Soil Moisture Retrieval from Sentinel-2 Data Using the Optical Trapezoidal Model (OPTRAM)

Main Article Content

Sidheshwar Raut
Neha Andure
Rajeshwari Pangarkar
Vaibhav Misal
Shafiyoddin Sayyad
Sandipan Sawant

Abstract

Surface soil moisture is a determining factor in developing hydrological and ecological processes, especially in the arid and semi-arid regions. Still, its accurate and extended special capacity estimation remains a challenge to the traditional ground-based techniques. In our study, we discuss prospects of the Optical Trapezoidal Model (OPTRAM) in the retrieval of soil moisture from high-resolution optical satellite data. The model exploits interactions among the vegetation indices and shortwave transformed reflectance to estimate the surface moisture without using thermal inputs. The applied over agricultural territory of semi-arid region near Beed, Maharashtra. In-situ data was collected at the same time when the satellite was over our region where we have employed gravimetric method with on ground observations in oven dry and the theta probe. To ensure cohesion between the two sets of the data, the results obtained from the gravimetric method were calibrated back to volumetric units for comparing with the results obtained from theta probe the comparison showed a good level of agreement (R²) of 0.8873, (RMSE) of 0.033 cm³/cm, indicating the reliability of the volumetric data obtained in the field. After this calibration, the OPTRAM retrieved soil moisture was compared to the Theta probe and used as the reference to validate latter. The comparison was strongly correlated with (R²) 0.8253 and (RMSE) 0.0437 cm³/ cm³ showing that OPTRAM has been able to capture the spatial variance of surface soil moisture under different field condition. These results emphasize on the model’s ability to effectively deliver reliable high-resolution estimates of soil moisture based on optical data retrieved from satellites. These results emphasize the stamina and scalability of the model for soil moisture retrieval, the present work takes up the need to assess the suitability of the Optical Trapezoidal Model (OPTRAM) for soil moisture retrieval using Sentinel-2 optical data. It is concerned with improving estimates of soil parameters in a diverse variety of land covers under a semi-arid landscape.

Article Details

Section
Articles