【作者】 何锋；

【导师】 孙丹峰；

【作者基本信息】 中国农业大学 ， 土地资源管理， 2014， 博士

【摘要】 区域土地利用/覆被系统变化产生的非点源污染压力日益增加,导致流域水质环境风险突出。本研究以北京山区流域为研究区,以多准则综合评价与土地利用系统协同方程开展农业非点源(以及农村居民点等分散无处理的点源扩散)污染风险分区,并利用半机理经验SPARROW模型分析流域水质总氮污染源,为解决缺资料地区的农村面源污染流域水土环境快速评价与区域水土资源管理提供技术方法和决策参考。本研究的主要结论有：(1)基于非点源产生、迁移、削减的各个过程研究,应用多准则分析的非点源污染评价方法,借鉴磷(P)指数及PNPI (Potential None-point Pollution Indicator)指标参数的选取原则,选择土地利用因子、径流因子、河流沟渠距离因子、土壤侵蚀、土壤渗透率5个因子,评价非点源污染物进入地表水体的潜在风险。通过改进的理想解法(TOPSIS)对5个因子客观赋权重,通过计算风险指数(NPA)对北京山区流域划分了潜在污染区(0-0.3)、轻度污染区(0.3-0.5)、中度污染区(0.5-0.7)、强度污染区(0.7-0.8)和重度污染区(0.8-1.0)五个等级的水污染源分区,且所占面积分别为42%、36%、14%、6%和2%(即面积分别为838.170、789.165、311.847、128.829、51.358km2).(2)基于各非点源环境风险区,应用序参量的特征值分析了农业土地利用系统的演化特点,探讨了农业土地利用系统耦合协调度在各非点源环境风险区的空间变化以及在密云县的时间演变过程。研究发现,主导北京山区流域土地利用系统的序参量为造林总面积、农村居民点边缘密度、恩格尔系数、灌溉面积、人口自然增长率。从空间变化来看,山区农业土地利用系统在潜在污染区中三个子系统的发展速度均为正,系统的序参量矢量指向第1象限,系统处于综合协调发展型。轻度污染区和中度污染区处于第Ⅱ象限,属于经济减速型。强度污染区和重度污染区处于第Ⅵ象限,属于生态调整型。进一步对协调演化状态的分析得出,潜在污染区处于0<α<45,对应协调演化类型Ⅱ。从时间变化来看,在1980-1994年内,密云县农业土地利用系统的序参量矢量指向第Ⅱ象限,系统处于经济减速型,1994年以后,流域农业土地利用系统的序参量矢量指向第1象限,系统处于综合协调发展型(社会、经济、生态协调发展)。(3)应用SPARROW模型,以密云水库流域为例,模拟TN污染物从产生至到达监测站点的传输及衰减过程。研究表明：SPARROW模型模拟精度R2为0.689,模型通过了统计性检验,且监测点模拟值与实测值之间的残差相对较小。对流域TN有显著贡献的因素为水产养殖,对TN传输具有重要影响的环境因子为土壤渗透率、土壤pH,对TN传输具有重要影响的环境过程为河段中的一级衰减反应。从水质来看,支流水质较好,干流水质较差。从空间分布来看,污染产率较高的子流域主要分布在各支流的源头河段,TN在子流域河段间的衰减表现出干流衰减比例大,支流衰减小的规律。密云水库上游流域总氮浓度在0.2-0.5mg/L之间,东帽湾及上游干流河段、密云水库周边流域、潮河遥桥峪水库上游干流河段流域总氮浓度大于1.5mg/L。总氮产率较高的子流域主要分布在各河流源头,最大产TN在4-8kg/ha.yr之间。对流域总氮来源进行解析,得出水产养殖最大贡献率分别在35%和65%。而土地利用增量流域分析结果表明,水产养殖用地预测增量流域产率在第90个百分位时为9.69kg/ha. yr。更多还原

【Abstract】 With the increasing pressures from regional land use/cover change for surface water quality, this study took Beijing northern mountainous areas as a case study, developed an evaluation approach for the rural non-point source pollution and an comprehensive evaluation by the multi-criteria analysis method basing on the total Nitrogen pollution. The SPARROW model was calibrated to simulate the components of the total Nitrogen pollution of the watershed. It provides references for the evaluation of the land and water environment of the rural non-point source pollution and the technical methods of the water and land resources management in areas where there are lack of data. The main conclusions are as follows.(1) Basing on the emerging, migrating, and reducing of the non-point source pollution, applying the multi-criteria analysis on the evaluation of the non-point source pollution, and combining the rules of factors selection for the indexes of P and PNPI, the5indexes of land use, water runoff, distance, soil erosion, soil permeability are chosen to evaluate the risks of the non-point pollution entering into the surface water. The modified TOPSIS method was employed to give objective weights for the5indexes. By calculating the risk index (NPA),5pollution categories are identified, which are:latent contaminated area (0-0.3), mild contaminated area (0.3-0.5), moderate contaminated area (0.5-0.7), severe contaminated area (0.7-0.8) and extremely severe area (0.8-1.0). The corresponding percentages of the areas are42%(838.170km2),36%(789.165km2),14%(311.847km2),6%(128.829km2) and2%(51.358km2).(2) Basing on the environmental risk zones of the non-point area, the characteristic value of the order parameter is used to analyze the evolution of the rural land use system. The spatial change of the coupling coordination degree of the agricultural land use system among the non-point pollution zones is discussed. Results show that the dominant factors affecting the order parameter of the agricultural land system for the Beijing mountainous area are:the foresting area, the edge density of the rural resident, the Engel coefficient, the irrigation area, the natural increase rate of the population.In terms of the spatial changes, the development speeds of the three sub-regions of the agricultural land use system are positive among the latent contaminated areas. The vector of the system order parameter points to the first quadrant, and the system is in the state of comprehensive development. The mild contaminated area and the moderate contaminated area are in the second quadrant, which are in the state of the decrease in economic. The severe contaminated area and the extremely severe area are in the sixth quadrant, which belongs to the type of ecological modification. The evolution state is further analyzed and results show that the a of the latent contaminated area is between0and45, and the corresponding evolution type is type Ⅱ. In terms of changes in time, between1980and1994, the coupling coordinate degree of the agriculture land use system in Miyun watershed area point to the second quadrant, and the system belongs to the type of economic decrease. After1994, the coupling coordinate degree of the agriculture land use system in Miyun watershed area point to the first quadrant, and the system belongs to the type of comprehensive development (harmonious development in society, economic and ecologic).(3) The SPARROW model was used to simulate the transmission and recession of the total nitrogen (TN) pollution from the beginning point of emergence to the ending point arriving at the investigation station using Miyun reservoir as a case study. Results show that the R squared of the SPARROW models is0.689, and the variations between the observed values and the simulated values are small. The contributed factor to the TN is aquaculture. The contributed environmental factors to TN are the soil permeability, the soil PH value, and the main environmental process affecting the TN transmission is the first level recession. In terms of the water quality, the quality is better in the branch stream than that in the main stream. In the spatial distribution, the watersheds with high polluting rate are in the original area of the branch stream, and the recession rates among the branch streams are higher in the main stream area, and lower in the branch streams.The TN density is between0.2and0.5mg/L in the upper stream of the Miyun reservoir watershed, while the TN density is larger than1.5mg/L in the upper main stream of Dongmaowan, area around Miyun reservoir and the upper main stream of the Chaoheyaoqiaoyu reservoir. The sub-regions with higher TN are distributed at the origins of each stream, with the largest production at4to8kg/ha.yr. The decomposition of the origins of TN, the factor of aquaculture contributes to up to35%and65%of the TN. The results of analysis on the incensement of the land use watershed show that the predicted value of watershed production of the land use for the aquaculture is9.69kg/ha.yr at the level of the90th percentile.更多还原