Novos marcadores genéticos foram caracterizados para jacaré-de-papo-amarelo (Caiman latirostris) pela construção de bibliotecas enriquecidas de DNA microssatélite. Um microssatélite foi desenvolvido a partir de uma biblioteca enriquecida de DNA microssatélite (ACC/TGG)n e 12 a partir de uma biblioteca enriquecida de DNA microssatélite (AC/TG)n. Esses marcadores foram testados em indivíduos da espécie Caiman latirostris e resultaram em sete novos microssatélites polimórficos. Adicionalmente quatro marcadores microssatélites de Alligator mississipiensis previamente transferidos para Caiman latirostris foram utilizados. Amostras de sangue jacarés-de-papo-amarelo originárias de várzeas associadas ao Rio Piracicaba e alguns de seus tributários no estado de São Paulo, Brasil, foram avaliadas quanto à variação genética entre populações e o parentesco entre indivíduos. Foi detectada variabilidade entre indivíduos originários de sitos diferentes, mesmo entre aqueles com pequena distância geográfica. Os resultados sugerem que os grupos amostrados em cada sítio são compostos predominantemente por indivíduos aparentados. Uma possível combinação de alta taxa de mortalidade e baixa taxa de natalidade pode ser a explicação do baixo número de indivíduos dispersos com sucesso por geração entre os sítios estudados. Esses marcadores podem auxiliar na compreensão dos processos metapopulacionais que aparentemente ocorrem na espécie. / New genetic markers were characterized for the broad-snouted caiman (Caiman latirostris) by constructing libraries enriched for microsatellite DNA. One microsatellite was developed from a (ACC/TGG)n enriched microsatellite DNA library and 12 from a (AC/TG)n enriched microsatellite DNA library. These markers were tested in Caiman latirostris individuals and resulted in seven new polymorphic microsatellites for the specie. Additionally four Alligator mississipiensis microsatellite markers previously transferred for Caiman latirostris were used. Samples from broad-snouted caimans from small wetlands associated with the Piracicaba River and some of its tributaries in the state of São Paulo, Brazil were used to study the genetic variation between populations and parentage between individuals. Genetic variability was detected among individuals from different sites, even those within a small geographic distance. The results suggest that the groups sampled at each site are composed predominantly of related individuals. A possible combination of high mortality and low natality rates in the fragmented Caiman latirostris populations may explain the low number of successfully dispersed individuals per generation observed between the sites studied. These markers might help to understand the metapopulation processes that are occurring within this species.

O avanço agrícola é uma das maiores causas da fragmentação de habitats, levando à diminuição e ao isolamento de áreas naturais. Assim, torna-se importante compreender qual papel da estrutura e dinâmica da paisagem na manutenção da biodiversidade local. Este estudo teve como objetivo descrever a estrutura de uma comunidade de mamíferos de médio e grande porte em paisagem fragmentada agrícola, considerando a estrutura e dinâmica da paisagem. O estudo foi realizado em unidades de paisagem (16 Km2) com matriz predominante de pasto (n = 2) e cana-de-açúcar (n = 3) (bacia do rio Corumbataí, São Paulo) de maio a outubro de 2010. O levantamento da comunidade de mamíferos foi realizado através de busca ativa por pegadas, em 15 transectos alocados nas margens de riachos, com distância percorrida padronizada em 200 m. No entorno de cada transecto foram gerados buffers (250, 500, 1000 e 2000 m de raio), para o cálculo de índices de estrutura (porcentagens dos usos do solo, densidade de drenagem, densidade de estradas e proximidade entre fragmentos) e de dinâmica (taxa anual de mudança e perfil da curva de mudança florestal) para cinco anos (1962, 1978, 1995, 2000 e 2008). A relação entre os índices de paisagem e a riqueza de espécies foi analisada através de um teste PCA (Análise de Componentes Principais), gerando um gráfico Biplot. Posteriormente, foi realizada regressão linear múltipla, para análise da influência da estrutura e dinâmica da paisagem sobre a riqueza de espécies e frequência de registros. Foram registradas 19 espécies, sendo 17 em unidades de cana-de-açúcar e 13 em unidades de pasto. A comunidade de mamíferos de médio e grande porte, presente nestas matrizes da bacia do rio Corumbataí, é representada, em grande parte, por espécies tolerantes a alterações ambientais da região. A espécie com maior frequência de registros foi Procyon cancrivorus, de hábito generalista, como a maior parte dos animais registrados. Pelo gráfico Biplot, o buffer de 1000 m foi o que melhor distinguiu os sítios amostrais em relação às matrizes de cana-de-açúcar e pasto. A riqueza apresentou relação positiva com a porcentagem de áreas florestais e densidade de drenagem. Em contrapartida, a riqueza mostrou relação negativa com a porcentagem de pasto. A riqueza e a frequência de registros não apresentaram diferença estatística significativa entre as matrizes e também não houve relação significativa entre os índices de paisagem e as variáveis dependentes. A similaridade entre a composição de espécies das comunidades amostradas nas matrizes foi de 57%. As relações entre a estrutura desta comunidade e a paisagem necessitam de mais esforços para serem melhor compreendidas, já que o método de levantamento utilizado neste estudo, assim como a escala espaço-temporal, não permitiram descrever tais relações. / Agricultural expansion is a major cause of habitat fragmentation, leading to a reduction and isolation of natural areas. It is therefore important to understand the role of agricultural landscape structure and dynamics in maintaining local biodiversity. We aimed to describe how the structure and dynamics of an agricultural landscape influenced the community of mid and large bodied mammals in the south of Brazil. From May to October 2010 we studied the mammal community in five 16 km2 landscape units located within the Corumbatai river basin, São Paulo. Landscape units contained a matrix of predominantly pasture (n = 2) and cane sugar (n = 3). We used track surveys along 15 transects (200m each) to sample the mammal community along river banks. To calculate indices of habitat structure and dynamics we generated buffers (250, 500, 1000 and 2000 m radius) around each of the transects. Within each of these distance buffers we calculated habitat structure indices (percentage of land use, drainage density, road density and proximity of fragments) and indices of habitat dynamics (annual rate of change and profile of the forest change curve) for five years (1962, 1978, 1995, 2000 and 2008). The relationship between landscape indices and species richness was analyzed through a PCA (Principal Component Analysis) and associated Biplot. Subsequently, multiple linear regression was performed to analyze the influence of landscape structure and dynamics on species richness and frequency of records. We recorded 19 species, 17 in cane sugar units and 13 in pasture units. The community of mid and large bodied mammals present in the habitat matrices of the Corumbataí river basin was represented mostly by generalist species tolerant of environmental changes. The most frequently recorded species was Procyon cancrivorus, a habit generalist. The PCA Biplot showed that the 1000m buffer was the one that best distinguished the sampling sites in relation to the sugar cane and pasture matrices. Species richness was positively related with the percentage of forested areas and drainage density. In contrast, richness was negatively related with the percentage of pasture. The richness and frequency of records showed no statistically significant difference between the two matrix types and there was no significant relationship between landscape indices and the dependent variables. The similarity between the species composition of the sampled communities in the matrices was 57%. Understanding the relationships between the composition of this community and the landscape structure and dynamics requires more efforts, as the survey method used in this study, and the spatiotemporal scale, do not permit the description of these relationships.

The fragmentation of the landscape provides, among others, the occurrence of edge habitats, implying different responses of bats in this condition. We aimed in this study to analyze the community structure of bats in edge and interior of forest environments and to characterize the activity of frugivorous species in an Atlantic Forest fragment of Sergipe, northeastern Brazil. The study was done at Refúgio de Vida Silvestre Mata do Junco (RVSMJ), municipality of Capela. The sampling occurred between November 2012 and October 2013, with monthly samples in the edge and inside the forest in two areas for three consecutive nights in each area between 1800 to 2400 p.m., using ten mist nets. With a sampling effort of 116,640 h.m² we captured 96 bats belonging to 12 species, of which Lophostoma brasiliense and Micronycteris schmidtorum represent new records to the RVSMJ and Atlantic Forest area in Sergipe. The parameters of richness, abundance and diversity did not show differences between the edge and interior. However the species composition differed between these environments and the bats had different edge-sensitivity, indicating that bats are affected by the formation of edge in the locality. Regarding the activity, the species were already active in the first hour of sampling and remained in continuous activity during the sample period. There was a high temporal overlap in activity between A. lituratus and D. cinerea (Øjk = 0,813) and between C. perspicillata and D. cinerea (Øjk = 0,734). The temporal activity pattern of the bat community at the RVSMJ showed relatively common characteristics to the activity of fruit bats, but the absence of differences between schedules and between species activities can be a reflex of the small sample size. / A fragmentação da paisagem proporciona, entre outros, a ocorrência de borda nos habitats, inferindo em respostas diferenciadas dos morcegos a essa condição. O presente estudo analisou a estrutura da comunidade de morcegos em ambientes de borda e interior e caracterizou a atividade das espécies frugívoras em um fragmento de Mata Atlântica de Sergipe, no nordeste brasileiro. O estudo foi realizado no Refúgio de Vida Silvestre Mata do Junco (RVSMJ), município de Capela. As coletas ocorreram entre novembro de 2012 a outubro de 2013, sendo amostrados mensalmente a borda e o interior da mata em duas áreas na localidade, durante três noites consecutivas em cada área entre as 18:00 e 24:00 hs, utilizando-se dez redes de neblina. Com um esforço amostral de 116.640 h.m² foram capturados 96 morcegos pertencentes a 12 espécies, das quais Lophostoma brasiliense e Micronycteris schmidtorum correspondem a novos registros para o RVSMJ e para área de Mata Atlântica em Sergipe. Os parâmetros de riqueza, abundância e diversidade não exibiram diferenças entre a borda e o interior. No entanto, a composição de espécies diferiu entre esses ambientes e os representantes deste grupo apresentaram distintas sensibilidades à borda, indicando que os morcegos são afetados pela formação de bordas na localidade. Com relação à atividade, as espécies foram ativas já na primeira hora de coleta e permaneceram com a atividade contínua durante o período amostrado. Houve uma alta sobreposição no padrão de atividade entre A. lituratus e D. cinerea (Øjk = 0,813) e entre C. perspicillata e D. cinerea (Øjk = 0,734). O padrão de atividade da quiropterofauna do RVSMJ apresentou características relativamente comuns à relatada para morcegos frugívoros, porém a ausência de diferenças entre os horários e entre a atividade das espécies pode ser reflexo do baixo número amostral.

The Allee effect corresponds to a positive relationship between population size and individual fitness. This positive relationship can cause thresholds, that is, critical population sizes below which the population becomes extinct. For species submitted to the Allee effect, the formation and cohesion of groups is therefore crucial to survival. Animals can achieve this collective behaviour through local interactions. Each individual interacts locally with conspecifics and, at the scale of the group, a unity of behaviour emerges: the animals move together, rest in the same place, or choose the same habitat patch to settle on. <p><p>We use a combination of mathematical modelling and experimental work to study certain mechanisms of collective behaviour. In particular we assess the extent to which different individual interactions can induce collective patterns and thereby influence the dynamics of dispersal and settlement of populations. First, we study the collective settlement induced by the arresting effect of a marker secreted by conspecifics. Then, two potential mechanisms for collective movement are examined: following the conspecifics and following a trail laid by conspecifics. Finally, we integrate explicit mechanisms of dispersal behaviour in a dynamic model involving a set of interconnected populations. This allows the study of the interplay between collective movements and Allee effects at the scale of the metapopulation.<p><p>Our work is inspired by the lifestyle of the two-spotted spider mite, Tetranychus urticae, a phytophagous pest of recognised agricultural importance. These subsocial mites live in aggregates on the leaf surface, protected by a collectively spun silk web. Experimental evidence suggests that its population dynamics are subject to the Allee effect. Moreover, these mites show a tendency to migrate collectively, which makes them an appropriate biological model. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

Black rhinoceroses (Diceros bicornis) are endangered; there are only ~5000 individuals remaining worldwide. Consequently, black rhinos are managed as a metapopulation in southern Africa. This means that even though several sub-populations of black rhinos are geographically separated, they are managed as a single population. This is achieved by interchanging genetic material between the sub-populations. The main aim of the metapopulation management approach is to increase black rhino numbers in the sub-region. The Rhino Management Group (RMG), formed in 1989, was tasked with setting scientifically defensible demographic targets for the remaining black rhino sub-populations in southern Africa. In the same year, the RMG set fecundity rate at 25% and stated that each black rhino sub-population should achieve a minimum growth rate of 5% per year. The black rhino population in the Hluhluwe-iMfolozi Park (HiP), KwaZulu-Natal, South Africa, is of international importance as it has historically been a significant donor population for the translocation of black rhinos to establish new subpopulations elsewhere in the sub-region. However, the black rhino population at HiP is performing poorly in terms of reproductive success. My study was initiated to assess the reproductive life-histories of the black rhino population at HiP and to determine whether the population was meeting the demographic targets set by the RMG. I also wished to assess the use of space by black rhinos in HiP as a potential explanation for the observed reproductive output. Detailed life-history information for known black rhinos (n = 120) was used to investigate reproductive success between 1998 and 2012. The mean age at sexual maturity (ASM) was determined as 12 years, which is higher than the ASM found in black rhino populations that are considered to be performing poorly (ASM = ~ 7 years). The intercalving interval (ICI) was three years, nine months; this was also longer than the ICI of most black rhino populations. Further, the 2012 black rhino population at HiP had a growth rate of 6.4%, and the population comprised 60% adults, 23% sub-adults and 13% juveniles. Such demographics are usually observed in stable populations and it is possible that poor habitat quality could be resulting in the delayed ASM and longer ICI’s at HiP. In addition, 5% of the population is removed annually as part of the Black Rhino Range Expansion Program (BRREP) to establish new sub-populations. This could also be negatively influencing the reproductive performance of the HiP population. To determine the home range and habitat use of black rhinos at HiP, field data from 13 individual adult black rhinos (n = 9 females and 4 males) over a two-year period (2011-2012) were analyzed. Black rhino female home ranges were slightly smaller (15.4 ± 12.9 km²) than the home ranges of males (23.3 ±14.0 km²) and both sexes utilized similar habitats in the centre of the park. In addition, both the male and female black rhinos were found closer to rivers and drainage lines than expected. However, the assessment of black rhino space use in HiP was limited by a small sample size, making inferences of the impact of space use on reproductive success difficult. Nevertheless, my study suggests that food availability likely influences the location and sizes of black rhino home ranges at HiP. I believe that my study forms the first step in understanding the factors that influence the reproductive success and habitat use patterns of the current black rhino population at HiP. The information produced by my study should enable park management to identify future research to be conducted at HiP, which will yield more robust data for making informed black rhino management decisions at HiP and elsewhere.

Thesis (M.Tech.: Information Technology)-Durban Institute of Technology, 2005 iii, iii, 85 leaves / Throughout Africa, growing human populations and resulting loss of wildlife habitat is a critical issue for most animal species. It is more and more common for privately owned small or medium sized farms to reintroduce wildlife on their land and such protected areas are fast becoming the only refuges available to wild animals. However a comprehensive understanding of the complex ecological processes taking place is vital for the effective management of restricted areas and the conservation of biodiversity. Due to the enormous complexity of an ecological system and the long periods of the related dynamics, it is very difficult to analyse the interaction between animals and plant populations without suitable computer models. In this thesis, the dynamics between elephant and trees (a major food source) are considered using computer simulations.

Um componente considerado crítico para o manejo bem sucedido de populações silvestres é a manutenção da variação genética. No intuito de avaliar a magnitude e a distribuição da variabilidade genética existente em populações de Caiman latirostris, nove populações distribuídas ao longo do eixo latitudinal da distribuição da espécie no Brasil e uma população em cativeiro foram estudadas com auxílio de onze locos microssatélites. A diversidade gênica média (He=h) e a heterozigosidade média observada (Ho) apresentaram valores elevados, 0,628 e 0,567, respectivamente, denotando existência de elevada variabilidade genética para esta espécie nas regiões de estudo. A população paulista mesmo estando na zona intermediária de distribuição geográfica no Brasil não possui a maior variabilidade genética. O valor FST estimado foi 0,270 e o RST foi 0,342. Ambas as medidas de diferenciação entre as populações foram significativas (P<0,05). As altas estimativas de FST e RST sugerem a ausência ou fluxo gênico restrito entre essas populações, exceção feita entre as populações de Natal (RN) e João Pessoa (PB), onde não houve diferenciação significativa entre as populações, sugerindo assim que há fluxo gênico entre elas, fato confirmado pelo coeficiente de parentesco. Pôde-se concluir neste trabalho que a distância genética entre a população do litoral é afetada pela existência da Serra do Mar como barreira geográfica pelo fato desta população apresentar as maiores diferenciações genéticas e não se agrupar a nenhuma população pelo método de agrupamento UPGMA, mesmo estando próxima das populações paulista. A distância genética entre as populações parece não acompanhar a distância geográfica, em termos de gradiente latitudinal (r=0,206). Entretanto quando retiramos a população da Ilha do Cardoso esta correlação aumenta significativamente (r=0,540), indicando haver um certo padrão espacial da variabilidade genética entre as populações. O coeficiente médio de parentesco foi baixo entre e dentro das populações estudadas. Com estes resultados podemos começar a entender a dinâmica e estrutura social de populações de Caiman latirostris, e quanto mais se compreende sobre a biologia destes animais mais precisa serão decisões visando condições que permitam a existência continua da espécie. / A component considered critical for the managment well succeed of wild populations it is the maintenance of genetic variation. In the intention of evaluating the extend and the distribution of the existent genetic variability in populations of Caiman latirostris, nine populations distributed along the latitudinal axis of the distribution of the species in Brazil and a population in captivity they were studied with aid of eleven locos microsatellite. The genic mean diversity (He=h) and the observed mean heterozygosity (Ho) across all loci for all populations ranged from 0,628 and 0,567, respectively, denoting existence high genetic variability. The population from São Paulo being in the intermediate zone of geographical distribution in Brazil doesn't possess the largest genetic variability. The value dear FST was 0,270 and RST it was 0,342. Both differentiation measures among the populations were significant (P < 0,05). The higher estimates of FST and RST suggested a absence or low gene flow among those populations, exception done between the populations of Natal (RN) and João Pessoa (PB), where there was not significant differentiation among the populations, suggesting a gene flow pattern among them, fact confirmed by the related coefficient. It could be concluded in this work that the genetic distance among the population of the coast is affected by the existence of the Mountain of the Sea as geographical barrier for the fact of this population to present the largest genetic differentiations and not to group the any population for the grouping method UPGMA, same being close of the populations from São Paulo. The genetic distance among the populations seems not to accompany the geographical distance, in terms of latitudinal gradient (r=0,206). However when we removed the population of Cardoso's Island this correlation it increases significantly (r= 0,540), indicating there to be a certain space pattern of the genetic variability among the populations. The medium coefficient of related was low among and inside of the studied populations. With these results we can begin to understand the dynamics and it structures social of populations of Caiman latirostris.

Thesis (M.Tech.: Information Technology)-Durban Institute of Technology, 2005 iii, iii, 85 leaves / Throughout Africa, growing human populations and resulting loss of wildlife habitat is a critical issue for most animal species. It is more and more common for privately owned small or medium sized farms to reintroduce wildlife on their land and such protected areas are fast becoming the only refuges available to wild animals. However a comprehensive understanding of the complex ecological processes taking place is vital for the effective management of restricted areas and the conservation of biodiversity. Due to the enormous complexity of an ecological system and the long periods of the related dynamics, it is very difficult to analyse the interaction between animals and plant populations without suitable computer models. In this thesis, the dynamics between elephant and trees (a major food source) are considered using computer simulations.

Elephants (Loxodonta africana, Blumenbach) are recognised as keystone herbivores and ecosystem engineers in African ecosystems due to the noticeable impact they have on plant communities. This impact can influence other animal taxa and ecosystem processes, especially within enclosed systems. I investigated the effects of elephants in four vegetation types and the cascade effect on three associated taxa and two ecosystem processes at five sites with elephants and five paired sites without elephants in the Eastern Cape Province, South Africa, between April 2005 and July 2007. In grassland habitats, the physical structure of the vegetation was neither enhanced nor degraded in the presence of elephants. Within the locally important Thicket Biome, elephant browsing transformed relatively homogeneous stands of vegetation into more heterogeneous units. However, although seeds from thicket plant species were recorded in elephant dung, elephants did not promote the viability and germination success of these plants. In addition, elephant foraging on aloes caused declines in their populations. By contrast, the size of bushclumps was not reduced in the presence of elephants in bushclump savanna. The associated insect, bird and mammal communities appeared to benefit from elephant foraging in all vegetation types assessed, whereas vegetation patch dynamics and soil surface processes were neither enhanced nor degraded in the presence of elephants. I conclude that, at current densities, elephants do not (in most cases) negatively affect plant and animal communities or ecosystem processes in enclosed reserves. However, elephants have only been present at each site for a relatively short period and it is likely that their impact will be cumulative, increasing over time in these closed systems. Thus, future research along a continuum of elephant density and time since re-introduction is recommended.

Um componente considerado crítico para o manejo bem sucedido de populações silvestres é a manutenção da variação genética. No intuito de avaliar a magnitude e a distribuição da variabilidade genética existente em populações de Caiman latirostris, nove populações distribuídas ao longo do eixo latitudinal da distribuição da espécie no Brasil e uma população em cativeiro foram estudadas com auxílio de onze locos microssatélites. A diversidade gênica média (He=h) e a heterozigosidade média observada (Ho) apresentaram valores elevados, 0,628 e 0,567, respectivamente, denotando existência de elevada variabilidade genética para esta espécie nas regiões de estudo. A população paulista mesmo estando na zona intermediária de distribuição geográfica no Brasil não possui a maior variabilidade genética. O valor FST estimado foi 0,270 e o RST foi 0,342. Ambas as medidas de diferenciação entre as populações foram significativas (P<0,05). As altas estimativas de FST e RST sugerem a ausência ou fluxo gênico restrito entre essas populações, exceção feita entre as populações de Natal (RN) e João Pessoa (PB), onde não houve diferenciação significativa entre as populações, sugerindo assim que há fluxo gênico entre elas, fato confirmado pelo coeficiente de parentesco. Pôde-se concluir neste trabalho que a distância genética entre a população do litoral é afetada pela existência da Serra do Mar como barreira geográfica pelo fato desta população apresentar as maiores diferenciações genéticas e não se agrupar a nenhuma população pelo método de agrupamento UPGMA, mesmo estando próxima das populações paulista. A distância genética entre as populações parece não acompanhar a distância geográfica, em termos de gradiente latitudinal (r=0,206). Entretanto quando retiramos a população da Ilha do Cardoso esta correlação aumenta significativamente (r=0,540), indicando haver um certo padrão espacial da variabilidade genética entre as populações. O coeficiente médio de parentesco foi baixo entre e dentro das populações estudadas. Com estes resultados podemos começar a entender a dinâmica e estrutura social de populações de Caiman latirostris, e quanto mais se compreende sobre a biologia destes animais mais precisa serão decisões visando condições que permitam a existência continua da espécie. / A component considered critical for the managment well succeed of wild populations it is the maintenance of genetic variation. In the intention of evaluating the extend and the distribution of the existent genetic variability in populations of Caiman latirostris, nine populations distributed along the latitudinal axis of the distribution of the species in Brazil and a population in captivity they were studied with aid of eleven locos microsatellite. The genic mean diversity (He=h) and the observed mean heterozygosity (Ho) across all loci for all populations ranged from 0,628 and 0,567, respectively, denoting existence high genetic variability. The population from São Paulo being in the intermediate zone of geographical distribution in Brazil doesn't possess the largest genetic variability. The value dear FST was 0,270 and RST it was 0,342. Both differentiation measures among the populations were significant (P < 0,05). The higher estimates of FST and RST suggested a absence or low gene flow among those populations, exception done between the populations of Natal (RN) and João Pessoa (PB), where there was not significant differentiation among the populations, suggesting a gene flow pattern among them, fact confirmed by the related coefficient. It could be concluded in this work that the genetic distance among the population of the coast is affected by the existence of the Mountain of the Sea as geographical barrier for the fact of this population to present the largest genetic differentiations and not to group the any population for the grouping method UPGMA, same being close of the populations from São Paulo. The genetic distance among the populations seems not to accompany the geographical distance, in terms of latitudinal gradient (r=0,206). However when we removed the population of Cardoso's Island this correlation it increases significantly (r= 0,540), indicating there to be a certain space pattern of the genetic variability among the populations. The medium coefficient of related was low among and inside of the studied populations. With these results we can begin to understand the dynamics and it structures social of populations of Caiman latirostris.