ESTIMATION OF SOIL MOISTURE FOR SUGARCANE CROP USING OPTICAL AND THERMAL REMOTE SENSING DATA

Abstract

Accurate estimation of soil moisture is crucial for effective irrigation management and yield optimization in sugarcane cultivation. This study aimed to assess the potential of optical and thermal remote sensing techniques for estimating soil moisture in sugarcane fields of Nagla Aimad, Roorkee, Uttarakhand, India. Using active microwave remote sensing to retrieve soil moisture from vegetated areas is a difficult process because surface and volume scattering from the underlying soil are combined in the scattering from the vegetated area. Furthermore, vegetation acts as a two-way attenuator for the signal scattered by the underlying soil. Consequently, to retrieve soil moisture, a method that can accurately depict the scattering behaviour from the vegetation-covered area must be used. Creating a statistical model to estimate soil moisture from a combination of ground-based measurements and spectral indices derived from satellites was the main goal. Landsat 9 satellite imagery covering the study region is used for this, with concurrent dates ranging from August 2023 to February 2024. The latitude, longitude, soil moisture content, and canopy temperature of the various sites were measured in the fields in accordance with the satellite pass times. Creating spatial maps of the distribution of soil moisture throughout the study region was another goal of this study. The process comprised combining satellite imagery from Landsat 9 with ground-truth data obtained with a Time Domain Reflectometry (TDR) meter. The measured soil moisture values were used to calculate and correlate several spectral indices, such as the Soil-Adjusted Vegetation Index (SAVI), Temperature Vegetation Dryness Index (TVDI), Enhanced Vegetation Index (EVI), and Normalized Difference Vegetation Index (NDVI). The findings showed that, when combined with SAVI in a multilinear regression model, TVDI showed the strongest correlation with soil moisture. The derived regression equation was used to estimate soil moisture. The estimated values and the observed soil moisture measurements were compared, and a satisfactory correlation (R2 = 0.66) was found. The study area was divided into low, moderate, and high soil moisture zones based on the estimated values of soil moisture. To identify spatial variations throughout the study area, spatial maps were created to visualize the distribution of soil moisture and spectral indices. The results show the potential for accurate soil moisture estimation in sugarcane fields through the integration of optical and thermal remote sensing techniques. This can lead to more sustainable agricultural practices and more effective irrigation water management.