Space use by animals can provide insight into the habitats in which they live. In predictable landscapes where high-quality habitat is distributed heterogeneously, fidelity can increase individual fitness because animals benefit from familiarity. We tested the hypothesis that space use and the level of site fidelity will reflect habitat productivity and the spatiotemporal distribution of resources. To test our hypothesis we examined site fidelity in 41 grizzly bears, Ursus arctos, during 2003–2006 in the Mackenzie Delta, in Canada's Arctic. We estimated annual home range size for males ( x ¯ = 1215 km 2 ) and females ( x ¯ = 680 km 2 ) using the 95% fixed kernel approach. We used three methods to examine the level of home range fidelity in consecutive years. Bear locations were largely spatially nonoverlapping between years, which represented a pattern of both low site and home range fidelity. Although distances between home range centres were small for both males ( x ¯ = 7.7 km ) and females ( x ¯ = 8.4 km ), mean percentage overlap was only 24.2%. Our observations suggest that bears remain philopatric to a region, however, position and configuration of the annual range drifts over time. The low fidelity that we observed suggests an adaptive strategy of space use for low-density populations in regions of low productivity where quality habitats are spatially and temporally heterogeneous. We suggest that traditional definitions of a home range may be too limited for Arctic grizzlies and propose that the multiannual home range is a more meaningful measure for describing space use.