In this work, we attempt to estimate the regional "time of emergence" of air quality climate penalties using an ensemble modeling framework based on the MIT Integrated Global System Model (MIT IGSM). With this framework, we drive offline atmospheric chemistry models (CAM-Chem and GEOS-Chem) using a suite of climate simulations assuming different underlying climate sensitivites and emissions policies over the period from the present to 2100. This ensemble approach allows us to study both the uncertainty in time of emergence due to natural variability as well as some degree of stuctural uncertainty based on model-simulated climate-air quality relationships.
We find that the air quality response to climate change can vary dramatically across different regions of the globe. To analyze these regionally-varying climate signals, we employ a hierarchical clustering technique to identify regions with similar seasonal patterns of air quality change. Our simulations suggest that the earliest emergence of ozone climate penalties would occur in Southern Europe (by 2035), should the world neglect climate change and rely on a "business-as-usual" emissions policy. However, even modest climate policy dramatically pushes back the time of emergence of these penalties - to beyond 2100 - across most of the globe. The emergence of climate-forced changes in PM2.5 are much more difficult to detect, partially owing to the large role that changes in the frequency and spatial distribution of precipitation play in limiting the accumulation and duration of particulate pollution episodes. Finally, we motivate future work which closes the feedback loop between atmospheric chemistry impacts and transient climate change as well as integrated assessment modeling using our approach to quantify the transient evolution of health and economic co-benefits yielded from climate mitigation.