Sensitivity of microwave remotely-sensed soil moisture to soil properties

Surface soil moisture is important in a number of disciplines including agricultural scheduling, water resource management, and weather forecasting. While conventional methods of surface soil moisture determination are labor intensive and of limited spatial scale, ground- and aircraft-based microwave remote sensing systems provide repetitive measurements over large areas. However, it is unclear how the remotely-sensed soil moisture reflects the variations in soil parameters used as input variables in soil moisture retrieval algorithms. Data from a multi-frequency passive microwave remote sensing experiment were used to study the sensitivity of an algorithm to soil texture, bulk density and surface roughness. The algorithm was relatively insensitive to texture, slightly sensitive to bulk density and highly sensitive to surface roughness (RMS). For the most part, the deviations in volumetric soil moisture from the baseline were more pronounced under wet conditions than under dry conditions, and these deviations were also independent of wavelength. This study implies that areal average values of texture variables and bulk density can serve as convenient inputs into soil moisture algorithms