{"id":{"repo_id":"oregon","oai_identifier":"oai:scholarsbank.uoregon.edu:1794/32744"},"canonical_url":"https://search.dev.ndltd.org/etd/oregon/oai:scholarsbank.uoregon.edu:1794/32744","repository":{"repo_id":"oregon","name":"University of Oregon","base_url":"https://scholarsbank.uoregon.edu/server/oai/request"},"display":{"title":"Simulation-Based Spatially Explicit Close-Kin Mark-Recapture for Genetic Monitoring of Wildlife Populations","abstract":"Population size is one of the most important pieces of information for monitoring conservation efforts and setting sustainable harvest limits, but it is often difficult and labor intensive to estimate. Close kin mark-recapture (CKMR) is a promising new method for estimating population size from genetic data that identifies related pairs of individuals in a sample and estimates population size from these pairs. Unlike widely used capture-recapture methods, CKMR does not rely on recapturing the same individual multiple times. This means that CKMR can be used when recaptures are difficult or impossible, such as studies involving lethal sampling, sampling at a single time point, or highly elusive species. However, a major shortcoming of current CKMR methods is that they do not model the complex spatial and social dynamics that are important in many species of conservation concern. In my dissertation, I develop a novel spatially explicit CKMR method that models limited dispersal, spatial bias in sampling, complex social structure, and uncertainty in kin estimation. I test the method on simulations and demonstrate that it is accurate and robust to misspecified population trend. I then apply the method to elephants in Kibale National Park in Uganda and explore the feasibility of using CKMR methods to improve monitoring. In Chapter II, I develop a novel spatially explicit CKMR method that uses a convolutional neural network trained with individual-based simulations. When tested on simulated populations, the method is able to accurately estimate population size from parent-offspring and half-sibling pairs, even when dispersal is limited and some areas of the landscape are more intensely sampled than others, a situation in which non-spatial methods are negatively biased. The method is also robust to unknown trend in population size. I apply the method to genetic data from elephant dung samples in Kibale National Park in Uganda and find that when kin pairs are assumed to be known without error, estimated population size agrees with previous estimates but with a much smaller confidence interval. In Chapter III, I extend the CKMR method to account for two common situations in populations of conservation concern: social structure with groups of related individuals traveling together and high uncertainty in kin estimation due to a small number of sequenced loci. I then test the feasibility of using these methods for genetic population monitoring of elephants in Kibale. With 14 sequenced microsatellite loci, kin estimation is highly uncertainty and contributes to uncertainty in CKMR estimates. However, the CKMR network with recaptures still performs better than capture-recapture. With a greater number of sequenced microsatellites (56), kin estimation has very low uncertainty and the network performs very well, even when recaptures are not included. The methods I developed in my dissertation make a valuable contribution to genetic monitoring of species with limited dispersal or complex social structure and demonstrate the power of spatially explicit simulation-based inference for conservation biology. This dissertation includes previously unpublished coauthored material.","abstract_html":"Population size is one of the most important pieces of information for monitoring conservation efforts and setting sustainable harvest limits, but it is often difficult and labor intensive to estimate. Close kin mark-recapture (CKMR) is a promising new method for estimating population size from genetic data that identifies related pairs of individuals in a sample and estimates population size from these pairs. Unlike widely used capture-recapture methods, CKMR does not rely on recapturing the same individual multiple times. This means that CKMR can be used when recaptures are difficult or impossible, such as studies involving lethal sampling, sampling at a single time point, or highly elusive species. However, a major shortcoming of current CKMR methods is that they do not model the complex spatial and social dynamics that are important in many species of conservation concern. In my dissertation, I develop a novel spatially explicit CKMR method that models limited dispersal, spatial bias in sampling, complex social structure, and uncertainty in kin estimation. I test the method on simulations and demonstrate that it is accurate and robust to misspecified population trend. I then apply the method to elephants in Kibale National Park in Uganda and explore the feasibility of using CKMR methods to improve monitoring. In Chapter II, I develop a novel spatially explicit CKMR method that uses a convolutional neural network trained with individual-based simulations. When tested on simulated populations, the method is able to accurately estimate population size from parent-offspring and half-sibling pairs, even when dispersal is limited and some areas of the landscape are more intensely sampled than others, a situation in which non-spatial methods are negatively biased. The method is also robust to unknown trend in population size. I apply the method to genetic data from elephant dung samples in Kibale National Park in Uganda and find that when kin pairs are assumed to be known without error, estimated population size agrees with previous estimates but with a much smaller confidence interval. In Chapter III, I extend the CKMR method to account for two common situations in populations of conservation concern: social structure with groups of related individuals traveling together and high uncertainty in kin estimation due to a small number of sequenced loci. I then test the feasibility of using these methods for genetic population monitoring of elephants in Kibale. With 14 sequenced microsatellite loci, kin estimation is highly uncertainty and contributes to uncertainty in CKMR estimates. However, the CKMR network with recaptures still performs better than capture-recapture. With a greater number of sequenced microsatellites (56), kin estimation has very low uncertainty and the network performs very well, even when recaptures are not included. The methods I developed in my dissertation make a valuable contribution to genetic monitoring of species with limited dispersal or complex social structure and demonstrate the power of spatially explicit simulation-based inference for conservation biology. This dissertation includes previously unpublished coauthored material.","abstract_has_math":false,"creators":["Patterson, Gilia"],"institution":"University of Oregon","degree_name":"Ph.D.","degree_level":"doctoral","degree_discipline":"Department of Biology","degree_department":null,"school":null,"contributors":[],"advisors":["Ralph, Peter"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026-06-18","date_published":"2026-06-18","updated_at":"2026-08-21T16:47:20Z","subjects":[],"languages":["en_US"],"rights":["CC BY-NC"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1794/32744","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"source_record":{"url":"https://scholarsbank.uoregon.edu/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Ascholarsbank.uoregon.edu%3A1794%2F32744","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ralph, Peter"]},{"key":"dc:creator","label":"Author","values":["Patterson, Gilia"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-06-18T14:28:44Z"]},{"key":"dc:date.issued","label":"Date","values":["2026-06-18"]},{"key":"dc:publisher","label":"Institution","values":["University of Oregon"]},{"key":"dc:type","label":"Dc Type","values":["Dissertation or thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Biology"]},{"key":"thesis:degree_level","label":"Degree Level","values":["doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Oregon"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["CC BY-NC"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1794/32744"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Population size is one of the most important pieces of information for monitoring conservation efforts and setting sustainable harvest limits, but it is often difficult and labor intensive to estimate. Close kin mark-recapture (CKMR) is a promising new method for estimating population size from genetic data that identifies related pairs of individuals in a sample and estimates population size from these pairs. Unlike widely used capture-recapture methods, CKMR does not rely on recapturing the same individual multiple times. This means that CKMR can be used when recaptures are difficult or impossible, such as studies involving lethal sampling, sampling at a single time point, or highly elusive species. However, a major shortcoming of current CKMR methods is that they do not model the complex spatial and social dynamics that are important in many species of conservation concern. In my dissertation, I develop a novel spatially explicit CKMR method that models limited dispersal, spatial bias in sampling, complex social structure, and uncertainty in kin estimation. I test the method on simulations and demonstrate that it is accurate and robust to misspecified population trend. I then apply the method to elephants in Kibale National Park in Uganda and explore the feasibility of using CKMR methods to improve monitoring. In Chapter II, I develop a novel spatially explicit CKMR method that uses a convolutional neural network trained with individual-based simulations. When tested on simulated populations, the method is able to accurately estimate population size from parent-offspring and half-sibling pairs, even when dispersal is limited and some areas of the landscape are more intensely sampled than others, a situation in which non-spatial methods are negatively biased. The method is also robust to unknown trend in population size. I apply the method to genetic data from elephant dung samples in Kibale National Park in Uganda and find that when kin pairs are assumed to be known without error, estimated population size agrees with previous estimates but with a much smaller confidence interval. In Chapter III, I extend the CKMR method to account for two common situations in populations of conservation concern: social structure with groups of related individuals traveling together and high uncertainty in kin estimation due to a small number of sequenced loci. I then test the feasibility of using these methods for genetic population monitoring of elephants in Kibale. With 14 sequenced microsatellite loci, kin estimation is highly uncertainty and contributes to uncertainty in CKMR estimates. However, the CKMR network with recaptures still performs better than capture-recapture. With a greater number of sequenced microsatellites (56), kin estimation has very low uncertainty and the network performs very well, even when recaptures are not included. The methods I developed in my dissertation make a valuable contribution to genetic monitoring of species with limited dispersal or complex social structure and demonstrate the power of spatially explicit simulation-based inference for conservation biology. This dissertation includes previously unpublished coauthored material."]},{"key":"dc:title","label":"Title","values":["Simulation-Based Spatially Explicit Close-Kin Mark-Recapture for Genetic Monitoring of Wildlife Populations"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ralph, Peter"],"dc:creator":["Patterson, Gilia"],"dc:date.accessioned":["2026-06-18T14:28:44Z"],"dc:date.issued":["2026-06-18"],"dc:description.abstract":["Population size is one of the most important pieces of information for monitoring conservation efforts and setting sustainable harvest limits, but it is often difficult and labor intensive to estimate. Close kin mark-recapture (CKMR) is a promising new method for estimating population size from genetic data that identifies related pairs of individuals in a sample and estimates population size from these pairs. Unlike widely used capture-recapture methods, CKMR does not rely on recapturing the same individual multiple times. This means that CKMR can be used when recaptures are difficult or impossible, such as studies involving lethal sampling, sampling at a single time point, or highly elusive species. However, a major shortcoming of current CKMR methods is that they do not model the complex spatial and social dynamics that are important in many species of conservation concern. In my dissertation, I develop a novel spatially explicit CKMR method that models limited dispersal, spatial bias in sampling, complex social structure, and uncertainty in kin estimation. I test the method on simulations and demonstrate that it is accurate and robust to misspecified population trend. I then apply the method to elephants in Kibale National Park in Uganda and explore the feasibility of using CKMR methods to improve monitoring. In Chapter II, I develop a novel spatially explicit CKMR method that uses a convolutional neural network trained with individual-based simulations. When tested on simulated populations, the method is able to accurately estimate population size from parent-offspring and half-sibling pairs, even when dispersal is limited and some areas of the landscape are more intensely sampled than others, a situation in which non-spatial methods are negatively biased. The method is also robust to unknown trend in population size. I apply the method to genetic data from elephant dung samples in Kibale National Park in Uganda and find that when kin pairs are assumed to be known without error, estimated population size agrees with previous estimates but with a much smaller confidence interval. In Chapter III, I extend the CKMR method to account for two common situations in populations of conservation concern: social structure with groups of related individuals traveling together and high uncertainty in kin estimation due to a small number of sequenced loci. I then test the feasibility of using these methods for genetic population monitoring of elephants in Kibale. With 14 sequenced microsatellite loci, kin estimation is highly uncertainty and contributes to uncertainty in CKMR estimates. However, the CKMR network with recaptures still performs better than capture-recapture. With a greater number of sequenced microsatellites (56), kin estimation has very low uncertainty and the network performs very well, even when recaptures are not included. The methods I developed in my dissertation make a valuable contribution to genetic monitoring of species with limited dispersal or complex social structure and demonstrate the power of spatially explicit simulation-based inference for conservation biology. This dissertation includes previously unpublished coauthored material."],"dc:identifier.uri":["https://hdl.handle.net/1794/32744"],"dc:language.iso":["en_US"],"dc:publisher":["University of Oregon"],"dc:rights":["CC BY-NC"],"dc:title":["Simulation-Based Spatially Explicit Close-Kin Mark-Recapture for Genetic Monitoring of Wildlife Populations"],"dc:type":["Dissertation or thesis"],"thesis:degree_discipline":["Department of Biology"],"thesis:degree_level":["doctoral"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Oregon"]},"updated_at":"2026-08-21T16:47:20Z"}