基于多源数据的喀斯特山区石漠化评价地理单元划分研究

发布时间：2018-09-05 15:12

【摘要】：石漠化不仅影响人类的生存环境,还制约地区的经济发展。西南地区作为石漠化的重灾区,多年来,有关专家和学者进行了大量的研究,在石漠化评价和综合治理等方面取得了大量的研究成果。在石漠化评价单元方面,目前的研究中仅限于行政区和格网,为更好的进行石漠化评价,本文以西南地区石漠化比较严重的关岭县为研究区,以石漠化评价地理单元划分为主线,运用主流的“3S”技术,层次分析法、对比分析法、面向对象分类等方法,开展面向对象的喀斯特山区高精度土地利用数据提取;发展基于多因子耦合关系的空间聚类分析,综合分析土地利用数据、岩性数据、降雨数据和坡度数据之间的关系,划分石漠化评价地理单元。主要研究内容和结论如下:(1)基于高分辨率遥感影像,运用面向对象分类方法,提取了研究区土地利用现状数据,通过野外验证分析,提取精度达到86.69%。(2)采用层次分析法,对每一个相关单元构建相应的层次结构模型,通过判断矩阵分析两两因子之间的关系,对模型的可行性进行了判断,赋予每个石漠化影响因子相应的权重,基于多因子耦合关系进行了空间聚类分析。(3)研究区石漠化评价地理单元主要划分为相关单元与不相关单元,相关单元基于土地利用和评分范围进一步划分。基于土地利用将研究区划分为林地单元区划、草地单元区划、旱地单元区划、园地单元区划和裸土地单元区划;基于石漠化评价单元指数划分为一到五个级别。(4)通过对研究区进行石漠化评价,评价结果显示:关岭县石漠化程度主要以轻度石漠化和中度石漠化为主,分布范围广,集中分布在关岭县西部。
[Abstract]:Rocky desertification not only affects the living environment of human beings, but also restricts the economic development of the region. As a serious disaster area of rocky desertification in southwest China, experts and scholars concerned have carried out a lot of research for many years, and have made a great deal of research achievements in evaluation and comprehensive control of rocky desertification. In the evaluation unit of rocky desertification, the current research is limited to administrative regions and grids. In order to better evaluate the rocky desertification, this paper takes Guanling County, which is a relatively serious rocky desertification area in southwest China, as the research area. Taking the division of geographic units for evaluation of rocky desertification as the main line, using the main techniques of "3s", AHP, comparative analysis, object-oriented classification, etc., the paper carries out the object-oriented high-precision land use data extraction in karst mountain areas. The spatial cluster analysis based on multi-factor coupling relationship is developed to analyze the relationship among land use data, lithology data, rainfall data and slope data, and to divide the geographic unit of evaluation of rocky desertification. The main research contents and conclusions are as follows: (1) based on high resolution remote sensing image, the data of land use status in the study area are extracted by using object oriented classification method. Through field verification and analysis, the extraction accuracy reaches 86.6949. (2) Analytic hierarchy process (AHP) is used. The corresponding hierarchical structure model is constructed for each related unit, and the feasibility of the model is judged by analyzing the relationship between two and two factors through the judgment matrix. Spatial clustering analysis based on multi-factor coupling relationship is carried out. (3) the geographic units of rocky desertification evaluation in the study area are mainly divided into related units and uncorrelated units, and the related units are further divided based on land use and scoring range. Based on land use, the study area is divided into forest land unit division, grassland unit division, dryland unit division, garden unit division and bare land unit division. Based on the evaluation unit index of rocky desertification, it is divided into one to five grades. (4) the evaluation results show that the degree of rocky desertification in Guanling County is mainly light and moderate rocky desertification, and its distribution is wide. Concentrated in the west of Guanling County.
【学位授予单位】：贵州师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X171

【参考文献】