基于ATI模型和TVDI模型的晋中土壤水分遥感反演研究

发布时间：2018-03-26 22:34

本文选题：土壤湿度　切入点：反演　出处：《山西农业大学》2015年硕士论文

【摘要】：土壤水是陆地和大气之间能量互换过程中的一个重要因素,是极其重要的环境因子。晋中地区地处黄土高原东缘,水资源较为短缺属于国际公认的极度缺水地区,干旱对农业造成了重要影响。遥感技术监测土壤湿度具有快速、宏观和动态等特点,且已经发展为大区域干旱监测评估和区域土壤水资源评价粮食安全问题解决的主要手段。本研究利用多源遥感影像反演晋中地区2014年5-9月土壤湿度,对快速、适时获取大范围土壤湿度空间分布信息,有效预防和应对干旱有理论意义和现实意义。晋中地区是山西省粮食、蔬菜、水果、畜牧品等的重要产区之一,农业居山西省领先位置生产条件十分优越。文化旅游业发达,是人口与经济密集分布区,气候属暖温带大陆性季风气候,季节变化明显,降雨在时空上分布不均,呈现东部丘陵地区降水多西部平原地带降水少趋势。遥感技术反演区域土壤水分的研究是定量遥感研究中的前沿方向,目前,在区域土壤水遥感监测研究中,基于表观热惯量和归一化植被干旱指数的反演模型应用较为广泛且精度较高,但是单一模型的反演往往会忽略模型的适用范围。当前山西省遥感监测的研究比较少且都是使用单一反演模型,急需补充这一领域研究。本文以山西省晋中市为研究区，以MODIS数据、TM8数据和实测土壤湿度数据为数据源,参照优势互补的设计思想,运用温度植被指数模型(TVDI)和表观热惯量模型(ATI)联合反演晋中5-9月土壤湿度,通过NDVI值分区将NDVI0.37区域采用归一化植被干旱指数(TVDI)方法反演土壤相对湿度,NDVI0.37区域采用表观热惯量模型(ATI)反演土壤湿度,并反演出2014年5-9月晋中每月土壤湿度,最后结合反演图像,对研究区土壤相对湿度的空间和时间分布特征进行分析,其中模型拟合结果均通过0.01显著性差异检验。研究结果表明：1晋中地区的土壤相对湿度在植物生长期间呈现两个由降到升的变化周期,第一个周期在5月到7月中旬,5-6月土壤相对湿度为下降,6-7月中旬土壤相对湿度为上升,7月中旬为土壤相对湿度峰值,6月初为生长期土壤湿度最低值,同样6月初也是降雨均值最低时期。第二个周期是在7月中旬到9月下旬,其中,7月中旬到8月中旬土壤相对湿度为下降,9月土壤相对湿度为上升期,峰值出现在8月初。2模型反演精度检验结果表明反演的土壤水分与实测的土壤水分具有良好的相关性,反演精度较高。
[Abstract]:Soil water is an important factor in the process of energy exchange between land and atmosphere, and is an extremely important environmental factor. Jinzhong region is located in the eastern edge of the Loess Plateau. Drought has an important impact on agriculture. Remote sensing technology is characterized by rapid, macroscopic and dynamic monitoring of soil moisture. And it has been developed as the main means to solve the food security problem of drought monitoring and assessment of regional soil water resources. This study uses multi-source remote sensing images to retrieve soil moisture in Jinzhong region from May to September 2014. It is of theoretical and practical significance to obtain the spatial distribution information of soil moisture on a large scale and to effectively prevent and cope with drought. Jinzhong region is one of the important producing areas of grain, vegetables, fruits and livestock products in Shanxi Province. Agriculture occupies the leading position in Shanxi Province in terms of production conditions. Cultural tourism is developed, population and economy are densely distributed, climate is continental monsoon climate in warm temperate zone, seasonal changes are obvious, and rainfall is unevenly distributed in time and space. The research of retrieving regional soil moisture by remote sensing is the forward direction of quantitative remote sensing research. At present, in the research of remote sensing monitoring of regional soil water, the trend of precipitation in the eastern hilly region is more than that in the western plain. The inversion model based on apparent thermal inertia and normalized vegetation drought index is widely used and has high accuracy. However, the inversion of a single model often ignores the scope of application of the model. At present, the research on remote sensing monitoring in Shanxi Province is less and uses a single inversion model, so it is urgent to supplement the research in this field. In this paper, Jinzhong City, Shanxi Province, is taken as the research area. Based on MODIS data (TM8) and measured soil moisture data, and referring to the design idea of complementary advantages, soil moisture was retrieved from May to September in Jinzhong by using TVDI-based model of temperature vegetation index (TVI) and apparent thermal inertia model (ATI). The NDVI0.37 region was divided into NDVI values and the normalized vegetation drought index (TVDI) method was used to invert the soil relative humidity and the apparent thermal inertia model was used to retrieve the soil moisture in the area of NDVI 0.37, and the monthly soil moisture in Jinzhong period from May to September 2014 was reversed. Finally, the spatial and temporal distribution characteristics of soil relative humidity in the study area are analyzed by using the inversion image. The results of model fitting all passed 0.01 significant difference test. The results showed that the soil relative humidity in Jinzhong area of 1: 1 showed two periods of change from decreasing to rising during plant growth. The first cycle was from May to the middle of July when the relative humidity of the soil decreased from June to the middle of July, and the peak value of the soil relative humidity in the middle of July and the beginning of June was the lowest value of soil moisture in the growing period. The second cycle was from mid-July to late September, in which the relative soil moisture decreased from mid-July to mid-August, and the soil relative humidity increased in September. The peak value appeared in the early August .2 model inversion accuracy test results show that the inversion of soil moisture and measured soil moisture has a good correlation, and the inversion accuracy is high.
【学位授予单位】：山西农业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：S152.7;S127

【参考文献】