Impact of climate related uncertainty on projections of US air quality and implications for extremes

Abstract:

Propagation of climate uncertainties to projections of the climate change impacts on air quality has not been systematically assessed, despite significant research advances in the understanding of the contributions of internal variability, model response, and policy to the total uncertainty in climate simulations. We use a coupled climate-air quality modeling ensemble to quantify and compare the contributions of major sources of climate uncertainty to uncertainty in 21^st century projections of O₃ and PM_2.5 pollution over the US. We find that in terms of the climate-induced change in air pollutant concentrations resulting from climate change alone, known as the air quality penalty, unforced internal variability accounts for the largest fraction of uncertainty in early-century projections, and contributes to uncertainty equally with model response and climate policy at mid-century. By end century, policy dominates total uncertainty with contributions of 3.8 ppb O₃ and 1.2 µg m^-3 PM_2.5 to estimates of annual U.S. population-weighted concentrations, followed by model response (1.6 ppb and 0.8 µg m^-3) and internal variability (0.8 ppb and 0.4 µg m^-3), with significant regional variations for all uncertainty sources. Climate model response uncertainty additionally increases the variance of projected impacts, leading to larger and more frequent air pollution extremes under high climate model response scenarios. This study highlights the role of uncertainty in projections of climate impacts on air quality, with a particular focus on climate model response and implications for extreme air pollution events.