This study investigated reservoir operation under climate change for a base period (1981–2000) and future period (2011–2030). Different climate change models, based on A2 scenario, were used and the HAD-CM3 model, considering uncertainty, among other climate change models was found to be the best model. For the Dez basin in Iran, considered as a case study, the climate change models predicted increasing temperature from 1.16 to 2.5°C and decreasing precipitation for the future period. Also, runoff volume for the basin would decrease and irrigation demand for the downstream consumption would increase for the future period. A hybrid framework (optimization-climate change) was used for reservoir operation and the bat algorithm was used for minimization of irrigation deficit. A genetic algorithm and a particle swarm algorithm were selected for comparison with the bat algorithm. The reliability, resiliency, and vulnerability indices, based on a multi-criteria model, were used to select the base method for reservoir operation. Results showed the volume of water to be released for the future period, based on all evolutionary algorithms used, was less than for the base period, and the bat algorithm with high-reliability index and low vulnerability index performed better among other evolutionary algorithms.

Utilizing new approaches to accurately predict groundwater level (GWL) in arid regions is of vital importance. In this study, support vector regression (SVR), Gaussian process regression (GPR), and their combination with wavelet transformation (named wavelet-support vector regression (W-SVR) and wavelet-Gaussian process regression (W-GPR)) are used to forecast groundwater level in Semnan plain (arid area) for the next month. Three different wavelet transformations, namely Haar, db4, and Symlet, are tested. Four statistical metrics, namely root mean square error (RMSE), mean absolute percentage error (MAPE), coefficient of determination (R²), and Nah-Sutcliffe efficiency (NS), are used to evaluate performance of different methods. The results reveal that SVR with RMSE of 0.04790 (m), MAPE of 0.00199%, R² of 0.99995, and NS of 0.99988 significantly outperforms GPR with RMSE of 0.55439 (m), MAPE of 0.04363%, R2 of 0.99264, and NS of 0.98413. Besides, the hybrid W-GPR-1 model (i.e. GPR with Harr wavelet) remarkably improves the accuracy of GWL prediction compared to GPR. Finally, the hybrid W-SVR-3 model (i.e. SVR with Symlet) provides the best GWL prediction with RMSE, MAPE, R2, and NS of 0.01290 (m), 0.00079%, 0.99999, and 0.99999, respectively. Overall, the findings indicate that hybrid models can accurately predict GWL in arid regions.