Evaluating MODIS Vegetation Indices as Ancillary Data in the Retrieval of Soil Moisture from Microwave Data

One approach to soil moisture retrieval from remotely sensed microwave data uses single frequency H polarization brightness temperature measurements. This algorithm requires ancillary information to estimate the effective temperature of the surface and the attenuation of the microwave signal by vegetation. When applied on a global basis using spaceborne microwave observations, such as AMSR-E, it is most efficient if the ancillary vegetation information is also based on satellite observations. At the present time, the best source of information on the spatial and temporal variations of vegetation biophysical properties is the Moderate Resolution Imaging Spectroradiometer (MODIS) on board the Terra and Aqua satellites. The MODIS land discipline group provides two types of vegetation indices (VI), the normalized difference vegetation index (NDVI) and the enhanced vegetation index (EVI). The first index normalizes the difference between the near infrared and red bands by their sum that cancels out a large proportion of signal variations attributed to calibration, noise and changing irradiance and atmospheric conditions. However, the NDVI can quickly saturate at high leaf biomass and is influenced by the properties of the soil background under sparse vegetation. The second index (EVI) was developed to optimize the vegetation signal while minimizing aerosol and canopy soil background sources of uncertainty. It is the vegetation water content (VWC) of the canopy that attenuates the microwave signal. Therefore, it is necessary to establish a relationship between the VIs and VWC. In limited tests, using AMSR-E 10 GHz H polarization data and one of the MODIS VIs (either NDVI or EVI) to estimate vegetation effect, the soil moisture retrieval results were mixed. Therefore, more specific studies were conducted to understand how NDVI or EVI are related to VWC under different vegetation biophysical conditions and to define the best choice of the VI for certain conditions. In this study, data collected for SMEX02 in Iowa, SMEX03 in Oklahoma, and SMEX04 in Arizona, representing densely, moderately and sparsely vegetated surface, will be used to evaluate MODIS VIs for their appropriateness in soil moisture retrieval from AMSR-E data.