Wavelet Packet-Genetic Programming: A New Model for Meteorological Drought Hindcasting

This study presents developing procedures and verification of a new hybrid model, namely wavelet packet-genetic programming (WPGP) for short-term meteorological drought forecast. To this end, the multi-temporal standardized precipitation evapotranspiration index (SPEI) has been used as quantifying parameter at two stations Ankara province, Turkey. The WPGP model comprises main steps. In first step, packet, which is generalization well-known transform, to decompose SPEI series into deterministic stochastic sub-signals. Then, classic genetic (GP) applied formulate sub-signal considering its effective lags. characterize component, different theoretical probability distribution functions were assessed, best one was selected integrate with GP-evolved function. efficiency cross-validated order autoregressive (AR1), GP, random forest (RF) models developed benchmarks in present study. results showed that robust superior AR1 RF, significantly increases predictive accuracy standalone GP model.