Polar bears (Ursus maritimus) are experiencing rapid and substantial changes to their environment due to global climate change. Polar bears of the southern Beaufort Sea (SB) have historically spent most of the year on the sea ice. However, recent reports from Alaska indicate that the proportion of the SB subpopulation observed onshore during late summer and early fall has increased considerably. Previous research suggests that the number of polar bears onshore is linked to sea ice conditions and the availability of subsistence-harvested whale carcasses, which are referred to as bone piles. My objectives were to determine the development and fitness consequences for SB polar bears that come onshore. Furthermore, I aimed to reveal the number of polar bears that come onshore and feed at whale carcasses. I used a combination of genetic and behavioral data collected on SB polar bears from 2010-2013 to determine if onshore behavior developed through genetic inheritance, asocial learning, or social learning. I found that onshore behavior was primarily transmitted via mother-offspring social learning. I used hair samples collected at bone piles near Kaktovik, Alaska from 2011-2014 and genetic capture-recapture techniques to estimate the annual number of polar bears that visited the bone piles and rates of apparent site fidelity to the bone pile. I estimated that as many as 146 (SE = 21) SB polar bears visited the bone piles near Kaktovik in 2012. Annual rates of apparent site fidelity to the bone pile for male polar bears ranged from 0.60 (SE = 0.07) to 0.61 (SE = 0.07), and female rates of apparent site fidelity was 0.69 (SE = 0.19). Lastly, I used quantitative fatty acid signature analysis to estimate the proportion of bowhead whale, ringed seal, bearded seal, and beluga whale in the diets of SB polar bears from 2004-2015. I revealed that polar bears achieved higher body condition by coming onshore and feeding at whale carcasses. Overall, my results indicated that SB polar bears are socially learning from their mother to come onshore and that this behavior is linked to increased body condition.

For the polar bear (<i>Ursus maritimus</i>), a terrestrial carnivore on the highest trophic level in the Arctic marine ecosystem, periods of nutritional restriction or fasting are a characteristic feature of an annual cycle. The overall objective of my thesis was to examine some of the effects of such a 'feast-or-fast' feeding pattern on the body composition, body size and reproductive performance of this Holarctic ursid. As a reproductive strategy, pregnant polar bears occupy maternity dens for up to 6 months between late summer and spring. While in dens, maternal nutrient stores meet all maintenance energy requirements and sustain the nutritional demands of gestation and early lactation. I quantified the nutritional costs of this prolonged 'reproductive fast' in polar bears and examined the effects of variation in maternal body condition on reproductive performance. While fasting, body mass decreased by 43% and of the change in body energy content 93% was attributable to loss of fat. Bears that were fatter priorto denning produced heavier cubs, which would be more likely to survive. Much of the variation in body condition prior to denning was accounted for by age, older females being in better condition. Using an index of milk quality that was closely related to daily milk energy yield, I investigated the independent effects of maternal body condition and age on lactation. Irrespective of condition, older bears tended to produce higher quality milk. While lactation in polar bears is clearly sensitive to body condition, these results also provide strong support for an age-specific increase in reproductive effort among females. Body size is typically a strong determinant of male reproductive success in polygynous mammals such as polar bears. Consequently, theory predicts that mothers in good condition should invest more in male than female offspring in-order to produce large adult males. An underlying assumption of this theory, however, is that early differences in body size among male offspring, such as those apparent by the end of maternal care, will affect their relative adult body size. I tested the validity of this assumption in polar bears and found that, in comparison to females, variation in body size among 2-year-old males was a weak determinant of adult size. I suggest that, in comparison to females, a longer period of growth after maternal care may predispose the adult size of males to a greater degree of environmentally mediated variation. This lack of persistence in relative body size would limit the ability of mothers to affect the adult size of their male offspring, and thus reduce the effectiveness of sex-biased maternal investment as a reproductive strategy in polar bears. One of the principle physiological adaptations enabling animals to go without food for prolonged periods seems to be a heightened ability to minimize the net catabolism of body protein. I quantified changes in the body composition of free-ranging polar bears during the ice-free season. In contrast to previous studies on fasting bears, catabolism of protein appeared to meet a significant proportion of maintenance energy demands in some individuals. (Abstract shortened by UMI.)

The principle objective of this investigation was to develop an understanding of the biochemical and physiological response of free-ranging adult polar bears (<i>Ursus maritimus</i>) to prolonged fasting. A body condition index was developed from two measures, total body mass and straight-line body length, and was used as a covariate in the analyses of all other data. Protein and amino acid catabolism and urea synthesis were significantly lower in fasting bears when compared to feeding bears, and in fat bears when compared to lean bears. The inference from these results is that the energy metabolism in both states (fasting and fat) is one in which lipid is the predominant fuel for energy and nitrogen is conserved. Nutritional state (feeding versus fasting) had no significant effect on the plasma concentrations of non-esterified fatty acid, glycerol, and ketone bodies, or on the plasma ratio of acyl-carnitine to free carnitine. Furthermore, acetoacetate concentration was below the level of detection (<196 [mu]mol/L) in all bears, and â-hydroxybutyrate concentration never exceeded 291 [mu]mol/L. These results suggest polar bears are able to regulate closely the synthesis, release, and use of lipid metabolites without significant alteration in their plasma concentrations. Fasting polar bears showed no evidence of essential fatty acid (EFA) deficiency; the proportions of the diet-derived EFA linoleic (18:2[omega]6) and á-linolenic (18:3[omega]3) acids in the plasma and adipose tissue of fasting polar bears were greater than that in feeding polar bears. Plasma triiodothyronine concentrations and rectal temperatures were lower in fasting bears captured during summer-fall than in feeding bears, which suggests metabolic rates were decreased during fasting to conserve body fuels. Liver glycogen concentrations were found to be higher in fasting polar bears than in feeding bears. Furthermore, the results from intravenous administration of glucose (glucose tolerance test) to polar bears indicated the rates of insulin secretion and clearance in polar bears were slow relative to rates reported for other mammals. The inference from these results is that polar bears are not as dependent on glucose for energy as are other mammals and, as a consequence, are more lax in regulating their body glucose stores.

Many types of ecological studies require identification of individual animals. I developed and evaluated an automated identification system for polar bears (Ursus maritimus) based on their whisker spot patterns. First, I measured the reliability of using whisker spot patterns for identification from polar bear photographs taken in western Hudson Bay. This analysis involved estimating the complexity of each whisker spot pattern in terms of its information content. I found that 98% of patterns contained enough information to be reliable, and this result varied little among three different observers. Based on these results, I implemented a computer-aided identification system for polar bears based on whisker spot pattern recognition. I used standard computer vision techniques to pre-process images and the Chamfer distance transform to compute similary scores between images. In addition, I evaluated the system by testing the effects of photographic quality and angle on system accuracy. I found that excellent and moderate quality/angle provided best results, with system accuracy of 90-95%. These findings suggest that individual identification of polar bears in the field based on whisker spot pattern variation is possible. Researchers studying polar bear behavior or estimating population parameters should benefit from this noninvasive technique.

Pregnant polar bears (Ursus maritimus) construct maternal dens out of snow in the autumn where they give birth to and raise altricial young. In recent years, there has been a decrease in polar sea ice extent and thickness, which has led to changes in denning behavior. One such change in the southern Beaufort Sea (SBS) is that polar bears are selecting maternal den sites on land, rather than on unstable sea ice. This change, coupled with expanding petroleum exploration along Alaska's North Slope, heightens the likelihood of bear-human interactions at maternal den sites. The purpose of this research was to 1) describe polar bears' post-den emergence behavior, establishing a benchmark for comparison to identify behavioral changes associated with climate change and disturbance, and 2) explore factors influencing the efficacy of a currently used den detection method, forward-looking infrared (FLIR). Maternal den sites were observed along Alaska's North Slope from March to April of 2009 and 2010. The mean length of stay at den sites post-emergence was 11.3 ± 7.5 d. The mean date of den emergence was 14 March; abandonment 26 March. Adult females were generally inactive (58.4% out-of-den time) with standing being the most prevalent activity (49.9%). Cubs were generally active (76.7%), playing more than any other activity (45.3%). Bears spent the majority of their time in the den (97.3% for adult females and 99% for cubs) with short bouts of intermittent activity (× = 7 min 42 s). We documented the death of one member of a triplet polar bear litter at its den site. All three cubs showed low activity levels relative to other cubs observed, and one died within one week of den emergence. Necropsy confirmed that the dead cub had a low body weight and was malnourished. Capture later confirmed that the two surviving cubs were also undersized. Triplet litters are often smaller and suffer higher mortality rates than singletons and twins. This cub was not only a triplet but also born following 2 y of record minimum sea ice extent, both of which may have played a role in this cub's death. Concurrent with the den emergence portion of this work, we conducted a separate study to identify limitations and optimal conditions for locating dens using FLIR. We took handheld FLIR images of three artificial dens under varied conditions. We tested variables hypothesized to influence detectability with linear models using a zero-inflated negative binomial distribution. Solar radiation, wind speed, and den wall thickness reduced the likelihood of detecting dens. The negative effect of wind speed on detectability increased with increasing distance. To maximize the efficacy of hand-held FLIR, den surveys should be conducted when solar radiation is <16 w/m2 (night) and when wind speed is <10 km/h (6 mph). Adherence to these guidelines will maximize the protection FLIR can afford to denning bears.

Human-bear interactions are an important consideration of bear biology, as interactions can lead to destruction of property as well as injury or death for both human and bear. Successful analysis of why these interactions occur can lead to appropriate preventative measures and mitigation of further conflict. Bryce Canyon National Park (BRCA) is comprised of relatively poor bear habitat, but a black bear population exists on the Paunsaugunt Plateau, on which the park occupies the eastern edge. Park managers expressed interest in learning more about bear movements and, specifically, bear use of anthropogenic features following a number of human-bear incidents located at backcountry campsites within park boundaries. By analyzing data from GPS radio-collared bears, trail cameras, existing literature, park incident reports and in-depth campsite assessments, we were able to show how bears are using both natural and anthropogenic features on the Bryce landscape. Campsites were assessed for bear habitat, displacement and encounter potential in order to establish an overall human-bear conflict potential. AIC model selection and resource selection functions using GPS collar data showed that bears selected for some anthropogenic features (campsites, springs), while actively avoiding others (trails, roads). Trail camera data, existing literature and park incident reports all pointed toward use of trails. We then considered all data sources used in the analysis and compiled rankings of human-bear conflict potential for each of the backcountry campsites within BRCA, and submitted a detailed report of findings, conclusions and recommendations to NPS personnel. Second, we investigated human-bear interactions at polar bear dens sites on Alaska's North Slope. As parturient female polar bears in the Southern Beaufort Sea subpopulation increasingly construct maternal dens on coastal land features rather than sea ice, they become more likely to interact with industry and other human activity. We wanted to understand what levels of human interaction could lead to disturbance of denning polar bears, and what types of responses were being exhibited by bears following those interactions. We subdivided potential disturbance stimuli into groups based on their size, motion and sound and the used AIC model selection techniques and multinomial logistic regression to analyze records of human-bear interactions at den sites ranging from 1975 through the present day. We found significant probabilities of varying levels of bear disturbance response among a number of stimuli and intensities. However, denning bear families were overall more tolerant of human activity near den sites than expected. Den abandonments were rare, and we documented no cases of reproductive failure following a disturbance event. We hope that our results from the analysis can be used to further enhance management of industry when operating in polar bear denning habitat.

Human-caused ecological change negatively affects the sustainability of many wildlife populations but may be especially challenging for large carnivores and ungulates. Long-term physiological stress may be an important mechanism linking ecological change with impaired health and reduced population performance in these groups. The determination of hair cortisol concentration (HCC) has recently demonstrated potential as a biomarker of long-term stress in humans and domestic animals, and may also represent a practical technique for use in free-ranging wildlife. The objectives of this research program were to: 1) develop and apply an accurate and reliable method for measuring cortisol levels in hair collected opportunistically or remotely from free-ranging grizzly bears (Ursus arctos), polar bears (Ursus maritimus), and caribou (Rangifer tarandus sp.), and 2) to evaluate the utility of HCC as a biomarker of long-term stress (and thus potentially useful conservation tool) in these threatened species. An enzyme-immunoassay (EIA) based technique for measuring HCC in non-human primates was successfully modified for use with small quantities (5-100 mg) of hair representative of samples which may be obtained through opportunistic (e.g. hunting, research captures, archives) or remote (e.g. barb wire snagging) methods in each species. HCC was determined in 151 free-ranging grizzly bears from Alberta, Canada (mean 2.84 pg/mg, range 0.62-43.33 pg/mg); 185 free-ranging polar bears from southern Hudson Bay, Canada (mean 0.48 pg/mg range, 0.16-2.26 pg/mg); in 12 captive Alaskan caribou (R. t. granti) (mean 2.31 pg/mg, range, 1.57-3.86 pg/mg) and 12 captive reindeer (R .t. tarandus) (mean 2.88 pg/mg, range 2.21-3.40 pg/mg) injected either with adrenocorticotropic hormone (ACTH) or saline; and in 94 free-ranging caribou (R. t. groenlandicus) from West Greenland (mean 2.21 pg/mg, range 0.60-6.90 pg/mg). Factors influencing HCC in each species were then explored including: 1) technical considerations for the prudent use of HCC analysis and 2) potential relationships between HCC, biological traits, health, and prevailing environmental conditions. Evidence revealed in this study suggests that, with further research, this technique may show potential as a practical conservation tool for use in free-ranging grizzly bear, polar bear, and caribou populations.

An international trade in endangered species of plants and animals is regulated by CITES (the Convention on International Trade in Endangered Species of Wild Fauna and Flora). Information on trades are available in CITES Trade Database. However, there are a lot of mistakes and discrepancies in this database that lead to inaccurate interpretation of data. The aim of this thesis was to: (1) identify mistakes in the database, record types and their possible ways of interpretation and (2) quantify the volumes of trade based on different ways of data interpretations in selected species of the family Ursidae. There were found 19 different combinations of missing data, except by unlisted unit and traded quantity specified only by one state of the trade and 5 types of record interpretation. Data were compared with the CITES annual records reported by the Ministry of the Environment of the Czech Republic. This thesis contains also recommendations for evaluation of international trade.