Soil moisture content estimation using GNSS reflectometry (GNSS-R)

Abstract

Global Navigation Satellite System-Reflectometry (GNSS-R) is defined as a bistatic radar system. GNSS-R has a unique potential in microwave remote sensing application. The focus of this paper is the feasibility of estimating soil moisture content using GPS-L1 scattered or reflected signals from the earth's surface. GPS signals are received and processed using ground based platform at 4m height. In this paper soil moisture is estimated from the surface power reflectivity which is the power ratio of the reflected signal and the direct signal. Analytical solution for the dielectric constant is resolved. The dielectric constant is measured and the physically suitable value of soil dielectric constant is related with the semi-empirical model of a constant parameters. Soil moisture content is retrieved by exactly knowing the soil textural composition of sand S and clay C in percentage. To validate the results estimated soil moisture values are compared with the in-situ measurements. Finally for error analysis three statistical tools are used, absolute relative error, Root Mean Square Error (RMSE), and Root Mean Bias (RMB). The results demonstrate that soil moisture varies and changed significantly after rainfalls occurred. However, it is also observed that values are inconsistent at higher elevation angles in some cases.