Implementing constrained multi-time approach with bootstrap analysis in ME-2: An application to PM2.5 data from...

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Chemical Engineering

Background:

• Traditional receptor models struggle with rotational ambiguity and limited source identification.
• Existing methods often fail to fully utilize aerosol compositional data with varying temporal resolutions.
• Accurate source apportionment of particulate matter is crucial for environmental and public health management.

Purpose of the Study:

• To enhance the multi-time receptor model by incorporating constraints for improved source identification.
• To implement a bootstrap technique for quantifying the uncertainty of constrained source apportionment solutions.
• To evaluate the performance of the advanced multi-time model using a comprehensive PM2.5 dataset.

Main Methods:

• Development and application of a constrained multi-time factor analysis model.
• Integration of aerosol compositional data with different time resolutions (hourly and daily).
• Implementation of a bootstrap resampling technique for uncertainty estimation.
• Chemical characterization of PM2.5 samples for elements, ions, and carbonaceous components.

Main Results:

• The advanced model successfully identified major PM2.5 sources, with traffic contributing 37% annually.
• Specific traffic emission processes, including exhaust and non-exhaust, were accurately characterized.
• Constrained model runs improved the identification of nitrates and biomass burning profiles.
• Bootstrap analysis provided reliable estimates of the uncertainty in source apportionment solutions.

Conclusions:

• The enhanced multi-time receptor model offers superior source identification and characterization capabilities.
• Integrating data of different time resolutions provides significant advantages over traditional methods.
• The model's ability to constrain factors and estimate uncertainty enhances the reliability of source apportionment results.