【作者】 张秋玲；

【导师】 陈英旭；

【作者基本信息】 浙江大学 ， 环境工程， 2010， 博士

【摘要】 我国水环境形势十分严峻,农业非点源污染已经成为水环境恶化和湖泊富营养化的重要原因。因此,控制农业非点源污染、保护生态环境对实现我国农业、资源和环境协调发展具有重要的理论和现实意义。太湖流域作为我国第三大淡水湖,随着经济的发展,社会水平的提高,目前水环境已经处于富营养化状态,农业非点源污染已经成为太湖水污染的主要来源之一,并将成为工业和城市点源污染得到控制后所要面对的最主要污染源。杭嘉湖地区作为太湖流域的八大水系之一,是浙江省水污染较为严重的地区之一,平原河网90%以上的监测断面不能满足水功能要求,京杭运河浙江境内全部河段水质均不能满足功能区水质目标。其中,农业非点源污染是其主要的污染源,大量化肥流失及其对水环境的影响日益严重并受到人们的关注。鉴于此,本研究结合运用水文学,地理学,景观生态学和环境学等在这一领域的已有研究成果,利用GIS技术,着重从环境学和景观生态学的角度,进行杭嘉湖区域的农业非点源污染模拟研究,分析不同土地利用和景观格局的动态变化,提出控制流域非点源污染的最佳管理技术体系和景观优化方案,对保护农业水环境具有现实意义。主要研究内容和结论如下：1、建立杭嘉湖区域非点源污染基础数据库。以RS与GIS技术为支持,构建了研究区数字高程模型(DEM),土地利用/覆被和土壤类型等一系列非点源污染的空间数据库；通过资料收集,调查统计和野外监测采样,结合实验室试验,构建研究区非点源污染的属性数据库。2、进行农业非点源污染源解析和杭嘉湖区域京杭运河嘉兴段小流域的污染物时空分布规律研究。以GIS技术为平台,对研究区的污染源进行分析,明确了农业非点源污染在研究区域所占的分量。结果表明,杭嘉湖地区农业化肥产生的氮磷含量远远大于生活面源和畜禽养殖,排放量大小顺序为农业化肥>畜禽养殖>生活面源。其中,嘉兴市产生的氮磷排放量相对比较大,对水环境影响比较大；NH4+-N的排放量相对集中在嘉兴和余杭。选取嘉兴双桥农场附近的京杭运河嘉兴段进行营养物时空分布规律研究。NH4+-N高峰值主要集中在4月,7月以及9月到10月期间,最高值在7月份。居民区,农田和养殖场附近的水体污染程度比较严重。3、分析和评价SWAT模型在杭嘉湖区域的适用性和可靠性。针对杭嘉湖区域中两种不同的地形(平原和山地)进行模型模拟研究,分别选取代表性小流域王江泾和杭长桥,对模型的关键性参数进行敏感性分析,结果显示：对径流影响最大的敏感参数是曲线数(CN2),土壤有效含水量(Soil_AWC),蒸发补偿因子(ESCO)。并且利用2002-2005年的实测数据对模型参数进行率定和验证,采用决定系数R2和Nash-Suttcliffe系数E对模拟结果进行评价,结果证明SWAT模型对杭嘉湖地区的非点源污染模拟具有一定的适用性,可用于研究区非点源污染的模拟研究。4、从时空变化、土地利用程度变化和土地利用类型转移等方面,对杭嘉湖的土地利用结构和动态变化进行分析研究。2000-2004年间,杭嘉湖地区土地利用类型主要以耕地和林地为主,五年期间二者的变化趋势一直是面积在减小。杭嘉湖区域西部山区的土地利用类型的主要变化特征为：2000-2004年间城镇用地面积增加最多,其次是旱地和水域。东部平原水网区的土地利用类型的主要变化特征为：2000-2004年间土地利用类型面积变化趋势类似于西部山区,城镇用地面积增加最多,其次是水域。而农村居民点、水田和旱地面积都在减少。2000-2004年间研究区域土地利用变化的数量很大,表现为城镇建设用地、水田、农村居民地和水域的变化。主要发生的土地利用类型转化为：水田转化为有林地、城镇用地、水域、旱地和农村居民地。5、根据提取的研究区域景观格局指数,分析了研究区景观格局动态演变过程。针对2000年和2004年的两期景观类型栅格图进行计算分析,斑块密度、斑块数目和形状指数呈上升趋势,最大斑块指数呈现下降趋势,说明整体景观有破碎化的倾向。6、针对杭嘉湖区域的土地利用类型的空间分布特征,东部平原河网区耕地和城镇用地占主要部分,起着“源”的作用,产生污染量相对比较大,所以,相对来说,东部平原水网区地表水质比较差。而西部山地丘陵区域大部分面积为林地,起着“汇”的作用,并且林地具有截留、过滤等作用,所以,相对于东部来说,地表水环境相对较好。7、针对杭嘉湖农业非点源污染特征,对非点源污染控制与治理的措施进行探讨性研究。结合景观生态学的观点,针对杭嘉湖地区特殊的景观组合,从水环境保护的需求出发,尝试性的探讨以保护水体质量为目标的景观格局优化方案。更多还原

【Abstract】 Lakes, water reservoirs, and streams, which are the most valuable sources of drinking water for the earth’s population, are vulnerable to pollution and degradation of water quality, particularly to eutrophication. Accelerated eutrophication of lakes and reservoirs is the most serious degradation of water quality in China during the last decades. With the effective control of industry point source pollution, the agricultural non-point source (NPS) pollution has already become the main pollution source in water environment nowadays, which resulted in Water body worsen and Lake deteriorated. Therefore, it has the important theoretical and practical significance to control agricultural NPS and protect the ecological environment for the development of the agriculture, resource and environment.With the development of economy and society, the issue on water eutrophication becomes one of the greatest challenges in the third largest fresh lake in China, the Taihu lake.90 percent of the river network in HangJiaHu region, Zhejiang province, one of the eight water system in the Taihu basin, do not meet the need for the water function. Also the water quality of the Jinghang canal in the Zhejiang province do not attain the water quality standard for grade III. Nitrogen and phosphorus losses in agricultural non-point source has been attracted more and more attention since it becomes one of major factors in resulting in eutrophication of surface waters after controlling the industry and urban point source. An interdisciplinary exploration of the intersection among hydrology, geography, landscape ecology and environment science is encouraged. We researched on the agricultural NPS simulation in HangJiaHu area with the integration of the Geographic information system (GIS) and environmental model. Moreover, we analyzed the dynamic change of land use and landscape pattern based on the point of view for environment science and landscape ecology. Finally, We advanced the best management technology system and optimized the landscape pattern which has the very realism significance for protecting the water environment. The main content and results are as follows:1、The foundation database of HangJiaHu region non-point source pollution simulation research is established. Based on the remote sensing and GIS, we built the spatial database including the Digital Elevation Model (DEM), two different periods land use/land cover type maps, soil type map. Moreover, we built the attribution database including the collecting data, investigation data and observed data outside and the data in the lab.2、We investigated the agricultural non-point source pollution in HangJiaHu region and analyzed spatial and temporal distribution of nitrogen and phosphorus pollutants in the typical Jiaxing segment of Jinghang canal. Based on the GIS technology, we analyzed the pollution sources and defined the agricultural NPS percent in the study area. Our results showed that nitrogen and phosphorus loss from agricultural fertilizer to the water environment is far larger than those of sewage and livestock. The losses order is agricultural fertilizer> livestock> sewage. In the HangJiaHu region, nitrogen and phosphorus loss from Jiaxing city to water environment is larger than the other areas. NH4+-N emissed mostly in Jiaxing and Yuhang. We selected the Jiaxing segment water of Jinghang canal as the experiment field where we surveyed the water pollution environment, monitored, sampled and laboratory experiments. The results showed the larger value of NH4+-N appeared in April, July and September to October in the temporal distribution. The largest value appeared in July. In the spatial distribution, the water in the vicinity of the residential areas, farms and breed fields were more seriously destroyed. The different land use types in conservation directly affect the different levels of nutrients for contributions of water pollution.3、Through SWAT model adjusting, simulating and verifying, the service ability and reliability of SWAT model are analyzed and evaluated in HangJiaHu region. In this paper, based on the landform, we divided Hangjiahu region into two parts, one as the west hill area of Hangjiahu, and the other as the east plain river network. Thus, the result showed the representation in the different landform. We considered sensitivity analysis for the specific parameters of the model, the results indicated:sensitive parameter which has the greatest influence on the runoff is Curve number (CN2), Soil available water content (Soil_AWC), Soil evaporation coefficient (ESCO). Moreover, based on the 2002-2005 observed data, these parameters were calibrated and validated for the surface runoff volume and nutrient loads in HangJiaHu region. Coefficient R2 and Nash-Suttcliffe coefficient E were selected to evaluate the simulation results. Simulated results closely matched with the observed data. It confirmed that SWAT model could simulate runoff and nutrient loads quite well, particularly in this complex region.4、Based on the spatial and temporal changes of land resources amount, land use degree and transition of land use types, this paper analyzed land use structure and dynamic change of land use types in the HangJiaHu region. Land use types in the HangJiaHu region focus on farmland and woodland which area diminished in 2000-2004. The main features of land use types in the western hill area of Hangjiahu region are:the increased area of urban land was the largest in the land use types whose areas increased, followed by Dry land and Water body between 2000-2004. The main features of land use types in the eastern plain area of HangJiaHu region are: Urban area has the highest increase in all the land use types for the increasing area, followed by Water body between 2000-2004. However, farmers’ settlement, Paddy fields and Dry land area reduced during this period. Land use changes were mainly Urban land, Paddy land and the Residential area in rural areas between 2000-2004. At the same time, the area of Water body changed largely with serious soil erosion. The transition between land use types has occurred mainly as:the paddy land were converted to Forest land, Urban land, Water body, Dry land and Farmers’ Settlement.5、As far as the landscape grid maps in 2000 and 2004 period is concerned, we selected the landscape indice such as patch density, patch number, shape index, and the largest patch index (LPI) to analyze the landscape pattern. The results showed patch density, patch number and shape index appeared the increase trend, the largest patch index (LPI) indicated the reduce trend. It suggests the whole landscape had a fragmentation trend. 6、Based on the spatial distribution pattern of land use types for HangJiaHu region, it showed the farm land and urban land in the eastern plain river network are the main land use type playing the role of "source". The amount of pollution from the "source" is relatively large. Therefore, the surface water quality of Eastern plain area was poor. While most of the mountainous and hilly areas in the western area of forest land, play the role of "load", which works with the intercept or filtering functions. Consequently, the surface water quality in the western is better than that of the eastern of the study area. The function of the "source" and "load" in the landscape has great impact on the nonpoint source. "Source" makes a function for making pollution and "load" make the function for the transportation. The results give a good advice for controlling the water pollution and protecting the water environment.7、As far as the HangJiaHu agricultural non-point source pollution characteristics is concerned, we analysed the best management technology system to measure non-point source pollution control which has been put forward. We also proposed the optimizing landscape project for the policy and management department integrated with the view of landscape ecology.更多还原