COMPARISON OF SOIL MOISTURE SIMULATION TECHNIQUES IN EERER SUB-BASIN, WABI-SHEBELE RIVER BASIN, ETHIOPIA

Abstract

Soil Moisture is a key component of the hydrological cycle and plays a crucial role in agricul- tural productivity, water resources management, and ecosystem health. The present study aims to simulate soil moisture using a SWAT semi-distributed hydrological model and Landsat 8 satellite images with a Thermal Optical Trapezoid Model (TOTRAM) in the Eerer Sub-basin, Wabi Shebele River Basin, Ethiopia. The Soil Water Assessment Tool (SWAT) was used to ex- amine the components of its water balance in both space and time. The parameters of Curve Number, Alpha base flow, groundwater delay, return flow threshold depth, and soil evaporation composition factor were shown to be sensitive for stream flow modeling based on the Global Sensitivity Analyses.Calibration and validation of SWAT using stream flow data to enable the water balance components such as soil moisture. Temporally the amount of rain fall in 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022 were 824.80mm, 781.65mm, 766.77mm, 464.71mm, 614.72mm, 864.55mm, 917.91mm, 834.05mm and 521.17mm which respons 116.23mm, 93.73mm, 72.04mm, 64.50mm, 74.12mm, 102.14mm, 124.37mm, 105.98mm,and 67.23mm soil moisture respectively. Spatially the highest soil moisture was observed in Eastern, Northwestern and the lowest soil moisture was observed in Northern and Southern part of the watershed. This study intended to apply the optical and thermal remote sensing data for esti- mating soil moisture in the Eerer sub-basin. Monthly simulation of stream flow modeling during the calibration with the SUFI2 algorithm provides a good result with a Nash Sutcliffe Efficiency (NSE) of 0.8 and R2 of 0.82. Besides stream flow data the SWAT model soil moisture product was validated by in situ soil moisture content performs well having NSE (0.56), R2 (0.80), RMSE (0.23), and PBIAS (-0.63). TOTRAM-based soil moisture estimates provide an overall perfor- mance of NSE, R2, RMSE, and PBIAS values of 0.64, 0.79, 0.039, and 22.52 respectively. The result indicates that validation of SWAT and TOTRAM models with in situ soil moisture data leads to acceptable accuracy with 0.031 cm3 cm-3, and 0.022 cm3 cm-3 standard errors respec- tively. The overall result indicates that the Landsat 8 image has a higher potential to simulate soil moisture comparable with Theta Prope