Soil Moisture Content estimation using GNSS-IR method in Tanzania

Abstract

Soil moisture is a crucial parameter in hydrology and meteorology, impacting water management and plant survival. Traditional methods of retrieving soil moisture, such as gravimetric, lack spatial-temporal variations and are laborintensive. Remote sensing offers global coverage but suffers from low spatial resolution. In the past decade, a new method, Global Navigation Satellite Systems Interferometric Reflectometry (GNSS-IR), has been developed. It provides continuous soil moisture measurements with improved spatial and temporal resolution but is yet to be employed in Tanzania. This study processed a total of 21 Continuously Operating Reference Stations (CORS) stations available in Tanzania, installed from various scientific projects since 2005. The suitability of these stations for the GNSS-IR method was evaluated based on the reflecting surface and the presence of an L2C signal capable of penetrating ~5 cm below the soil. The evaluation leaves only one site (OLO6), suitable for soil moisture determination. One year of daily L2C Signal to Noise Ratio (SNR) data for OLO6 was processed using the GNSSREFL software. The results were validated using Soil Moisture Active Passive (SMAP) satellite data. The GNSS-IR method showed good agreement with SMAP mostly during the dry season, with RMSE 0.038 cm³/cm³ and SDD 0.036 cm³/cm³, compared to the wet season (when soil moisture is above 0.25 cm³/cm³). Despite its limitations during the wet season, this study provides valuable insights into the GNSS-IR method for soil moisture estimation in Tanzania.