【作者】 王培；

【导师】 马友华； R.Anlauf；

【作者基本信息】 安徽农业大学 ， 土壤学， 2008， 硕士

【摘要】 农业面源污染目前是造成我国水环境污染的主要原因之一,且农业面源污染是一种间歇性,随机性、不确定性很强的复杂过程,估算其负荷难度较大。通过农业面源污染数学模型与地理信息系统(GIS),遥感(RS)的结合对农业面源污染进行计算机模拟,分析其时空变化规律,是农业面源污染控制研究的有效手段。本文全面阐述了国内外农业面源污染模型的发展,在分析基于GIS的SWAT(SoilWater Assessment)模型的方法技术原理的基础上,以德国Nette流域为研究区域进行流域水文及养分迁移的分布式计算机模拟。主要目的是研究养分迁移的分布式模拟方法并进行区域适用性验证。主要内容如下:■分析SWAT模型原理技术方法,模型运行特点,对土地利用数据,土壤数据气象数据等分别进行相关标准的转化与数据转换,按照模型运行所需的数据结构特点建立一套农业面源污染基础信息库。利用数字高程模型(DEM)提取流域河网,流域边界,进行子流域的划分,并进行水文响应单元(HYU)的划分,完成流域的空间离散化和参数化。■在流域出口收集相关水文水质监测资料数据,选取2001-2003年段的实际监测数据为参照分别对模型水文与氮磷模拟进行参数敏感性分析,通过对敏感性参数的调整对水文模拟与氮磷模拟进行校准,得到一套较为合适的模型参数。■分析模型结果并进行空间分析与模型适用性分析,在空间尺度上氮磷流失基本呈由北向南逐渐汇集的特点,主要的氮磷输出集中在中部,南部的农用地与农村居民点区域,北部林地区域较少。以1996-2006年段的流域出口水文水质监测数据与模型模拟结果进行比较,水文模拟取得了较高的精度,硝态氮、矿质磷,氮磷负荷基本反应流域实际流失状况,但个别年份误差较大,由此表明SWAT模型对流域产水,产沙的长期变化可以较好的预测,但在特殊气象条件下的地理过程中,模拟结果存在误差。对于氮磷迁移的过程模拟,其适用性有待进一步研究。更多还原

【Abstract】 One of the major reasons,which is agricultural non-point source pollution,causes water environment pollution in China,and this kind of pollution is intermittent,random,strongly uncertain and complex process.It is really difficult to estimate its value of load.It is the effective method to control and research agricultural non-point source pollution,through the combination of digital model of agricultural non-point source pollution,GIS and RS. This method is simulated by the computers,and analyzes the regulation of space-time changing.This thesis completely demonstrates the development of agricultural non-point source pollution in domestic and overseas.Analyzing SWAT model which is based on GIS software with ArcSWAT technology principle,and doing research in the Nette drainage area to get the simulation of watershed hydrological and the distributed simulation methods of nutrition immigrating.The main purpose is to research the distributed simulation methods of nutrition immigrating and verify regional applicability.■Analyzing the SWAT model’s principle,technology and method.The characters of model when it works.Transforming the soil applying data,soil data and other relative data with the standard ways.According to the character of data’s construction needed by the model working,,establishing one set of basic agricultural non-point source pollution information database.With DEM, extracting watershed network and watershed boundary,dividing the sub watershed rivers,and then making HYUs divided.Finally,discrete and parameterize the space of watershed.■Collecting relative monitoring data about hydrologic and water quality from the basin outlet.Selecting the true data from 2001-2003 monitoring,and comparing with watershed model and simulation of nitrogen and phosphorus.Analyzing the parameter in the sensitive methods.Via this way,to rectify the simulation of nitrogen and phosphorus,at least,getting one set of relative appropriate model parameters.■Analyzed the results of model,space and application of model,the character of lacking nitrogen and phosphorus is from North to South in the dimension of space. The main reason is that the nitrogen and phosphorus is gathered into middle area, and South is farm land and farmer resident,and the North is less forest areas. Compared with hydrologic and water quality from the basin outlet monitored from 1996 to 2006,the model simulation gets the more accuracy.The load of nitrate-N, mineral-P and N-P basically reflects the losing status in that watershed,however, there are several months existing great errors.From the point of this view,SWAT model can predict the watershed producing water and sands very well,but there also be some errors existing in certain climate situation controlling geography process.To the immigrating nitrogen and phosphorus,we have to further research its application.更多还原