Spatiotemporal Modeling for Wildlife Demographic Analysis: Bridging Analysis to Waterfowl Conservation
Loading...
Authors
Lohman, Madeleine
Issue Date
2025
Type
Dissertation
Language
en_US
Keywords
Agriculture , Bayesian , Population ecology , Waterfowl
Alternative Title
Abstract
Examining variation in ecological systems is critical to understanding the fundamental demographic processes (e.g. reproduction, survival, growth, and dispersal) that govern populations and manage them in an increasingly altered world. Population dynamics often respond to environmental alterations, such as habitat fragmentation, climate variability, and resource changes, as individuals cope with shifting stressors. I demonstrate the use of spatially explicit models to estimate demographic rates and their relationships to environmental conditions in a highly variable system. In doing so, I clarify the changing spatial nature of this system and highlight conservation issues for midcontinent mallard populations. My second chapter serves as a guide to using spatially explicit models in population ecology. In a comprehensive literature review, I discuss spatial autocorrelation in demographic models and show how spatial models can mitigate this problem while leading to a greater ecological understanding of different systems. With three examples, I estimate spatial variation in survival and show how spatial models contribute to population ecology by reducing autocorrelation in residuals, smoothing and interpolating data, and providing insight into the spatial dynamics of ecological processes. In my third chapter, I use a conditional autoregressive model to estimate survival and harvest mortality from 1974 to 2023 of female and male mallards at the adult and juvenile age stages. Specifically, I studied populations in the Prairie Pothole Region (PPR) of the northern Great Plains. This area has seen dramatic and widespread landscape changes in recent decades with the change of agricultural patterns and climate. These spatial and temporal gradients of land use, combined with natural heterogeneity on the landscape, allow us to see how changing environmental conditions affect broad-scale population dynamics. Results showed substantial variation across time and space for survival and harvest mortality for all age and sex classes. Juvenile survival had a positive relationship with environmental conditions associated with higher recruitment, while adult survival had a negative relationship with the same variables. In the same vein, adult survival increased with habitats associated with lower recruitment rates, while juvenile survival decreased. These results also suggested a worrying trend: both adult and juvenile females exhibited declining survival probabilities throughout the time series. My fourth chapter was driven by these findings. Changes in sex-specific survival rates can lead to changes in population sex ratios and lead to substantial shifts in population structure. These changes in survival rates can result from sex-specific environmental effects on males and females, often via reproductive investments and risk. I estimated the relationships between sex-specific survival differences and environmental factors in mallard breeding grounds in the PPR. I found a rapid increase in survival differences, with males surviving at increasingly higher rates than females at the adult and juvenile age stages. Environmental covariates associated with higher quality breeding habitats were correlated with increases in sex-specific survival differences. These trends, resulting from land use and climate change in the region, could result in declining midcontinent mallard populations. Together, these results indicate a population that is responding to an extremely variable environment across time and space. The use of spatially explicit models in this research shows the spatial nature of mallard population dynamics, as well as where and how conservationists can target resources towards this population.