Grizzly bear (Ursus arctos) movements and habitat use predict human-caused mortality across temporal scales

While the location of wildlife mortalities provides some insight on the cause of death, identifying the risk factors associated with mortality events and in which cases these factors result in death requires information on individual behaviour prior to death. With access to a long-term database of grizzly bear (Ursus arctos Linnaeus, 1758) GPS location data, we investigated how behaviour differed between individuals that died of anthropogenic causes and those that survived across different temporal scales. We analyzed movement (diurnality and daily displacement) and habitat use (modelled risk and habitat quality) of grizzly bears residing in Alberta, Canada from 2005-2021 to determine if grizzly bears that died and grizzly bears that survived differed in these behaviours 2-4 years, 1 year, and 1 week prior to death, and whether patterns changed over time. We found that diurnality increased in the last year of life, while displacement increased in the last week of life, with differences becoming greater nearer the day of death. Grizzly bears that died used high risk and low-quality habitat at all time scales, and these behaviours increased as death approached. Our analysis suggests that grizzly bear mortalities do not occur randomly but happen at times when individuals exhibit high risk behaviours. This information can be used to make management decisions related to habitat management, road use, and human access.