Accuracy of radio-tracking to estimate activity and distances walked by European bison in the Bialowieza Forest, Poland

To determine the accuracy of radio-tracking estimates of activity and distances walked by European bison Bison bonasus (Linnaeus, 1758), we observed bison while simultaneously conducting radio-tracking in the Białowieża Forest (Poland). Single radio-tracking estimates often did not represent the actual activity of the bison, but the mean time spent active calculated from radio-tracking and direct observations did not differ significantly. Radio-tracking provided overestimates of the distances walked by bison when the animals walked between 0 and 25 m and underestimated the distances walked by bison if they walked further than 25 m in a radio-tracking interval. Activity was best estimated when radio-trackers were between 200 and 300 m from the bison. The mean activity calculated with a sampling interval of 15 min needed about 8 days of radio-tracking and the mean of distances walked about 35 tracking days to stabilise. We concluded that collars without activity sensors can measure the mean activity of bison but cannot represent the length of their activity bouts or the type of their activity and that our radio-tracking estimates of distances walked by bison needed to be corrected before further analyses.