Home WakeSpace Scholarship › Electronic Theses and Dissertations

Using Bayesian Hierarchical Models to Study the Spatial and Temporal Distribution of Mammals in Serengeti National Park

Electronic Theses and Dissertations

Item Files

Item Details

abstract
In this thesis, I work with camera trap data from the Snapshot Serengeti project in Serengeti National Park in Tanzania Africa. This data consists of the time and location where species were observed during the year 2012. Occupancy models are used to analyze this type of data, but they require discretization of time. Thus, we have to choose the width of time intervals to be analyzed. The goal of this research is to develop methodology to compare results based on different time interval widths. We have created occupancy models to predict the presence of species based on environmental variables while accounting for spatial and temporal dependence. We develop a join statistic to measure the spatial dependence of data. Then we create a metric that compares the spatial clustering that exists when the data is compared at different time intervals. This shows how much the spatial dependence changes based on the time window that is chosen. We begin by studying the change in spatial clustering for a single species depending on the time interval, then we extend our work to study the dependence between two species.
subject
contributor
Harris, Richard Trafford (author)
Hepler, Staci (committee chair)
Norris, James (committee member)
Erhardt, Robert (committee member)
Anderson, Todd M (committee member)
date
2018-05-24T08:35:43Z (accessioned)
2018-05-24T08:35:43Z (available)
2018 (issued)
degree
Mathematics and Statistics (discipline)
identifier
http://hdl.handle.net/10339/90679 (uri)
language
en (iso)
publisher
Wake Forest University
title
Using Bayesian Hierarchical Models to Study the Spatial and Temporal Distribution of Mammals in Serengeti National Park
type
Thesis

Usage Statistics