Professor University of Montana Missoula, Montana, United States
Abstract: The management of gray wolves (Canis lupus) across North America requires accurate population estimation techniques. Previous research has also highlighted the need for measuring wolf density and distribution due to their ecological effects, notably on ungulate prey, including deer, elk, and endangered caribou. Gray wolves pose a challenge for researchers due to their cryptic and group living nature, while also often being naturally individually unidentifiable. Recently developed generalized Spatial-Mark-Recapture (gSMR) models, however, provide new statistical techniques for estimating abundance of group living species, when some individuals of a population are individually identifiable via tagging, radiocollaring, etc. These gSMR models have yet to be adapted to gray wolves but show promise in their application to other cryptic and social wildlife. Wolves have been radiocollared in Banff National Park (BNP; n = 24 wolves; average years monitored = 1.95 years; range = 1-3 years) from 3 - 6 wolf packs since 2011. Additionally, Parks Canada has deployed and maintained one of the largest remote wildlife camera networks in North America since 2011, with over 660 unique camera locations (average of 288 camera locations/year; range = 99 – 390). From 2011 – 2022, there have been over 60,000 wolf detection events with over 2,000 marked individual (i.e., radiocollared) detections. We used these data to develop a two-stage gSMR model in which we leverage GPS telemetry data and camera detections over the last 11 years (2011-present) to estimate detection parameters (i.e., detection probability, and sigma [an individual space use metric]) of radiocollared wolves. Then, we used these detection parameters of marked individuals to develop a second SMR model that treats all detected individuals as unmarked, providing an estimate of density and spatial distribution across BNP. Preliminary results suggest both promise, and challenges in the application of gSMR to estimate wolf density and distribution. Despite the challenges, we expect gSMR models to improve the way in which researchers monitor wolf populations because of the growing adoption of remote wildlife camera networks in combination with traditional radiocollaring. These gSMR methods also allow ecologists to provide an estimate of distribution, in addition to wolf density, something that has proved challenging with previous remote camera studies.
