My study focused on determining home ranges, movements, activities, habitat use, and habitat linkages of black bears during the pre-construction phase of the research. Field personnel collected information on bear locations and movements from June 2000-June 2001 on the 12,240-ha treatment area (the area of highway construction), and a 12,266-ha control area (no highway construction). These data will serve as the baseline for comparisons with data collected after the anticipated completion of the new highway in 2005.

Project personnel collected 1,909 daily telemetry locations and 2,569 hourly telemetry locations on 35 radiocollared bears (9 M: 26 F). Based on the 95% fixed-kernel home range method, home ranges on the treatment area averaged 5.3 km2 for females (n = 9) and 70.7 km2 (n = 4) for males. The average annual home range for females on the control area was 3.1 km2 (n = 14), and 14.2 km2 for males (n = 3). Female home ranges on the treatment area were larger than on the control area (Z = 2.87, P = 0.003), with extensive home-range overlap on both areas.

Although daily activity patterns did not differ for spring or summer, I observed a disproportionate number of active observations during fall (X2 = 17.4, df = 3, P < 0.001), particularly in the evening. Land cover was not associated with daily activity patterns. Hourly movement rates differed between the study areas, with greatest rates on the control area. Furthermore, hourly movements of females on the control area were greater than those of males.

I used the weights-of-evidence technique for analysis of habitat selection; this technique is a discrete, multivariate method for combining spatial data to predict habitat use. The final model was based on 1,811 daily telemetry locations and included the variables forest cohesion, forest diversity, and forest-agriculture edge density (all measured at a 0.20-sq km scale). The overall conditional independence ratio of the model was 0.97, indicating that one of the primary model assumptions was met. The model performed well; the highest predicted probability category included only 23.8% of the study area but contained 56% of the bear locations reserved for model testing. Contrast values indicated that forest cohesion and forest-agriculture edge density were the most influential variables to predict black bear habitat use.

I used predicted probabilities of bear occurrence from the weights-of-evidence model in a least-cost-path analysis to delineate habitat linkages for the 2 study areas and for the Albemarle-Pamlico Peninsula region. Habitat linkages on the treatment area and between the 2 study areas coincided with the underpass locations of the new highway. Regionally, habitat linkages often converged near underpass sites as well, suggesting that these underpasses may have both regional and local importance.

Results of my study provide baseline data on the spatial ecology of black bears prior to construction of the new highway. Changes in home ranges, movements, and habitat use based on comparisons with post-construction data will be useful to assess whether the new section of highway affects the ecology of the resident black bear population.