Daily rainfall estimates considering seasonality from a MODWT-ANN hybrid model

Abstract Analyses based on precipitation data may be limited by the quality of data, size available historical series and efficiency adopted methodologies; these factors are especially limiting when conducting analyses at daily scale. Thus, methodologies sought to overcome barriers. The objective this work is develop a hybrid model through maximum overlap discrete wavelet transform (MODWT) estimate rainfall in homogeneous regions Tocantins-Araguaia Hydrographic Region (TAHR) Amazon (Brazil). Data from Climate Prediction Center morphing (CMORPH) satellite products National Water Agency (ANA) were divided into seasonal periods (dry rainy), which train for forecasting. results show that had good performance forecasting using both databases, indicated Nash–Sutcliffe coefficients (0.81–0.95), thus, considered potentially useful modelling rainfall.