Search for a command to run...
As an important parameter in recent and numerous environmental studies, soil moisture (SM) influences the exchange of water and energy at the interface between the land surface and atmosphere. Accurate estimate of the spatio-temporal variations of SM is critical for numerous large-scale terrestrial studies. Although microwave remote sensing provides many algorithms to obtain SM at large scale, such as SMOS and SMAP etc., resulting in many data products, they are almost low resolution and not applicable in small catchment or field scale. Estimations of SM from optical and thermal remote sensing have been studied for many years and significant progress has been made. In contrast to previous reviews, this paper presents a new, comprehensive and systematic review of using optical and thermal remote sensing for estimating SM. The physical basis and status of the estimation methods are analyzed and summarized in detail. The most important and latest advances in soil moisture estimation using temporal information have been shown in this paper. SM estimation from optical and thermal remote sensing mainly depends on the relationship between SM and the surface reflectance or vegetation index. The thermal infrared remote sensing methods uses the relationship between SM and the surface temperature or variations of surface temperature/vegetation index. These approaches often have complex derivation processes and many approximations. Therefore, combinations of optical and thermal infrared remotely sensed data can provide more valuable information for SM estimation. Moreover, the advantages and weaknesses of different approaches are compared and applicable conditions as well as key issues in current soil moisture estimation algorithms are discussed. Finally, key problems and suggested solutions are proposed for future research.