Heatwaves have substantial but poorly quantified impacts on surface ozone photochemical regimes. As heatwaves of increasing severity occur, communities face more serious exposure to ozone, necessitating a more comprehensive understanding of the impact of heatwaves on the nonlinear response of ozone to its precursors for guiding policies in emission reductions. Here we estimate the spatiotemporal evolution of global ozone chemistry based on machine learning and in situ observations and show that emission changes and heatwaves alter ozone photochemical regimes, leading to diverse ozone changes across regions. Sustained emission reductions in East Asia decreased the ozone formation sensitivity to formaldehyde (HCHO) and fine particulate matter (PM2.5), counteracting the adverse high-temperature effect. Quantified results reveal that heatwaves increased the sensitivity of ozone to HCHO and PM2.5, enhancing their positive contributions and causing increased ozone trends across most regions, with a global average anomaly of 9.4 μg/m3. Meanwhile, heatwave-induced PM2.5 anomalies concentrated in wildfire-risk zones, coupled with increased HCHO, elevated downwind ozone levels. Specifically, the effects in wildfire-endangered western Canada and heatwave-exposed southeastern United States contributed to a chemically driven ozone increase of 0.18 μg/m3/month in Northern America. Our results demonstrate that more targeted and substantial regulation of volatile organic compounds will be beneficial in mitigating future intensifying climate penalty effects.
Keywords: deep learning; global; heatwave; ozone changes; photochemical regime.