Modeling About Hydrochemical Variation Of Tangpu Reservoir Watershed And Total Pollution Amount Control

Reservoirs can be used for irrigation, power generation, flood control, fishing and supply of urban drinking water. Since1980s, the eutrophication of lake or reservoir which is caused by the excess of total nitrogen and total phosphorus tend to be normal in Taihu, Guanting Reservoir, etc. And the eutrophication really constrains the use of water in reservoirs. In this paper, the study in Tangpu Reservoir watershed is developed. Firstly the data of maps and statistics used in SWAT model were collected and analyzed; secondly the non-point source pollution was represented by parameters in SWAT model; thirdly a sensitivity analysis was adopted by SWAT self-sensitivity module; finally the calibration and validation of predicted values were developed. Based on the accomplishment of non-point source model, the effects attributed to different landuse types or different sources were quantified. So people can make change to reduce the inputs of nitrogen and phosphorus according to the analysis. What’s more, the concept of uncertain water environmental capacity was developed to control the total quantity of pollution. In the end, the measures to improve the quality of water body were made based on the target of environmental capacity and were coupled with SWAT model to predict the change. The whole research represents a whole performance of non-point source pollution in a watershed containing reservoir and a full system to control the non-point source pollution, providing an important reference to control total quantities of nitrogen and phosphorus.The main research results and conclusions of the dissertation are as follows:(1) NSE and R2of runoff and nutrients (TN and TP) in the periods of calibration and validation are all satisfied, so the model can represent the true progress of non-point source pollution in Tangpu Reservoir watershed.(2) Monthly river TN/TP load was strongly and positively related with river water discharge (p<0.05), respectively, with low TN and TP loads generally occurring in May and September. The corresponding runoff is little either and the TN/TP loads varied with the fluctuation of runoff. In the simulation period, the minimum load of total nitrogen and total phosphorus are5.118ton per month (October) and 0.001ton per month (October), respectively. And the maximum load of total nitrogen and total phosphorus are236.090ton per month (June) and0.001ton per month (June), respectively. So we should pay more attention to the non-point source pollution in June.(3) Annual TN export coefficient (kg ha-1a-1) for fourteen sub-watershed varied between10.516and16.760kg ha-1a-1, the critical source area is the sixth sub-watershed; annual TP export coefficient (kg ha-1a-1) is between0.007and0.687kg ha-1a-1, the critical source area is the first sub-watershed. The main reasons for severity of non-point source pollution are the intensity and ways of human activities. The critical factors in the key area about non-point source pollution in Tangpu Reservoir are the extent for exploiting and type of land use.(4) As to the TN export coefficient in different landuse type, orchard(83.109kg hm-2a-1)> paddy fields(47.229kg hm-2a-1tea garden(25.175kg hm-1a-1)>dry land(20.495kg hm-2a-1)> residential land(17.059kg hm-2a-1)>forest(6.998kg hm-2a-1); TP export coefficient in different landuse type:paddy fields(3.166kg hm-2a-1)> dry land (2.418kg hm-2a-1)> residential land (1.561kg hm-2a-1)> tea garden (0.016kg hm-1a-1)> orchard (0.011kg hm-2a-1)>forest(0.005kg hm-2a-1). The application of fertilizers contributed greatly to the pollution of nitrogen and phosphorus as well as the backgrounds value.(5) The results of uncertainty analysis showed that the TN/TP environmental capacities subject to normal distribution. And the simulation of TP environmental capacity resulting from the uncertainty analysis is23.337t/a (90%confidence level), while the simulation of TN environmental capacity is468.125t/a(90%confidence level). So the TP loads do not need to be controlled but the TN loads should be reduced by some measures like reducing50%fertilizers and letting rice field revert to forest. The spearman rank correlation coefficient of the input parameters shows that the replenishment coefficient for Dillon model is the most sensitive input parameter to calculate the TP environmental capacity. The replenishment coefficient is calculated by annual reservoir inflow and volume, additionally, reservoir volume is referred from the storage curve according to the corresponding water level, so the uncertainty of the capacity model could be minimized by reducing the measurement errors of reservoir inflow and water level.