Courtesy Dr Scott Rupp (UAF): predictions of cumulative area burned using actual data (CRU) through 2002 for the top 5 GCM climate models for Alaska, and a composite model (green line). The black line from 1950 to 2006 shows the actual fire history data compared to the green line of simulated data.

These maps show how vegetation may change between 2000 and 2025 for the Kanuti National Wildlife Refuge. This example simulation (not calibrated for this particular subarea) shows a 30% decline in white spruce and a 25% decrease in black spruce. Deciduous increases by approximately 40%.

In mid-February about 35 fire staff, refuge biologists, refuge managers and other Fish and Wildlife Service (FWS) employees met to see how refuges may change over the next 100 years as fire frequency and disturbance changes with the climate.  Drs Scott Rupp and Paul Duffy presented results of research at the University of Alaska Fairbanks (UAF), funded in part by a Joint Fire Science Program grant administered by Karen Murphy, Regional Fire Ecologist for FWS.  We saw an overview of how the newest global climate models have been downscaled for the Arctic and Alaska and the first results of a state-wide analysis to see how the top 5 climate models are likely to affect fire regimes and the vegetation on the landscape over the next 100 years.  The results confirm the general expectation of increasing fire occurrence and fire size into the future but we also learned how crucial these next 20 years may be in setting the landscape patterns far into the future.                              

Each of the 12 fire-affected refuges in Alaska were also given the subset of the state-wide data to run climate and landscape simulations for their refuges.  The landscape simulation model, Boreal ALFRESCO, shows general vegetation patterns (tundra, white spruce, black spruce and deciduous) that change as a function of fire occurrence, climate, and time since the last disturbance.  These broad classes can be used to evaluate the potential availability of habitat for species dependant on different seral stages of vegetation.  The model can also be used to test potential changes to the landscape as a result of changes to fire management.  Twenty biologists and fire management staff spent a day of hands-on training learning how the model operates and running some test simulations for their refuges.  These refuge-scaled Boreal ALFRESCO models are the first of several products that are designed to help refuges, and others, understand how our management actions may change the landscapes of the future under different climate scenarios.