97th ESA Annual Meeting (August 5 -- 10, 2012)

PS 75-105 - Modeling climate change effects on the ecotone between forest and grassland at Wind Cave National Park, South Dakota

Thursday, August 9, 2012
Exhibit Hall, Oregon Convention Center
David A. King, Biological and Ecological Engineering, Oregon State University, Corvallis, OR, Dominique M. Bachelet, Conservation Biology Institute and Oregon State University, Corvallis, OR and Amy Symstad, USGS Northern Prairie Wildlife Research Center, Hot Springs, SD
Background/Question/Methods

Climate change will affect reserves set aside to protect natural resources. Like other land management agencies, the National Park Service requires information on the possible long-term impacts of climate change and how management practices may counter or increase these impacts.  To address these concerns, we applied a dynamic global vegetation model (MC1) to Wind Cave National Park, which lies on the ecotone between ponderosa pine forests and prairies in the southern Black Hills of South Dakota.  MC1 simulates vegetation distribution, carbon, water and nutrient cycling, as well as wildfire in a highly interactive manner, using climate and soils drivers.  We parameterized the model to approximate recent stream flows and the historical balance between forests and grasslands.  We then ran 100-year-long simulations into the future using statistically downscaled climate projections from the CSIRO Mk3, Hadley CM3 and MIROC 3.2.Medres general circulation models (GCMs), run with the A2 emission scenario.  These inputs were chosen as representative of the range of GCM projections.

Results/Conclusions

Projected temperatures increased substantially for all three projections, but more so for Hadley and especially MIROC than for CSIRO.  Annual precipitation increased for CSIRO, changed little for Hadley and decreased for MIROC.  Fire frequency increased for all three future projections due to increased drying of fire fuels during the summer season. More frequent fires reduced the woody biomass of the Park; more so for the CSIRO projection. In this latter case, grasslands partially replaced forests despite the fact that the CSIRO future climate caused the smallest increase in fire frequency. This counterintuitive result was associated with an increase in fine fuel loads (due in part to an increase in deciduous trees) that resulted in hotter projected fires, and therefore greater tree mortality.  Increasing the level of grazing reduced grass-derived litter, but had little effect on forests, where tree-derived litter dominates fine fuel loads.  Simulations with complete fire suppression resulted in a gradual increase in tree biomass across the grasslands that may be expected in areas within ~100 m of seed trees, where regeneration is currently observed at Wind Cave.