<p> Like many imperiled fishes, the endangered Pahrump Poolfish (<i> Empetrichthys latos latos</i>) is managed in <i>ex situ</i> refuges. I investigated life history characteristics of females from two such populations at Lake Harriet and Shoshone Stock Pond. Lake Harriet is a relatively large lake with low fish densities located at relatively low elevation and low latitude, while Shoshone Stock is a small pond with high fish densities at a higher elevation and latitude. Females from the Lake Harriet population were larger, and had greater fat content, reproductive allocation, and &lsquo;clutch&rsquo; size than females from the Shoshone Pond population. This divergence, which occurred in three decades, may result in a phenotypic mismatch if the fish are used as a source for restocking their native habitat or stocking new refuges. </p><p> Poolfish conservation may require establishing new populations; however, many sites are inhabited by non-native fish and/or other protected fish species. Thus, managers may wish to consider establishing multi-species refuges that may even already include undesirable species. I established experimental communities that included allopatric and sympatric communities of Poolfish, Amargosa Pupfish (<i>Cyprinodon nevadensis</i>), and invasive Western Mosquitofish (<i>Gambusia affinis</i>). Pupfish persisted in sympatry with both poolfish and mosquitofish, but had higher juvenile production when maintained in allopatry. By contrast, poolfish juvenile production was high in allopatry, but virtually absent in the presence of other species. </p><p> To evaluate the generality of these findings, I established experimental allopatric and sympatric communities of poolfish or pupfish with mosquitofish from two populations that differed in body size: Garrett mosquitofish were approximately twice the mass of Wabuska mosquitofish. Poolfish juveniles had high survival in allopatry, but produced virtually no juveniles when sympatric with either of the two mosquitofish populations. Pupfish juvenile survival was higher in allopatry than sympatric with Garrett mosquitofish, which in turn was higher than sympatric with Wabuska mosquitofish. These results were consistent with the earlier experiment suggesting that poolfish were functionally extirpated but pupfish maintained substantial production in the presence of mosquitofish. These findings suggest that poolfish should be maintained in single species refuges, but that multi-species refuges may protect imperiled pupfish species.</p>

<p> The white shrimp <i>Litopenaeus setiferus</i> is harvested throughout its range in the western Atlantic Ocean and is particularly important in supporting a commercial fishery in the northern Gulf of Mexico (nGoM) concentrated in Texas and Louisiana. Mortality rates of young (juvenile) white shrimp in estuarine nursery areas may have a large effect on the adult population. Understanding the temporal and spatial variation in mortality rates and other population parameters (i.e., density, growth rate, and secondary production) is important for the management of white shrimp in determining how estuaries function as nursery areas and can also be useful in identifying coastal habitats that best support white shrimp populations. My objective was to examine and describe the population dynamics, with a focus on mortality, of juvenile white shrimp populations in Sabine Lake, an estuary of the nGoM. I estimated density, growth, mortality, and secondary production in three areas along the estuarine salinity gradient. Most of these metrics were generally higher in areas of high salinity and total secondary production was relatively high in all three areas. Approximately half of the potential fish predators collected in my study area had preyed upon juvenile penaeid shrimps based on an examination of fish gut contents. Given their abundance within estuaries of the nGoM, these fish predators may have a large impact on populations of juvenile white shrimp. Mortality rates of juvenile white shrimp estimated with mark-recapture and length-frequency data were relatively similar to each other and within the range of the few values previously reported. Two commonly used age-based mortality estimators originally developed for fishes provided reasonably unbiased mortality rates for juvenile penaeid shrimps, although both estimators provided relatively high biased estimates at the lowest mortality rates considered. </p>

<p> Animals that live in groups face a dual challenge of effectively exploiting their environment while at the same time maintaining cohesion with other group members. Maintaining cohesion requires group members to come to consensus about when and where to move, despite the fact that they may not always agree. In this thesis, I investigate how individuals in groups make movement decisions, and how these individual decisions scale up to group-level properties. Using a laboratory experiment with golden shiners (<i>Notemigonus crysoleucas </i>), I first investigate the interaction network over which information spreads, finding that decisions are better predicted by whom individuals can see rather than whom they are close to, with potential consequences for the global spread of information (Chapter 2). I then investigate collective movement behavior in the wild using high-resolution GPS data from members of a troop of olive baboons (<i>Papio anubis</i>). I first show that baboons are consistent in the spatial positions they occupy within the group, and that the observed patterns may be understood based on a very simple mechanism by which individuals maintain cohesion with different numbers of their neighbors (Chapter 3). By quantifying how group members move relative to one another, I then show that baboon movement decisions are consistent with a shared decision-making process, rather than despotic leadership by dominant individuals, and that the patterns of decision-making are consistent with simple models of collective motion (Chapter 4). Finally, by incorporating a fine-scale, three-dimensional reconstruction of the habitat through which the baboons move, I show that habitat structure, in addition to social factors, also exerts an important influence on individual movement decisions, resulting in changes in the emergent structure and movement of the group (Chapter 5). Taken together, these results highlight that by combining high-resolution animal tracking, remote sensing, and analytical methods, we can begin to extend our understanding of collective animal movement from laboratory studies to complex animal societies living in the wild.</p>

<p> Life-history research explores how environmental variation and physiological constraints shift the energy allocation decisions to competing functions of the body among individuals and populations. Variation in the energy available across the environmental landscape is of particular interest as it dictates the amount energy available for organisms to acquire and allocate to their life-history. Additionally, individuals vary in their ability to capture and assimilate energy from the environment. This dissertation seeks to understand how environmental variation in energy availability and physiological constraints of attaining this energy shape allocation decisions among individuals and populations. I examined diamond-backed watersnakes (Nerodia rhombifer) from five populations that differed in their access to energy resources of prey. I compared measures of reproduction and growth to detect the presence of differential allocation decisions. I tested whether differences in allocation among individuals and populations arose as a constraint of their ability to assimilate energy resources in the environment, or as a function of the variation in prey characteristics across the landscape. Lastly, I measured how variation in reproductive investment may limit energy acquisition among individuals.</p>

<p> Seabird populations are declining worldwide. Two major causes of decline are non-native predators and seabird (indirect or direct) overexploitation; these have been suspected but not assessed in the Caribbean region, including in the southern Grenadine islands. Using cameras, questionnaire surveys, and nest monitoring, I explored the potential of invasive predator presence and seabird harvest in affecting nesting productivity of five species in five uninhabited islands during 2014&ndash;2017. Overall, nesting productivity increased (n = 567 nests) despite the confirmed presence of non-native rats. However, my survey of Grenada residents (n = 32 responses) and physical evidence found on the islands (e.g., gun shells) suggest that seabird harvest still occur and may be responsible for previous declines. Additionally, most contributors to harvest did not seem aware of any seabird-protective laws. Therefore, I recommend establishing a community-based monitoring program that educates and empowers residents to prevent seabird harvest and continue nest monitoring. </p><p>

<p> Western Screech-Owls (<i>Megascops kennicottii</i>) are small, nocturnal owls that occur year-round throughout most of the Trans-Pecos region in West Texas, and are associated with deciduous woodlands and riparian habitats. Roadside nocturnal callback surveys, were established along well-defined trails within the Davis Mountains Preserve, and took place from May until August for two years to ascertain the relative abundance of Western Screech-Owls within the property. Abundance indices (birds detected/visit/station) were calculated for both survey seasons. Radiotelemetry was incorporated to determine certain life-history information, while habitat was characterized from both survey and telemetry sites. Occupancy models with detection probabilities were created using Program PRESENCE based on collected presence-absence data. Based on criteria for suitable screech-owl habitat previously defined in the literature, several models were also generated within ArcGIS to compare with the field data and to help predict areas of occupancy within the property. Owl abundance averaged 1.05 owls/point in 2013 and 0.82 owls/point in 2014. Six owls were marked during the study, and 3 of 6 locations were confirmed as daytime roosting sites. Statistically, there were no significant differences between occupied survey sites and telemetry sites, with only the percentage of canopy being significantly different with a mean difference of 9.32, <i> t</i>₄₂ = -2.36, <i>p</i> = 0.023. Both owls and transmitters were discovered in dense juniper-oak mottes on north-facing slopes, indicating a preference for dense, short, closed canopy areas for roosting. Overall vegetative ground cover for the two seasons averaged 54% and 48%, respectively. It was concluded that none of the included covariates in the models influenced either occupancy or detection probabilities, though detectability demonstrated an overall trend based on time of night. GIS models did not reveal any variable that strongly influenced owl occupancy but did correspond with results from previous literature.</p>

Cormorants are piscivorous birds with a daily food intake (DFI) of approximately 500 g. They are a protected species under the Wildlife and Countryside Act 1981. In the UK, the number of over-wintering, inland cormorants increased steadily between 1970 and 1987, at a rate of between 5 and 10 % per annum. An increase of 74 % occurred between winter 1987/88 and 1990/91, and the population is still believed to be rising. The population growth was observed in all regions of the UK, on all habitat types. As cormorants exploited new habitats, ornithologists welcomed their increased presence. This contrasts with the views of angling bodies, who assert that the presence of cormorants, feeding daily on their fisheries, has a damaging impact on fish stocks with inevitable financial losses. Due to a lack of effective non-lethal control methods, the angling bodies wish to see the cormorants removed from the protected species list so their inland numbers can be controlled. Ornithologists insist that there is no scientific evidence proving cormorants are damaging to inland fisheries and so are opposed to any culling. A review of previous cormorant studies was undertaken to evaluate information on their ecology, feeding behaviour and predation impact. The general conclusion was no study had been able to prove cormorant predation damages fish populations, because few studies had moved beyond determining the mass of fish removed by the birds over the particular study period. No assessment had been made of the impact of that fish removal on the fish population dynamics and the angling performance of the fishery. This highlighted the requirement for research into the impact of cormorant predation on inland fisheries. This study was formulated to estimate cormorant predation impact on fisheries in a more realistic and robust manner than had previously been undertaken. The principal objective of the study was to integrate fish population and cormorant feeding dynamics data on specific fisheries (study sites) in such a way as to quantify, where possible, the full impacts of the cormorant predation. This required the following criteria at each study site: - evaluation of the historical status of fish and cormorant populations; - determination of the population and community dynamics of the fish stocks; - analysis of the angling effort and angling performance; - identification of the species, and estimation of the numbers and sizes of fish consumed and wounded by cormorants, and comparison with the numbers and sizes of the fish populations present; - determination of the occupancy on, and use by, cormorants at the selected sites. The work programme ran between September 1995 and July 1998, covering three winters of cormorant predation. At each study site, the cormorant feeding dynamics were assessed by detailed feeding observations and cormorant counts. This enabled data to be collected on the species, size and amount of fish being ingested during each foraging bout, and the diurnal and seasonal patterns of cormorant occupancy. As feeding observations were unable to completed at each site everyday, a modelling system was designed, using a Monte Carlo Simulation (MCS), to estimate the number and mass of fish being removed from the site over the whole winter period. The fisheries data were collected by electric fishing, seine netting, hydro-acoustics and angler catch analysis. The actual methods used at each site were dependent upon the physical conditions present. The data were analysed for fish population dynamics, including length frequency of species, year class strength, natural mortality rate and growth indices; and for angling performance, including catch per unit effort and the relative importance of species. Combining site-specific data for the fish species composition, and the length frequency distribution from fisheries surveys and the cormorants' diet, allowed preliminary predation impact assessment. Reconstruction of life tables from the fisheries data allowed integration of the cormorant feeding data from the Monte Carlo Simulation to assess impact in terms of the numbers of fish consumed on subsequent population densities. This enabled the status of the fish population at each study site to be shown, with and without cormorant predation over the three-year period, resulting in a detailed predation impact assessment. The fisheries studied were located in two regions of the UK, the Midlands and the North West of England. This enabled the research to be completed in two distinct geographical areas, with known and established over-wintering cormorant roosts. The Midland study sites were Holme Pierrepont Rowing Course, Colwick Park Trout Lake and the River Trent. The North West study sites were the lower River Ribble and Grimsargh number 3 Reservoir. These sites encompassed cyprinid and salmonid fish populations, and covered riverine and lacustrine fisheries.

Thesis (Ph. D.)--University of Iowa, 1917. / Bibliography: p. 37.

<p> Community ecology investigates questions related to the density, growth/decline and movements of species over time in given geographical regions. This study investigated similar questions regarding communities of endemic bat species in Connecticut. White-nose syndrome (WNS) has recently killed millions of bats in New England, yet few large-scale conservation efforts have occurred in Connecticut and few data have been published on the status of Connecticut bats post-WNS. This study aimed to: 1) survey bats persisting in WNS regions to document whether changes have occurred in species biodiversity, richness, distribution and habitat use; and 2) measure seasonality effects from summer through pre-hibernal months. Bat presence and activity were recorded using bat detectors set in grassland and forested habitats, near bodies of water and near anthropormorphic and geologic structures across Connecticut. Bioacoustics data have been analyzed by using Sonobat^&copy; software. Combined, these data show that bat activity varies significantly across habitat type (p = 0.02) and over seasons (p = 0.05). Additionally, these data provide insight regarding relationships between individual species, and clumped species groups, with habitat types and across seasons. Ultimately, these data show how bat communities have changed over time in a post-WNS environment. Combined, these data can help drive future wildlife conservation, outreach, education and management practices.</p><p>

<p> Prey species defend themselves behaviorally and morphologically, and often utilize varied antipredator strategies against dissimilar predator types (i.e. terrestrial vs. aerial). Striped skunks <i>(Mephitis mephitis) </i> spray noxious secretions at predators and advertise this danger with deterrent behaviors and black-and-white aposematic coloration. Evidence suggests skunks are effective at deterring terrestrial mammalian predators but are vulnerable to aerial predators; how skunks assess the risk posed by different predator types, however, has not been examined empirically. I recorded the behavioral responses of skunks to audio playbacks of coyotes and great horned owls (the primary terrestrial and aerial predators of skunks, respectively), and peregrine falcons and white noise as controls, as well as to a visual remote controlled model. Skunks engaged in vigilance and running away more often in response to owl vocalizations, suggesting skunks perceive owls as more threatening relative to coyotes. Skunks were more likely to foot stomp and run away in response to the remote controlled model compared with coyote vocalizations, implying visual cues were perceived as more risky than audio cues. This study elucidates how a well-defended mammal can determine which perceived threat is most risky and alter its behavior when its main defense strategy is not successful against all predator types.</p><p>