A time domain solution of the Modified Temperature Vegetation Dryness Index (MTVDI) for continuous soil moisture monitoring

This study presents indexes developed from remote sensing observations for continuous soil surface moisture monitoring. The Modified Temperature Vegetation Dryness Index (MTVDI) proposed in our earlier work is developed using a time domain solution method rather than the traditional spatial domain solution method, because of limitations arising from a traditional temperature-vegetation index (TVX) method. In the procedure herein proposed, the MTVDI parameterization scheme is transformed from regional scale to pixel scale, and the boundary conditions defining maximum water stress are determined pixel by pixel using the surface energy balance principle. All the parameters required for the retrieval of MTVDI are pixel specific, thereby avoiding the limitations of the traditional spatial domain solution method. Previous studies demonstrate the applicability of the TVX method with data from only a few days of clear sky conditions. In contrast, our proposed MTVDI is demonstrated using Moderate Resolution Imaging Spectroradiometer (MODIS) products over the course of a full year, and the soil moisture status over the Southern Great Plains (SGP) region is monitored continuously. Finally, the volumetric surface soil water content (theta(nu)) is estimated using a calibration procedure from the MTVDI retrievals. The results show that the accuracy of both MTVDI retrievals and theta(nu) estimates obtained in this work has reached a level comparable with those produced in previous studies. Across all sites, the correlation coefficient (r) between the MTVDI and theta(nu) measurements is 0.60. The values of r, mean absolute error (MAE) and root mean square error (RMSE) for theta(nu) estimates are 0.75, 0.019 m(3) M-3 and 0.025 m(3) m(-3), respectively.