Declines and extinctions of amphibian populations are a global dilemma with complex local causes, which should be viewed in the context of a much larger biodiversity crisis. As other animal groups, amphibians with restricted distributions, such as island endemics, are thought to be more vulnerable to environmental change and susceptible to population declines. In the New Zealand archipelago, the only four native species of frogs (Leiopelma hochstetteri, L. archeyi, L. hamiltoni and L. pakeka) are classified as threatened. In particular Leiopelma hochstetteri, the most widespread and abundant endemic frog species in New Zealand, now survives only in spatially fragmented populations as a result of direct or indirect human activity. Hence, it is recognised as threatened and fully protected by legislation. In the last fifty years, some L. hochstetteri populations have been studied, providing descriptive information, which may be used to assess the current status (increasing, stable or declining) of previously or never monitored populations. This thesis examines the diet and trophic level, the effects ship rats (Rattus rattus) as well as the distribution and abundance of L. hochstetteri on a habitat-use context, to provide a basis for evaluating conceivable decline-agents, and to establish a platform to design directed conservation strategies. The Waitakere Ranges are considered a Leiopelma hochstetteri conservation management unit, on which L. hochstetteri has been previously studied. This area consists of a series of hills that run roughly north–south, which are mostly covered in regenerating indigenous vegetation. Today, 60% of the Waitakere Ranges fall within a Regional Park, which together with its surrounding residential areas is afforded protection to minimise the effects of development on the region. The accessibility and conservation character of this area makes it an ideal area for the study of L. hochstetteri populations. As a first step to characterise the diet and trophic level of L. hochstetteri within streams in the Waitakere Ranges, Auckland, stable carbon and nitrogen isotope analyses were undertaken on a variety of sympatric terrestrial and aquatic plant and animal species, including adult frogs. These results showed that: 1) aquatic and terrestrial food webs were linked by terrestrial inputs into the stream; 2) invertebrate and vertebrate predators separated well into distinct trophic groups; and 3) L. hochstetteri occupied an intermediate trophic position among predators, with a diet, at least as an adult, comprising terrestrial invertebrates. Shortfin eels and banded kokopu were identified as potential predators of L. hochstetteri, but data for rats were inconclusive. The inconclusiveness of these trophic studies, with regard to the effects of ship rats on L. hochstetteri populations, lead me to evaluate the influence of a seven-year ship rat management operation on frog abundance. To achieve a reliable evaluation, the habitat characteristics that had significant influence on frog abundance were identified. Then, it was confirmed that the study areas represented similar habitats in terms of those variables, and finally the effect of the pest-management activities was evaluated. Presence/absence of pest-management operations did not have a significant effect on frog abundance. These results, together with the results of the diet and trophic level analyses, suggested that ship rats do not represent a significant threat for this frog species, at least in the Waitakere Ranges. The results of distribution and abundance investigations indicated that in the Waitakere Ranges frogs are currently widely distributed, relatively abundant and that recruitment has occurred at least in the last ten years. Additionally, in order to identify associations between habitat characteristics and frog distribution and abundance, reliable and specifically designed monitoring methodologies were developed. Although this frog is known to occur in wet areas adjacent to shaded streams in forested catchments, quantitative ecological data previously did not exist to enable characterisation of its habitat. Here, novel data were reported on the current distribution and habitat requirements of this species in the Waitakere Ranges. Statistical modelling demonstrates frogs most likely occur in small, erosive streams with coarse substrates and cold waters, surrounded by mature or undisturbed riparian vegetation, where higher abundances of frogs may be found in steep areas with stable patches of cobbles and boulders lying against larger stream bed elements within the stream channel. Anthropogenic activities, such as clearing or logging, and upstream disturbances that potentially increase silt input into streams were identified as threats to these frog species. Finally, the habitat-use information gathered during this investigation was utilised to develop a spatial decision support system (SDSS) as a tool to assess the quality and quantity of habitat available to L. hochstetteri populations associated with the Auckland Region. These results have important implications for the conservation of New Zealand native frog species and riparian stream habitat.

Current research is necessary to focus management of white-tailed deer (Odocoileus virginianus) in the agricultural Midwest, especially given the novel presence of chronic wasting disease in the region. My objectives were to: 1) examine the potential effects of weather and row-crop harvest on daily harvest of white-tailed deer by archery, and individual hunter variables (e.g., age, weapon preference, preference of hunting method) on individual hunter efficiency and success in 2 regions of Illinois (i.e., east-central and southern Illinois); 2) estimate white-tailed deer densities using direct (i.e., spotlighting deer from road transects) and indirect (i.e., counting pellet groups on randomly-placed transects) techniques across 3 study areas in the midwestern U.S.; 3) quantify sex, age, and season-specific survival and dispersal rates of white-tailed deer in east-central Illinois; and 4) assess habitat selection of white-tailed deer during the summer months in east-central Illinois. Knowledge of factors affecting hunter efficiency and success is vital given current trends in hunter attitudes and behaviors that are leading to a reduction in the efficacy of white-tailed deer hunting as a population management tool nationwide. Numerous techniques of density estimation for white-tailed deer have been developed, including aerial surveys, mark-recapture or resight methods, pellet counts, and thermal infrared imaging surveys. Distance sampling has shown great potential for estimating white-tailed deer density at a reduced cost relative to traditional survey techniques and may be useful within the agriculturally-dominated Midwest. Survival and dispersal rates are pertinent model parameters when examining chronic wasting disease (CWD) geographic spread. Structural changes in agricultural landscapes are frequent (i.e., temporally) and ecologically abrupt (i.e., spatially) due to crop harvest and rotation and human development. Providing sex-specific data on site selection would benefit agencies and private land owners by allowing compartmental focus for selective management strategies for either male or female white-tailed deer. Following the 2006 hunting season, I queried white-tailed deer hunters regarding factors potentially affecting hunter efficiency and success in east-central and southern Illinois using a mail-in survey (n = 2,000). I also examined the influence of weather and row-crop harvest progress on daily harvest of white-tailed deer by archery hunting. Hunter efficiency ( ¡À SE= 0.12 ¡À 0.01 deer/day) and hunter success (range = 1.25-1.39 deer/hunter) were similar between study areas. No relationship (r2 <0.01) was detected between respondent age and hunter efficiency, but had a weak influence (r2 = 0.006) on hunter success. Respondents that preferred shotguns, used 1 weapon, and those that preferred still hunting had 62%, 58%, and 52%, respectively, greater (P ¡Ü 0.001) mean hunter efficiency than those in the lowest group within their particular categories. There was no apparent difference (P ¡Ý 0.087) in hunter efficiency across categories of area familiarity, number of hunting methods used, and scouting hours, or categories related to access and use of reconnaissance tools. Respondents that had ¡Ý11 years of area familiarity, preferred archery hunting, used ¡Ý 3 weapons, used ¡Ý3 hunting methods, scouted ¡Ý30 hours, and preferred tree stands had 51%, 45%, 62%, 35%, 61% and 41%, respectively, greater (P ¡Ü 0.001) mean hunter success than those in the lowest group within their particular categories. Access and use of GIS did not appear to affect (P = 0.376) hunter success. Respondents that had access and used topographic maps, aerial or satellite photographs, or GPS had 35%, 34%, and 29% greater (P ¡Ü 0.049), respectively, hunter success than those in the lowest group within their particular categories. Crop harvest progress did not (P = 0.780) appear to affect daily harvest of white-tailed deer by archery hunting. Three models of weather impacts on daily harvest of white-tailed deer by archery hunting had AICc <2. The most parsimonious models¡¯ covariates were MaxWSP, MSLP, and WDSP1, with MaxWSP (¦Â = -0.005) having a negative influence and MSLP (¦Â = 0.00007) and WDSP1 (¦Â = 0.006) having a positive influence on daily harvest of white-tailed deer by archery. I compared direct- and indirect distance sampling techniques for estimating white-tailed deer densities on study areas in east-central Illinois (ECI), southern Illinois (SI, 2007 only), and northern lower peninsula of Michigan (MI) during winter 2007-08. Density estimates obtained via indirect distance sampling for MI, ECI, and SI were 6.1-12.7, 11.2-15.8, and 15.4 deer/km2, respectively. Density estimates obtained via direct distance sampling for MI, ECI, and SI were 18.3-25.2, 14.4-18.1, and 19.0 deer/km2, respectively. Upon examining confidence interval (CI) overlap between direct- and indirect distance sampling techniques by year and study area, only the MI study had non-overlapping CI values. An examination of sites used by deer in summer and quantification of their survival and dispersal rates were conducted in east-central Illinois. From December 2005 to September 2009, I monitored 105 white-tailed deer for 35,478 radiodays for survival and dispersal analysis. I used Program MARK to estimate rates of annual survival, seasonal survival, and dispersal for fawns, yearlings, and adults. I measured habitat variables at sites used by white-tailed deer in summer and random locations in east-central Illinois, examining potential differences in site selection by sex and at multiple scales. Male and female full-season (winter/spring [16 Dec¨C14 May], summer [15 May-30 Sep], fall/winter [1 Oct¨C15 Dec]) survival rate ranged from 0.56 to 0.95 and 0.84 to 0.95, respectively. Male survival was lower than that of females during the fall/winter season for a model not accounting for overdispersion. Averaging across parsimonious models, the dispersal rate for yearling and fawn males and yearling and fawn females were 0.44 ¡À 0.07 and 0.41 ¡À 0.07, respectively. Adult male dispersal rate was 0.46 ¡À 0.15 and no adult females dispersed. Slight differences were observed in habitat variables at sites used by male and female white-tailed deer, but hypotheses of sexual segregation between the sexes were not supported. Males were never observed in developed landcover. Sex differences in the use of developed landcover approached significance with females using sites with that landcover 131% more than males; they also used sites with 87% higher patch density of wetland landcover than males. Based on the best-fit AIC model, overall (i.e., both males and females) site selection of white-tailed deer was influenced by patch density of agriculture landcover, percentage of fallow field landcover, disjunct core area of the landscape, upper visual obstruction, and percentage of shrubs. Increases in upper visual obstruction and percent of shrub increased the likelihood of a site being selected. Decreases in patch density of agriculture landcover, percentage of fallow field landcover, and disjunct core area of the landscape increased the likelihood of a site being selected. This study provides updated information regarding white-tailed deer demographics, habitat use, survey methods, and hunter behavior to white-tailed deer managers in the Midwest. The data produced from mail-in surveys may provide knowledge of influences on hunter efficiency and success that may be used to manipulate deer harvest under a declining number of hunters in the U.S. Consistent estimates of density between indirect- and direct distance sampling shows utility for use of direct distance sampling within agriculturally-dominated regions of the Midwest. Fine-scale management by private landowners may benefit from my data by demonstrating a preference for a particular landscape signature by deer during the summer.

Scaphirhynchus sturgeons are species of concern throughout their ranges. To revitalize their populations, a thorough understanding of their habitat use during all life stages is critical. However, limited information exists about fine-scale habitat use during early life. To determine the feasibility of using radio telemetry to monitor habitat use in the field, I assessed growth and survival of age-0 sturgeon tagged with 0.2 g, non-functioning telemetry tags during an eight-week period in a controlled setting and then tested field performance in the Middle Mississippi River (MMR). Three treatments were evaluated: 1) control, 2) internally implanted tags, and 3) externally attached tags. Growth and survival varied across treatment groups (all comparisons P<0.05). Sturgeon with internally implanted tags grew slowly and had low survival, while sturgeon within the control group and those with external tags grew faster and had higher survival (~100%). A trial was conducted to determine swimming performance of each of the treatment groups, which resulted in no differences in critical swimming speed among the tag types. Based on these results, a field-based telemetry evaluation was performed to determine if it is feasible to monitor fine-scale habitat use. Four wild, age-0 Scaphirhynchus sturgeon were externally tagged with functioning radio transmitters in the MMR. After less than 24-h, the tagged fish were no longer detectable, likely due to environmental conditions (e.g., extreme depths and high conductivity). Thus, efforts should be directed to creating a small telemetry transmitter that uses technology (e.g., ultrasonic) that transmits well in deep rivers while maintaining a minimum tag weight and maximizing battery life. Fine scale habitat use of age-0 sturgeon may ultimately be able to be quantified in the MMR and other deep rivers in the future, if technologies improve.

In the greater Central Hardwood Region, advance regeneration of oak (Quercus spp.) and hickory (Carya spp.) has been in decline for several decades. Facilitated in part by an abrupt change in disturbance regime, coupled with an increase in herbivore density, the existing mid-successional, mast-producing species are being outcompeted by late-successional, mesophytic species. Oak-hickory forests provide keystone resources for a diverse forest wildlife community, and a decline in its dominance will likely impact habitat use and occupancy patterns in the mammal community, but to what extent is unclear. During May-August 2015-2016, I deployed 150 remotely-triggered camera traps in Trail of Tears State Forest (TTSF), Union County, Illinois to investigate single-season, site occupancy patterns and detection probabilities as a function of forest composition and structure for 3 mammals (eastern gray squirrel [Sciurus carolinensis], raccoon [Procyon lotor], and white-tailed deer [Odocoileus virginianus]). I collected microhabitat data at each camera-site and utilized a GIS application to estimate spatial relationships among anthropogenic features and camera-sites. I recorded 404 photographs of 11 endothermic species during 3927 days of survey effort, with white-tailed deer, raccoons, and eastern gray squirrels as the most detected species, respectively. Detection probability of eastern gray squirrels was best explained by the global detection model, indicating no covariate measured explained the variation in detection rates. Raccoon detection probability was best described by a negative relationship with the average temperature recorded during survey period. The best-fitting detection model for white-tailed deer indicated detection probabilities declined throughout the sampling period and across seasons. Eastern gray squirrel site occupancy models received little support, however, ecological land type phase was the most supported model. The best fitting habitat model described a negative relationship between eastern gray squirrel site occupancy probability and coarse woody debris volume. For raccoons, no model with habitat covariates was better fitting than the null model. Raccoon occupancy probability increased with maximum DBH at a site, ground cover, and beech-maple importance values, but decreased with oak-hickory importance values. White-tailed deer occupancy was most positively influenced by ground cover and oak-hickory importance values, but decreased with distance to forest edge, number of understory stems, and beech-maple importance values. My research provides empirical evidence to predictions made regarding the impact of a decline in oak dominance across the Central Hardwood region on a portion of the region’s mammal community. Shifts to late-successional conditions in the Central Hardwood region will likely continue and magnify if forest management approaches continue to minimize the frequency and occurrence of large, anthropogenic disturbances to the forest overstory. A mosaic of forest conditions will be needed to best support a diverse and complete mammal community across the region.

Os efeitos da perda de hábitat e da fragmentação já estão bem estabelecidos na literatura, porém pouco se sabe sobre como o uso do habitat pode ser modificado em função de alterações na quantidade e na qualidade da mata. Este trabalho tem como objetivo compreender melhor como a perda e modificação do hábitat afeta o uso do espaço pelas aves de subosque na escala da comunidade e população. Para isso, utilizamos um banco de dados de aves, que inclui informação sobre captura e recaptura de 2.121 indivíduos, coletados em um gradiente de cobertura florestal, com fragmentos de 2,85 ha a 10.000 ha, localizados em Mata Atlântica no estado de São Paulo. As análises foram feitas por espécie, por grupos de espécies de acordo com guilda trófica e grau de sensibilidade a alterações antrópicas, e para a comunidade de forma geral. Para testar se a qualidade da vegetação influencia a abundância e frequência de uso do espaço, utilizamos dados de três áreas de florestas contínuas, com diferentes níveis de alteração (floresta secundária, mata com corte seletivo e mata madura). Observamos que, apesar de haver uma queda na abundância total da comunidade em fragmentos pequenos, há um aumento na frequência de uso do ambiente, evidenciado pelo aumento na taxa de recaptura com a perda de habitat. O mesmo padrão foi encontrado para aves com média sensibilidade, frugívoras e insetívoras. Aves com baixa sensibilidade, no entanto, mostraram maiores abundancia e taxa de recaptura em fragmentos menores e áreas mais desmatadas. Verificamos também que a frequência de uso do espaço foi maior em floresta secundária, bem como a abundância de aves. Nossos resultados sugerem que além de conduzir a uma perda de espécies e de indivíduos, a perda de hábitat também leva ao aumento na intensidade do uso do habitat. Este aumento se dá em particular para espécies de média sensibilidade, uma vez que espécies com alta sensibilidade estão ausentes de fragmentos pequenos. Observamos que, além da matriz como barreira à dispersão, a própria qualidade da mata se apresenta como um fator que modifica o uso do habitat. Esta variação no uso do habitat pode impedir que as mesmas desenvolvam seu papel de dispersores de sementes entre os fragmentos. Por fim, estudos sobre uso do espaço hoje praticamente não abordam a intensidade com que o hábitat é utilizado pelos indivíduos, sendo uma informação importante que pode ser facilmente estudada através de dados de captura e recaptura, geralmente subutilizados / The effects of habitat loss and fragmentation area well established in the literature, however not much is known about how habitat use can be modified due to alterations in habitat quantity and quality. This work aims to understand how habitat loss and changes to habitat quality affects the habitat use by understory birds at the community and population level. We analysed a database, which includes capture and recapture data from 2.121 individuals, of birds sampled along a gradient of Atlantic Forest cover and fragment size varying from 2,85 ha to 10.000 ha in the São Paulo State. We analysed data by species, by groups of species according to their trophic guilds and sensitivity to habitat change, and at the community level. To test if the quality of vegetation affects species abundance and the intensity of habitat use, we used data from three continuous forests, with different levels of habitat modification (secondary forest, forest selective logged and mature forest). Our results show that, although the total species abundance decreased in small fragments, the intensity of habitat use increases, as shown by the increase in recapture rates with habitat loss. This pattern was observed for birds with high and medium sensitivity, frugivorous and insectivorous. Species with low sensitivity, however, showed higher species abundance and recapture rate in small fragments. We also show that the frequency of habitat use was higher in secondary forest, as was the abundance of individuals. Our results suggest that habitat loss not only leads to smaller populations, but it also modifies the intensity of habitat use. This pattern was mostly observed for species with medium levels of sensitivity, as highly sensitive species were no longer present in small patches. Our results indicate that the matrix may act as a barrier for dispersal, but habitat quality may also further influence habitat use. This variation in habitat use can affect birds ability to disperse seeds between the fragments. Finally, studies of habitat use almost never focuses on the intensity of habitat use, even though this information can be easily evaluated using capture and recapture data, with the latter being usually underused

Rotas por veículos foram realizadas para que a riqueza, a abundância e o uso do habitat por rapinantes diurnos no Parque Estadual Veredas do Peruaçu (30.702 ha), Minas Gerais, fosse estudada. Onze rotas foram selecionadas e distribuídas em dois habitats: cerrado (n=6) e uma associação entre vereda e cerrado (n=5). O estudo foi realizado entre setembro de 2010 e julho de 2011 por três observadores na carroceria de um veículo 4x4 dirigido a uma velocidade média de 20km/h, totalizando 2772 km percorridos. Ao longo das transecções foram obtidos 681 registros (0,24 rapinantes/km), referentes à 20 espécies, contudo, Falco peregrinus foi registrado ao longo das amostragens ad libitum e elevou a riqueza. As cinco espécies mais comuns foram Heterospizias meridionalis, Caracara plancus, Milvago chimachima, Rupornis magnirostris e Falco femoralis, as quais totalizaram aproximadamente 20% da riqueza local e 80% de todos os registros. As cinco espécies mais raras (Leptodon cayanensis, Elanus leucurus, Accipiter bicolor, Micrastur semitorquatus e Falco sparverius), juntas, representaram menos de 1% das detecções. As aves de rapina foram mais abundantes ao longo da estação chuvosa (n=438), se comparada à seca (n=243). As veredas apresentaram maior riqueza (19 contra 10 taxa) e quase três vezes mais registros (n=466; 0,36 rapinantes/km) que o cerrado (n=215; 0,14 rapinantes/km). O número de detecções computadas para as manhãs foi maior do que as anotadas para as tardes. Baseado no &Delta;AICc, os modelos que melhor explicaram a riqueza e abundância dos rapinantes continham todas as três variáveis independentes (habitat, estação e período do dia). Rotas por veículos aparentaram ser indicadas para estudos com rapinantes em habitats similares a savanas. O cerrado local apresentou menor riqueza e abundância se comparado às veredas, entretanto, o habitat ocupa grandes proporções e deve possuir importante papel na conservação dos rapinantes. A vereda delimita o parque ao norte e nela foram registradas espécies restritas ao ambiente, como a águia-cinzenta (Urubitinga coronata), taxon mundialmente ameaçado de extinção. Tal fato eleva a importância do habitat para o grupo e, consequentemente, para toda a comunidade. Sugere-se que o parque tenha sua área aumentada em sua porção norte, considerando que, adicionalmente, grande parte da água corrente provém desta porção. / Raptors were surveyed by road transects in order to assess local species richness, abundance and habitat use in Parque Estadual Veredas do Peruaçu (30,702 ha), a reserve that lies in the Cerrado Region of southeastern Brazil. Eleven road transects of 7km each were delimited in two different habitats: cerrado (n=6) and an associated vereda-cerrado (n=5). Transects were sampled between September 2010 and July 2011 by three observers on the back of a 4WD pickup truck, driven at an average speed of 20km/h - total of 2772 km. There were computed 681 records along road transects (0.24 raptors/km), belonging to 20 species. One other species (Falco peregrinus) was recorded exclusively along ad libitum sampling. The five most common raptors were Heterospizias meridionalis, Caracara plancus, Milvago chimachima, Rupornis magnirostris and Falco femoralis, which totalized approximately 20% of local richness and 80% of all records. The five rarest raptors in the study area (Leptodon cayanensis, Elanus leucurus, Accipiter bicolor, Micrastur semitorquatus and Falco sparverius), together, represented less than 1% of all records. Raptors were more abundant during rainy season (n=438) than during dry season (n=243). The veredas presented higher species richness (19 against 10 taxa) and almost three times more records (n=466; 0.36 raptors/km) than the cerrado (n=215; 0.14 raptors/km). Mornings held higher number of records when compared to afternoons. Based on &Delta;AICc the models that better explained species richness and raptors\' abundance had all three explanatory variables (habitat, season and day period) considered. Road transects seemed to be recommended for surveys in savannah like habitats. The local cerrado presented lower raptor richness and abundance when compared to the vereda, however it occupies most of the region and it should have an important role on raptors\' conservation. The vereda delimit the reserve northern boundaries and it holds some locally restricted taxa, like the Crowned Eagle (Urubitinga coronata) a species globally threatened of extinction. This makes the habitat even more important for these species and, consequently, to local ecological community. The fact that the vereda is protected only by one margin is a worrying issue and it is suggested that the reserve boundaries augment northwards, once important water supply also comes from there.

Um dos primeiros passos para a conservação e uso racional de um ecossistema é o inventário da sua fauna. A região Neotropical abriga 24% das espécies de mamíferos do planeta, ou seja, cerca de 1.340 espécies, das quais 199 ocorrem no Cerrado. Os objetivos deste estudo foram conhecer a fauna de mamíferos de médio e grande porte da Estação Ecológica de Itirapina (EEI), um remanescente de cerrados abertos na região central do estado de São Paulo, e verificar se há um padrão de ocupação de habitats por parte dessas espécies. Foram obtidos 75 registros com a contagem visual e 77 com as armadilhas fotográficas detectando nove espécies pelo método de contagem visual e dez pelo método de armadilhas fotográficas, totalizando 13 espécies. Três espécies foram avistadas por terceiros, totalizando, com estudos anteriores, 22 espécies para a EEI. Em nenhum dos métodos foi atingida a assíntota nas curvas de rarefação de espécies, o que sugere que cada método deve resultar em um maior número de espécies com maior tempo de amostragem. A taxocenose de mamíferos da EEI foi comparada a outras taxocenoses de Cerrado do Brasil, sendo que a área das localidades parece determinar em grande parte (63%) a riqueza de espécies de mamíferos de médio e grande porte. Uma análise de agrupamento utilizando a composição destas resultou em dois grupos distintos: um contendo as localidades de área relativamente menor e do sudeste e outro contendo as demais localidades, com áreas iguais ou maiores a 10.500 ha. A EEI mostrou maior similaridade com a gleba Pé do Gigante do Parque Estadual de Vassununga, SP, cujas riquezas foram as menores. Quanto ao uso do ambiente na EEI, os resultados obtidos pelos dois métodos de amostragem seguiram o mesmo padrão geral, onde tanto o número de espécies quanto o de registros foi maior na fisionomia mais aberta e decresce nas fisionomias mais fechadas. O teste de &#967;² para o número de registros em cada fisionomia foi significativo para os dois métodos, indicando que os mamíferos não estão distribuídos de acordo com a disponibilidade das fisionomias. Os resultados dos testes foram significativos apenas para Euphractus sexcintus e Lepus europaeus, mas em apenas um dos métodos de amostragem. O campo limpo foi a fisionomia que resultou nos maiores valores de &#967;² parcial em todos os testes, e aparentemente tem grande importância para a taxocenose de mamíferos de médio e grande porte da EEI. / A crucial step for the conservation of any ecosystem is a complete faunal survey of the area. The Neotropics harbor 24% of the world’s the mammalian species, approximately 1.340 species, 199 of wich occur in the Cerrado. The objectives of this study were to sample the mammals of medium and large size at the Estação Ecológica de Itirapina (EEI), a Cerrado remnant in the center of the State of São Paulo, south-eastern Brazil, and to verify patterns of habitat use in these species. Two different sampling methods were used, visual censuses and camera-trapping. Seventy five records were obtained using visual censuses and detecting nine species and seventy seven by camera-trapping, detecting 10 species, totalling 13 species. Aditional three species were sighted by other researchers, raising the number to 22 species for EEI, with other two studies. None of the methods resulted in asymptotic rarefaction curves, which suggests that both methods will detect mores species with a longer sampling time. The species richness and composition of the EEI mammal assemblage was compared to other five Cerrado assemblages and the area of surveyed localities seems to determine, in great part (63%) the mammal species richness. The cluster analysis resulted in 2 main groups, one containing the localities with relatively small areas from the southeast Brazil, and the second group with the remaining localities, with areas equal or greater than 10,500 ha. EEI was most similar to the Pé do Gigante Cerrado Fragment in Parque Estadual de Vassununga, São Paulo, and both fragments exhibited smallest species richness. What regards to habitat use results of both methods suggest the same general pattern, where both the number of records and the number of species are greater in open physiognomies and decrease in more closed habitats. A &#967;² test for the number of records in each physiognomy was significant for both methods, confirming that Cerrado mammals are not distributed according to the availability of the physiognomies. Taking into account each species individually, only Euphractus sexcinctus and Lepus europaeus showed significant deviation from expected, and only for one of the sampling methods (visual census for E. sexcinctus and camera trapping for L. europaeus). In all tests the campo limpo physiognomy showed the highest partial &#967;² values, and seems to have great importance to the medium and large size mammal assemblage of the EEI.

The central objective of this thesis is to better understand early life history dynamics of salmonids in large regulated rivers. I studied spawning, incubating, and age-0 life stages of rainbow trout in the Lee’s Ferry reach of the Colorado River below Glen Canyon Dam, AZ. My first objective was to evaluate the effects of hourly fluctuations in flow on nearshore habitat use and growth of age-0 trout. Catch rates in nearshore areas were at least 2- to 4-fold higher at the daily minimum flow compared to the daily maximum and indicated that most age-0 trout do not maintain their position within immediate shoreline areas during the day when flows are high. Otolith growth increased by 25% on Sundays in one year of study, because it was the only day of the week when flows did not fluctuate. My second objective was to evaluate the effects of flow fluctuations on survival from fertilization to a few months from emergence (early survival). Fluctuations were predicted to result in incubation mortality rates of 24% in 2003 and 50% in 2004, when flow was experimentally manipulated to reduce trout abundance, compared to 5% in 2006 and 11% in 2007 under normal operations. Early survival increased by over 6-fold in 2006 when egg deposition decreased by at least 10-fold. Because of this strong compensatory dynamic, flow-dependent incubation mortality in experimental years was likely not large enough to reduce the abundance of age-0 trout. My final objective was to determine how flow, fish size and density effects habitat use, growth, and survival of age-0 trout. Apparent survival rates from July to November were 0.18 (2004), 0.19 (2006), and 0.32 (2007). A stock synthesis model was developed to jointly estimate parameters describing early life history dynamics, and indicated that early survival was lower for cohorts fertilized during the first half of the spawning period and was negatively correlated with egg deposition, that movement of age-0 trout from low- to high-angle shorelines increased with fish size, and that survival varied by habitat type and over time in response to flow changes from Glen Canyon Dam. / Science, Faculty of / Zoology, Department of / Graduate

Informing monitoring and surveillance efforts for invasive species is crucial for effective management to minimize negative ecological, economic, and social impacts. Management responses to the invasion of silver carp (Hypophthalmichthys molitrix) and bighead carp (H. nobilis)—collectively termed bigheaded carps—in the upper Illinois waterway involve removal efforts to maintain low abundance near the invasion front and surveillance to detect any spread farther upriver. However, the efficiency and effectiveness of these efforts may be limited by the low abundance and patchy distributions of bigheaded carps near the invasion front, as well as the spatial scope of the gears used to sample them. My thesis applied mobile hydroacoustic sampling to understand the factors influencing the distributions of silver carp near the invasion front in the upper Illinois waterway, and factors influencing the observability and detectability of fishes with this gear. These results will be useful to inform removal and surveillance efforts in the upper Illinois waterway and reduce the risk of further upstream invasion into the Great Lakes. To assess factors influencing abundance, bi-monthly (March – October) mobile hydroacoustic sampling was conducted from 2017-2018 and paired with continuous environmental variable data collection in the Dresden Island and Marseilles pools, the upper two invaded pools of the Illinois waterway. Environmental variables collected were water temperature, dissolved oxygen, total dissolved solids, turbidity, chlorophyll-a concentration, blue-green algae concentration, and depth. Logistic and multiple regressions were performed within pool during each sampling period to describe factors associated with silver carp presence, and factors associated with their densities in areas where they were present. Silver carp densities fluctuated across space and time, suggesting environmental variables influenced distributions. Depth was the most consistent predictor of silver carp presence and density, with shallower depths associated with higher abundance. Specifically, in summer months, areas in large backwater habitats with lower temperature, higher dissolved oxygen, and higher chlorophyll-a concentration tended to hold more silver carp. These results suggest removal efforts should be focused on shallower depths within sites, and should target areas in backwaters with lower water temperature, higher dissolved oxygen, and higher chlorophyll-a concentrations during summer months. To inform bigheaded carp surveillance efforts upstream of the invasion front, the influence of factors of vessel speed, wind speed, and beam compensation on the observability (the proportion of the true abundance of fish that fall within the acoustic beam) and detectability (the proportion of observable fish that are detected) of fishes in large rivers with mobile horizontal hydroacoustic sampling was also evaluated. To accomplish this, 23 sets of paired trawl floats (buoys) were deployed in a 2.1 km stretch of the Dresden Island Pool, and four sets of replicate hydroacoustic surveys were conducted past them at three vessel speeds. Mean wind speed during each survey was obtained, and proportion of buoys observed and detected were quantified under two different maximum beam compensation values (6 dB and 12 dB). Results revealed significant interactions between vessel speed and wind speed, with observability increasing with wind speed at slow and fast vessel speeds, but decreasing at intermediate vessel speeds. Detectability decreased with greater wind speed across all vessel speeds, with more pronounced effects at slower vessel speeds. Significant positive effects of beam compensation occurred for both observability and detectability. These results suggest, to improve detection of bigheaded carps, sampling should be focused on calm days under moderate vessel speeds when analyzing with standard beam compensation values. Broadening beam compensation was shown to increase observability and detectability, but needs further investigation. Ultimately, these studies provide valuable information to assist management efforts to control bigheaded carps near the invasion front in the upper Illinois waterway, and detect any individuals that may spread farther upriver towards the Great Lakes.

The postbreeding period in migratory bird species is an important, but often neglected,area of knowledge. From May-October of 1993-95, I studied the breeding andpostbreeding ecology of 61 adults and postfledging movement of 43 juveniles in aradio-tagged population of Wood Thrushes (Hylocichla mustelina) on the U. S. MarineCorps Base, Quantico, Virginia (38 30' N, 77 25' W). Fledglings became independent 0 0from their parents at 28-36 days post hatching and dispersed 307-5300 m from their natalsites to join flocks of conspecifics. About half (46%) of the young birds stayed in onedispersal site until migrating, but the rest visited other sites. In 40 instances, 15fledglings moved up to 6 km out of the dispersal site and, after 1-5 days, returned to thelocality occupied before initiating the movement. After dispersal, fledglings' positions(n = 556) occurred in (1) second growth and sapling stage sites at the edge of forested areas [52%], (2) gypsy moth (Lymantria dispar) damaged deciduous forest [21.8%], (3)Virginia pine (Pinus virginianus) forest that had a heavy understory of young deciduous trees and an open canopy [15.6%], and (4) mature mixed forest [10.6%]. Most fledglings(73%) left the Marine Base in September at the mean age of 81 days. After finishing breeding, adult Wood Thrushes underwent molt that extended from late July to early October. Flight-feather molt lasted on average 38 days and may have impaired flight efficiency in some individuals. Of 30 observed adults, 15 molted in the same area where they nested, and 15 moved 545 to 7290 m from their nesting sites. Molting sites were located in areas with a larger number of pines, less canopy cover, fewer trees with dbh>38 cm, and a denser understory strata than nesting sites (P < 0.1). My data suggest that a conservation strategy that focuses on identifying and protecting nesting habitat in the temperate region, although important, is incomplete at best if the events and needs during the post reproductive and post fledging periods are not considered. / Ph. D.