Effectiveness of Emission Controls on Atmospheric Oxidation and Air Pollutant Concentrations: Uncertainties due to Chemical Mechanisms and Inventories

Mingjie Kang, Hongliang Zhang, Qi Ying2025 · DOI

Abstract

In this study, three photochemical mechanisms of varying complexity from the Statewide Air Pollution Research Center (SAPRC) family and two widely used anthropogenic emission inventories are employed to quantify the discrepancies in the predicted effectiveness of nitrogen oxides ($NO_x$) and volatile organic compound (VOC) emission controls on ozone ($O_3$), secondary inorganic aerosols (SIA), and hydroxyl ($OH$) and nitrate ($NO_3$) radicals using the Community Multiscale Air Quality (CMAQ) model. For maximum daily average 8-hour $O_3$ ($O_3$-8 h), relative reductions predicted using different emission inventory and mechanism combinations are consistent for up to 80% $NO_x$ or VOC reductions, with maximum differences of approximately 15%. For secondary inorganic aerosols (SIA), while the predicted relative changes in their daily average concentrations due to $NO_x$ reductions are quite similar, very large differences of up to 30% occur for VOC reductions. Sometimes even the direction of change is different. For the oxidants $OH$ and $NO_3$ radicals, the uncertainties in the relative changes due to emission changes are even larger among different inventory-mechanism combinations, sometimes by as much as 200%. Our results suggest that while the $O_3$-8 h responses to emission changes are not sensitive to the choice of chemical mechanism and emission inventories, using a single model and mechanism to evaluate the effectiveness of emission controls on SIA and atmospheric oxidation capacity may have large errors. For these species, the evaluation of the control strategies may require an ensemble approach with multiple inventories and mechanisms.

Introduction

Air quality management relies heavily on Chemical Transport Models (CTMs) to simulate the impact of emission control strategies. Policymakers use these simulations to determine the necessary reductions in nitrogen oxides ($NO_x$) and volatile organic compounds (VOCs) to meet air quality standards. However, the internal "recipes" of these models—specifically the chemical mechanisms and emission inventories—contain inherent uncertainties. This study investigates how these uncertainties affect the predicted effectiveness of pollution controls in major Chinese megacities.

Methodology

The researchers utilized the Community Multiscale Air Quality (CMAQ) model (version 5.0.2) to conduct sensitivity tests across five major urban centers in China, including Beijing and Shanghai. The study focused on two primary variables:

  1. Chemical Mechanisms: Three versions of the SAPRC mechanism (CS07, S11, and S18) were compared, representing a range of chemical complexity.
  2. Emission Inventories: The Multi-resolution Emission Inventory for China (MEIC, 2017) and the Regional Emission inventory in ASia (REAS 3.1, 2015) were evaluated.

The team simulated emission reductions ranging from 20% to 80% for both $NO_x$ and VOCs to observe the resulting changes in ozone ($O_3$), secondary inorganic aerosols (SIA), and key atmospheric oxidants ($OH$ and $NO_3$ radicals).

Key Findings

Robustness of Ozone Predictions

A significant finding is that the response of 8-hour ozone ($O_3$-8h) to emission reductions is remarkably consistent across different model configurations. Regardless of the chemical mechanism or emission inventory used, the predicted relative reductions in ozone remained within a 15% margin of error. This suggests that current modeling frameworks are reliable for designing ozone-specific control strategies.

Uncertainty in Secondary Inorganic Aerosols (SIA)

While predictions for SIA response to $NO_x$ cuts were relatively stable, the response to VOC reductions showed significant discrepancies. Differences of up to 30% were observed between different model combinations, and in some instances, the models even disagreed on whether SIA concentrations would increase or decrease. This variability is largely attributed to how different chemical mechanisms simulate the interaction between the $NO_3$ radical and organic compounds.

Atmospheric Oxidation Capacity

The most substantial uncertainties were found in the response of atmospheric "detergents" — the $OH$ and $NO_3$ radicals. These radicals are responsible for the atmosphere's self-cleaning capacity. The study revealed discrepancies of up to 200% in how these radicals respond to emission changes depending on the inventory-mechanism pair. Such high uncertainty is critical, as these radicals influence the formation of various other pollutants.

Conclusion and Policy Implications

The study highlights a critical limitation in current air quality modeling: while ozone predictions are robust, the effectiveness of controls on SIA and atmospheric oxidation capacity is highly sensitive to model configuration. Relying on a single model setup for multi-pollutant control strategies carries significant risk.

To provide more reliable guidance for complex pollutants, the authors advocate for an "ensemble" modeling approach. By utilizing multiple chemical mechanisms and emission inventories, researchers can provide a range of possible outcomes, allowing policymakers to account for scientific uncertainty when designing environmental regulations.

Summary updated

Abstract

In this study, three photochemical mechanisms of varying complexity from the Statewide Air Pollution Research Center (SAPRC) family and two widely used anthropogenic emission inventories are employed to quantify the discrepancies in the predicted effectiveness of nitrogen oxides (NOx) and volatile organic compound (VOC) emission controls on ozone (O3), secondary inorganic aerosols (SIA), and hydroxyl (OH) and nitrate (NO3) radicals using the Community Multiscale Air Quality (CMAQ) model. For maximum daily average 8-hour O3 (O3-8 h), relative reductions predicted using different emission inventory and mechanism combinations are consistent for up to 80 % NOx or VOC reductions, with maximum differences of approximately 15 %. For secondary inorganic aerosols (SIA), while the predicted relative changes in their daily average concentrations due to NOx reductions are quite similar, very large differences of up to 30 % occur for VOC reductions. Sometimes even the direction of change (i.e., increase or decrease) is different. For the oxidants OH and NO3 radicals, the uncertainties in the relative changes due to emission changes are even larger among different inventory-mechanism combinations, sometimes by as much as 200 %. Our results suggest that while the O3-8 h responses to emission changes are not sensitive to the choice of chemical mechanism and emission inventories, using a single model and mechanism to evaluate the effectiveness of emission controls on SIA and atmospheric oxidation capacity may have large errors. For these species, the evaluation of the control strategies may require an ensemble approach with multiple inventories and mechanisms.
BibTeX
@misc{kang2025effectiveness,
  author = {Mingjie Kang and Hongliang Zhang and Qi Ying},
  title = {{Effectiveness of Emission Controls on Atmospheric Oxidation and Air Pollutant Concentrations: Uncertainties due to Chemical Mechanisms and Inventories}},
  year = {2025},
  doi = {10.5194/egusphere-2025-255},
}
Figure