Prediction of groundwater table and drought analysis; a new hybridization strategy based on bi-directional long short-term model and the Harris hawk optimization algorithm

Abstract In the present study, a new hybridization strategy for predicting groundwater table (GWT) and drought analysis is presented. Therefore, hybrid of bi-directional long short-term model (BLSTM) Harris hawk optimization (HHO) algorithm, namely BLSTM–HHO applied. this lagged data GWT are used as input, whereas current output. Additionally, standalone BLSTM, (LSTM), artificial neural networks (ANN), Seasonal Autoregressive Integrated Moving Average (SARIMA), (ARIMA) employed benchmark simulating algorithms. The results show that algorithm has more accuracy than other investigated algorithms based on different evaluation criteria such relative root mean squared error (RRMSE), Nash–Sutcliffe coefficient (NSE), refined Willmott index (dr). prediction (from 2018 to 2022) in all three aquifers decline (?5.40 m Brojen aquifer, ?7.23 Javanmardi ?5.81 Shahrekord aquifer). Accordingly, by ground resource (GRI) areas shows expected be continued next 5 years with an increasing magnitude severity.