摘要: In order to have effective agricultural production the impacts of drought must be mitigated. An important aspect mitigating is an method forecasting future events. this study, three methods short-term for short lead times are explored in Awash River Basin Ethiopia. The Standardized Precipitation Index (SPI) was index chosen represent basin. following machine learning techniques were study: artificial neural networks (ANNs), support vector regression (SVR), and coupled wavelet-ANNs, which pre-process input data using wavelet analysis (WA). forecast results all compared two performance measures (RMSE R 2 ). study indicate that network (WA-ANN) models most accurate SPI 3 (3-month SPI) 6 (6-month values over 1 months