Interchange and Overlap Among Four Adjacent Arctic Caribou Herds

Barren ground caribou (Rangifer tarandus granti) are distributed in herds that seasonally use specific geographic regions within an annual range, with varying levels of fidelity during different periods (e.g., calving, insect relief, wintering). As a result, caribou management is generally tailored to individual herds that often range across administrative boundaries. Herd ranges can shift over time, seasonal ranges of adjacent herds often overlap, herds merge, and there is often little genetic differentiation among adjacent herds. If substantial herd interchange occurs, it would have important management implications by influencing estimates of herd size, herd composition, and harvest rates. We compiled satellite telemetry data from 2003–2015 for 4 large arctic caribou herds to quantify herd interchange rates. We calculated a metric of herd interchange based on the relationship of caribou locations to typical weekly herd ranges (all yrs combined) and the distance to other radio-collared caribou from each of the 4 herds (yr specific). Although herd membership cannot always be clearly defined based on location, this metric provides an objective measure of the strength of evidence of herd membership that can be used to make comparisons among herds and time periods. We also calculated herd overlap and quantified how it varied throughout the year. Herd interchange was rare in the 2 larger herds, generally occurring when caribou overwintered with an adjacent herd, whereas herd interchange from the 2 smaller herds was more frequent and could last longer than a year. Although sample sizes were limited, there were no clear patterns in herd interchange with year or annual herd size. The 2 smaller herds had large seasonal overlap with adjacent herds, suggesting that herd interchange may be related to spatiotemporal herd overlap and relative herd size. Our results can help managers understand herd interchange and overlap to make management decisions, interpret research results, and develop more accurate population models.