Dissertation/Thesis Abstract

Spatial ecology of grizzly bears in northwestern Montana and estimating resistance to gene flow
by Graves, Tabitha Ann, Ph.D., Northern Arizona University, 2012, 157; 3509852
Abstract (Summary)

Landscape characteristics influence the abundance, gene flow, and dispersal of wildlife. I examined the influence of landscape characteristics on grizzly bear abundance, identified expectations for isolation by distance patterns of gene flow, and evaluated a method identifying the influence of landscape on gene flow.

I found that local grizzly bear abundance is influenced by both top-down and bottom up processes and by both recent and historical influences. Bottom-up influences (habitat) include the amount of mesic and meadow-shrub vegetation types. Top-down influences (mortality risk) include land ownership and the accompanying relative security. Abundance in this ecosystem is still influenced by the historical distribution of bears.

Isolation by distance, the pattern of genetic distance increasing with geographic distance, is the null model contrasted with the hypothesis that landscape influences gene flow. Although substantial research exists on expected patterns of isolation by distance between populations, inter-individual isolation by distance patterns have not been described. Using simulations, I determined that isolation by distance between individuals is non-linear under restricted dispersal and becomes more linear, but less strong as dispersal increases. Mutation rate and time also influence isolation by distance patterns. Because isolation by distance effects occur over shorter distances with limited dispersal, studies using a linear null model may fail to identify the influence of landscape characteristics. Null models created with realistic dispersal and mutation patterns should be used to correctly account for isolation by distance.

Resistance reflects how difficult it is for genes to move across an area with certain attributes versus an area with attributes that best enhance gene flow. Landscape geneticists use Mantel correlations to compare genetic distances with ecological distances under several alternative sets of resistance parameters to determine which variables and resistance parameters best match the observed ecological distance. I developed a statistical approach for estimating resistance parameters. I also demonstrated that for a wide range of simulated scenarios, Mantel correlations between inter-individual genetic distances and ecological distances lead to biased and imprecise resistance parameter estimates. A model derived from the underlying process of individual movement would be more appropriate and may yield unbiased estimates.

Indexing (document details)
Advisor: Beier, Paul
Commitee: Chambers, Carol L., Dickson, Brett G., Wilder, Jason A.
School: Northern Arizona University
Department: School of Forestry
School Location: United States -- Arizona
Source: DAI-B 73/09(E), Dissertation Abstracts International
Subjects: Wildlife Conservation, Genetics, Conservation
Keywords: Abundance, Gene flow, Grizzly bears, Isolation by distance, Isolation by resistance, Landscape genetics, Least cost path
Publication Number: 3509852
ISBN: 9781267367761