Relations between Land Use and Stream Nutrient Concentrations for Small Watersheds in the Georgia Piedmont

Abstract

We have been sampling nutrient concentrations in 17 headwater streams within the South Fork Broad River (SFBR) watershed on a monthly basis since November 2001. The streams were classified as either developed (n=4), agriculture/pasture (n=4), mixed land use (n=6) or forested (n=3) based on information from the National Land Cover Data (NLCD) database. The lowest mean nitrogen concentrations were observed in streams draining forested watersheds. A set of landscape indicators explained 93 % of the spatial variability in the total nitrogen concentrations which suggests that watershed land use has an important effect on stream nitrogen levels. Total phosphorus varied seasonally at some of the sites, with high concentrations observed during the summer and low concentrations in winter. Only 4 out of 14 non-forested sites showed higher mean total phosphorus concentrations and no relationships were found between the landscape indicators and the total phosphorus concentrations in the streams. We suggest that, in contrast to what was observed for nitrogen, instream sources (e. g. decaying plant material), in-stream cycling, or strong pollution sources (e. g. manure, septic tanks) are more important than watershed land use in determining the total phosphorus concentrations in these small streams.