Constraining Atmospheric Sulfate Formation in the Northeastern US: Insights from Sulfur Isotopes

Sulfur dioxide (SO₂) is a critical atmospheric trace gas that poses significant environmental and human health risks. Once emitted, SO₂ is primarily oxidized to particulate sulfate (pSO₄) through gas-phase or aqueous-phase reactions. Despite an over 80% reduction in SO₂ emissions since implementing the Clean Air Act and its amendments, pSO₄ concentrations have not decreased as expected, necessitating further investigation. Current atmospheric chemistry models, which guide regulatory efforts, often overestimate SO₂ concentrations and underestimate pSO₄ concentrations, indicating potential missing sources or under-constrained SO₂ oxidation pathways. Isotopic measurements of sulfur (δ³⁴S) have been proven to be powerful tools for elucidating the oxidation pathways involved in pSO₄ formation.

In this study, particulate matter (PM₁₀) samples from the northeast US, collected between 2005 and 2015—a period of substantial SO₂ emission reductions—were analyzed for sulfur isotope variations (δ³⁴S) of pSO₄. The δ³⁴S(pSO₄) values averaged 5.3±1.0‰ (n=72), with no significant long-term variation or seasonality. These sulfur isotope data were then modeled using the US EPA Community Multiscale Air Quality Modeling System (CMAQ) with the sulfate tracking mechanism (STM) script to evaluate the model’s accuracy in simulating SO₂ oxidation chemistry. Our results indicate that the modeled sulfate oxidation ratio was biased low, and δ³⁴S was biased high during the summer months, suggesting a missing source of pSO₄ production with a characteristically low δ³⁴S value. Based on the current understanding of the isotope effects of major SO₂ oxidation pathways, the model appears to underestimate transition metal ion chemistry significantly during the summer months. This discrepancy is likely due to the absence of SO₂ heterogeneous reactions in the model chemistry. Overall, our findings refine air quality model simulations of pSO₄ chemistry, potentially leading to more accurate predictions of the impacts of future emission changes on its atmospheric burden.