ABSTRACT
The availability of accurate reference evapotranspiration (ETo) data is crucial for developing decision support systems for optimal water resource management. This study aimed to evaluate the accuracy of three empirical models (Hargreaves-Samani (HS), Priestly-Taylor (PT), and Turc (TU)) and three machine learning models (Multiple linear regression (LR), Random Forest (RF), and Artificial Neural Network (NN)) in estimating daily ETo compared to the Penman-Monteith FAO-56 (PM) model. Long-term data from 42 weather stations in Florida were used. Moreover, the effect of ETo model selection on sweet corn irrigation water use was investigated by integrating simulated ETo data from empirical and ML models using the Decision Support System for Agrotechnology Transfer (DSSAT) model at two locations (Citra and Homestead) in Florida. Furthermore, a linear bias correction calibration technique was employed to improve the performance of empirical models. Results were consistent in that the NN and RF models outperformed the empirical models. The empirical models tended to underestimate and overestimate small and high daily ETo values, respectively, with the HS model exhibiting the least accuracy. However, calibrated PT and TU models performed comparably to the ML models. Results also revealed that using an inappropriate ETo model could lead to over-irrigation by up to 54 mm during a single crop season. Overall, ML models have proven reliable alternatives to the PM model, especially in regions with access to long-term data due to their site-independent performance. In areas without long-term data for ML model training and testing, calibrating empirical models is viable, but site-specific calibration is needed. It is important to highlight that distinct plant species exhibit varying transpiration characteristics and, consequently, have different water requirements. These differences play a pivotal role in shaping the overall impact of ETo models on crop water use.
ABSTRACT
Erosion of soil refers to the process of detaching and transporting topsoil from the land surface by natural forces such as water, wind, and other factors. As a result of this process, soil fertility is lost, water bodies' depth is reduced, water turbidity rises, and flood hazard problems, etc. Using a numerical model of erosion rates and erosion risks in the Jejebe watershed of the Baro Akobo basin in western Ethiopia, this study mapped erosion risks to prioritize conservation measures. In this study, the Revised Universal Soil Loss Equation (RUSLE) model was used, which was adapted to Ethiopian conditions. To estimate soil loss with RUSLE, the rainfall erosivity (R) factor was generated by interpolating rainfall data, the soil erodibility (K) factor was derived from the soil map, the topography (LS) factor was determined from the digital elevation model (DEM), cover and management (C) factor derived from the land use/cover data, and conservation practices (P) factor generated from digital elevation model (DEM) and land use/cover data were integrated with remote sensing data and the GIS 10.5 environment. The findings indicated that the watershed annual soil loss varies from nearly 0 on a gentle slope of forest lands to 265.8 t ha-1 year-1 in the very steep slope upper part of the watershed, with a mean annual soil loss of 36.2 t ha-1 year-1. The total annual soil loss in the watershed is estimated to be around 919,886.5 tons per year. To minimize the amount of soil erosion in the watershed that had been most severely affected, we identified eight conservation strategies that could be implemented. These strategies were based on the participatory watershed development (PWD) principles established by the Ethiopian government and the severity of the erosion in the watershed. The study's findings showed that a GIS-based RUSLE soil erosion assessment model can provide a realistic prediction of the amount of soil loss that will occur in the watershed. This tool can also help identify the priority areas for implementing effective erosion control measures.
