An ensemble deep learning approach to spatiotemporal tropospheric ozone forecasting: A case study of Tehran, Iran

This study proposes a novel framework for spatiotemporal forecasting of Ground-level Ozone Concentration (GOC) using advanced machine learning techniques, including Artificial Neural Networks (ANN), Regularized Extreme Learning Machines (RELM), Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), and their wavelet counterparts. Data from eight air quality stations in Tehran, Iran, were employed for training and testing. The WBiLSTM model offered the most precise predictions, followed by WLSTM and WGRU. The forecasts were obtained by WRF inputs without using time-lagged GOC observations into ML models. The combination of the WDDFF approach and selective wavelet decomposition has the potential to enhance ML models for GOC forecasting.

Recommended citation: Rezaali, M., Jahangir, M. S., Fouladi-Fard, R., & Keellings, D. (2024). An ensemble deep learning approach to spatiotemporal tropospheric ozone forecasting: A case study of Tehran, Iran. Urban Climate, 55, 101950.
Download Paper