Impacts of hydrologic soil properties on drought detection with MODIS thermal data

Remote sensing data from the Moderate Resolution Imaging Spectroradiometer (MODIS), a climatic water budget model, and the STATSGO database were used within a GIS environment to determine the influences of hydrologic soil properties on soil moisture and thermal emission in western-central Kansas for a dry year, 2000. Two important variables, water-holding capacity (WHC) and hydrologic soil group (HSG), were controlled in our water budget experiment to evaluate their impacts on soil moisture content (SMC) changes throughout the period. Results showed that HSG affected drought detection and occurrence very little, but WHC variations explained most local variations of soil moisture content. As a strong indicator of relative soil moisture deficit, the Standardized Thermal Index (STI) patterns were also influenced by WHC. Generally, the earlier the soil moisture content drops below 40%, the earlier the STI reaches a threshold value of 0.2 or higher. Vegetation responses to thermal detection lagged behind the STI by up to 8 weeks, which was computed by comparing the STI and Normalized Difference Vegetation Index (NDVI) deviation from a 10-year mean. The spatial pattern of lag-times was not apparent, but lag-times were correlated with a WHC component. (C) 2003 Elsevier Inc. All rights reserved.