Black bear home ranges in a fragmented Florida landscape at various sampling intervals

Background/Question/Methods

How animals use space is an important consideration for species management and conservation. This is especially true for large carnivores because many have large space requirements yet occupy fragmented habitats in human-dominated landscapes. Radio-tracking animals using a global positioning system (GPS) allows researchers to collect high resolution location data but analysis of space and habitat use has been challenging because this data tends to be highly autocorrelated. The Florida black bear (Ursus americanus floridanus) population is thought to be increasing and expanding into fragmented areas. GPS collars were placed on 17 (6F:11M, ages 1 to 9 years) black bears in north-central Florida and tracked for 5,362 bear-days at intervals of 30 minutes during dawn and dusk and 2 hours otherwise. We estimated the bears’ home ranges (HR) using minimum convex polygon (MCP), kernel density estimator (KDE) and dynamic Brownian bridge movement models (dBBMM; designed to analyze GPS data) applied to data sampled at 2, 4, 8,12 and 24 hour intervals, and asked: Do these methods provide similar HR estimates? Does sampling interval affect HR size; if so, are estimates affected similarly? Are HRs of males and females affected differently when changing sampling intervals, and are these effects consistent across methods?

Results/Conclusions

The average HR size was larger for males than for females and this trend was consistent across all three methods and all sampling intervals. Prior to sampling the data, dBBMM resulted in the smallest HR for both males and females; MCP and KDE provided larger estimates but were comparable to each other. As the sampling interval was increased (with a corresponding decrease in sample size), MCP HR areas for both males and females decreased whereas those obtained from KDE and dBBMM increased. Regardless of estimation method, HR sizes did not differ significantly among females of different ages but sub-adult (2 to 3 years of age) males had a larger HR than adult males (≥4 years). These results are consistent with published accounts that young males cover large areas as they disperse from their natal range whereas females typically settle near their mothers. For most bears, home ranges obtained from dBBMM consisted of many small, disconnected polygons, which reflects the potentially localized use of habitat patches on our study site. MCP and KDE were not as robust at identifying patchy use of fragmented habitat in our study site.