SO2 emission controls, combined with modestly increasing ammonia, have been thought to generate aerosol with significantly reduced acidity for cases in which sulfate is partially substituted by nitrate. However, neither expectation agrees with decadal observations in the southeastern USA, suggesting that a fundamentally different response of aerosol pH to emissions changes is occurring. We postulate that this nitrate substitution paradox arises from a positive bias in aerosol pH in model simulations. This bias can elevate pH to a level at which nitrate partitioning is readily promoted, leading to behavior consistent with nitrate substitution. CMAQ simulations are used to investigate this hypothesis; modeled PM2.5 pH using 2001 emissions compare favorably with pH inferred from observed species concentrations. Using 2011 emissions, however, leads to simulated pH increases of one unit, which is inconsistent with observations from that year. Nonvolatile cations (K+, Na+, Ca+2, and Mg+2) in the fine mode are found to be responsible for the erroneous predicted increase in aerosol pH of about 1 unit on average over the USA. Such an increase can induce a nitrate bias of 1-2 μg mĝ-3, which may further increase in future projections, reaffirming an otherwise incorrect expectation of a significant nitrate substitution. Evaluation of predicted aerosol pH against thermodynamic analysis of observations is therefore a critically important, but overlooked, aspect of model evaluation for a robust emissions policy.

. © 2018 Copernicus GmbH. All rights reserved.