Assessment of grass/shrub habitat fragmentation in the Whitewater Watershed using GIS and Spatial Linear Regression to model Sensitive Species Population

Fragmentation analysis of the Whitewater watershed, in southeast Minnessota, revealed 4 structural measures of grass-shrub habitat that were significant predictors of sensitive species population densities. Models were developed using simple linear regression and further refined to incorporate spatial autocorrelation using a Moran’s test. Significant variables were divided into a five class ordinal model based on Jenks Optimization Method. Ordinal values were summed to determine an overall measure of subwatershed restoration potential. Results suggest that grass-shrub habitat should be restored in 0.6 ac patches equally dispersed about the landscape to optimize sensitive species densities.