The Characteristic Of Non-point Source Pollution And Identification Of Critical Source Areas For Plain Area

The North China Plain (NCP) is one of China’s most important social, economic, and agricultural regions. Increasing food demands associated with intensive agriculture production and expansion of irrigated land had led to eutrophication of surface water and serious shortage of water resources. Thus, to investigate the critical sources of phosphorus (P) loss and improve the P management strategies in North China Plain are important for the safety of water resources in this region. But for special characteristics of plain area, like homogeneous land use types, well-developed irrigation stream network, small rainfall-erosion, etc., existing methods were often limited by the departure from local reality or high error or the complexity of model formulation. In this study, based on the data which collected from Wuqing district of North China Plain, we put forward a new method of identification critical sources for plain area, and concluded that:1) Understanding the temporal and spatial variations and major sources of water pollution are important in the water resource management. In Wuqing District, the major pollutants in Wuqing section of Beiyunhe River were organic pollutants, ammonia, nitrate, pH and phosphorus. In short, point source was still the major source of pollution in the Wuqing section of Beiyunhe River. But phosphorus was origin from non-point source pollution, so it’s became the major control element of non-point source pollution.2) Runoff is the most important path way of the non-point source pollution. In this study, we investigated runoff characteristics of different pollutions from different land use types, farmland, forest, village, et al. The results showed the event mean concentrations (EMCs) of total dissolved N in forestland, farmland and village were3.14mg·L-1,12.68mg·L-1and17.81mg·L-1, respectively. Organic N, NH4+-N and NO3--N were the dominant species of dissolved N, accounting for33.6%,28.9%and35.0%of the total dissolved N, respectively. The N species with high proportion in the total dissolved N were organic N and NH4+-N in forest land, organic N and NO3--N in both farmland and village. EMC of total phosphorus in the order:farmland (0.44mg/L)> village (0.22mg/L)> forestland (0.17mg/L). The phosphorus species with high proportion in the total phosphorus was particle P (accounting for75%) in forestland, dissolve P (79%) in farmland, and particle P (48%) and dissolve P (52%) in village. Overall, due to weak of soil-erosion, the migration of contaminants is in the form of dissolved-based.3) A GIS-based method (Loss Index-LI) had been developed to evaluate the P loss potential at a district level. Three factors were formulated to characterize the annual event mean concentration(C), mean annual runoff volume (Q) and weight of potential pollution (α). Based on measured-data, and interpolation in GIS environment with other spatial datasets, this method generated maps that could be easily interpreted to support decision-making process. The maps of P loss potential were helpful for examining the regional pattern of diffuse P loss, and could facilitate the decisions of NPS pollution management at the district or town level.The pollution losses varied considerably depending on land use and farming practices. Land cover/use is the major controlling factor of nutrient pollution occurrence. But for special characteristics of plain area, like homogeneous land use types, existing methods of computing annual event mean concentration(C) were limited by the departure from the need of management. In this paper, the solution of homogeneous land use types was decomposition the major land use types. For smaller proportion of land units, the value was assignment by monitoring data. But for larger proportion land units, the values were assignment according the content of phosphorus in the soil.Mean annual runoff volume (Q) was estimated by the method of SCS-CN.For plain area, as well developed of agriculture production, irrigation canals were also very densely distributed. This has caused great difficulties to determining weight of potential pollution (α). In this study, we use method of Kernel Density to calculate the density of river network in the neighborhood of each output raster cell. This method not only takes into account the distance to the stream, but also considered the role of river density. That means the risk of the loss of non-point source pollutants was reduced with the increase of the distance from the river, and rose with the increase of drainage density.Finally, this method had been applied to Wuqing district, and concluded that the critical source areas of phosphorous, which indentified by LI, accounted for9.58%of area, but the load accounted for27.84%of the total.97.91%source area was farmland, rural residential and urban residential areas accounted for only1.93%and0.16%respectively. Therefore, to strengthen the management of agricultural production is the key to reduce non-point source pollution in the plain district.In short, the results show that this process can overcome the defects of subjective and limitation in applying, and can provide a reference to non-point source pollution assessment and management for similar area.