Prototype radiocollars incorporating global positioning system (GPS) receivers were attached to 8 adult female black bears (Ursus americanus) in March (5) and June-July (3) of 1994 in northern Ontario, Canada. Collars contained a 6-channel GPS receiver, an independent very high frequency (VHF) tracking beacon, an activity sensor, and a computer-memory module. The total package weighed 1.3 kg. The GPS receiver accessed the coarse/acquisition (C/A) code, so accuracy of system horizontal position was expected to be 100 m (i.e., 95% of fixes will be within 100 m of true position). To test the accuracy of location estimates, an additional collar was placed at a reference location during the study period. Units were set to acquire GPS fixes at 3-hour intervals and stored latitude, longitude, time, date, fix quality (horizontal dilution of precision, [HDOP]), fix status (no fix, 2-dimensional [2D], or 3-dimensional [3D] fix), and activity count information in a memory module. Projected life of the unit was 4.5 months with storage capacity for a minimum of 1,000 records. One collar malfunctioned, 1 collar was not recovered, and 2 bears died of natural causes before significant data were acquired, but 4 collars successfully acquired data from 15 June to 13 August 1994. Two collars were recovered from live-trapped bears in September 1994, and 2 were recovered during routine den visits in March 1995. There were no detectable collar-induced injuries, nor any detectable difference in behavior compared to bears with conventional radiocollars. Three of the females successfully raised cubs to the yearling stage by March 1995; the fourth female had a normal litter of cubs-of-the-year in March 1995. The reference collar acquired fixes 99% of the time with 95% confidence intervals for the locations of 73 m on the east-west axis and 103 m on the north-south axis, with no bias away from the reference location. Collars on 3 bears in open habitat (recent burns, young forest stands) acquired fixes 50%, 50%, and 65% of the time, whereas the collar on the bear under more closed canopy acquired fixes only 32% of the time. However, the fine-grained analysis of fixes and satellite-derived landcover did not show a significant difference between the fixes obtained by GPS and conventional radiotelemetry under different landcover types.
- Volume 10
- Issue
- Pages 209-217
- Publication Date 1 January 1998
- DOI
- File Size 528.96 KB
- Download