Conn, Paul Bayne, authorWhite, Gary C., advisor2024-03-132024-03-132007https://hdl.handle.net/10217/237660State and federal agencies often collect hunter harvest data at check stations. When age- and sex-classes can be determined at the time of harvest, such data provide a wealth of information about population structure. For instance, such summaries are used extensively in quantitative fisheries stock assessment. However, statistically defensible approaches for using age-at-harvest data to monitor terrestrial wildlife populations have not appeared until quite recently, and are deficient in several respects. The primary focus of this dissertation is on developing better methods for analyzing wildlife age-at-harvest data, and on applying these methods to real and hypothetical populations. Chapter one starts by developing statistical methods necessary for fitting population dynamics models to age-at-harvest data. As an example, I analyze marking and harvest records from female black bears (Ursus americanus) in Pennsylvania. In chapter two, I describe numerical implementation issues, as well as results from several extensive rounds of simulation testing. I show that Markov chains will typically need to be quite long to accurately summarize the posterior distribution of model parameters. Nonetheless, estimators are shown to display little bias, to have satisfactory credible interval coverage, and to have a high degree of precision. I show that abundance estimators are quite robust to aging errors, although using data from marked animals twice may lead to overstated measures of precision. In chapter three, I conduct a power analysis to determine if it would be feasible to monitor black bear in Colorado with age-at-harvest and radio telemetry data. My focus in this chapter is on detecting and estimating population trend for varying levels of effort. I show that five year studies are typically too short for all anticipated levels of marking effort, but that ten year studies can yield meaningful estimates of population trend. In chapter four, I address methods that can be used to correct age-at-harvest data for misclassification errors. When the aging criterion is inexact, it is possible to correct for errors if additional information is available on error rates. I illustrate proposed methodology with a black bear dataset from Pennsylvania.born digitaldoctoral dissertationsengCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.age-at-harvestwildlife monitoringecologyforestryTextPer the terms of a contractual agreement, all use of this item is limited to the non-commercial use of Colorado State University and its authorized users.