Soil moisture retrieva‎l using radar remote sensing and polarimetric methods at the farm scale

2/20/2024 12:00:00 AM

Soil moisture is an important parameter that affects environmental processes and the hydrological cycle. Furthermore, this parameter plays a key role in fields such as agriculture, meteorology, floods, drought. In this research, a framework based on the polarimetric decomposition of radar images in interaction with optical-thermal images is proposed for soil moisture retrieva‎l. This study includes two phases: 1- Approximation of soil moisture at the farm level based on optical-thermal images of Landsat-8 satellite 2- Retrieva‎l of soil moisture by the dual polarized basis decomposition model with the help of approximate soil moisture that is obtained by optical-thermal model. In the proposed basis decomposition method, the 3x3 coherence matrix is converted into a 2x2 matrix. In this research, two mechanisms: volume scattering and surface scattering are considered. Furthermore, in order to model surface scattering, Bragg matrix has been used. The obtained 2x2 matrix is able to use the dual images (VV-VH) of the Sentinel-1 satellite. Carlson triangular model has been used to approximate soil moisture using optical-thermal images. Three indices NDVI, NDMI and MSI have been used in the optical-thermal model. Two ground soil moisture data sets are used in this study: 1- CAF farm soil moisture data located in the United States of America 2- RISMA soil moisture data located in Canada. In this study, the optical-thermal model is first calibrated in the CAF study site, then this model is applied in the CAF farm and the Canadian study site; By applying the optical-thermal model on these two sites, the maximum and minimum soil moisture at the farm level has been obtained, which are converted into the dielectric coefficient using the Topp model; The maximum and minimum of the dielectric coefficient are entered into the base decomposition model. Additionally, calibrating the optical-thermal model has been done on two dates.The lowest RMSE value obtained for the radar model (base decomposition model) is 3.33%. Also, the highest RMSE value of the radar model is estimated at 11.21%. The lowest RMSE value obtained for the optical-thermal model is 4.04%. Also, the highest RMSE value of the optical-thermal model is estimated to be 9.68%. If we enter the average soil moisture obtained from the three indices NDVI, NDMI and MSI into the radar model, the highest and lowest RMSE values of the optical-thermal model are 7.43% and 4.07%. Furthermore, in this case, the highest and lowest RMSE values for the radar model are 10.04% and 4.11%. In this study, in addition to CAF and RISMA data sets, ground data related to Iran have been used. This farm is located in the Kurdistan province. Soil moisture and surface roughness have been measured in this farm. The method of measuring soil moisture in this farm is weight method. The total number of soil moisture measurement points is 21 points. Soil moisture has been measured three times in this farm. The optical-thermal model is calibrated on the second date. The maximum and minimum RMSE values of the optical-thermal model for this farm are 4.32% and 3.06%. Furthermore, the maximum and minimum RMSE values of the radar model for this farm are 5.54% and 2.93%.

رطوبت خاك , سنجش از دور , لندست-8 , سنتينل-1

soil moisture , remote sensing , landsat-8 , sentinel-1

Mohammad Moghaddas

Dr. Barat Mojaradi - Dr. Hosein Alizadeh