Estimation of soil surface moisture and planning of sugarcane irrigation using thermal trapezoidal model

Document Type : Research Paper


1 Head of Remote Sensing and GIS Office, Sugarcane Development Research and Training Institute, Ahvaz, Iran

2 Associate Professor, Department of Remote Sensing and GIS, Faculty of Geography, University of Tehran, Iran

3 Professor, Department of Irrigation and Drainage, Faculty of Water Science Engineering, Shahid Chamran University, Ahvaz, Iran


Soil moisture is one of the key parameters in water resources studies and irrigation remote planning. Measuring soil moisture on a large scale is very time-taking and costly. Traditional methods of measuring soil moisture at farm level cannot show the spatial changes of moisture in the best way. Various new methods have been developed to use satellite data to model soil moisture based on thermal images. This study was conducted in 2020 with the aim of investigating the ability of thermal satellite imagery to estimate soil moisture and to plan irrigation rounds of land irrigation and sugarcane industry of Amirkabir located in the south of Khuzestan province. For the purpose of sugarcane season, soil moisture content was calculated for 9 crossings of Landsat 8 satellite and evaluated using 180 ground control points, and also daily irrigation data of 32 farms 25 hectare were recorded during the study period. The results showed that the accuracy of the model is suitable for estimating soil moisture with the measured values at the farm level. The mean square error root (RMSE) was 4.45%, the mean square root of normalized error (NRMSE) was 12.9% and the coefficient of explanation (R²) was 0.82. Also, the results of soil moisture in planning the irrigation round of sugarcane fields showed that the thermal trapezoidal model is effective due to the use of thermal bands to environmental factors such as relative air moisture percentage, average air temperature, pest (leaf dryness) and plant temperature, which causes errors in irrigation planning of sugarcane fields. The mean square root of normalized error (NRMSE) for soil moisture stress class was 24.32%, irrigation time class was 22.20%, average moisture class was 11.7%, high moisture class was 13.20% and irrigation class was 8.86%.


Main Subjects