【作者】 吴静；

【导师】 张德罡；

【作者基本信息】 甘肃农业大学 ， 草业科学， 2013， 博士

【摘要】 草地分类是草地科学研究的基础理论问题。目前草地分类系统多属经验性分类，难以客观、准确地反映草地类型分异的实质。草原综合顺序分类系统(CSCS)在世界草地分类系统中处于领先地位，其重要贡献是在基本单位“类”一级实现了定量分类。但是，与其理论体系相比，CSCS具体的划分手段还远未匹配，而且由于数据限制，出现以点代面，以相邻生境代实际生境的问题,对CSCS科学严谨性和实际推广应用提出了挑战，减弱了其实用性。本研究提出在CSCS“类”的划分中，用面状的、由定量遥感反演的土壤水热数据替代点状的、地面站点的大气水热数据作为分类指标的解决思路。在研究草地类型与土壤水热条件规律性联系的基础上，提出利用土壤水热条件划分草地类型的理论依据，拟定分类方案。旨在为增强草地分类指标的确限性、提高草地分类定量化研究水平提供新思路，推进CSCS的实用化进程。在草地发生学原理指导下，在草原综合顺序分类中引入甘肃省2008年每日1km分辨率的MODIS地表温度产品(MYD11A1)和0.5km分辨率的MODIS地表反照率产品(MYD09GA)，反演土壤水分和地表年积温，划分草地的热量级和湿润度级，并对甘肃省草地进行分类，以野外调查数据为相对真值验证了结果，评价了分类精度。主要研究结果如下：(1)甘肃省天然草地横跨寒冷-寒温-微温-暖温-暖热五个热量级，极干-干旱-微干-微润-湿润-潮湿六个湿润度级，共26个草地类，其中暖温干旱暖温带半荒漠类、微温干旱温带半荒漠类和寒温潮湿温性针叶林类是甘肃省最主要的几种草地类型，占全省面积的40%以上；草地类的分布呈现出明显的垂直地带性，划分结果基本符合甘肃省干旱、半干旱的区域气候特征、北半球中纬度的区域位置特征以及山地丘陵为主的地貌特征；(2)研究基于草地发生学原理，首次在草原综合顺序分类法中利用遥感数据，引入MODIS每日温度和每日地表反射率数据，反演地表温度和土壤水分参数，划分草地类型,结果可覆盖广大无人区，在草地分类中有更大的潜力；(3)研究在一定程度上减少了以往草原综合顺序分类对气象站点分布和插值方法的依赖性，解决了综合顺序分类法中站点数据向区域数据转换这一难题，改善了点数据外推的边界模糊问题，为草原综合顺序分类提供了一种新方法；(4)在草原综合顺序分类基本单位“类”的划分中，建立了适用于3S技术的水热指标，实现了两个替代：用土壤水热条件替代大气水热条件，用面状数据替代点状数据；(5)以2005年和2006年甘肃省草地实地监测数据为相对真值，首次在草原综合顺序分类法中引入混淆矩阵对草地分类结果进行了精度分析，总体分类精度为70.11%，Kappa系数为0.67。(6)以交互式数据语言（IDL）为基础，设计与开发了土壤水分遥感反演平台（SMIP），处理了2100余景遥感影像，实现了海量遥感数据的投影转换、拼接、裁剪、数据补偿、地表参数反演和验证工作，将以往需要以年为单位进行的数据处理工作缩短到了一个月。更多还原

【Abstract】 Grassland classification is a fundamental need of grassland science. Meanwhile it is alsoa challenge to develop a comprehensive grassland classification system because of themultivariable and multi-functional features of grassland ecosystem.The Comprehensive and Sequential Classification System of Grassland (CSCS), one ofwell known grassland classification systems in China and even over the world, involves ahierarchy of three classification levels(class-subclass-type, class is the basic level) and isadvanced in quantification indicators.However, there are at least two aspects need to be improved at the basic classificationlevel of CSCS:1) the grasslands are grouped into classes according to the data involvingannual precipitation and accumulative temperature, which are collected from meteorologicalstations.These data reflect the near-surface atmosphere hydrothermal conditions instead of theactual habitat of grasses;2) the data of precipitation and temperature from ground observationcan only present the conditions within a small area, but they are used through extrapolation toa larger region.In order to resolve the problems, the areal data of land surface temperature and soilmoisture are introduced by quantitative remote sensing as main data sources for the basicclassification level of CSCS to replace the parameters of precipitation and atmospheretemperature from ground observation. In this paper, the MODIS land surface temperatureproduct (MYD11A1, daily with1km resolution) and MODIS land surface reflection product(MYD09GA, daily with0.5km resolution) of Gansu Province in2008were used to invertsoil moisture based on Thermal Inertia Model with the help of a Soil Moisture InversionPlatform (SMIP) developped from ENVI/IDL. Then, the annual accumulative land surfacetemperature (＞0℃Σθ) and annual sum of soil moisture were carried out, moreover, fit withannual accumulative temperature (＞0℃Σθ′) and precipitation data from meteorologicalstations respectively. Thermal zones were determined by temperature and humidity zones byK-value (moisture index), grassland class was obtained by coupling the thermal zones andhumidity zones. Finally, the grassland types were verified through the field investigation andaccuracy assessment was tested with confusion matrix.The main results are as follows: 1) the grassland in Gansu occupies five thermal zones (Frigid-Cold temperate-Cooltemperate-Warm temperate-Warm-Subtropical) and six humidity zones (Extrarid-Arid-Semiarid-Subhumid-Perhumid);2) there are26possible types present in Gansu Province, and three grassland classes thatcover the largest area in Gansu are Warm temperate-arid warm temperate zonal semidesert(ⅣB11), Cool temperate-arid temperate zonal semidesert (ⅢB10) and Cold temperateperhumid taiga forest (ⅡF37), with the total area of these three is17.83million ha,accounting for44.43%of the total grassland area in Gansu;3) the geographical distribution of grassland type indicates significantly vertical zonalitypattern: with the altitude decreasing, frigid series grassland, cold temperate series grassland,cool temperate series grassland, warm temperate series grassland and warm series grasslanddistribute successively from southwest to northeast, which fit the terrain of Gansu Province;4) accuracy assessment shows: the overall accuracy of grassland classification is70.11%and the kappa coefficient is0.67. The research solved the problem of transforming fromscattered site data to regional polygon data in CSCS and the problem of uncertain borderlinein punctate data extrapolation, and provide a new approach to the utilization of CSCS, whichcould carry forward the practical progress of CSCS.更多还原