|Abstract or Summary
- Understanding the effects of habitat disturbance on a species' habitat selection patterns, and demographic rates, is essential to projecting the trajectories of populations affected by disturbance, as well as for determining the appropriate conservation actions needed to maintain those populations. Greater sage-grouse (Centrocercus urophasianus) is a species of conservation concern in western North America. The distribution of the species has been reduced by approximately half since European settlement, with concurrent and continuing population declines across its occupied range. The primary threats to the species are habitat alteration and loss, caused by multiple factors. In the western portion of its distribution, increasing wildfire activity is a primary cause of habitat loss and degradation. Single wildfires in this area may now reach extremely large sizes (>100,000 ha), and wildfires have been linked to local population declines. However, no published studies, to date, have examined the immediate effects of large-scale wildfire on sage-grouse habitat selection and demographic rates, using modern telemetry methods.
I studied the habitat selection patterns, nest success, and survival of adult, and yearling female sage-grouse, captured within or near the Holloway fire, using state-of-the-art GPS-PTT telemetry methods. The Holloway fire burned ~187,000 ha of highly productive sage-grouse habitat in August, 2012. My study began during the first spring post-fire (March, 2013), and continued through February, 2015. I monitored seasonal habitat use patterns, and site-fidelity of sage-grouse, and modeled third-order seasonal resource selection, using mixed effects resource selection
functions, in relation to characteristics of the post-fire habitat mosaic, terrain, mesic habitat availability, and herbaceous vegetation regeneration. I described sage-grouse nesting habitat use, nesting effort, and modeled daily nest survival in relation to temporal patterns, patch scale vegetation, biological factors, and landscape-scale habitat composition. I modeled adult and yearling female sage-grouse survival in relation to temporal patterns, biological factors, and landscape-scale habitat composition.
Female sage-grouse primarily exhibited a three range seasonal movement pattern, with differentiation between breeding-nesting-early brood-rearing habitat (mean use dates: 8 Mar - 12 Jun), late brood-rearing-summer habitat (13 Jun - 20 Oct), and winter habitat (21 Oct - 7 Mar). However there was variation in seasonal range behavior among individuals. Sage-grouse exhibited considerable fidelity to all seasonal ranges, for individuals which survived >1 yr, mean distance between seasonal range centroids of the same type were 1.80 km, 1.65 km, and 3.96 km, for breeding ranges, summer ranges, and winter ranges, respectively. Within seasonal ranges, sage-grouse exhibited third-order resource selection patterns similar to those observed for populations in undisturbed habitats. Sage-grouse, at the population level, selected for level terrain throughout the year. During the breeding season sage-grouse selected for areas with increased amounts of intact sagebrush land-cover within a 1-km² area around used locations, areas of increased NDVI values within a 6.25-km² area, an amount of mesic habitat within a 6.25-km² area roughly equal to that available on the landscape, and mid-level elevations. During summer, sage-grouse, at the population level, selected for an areas with an intermediate density of burned-intact habitat edge within a 1 km² area, areas of increased NDVI values within a 6.25-km² area, intermediate distances to mesic habitat, and high elevations. During winter, sage-grouse, at the population level, selected for increased amounts of intact sagebrush land-cover within a 0.089-km² area, areas with decreased variation in NDVI within a 0.089-km² area, an amount of mesic habitat within a 6.25-km² area roughly equal to that available on the landscape, and intermediate elevations. There was considerable variation in third-order resource selection patterns among individuals during all seasons.
Sage-grouse nest success was consistently low during the study (2013: 19.3%, 2014: 30.1%), and nest initiation rates were average to high (2013: 1st nest initiation = 90.5%, 2nd nest initiation = 23.1%; 2014: 1st nest initiation = 100%, 2nd nest initiation = 57.1%). Daily nest survival rates were influenced by an interaction between year and nesting attempt, and by forb cover within 5 m of the nest. Nest survival over the incubation period was consistently low for 1st and 2nd nests during 2013, and for 1st nests during 2014 (range: 0.131 - 0.212), but increased to 0.744 for 2nd nests during 2014. Forb cover within 5 m of the nest had a positive effect on daily nest survival rates, with a 1% increase in forb cover increasing the probability of a nest surviving a given day by 1.02 times.
We did not detect strong direct effects of habitat or biological characteristics on survival of adult and yearling female sage-grouse. Rather, survival varied by month with lowest survival occurring in April and August of each year, and highest survival occurring during the winter. While patterns of monthly survival were similar between years, there was a strong, negative additive effect on survival which extended from the beginning of the study (March, 2013), through the end of the first post fire growing season (July, 2013). Although monthly survival increased following the end of the 1st post-fire growing season, yearly survival over both the 1st and 2nd biological years post-fire was low (March 2013 - February 2014: 24.0%; March 2014 - February 2015: 37.9%).
These results indicate that female greater-sage grouse do not respond to wildfire related habitat disturbance through emigration, and rather continue to attempt to exist and reproduce in habitats disturbed by wildfire during the immediate years following a fire. While, due to site-fidelity, sage-grouse are not able to leave wildfire affected seasonal ranges, within those seasonal ranges they still attempt to utilize habitat components which most closely match their life-history requirements. However, this behavior appears to have an acute fitness cost to individuals, with reduced nesting success and survival of individuals utilizing fire-affected habitats during the first two years post-fire. This reduction in demographic rates likely explains observed sage-grouse population declines following wildfire, and indicates that these population declines are not the result of sage-grouse emigration away from fire-affected leks, but rather a true decline in the number of individual sage-grouse on the landscape following large-scale wildfire.