Simulation and Validation of Non-point Source Nitrogen and Phosphorus Loads under Different Land Uses in Sihu Basin, Hubei Province, China

AbstractMany water quality models have been extensively used to evaluate non-point source pollution, such as SWMM (Storm Water Management Model), AnnAGNPS (Annualized Agricultural Non-point Source), HSPF (Hydrological Simulation Program– Fortran), SWAT (Soil and Water Assessment Tool) and so on. However, these models’ construction tended to be complex, only used well in the condition of many parameters and normative data can be obtained. There were much problem to use these models in tidal plain, where topography were flat, stream network were complex and controlled by many water conservancy systems. The L-THIA (Long-Term Hydrologic Impact Assessment) model can be used in the areas with uncompleted data, evaluate the effects of land use on non-point source pollution, because it only needs land use, soil types and some rainfall data. But the output results of L-THIA model are the annual average runoff volume and pollution load, which cannot be obtained easily in in-site observation. In this paper, one of typical tidal plain, Sihu basin in Hubei province of China, was studied. The annual average runoff volume and the non-point source nitrogen and phosphorus load of each grid (16m*16m) and the spatial distribution were estimated by L-THIA model. The average load of non-point source nitrogen and phosphorus at each exit of the units were calculated based on the suppose that the pollution load is inverse to the distance of water flow. Model validation was executed by comparing calculated results and observed water quality data for study area. The results demonstrated that there were highly significant linear correlation between the calculated average load of non-point source nitrogen and phosphorus and the measured concentrations of TP and TN (TN: R=0.828; TP: R=0.957). These suggested that the average load of non-point source nitrogen and phosphorus were reliable to evaluate non-point source pollution. Moreover, the simulation results illustrated that the non-point source nitrogen and phosphorus loads under different land uses were quite different. The impact of dry lands on the non-point source nitrogen and phosphorus was the highest; that of paddy fields and residential areas were secondary; and the impact of woodlands to non-point source nitrogen and phosphorus was the minimum