【作者】 付强；

【导师】 孙九林； 孔云峰；

【作者基本信息】 河南大学 ， 地图学与地理信息系统， 2013， 博士

【摘要】 中国是水资源严重不足的国家，目前也面临着水资源短缺和水环境污染的双重压力。点源污染和非点源污染共同作用导致了水环境的恶化，而随着点源污染防治水平的提高，非点源污染尤其是农业非点源污染已成为水环境污染的主要污染源之一。2010年公布的第一次全国污染源普查公报表明，畜禽养殖污染源是我国农业非点源污染的主要来源之一。然而，在中国的农业非点源污染管理与控制工作中存在着基础数据缺乏、实地实验区域覆盖不全、流域尺度定量化分析不足等突出问题。因此，在中国畜禽养殖污染研究中，基本数据获取与核算、结合土地利用方式特别是耕地进行产污分析、全国畜禽养殖产污规律等是研究的核心环节。鉴于此，本文尝试基于空间分析、数据驱动的中国畜养产污综合区划方法研究。研究目的是提出一种借助现有统计和调查资料进行全国范围畜养产污的分区方法，并进一步选择典型区探索区域影响因素及二级分区划分的方法。在国内外文献综述的基础上，本文的研究思路是：首先分析农业非点源污染机理，并构建畜养产污综合区划指标体系；然后，采用等值图分级、分布图系、重心曲线及分级合并的方法，参考中国自然区划、中国畜牧业综合区划、中国综合农业区划和畜养污染减排核算分区进行中国畜养产污的区域区划；最后，以东北区为典型区核算多尺度的指标数据，借助数理统计和空间分析方法进行主要影响因素研究，并探索基于主导因素的类型区划方法。研究取得了以下成果：（1）构建了中国畜养产污综合区划指标体系，包括自然条件、农业基础条件、农业社会经济条件、农业种养情况和产污情况等五类，其中共12个二级指标和31个三级指标；（2）设计了定量分析与定性分析相结合、区域区划与类型区划相结合的产污综合区划方案；（3）得到了产污综合区划一级分区方案，包括甘新蒙区、北蒙黑区、青藏区、东北区、黄土高原区、内蒙及长城沿线区、黄淮海区、西南区、长江中下游区、华南区等十个一级区；（4）以东北区为例，采用多尺度指标数据，通过多种统计方法筛选了东北区畜养产污的影响因素，并探索了各种因素的空间分异规律。基于以上研究成果，本文形成了下述结论：（1）本研究构建的中国畜养产污综合区划指标体系是合理的。在进行中国畜养产污综合区划及相关研究时，应综合考虑自然条件、农业基础条件、农业社会经济条件、农业种养情况和产污情况指标。（2）本研究设计的中国畜养产污综合区划方案是可行的。在进行全国综合区划研究时，建议结合定量分析与定性分析，在宏观尺度的区域区划中参考自然因素，在中观尺度的类型区划中参考经济因素。（3）采用多尺度指标数据，借助多种统计和分析方法能够进行影响因素和区划指标空间分异规律的研究。在进行畜养产污综合区划研究时，可根据不同的研究范围和不同的研究目标，选择合适的研究尺度进行分区规律的探索。更多还原

【Abstract】 China is one of the countries with a serious shortage of water resources. It is also facing the dual pressuresof both water shortage and water pollution. The water environment deterioration caused by non-pointsource (NPS) pollution as well as point source (PS) pollution, and with the PS pollution under control, NPSpollution, especially agricultural NPS (AGNPS) pollution has become one of the major source of waterenvironment pollution. The first national pollution census bulletin published in2010shows us the livestockand poultry sources is one of the main source of AGNPS pollution.However, there exist some prominent problems in the AGNPS pollution management and control works,such as lack of basic data, insufficient regional coverage of the field experiments, inadequate watershedscale quantitative analysis, and so on. Therefore, core part of the livestock and poultry breeding(LPB)pollution researches are coming, such as the basic data acquisition and calculation, and considering landuse patterns in pollutant generation analysis, especially arable land, and the spatial patterns of the nationallivestock and poultry breeding pollutant generation, and so on. In view of this, this paper focuses on thecomprehensive regionalization method of LPB pollutant generation in China, attempts to use spatialanalysis and data-driven approach. The objective of the study is to propose a regionalization method forLPB pollutant generation at a nationwide level. The method is based on existing statistics and survey data.And then it uses a typical region to explore the region’s influencing factors and influencing-factors-basedregionalization method.On the basis of literature review in domestic and international, the research mentality of this paper are:First analyse the mechanism of the AGNPS pollution, and constructs the indicator system for LPB pollutant generation regionalization; then using contour maps, thematic map series, related gravity centers curve andclassification merge methods, and reference the physical regionalizaiton of China, the comprehensiveregionalization of livestock in China, the comprehensive agricultural regionalization of China and thecurrent districting scheme of pollutants emission reduction and calculation, to support the regionalization ofLPB pollutant generation in China; And then, this paper analyse the dominant factors in a typicalregion(Northeast region), with a variety statistical and spatial analysis methods, calculating data at differentscales. Finally, the dominant factors are used to explore the type districting scheme of the typical region.The study achieves the following results:(1) This paper constructs the indicator system for LPB pollutant generation regionalization, includingfive categories of natural environment, basic agricultural environment, agricultural socio-economicenvironment, agricultural planting and breeding situation, and AGNPS pollution generation situation indexsystem. The five index system has12secondary indicators and31third-level indicators.(2) This paper design a research scheme for comprehensive regionalization of LPB pollutant generation,combining quantitative analysis with qualitative analysis, and combining regional districting with typedistricting.(3) First-level comprehensive regionalization scheme, including ten regions as followings:Gansu-Xinjiang-Inner Mongolia region, Northern Inner Mongolia-Heilongjiang region, Qinghai-Tibetregion, Northeast region, Loess Plateau region, Inner Mongolia and along the Great Wall region,Huang-Huai-Hai region, Southwest region, the Middle and Lower Reaches of the Yangtze River region andSouth China region.(4) This paper select the Northeast region as a typical region, and utilizes multi-scale data and manystatistical method to screen the influencing factors of LPB pollution in this region. In the end, this paper explores the districting scheme of the Northeast region by use of the dominant factors’ GWR coefficient.Based on the above research results, this paper’s conclusions as follows:(1) The comprehensive regionalization indicator system constructed in this study is suitable for the actualsituation in China. When carrying out comprehensive regionalization of LPB pollution in China and relatedresearch, the natural environment, basic agricultural environment, agricultural socio-economic environment,agricultural planting and breeding situation, and AGNPS pollution generation situation indicators should betaken into account.(2) The comprehensive regionalization scheme of LPB pollutant generation designed in this research ispracticable. When carrying out a comprehensive nationwide regionalization, it is recommended thatcombining with quantitative analysis and qualitative analysis, and reference to natural factors in themacro-scale regionalization, and reference to economic factors in meso-scale type districting.(3) Using multi-scale indicators data, and with the help of a variety of statistical and analytical methods,it can be able to carry out a study of the influencing factors and spatial patterns of regionalizationindex.When carrying out comprehensive regionalization of LPB pollution research, it is recommended thataccording to the different regional scale and different research objective to choose a suitable scale.更多还原