A multiregional, multimodel examination of the sensitivities, mechanisms, and consequences of the ozone climate penalty

Abstract:

Tropospheric ozone (O₃) pollution levels have widely observed correlations to daytime surface temperatures, especially in highly polluted regions. This nonlinear correlation occurs through a variety of temperature-dependent mechanisms related to O₃ precursor emissions, lifetimes, and reaction rates, making the reproduction of temperature sensitivities – and the associated human health risks connected to increased pollution – a problematic task. Here we examine the observed summertime O₃ climate penalty across several regions worldwide using quantile regression, allowing for an examination of temperature sensitivities across the O₃ response distribution. Through this methodology, we show significant differences between the sensitivities of high vs. low O₃ percentiles, suggesting that climate penalties based on means and ordinary least squares regressions may in fact underestimate the magnitude of expected changes to extreme O₃ events. We then examine these relationships as reproduced by the global chemical transport model GEOS-Chem, perturbing individual mechanisms within the model to both highlight possible explanations for model/observation discrepancies, and also better understand the regionality and magnitude of individual mechanism contribution to the observed penalty. Finally, we explore the future consequences of the climate penalty as forecast by commonly used global climate models, proposing corrections based on the quantile and regional differences from observed sensitivities.