微波与光学遥感协同反演旱区地表土壤水分研究

土壤水分是水文循环中的关键因素,尤其对旱区的生态环境具有十分重要的意义。微波遥感是反演土壤水分的有效手段,而植被是影响土壤水分反演精度的重要因素。因此,对土壤水分的反演需要考虑植被的影响。本文以内蒙古乌审旗为研究区,利用Radarsat-2雷达数据与TM光学数据,对旱区稀疏植被覆盖地表土壤水分反演进行研究。利用TM数据,分别选取NDVI和NDWI指数对植被含水量进行反演,通过水云模型消除植被层对土壤后向散射系数的影响;在此基础上,根据研究区地表植被特性,提出一种基于AIEM 模型的反演土壤水分的改进算法,反演了不同粗糙度参数、不同极化（VV极化和HH极化）条件下的研究区土壤水分。反演结果与野外实测数据的对比结果表明,本文提出的基于地表植被特性的土壤水分改进算法,具有更好的适应性;土壤水分反演模式\begin{array}{c}M{v}_{\mathit{\sigma vv}\left(1\right)}^{\mathit{lh}}\end{array}（VV极化方式下采用NDVI去除植被影响的反演模式）更适合于旱区考虑稀疏植被覆盖影响的地表土壤水分的反演。

Inversion of Soil Moisture in Arid Area Based on Microwave and Optical Remote Sensing Data

Soil moisture is a key factor in the dynamics of hydrological cycle, and it also plays an important role in the ecological environment, especially in terms of an arid area. Microwave remote sensing technology is an effective technique that has been used to extract the soil moisture. However, the impact that vegetation imposes on the process and result of soil moisture retrieval is so influential that cannot be ignored. Therefore, it is necessary to establish a soil moisture retrieval model for the arid area with vegetation being taken into consideration. By taking the Uxin Banner of Inner Mongolia as a case study, the main objective of this paper is to develop different soil moisture retrieval models that are suitable for surfaces that are covered by sparse vegetation in the arid area based on Radarsat-2 and TM remote sensing data. The NDVI and NDWI indices extracted from the TM data have been used to calculate the vegetation water content. And subsequently, the impact of vegetation to the backscattering is removed by applying the water-cloud model. Furthermore, according to the characteristics of the vegetation in the study area, an improved algorithm based on the AIEM model is proposed to retrieve the soil moisture under different surface roughness parameters and polarization modes (VV and HH). Different from the existing algorithms which only utilize the backscattering coefficient within a single treatment for the soil moisture inversion, the improved algorithm comprehensively utilizes the backscattering coefficient before and after applying the correction which is conducted by the water-cloud model. After comparing the inversion results with the field in situ data, the results show that the improved soil moisture inversion algorithm based on the vegetation characteristics has presented a better adaptability. The soil moisture inversion model \begin{array}{c}M{v}_{\mathit{\sigma vv}\left(1\right)}^{\mathit{lh}}\end{array} (this inversion model removes the vegetation influence by using NDVI under the VV polarization mode) is more suitable for the soil moisture inversion in arid area while considering the influence of sparse vegetation.