内蒙古典型针茅-羊草荒漠草原土壤水分监测与校验研究

Remote sensing method is the main technical means to acquire a wide range of soil moisture content. But for a long time, it has always been questioned because of too much dependence on the point measurement data, the point data’s lack of representative of regional scale and the difficulties to be validated effectively from the scale of pixel. In this study, soil moisture content of pixel scale is to be monitored using cosmic-ray neutron method in the typical steppe desert of central Inner Mongolia. What’s more, the influence of precipitation and vegetation and the spatial distribution characteristics of measuring footprint for cosmic-ray neutron method are to be analyzed further. Meanwhile, a joint retrieval algorithm of microwave and optics is to be established based on deep analysis of their soil moisture retrieval mechanism to acquire wider range of soil moisture content. Mutual validations of soil moisture retrieval results from different scales including point measurement method with FDR, cosmic-ray neutron method and remote sensing method are conducted finally combining the spatial distribution characteristics of measuring footprint for cosmic-ray neutron method. The research results of the project will provide effective means for coupling point and area measurements of soil moisture, as well as building and validating of remote sensing retrieval model, thus improving the monitoring accuracy of soil moisture and developing and enriching the techniques of soil moisture monitoring.

遥感法是解决大范围土壤水分获取的主要技术手段，但长期以来因过于依赖站点观测数据、点上数据缺乏区域尺度代表性且难以开展像元级的有效验证而一直受到质疑。本研究选择内蒙古中部典型荒漠草原为研究区，基于宇宙射线中子法获取匹配像元尺度的土壤水分观测数据，并开展降水与植被对测量结果的影响以及中子足迹空间分布特征等的分析研究；同时在深入分析微波与光学反演土壤水分机理的基础上，发展一种微波与光学联合反演算法，获得更大范围土壤水分监测数据；最终结合宇宙射线中子空间足迹特征，开展FDR点观测法、宇宙射线中子法、遥感法不同尺度土壤水分监测结果的相互校验分析。项目研究成果将为实现土壤水分点、面观测的有机耦合，为遥感法模型构建及结果验证提供有效的手段，提高空间尺度土壤水分监测的精度，从而发展和丰富土壤水分监测的技术方法。

遥感法是解决大范围土壤水分获取的主要技术手段，但长期以来因过于依赖站点观测数据、点上数据缺乏区域尺度代表性且难以开展像元级的有效验证而一直受到质疑。本研究选取内蒙古中部典型荒漠草原上东河小流域为研究区，主要开展研究和成果如下：利用已有宇宙射线中子土壤水分观测系统设备条件，设计了土壤水分多尺度协同观测实验室方案；系统分析空气水汽、降水、植被等因素对监测结果的影响，开展宇宙射线中子法（CRS）适宜性分析；研究了宇宙射线快中子水平与垂直方向足迹空间分布特征，探讨了不同土壤水分尺度转换模型，构建了一种结合遥感数据的宇宙射线中子仪空间特征模拟方法；研发了基于光学、光学和遥感协同的土壤水分机理和机器学习反演模型。进一步对比分析了FDR与CRS监测结果，系统开展不同尺度土壤水分观测结果的耦合验证。项目成果实现了土壤水分点、面观测的有机耦合，积累了区域土壤水分验证序列数据，发展遥感土壤水分反演和验证的新方法，实现区域高空间分辨率土壤水分定量反演，对于定量研究植被-土壤-大气系统水分循环效应、探讨尺度问题等提供重要科学参考。