Modeling Spatial and Temporal Dynamics of Montane Meadows and Biodiversity in the Greater Yellowstone Ecosystem
Montane meadow communities can function as early indicators of change because they are highly sensitive to variations in precipitation and temperature. However, before an accurate estimation of directional change rates may be made with confidence, the seasonal and interannual rates of change inherent to a system must be quantified. We are using a time series of satellite multispectral imagery for monitoring the extent, condition, and spatial pattern of montaine meadows in the Greater Yellowstone Ecosystem on a seasonal and interannual time scale. Spectrally-based, spatially-explicit models were developed for six meadow types using a GIS to stratify the study area by topography and geology.
The objectives of this research are to 1) quantify the spatial and temporal variability in montane meadow communities; 2) develop a spectrally-based spatially-explicit model for predicting plant and animal species diversity patterns in montane meadows; and 3) test the spectrally-based spatially-explicit model for predicting plant and animal species diversity patterns in montane meadows.
Field sampling is being used to collect data on the distribution of plant, bird, and butterfly species. We have sampled for two years in two regions of the ecosystem: the northern part of the ecosystem, hereafter termed the Gallatin study area, included the Gallatin National Forest and northwestern portion of Yellowstone National Park; the southern part of the ecosystem, hereafter termed Teton study area, included Grand Teton National Park.
Twenty-five sample sites were located in the Tetons and thirty sample sites were located in the Gallatins. Birds, butterflies, and plants were surveyed at each of the sites. The bird community showed a 47-59% similarity between the Gallatins and the Tetons. Bird species composition in the hydric meadow type was the most accurately predicted in both study areas. The butterfly community showed 60-65% similarity in species composition between the two sampling areas. If meadows were collapsed into three categories rather than five, meadow types at the two extremes of the gradient were 90-100% predictable while the mesic (middle gradient) meadow type was less easily predicted.