Leslie A. Hansen

ANIMAL-RELATED VEHICLE ACCIDENTS AT THE LOS ALAMOS NATIONAL LABORATORY, NEW MEXICO

Abstract Large animals occurring on Los Alamos National Laboratory (LANL) and adjoining property pose a number of concerns to area residents and government agencies. Some of these concerns are animal-related accidents that can result in human injuries and fatalities, property damage, and a loss of an economically viable resource (game). We analyzed animal-vehicle accident data with respect to time, season, location, and species for accidents occurring on LANL property and analyzed site characteristics of accident hotspots. We observed a significantly greater number of vehicleelk (Cervus elaphus) accidents during winter compared to summer and spring and a greater number of elk and mule deer (Odocoileus hemionus) vehicle accidents during late afternoon and evening hours compared to morning and afternoon hours. We estimated a cost of

Population and genetic outcomes 20 years after reintroducing bobcats (Lynx rufus) to Cumberland Island, Georgia USA.

136,500 per year associated with animal–vehicle accidents occurring on LANL property (excluding medical costs). Slope and vegetation height were the best predictors of the status of an area as a hotspot or a control site. These data will be used in public education efforts and to develop mitigation measures to reduce the potential for accidents.

Evolutionary Computation and Post-wildfire Land-cover Mapping with Multispectral Imagery

Abstract In 1988–1989, 32 bobcats Lynx rufus were reintroduced to Cumberland Island (CUIS), Georgia, USA, from which they had previously been extirpated. They were monitored intensively for 3 years immediately post‐reintroduction, but no estimation of the size or genetic diversity of the population had been conducted in over 20 years since reintroduction. We returned to CUIS in 2012 to estimate abundance and effective population size of the present‐day population, as well as to quantify genetic diversity and inbreeding. We amplified 12 nuclear microsatellite loci from DNA isolated from scats to establish genetic profiles to identify individuals. We used spatially explicit capture–recapture population estimation to estimate abundance. From nine unique genetic profiles, we estimate a population size of 14.4 (SE = 3.052) bobcats, with an effective population size (N e) of 5–8 breeding individuals. This is consistent with predictions of a population viability analysis conducted at the time of reintroduction, which estimated the population would average 12–13 bobcats after 10 years. We identified several pairs of related bobcats (parent‐offspring and full siblings), but ~75% of the pairwise comparisons were typical of unrelated individuals, and only one individual appeared inbred. Despite the small population size and other indications that it has likely experienced a genetic bottleneck, levels of genetic diversity in the CUIS bobcat population remain high compared to other mammalian carnivores. The reintroduction of bobcats to CUIS provides an opportunity to study changes in genetic diversity in an insular population without risk to this common species. Opportunities for natural immigration to the island are limited; therefore, continued monitoring and supplemental bobcat reintroductions could be used to evaluate the effect of different management strategies to maintain genetic diversity and population viability. The successful reintroduction and maintenance of a bobcat population on CUIS illustrates the suitability of translocation as a management tool for re‐establishing felid populations.

Microsatellite markers for the Mexican spotted owl (Strix occidentalis lucida)

The Cerro Grande/Los Alamos wildfire devastated approximately 43,000 acres (17,500 ha) of forested land, and destroyed over 200 structures in the town of Los Alamos. The need to monitor the continuing impact of the fire on the local environment has led to the application of a number of advanced remote sensing technologies. During and after the fire, remote-sensing data was acquired from a variety of aircraft- and satellite-based sensors, including Landsat 7 Enhanced Thematic Mapper (ETM+). We now report on the application of a machine learning technique to the automated classification of land cover using multispectral imagery. We apply a hybrid genetic programming/supervised classification technique to evolve automatic feature extraction algorithms. We use a software package we have developed at Los Alamos National Laboratory, called GENIE, to carry out this evolution. We use multispectral imagery from the Landsat 7 ETM+ instrument from before and after the wildfire. Using an existing land cover classification based on a Landsat 5 TM scene for our training data, we evolve algorithms that distinguish a range of land cover categories, along with clouds and cloud shadows. The details of our evolved classification are compared to the manually produced land-cover classification.