Regional Heatwave Prediction Using deep learning based Recurrent Neural Network

The increased frequency and severity of heatwaves are a result of global warming’s increasing temperatures. The performance of various reliable prediction models has decreased because of changes in the world environment. LSTM neural networks, among other deep learning-based recurrent neural network algorithms, are used to develop a reliable model for the prediction of heatwave maximum temperature. In this research, we have developed a percentile-based threshold over the predicted maximum temperature to forecast heatwaves using LSTM based predictive model. The LSTM algorithm was applied in this study to determine the approximate maximum temperature for a severe heatwave. The obtained results demonstrate that the proposed percentile approach based on the deep learning LSTM model can solve this issue quickly and effectively.