Assessing Surface Soil Moisture in Arid and Semiarid Rangelands Using NDVI and Meteorological Parameters

Soil moisture and its temporal and spatial variations are one of the critical parameters in climatic, ecological, and hydrological modeling. Overall and consistent views on different parts of an area, repetitive satellite images from inaccessible areas and over large zones, along with regular periodic data are among important outcomes of the Remote Sensing (RS). Rangelands are highly dependent on the availability of water and therefore the vegetation indices can be used for tracking the changes in soil moisture. This study analyzes the correlation between the field measurements of soil moisture obtained using gravimetric sampling and the coincided and lagged MODIS-NDVIs based on the daily meteorological parameters during the growing season (April- August) from 2003 to 2005 in arid and semi arid rangeland in Khorasan Province, north eastern Iran. Results show that the surface soil moisture in these rangelands is moderately correlated to the coincided and lagged NDVI and can be estimated using NDVI during the growing season. Stronger relations can be obtained for surface soil moisture data lagged by 2 weeks with respect to the vegetation index. Considering wind and cumulative evaporation factors represent better temporal and spatial estimation for the surface soil moisture in the study area.