NONLINEAR RESPONSE OF A MIXED LAND-USE WATERSHED TO NITROGEN LOADING

Nonpoint source pollution comprises a significant portion of the nutri ent loads that adversely affect the water quality of freshwater and ma rine ecosystems. Current research indicates that reductions in nonpoin t source loading of nitrogen to a basin should be considered in manage ment decisions if significant nitrogen export reduction is to be achie ved. Managers need to know the percent reduction in nitrogen loading n ecessary to achieve a desired reduction in the export. What makes this task difficult to achieve is that watersheds respond non-linearly bec ause of the nature of hydrogeochemical processes that operate in a spa tially variable land use, soils, geology and vegetation environment. T he non-linear response of a watershed to nitrogen loading has been add ressed through field monitoring and mathematical modelling. This study included the monitoring of a small, multiland use watershed for two y ears, the development of a mathematical model that can simulate the mo vement of nitrogen and the evaluation of the non-linear response of th e watershed to nitrogen loadings. The site was the Muddy Brook watersh ed located in north-central Connecticut, U.S.A. The drainage area is 3 60 ha. The watershed land use consists oi. 54% unmanaged forest, 23% r esidential areas, 12% grass and hay lands and]IB maize fields. The Nut rient Transport and Transformation (NTT) model was used to assess the fate and transport of nitrogen in the watershed. The model is a distri buted code that can simulate both the hydrology and chemistry of water sheds. Two management scenarios were simulated to assess the non-linea r response of the Muddy Brook watershed. The first scenario simulated the possibility of reducing the nitrogen loading generated from a resi dential area by eliminating the septic tank input. The results indicat ed that there is going to be a 25% reduction in nitrogen export from t he watershed if that scenario was followed. This dramatic reduction in export primarily results from the residential development being very close to the stream and then is a preferential flow that short-circuit s the groundwater to the stream. On the other hand, an 100% reduction in fertilizer and manure input to the agricultural land will only resu lt in less than 1.5% reduction in nitrogen export after a three year o f simulation period. These simulations illustrate the non-linear natur e of the Muddy Brook watershed response to nitrogen loadings and can b e used to guide managers in selecting appropriate best management prac tices to improve the water quality of the stream. (C) 1998 Elsevier Sc ience B.V.