We studied aspen (Populus tremuloides) and elk (Cervus elaphus) response to a 2008 prescribed fire (1200 ha) in Waterton Lakes National Park, AB. The fire was set in an aspen parkland in elk winter range, to restore aspen and rough fescue (Festuca spp.). This site contains 1-2 wolf (Canis lupus) packs. Aspen provide a key elk food, and elk are the dominant herbivore by biomass. Previous research in non-wolf systems found a positive relationship between aspen sprouting, fire severity, and elk browse, with little to no aspen recruitment. We hypothesized that in an elk-wolf-aspen system, elk would avoid areas of highest fire severity, due to the high predation risk created by thick vegetation and coarse woody debris at such sites. We measured aspen and elk response to the fire annually for 6 years (2008-2013). We tested our hypothesis using an information theoretic approach, generating predictive models to examine the influence on the proportion of aspen browsed of fire severity (snag basal area), site index (site productivity), shrub cover (proportion), aspen sprout density (per ha), and distance to edge of the aspen parkland (m). We used GPS-collar data to confirm elk activity and camera traps to confirm wolf activity.
Results/Conclusions
We found that elk browse declined sharply as distance from the edge of the parkland increased (p=0.0008). Fire severity (p=<.0001), aspen sprout density (p=<.0001), and site index appeared in the best-fitting models < 4 Δ AIC values, with increases in these variables linked to a decrease in elk browse. Elk avoided areas with the highest aspen sprout density, productivity, or fire severity. Elk avoidance of sites that had the greatest food resources may be associated with avoidance of wolf predation. We recommended a re-burn of this site to control aspen expansion onto the grassland and to maintain aspen vigor. We further recommended that the re-burn be implemented using a multifactorial experimental design. Park natural resources managers implemented a full (1200 ha) re-burn in 2015, in which they manipulated fire severity within an experimental framework and plan a second re-burn in 2017 as part of long-term research on the effects of prescribed fire on aspen and grassland restoration in this system.