Retrieving Soil Moisture Using Spectral Mixture Analysis of Landsat8 and the Moderate Resolution Imaging Spectroradiometer北大核心CSCD

【Objective】Soil moisture plays a critical role in many hydrological and ecological processes and understanding its spatiotemporal changes is imperative to ecosystem management and irrigation design. This paper presents and tests a method to estimate soil moisture distribution using satellite technologies.【Method】We took Guangli Irrigation District at Jiaozuo in Henan Province as our study area. The Landsat-8 and MODIS imageries from the area were used to retrieve soil moisture with the apparent thermal inertia and vegetation water supply index, respectively. Using the abundance of vegetation and soil moisture decomposed by the spectral mixture analysis as weighting factors, we retrieved the soil moisture distribution in the area using the two methods.【Result】The correlation coefficient between the measured soil moisture and the retrieved using the vegetation water supply index method and the apparent thermal inertia method was 0.47 and 0.51 respectively. We found that combining the two methods could improve the correlation coefficient between the measured and retrieved soil moisture to 0.73. 【Conclusion】Our results showed that the proposed method is reliable and can adequately estimate soil moisture in irrigation district grown with a diverse of crops. © 2019 Authors. All rights reserved.