Dielectric behavior of wet soils in the 1--36 GHz microwave range: Modeling and validation

Abstract

Earth observation by the Advanced Microwave Scanning Radiometer 2 (AMSR2) aboard the Global Change Observation Mission 1st-Water (GCOM-W) satellite is a promising method for monitoring the global daily soil moisture. The brightness temperatures observed at 10 and 36 GHz are used to estimate the soil moisture through a dielectric model that relates water content and permittivity of soils, for which the Dobson model has been widely used. In this study, the performance of the Dobson model was compared with those of other models and a novel model was proposed. The models were validated using the soil samples taken in Japan and Cambodia measured at 25 MHz -- 4 GHz and that of Toyoura Sand and bentonite measured at 1 -- 18 GHz. Large differences were observed among estimates by Dobson, Topp, Wang-Schmugge, and Mironov models. The estimates by the Wang-Schumugge and Mironov models were in fair agreement and comparable with the measurements. However, there was a large difference in the estimated quantity and permittivity of bound water between the two models, and the permittivity for both bound and free soil water estimated using the Mironov model was higher than that estimated by the other models. In our novel model, the quantity of bound water was defined as the water content at air-drying under a relative humidity of 50 %. The permittivity of bound water was obtained under a matric potential of -500 MPa. The matric potential was calculated using a pedotransfer function and the van Genuchten model, and the wet-soil permittivity was obtained using Birchak's mixing model. The estimated permittivity was fairly accurate for the soil types and frequencies examined in this study. The proposed model is an effective means of applying soil physics concepts to improve soil moisture estimates by a satellite for the wide range of soil types around the world.