Causes of model biases in simulating inorganic aerosol composition during KORUS-AQ and implications for the estimate of transboundary pollution

Abstract:

East Asia is a region of increasing economic growth which has led to severe PM_2.5 pollution in urban areas. The joint NASA-NIER Korea-United States Air Quality (KORUS-AQ) field campaign in May-June 2016 provided an extensive dataset of ground, airborne, and remote sensing observations to test model simulations of PM_2.5 pollution transport and potential control measures. During KORUS-AQ, a period of haze was observed in which PM_2.5 rapidly increased to the highest levels observed during the campaign. This increase is associated with increasing inorganic aerosol. While a portion of this increase is due to long-range transport, there is observational evidence that aerosol formation from local precursors was also enhanced. This suggests that domestic policy measures could have a greater than expected influence on controlling PM_2.5. However, models have difficulty reproducing PM_2.5 levels, particularly the composition of secondary inorganic aerosol. Models generally underestimate sulfate and overestimate nitrate and fail to represent the peak levels of PM_2.5 during haze events. These biases have been chiefly attributed to errors in chemical mechanisms and model meteorology, not issues with underlying emissions inventories. Here, we use observations from KORUS-AQ interpreted by the GEOS-Chem chemical transport model to improve the model’s ability to reproduce secondary inorganic aerosol concentrations during the campaign and explore mechanisms to improve model biases during the haze event. We assess the fraction of inorganic aerosol from the improved model simulation that results from transboundary transport during KORUS-AQ and test the model sensitivity to potential emission reduction measures that could improve air quality in Seoul during different meteorological periods including haze episodes.