【作者】 何栋材；

【导师】 徐中民；

【作者基本信息】 西北师范大学 ， 人文地理学， 2009， 博士

【摘要】 张掖市甘州区位于西北内陆地区,黑河的中上游,由于气候、地表物质、植被、土壤和人类活动的长期作用而形成了独特的绿洲农业。近年来,随着流域管理的展开,张掖市甘州区需向地处黑河下游的额济纳旗分水,当上游莺落峡来水量为15.8亿m3时,要向下游的正义峡下泄9.5亿m3水量。于是张掖市甘州区的缺水问题凸显了出来。而水资源一向居于由水-土-气-生-人类社会经济活动五大单元组成的复杂巨系统的核心,缺水必然影响其余组成单元。传统的水资源评价,拘囿于水资源系统内部,人为割裂了其与人类社会经济系统之间的内在的本质必然联系。为了补其不足,引入水贫困的概念。这一概念是建立在一般的贫困理论、可持续生计理论和地理环境决定论的基础之上。为了较为准确测量甘州区的水贫困状况,引入英国牛津大学地理学院生态水文研究中心(CEH)开发的WPI综合指数的方法,此方法将WPI分解为五个一级指标,每个一级指标又分解为3-10个二级指标。结合张掖市甘州区的水贫困现状,采用与CEH相同的一级指标,考虑到数据的可得性,结合甘州区水资源开发、利用、保护和管理实际,确定了25个二级指标。以实地调查、问卷调查和查阅文献、年鉴和报告的方式,分别获得第一手数据和第二手数据。经过同趋势化数据处理,用均衡法求得甘州区8个灌区的水资源状况(resources),供水设施状况(acess),利用效率和结构(use),用水能力(capacity)和水环境状况(environment)5个分指标的得分,在此基础上,汇总求得8个灌区的WPI总得分。在水资源分指标中,乌江灌区得74分,花寨灌区得9分;在供水设施分指标中,甘浚灌区得61分,安阳、乌江同得29;在利用效率与结构分指标中,盈科灌区得85分,安阳灌区得9分;水的利用能力分指标中,盈科灌区得79,花寨灌区得分为0;水环境分指标中,盈科得97,安阳得分30。在WPI总得分中,盈科灌区以72.6居第一位,安阳灌区以16.2分居末位。为了了解各分指标及WPI的空间分布,应用arcviewGIS3.2做了各分指标及汇总后的各灌区的WPI空间分布图。由于WPI分析方法中采用了25个变量,致使变量数目过多,加之许多变量之间高度相关。为了进一步了解25个变量中究竟哪些变量的贡献大,应用PCA方法对此25个变量进行了主成份分析。经过对原始变量的同趋势化和标准化处理,根据各自的方差贡献率和累积方差贡献率,从原始25个变量中提取了4个主成份,在求得每个主成份得分表达式的基础上,根据各主成份的方差贡献率求得主成份综合评价方程式,并据此求得甘州区8个灌区的主成份综合评价得分并进行了主成份得分排名,这一结果与WPI综合指数法求得的结果相比大同小异,其它灌区的得分位序没有变化,仍然是盈科(1)、大满(2)、西干(3)、甘浚(4)、花塞(7)、安阳(8),只是在WPI中,上三(5),乌江(6),在主成份分析中换成了乌江(5),上三(6)。为了更进一步了解25个变量对主成份得分的影响,以主成份得分为因变量,以25个原始变量为自变量,进行主成份回归,回归的结果,18个变量被系统自动剔除,另外7个变量与主成分得分之间有确定的数量关系。为了探寻影响WPI的主要因素,以WPI得分为因变量,以各灌区水管人员总数、农民人均纯收入和各灌区的社会资本为自变量进行多元回归分析,结果为社会资本与WPI之间的线性关系不显著,而农民人均纯收入和灌区尺度的水管人员总数与WPI之间具有确定和显著的线性关系。究竟被剔除的18个变量与主成份得分之间以及社会资本与WPI之间是何种关系,需要在今后的研究中做深入的探讨。更多还原

【Abstract】 Ganzhou,Zhangye City, locates at the north-western inland areas,belongs to the upper and middle watershed of Heihe river, an unique oasis agriculture formed for long-term roles of climate , surface material, vegetation, soil and human activities. With the watershed management in recent years, Ganzhou area spills its water to the downstream, Ejinaqi, when water from Yingluoxia, the upperstream reaches to 1.58 billion m3, it discharges 950 million m3 to the downstream, So water shortage becomes an serious issue in this area. It is known that water resources occupys the core in the complex giant system compsoed by five modules, water - soil -gas– living creature- human socioeconomic activities, water shortage will inevitably affect the rest of the component units. Traditional water resources assessment only focus on water resources internal system and neglects its intrinsic relation with human being’s socio-economic system. Therefore,the concept of water poverty is introduced. The concept is on the basis of the general theory of poverty, sustainable livelihoods and geographical environment. For more accurate measurement of water poverty in Ganzhou area, WPI composite index approach is used, which is developed by University of Oxford Ecohydrological Research Center (CEH), the United Kingdom Institute of Geography. this method is to divide WPI into five level-1 indicators, each level-1 indicators is also divided into 3 to 10 levle-2 indicators. With the current situation of water poverty in Ganzhou area in Zhangye City,the same CEH level-1 indicators are used , taking into account the availability of data, combined with water resources development, utilization, protection and management of the actual situation in this area, 25 secondary indicatorsare are identified. By field survey, questionnaire and literature review, Yearbook and reports, respectively got the first-hand and secondary data. with the trend data and by use of balanced method gains the 5 sub-indicator scores on water resources, water supply facilities acess, utilization efficiency and structure, water capacity and water environment situation in eight irrigation areas among Ganzhou,Zhangye City and in this basis, the WPI total score is obtained for the eight irrigation areas. Among the sub-indicator of water resources,Wujiang irrigation area scores 74 pionts, Huazai irrigaiton area is 9 points; Among the sub-indicator of water supply facilities, Ganjun is 61 points, Anyang and Wujiang is the same score 29 point; For the sub-indicator of utilization efficiency and structure, Yingke irrigation area scores 85,Anyang is 9; For water capacity Yingke is 79, Huazai is 0; Among sub-indicator for water environment, Yingke is 97,Anyang is 30. For the total score of WPI, Yingke ranks the top with 72.6 point, Anyang is on the bottom with 16.2 point. In order to know the sub-indicators and the spatial distribution of WPI, arcviewGIS3.2 is applicated to get WPI spatial distribution map of every irrigation area. In WPI analytical methods 25 variables will have to be used, 25 variables is an excess number, in addition many variables have a high degree of correlation, In order to know which variables contributes more , PCA method is used to analyse the principal component of the 25 variables. By process of trending and standardization of original variables, with contribution rate of variance and cumulative variance, extracted four principal components from the original 25 variables. In the basis of the principal component scores caculation,by the variance contribution rate of each principal component to achieve comprehensive evaluation equation for the principal component,by this way achieved the comprehensive evaluation of the principal component scores in the 8 irrigation areas in Ganzhou area and ranked them. The result is consistent with what gained by WPI composite index , the scoring rank of other irrigated areas has not changed, Yingke still ranks the first, Daman the second, Xigan is the third, GanJun ranks the fourth , Huazai is the seventh, Anyang is on the bottom , only just in the WPI, Shangsan ranks the fifth, Wujiang ranks the sixth , in the principal component analysis Wujiang the fifth and Shangsan ranks the sixth. In order to further understand the 25 variables impact on principal component scores , input principal component scores as the dependent variable, the 25 original variables as variables for principal component regression, the results shows 18 variables were automatically removed, the other 7 variables determine the quantity of principle component scores. In order to explore the main factors affect the WPI ,input WPI scores as the dependent variable , the total number of staff in each irrigation areas, the per capita net income of farmers and the social capital in each irrigation area as variables to carry out multiple regression analysis, the results shows that the rlation between social capital and WPI linear was not significant, while the per capita net income of farmers and water pipes scale with the total number of staff in irrigation area have identified significant with WPI linear. What the relation for the 18 removed variables and principal component scores as well as relation between social capital and WPI are still for future research.更多还原