Estimating surface soil moisture with the scanning low frequency microwave radiometer (SLFMR) during the Southern Great Plains 1997 (SGP97) hydrology experiment

Abstract

The scanning low frequency microwave radiometer (SLFMR) was used to map surface soil moisture (0–5 cm depth) during the Southern Great Plains 1997 (SGP97) hydrology experiment. On June 29, July 2, and July 3, surface soil moisture maps with a pixel resolution of 200 m were obtained using a soil moisture retrieval algorithm, developed for L-band (1.4 GHz frequency, 21 cm wavelength) passive microwave data. In comparison with the 800 m resolution data from the electronically scanned thinned array radiometer (ESTAR), the higher resolution SLFMR data required a more site specific calibration. After calibration root mean square difference (RMSD) between model and observed surface soil moisture observations were on the order of 5%. Although the higher pixel resolution generally provided brightness temperatures of individual fields, it is also meant the greater spatial variability in land cover properties (primarily vegetation cover) were affecting the microwave observations and had to be accounted for in the soil moisture algorithm. Parameters in the soil moisture algorithm required local recalibration, particularly for the heavily vegetated fields, in order to account for vegetation effects on the microwave brightness temperatures. Thus having microwave data at resolutions that differentiate field boundaries with sharp contrasts in vegetation cover amounts will likely require greater variation in parameter values (and more uncertainty) be assigned to the soil moisture algorithm than at coarser resolutions. This result indicates that parameter values in the soil moisture algorithm may be resolution dependent under certain land cover conditions, particularly at resolutions that discriminate field boundaries.