【作者】 阮晓红；

【导师】 金忠青；

【作者基本信息】 河海大学 ， 水力学及河流动力学， 2002， 博士

【摘要】 本论文结合国家自然科学基金项目——“农业面源污染负荷的产生及土壤背景参数的影响(49973029)”和近三年来所从事的科研及生产课题，对农业非点源营养类污染因子磷素和氮素在土壤环境和受纳水体环境中的迁移转化规律，农业非点源污染负荷及城市非点源污染负荷的定量化评价方法、非点源污染负荷对地表水体和地下水体的环境影响进行了较系统深入的研究，提出了相应的定量化评价方法。一定的土壤背景参数影响下的营养源(N、P)潜在流失量预测模型；建立了平原河网区农业非点源污染负荷模型、城市非点源污染负荷模型；一维平原河网地表水水质响应模型、一维垂向渗透地下水水质响应模型。本论文的主要成果有： 1、土壤背景参数对土壤中营养类污染物磷素和氮素的流失性影响分析 近年来，由于人工神经网络在非线性拟合方面的优势，使其在复杂系统的模拟中的应用越来越广泛。本文将此方法首次应用于对土壤中非点源营养类污染物磷素的潜在流失量预测，提出多变量插值的径向基函数(RBF)较之环境模拟中常用的反向传播网络(BP)在土壤磷素的流失性预测方面具有更好的拟合性。 2、非点源污染负荷定量化评价方法： (1)建立了城市径流污染负荷模型。模型考虑了城区土地利用的不同类型、街区非点源污染负荷的积累速率、干旱期污染负荷在雨水管网中的沉降过程，并分别建立了干旱期非点源沉降过程模拟模型、暴雨期径流过程的非点源模拟模型、污水管网非点源排放过程模拟模型，公式概念清晰，计算方便。 (2)建立了平原河网区农田径流污染负荷模型。模型考虑旱地径流、旱作径流、稻田排水、稻田渗漏等不同的土地利用类型下径流污染负荷特征，首次将产、汇流理论中陆域宽度的概念应用于农业非点源污染负荷的时、空分配，算例证明这种非点源污染负荷的分配方法对提高水质模型的预测精度有明显效果，为水质模型的实时预报提供了理论基础。 3、非点源污染负荷对地表及地下水环境的影响 (1)将水质分析模拟程序(WASP5)成功地应用于平原河网营养源污染物质组分的模拟及预测，并对模型作了进一步的改进和完善：将一维河网水量模型与水质分析模拟模型中的水质模拟模块相耦合，水质分析模拟模型与污染负荷模型相耦合，使水质模拟更方便、更具可操作性。 (2)建立了非点源污染负荷影响下的一维垂向渗透地下水水质响应模型，利用多孔介质中物质运移的基本理论推导了上述公式，利用实测资料进行了模型的验证和回灌区非点源对地下水的水环境影响分析。更多还原

【Abstract】 Based on the National Natural Science Foundation Program, the Produce of Non-point Source(NPS) Pollutant Load and the Influence of the Background Parameters of Soil (49973029) and the relative research work and production projects recently, this paper made systematic studies on the transportation and transformation principles of phosphorous and nitrogen from agricultural NFS in soil and receiving water, quantitative approaches for agricultural and urban NFS pollutant load assessment, and the effects of NFS pollutant load on surface and groundwater. And the corresponding quantitative methods were presented. Models developed in this paper are: agricultural NFS pollutant loading model in river plain area, urban NFS pollutant loading model; one-dimension surface water receiving model, one-dimension groundwater receiving model on vertical penetration, and one prediction model for nutrient loss with the influence of background parameters of soil. The details are as following:1. The influences of background parameters of soil on the loss of nutrient in soilRecently, the Artificial Neural Network (ANN) has been used excessively in complex systems because of its advantages in nonlinear equating. This approach was first used in analysis and determination of background parameters affecting the loss of nutrient factor in soil and the potential loss of nutrient from soil.2. Quantitative approaches for NFS pollutant loading assessment(1) Pollutant loading model for urban runoff was developed. Simulation models for sedimentation of NFS in dry season, NFS from storm runoff process, and discharge process of NFS from sewer were established with respect to urban land use, accumulation rate of NFS pollutant loading from streets, and sedimentation of pollutant load during dry season in storm drain. And different formulae might be applied in different NFS.(2) Pollutant loading model for agricultural runoff was developed. The basic theories of runoff producing and confluence were used in spatial and temporal distribution of rural NFS pollutant load for the first time, considering characteristics of runoff pollutant load occurring in various land uses. Calibration was completed using experimental data and receiving water models. The results showed that the applied method might improve the prediction accuracy of water quality model.3. The effects of NFS pollutant load on surface water and groundwater(1) Water quality modeling program (WASPS) was applied successfully to simulate and predict pollutant constituents of nutrient sources in river plain. Using one-dimension water net work quantity model, coupled with one part (water quality modeling) of water quality simulating model, and combining water quality simulating model with pollutant loading model can make it more easy and accessible to simulate water quality. Thus, this program was improved.(2) One-dimension groundwater quality model on vertical penetration with the influence of NFS pollutant load was developed. Based on the primary theories of mass transportation in porous media, the related formulae were derived. And this model was calibrated using the measured data.更多还原