【作者】 董战峰；

【导师】 王金南； 毕军；

【作者基本信息】 南京大学 ， 环境科学， 2010， 博士

【副题名】以COD为例

【摘要】 中国的水污染物总量分配在本质上属于一种污染物管理的目标总量模式。目前,国家到各省市区的污染物总量分配方案制定过程主要是基于各省市区的经济表现,采取国家与各省市区间协商、以及行政主管部门“拍板”的模式来制定,主观因素较强,各省市区对这种分配方法争议也较大。如何设计在方法上科学合理、在实务上能为各省市区普遍接受的分配方案是中国水污染物总量管理面临的巨大挑战。本研究旨在以中国“十二五”时期的COD总量从国家到各省市区分配为案例,研究确定COD总量分配的指标集,构造不涉及主观决策因素的污染物总量分配模型,以及COD排放与人均GDP等经济变量的关联模型,探讨设计的分配计量模型用于污染物总量分配的可行性,以及与分配方案关联的各省市区的经济指标的动态变化特征,从而为我国的水污染物总量控制管理提供理论方法和技术上的支持。首先,本研究分析了COD总量分配的指标集和各省市区在各指标集下的异质性表现。主要包括确定COD总量分配要素、分配指标集、设计分配规则、考察各分配指标的变异特征,以及各地区在分配指标下的异质性表现特征。指出总量分配应采用系统分析思维,以分配对象的异质性为基础,以分配的公平性为前提,紧密围绕COD削减和促进环境质量改进来设计。研究指出影响各省市区COD排放和削减的要素主要包括经济结构和规模要素、人口要素、技术进步要素、污染治理要素、水环境质量要素和水资源禀赋要素,在对这些要素的表征指标进行系统分析的基础上,初步筛选确定了COD总量分配的指标集,进而利用相关性分析方法,最终确定了COD总量分配指标集。该指标集由七项指标构成,分别为：人均GDP、人均污染排放强度、水污染重点行业增加值比重、水环境压力指数、单位国土面积水资源量、工业废水COD处理率、城镇生活污水COD处理率,其中前五项指标为总量分配的正向指标,后两项指标为反向指标。对各分配指标的变异系数测量表明,不同指标的变异程度明显,其中以水环境压力指数、单位国土面积水资源量、人均GDP三个指标的变异程度相对较大,这意味着这些指标在总量分配方案设计时的相对重要性程度较大。对各分配指标的相对差异指数的测量表明,各省市区的指标差异性表现非常明显,说明不同省市区的异质性水平呈离散分布特征,本研究为COD总量分配的定量测度提供了定性基础信息。其次,本研究基于信息熵和Gini系数理论,分别构建了COD总量分配计量模型,两种模型的分配结果差异明显。其中,基于信息熵的分配方法与国家“十二五”时期的水环境总量管理需求一致性较好,而基于Gini系数的分配方法则与其关联性较差。进一步利用“十一五”规划的COD总量分配方案对两种方法进行模拟分析,也得出同样的结论,而且,基于信息熵理论构建的计量模型的分配结果也与前述采用变异系数和相对差异指数的定性分析结论相一致。这说明基于信息熵的分配方法较基于Gini系数的分配方法更适用于国家COD的总量分配。利用该分配方法得到的“十二五”时期COD总量分配方案与“十一五”COD总量分配规划方案进行比对分析的结果表明,“十二五”期间,以东部地区和山西、广西等重点省市区减排为主的格局仍未转变过来,尽管这些地区的削减率在总体下降,以及削减率向中国的西部和南部增调。最后,对中国30个省市区的第二产业比重、人均GDP的变化与COD排放量之间的面板数据分析表明,两个经济变量与以COD排放指标为表征的环境变量间存在长期稳定关系。通过数值拟合法构造各省市区的人均GDP、第二产业比重与COD排放量的计量模型的分析结果表明：总体上,“十一五”至“十二五”时期,东部省市区的第二产业比重呈下降趋势,西部的大部分省市区的第二产业比重则仍在增长；各地的增加幅度差异较大,西部和中部地区以内蒙古和湖北两地增幅最大,而河北、江西、广西等地则相对较小,东部地区以江苏增幅最大；人均GDP增长率变化的东、中、西分布格局并不明显,大部分省市区分布在20%-40%区间,一些省市区的增幅相差较大,湖北、云南两地的人均GDP增加1倍多,而北京、河北、甘肃、青海四省市的增幅则相对较小,低于15%。以单位COD的GDP贡献来表征各省市区的排放效率,对该指标在时间趋势上进行回归拟合,构造排放效率变化与时间的单变量反应函数,发现该函数在时间趋势上符合指数型特征,但应用于“十二五”时期的趋势分析存在较大误差。这也意味着国家COD总量分配方案设计时,应抛弃过去过度重视各省市区GDP表现的做法,应该选取客观反映区域异质性的指标来设计分配方案。更多还原

【Abstract】 Total amount allocation of the water pollutants in China is in essence belongs to a kind of target-set pollutant management models. Currently, total amount allocation scheme design mainly considers the respective economic performance of the provinces, municipalities and districts (namely, provincial regions), and was realized by the negotiation between the MEP and the environmental departments of the provincial regions, more subjective decisive factors are involved in the process of allocation scheme design, this brought tremendous dispute amongst the provincial regions. So, how to design a scientific and reasonable scheme that is easy to be accepted by the provincial regions in fact put forwards a big challenge to the China’s water pollutant total amount management. This work aims to take the COD allocation from the national to the provincial regions as a case, to explore the allocation indicators, the heterogeneous characteristics of the provincial regions, to construct the pollutant total amount allocation models with no subjective decision factors involved and the correlation calculation models between the COD emission parameter and the economic parameters like per capita GDP, to analyze the feasibility of the constructed allocation method, and the dynamic changing characteristics of the economic parameters in correlation to the allocation scheme, and eventually to provide the theoretic and technical supports to the China’s pollutant total amount control management.Firstly, the work analyzes the COD total amount allocation indicators and the heterogeneous characteristics of the provincial regions under the respective allocation indicators. Including determining the key allocation factors and indicators, designing allocation rules, exploring variance extent of each indicator and heterogeneous manifestation of each provincial regions under the seven respective allocation indicators. The work states that the allocation scheme design should be taken a systematic idea and taken heterogeneous characteristics of the provincial regions as the bases, taken the fairness as the prerequisite, should facilitate COD reduction and boost the improvement of the environmental quality. The results showed that factors influencing COD emission and reduction mainly including economic scales and structure, population, technical improvements, pollution control capabilities, water quality and natural resources endowment. With a systematic analysis of the indicators used to characterize the factors, the paper determines the indicator systems initially, by the statistic correlation test with SPSS software, seven indicators were selected to consist of the COD total amount allocation indicator system, which are per capita GDP, per capita COD emission intensity, added value proportion of the water pollution key industries compare to GDP, pressure index of the water environment, water resources quantity of the per unit of land. The former five indicators are positive indicators, the later two are negative indicators. Calculation results of the variance coefficient of the different COD allocation indicators showed that variance extent of each indicator is obvious, and that of the pressure index of the water environment, water resources quantity of the per unit of land, and per capita GDP is relatively higher, which means that the three indicators are relatively important in the process of the COD total amount allocation scheme design. Calculation results of the relative difference index of the different allocation indicators showed that the manifestation of the provincial regions differs greatly, which means that the various provincial regions’ heterogeneity emerges a discrete distribution. In conclusion, the paper states that the indicators difference and the manifestation difference under different indicators of the various provincial regions should be placed more emphasis in the process of the COD total amount allocation scheme design.Secondly, the paper introduced the information entropy and Gini coefficient theory into the pollutant total amount allocation fields, and COD total amount allocation econometric models were designed, the allocation results with the two models differ obviously. The allocation results based on information entropy keep well consistency with the strategic requirements of the China’s water environment management in the 12th period, but the allocation results based on Gini coefficient theory do not show the consistency. With a further comparative analysis of the two methods using the COD total amount allocation scheme in the 11th five environmental planning, the same conclusion could be obtained. Moreover, the allocation results based on the information entropy keep well consistency with the qualitative analysis results of the variance coefficient and relative difference indexes method, these indicate that the allocation method based on information entropy is more applicable to the China’s COD total amount allocation. Comparative analysis between the COD total amount allocation scheme in the 12th five period based on the information entropy theory and the planned COD total amount allocation scheme in the 11th five period show that overall spatial pollution reduction pattern in the eastern China and the key provincial regions like Shanxi and Guangxi will not change, though the reduction ratios in these regions are keeping decreasing and the reduction ratios in the western and southern China begin to increase.Thirdly, panel data analysis to the secondary industry proportion and per capita GDP with COD emission amount in the 30 provincial regions shows, that long term steady relation between the two economic parameters and COD emission amount exists. With analysis of the constructed equation by the statistic data simulation of the economic parameters and environmental parameters, the paper concludes that totally proportion of the secondary industry will decrease in the eastern provincial regions, but most provincial regions in the western China will increase, the increasing level differs great amongst the provincial regions, that in the Inner Mongolia and Hubei is relatively higher, and that in Hebei, Jiangxi, Guangxi and Henan province etc is relatively lower, that in Jiangsu is the highest in the eastern China. Changing pattern of the Per capita GDP increasing ratio in the eastern, central and western China in not obvious, per capita GDP increasing ratio distribution in the most provincial regions lies in the range of 20%-40%, but that in some provincial regions differs great, as that in Hubei, Yunnan is above 100%, but in Beijing, Hebei,Gansu, Qinghai is lower than 15%. Construction of single variable responsive function with emission efficiency characterized with the GDP contribution of per unit of COD emission and the time parameter through simulating the emission efficiency in time trend of the 30 provincial regions was realized, further analysis of that function could be found it follows a exponential change, but errors happens when using the function to analyze the trend of per unit of COD emission, that means when design the COD total amount control scheme should get rid of the idea of overemphasizing the GDP performance of the provincial regions, and indicators reflecting the regional heterogeneity should be chosed to be used in the process of the COD total amount allocation scheme design.更多还原