Habitat modeling can assist in managing potentially widespread but poorly known biological resources such as the federally endangered Delmarva fox squirrel (DFS; Sciurus niger cinereus). The ability to predict or identify suitable habitat is a necessary component of this species' recovery. Habitat identification is also an important consideration when evaluating impacts of land development on this species distribution, which is limited to the Delmarva Peninsula. The goal of this study was to build a predictive model of DFS occurrence that can be used towards the effective management of this species. I developed 5 a'priori global models to predict DFS occurrence based on literature review, past models, and professional experience. I used infrared photomonitors to document habitat use of Delmarva fox squirrels at 27 of 86 sites in the southern Maryland portion of the Delmarva Peninsula. All data were collected on the U.S. Fish and Wildlife Service Chesapeake Marshlands National Wildlife Refuge in Dorchester County, Maryland. Preliminary analyses of 27 DFS present (P) and 59 DFS absent (A) sites suggested that DFS use in my study area was significantly (Wilcoxon Mann-Whitney, P < 0.10) correlated with tree stems > 50 cm dbh/ha (Pmean = 16 + 3.8, Amean = 8+ 2.2), tree stems > 40 cm dbh/ha (Pmean = 49 + 8.1, Amean = 33 + 5.5), understory height (Pmean = 11 m + 0.8, Amean = 9 m + 0.5), overstory canopy height (Pmean = 31 m + 0.6, Amean = 28 m + 0.6), percent overstory cover (Pmean = 82 + 3.9, Amean = 73 + 3.1), shrub stems/ha (Pmean = 8068 + 3218, Amean = 11,119 + 2189), and distance from agricultural fields (Pmean = 964 m + 10, Amean = 1308 m + 103). Chi-square analysis indicated a correlation with shrub evenness (observed on 7% of DFS present sites and 21% of DFS absent sites). Using logistic regression and the Information Theoretic approach, I developed 7 model sets (5 a priori and 2 post hoc) to predict the probability of Delmarva fox squirrel habitat use as a function of micro- and macro-habitat characteristics. Of over 200 total model arrays tested, the model that fit the statistical, biological, and pragmatic criteria postulated was a post hoc integrated model: DFS use = percent overstory cover + shrub evenness + overstory canopy height. This model was determined to be the best of its subset (wi = 0.54), had a high percent concordance (>75%), a significant likelihood ratio (P = 0.0015), and the lowest AICc value (98.3) observed. Employing this predictive model of Delmarva fox squirrel occurrence can benefit recovery and consultation processes by facilitating systematic rangewide survey efforts and simplifying site screenings. / Master of Science

This thesis explores the background of the Klamath Basin Restoration Agreement, a 250 page water use agreement among irrigators, Indian tribes, fishermen, environmental groups, federal and state agencies in the Klamath Basin. The agreement is contextualized in terms of water rights law, Indian law and the Endangered Species Act. The specific details of the agreement are explored. Finally, this thesis evaluates the agreement's merits and suggests a path forward for the agreement to become law.

Thesis (MSc)--Stellenbosch University, 2003. / ENGLISH ABSTRACT: The geometric tortoise, Psammobates geometricus, found in the Western Cape, South Africa, is one of the world's rarest tortoises. At present, it is only known from five isolated geographical areas, one being the Worcester-Tulbagh Valley. The primary aims of my study were to determine the status of geometric tortoise subpopulations along the Nl highway in the Worcester Valley, to determine the population structure of these subpopulations and, if necessary, to provide recommendations for the mitigation ofNI highway-induced impacts on geometric tortoise subpopulations along the road. All Renosterveld remnants along the Nl between Du Toitskloof and Worcester were surveyed for the presence of geometric tortoises, and tortoise density and population structure were determined for each remnant that supports tortoises. Two geometric tortoise subpopulations on opposite sides of the Nl were identified. In the larger remnant of the two, females dominated both the subadult and adult classes, whereas in the other remnant male frequency was slightly higher than female frequency for the adult class only. In both remnants, adult females were significantly larger than adult males in mean carapace length. Mean body mass for adult females and males also differed significantly in both remnants. Analysis of habitat quality showed that both remnants supporting the geometric tortoise are seriously infested by encroaching indigenous and alien plants. The barrier effect of the Nl highway can be mitigated by constructing underpasses, linking the two remnants on opposite sides of the road. Furthermore, road mortality of tortoises can be avoided by erecting tortoise proof fencing along appropriate sections of the highway in the Worcester Valley. By linking the remnants, the largest single area available to the geometric tortoise in the Worcester Valley will also be formed. An additional aim of my study was to investigate the correspondence of geometric tortoise density in individual habitat remnants to the West-to-East change in vegetation composition/characteristics occurring along the Nl in the Worcester Valley. Vegetation composition/characteristics were thus determined in all remnants also surveyed for tortoises. The West-to-East change in vegetation composition/characteristics of the study area indicates that there is a transition from Renosterveld in the West to karroid veld in the East. / AFRIKAANSE OPSOMMING: Die geometriese skilpad, Psammobates geometrieus, wat in die Wes-Kaap, Suid- Afrika, voorkom, is een van die wêreld se skaarste skilpaaie. Vandag word hierdie skilpad net nog in vyf geïsoleerde geografiese streke gekry, een van hulle die Worcester-Tulbagh Vallei. Die primêre doel van my studie was om die status van geometriese skilpad subpopulasies langs die Nl snelweg in die Worcester Vallei te ondersoek en om die populasiestruktuur van hierdie subpopulasies te bepaal. Verder was die doelook, om aanbevelings te maak oor hoe die impakte van die pad op die geometriese skilpad subpopulasies langs die Nl verminder kan word. Al die Renosterveld oorblyfsels langs die Nl tussen Du Toitskloof en Worcester, en binne die bekende gebied van die geometriese skilpad, is deursoek vir geometriese skilpaaie. Populasiedigtheid en populasiestruktuur van die skilpaaie is bepaal vir elke oorblyfsel wat skilpaaie bevat. Twee geometriese skilpad subpopulasies is gekryaan teenoorgestelde kante van die Nl. In die groter oorblyfsel, het wyfies albei die onvolwasse en volwasse klasse gedomineer. In die ander oorblyfsel is gevind dat die frekwensie van mannetjies ietwat hoër was as die frekwensie van wyfies, maar vir slegs die volwasse kategorie. Binne albei oorblyfsels is gevind dat volwassse wyfies betekenisvol groter in gemiddelde doplengte is as volwasse mannetjies. Gemiddelde liggaamsgewig het ook betekenisvol verskil tussen volwasse wyfies en mannetjies in albei oorblyfsels. 'n Analise van habitatkwaliteit het gewys dat beide oorblyfsels wat die geometriese skilpad ondersteun grootliks ingeneem is deur inheemse indringers, sowel as uitheemse plante. Die versperrings effek van die Nl snelweg kan verminder word deur duikwege te skep wat die twee ooblyfsels, op teenoorgestelde kante van die pad, verbind. Bowendien, deur 'n ondeurdringbare heining langs passende dele van die snelweg in die Worcester Vallei op te rig, sal skilpadmortaliteit vermy word. Deur oorblyfsels te verbind, salook die grootste enkele area, beskibaar aan die geometriese skilpad in die Worcester Vallei, gevorm word. 'n Addisionele doel van hierdie studie was om veranderinge in vegetasie langs 'n wes-na-oos gradient binne die studie area te korreleer met die teenwoordigheid van skilpaaie langs hierdie gradient. Vegetasie samestelling/eienskappe is dus bepaal vir elke oorblyfsel wat deursoek is vir skilpaaie. Die wes-na-oos verandering in vegetasie samestelling/eienskappe binne die studie area wys dat daar 'n oorgang van Renosterveld in die weste na 'n karooagtige veld in die ooste is.

published_or_final_version / Environmental Management / Master / Master of Science in Environmental Management

Ce mémoire visait à déterminer les pratiques et les facteurs favorables au succès des réintroductions du carex faux-lupulina (Carex lupuliformis), une espèce vivace menacée poussant dans divers types de milieux humides ouverts du nord-est de l’Amérique du Nord. Pour répondre à cet objectif, des expériences de germination en serres ont été réalisées, et la survie, la vigueur et le microhabitat de chaque plant de carex faux-lupulina connu au Québec (sauvage ou réintroduit) ont été mesurés. Nos résultats montrent que le carex faux-lupulina est une espèce facile à propager ex situ. Une stratification froide (4°C) des semences dans du sable humide pendant un mois induit des taux de germination élevés. Nos résultats montrent également que des pucerons exotiques, ainsi qu’une transplantation dans des habitats inadaptés aux besoins de l’espèce en termes d’humidité du sol et de lumière disponible, ont causé la mort prématurée d’un grand nombre de transplants. Notre programme de réintroduction a permis d’établir une nouvelle population de l’espèce au Québec, et a potentiellement permis de sextupler la taille des populations de la province. Globalement, cette étude a permis de mieux comprendre l’écologie du carex faux-lupulina, et ainsi de faciliter la restauration de ses populations. / This study aimed to determine the practices and factors influencing the success of the reintroductions of false hop sedge, an endangered perennial species growing in North American open wetlands. To do so, germination experiments were performed, and the survival, vigour and microhabitat of each false hop sedge plant known in Québec (wild or reintroduced) were measured. Our results show that false hop sedge is easy to propagate ex situ. A cold stratification (4°C) of seeds in wet sand for one month induced higher germination rates. Our results also show that exotic aphids, as well as a transplantation in habitats that did not meet the false hop sedge requirements in terms of soil moisture and available light, caused the premature death of many transplants. Our reintroduction program allowed the establishment of at least one new false hop sedge population in Québec, and potentially contributed to sextupling the size of the provincial population. Overall, this study allowed us to better understand the ecology of false hop sedge, and thus facilitate the restoration of its populations.

Inbreeding depression is one of the factors that can increase the risk of extinction of small populations, and therefore understanding its effects is currently an important issue in conservation biology. Until recently, few studies on inbreeding depression were carried out in wild populations. These recent studies have highlighted the variability in detecting inbreeding depression among natural populations and the multitude of factors that can influence its expression. Many of the factors affecting inbreeding depression in wild populations remain largely unexplored and most of the recent studies in this area have tended to focus on incidents of inbreeding in populations with a history of large population size. The aim of this study is to investigate the relative importance inbreeding depression has had on individual fitness parameters in a population of New Zealand's Stewart Island robins Petroica australis rakiura introduced to Ulva Island. This island population has historically gone through several population bottlenecks. Four main factors that potentially influence the rate of inbreeding and the extent of inbreeding depression, were investigated: environmental variability, life history stage, genetic load and dispersal. Generalized Linear Mixed Modelling was first used to determine how weather affected nest survival. Weather effects were then incorporated into models containing demographic factors to control for environmental variability, and finally parental, maternal and paternal inbreeding co-efficients (=f) were added to models to determine the relative importance of inbreeding depression. Interactions between inbreeding depression and environmental factors were explored. Three different life history stages were compared to determine the differences in inbreeding depression at each stage as well as cumulative effects over time. The genetic load of the population was estimated using lethal equivalents allowing for standardised comparison of inbreeding depression with other species. The likelihood of inbreeding in the population was also explored by investigating the factors affecting dispersal patterns and evaluating evidence for inbreeding avoidance. Inbreeding depression was found to be mild in the robin population. Weather did not have strong effects on nest survival or interactions with inbreeding. Female age was the only factor interacting with inbreeding, with younger inbred females experiencing significantly reduced offspring juvenile survival. Parental and paternal f did not significantly affect brood survival at any life history stage, however, maternal f showed significant effects on nest juvenile survival with the strongest effect occurring when survival was examined cumulatively over all life history stages. The Stewart Island robin had a relatively low lethal equivalent value compared to the closely related North Island robin and other avian species. This difference was associated with the Stewart Island robin having a low genetic load, most likely due to historical genetic purging during periods of population bottleneck. The Ulva Island robin population did not appear to be avoiding inbreeding through dispersal. Dispersal distance was most strongly influenced by the location of the natal nest of the dispersing offspring. In conclusion, the genetic history of the population was likely to have had the strongest impact on the severity of inbreeding depression in the Ulva Island robin population. The results of the thesis highlight the need to examine a number of factors to be able to explain variability in inbreeding depression among populations.

The current environmental status of the majority of Bermuda’s one hundred sixty-six known caves was investigated. This survey replicated a historical cave study performed in 1983, wherein each was analyzed for positive and negative features. Statistical analysis of the data showed a significant difference between the historical and current survey ratings, with an overall decrease in environmental status. A water quality study was performed on twenty different caves with sea level pools in various locations around the island. Nitrate, nitrite, ammonia, and phosphate levels were measured from varying depths in these caves. Fifteen of these caves were also tested for the presence of fecal bacterial contamination. High nitrate levels were discovered in some of the caves, particularly in surface samples. Additionally, bacterial contamination was detected in some caves. No obvious relationship between cave size or location and contamination existed for any of the pollutants sampled. Three separate caves from this group were dived and analyzed using a Hydrolab Sonde 3 Multiprobe Logger to acquire in situ water column data including depth, temperature, pH, salinity, and dissolved oxygen. Each cave studied had its own unique trends in hydrology at varying depths in the water column. A later water sampling study with a randomized experimental design was created and caves were divided into four classes based on size and location. Surface and subsurface samples were gathered from twelve randomly selected caves, three from each class. Each sample was analyzed for nitrate, nitrite, and ammonia concentrations. The results were analyzed using multiple analysis of variance statistics. A significant difference between the nitrate concentrations in the surface and subsurface water samples was discovered. None of the other comparisons were statistically significant. To represent the data visually, a Bermuda Cave and Karst Information System (BeCKIS) was created using the environmental survey data and water quality information. Some of the maps generated highlighted regions where negative environmental impacts on caves were concentrated geographically, thus demonstrating how this geographic information system could be used as a conservation tool.

Eastern Sand Darter (Ammocrypta pellucida) is listed as Threatened under the Canadian Species at Risk Act. Canadian populations are declining primarily due to the siltation of sandy depositional areas, the preferred habitat of the species. Little other relevant biological information is available for most Canadian populations and only limited information is available for populations in the United States. To supplement the paucity of information, this study collected biological information on A. pellucida during field surveys in 2006- 2007 from 10 sites located around the Big Bend Conservation Area in the lower Thames River, Ontario, Canada. Collected data were used to estimate critical life history traits including: longevity, fecundity, clutch size and number, growth, survival, age-at-first-maturity and cohort age structure. Longevity was 3+ years, with age-at-first-maturity being 1+ for both sexes. A minimum of 2 clutches, were laid per year with an average clutch size of 71 eggs. Average density within in the study area was 0.36 ± 0.11 A. pellucida/m². Quantitative comparison of lower Thames River biological information with a more southerly A. pellucida population in the Little Muskingum River, Ohio, demonstrated little latitudinal variation between the populations. Data comparison suggests that localized environmental factors are affecting biological characteristics, in particular water temperature that may be controlled by differences in riparian cover and/or groundwater input. Field derived life history information was used to create a Leslie matrix model which was used for population viability analysis. Perturbation analyses of reproductive scenarios involving changes in clutch numbers and size and age-at-first maturity found large variations in the finite rate of population growth. Elasticity analyses further indicated that 0+ survival and 1+ fertility were the limiting life history parameters. Thus allowing fish to survive until first reproduction would have the largest overall impact on improving population viability. Inclusion of environmental stochasticity in the model facilitated estimation of extinction probabilities in the range of 0.13 to 0.21 within 100 years. Based on the above, it is recommended that management activities for protection and restoration of A. pellucida focus on habitat protection of nursery and spawning areas.

Eastern Sand Darter (Ammocrypta pellucida) is listed as Threatened under the Canadian Species at Risk Act. Canadian populations are declining primarily due to the siltation of sandy depositional areas, the preferred habitat of the species. Little other relevant biological information is available for most Canadian populations and only limited information is available for populations in the United States. To supplement the paucity of information, this study collected biological information on A. pellucida during field surveys in 2006- 2007 from 10 sites located around the Big Bend Conservation Area in the lower Thames River, Ontario, Canada. Collected data were used to estimate critical life history traits including: longevity, fecundity, clutch size and number, growth, survival, age-at-first-maturity and cohort age structure. Longevity was 3+ years, with age-at-first-maturity being 1+ for both sexes. A minimum of 2 clutches, were laid per year with an average clutch size of 71 eggs. Average density within in the study area was 0.36 ± 0.11 A. pellucida/m². Quantitative comparison of lower Thames River biological information with a more southerly A. pellucida population in the Little Muskingum River, Ohio, demonstrated little latitudinal variation between the populations. Data comparison suggests that localized environmental factors are affecting biological characteristics, in particular water temperature that may be controlled by differences in riparian cover and/or groundwater input. Field derived life history information was used to create a Leslie matrix model which was used for population viability analysis. Perturbation analyses of reproductive scenarios involving changes in clutch numbers and size and age-at-first maturity found large variations in the finite rate of population growth. Elasticity analyses further indicated that 0+ survival and 1+ fertility were the limiting life history parameters. Thus allowing fish to survive until first reproduction would have the largest overall impact on improving population viability. Inclusion of environmental stochasticity in the model facilitated estimation of extinction probabilities in the range of 0.13 to 0.21 within 100 years. Based on the above, it is recommended that management activities for protection and restoration of A. pellucida focus on habitat protection of nursery and spawning areas.

Wildlife biologists and land managers tasked with the recovery of the endangered Lower Keys marsh rabbit (LKMR; Sylvilagus palustris hefneri) were in need of a current population estimate as well as a method to estimate the LKMR population annually. Habitat loss and fragmentation from population growth and development have threatened the existence of the LKMR. Establishing and understanding long-term habitat availability for the LKMR is important for determining causes of historical population declines as well as designing and implementing successful recovery plans. I conducted a range-wide pellet survey and a mark-recapture study to estimate the LKMR population. I evaluated the fit of 5 models and considered the variation in behavioral response model the best model. I correlated (r2 = 0.913) this model's rabbit abundance estimates to pellet density in 11 patches and generated a range-wide population estimate of 317, a western clade population of 257, an eastern clade population of 25, and translocated LKMR populations of 35 and 0 on Little Pine and Water keys, respectively. This prediction equation provides managers a quick, efficient, and non-invasive method to estimate LKMR abundance from pellet counts. To quantify the amount of habitat loss and fragmentation that occurred over the last 50 years, I systematically delineated and compared potential LKMR habitat using 1959 and 2006 aerial photographs. Additionally, I investigated if other factors could have reduced the amount of suitable habitat available for the LKMR with a comparison of habitat loss and fragmentation on a developed island and an undeveloped island. Range-wide, I found that number of habitat patches increased by 38, total class area decreased by 49.0%, and mean patch size decreased by 44.3%. Mean shape index increased by 4.2% and mean proximity index decreased by 13%. Both the 1959 and 2006 connectance indices were low while the 2006 set decreased 12.1%. I observed the same patterns of habitat loss and fragmentation on both the developed and undeveloped islands as I did in the range-wide landscape analysis. I found that LKMR habitat has declined in area and become more fragmented over the last 50 years. Habitat loss and fragmentation by development have directly endangered the LKMR; however, sea level rise and woody encroachment also could have historically caused habitat loss and fragmentation. Although development in LKMR habitat was halted, sea-level rise and woody encroachment could continue to alter LKMR habitat.