The influence of landuse on aquatic macroinvertebrates in streams and rivers of South Carolina

The South Carolina Department of Health and Environmental Control (SCDHEC) began conducting bioassessments in 1974.  Since then there has been much landscape modification in South Carolian (SC) as a result of population growth and development. We analyzed the relationships of watershed landuse, human population density, and road density on the SCDHEC bioclassification score and bioclassification for 190 watersheds in the Piedmont Ecoregion of SC.  Spearman correlations between bioclassification score were relatively strong for most landscape variables: percent developed land use (r= -0.46, p<0.0001) percent undeveloped land use (r= 0.40, p<0.0001), population density (r= -0.41, p<0.0001), and road density (r= -0.44, p<0001). There was little correlation between bioclassification score and percent agricultural land use (R=0.15, p=0.04). Results from ANOVA indicated that percent developed, percent undeveloped, population density, and road density varied significantly across bioclassification categories (p<0.001). Duncan’s pairwise comparisons were used to test for post hoc differences in landscape variables between bioclassification categories. The Poor/Fair category could be distinguished from Good/Fair, Good, and Excellent categories for all landscape variables except percent agriculture, which was not a good predictor of bioclassification categories (p=0.03).  Results were similar for stream condition with all landscape variables except percent agriculture varying significantly across stream condition (Impaired, Unimpaired).  There was a wide range of watershed development for streams that were classified as impaired (3%-89%).  However, the maximum amount of development that was associated with unimpaired streams was 21%.