Explaining spatial and temporal variation in the abundance of species is one of the primary goals of ecology. Habitat selection, the behaviour that organisms use to choose habitat patches that maximize fitness, can explain patterns in abundance between patches at small spatial scales within the dispersal capacity of the species. However, habitat selection models assume there is a reduction in individual fitness as population density increases due to increased competition between individuals. Ectotherms, which often select habitats based on temperature, a density-independent resource, may not display density-dependent responses if temperature limits energy assimilation more than finite food resources limit energy acquisition. As predicted by their dependence on environmental temperatures, some ectotherms select habitat largely independently of population density when temperatures are far from the optimal temperature for performance. But, is density-dependence prevalent in ectotherm populations when temperatures are close to the optimal temperature for performance? Habitat selection models also assume that all individuals of a population exhibit the same strategy for maximizing fitness through habitat selection. However, differences in morphology and behaviour (e.g., reproductive strategy) can modify the optimal habitat selection strategy for different phenotypes. Finally, observed patterns in habitat selection and abundance can also be modified by competition with other species. Quantifying the relative importance of these different factors that affect habitat selection behaviour will improve our ability to predict the spatial distribution and relative abundance of organisms. The objective of my thesis was to explain spatial variation in the abundance of ectotherms, using the ornate tree lizard (Urosaurus ornatus) as a study species. In chapter one, I tested whether density-dependent habitat selection explained patterns in abundance and fitness of lizards between two habitats differing in suitability. In chapter two, I tested whether density dependent habitat selection in tree lizards was caused by intraspecific competition for food that limited body size and growth. In chapter three, I tested whether variation in reproductive strategy, as indicated by throat colour phenotype, affected space use and habitat selection in male tree lizards. Finally, in chapter four, I tested whether interspecific competition with another lizard species affected habitat selection, fitness, and abundance of tree lizards. My thesis emphasizes the importance of intraspecific competition in shaping patterns of habitat selection and abundance in terrestrial ectotherms. I show that habitat selection is strongly density-dependent despite differences in thermal quality between habitats. I show that density-dependent mortality and growth lower the fitness of individuals when populations reach high densities, and this likely caused habitat selection to be density-dependent. Despite this evidence for density-dependent habitat selection, I show considerable variation between individuals in habitat selection and space use. Males with different throat colour phenotypes select habitats differently, demonstrating that variation in morphology can influence habitat selection patterns within a population. Finally, I show that interspecific competition with another lizard affects space use and how frequently tree lizards switch habitats, but this does not lead to differences in fitness or in the relative abundance of tree lizards in habitats. Therefore, intraspecific competition for resources was the dominant force shaping the relative abundance of tree lizards in different habitats.

Loggerheads (Caretta caretta) are listed as Threatened and green turtles (Chelonia mydas) are listed as Endangered under the United States Endangered Species Act. While green turtle nest production in Florida has increased markedly in recent years, loggerhead nest production has followed a more tenuous path. Reasons for these differences are unknown. Limited demographic information is available for these species of conservation concern. I used Barker models, which incorporated mark-recapture, live-resight and dead recovery data, implemented in Program MARK. These models were used to estimate apparent survival for immature loggerhead (

Competition among individuals over shared resources reveals asymmetries in quality resulting in the formation of dominance hierarchies. These hierarchies act as a mechanism for social selection by partitioning resources among group-living animals. The following chapters describe my dissertation research which investigates the factors contributing to competitive asymmetries among broodmates as well as the short- and long-term consequences of the early social environment for the cooperatively breeding red-cockaded woodpecker (Picoides borealis). My research revealed that fledgling red-cockaded woodpeckers form male-biased, linear dominance hierarchies. Among fledgling males,, high relative nestling condition strongly predicted fledgling dominance, and this condition–rank relationship persisted through independence. Male nestlings are slightly larger and heavier than females; however, the sexual size dimorphism in mass is only present in mixed-sex broods, suggesting that the subtle structural size advantage gives males a competitive advantage over their sisters. Conflict rates among siblings increased with decreasing targeted feeding rates, and dominant fledglings were able to secure more food from provisioning adults through scramble competition. First-year survival favored males over females and dominant males over subordinates. Females were more dispersive overall than males, and subordinate males were more likely to disperse than dominants. The social environment prior to fledging influenced male dispersal decisions and subordinates delayed dispersal in the spring in situations where all dominants died over the winter. The probability of delayed dispersal in females was higher for females raised without brood-mates in one of two populations included in a long-term demographic data analysis. The availability of breeding vacancies may explain the differences in female dispersal behavior according to social environment between these populations. This research contributes to a greater understanding of the relative contribution of intrinsic benefits versus extrinsic constraints as an influence on delayed dispersal decisions in red-cockaded woodpeckers. Inter- and intra-sexual social rank is correlated with individual access to natal food resources and the probability of first-year survival. The intrabrood variation in dispersal strategies driven by social rank is sufficient to regularly produce both dispersal strategies among males and provides additional support that delaying natal dispersal is the preferred strategy for males in this system. / Ph. D.

In this report mark-recapture data are analyzed with the use of weighted log-linear models, mosaics, and computer drawings of fish. The data are from salmon hatcheries, subsequent returns to the hatcheries and commercial catches of salmon. The log-linear models were used to study the effects of several variables on catches and returns. It is shown that these variables may have opposite effects depending on the brood year of the fish, that hatchery returns and commercial catches do not respond in the same way to the variables, and that more research is needed to determine the causes of the brood year differences. The mosaics and fish drawings were used to study the migration of the salmon in the ocean. The results confirm that chinook salmon decrease their food intake during the return trip to the hatchery, and they are consistent with theories of ocean migration of other species of salmon. / Thesis / Master of Science (MSc)

Large carnivores are in decline throughout their range primarily due to anthropogenic influences. This is a concern because these species have high social, economic and ecological value. African lion (Panthera leo) populations are in decline in most parts of the continent and this has been attributed largely to habitat loss, habitat fragmentation, wire snare poaching, black market trafficking in lion parts, retaliatory killings, and poorly regulated trophy hunting. In order to implement effective lion conservation actions, the effect of each factor on lion demography or population dynamics must be well understood. In the past, most studies have used indirect methods to quantify the effects of hunting on lion demography. The temporary ban on lion hunting in Zambia allowed me to study directly how removing a key source of mortality, for males in particular, changed the demography. Using data collected from 2008-2015, I studied how the lion population in and around South Luangwa National Park, Zambia responded to a 3-year moratorium on lion trophy hunting implemented in 2013. For the duration of the study ban, I monitored 386 known individuals in 19 prides and 15 male coalitions. Reproductive activity as represented by cub production appeared to improve after the hunting ban and the number of individually identifiable lions in the study area steadily increased from 88 in 2008 to 197 in 2015. Using a Cormack-Jolly-Seber model, I found that apparent annual survival increased in the absence of hunting for all adult male age classes. The temporary ban on hunting was lifted in 2016 and I recommend that quotas remain conservative in order to allow more time for the population to recover, particularly in the adult male age classes. Better monitoring protocols should also be implemented to promote compliance with hunting regulations.

Master of Science / Department of Biology / Brett K. Sandercock / The migratory period is a critical time for birds. Population declines in long-distance migratory birds have been associated with trophic mismatches among climate change, timing of food availability, and timing of migratory movements. Studies on migratory songbirds have been limited to eastern North America and Europe, and migration strategies of birds may differ along the Pacific flyway. We evaluated the stopover ecology of five species of migratory songbirds at a coastal site in northern California. We found variation in changes in timing of spring migration, but consistent changes in timing of autumn migration over a 22-year period. Timing of spring migration advanced for Swainson’s Thrushes (Catharus ustulatus) and Yellow Warblers (Setophaga petechia), was compressed for Pacific-slope Flycatchers (Empidonax difficilis) and Wilson’s Warblers (Cardellina pusilla), but was protracted for Orange-crowned Warblers (Oreothlypis celata). In contrast, timing of autumn migration was delayed for Pacific-slope Flycatchers, Orange-crowned Warblers, and Yellow Warblers, but was protracted for Wilson’s Warblers. Warm, wet conditions were associated with advanced spring migration, and warm, dry conditions were associated with delayed autumn migration. Changes in timing of migration related to climate conditions were strongest for Pacific-slope Flycatchers and Orange-crowned Warblers. Stopover duration of our five study species was longer than songbirds in eastern flyways, and on average, was longer in spring than autumn. Pacific-slope Flycatchers and Yellow Warblers had longer stopovers in spring, whereas Swainson’s Thrushes and Wilson’s Warblers stopped had longer stopovers in autumn. Birds captured at low body mass had longer stopovers in spring and autumn. Migratory birds in western North America may have different migration strategies because of differences in climate and geography, but more study is needed to discover migratory routes and compare our results to other stopover sites along the Pacific flyway. Understanding differences in migration strategies of different populations of the same species are important in directing conservation efforts, especially in light of ongoing climate change.

O bagre cego de Iporanga, Pimelodella kronei, da região do Parque Estadual Turístico do Alto Ribeira-PETAR, SP, foi o primeiro peixe troglóbio descoberto e descrito no Brasil e um dos mais estudados até hoje. Populações de bagres cegos atribuídas a essa espécie foram encontradas, até o momento, em cavernas do Sistema das Areias (Areias de Cima - localidade-tipo e abrigando a maior população conhecida -, Areias de Baixo e Ressurgência das Areias da Água Quente, esta aparentemente com uma população periférica), na Caverna Córrego Seco, na Ressurgência de Bombas e em cavernas do Sistema Alambari (caverna Alambari de Cima - população aparentemente extinta -, e Abismo do Gurutuva). Os bagres cegos do Sistema Areias foram excessivamente coletados na década de 1970, o que teria provocado um declínio populacional acentuado, registrado por Trajano (1987) 10 anos mais tarde. O presente estudo aborda a ecologia de P. kronei, com ênfase na população das Areias, utilizando métodos comparáveis aos de Trajano (op. cit.), para fins de monitoramento. Os aspectos analisados nas Areias foram: tamanho da população, crescimento individual e deslocamentos no habitat, a partir de marcação e recaptura individual; estrutura da população em termos de distribuição das freqüências de tamanhos (comprimento-padrão), peso e fator de condição (isométrico e alométrico); troglomorfismos (graus de redução de olhos e de pigmentação melânica). Complementarmente, exemplares das Areias, Bombas, Córrego Seco e Gurutuva foram comparados quanto à morfometria. A população de P. kronei das Areias mostrou evidências de estabilidade através da recuperação da estrutura de peso e comprimento-padrão. Assim como observado para outras espécies de siluriformes troglóbios, o crescimento individual é altamente heterogêneo, com casos de crescimento negativo. Os parâmetros da equação de Von Bertallanfy revelaram valores superiores aos de Trajano (1987), resultando em uma longevidade média de 15-20 anos. Os bagres estudados na Areias de Cima mostraram tendência de deslocamentos rio acima, em direção ao ponto mais distal no Córrego Grande (mais próximo à entrada de alimento externo), provavelmente para alimentação e reprodução. É possível que existam diferenças entre a população de Bombas e as demais localidades, sendo necessária uma investigação mais aprofundada, com métodos mais discriminatórios. / The blind catfish from Iporanga, Pimelodella kronei, from Parque Estadual Turístico do Alto Ribeira- PETAR, SP, was the first troglobitic fish discovered and described in Brazil. It is also one of the most investigated species so far. Populations of blind catfish attributed to this species have been found in caves belonging to the Areias System (Areias de Cima - type-locality, sheltering the largest known population - , Areias de Baixo and Ressurgence of Areias of Águas Quente, which apparently shows a peripherical population), besides Córrego Seco Cave, Bombas Ressurgence and caves of the Alambari System (Alambari de Cima - an apparently extinct population - , and Gurutuva). The blind catfish of Areias system were over-collected during the 1970s, which may have led to an accentuated population decline, as recorded by Trajano (1987) ten years later. The present study focus on the ecology of P. kronei, with emphasis on the Areias population, by using methods comparable to those by Trajano (op. cit.) for monitoring purposes. The aspects studied in the Areias caves were: population size, individual growth and movements in the habitat, based on individual marking and recapture; population structure in terms of distribution of size (standard length), weight and condition factor (isometric and alometric) frequencies; troglomorphism (levels of eye reduction and melanic pigmentation). In addition, samples from Areias, Bombas, Córrego Seco and Gurutuva were morphometrically compared. The P. kronei population of Areias showed evidence of stability through the regain of the size and weight structure. Like the observed for other troglobitic siluriforms, the individual growth in P. kronei is highly heterogeneous, with frequent cases of negative growth. The parameters of the Von Bertallanfy equation calculated were higher than those by Trajano (1987), resulting in a higher expected longevity - 15 to 20 years. The blind catfish from Areias presented a tendency to move upstream, approaching the distal end of Córrego Grande (closer to the input of epigean nutrients), probably for feeding and reproduction. There may be some differences in the Bombas populations of other locations, to be investigated using more discriminatory methods.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Estudos envolvendo estimativas de parâmetros populacionais de cetáceos baseados em dados de foto-identificação têm crescido ao longo dos anos, possibilitando estimar parâmetros como abundância, sobrevivência, probabilidade de captura e padrões de emigração. No Brasil, ainda são poucos os estudos com cetáceos que utilizam essa abordagem, contudo, eles estão aumentando, principalmente com a espécie Sotalia guianensis (van Bénéden, 1864), conhecida como boto-cinza. O objetivo desse trabalho foi estimar a sobrevivência, a probabilidade de captura, a emigração e a abundância da população de botos-cinza na Baía de Paraty (RJ; 2307S 4433W), utilizando modelos de marcação-recaptura baseados em dados de foto-identificação, entre os anos de 2005 e 2009. Os cruzeiros foram realizados entre agosto de 2005 e julho de 2009 com uma embarcação de 26 pés e motor de 200hp. As fotografias das nadadeiras dorsais foram tomadas com a utilização de uma câmera digital Canon 20D equipada com lente de zoom 100-400mm. Foram realizadas 60 saídas de campo na Baía de Paraty totalizando 172 horas de esforço no campo, sendo 72 horas de esforço fotográfico. Nas 42 saídas com observação de botos-cinza foram efetuadas 27.892 fotografias e, destas, 6.612 (23,7%) foram consideradas adequadas e utilizadas na identificação dos indivíduos de boto-cinza. Um total de 621 indivíduos foi catalogado com base em marcas permanentes na nadadeira dorsal. Os históricos individuais de captura foram analisados no programa MARK. A estimativa de sobrevivência variou de 0,82 ( 0,05) a 0,85 ( 0,05) e a probabilidade de captura variou de 0,16 ( 0,04) a 0,40 ( 0,06). O número de indivíduos marcados na população estimado pelo modelo de Jolly-Seber foi de N = 586 ( 114,66) (IC: 361-811). A proporção de animais marcados na população para o modelo de JS foi de 0,66 ( 0,03). A partir desse valor, foi possível obter uma abundância total de 884 indivíduos ( 176,8) (IC: 602-1296). A probabilidade de emigração temporária (γ``) e de permanecer fora da área (γ`) foi de 0,43 ( 0,07) e a probabilidade de captura/recaptura variou de 0,01 ( 0,01) a 0,42 ( 0,05). A probabilidade de permanência na área (1 - γ``), assim como a probabilidade de retorno da emigração temporária (1 - γ`) foi 0,57. Apesar dos movimentos significativos de indivíduos transientes, uma vez que a área amostrada é menor do que as áreas de vida que se tem estimadas para o boto-cinza, os dados indicam uma residência moderada dos botos-cinza. A abundância anual total variou de 325 para o ano de 2005 a 448 para o ano de 2009. A análise de poder determinou que seriam necessários nove anos de estudo para observar uma tendência de queda de 5% no tamanho da população. Os resultados reforçam que a Baía de Paraty abriga uma das maiores populações já estimadas para Sotalia guianensis.

I consider statistical modelling of data gathered by photographic identification in mark-recapture studies and propose a new method that incorporates the inherent uncertainty of photographic identification in the estimation of abundance, survival and recruitment. A hierarchical model is proposed which accepts scores assigned to pairs of photographs by pattern recognition algorithms as data and allows for uncertainty in matching photographs based on these scores. The new models incorporate latent capture histories that are treated as unknown random variables informed by the data, contrasting past models having the capture histories being fixed. The methods properly account for uncertainty in the matching process and avoid the need for researchers to confirm matches visually, which may be a time consuming and error prone process. Through simulation and application to data obtained from a photographic identification study of whale sharks I show that the proposed method produces estimates that are similar to when the true matching nature of the photographic pairs is known. I then extend the method to incorporate auxiliary information to predetermine matches and non-matches between pairs of photographs in order to reduce computation time when fitting the model. Additionally, methods previously applied to record linkage problems in survey statistics are borrowed to predetermine matches and non-matches based on scores that are deemed extreme. I fit the new models in the Bayesian paradigm via Markov Chain Monte Carlo and custom code that is available by request.

Pathogenic fungi are increasingly associated with epidemics in wildlife populations and represent a significant threat to global biodiversity. Snake fungal disease is an emerging disease caused by the fungus, Ophidiomyces ophiodiicola, and appears to be widespread in the eastern United States. Yet an evaluation of field diagnostics, and an understanding of the population-level consequences of the disease, are lacking. First, I evaluated the use of clinical signs to predict the presence of O. ophiodiicola across season and snake habitat affiliation (aquatic or terrestrial) and I compared two sampling methods to see if collection method impacts PCR result. Overall, snakes with clinical signs had a higher probability of testing positive regardless of season or habitat association. However, terrestrial snakes had a lower overall probability of testing positive for O. ophiodiicola compared to aquatic snakes. I found no significant difference between sampling methods. Second, I used Passive Integrated Transponder (PIT) telemetry, and multistate capture-mark-recapture modelling to determine if SFD affects the short-term survival, movement, and behavior of wild snakes. I found no difference in short-term survival for snakes with SFD. Snakes with SFD spend more time surface-active and have lower permanent emigration and temporary immigration rates than snakes without SFD.