Factors contributing to process variance in annual survival of female greater sage-grouse in Montana

Populations of Greater Sage-Grouse (Centrocercus urophasianus) have declined by 69-99% from historic levels, and information on population dynamics of these birds at a landscape scale is essential to informed management. We examined the relationships between hen survival and a suite of landscape-scale habitat and environmental conditions. We radio-marked 237 female Sage-Grouse and measured 426 vegetation plots during 2001-2004 at four sites in a 3200-km² landscape in north-central Montana, USA. We used program MARK to model monthly survival rates for 11 seasonal intervals. There was strong support for the best-approximating model (${\rm{AIC}}_{\rm{c}} $weight = 0.810), which indicated that (1) hen survival varied by season within years and by year within seasons, (2) nesting hens had higher nesting-season survival than non-nesting hens, and (3) individuals at one site had lower hunting-season survival than at other sites. We observed considerable variation in hen survival. Process variation was 0.255, with an expected range of annual survival of 0.12 to 1.0. The ratio of process to total variation was 0.999, indicating that observed variation was real and not attributable to sampling variation. We observed a nearly fourfold difference in maximum and minimum annual survival, ranging from 0.962 ± 0.024 (mean ± se) for nesting hens in 2001- 2002 to 0.247 ± 0.050) for non-nesters in 2003-2004. Low annual survival in 2003 resulted from the compounded effects of a West Nile virus outbreak in August and a severe winter in 2003-2004. Increased hen mortality associated with severe winter weather contrasts with prior beliefs that Sage-Grouse populations are typically unaffected by winter weather conditions and underscores the importance of protecting winter sagebrush (Artemisia spp.) habitats.