【作者】 胡喜生；

【导师】 吴承祯； 洪伟；

【作者基本信息】 福建农林大学 ， 生态学， 2012， 博士

【摘要】 土地生态系统服务功能评价是目前国际上生态学和经济学界研究的热点。国外对土地生态系统服务功能的评价，绝大部分都是沿用Costanza等关于全球生态系统服务功能价值的估算方法，国内主要还是采用谢高地等的中国陆地生态系统单位面积服务价值表对生态系统服务功能价值进行评价。我国正处于经济转型的特殊发展阶段，生态环境与社会经济发展的矛盾日益凸现。由于受经济发展水平和社会文明程度等条件的限制，土地生态系统服务功能评价的研究相对滞后，主要存在如下问题：对土地生态系统服务价值的评估处于静态评估阶段，忽略了不同历史时期和不同生态资源稀缺程度下人们对生态资产的支付意愿不同；评估模型未考虑建设用地的服务价值，一般都假设为零；主要分析某个时间段某个区域或某种类型生态系统服务功能的价值总量，而对空间分异特征的研究较薄弱；对土地生态系统服务功能价值变化内在机制，尤其是在人类活动（如城镇化过程）对土地生态系统服务功能的影响机制的研究鲜见报道。鉴于此，本文在借鉴国际、国内相关研究成果基础上，以海峡西岸经济区中心城市—福州市为研究对象，借助福州市历年土地利用变更数据以及1985年和2007年两期SPOT5遥感影像图等数据源，运用改进的土地生态系统服务功能价值评估模型对福州市的土地生态系统服务功能价值进行核算，在此基础上，分析了福州市土地生态系统服务功能价值的时空变异特征规律，探讨了土地生态系统与城市化系统之间的耦合关系。主要研究内容和结果如下：（1）根据Costanza等人以及谢高地等人的研究成果，结合福州市的实际情况，并运用防治和替代成本法估算城市建设用地对生态系统服务功能的影响价值，修正土地生态系统服务静态总价值。在此基础上，对研究区的生态系统服务功能价值变化情况进行对比分析。结果表明：1985年、1996年、2007年土地利用程度综合指数逐渐增大，分别为2.14、2.19和2.21，而土地生态系统服务价值却逐渐降低，1985年为8748708.78万元，1996年为7456746.33万元，2007年为7438150.86万元。在各种土地生态系统类型中，历年林地生态系统产生的服务价值均最大，约占总价值的71.21%~84.04%，且比例呈逐年增加的趋势。在各项服务功能中，土壤形成与保护、气体调节和水源涵养产生的价值最大，三者合计约占总价值的46%。（2）在总结了土地生态系统服务功能价值估算方法的基础上，对估算方法的修正进展进行了较系统的分析，指出评估模型必须考虑空间异质性、社会发展阶段和资源稀缺性等因素。结合福州市“山多地少”的实际情况，以耕地和林地两者的生物量来修正区域差异对土地生态系统服务价值的影响，构建了空间异质系数，以恩格尔系数和城镇化率共同构建了社会发展系数，以人口密度构建了资源稀缺系数，从而构建了土地生态系统服务功能价值动态估算模型。基于改进的动态估算模型，以福州市为例，估算出2009年土地生态系统服务功能的静态价值为389.48×108元，相当于当年GDP的14.77%；土地生态系统服务功能的动态价值为270.32×108元，相当于当年GDP的10.25%；福州市单位面积土地生态系统服务功能动态价值的空间分布为：市区＞福清市＞闽侯县＞平潭县＞长乐市＞连江县＞闽清县＞永泰县=罗源县；而人均土地生态系统服务功能动态价值的空间分布为：永泰县＞闽清县=闽侯县＞罗源县＞福清市＞连江县＞市区＞平潭县＞长乐市。改进后的动态估算模型计算结果更符合福州市各县（市）区的实际，可为区域土地生态系统保护和经济发展综合决策提供更加及时、准确的有益信息，对区域的生态环境建设和保护具有重要的现实意义。（3）应用目前评价非使用价值较成熟的条件价值法(CVM)，首先对该方法可能存在的偏差进行分析并提出处理措施，在此基础上，以福州市中心城区公园绿地和内河生态系统为研究对象,采用实地调研的方式，运用支付意愿问卷调查的形式，结合数据相关分析和回归分析的方法，分析了公园绿地和内河生态系统服务功能的支付意愿及其影响因素。共发放支付卡式CVM问卷1000份，回收有效问卷836份，获得支付意愿的分布形态和规律。结果表明：福州市中心城区公园绿地和内河生态系统的服务功能总价值为21431.39×10~4元·a-1，占中心城区2009年生态系统服务功能总价值515097.98×104元·a-1的4.2%，其值不可忽视；通过多元统计分析得到在公园绿地和内河生态系统恢复计划调研中，受访者的支付意愿主要受到收入水平、职业、到公园（内河）距离等因素的影响。进一步对“问卷内容依赖性”的问题进行考察，结果表明“部分—整体效应”和“嵌入性效应”在本研究案例中存在，而“顺序效应”不存在，同时在本研究中存在“均等效应”新现象。（4）运用分室理论对未来30a福州市在不同干扰程度下土地生态系统服务功能价值变化进行了模拟分析。结果表明：在1985~2007年期间，福州市土地利用景观结构发生了较大的变化，其中发生变化的土地利用景观面积为46578.45hm~2，占福州市土地利用景观总面积的4%；不同干扰程度下土地生态系统服务功能价值变化模拟结果表明了福州市土地生态系统服务功能价值随着干扰强度的增大而降低，从而提出了城市化快速进程下土地利用对策。（5）根据1993、2004、2009三个时期福州市各种土地利用类型的面积以及前文所提供的单位面积生态系统服务价值，分别计算了福州市9个县市（区）的生态系统服务价值。结果表明：①福州市各县市（区）的生态系统服务价值存在较大的差异，其中永泰县、闽候县和福清市的生态系统服务价值位居前三，而市区、长乐市和平潭县的生态系统服务价值位居倒三；从单位面积服务价值来看，平潭县、连江县和福清市等沿海区域较高，而永泰县、闽清县和罗源县等山区城市较低。②根据福州市各县市（区）用地利用结构将福州市生态功能划分为3个区：人工调控生态区、半自然半人工恢复生态区和自然调控生态区。（6）根据“力矩平衡点”法计算得到福州市土地生态系统服务功能价值分布重心，结果表明：福州市不同类型土地生态系统服务功能价值的重心均分布在福州市的行政中心和地理中心附近。总价值重心从1993年至2004年有由东向西、由南向北移动的趋势；从2004年至2009年有由西向东的趋势，而在南北方向变化不大。（7）由于以土地利用面积指标计算结构信息熵忽略了不同土地利用类型功能（生态功能）的差异，可能导致结果的偏差。本文以土地生态系统服务功能价值指标替代面积指标计算土地利用结构信息熵，结果表明：福州市各县市（区）土地生态系统服务功能结构信息熵由大到小依次为：长乐市＞福清市＞连江县＞平潭县＞闽候县＞市区＞罗源县＞闽清县＞永泰县；福州市各土地利用类型生态系统服务空间结构信息熵值由大到小依次为未利用土地＞草地＞耕地＞建设用地＞林地＞水域＞园地；通过Pearson相关分析表明林地、水域、草地和建设用地等土地利用类型是影响福州市土地生态系统服务功能结构信息熵的主要因素。采用逐步多元回归分析法进行城镇化对功能结构信息熵的影响因素分析表明：人口密度、城镇居民人均可支配收入和第三产业占GDP百分比是最显著的3个影响因子。（8）基于改进的土地空间分布洛伦兹曲线，绘制福州市不同时期的各类型土地生态服务功能空间结构、不同利用强度土地生态系统服务功能空间和生态经济协调等的洛伦兹曲线，结果表明：①水域和建设用地生态系统服务功能空间分布较不均衡，而其它类型土地生态系统服务功能空间分布均处于相对合理及以上状态；各类型土地生态服务功能的人均指标空间分布洛伦兹曲线曲率均较高，只有均林地和园地的人均生态服务功能空间分布处于“警戒线”以内，而耕地、草地、未利用土地、建设用地和水域的人均生态服务功能价值空间分布均处于“差距偏大”或以上状态。②不同利用强度土地生态系统服务功能空间分布洛伦兹曲线及基尼系数结果表明：土地利用强度2的生态服务功能空间分布洛伦兹曲线均几乎接近绝对均匀线，空间分布均衡且稳定；土地利用强度3的生态服务功能空间分布处于“绝对平均”状态；土地利用强度1的空间分布处于“比较平均”状态；而土地利用强度4的生态服务功能空间分布在“警戒线”左右徘徊。③以福州市2009年各县（市）区GDP占全市GDP的累计百分比为横轴，以2009年各县（市）区土地生态系统服务功能价值占全市土地生态系统服务功能总价值的累计百分比为纵轴，绘制洛伦兹曲线。结果表明洛伦兹曲线曲率较大，远离绝对平均线，基尼系数为0.613，集中系数为1.480，生态经济对比处于高度不协调状态。（9）运用STIRPAT模型研究福州市土地生态系统服务功能价值变化的驱动力因素。结果表明：STIRPAT模型能较好地拟合土地生态系统服务功能价值与社会经济发展指标间的关系，反映人口、富裕度、产业结构和城市化率等对服务功能价值变化的影响；影响土地生态系统服务功能价值（尤其对人均服务功能价值）的最主要因素是人口的增长。（10）利用回归分析探讨了土地生态系统服务价值与人口、经济发展、社会发展之间的关系，结果表明：福州市生态系统服务价值与人口、城镇化率、GDP以及人均GDP之间存在着较好的线性正相关关系，相关系数R均大于0.580，福州市生态系统服务价值与城镇化率的线性相关度＞与GDP的相关度＞与人均GDP的相关度＞与人口的相关度；而福州市人均生态系统服务价值与人口、城镇化率、GDP以及人均GDP之间存在着较好的线性负相关关系，相关系数R均大于0.879，且福州市人均生态系统服务价值与人口的线性相关度＞与GDP的相关度＞与人均GDP的相关度＞与城镇化率的相关度。（11）将研究期分为2000-2009年、2000-2004年和2005-2009年3个时间段，分别计算3个时间段的生态经济协调度，结果表明：当以单位面积作为计算指标衡量时，2000-2009年、2000-2004年和2005-2009年3个时间段的生态经济协调度EEH值分别为0.006、-0.0003和0.014，表明研究区处于生态经济轻度协调状态，其中前期处于轻度冲突状态，而后期处于轻度协调状态；当以人均作为计算指标衡量时，2000-2009年、2000-2004年和2005-2009年3个时间段的生态经济协调度EEH值分别为-0.104、-0.057和-0.155，表明研究区处于生态经济轻度冲突状态，且后期比前期冲突更加严重。连江县和平潭县的生态经济处于中度协调水平；市区的生态经济处于轻度冲突水平；其他县市（区）的生态经济均处在轻度协调的边缘。（12）采用6种序参量上下限确定方案分别计算研究区的耦合度和耦合协调度值，并以其平均值代表研究区的耦合度和耦合协调度最终值。结果表明：福州市历年的耦合度数值均较大，介于0.65~1之间，只有2000年处于磨合阶段，而后进入高水平的耦合阶段；耦合协调度和动态耦合模型的计算结果与耦合度计算结果基本一致。更多还原

【Abstract】 For about30years the valuation of ecosystem services has been the focus of ecologicaleconomics and environmental economics, and China began the study since later1990s. In foreignresearches of land ecosystem services estimation，mostly follow the method of the Costanza’s;While in domestic studies, mostly follow the method of Xie Gao-de’s land unit area of theecosystem service value. China is in the special stage of the economic transformation, the conflictbetween ecological environment and socio-economic development is becoming increasinglyapparent. Due to the level of economic development and social civilization degree, the assessmentof land ecosystem services has been lagged behind. There are several problems exiting in this field.Firstly, assessment of land ecosystem service value is in static evaluation stage, ignoring thewillingness to pay of the different historical periods and different ecological resources scarcities andecological assets. Secondly, assessment model does not consider the services value of constructionland, which are generally assumed to be zero. Thirdly, most of researches analyzed of the totalecosystem service in a special region during a special period, while the spatial variationcharacteristics studies of ecosystem service is relatively weak. Fourth, internal mechanism of thechanges in the value of the land ecosystem services, especially the impact of human activities (suchas the process of urbanization) on the land ecosystem services mechanism is rarely reported. Inview of this, on the basis of international and domestic researches, taking the central city of theEconomic Zone on the Western Coast-Fuzhou as the research object, it use calendar year land usechange data as well as the1985and2007two SPOT5remote sensing image map of Fuzhou city asdata sources, then several parts of researches have been carried out in this thesis, main contents andresults are as follows:（1）According to global ecosystem and Chinese terrestrial ecosystem services valueevaluation results obtained by Costanza et al. and Xie Gao-di et al., and the state of the study area,this paper used substitution cost methods and prevented cost methods to estimate the influencingvalues of constructive lands on ecosystem services to analyze the change of ecosystem servicesvalue in Fuzhou. The results showed that the degree of land use increased, while the ecosystemservice values decreased, which was87,487.0878million RMB in1985，74,567.4633million RMBin1996,74,381.5086million RMB in2007. Woodland contributed over70％of the total servicevalue among the land use types．The value of soil formation and disposition, air regulation and airwater conservation occupied about46%of the total service value.（2）The evaluation of land ecosystem service is the hotspot of ecological economics andenvironmental economics. On the basis of reviewing the evaluation methods of ecosystem servicefunction valuation, analyzing the research progress of the evaluation methods correction, it putforward that the factors such as spatial heterogeneity, social development and resource scarcityshould be taken into account in the ecosystem service evaluation model. Based on the former studies and the characteristics of the research area, it took both cultivated and forest lands’ biomassto build coefficient of spatial heterogeneity, both Engel’s coefficient and urbanization to buildcoefficient of social development, population density to build coefficient of resource scarcity. Thenthe improved dynamic evaluation models were make up from the three coefficients. It took FuzhouCity as the study area. The effects showed that the annual static value of the land ecosystemservices in the city was estimated as38.948billion yuan, equal to14.77%of the local GDP in2009.The annual dynamic value was27.032billion yuan. The results showed that the output of theapproved model was more in accordance with the actual situation of all counties. Which couldprovide timely, exact and helpful information in decision making of land ecosystem conservationand social development, and had practical significance for land ecosystem construction andenvironment protection.（3）The relatively mature method-Contingent Valuation Method (CVM) was used to beassociated with a questionnaire survey to evaluate the valuation of the ecosystem services of publicgreen spaces and rivers in center area of Fuzhou city. Using field research methods and thequestionnaire survey with the form of willingness to pay (WTP), we evaluated the influencingfactors to pay for the public green spaces and rivers ecosystem services combined with correlationanalysis and regression analysis. As a case study,1000payment card CVM questionnaires weresurveyed and836as useful quesionnaires were feed back. Distribution shape and rule of willingnessto pay were probed. The results showed that total value of the public green spaces and riversecosystem services was214.31million yuan, accounting for4.2%of the total value of the centralcity. T hrough multivariate statistical analysis in the case of public green spaces and riversecosystem recovery, some key factors were found to be quite important to WTP, such as the incomelevel, occupation and the distance between home and study area. Then the content dependency wereresearched in this thesis. The results showed that the “part-whole effect” and the”embedding effect" in this case study existed, but”sequencing effects "did not. There was a new effect as well, thatwas “equal effect”.（4）The compartment theory was applied to simulate land ecosystem service value dynamicsunder different disturbance scenarios in the next three decades. The results showed that landscapestructure has undergone large changes, among which the changing land area was4.66million hm2from1985to2007, accounting for4%of the total area of Fuzhou. In different degree ofinterference, land ecosystem services changes were simulated using compartment theory. Theresults showed that the land ecosystem services value decreased with the increase of disturbanceintensity. On the basis of which, it proposed some land use strategies during the process of rapidurbanization.（5）According to each types of land use area of1993,2004and2009in Fuzhou city, andusing the unit ecosystem services value supplied by the former study, it calculated ecosystemservice value of the nine counties of Fuzhou. The results showed that:①There were large differences of ecosystem service value existing in the nine counties, in which the ecosystem servicevalue in the top three were Yongtai County, Mminhou County and Fuqing City, while, themunicipal districts, Changle City and Pingtan County were ranked down; from the service value ofper unit area, the coastal areas of Pingtan County, Lianjiang County and Fuqing City were higher,while the mountain area of Yongtai County, Minqing County and Luoyuan County were lower.②Based on the former study, Fuzhou was divided into three districts: the artificial regulation ofecological zones, semi-natural artificial restoration of ecological zones and natural regulation ofecological zones.（6）Using torque equilibrium point method to calculate the distribution center of Landecosystem services value. The results showed that the center of gravity of the different types of landecosystem services values were located near the administrative center and the geographic center ofFuzhou City. The moving trends of the gravity center of total value were from east to west,andsouth to north during1993to2004; while the trends were from west to east during2004to2009,and little change in the north-south direction.（7）Shannon entropy of land ecosystem services structure in each counties sequenced:Changle City, Fuqing City, Liangjiang County, Pingtan County, Minhou County, the City districts,Luoyuan County, Minqing County and Yongtai County. Shannon entropy of each types of landecosystem services structure sequenced: unused land, grass land, arable land, construction land,forest land, water land, and garden land. Pearson correlation analysis showed that the forest lands,waters land, grass land and construction land and other types were the main factors that affectingthe structure of information entropy of land ecosystem services. Using stepwise multiple regressionto analyze the impact of urbanization on information entropy of functional structure, it showed thatpopulation density, per capita disposable income of urban residents and the proportion of tertiaryindustry to GDP were the most significant three impact factors.（8）The Lorenz curve and Gini coefficient of various types of land ecosystem services indifferent periods in Fuzhou were studied in this paper. The results showed that:①the ecosystem services spatial distribution of waters land and construction land were notrelatively balanced in each counties, while the spatial distribution of other types of land were in thereasonable state. The curvature of the various types land Lorenz curves were high in differentperiods In fuzhou, when taking per capita land ecosystem services into account. Wood land andgarden land were the only two types of land which’s spatial distribution of per capita ecologicalservice functions were within the cordon, while the spatial distribution of arable land, grass land,unused land, construction land and water land per capita value of ecosystem services were all in thestate of “too large gap” or above.②The Lorenz curve and Gini coefficient of ecosystem services of different land use intensityin different periods in Fuzhou were studied in this paper. The results showed that the ecosystemservices Lorenz curve of land use intensity2was almost close to the absolutely uniform line, which’s spatial distribution was balanced and stable; the ecosystem services spatial distribution ofthe land use intensity3was in the state of "absolute average"; the land use intensity1was in theaverage state; while the land use intensity4was wandering around in the "cordon".③Using the cumulative percentage of the counties GDP against the Fuzhou city’s GDP in2009as the horizontal axis, and the cumulative percentage of the counties land ecosystem servicesvalue against the Fuzhou city’s in2009as the vertical axis, it drew the Lorenz curve. The resultsshowed that the curvature of the Lorentz curve was far away from the absolute average line, and theGini coefficient was0.613, the concentration factor was1.480, which meant ecological system andeconomic system were in a highly uncoordinated state.（9）Using STIRPAT model to study driving force factors of the Land ecosystem servicesvalue. The results showed that the STIRPAT model was fitted well to fit the relationship betweenthe land ecosystem services value and socio-economic indicators, which reflected that thepopulation, affluent degree, industrial structure and the urbanization rate would affect land ecologyservice value. To the land ecology service value (per capita service value in particular), the mostimportant factor was population growth.（10）The relationships between land ecosystem service value and population, economic,social were explored here using regression analysis. The results showed that land ecosystem servicevalue has good relationship with population, urbanization rate, GDP and per capita GDP, and all thecoefficient R were greater than0.580. The most relevant fator to land ecosystem service value wasurbanization rate, the second was GDP, the third was per capita GDP, and the fourth was population.While per capita ecosystem services value has linear negative correlation with population,urbanization rate, GDP and per capita GDP, all the coefficient R were greater than0.879. The mostrelevant fator to per capita ecosystem services value was population, the second was GDP, the thirdwas per capita GDP, and the fourth was urbanization rate.（11）The study period was divided into three periods,2000-2009,2000-2004and2005-2009,the coordination value of eco-economic were calculated for the three periods. The results showedthat: when the unit area as the calculation indicators to measure, the eco-economic coordinationvalue of the three periods were0.006,-0.0003and0.014, which meant that the ecological andeconomic in the study area was in mild coordination state, among which the early period was in themild state of conflict, but the late period was in the mild coordination state. When the per capita asthe calculation indicators to measure, the eco-economic coordination value of the three periods were-0.104,-0.057and-0.155, which meant that the ecological and economic in the study area was inmild conflict status, and the latter period was more serious than the former period. Among the ninecounties, the ecological and economic of Liangjiang County and Pingtan County were in amoderate level of coordination state; municipal districts was in a mild levels of conflict; while theother counties were in edge of the mild coordination.（12）Six kinds of upper and lower limits to order parameter were adopting in this paper, and then each kinds of coupling degree and coupling coordination degree according to respectiveupper and lower limits were calculated. Using there average represented the coupling degree andcoupling coordination degree of the study area, it showed that the values were high, which rangeswere between0.65and1, among which the relationship between ecology and economic in2000was in the run-in phase, and then it went into a high level of coupling phase. The results of couplingcoordination and dynamic coupling model were in agreement.更多还原