A Dual-Branch Transformer with Adaptive Residual Correction for Improving High-Ozone Forecast Skill
Abstract
Near-surface ozone (O3) pollution is a growing environmental concern, particularly in the Beijing–Tianjin–Hebei (BTH) region, one of China’s most densely populated megacity clusters experiencing increasingly severe O3 episodes. Existing data-driven forecasting models systematically underestimate high-concentration events and offer limited lead times. To reveal the meteorological drivers of extreme O3 episodes, we conducted composite anomaly analysis over 2019–2023 and identified the dominant meteorological mechanism as a coupled pattern of mid-tropospheric anticyclonic circulation with high temperature, low humidity, and deep subsidence inversion, which suppresses vertical diffusion while southerly advection drives rapid near-surface O3 accumulation. Motivated by meteorological diagnostics, we proposed ARC-Net, a Transformer-encoder-based Adaptive Residual Correction Network that ingests numerical weather prediction data from the European Centre for Medium-Range Weather Forecasts (ECMWF) and air quality observations to produce hourly O3 forecasts up to 240 h (10 days) ahead. The model features a dual-branch regression-classification architecture enhancing feature discrimination at high concentrations and an Adaptive Residual Correction module that dynamically calibrates outputs through a triple-gating mechanism conditioned on pollution-level priors. In independent forecast tests for the year 2023 across 13 cities in the BTH region, ARC-Net achieved R2 = 0.879 and a root mean square error (RMSE) of 17.03 μg/m3 at 0–24 h, retaining R2 = 0.749 and RMSE = 24.57 μg/m3 at 0–240 h. For extreme episodes (maximum daily 8 h average ozone (MDA8_O3) ≥ 215 μg/m3), the Critical Success Index improved by 63.9% over the baseline, and RMSE decreased by 33.15% within the 215–265 μg/m3 range in a representative case. These results indicate that meteorology-guided predictors combined with adaptive residual correction can partially alleviate high-O3 underestimation and provide practically useful medium-range warning skill.