2007 BES Annual Meeting Presentation and Poster Abstracts
Characterizing Surface Moisture Conditions in Maryland Climate Division 6 Using Moderate Resolution Satellite Imagery
Co-Authors: Lawrence E. Band, Aaron Moody, Jingfeng Xiao
Abstract: Satellite remote sensing of land surface conditions provides the potential to develop an important set of tools to study detailed space/time patterns of hydrologic storage and flux. Previous work has concentrated on using combinations of thermal and optical remote sensing information to estimate moisture status resulting from the interaction of land cover and meteorological forcing. We make use of a Temperature-Vegetation Dryness Index (TVDI), calculated using Moderate Resolution Imaging Spectroradiometer (MODIS) imagery to investigate 2002 growing season soil moisture and evaporative flux patterns in Maryland Climate Division 6, where the BES study sites are located. An antecedent precipitation index (API) was derived from NEXRAD Doppler radar imagery for the same time period and locations. Variables describing the fraction of each pixel occupied by land use classes were produced using MODIS land cover data. Linear models of TVDI as a function of API and the land use variables were constructed. Adjusted R2 values for most models were between 0.5 and 0.7. Both the API variable and the ensemble of LULC variables are significant contributors to the modelsí fit, demonstrating that TVDI is a function of both precipitation and land use patterns. Field samples of soil moisture were available at two locations (suburban and forested) within the study climate division, and these sampled values also correlated well with TVDI values for the same locations, demonstrating that TVDI can be a useful indicator of surface moisture status at locations where these land uses are present.