A novel hybrid modeling approach based on empirical methods, PSO, XGBoost, and multiple GCMs for forecasting long-term reference evapotranspiration in a data scarce-area

Abstract

Estimation of the Reference Evapotranspiration (ET_o) is critical in water resources management under climate change, especially in arid and semi-arid regions. Thus, estimating baseline ET_o poses significant challenges, particularly in inadequate climatological monitoring regions. In this study, a hybrid modeling approach based on the incorporation of empirical models, Particle Swarm Optimization (PSO), and XGBoost algorithm (Empirical-PSO-XGBoost) was developed and evaluated to forecast ET_o under limited climate variables. The results showed the Empirical-PSO-XGBoost outperformed the purely calibrated empirical and Temperature-PSO-XGBoost models for estimating monthly (daily) ET_o with NSE reaching 0.99 (0.86) and 0.98 (0.67) for the calibration and validation phases, respectively. Besides, up to 63 CMIP6 projections were coupled with Empirical-PSO-XGBoost for forecasting the long-term ET_o under SSP245 and SSP585 climate change scenarios. Thus, the simulation showed a significant increase in ET_o and seasonal patterns compared to the baseline ET_o where the change in range of [+5, +10] % is associated with probability values of 0.65 and 0.78 for SSP245 and SSP585, respectively. Overall, the developed framework is useful for implementing adaptation strategies to mitigate climate change effects on water resource allocation and agricultural management. It provides the ET_o associated with Exceedance probability for each month which is useful for assessing the water availability-related-risk in scheduling irrigation and sowing date of crops.