Theses and Dissertations at Montana State University (MSU)
Permanent URI for this communityhttps://scholarworks.montana.edu/handle/1/732
Browse
13 results
Search Results
Item Smart wildlife monitoring: evaluating a camera trap enabled with artificial intelligence(Montana State University - Bozeman, College of Agriculture, 2024) Kaltenbach, Taylor Louise Gregory; Chairperson, Graduate Committee: Jared T. Beaver; Jeffrey C. Mosley (co-chair)Wildlife-livestock conflicts, including depredation, disease transmission, and resource competition, present significant challenges to both the ecological and economic aspects of ranching operations. These conflicts can undermine the sustainability of ranching operations as well as the conservation of wildlife in working landscapes. Leveraging timely and precise data on wildlife activity, distribution, and their interactions with livestock are crucial for enhancing ongoing conflict mitigation efforts and to help sustain wildlife on working landscapes. I evaluated the potential of an artificial intelligence (AI)-enabled camera trap to limit false positive images and provide real-time monitoring of wildlife presence while reducing data overload. In Study 1, I compared the performance of a prototype, edge AI-enabled camera trap (Grizzly Systems) with 2 traditional, non-AI camera traps (Browning and Reconyx) at 8 sites across 3 ranches in south-central Montana, USA, from mid-June through mid-September 2023. I also evaluated the influence of site-specific environmental conditions, including air temperature, wind speed, cloud cover, and vegetation type on camera trap performance. The Grizzly Systems camera trap captured fewer false positive images but exhibited a higher rate of missed detections compared to the Browning and Reconyx camera trap models. Across all 3 camera trap models, the probability of positive detections declined with warmer air temperatures and greater wind speeds. In addition, warmer air temperatures positively influenced missed detections by Reconyx and Grizzly Systems camera trap models, but warmer air temperatures negatively influenced missed detections by Browning camera traps. In Study 2, I compared the performance of a cellular-connected AI-enabled Grizzly Systems camera trap, equipped with an automated image processing and notification reduction workflow, to a traditional, non-AI, cellular-connected Reconyx camera trap at 2 sites in south-central Montana, USA from mid-April to mid-June 2023. The AI-enabled, cellular-connected Grizzly Systems camera trap successfully sent real-time notifications of wildlife presence and transmitted significantly fewer false positive images than the cellular-connected Reconyx camera trap. However, the Grizzly Systems camera trap sent substantially fewer notifications of positive detections than the Reconyx camera trap, which are likely attributed to missed detections by the Grizzly Systems camera trap.Item Evaluating bear management areas in Yellowstone National Park(Montana State University - Bozeman, College of Letters & Science, 2022) Loggers, Elise Ahlenslager; Chairperson, Graduate Committee: Andrea Litt; This is a manuscript style paper that includes co-authored chapters.A growing body of research suggests large predators change their behavior near humans in ways that parallel how prey respond to predators; when outdoor recreation increases, avoiding humans becomes more difficult. Restricting human access to reduce detrimental effects of human-wildlife interactions can be an attractive management tool, however, rarely is the efficacy of such measures assessed. In 1982, Yellowstone National Park began instituting short-term, annual restrictions to areas of the backcountry containing important food resources for grizzly bears (Ursus arctos). These areas -- Bear Management Areas (BMAs) -- were intended to reduce human-caused disturbance of foraging bears and improve visitor safety. We sought to assess whether grizzly bears: 1) preferred BMAs with access restrictions more than other areas in YNP and 2) changed their response to sporadic (trail) and predictable (campsite) recreation sites depending on BMA access restrictions. We modeled resource selection of grizzly bears with step-selection functions, based on GPS locations from male and female bears collected from 2000 to 2020. Our analyses demonstrated that grizzly bears differentially selected BMAs, compared to areas outside BMAs, and that selection changed with sex and season. Bears likely prefer BMAs for the resources they contain more than to avoid people as only males changed their selection of BMAs based on access restrictions. Males avoided hiking trails during the day, but preferred trails at night. Females changed their selection of trails depending on human access restrictions and avoided trails in unrestricted BMAs. Combined with previous work, results suggest bears capitalize on the environment to avoid human presence, often with sex-specific strategies. For sporadic recreation, males temporally avoid the perceived risk of people whereas females spatially avoid the perceived risk of people. Although lower-intensity activities often are thought of as compatible with conservation, such recreation may be cryptic, but important, drivers of behavioral change in wildlife.Item High net worth ownership regimes in critical conservation areas: implications for resource governance(Montana State University - Bozeman, College of Letters & Science, 2020) Epstein, Kathleen Elizabeth; Chairperson, Graduate Committee: Julia Hobson Haggerty; Julia H. Haggerty and Hannah Gosnell were co-authors of the article, 'Super-rich landowners in social-ecological systems: opportunities in affective political ecology and life course perspectives' in the journal 'Geoforum' which is contained within this dissertation.; Julia Hobson Haggerty and Hannah Gosnell were co-authors of the article, 'With, not for, money: ranch management trajectories of the super-rich in greater Yellowstone' submitted to the journal 'Annals of the American Association of Geographers' which is contained within this dissertation.; Dissertation contains a article of which Kathleen Elizabeth Epstein is not the main author.Despite the expanding financial power of the global super-rich and their expansive control over natural resources as proprietors of an increasing number of large agricultural properties, geographers have only just begun to assess the influences of wealthy landowners on systems of environmental management. In this dissertation, I examine a set of ownership dynamics related to the acquisition of ranchland properties by high net worth (HNW) individuals in the Greater Yellowstone Ecosystem, a charismatic conservation area in the Northern Rockies, USA. The dissertation deploys a mixed methods approach informed by social-ecological systems theory and insights from the literature on political ecology of the American West to assess HNW ownership regimes at the landscape and property scales from the perspective of an iconic regional resource institution: state-led elk management. The work follows a central conceptual logic related to the evolution of HNW land management, namely that ranch owners and properties interact with local ecologies, social actors, and resource institutions in ways that influence land use strategies and practices over time and space. At the landscape scale, patterns of land-use intensification (e.g., increased use of irrigation) have converged with growing diversification (e.g., increased residential development), to make elk management more complex, as elk encounter a range of push and pull factors across a shifting and diverse landscape of land-use values and practices. A defining characteristic of the trajectory for ranches of the super-rich is that HNW landowners ranch with, as opposed to for, money, though multiple social-ecological factors (markets, property lines, legal institutions, and unpredictable rangeland socio-ecologies) also shape HNW landowners' abilities to realize management goals and visions. Where HNW ownership regimes intersect with shifts in the political and moral economy, conflicts related to public access to wildlife on private lands have emerged. In this context, the work of wildlife managers requires adaptive strategies as wildlife management has become more about managing people - and the psychosocial outcomes of conflict - than managing wildlife. Ultimately, this research argues that the challenges HNW ownership regimes pose for resource governance require strategic engagement with the broader structures of wealth concentration and resource control that have enabled them.Item Wolverine habitat quality, connectivity, and prioritization at the landscape scale(Montana State University - Bozeman, College of Letters & Science, 2019) Carroll, Kathleen Anne; Chairperson, Graduate Committee: Andrew J. Hansen; Andrew J. Hansen, Robert M. Inman and Rick L. Lawrence were co-authors of the article, 'Comparing methods to disentangle habitat predictors for wolverines in the southern extent of their distribution' which is contained within this dissertation.; Andrew J. Hansen, Robert M. Inman, Rick L. Lawrence and Andrew B. Hoegh were co-authors of the article, 'Testing landscape resistance layers and modeling connectivity for wolverines in the western US' which is contained within this dissertation.; Robert M. Inman, Andrew J. Hansen, Kevin Barnett and Rick L. Lawrence were co-authors of the article, 'Prioritizing metapopulation connectivity for wolverines' which is contained within this dissertation.The core of conservation biology is understanding how to mitigate the impacts of anthropogenic activities on species. These impacts are particularly detrimental to isolated and small populations, which face extirpation or extinction without immediate conservation action. For small and isolated populations, protecting connective habitat (e.g., corridors) and facilitating movement is key. Corridor identification requires rigorous planning and appropriate statistical choices to ensure that resulting conservation actions are defensible and best support ecological processes. This manuscript asks: 1) how do different, commonly used statistical methods inform our understanding of species resource selection across scale and between sexes, 2) how does landscape resistance and connectivity differ between resident and dispersing individuals, and 3) what information is important to include in a systematic conservation plan to best support on-the-ground conservation between land trusts, landowners, and other practitioners under future climate change conditions. To address each of these questions we focused on wolverines (Gulo gulo), which exist as isolated metapopulations across the western contiguous United States. Our key findings included that 1) the importance of habitat variables differ only slightly by sex, across selection scales, and across analysis methods, 2) dispersing animals are less sensitive to habitat quality compared to resident animals, and 3) including information that both helps mitigate potential threats and preserves ecological processes is the best approach for connectivity conservation planning. This work represents the most comprehensive wolverine connectivity conservation analyses to date. This research suggests that examining multiple approaches and validating results is critical to generating rigorous and defensible conservation decisions are being made for wolverines, although more studies are needed to validate this in other species. Taken together, this research provides land managers, policy makers, and scientists with guidance for future connectivity analyses, conservation action for wolverines, and a research framework that can be applied to additional species of conservation concern in isolated populations.Item Permeability of three-strand electric fences by black bears and grizzly bears(Montana State University - Bozeman, College of Agriculture, 2018) Johnson, Brittani Justine; Chairperson, Graduate Committee: Lance McNewElectric fencing has been used to deter bears in North America for several decades. Producers have turned to a design of a minimum of five-wire electric fence as their primary solution to reduce livestock depredation and to reduce raids of chicken houses and beeyards. However, these designs are expensive and reducing the number of wires used in a design to three wires would be beneficial. Scientific evaluations of the efficacy of three-wire electric fencing at deterring bears is lacking. In 2015 and 2016, I conducted a study in the Blackfoot Valley of Montana to evaluate the efficacy of rapid-deployment electric fencing designs in deterring bears from baited enclosures. Baited enclosures of two fencing configurations were established at 20 sites in the study area. Each enclosure was systematically energized and unenergized for 3-day periods; and passage into the enclosure was monitored with trail cameras to provide information on effectiveness and permeability. I recorded 134 visits by bears to fenced enclosures during the study seasons of 2015 and 2016. Of these visits, 78 occurred in 2015 and included 57 black bears and 21 grizzly bears. Fifty-six visits occurred in 2016, including 34 black bears and 22 grizzly bears. Black bears and grizzly bears were successful at passing the short fence 48% (95% CI: 32.0 -- 63.6) and 23% (5.0 -- 53.8) of the time, respectively, when it was not energized. When the short fence was energized, black bears were 7% (0.2 -- 33.9) successful in passing, whereas grizzly bears were successful in 25% (5.5 -- 57.2) of attempts. When not energized, both species successfully passed the tall fence design in 58% (95% CI: 27.7 -- 84.8) of attempts. Black bears and grizzly bears successfully entered energized enclosures with tall fences in 30% (95% CI: 13.2 -- 52.9) and 0% (95% CI: 0.0 -- 45.9) of attempts, respectively. Both fence types deterred bears from entering baited enclosures and all fences allowed less than perfect access when unenergized, suggesting that even minimalistic configurations of electric fences may act as barriers to black and grizzly bears. Further study evaluating the effects of rapidly increasing construction of electric fencing is needed to assess landscape level effects on bear movement and habitat selection.Item Human wildlife interactions in the Gallatin Range, Yellowstone National Park, 1973-1974(Montana State University - Bozeman, College of Letters & Science, 1976) Chester, James MichaelItem Modeling the restoration of a metapopulation : implications for resource management(Montana State University - Bozeman, College of Letters & Science, 1996) Berkson, Jim M.Item Waterfowl and hunter use of Freezout Lake game management area, Teton County, Montana(Montana State University - Bozeman, College of Letters & Science, 1979) Lorang, Kenneth DavidItem Elk pregnancy, production, and calf survival in the South Fork of the Flathead River, Montana(Montana State University - Bozeman, College of Letters & Science, 1998) Kastler, Michele AnnItem Using noninvasive genetic sampling methods to estimate demographic and genetic parameters for large carnivore populations in the Rocky Mountains(Montana State University - Bozeman, College of Letters & Science, 2012) Sawaya, Michael Allen; Chairperson, Graduate Committee: Steven Kalinowski; Toni K. Ruth, Scott Creel, Jay J. Rotella, Jeffrey B. Stetz, Howard B. Quigley, and Steven T. Kalinowski were co-authors of the article, 'Evaluation of noninvasive genetic sampling methods for cougars in Yellowstone National Park' in the journal 'Journal of wildlife management' which is contained within this thesis.; Jeffrey B. Stetz, Anthony P. Clevenger, Michael L. Gibeau and Steven T. Kalinowski were co-authors of the article, 'Estimating grizzly and black bear population abundance and trend in Banff National Park using noninvasive genetic sampling methods' in the journal 'PLoS ONE' which is contained within this thesis.; Anthony P. Clevenger, and Steven T. Kalinowski were co-authors of the article, 'Wildlife crossing structures connect ursid populations in Banff National Park' in the journal 'Conservation biology' which is contained within this thesis.; Steven T. Kalinowski, and Anthony P. Clevenger were co-authors of the article, 'Gene flow at wildlife crossing structures in Banff National Park' in the journal 'Molecular ecology' which is contained within this thesis.Healthy carnivore populations are important to maintaining ecosystem balance, but many species are declining globally at disturbing rates due to anthropogenic causes. To effectively manage and conserve carnivores, wildlife managers must be able to obtain reliable estimates of population parameters. Noninvasive genetic sampling (NGS) methods such as hair or scat collection offer new and exciting alternatives to traditional carnivore research methods involving capture, drugging, and handling of animals; however, the potential of NGS methods to answer applied ecological questions has not been fully realized. The main objective of my doctoral research was to develop and apply NGS methods to estimate demographic and genetic parameters for large carnivore populations in the Rocky Mountains. First, I evaluated two NGS methods, hair snares and snow tracking, for cougars (Puma concolor) in Yellowstone National Park. I developed a method to collect hair while following cougar tracks in snow to bed sites and natural hair snags (e.g. thorn bushes, branch tips) from which I demonstrated that samples collected using NGS can provide reliable information on cougar population abundance. Next, I compared the ability of two NGS methods, hair traps and bear rub surveys, to estimate population abundance and trend of grizzly (Ursus arctos) and black bears (U. americanus) in Banff National Park. I found that bear rubs performed better than hair traps for estimating grizzly bear abundance and population growth rates, whereas hair traps worked better than bear rubs for black bears. I then used NGS to examine demographic and genetic connectivity at wildlife crossing structures along the Trans-Canada Highway that bisects Banff National Park. I compared genetic data collected from the bear populations surrounding the highway to data collected at the crossing structures using a novel hair sampling system. This comparison allowed me to show that wildlife crossing structures provided demographic connectivity for bear populations and maintained sufficient gene flow across the highway to prevent genetic isolation. In short, I have demonstrated the power of using an array of NGS methods, alone or in combination, to estimate abundance, gene flow, genetic structure, migration, and population growth rates for large carnivores in the Rocky Mountains.