Date of Award

Winter 2008

Project Type


Program or Major

Natural Resources and the Environment: Wildlife Ecology

Degree Name

Master of Science

First Advisor

Peter Pekins


Abundance estimates for black bears (Ursus americanus) are an important tool for effective management. Recent advancements in DNA technology have enabled genetic tagging mark-recapture population estimates using DNA from hair samples. I conducted a population estimate using genetic tagging in 2 study sites presumed to have different bear densities in northern New Hampshire (Pittsburg and Milan). To test repeatability, I conducted the genetic tagging estimates in 2 consecutive years. I also compared these estimates to those derived from traditional methods used by the New Hampshire Fish and Game Department (NHFG) using hunter harvest and mortality data. I found that the density estimates produced from the genetic tagging methods were consistent in the 2 years, and were similar to those derived from traditional methods. In 2006, the estimated number of bears in Pittsburg (315 km2) was 70, corresponding to a density of 0.16-0.28 (95% CI) bears/km2 . In 2007, the Pittsburg (400 km2) estimate was similar: 78 bears with a density of 0.15-0.24 bears/km2. In Milan (440 km2) during 2006, the estimated number of bears was 106 corresponding to a density of 0.13-0.35 bears/km2. The 2007 Milan estimate (371 km2) was similar with 99 bears and a density of 0.19-0.34 bears/km2. While the traditional methods may be appropriate and more cost effective for density estimation at a regional scale, I found that the genetic tagging methods were able to detect demographic variation at a local scale. In addition to generating population estimates, I used the genetic information to identify population and spatial genetic structure and to determine if landscape features such as roads and rivers caused resistance to gene flow. I tested for population distinction using the program STRUCTURE, FST values, and a mean relatedness function. I used a Mantel test of isolation by distance and spatial autocorrelation for the spatial analyses. To assess landscape resistance, I used an analysis of mean relatedness between subpopulations divided by landscape features. Through consensus, I found that the 2 study sites were genetically distinct (F ST = 0.024, P = 0.05). I also found a positive relationship between genetic and geographic distance (R = 0.13, P>0.0001), and that females showed spatial autocorrelation through 5 km. Regarding landscape resistance to gene flow, I found that the presence of Route 3 in Pittsburg did not cause genetic differentiation between subpopulations on either side of the road, while the Route 16-Androscoggin River corridor in Milan influenced the genetic population structure of females.