Effects of land-use change on nutrient discharges from the Patuxent River watershed

Abstract

We developed an empirical model integrating nonpoint source (NPS) runoff, point sources (PS), and reservoir management to predict watershed discharges of water, sediment, organic carbon, silicate, nitrogen, and phosphorus to the Patuxent River in Maryland. We estimated NPS discharges with linear models fit to measurements of weekly flow and 10 material concentrations from 22 study watersheds. The independent variables were the proportions of cropland and developed land, physiographic province (Coastal Plain or Piedmont), and time (week). All but one of the NPS models explained between 62% and 83% of the variability among concentration or flow measurements. Geographic factors (land cover and physiographic province) accounted for the explained variability in largely dissolved material concentrations (nitrate [NO3], silicate [Si], and total nitrogen [TN]), but the explained variability in flow and particulates (sediment and forms of phosphorus) was more strongly related to temporal variability or its interactions with land cover and province. Average concentrations of all materials increased with cropland proportion and also with developed land (except Si), but changes in cropland produced larger concentration shifts than equivalent changes in developed land proportion. Among land cover transitions, conversions between cropland and forest-grassland cause the greatest changes in material discharges, cropland and developed land conversions are intermediate, and developed land and forest-grassland conversions have the weakest effects. Changing land cover has stronger effects on NO3 and TN in the Piedmont than in the coastal Plain, but for all other materials, the effects of land-use change are greater in the Coastal Plain. We predicted the changes in nutrient load to the estuary under several alternate land cover configurations, including a state planning scenario that extrapolates current patterns of population growth and land development to the year 2020. In that scenario, declines in NPS discharges from reducing cropland are balanced by NPS discharge increases from developing an area almost six times larger than the lost cropland. When PS discharges are included, there are net increases in total water, total phosphorus, and TN discharges.