96th ESA Annual Meeting (August 7 -- 12, 2011)

OOS 23-3 - Fire Regime Dynamics in BLM rapid ecological assessments

Wednesday, August 10, 2011: 8:40 AM
12A, Austin Convention Center
Bob Unnasch, Sound Science and Jon Hak, Conservation Services, NatureServe, Boulder, CO
Background/Question/Methods

In the arid west, wildfire is the key natural disturbance for many terrestrial ecosystem types that serve as Conservation Elements (CEs) for the Bureau of Land Management (BLM).  Historic and current land uses have resulted in significant departure from natural fire frequency and intensity.  These changes are exacerbated by multiple Change Agents (CA) including invasive annual grasses and will accentuate shifts in CE extent and distribution resulting from climatic alterations.  As part of the process of Rapid Ecological Assessment for BLM, we are evaluating multiple CAs in both a spatial and temporal context to examine the effects of land management decisions and climatic variation on CE distribution and landscape mosaic structure across the Central and Mojave Basin and Range ecoregions.  We used both tabular and spatial models to predict fire regime departure at the sub-watershed scale (i.e., HUC 10 units).  Integrating the fire regime models with predicted changes in climate envelopes provides a clearer understanding of how these ecological systems are likely to respond to multiple stresses.

The natural range of variability for each CE was initially derived from LANDFIRE Vegetation Dynamics Development Tool (VDDT) models.  The VDDT allows for the development of probabilistic quantitative model of CEs consisting of multiple ecological states with both deterministic and probabilistic drivers.  For any defined suite of drivers, the models predict the relative abundance of each state within a defined geography.   NRV represents the variation of these abundances after 500 simulations. 

Results/Conclusions

These models were then modified to include uncharacteristic states – states that have emerged over the past 100-150 years as a result of human land use changes and the invasion of exotic annual grasses.  These models were parameterized so that the resulting modeled mosaic structure resembled current land cover determined from the classification of satellite imagery.  Predicted changes in temperature and precipitation were then utilized to estimate changes in fire return intervals, fire severity, and probability of transitioning into a different ecological system type.    These results were then summarized by HUC 10 watershed unit to examine changes in CE distribution and structure across each ecoregion. This methodology is just now being established for application to all fire-driven vegetation types within each ecoregion.