Michigan State University
Data integration in population and community ecology using hierarchical modeling
Abstract
dc:descriptionIn this dissertation, I develop and apply methods for data integration using hierarchical modeling to estimate the status, trends, and demography of wildlife populations and communities. I use multi-level statistical and mathematical models to explicitly link observed data to latent ecological processes. By separately modeling observational and ecological processes, I can integrate multiple disparate data sources into a unified framework to estimate ecologically relevant population and community parameters, often in the context of wildlife conservation. In Chapter One, I apply a multispecies hierarchical distance sampling model to assess the effect of management actions on a carnivore community in the Masai Mara National Reserve, Kenya. I assess variation in species-level responses to passive management, resulting in human disturbance and apex predator declines. In Chapter Two, I develop an integrated distribution model that uses distance sampling and presence-only data to jointly estimate species abundance. I apply this model to a case study on black-backed jackals (Canis mesomelas) to evaluate the effects of anthropogenic disturbance on the distribution of jackals across the Masai Mara National Reserve. In Chapter Three, I evaluate status and trends of species in a forest dwelling duiker community using detection-nondetection data. I develop a multispecies dynamic N-occupancy model to estimate species-level abundance, demographic parameters, and quasi-extinction probabilities. In Chapter Four, I create a spatiotemporal integrated model to estimate the effects of weather conditions on monarch butterflies (Danaus plexippus) during spring migration. Each chapter illustrates a unique application of data integration in wildlife ecology, either by combining data on multiple species to estimate population and community-level parameters or by combining disparate data sources on a single species to estimate demography and other population-level parameters. Data integration is a powerful framework that leverages all available information to address pressing conservation challenges.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Farr, Matthew T.
- Contributors dc:contributor
-
- Zipkin, Elise F.
- Finley, Andrew O.
- Holekamp, Kay E.
- Roloff, Gary J.
Subjects
dc:subject × 6Rights
dc:rights- Statement dc:rights
-
- Attribution 4.0 International
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
-
etd:50127
isbn:9798471109773
oclc:on1294308312
oclc:1294308312
umi:28771712
local:Farr_grad.msu_0128D_18707