Near-surface soil moisture content measurement by GNSS reflectometry: An estimation model using calibrated GNSS signals

The microwave signal of GNSS L band is very sensitive to the electromagnetic property of soil surface, through measuring the reflected electromagnetic energy from soil, the relationship between power of GNSS reflected signals and soil dielectric constant can be built, and then soil moisture can be estimated. In order to improve estimation accuracy, both the direct and reflected GNSS signals need to be calibrated to eliminate the signal error. A soil moisture estimation model utilizing calibrated GNSS L band reflected signals is proposed in this paper, and the estimation accuracy is validated using SMEX02 data. The error of areas with different NDVI is calculated. The error of soil moisture estimation is 7.04% for bare soil condition of the verify areas. It is shown that this model is suitable for bare soil condition or areas with low vegetation coverage. For further research, in the case of soil with high vegetation coverage, the model should be modified through adding the vegetation effect into it.