Thursday, August 7, 2008: 9:50 AM
202 B, Midwest Airlines Center
James H. Thorne, Department of Environmental Science and Policy, University of California, Davis, Davis, CA
Background/Question/Methods The Wieslander Vegetation Type (VTM) maps provide detailed landcover data from the 1930s for 1/3 of California. Comparison of VTM maps to modern landcover maps permits assessment of both change in total area occupied and in ecotonal dynamics. To standardize scales between maps, we resampled both maps to 300m grid, and assigned the majority landcover type as the label for each cell. This permits development of a landcover transition matrix, representing change in dominant landcover through time. We examined area-based changes in the San Francisco Bay Area (SFBA) to explore the impact of urbanization. We examined area and ecotone changes in the Sierra Nevada mountains for evidence of climate change by resampling historical climate trends to the same cell framework as the vegetation, and comparing what vegetation change was associated with what changes in monthly nighttime temperature lows and seasonal precipitation.
Results/Conclusions
The greatest change in the SFBA region was urban expansion. The great majority of this change occurred on lands that had already been cleared for agriculture, although some urbanization of natural vegetation types was also measurable. This trend is similar to what is now occurring in California’s central valley, and suggests potential future food security risks, if appropriate zoning is not developed to protect agricultural lands.
Some shifts in the lower ecotones of vegetation types are evident in the Sierra Nevada. This talk will present upslope retraction of ponderosa pine forests, fir-dominated forests, and subalpine conifer forests. We also present evidence of expansion of hardwoods in regions where such expansion is expected under future climate conditions. Changes discussed are consistent with expected change under future shifts in climate, and provide an opportunity to test species distribution models with two independent, real world datasets.
