Spatial Evaluation of Soil Moisture (SM), Land Surface Temperature (LST), and LST-Derived SM Indexes Dynamics during SMAPVEX12

Abstract

Downscaling microwave soil moisture (SM) with optical/thermal remote sensing data has considerable application potential. Spatial correlations between SM and land surface temperature (LST) or LST-derived SM indexes (SMIs) are vital to the current optical/thermal and microwave fusion downscaling methods. In this study, the spatial correlations were evaluated at the same spatial scale using SMAPVEX12 SM data and MODIS day/night LST products. LST-derived SMIs was calculated using NLDAS-2 gridded meteorological data with conventional trapezoid and two-stage trapezoid models. Results indicated that (1) SM agrees better with daytime LST than the nighttime or the day-night differential LST; (2) the daytime LSTs on Aqua and Terra present very similar spatial agreement with SM and they have very similar performances as downscaling factors in simulating SM; (3) decoupling effect among SM, LST, and LST-derived SMIs occurs not only in very wet but also in very dry condition; and (4) the decoupling effect degrades the performance of LST as a downscaling factor. The future downscaling algorithms should consider net surface radiation and soil type to tackle the decoupling effect.

Monitoring LST-NDVI Relationship Using Premonsoon Landsat Datasets

“…NDVI acts as a determining factor of LST [30], and some studies used the LST-NDVI correlation to evaluate the distributional pattern of LST [31][32][33][34][35][36]. A lot of recent studies assess the LST-NDVI correlation in multidimensional approach [2,32,[37][38][39][40][41][42][43][44][45][46][47][48].…”

Enhancing Spatial Resolution of SMAP Soil Moisture Products through Spatial Downscaling over a Large Watershed: A Case Study for the Susquehanna River Basin in the Northeastern United States

“…Thermal inertia describes the resistance of soil to temperature change and it has positive relation with SM, which implies that wetter soil has higher thermal inertia which in turn leads to decrease in diurnal LST range [63], whereas, in the context of ET, the relationship between LST and SM varies depending on ET regimes: water-or energy-limited. In water-limited environments, increase in SM leads to decrease in LST, whereas in energy-limited environments, their relationship is not significant [64,65]. Similarly, brightness temperature is among the predictors which influence SM variability.…”

Series or Parallel? An Exploration in Coupling Physical Model and Machine Learning Method for Disaggregating Satellite Microwave Soil Moisture

“…For example, the well-known change detection method is based on a linear relationship between SM and radar backscatter data [24]. The Polynomial Fitting method is based on the approximate polynomial relationship between SM and Land Surface Temperature (LST), Normalized Differential Vegetation Index (NDVI), surface net incoming radiation (Rn), and so on [29]. Additionally, more complex machine learning methods were employed such as the Bayesian merging method [30], Fourier transformation method [31], General Regression Neural Network, Artificial Neural Net-work (ANN), Random Forest (RF), Support Vector Regression (SVR), and Deep Learning methods [32][33][34][35][36], etc.…”