Masters Theses

Date of Award

8-2003

Degree Type

Thesis

Degree Name

Master of Science

Major

Wildlife and Fisheries Science

Major Professor

Frank T. van Manen

Committee Members

Joseph D. Clark, Lisa I. Muller, Shih-Lung Shaw

Abstract

The Arkansas Game and Fish Commission (AGFC) is considering translocation to expand its established elk population. I conducted a feasibility study that identified potential restoration sites in Arkansas to reduce the probability of reintroduction failure. I developed 2 landscape-scale predictive models using geographical information system (GIS) technology to identify potential elk restoration sites in Arkansas, one to identify suitable elk habitat and the other to assess the potential for elk-human conflict. I assessed winter habitat for elk using empirical data consisting of 239 elk-group locations collected from helicopter surveys in the Buffalo National River area. Those surveys were conducted by the AGFC in February-March, 1992-2002. A suite of 9 habitat variables were developed to characterize the habitat and landscape conditions associated with those elk-group locations. Variables were generated at multiple spatial scales, representing different orders of habitat selection, so that I could select the most appropriate scale to evaluate each variable. From those data, I then applied the Mahalanobis distance statistic to evaluate winter habitat suitability in Arkansas based on 90- x 90-m pixels. Lower Mahalanobis distance values indicated a greater similarity to the habitat conditions associated with the elk-group locations. More suitable elk habitat was associated with areas of high landscape heterogeneity, heavy forest cover, and gentle sloping ridge tops and valleys in western and northwestern Arkansas, where human population and road densities also were relatively low. Areas of intensive agriculture in the Mississippi River Delta generally were least suitable. I tested model performance by recording the frequency of occurrence of elk scat within 19 fixed-width transects surveyed in March 2002. Linear regression analysis indicated that the frequency of scat occurrence increased with decreasing mean Mahalanobis distance values (F = 9.65, P = 0.039). Those results suggest that elk presence was more likely in areas predicted by the model to be more suitable habitat. Finally, I assessed the potential for elk-human conflict in Arkansas with a GIS adaptation of the Analytical Hierarchy Process. Five elk experts in Arkansas ranked the relative importance of 8 criteria that could influence the potential for elk-human conflict in a series of pairwise comparisons. Those rankings were then applied in a weighted linear summation of 8 variables representing those criteria, resulting in a single map delineating the relative potential for elk-human conflict. Public land forage availability was determined to have the strongest influence on the potential for elk-human conflict, contributing 33% to the overall conflict potential, followed by human population growth rate (22%) and the amount of private land in row crops (18%) . Elk-human conflict potential in Arkansas ranged from 0.14 to 0.72 (x = 0.54 ± 0.009). Conflict potential was classified as low (:S 0.49), medium (0.49- 0.59), and high (> 0.59), representing intervals of 6 standard deviations from the mean conflict potential value for radio-locations of nuisance elk cows. I combined contours of those conflict potential intervals with the winter habitat suitability model to identify regions where suitable elk habitat corresponded with low potential for elk-human conflict. Those regions mainly were associated with public lands in western and northwestern Arkansas. Large, contiguous patches of suitable habitat within areas of low elk-human conflict potential tended to correspond with public and private land boundaries in northern and northwestern Arkansas. The combined map provides a tool for natural resource managers to identify and rank potential elk restoration areas in Arkansas.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS