A method to quantitatively apportion pollutants at high spatial and temporal resolution: the Stochastic Lagrangian...

Area of Science:

• Atmospheric chemistry and physics
• Environmental science
• Computational modeling

Background:

• Accurate source apportionment is crucial for understanding and mitigating air pollution.
• Existing methods may struggle with complex atmospheric processes like dispersion and chemical transformation.
• Quantifying contributions from various sources is essential for effective environmental policy.

Purpose of the Study:

• To introduce and validate the Stochastic Lagrangian Apportionment Method (SLAM) for quantifying upwind pollutant contributions.
• To apply SLAM for source apportionment of ammonia (NH3) and ammonium particulates (p-NH4(+)) in southern Ontario.
• To compare SLAM results with a traditional 'brute force' method (BFM).

Main Methods:

• Utilizing a time-reversed Lagrangian particle dispersion model.
• Simulating chemical changes forward in time while tracking distinct sources.
• Incorporating turbulent dispersion, chemical transformations, and depositional losses.
• Minimizing numerical diffusion for accurate source localization.

Main Results:

• Agricultural emissions were identified as the dominant source of NH3 and p-NH4(+) in southern Ontario.
• The source region for NH3 was localized, while p-NH4(+) originated from a wider area, including the American Midwest.
• SLAM results closely matched BFM for NH3 but showed higher p-NH4(+) concentrations, suggesting improved capture of indirect effects.

Conclusions:

• SLAM provides a robust method for atmospheric pollutant source apportionment, accounting for complex physical and chemical processes.
• The study highlights the significant impact of agricultural emissions on air quality in the region.
• Future work may involve integrating SLAM with inverse methods to further refine uncertainty quantification.