss at tropical forests. We applied a systematic literature review searching for empirical evidences that edge effects can drive biomass loss in tropical forests (Chapter 2). Our findings highlighted the gap of knowledge about the pattern and process related to biomass loss in tropical forests. To strengthen this understanding, we formulated a conceptual model linking landscape structure and patch-level attributes to severity of edge effects affecting aboveground biomass. Our model hypothesizes that habitat amount, isolation, time since edge creation, and the synergism between edge distance, patch size, and matrix type are the main drivers of biomass loss in anthropogenic tropical forests. We thus used a large plant dataset (18 503 trees ≥ 10 cm dbh) from 146 sites distributed across four Mexican and four Brazilian rainforest regions to test our conceptual model predictions, specifically the influence of forest cover, site isolation, edge distance, patch size and type of matrix on biomass (Chapter 3). We observed that carbon-rich sites presented species that are typical of old-growth forests (shade-tolerant, large-seeded, zoocoric) contrasting to carbon-poor sites composed by disturbed-adapted species (pioneer occupying the understory). Large shade-tolerant trees (≥ 40 cm dbh) were impacted severely by the combination of forest loss and edge effects. Edge distance, patch size, and the amount of open-matrix strongly influence small shade-tolerant trees (≤ 20 cm dbh). Although our results do not fully corroborate the initial predictions of the conceptual model, they support the idea that landscape composition interact with patch structure and ultimately impacts biomass stocks in fragmented tropical forests. Finally, we further investigated if the disturbance level of the region influences plant-structure responses to forest loss (Chapter 4). Biomass, but not plant density, was affected by forest loss in regions with intermediate disturbance levels, i.e. regions showing a combination of moderate deforestation (20-40% of remaining forest cover) disturbed during the past 30-60 years, high defaunation but harboring relictual populations of large-mammals, and areas mostly composed by heterogeneous matrices. In general, our findings highlight that both landscape composition and patch structure are the main drivers of biomass loss in Neotropical forests, and that the landscape context must be considered to obtain more reliable estimations of carbon emissions due to forest degradation. Landscape planning (e.g. restoration of forest cover) should be included in conservation strategies in order to sustain carbon storage. Moreover, we advocate that conservation initiatives will be less costly and more effective if implemented in areas under intermediate disturbance levels / Apesar das florestas tropicais serem a mais importante fonte mundial de carbono da porção terrestre do globo devido ao armazenamento de carbono na biomassa acima do solo, elas são também o alvo primário do desmatamento. A conversão das florestas Tropicais em áreas antropogênicas pode interromper o fluxo biológico e também levar a severas mudanças microclimáticas na borda dos fragmentos. A combinação desses efeitos pode engatilhar profundas mudanças na composição da vegetação através tanto da mortalidade de espécies sensíveis à fragmentação como também pela proliferação de espécies adaptadas distúrbios, com impactos finais nos estoques de carbono. Assim, o maior objetivo desse estudo foi compreender o papel dos distúrbios induzidos pelo homem na modulação da dimensão da perda de biomassa em florestas Tropicais. Nós aplicamos uma revisão sistemática da literatura procurando por evidências empíricas de que o efeito de borda pode levar a perda de biomassa em florestas tropicais (Capítulo 2). Nossos resultados destacam a lacuna de conhecimento entre padrões e processos relacionados à perda de biomassa em florestas Tropicais. Para fortalecer esse conhecimento, nós formulamos um modelo conceitual conectando estrutura da paisagem e atributos na escala do fragmento à severidade do efeito de borda, e assim afetando a biomassa acima do solo. Nosso modelo hipotetiza que a quantidade de hábitat, o isolamento, o tempo desde a formação da borda e o sinergismo entre tamanho do fragmento, distância da borda e tipo de matriz são os principais condutores de perda de biomassa em florestas Tropicais antropogênicas. Utilizando um grande banco de dados (18 503 árvores ≥ 10 cm dap) provenientes de 146 locais distribuídos em quatro regiões de floresta úmida no México e quatro no Brasil, nós então testamos as predições do nosso modelo conceitual. Especificamente, a influência da cobertura florestal, isolamento, distância da borda, tamanho do fragmento e tipo de matriz sobre a biomassa (Capítulo 3). Nós observamos que áreas com muito carbono apresentaram espécies típicas de florestas maduras (tolerantes ao sombreamento, zoocóricas, com sementes grandes) contrastando com áreas com pouco carbono compostas por espécies adaptadas à distúrbio (pioneiras ocupando o sub-bosque). Árvores grandes tolerantes ao sombreamento (≥ 40 cm dap) foram impactadas severamente pela combinação de perda de cobertura florestal e efeitos de borda. Distância da borda, tamanho do fragmento e a extensão da área de matriz aberta influenciaram fortemente as árvores pequenas tolerantes a sombreamento (≤ 20 cm dap). Apesar dos nossos resultados não corroborarem completamente as predições iniciais do nosso modelo conceitual, eles dão suporte à ideia de que a composição da paisagem interage com a estrutura do fragmento com impactos finais nos estoques de biomassa em florestas Neotropicais. Por fim, nós investigamos se o nível de distúrbio da região pode influenciar nas respostas da estrutura da vegetação à perda de cobertura florestal. Biomassa, mas não a densidade de indivíduos, foi afetada pela perda de cobertura florestal em regiões com nível intermediário de distúrbio, i.e. regiões apresentando uma combinação de níveis moderados de desmatamento (20-40% de cobertura florestal remanescente) em que a perturbação ocorreu ao longo dos últimos 30-60 anos, com alto grau de defaunação mas ainda abrigando populações relictuais de grandes mamíferos e, em sua maioria, compostos por uma matriz heterogênea. Em geral, nossos resultados destacaram que tanto a composição da paisagem como a estrutura do fragmento são os principais condutores de perda de biomassa em florestas Neotropicais e que o contexto da paisagem deve ser considerado para se obter estimativas mais confiáveis de emissão de carbono devido à degradação florestal. O planejamento da paisagem (e.g. restauração da cobertura florestal) deve ser incluído em estratégias de conservação em ordem de sustentar o armazenamento de carbono. Além disso, nós defendemos que iniciativas de conservação serão menos custosas e mais efetivas se implementadas em áreas sob níveis intermediários de distúrbio