Estimation of soil moisture status based on TVDI and ATI in Heihe River basin

Abstract:

Based on TVDI (Temperature-Vegetation Dryness Index) and ATI (Apparent Thermal Inertia), this study estimated the soil moisture status in Heihe river basin (northwestern of China) using the vegetation, surface temperature and surface reflectance products in 2012 from MODIS. When calculating the dry edge and wet edge in temperature-vegetation triangle framework, a step-width of 0.02 in NDVI was set and the maximum and minimum temperatures in each interval were found. A fitting line was derived based on these points using the least square method. The results shows that the values of R² of the dry edge is larger than that of the wet edge. The average R² of dry edges is 0.89, while the average R² of wet edges in only 0.74. The slope of dry edges reaches its peak absolute value of -61.51 in DOY 97 (day of year) in 2012. To divide the whole basin into different parts with different degrees of soil moisture status, defined intervals and natural breaks methods were used for TVDI and ATI respectively. The majority of the upstream of Heihe river is comparable or general wet region, with the downstream of the river basin being the dry region. The result of TVDI shows that the midstream of the river belongs to the dry region, as same as its surrounding areas. However, the result of ATI shows the midstream of the river is much wetter. The comparison between the in-situ precipitation data and TVDI and ATI indicates that TVDI and ATI both respond to the precipitation. Only the remote sensing data is need to estimate the soil moisture status based on the TVDI and ATI methods, which can provide straightforward approaches to estimate soil moisture and monitor drought.