The Capability of Deep Learning Model to Predict Ozone Across Continents in China, the United States and Europe

Data‐driven methods have been extensively applied to predict atmospheric compositions. Here, we explore the capability of a deep learning (DL) model to make ozone (O3) predictions across continents in China, the United States (US) and Europe. The DL model was trained and validated with surface O3observations in China and the US in 2015–2018. The DL model was applied to predict hourly surface O3over three continents in 2015–2022. Compared to baseline simulations using GEOS‐Chem (GC) model, our analysis exhibits mean biases of 2.6 and 4.8 μg/m3with correlation coefficients of 0.94 and 0.93 (DL); and mean biases of 3.7 and 5.4 μg/m3with correlation coefficients of 0.95 and 0.92 (GC) in Europe in 2015–2018 and 2019–2022, respectively. The comparable performances between DL and GC indicate the potential of DL to make reliable predictions over spatial and temporal domains where a wealth of local observations for training is not available. Machine learning techniques have been extensively applied in the field of atmospheric science. It provides an efficient way of integrating data and predicting atmospheric compositions. Here, we explore the capability of a deep learning (DL) model to make ozone (O3) predictions across continents in China, the United States (US) and Europe. The DL model was trained and validated with surface O3observations in China and the US in 2015–2018. We then applied the DL model to predict surface O3concentrations in China, the US and Europe in 2015–2022. Our analysis exhibits comparable performances between DL and chemical transport models for surface O3predictions in Europe. This indicates the potential of DL models to extend predictions across spatial and temporal domains. Deep learning exhibits acceptable capabilities of spatial and temporal extrapolations for surface O3predictionsComparable performances between deep learning and GEOS‐Chem models with respect to independent O3observationsGood performance of deep learning for rapid hourly O3predictions Deep learning exhibits acceptable capabilities of spatial and temporal extrapolations for surface O3predictions Comparable performances between deep learning and GEOS‐Chem models with respect to independent O3observations Good performance of deep learning for rapid hourly O3predictions