Identification of PM sources by principal component analysis (PCA) coupled with wind direction data

Area of Science:

• Environmental Chemistry
• Atmospheric Science
• Geochemistry

Background:

• Particulate matter (PM) pollution poses significant environmental and health risks.
• Accurate identification of PM sources is crucial for developing effective air quality management strategies.

Purpose of the Study:

• To evaluate the combined effectiveness of principal component analysis (PCA), multi-linear regression (MLRA), and wind direction data for PM source apportionment.
• To identify and characterize distinct PM sources in a highly industrialized region of Northern Spain.

Main Methods:

• Analysis of PM data from three grain-size fractions.
• Application of PCA to identify independent PM sources.
• Integration of MLRA and wind direction data to determine source contributions and spatial distribution.

Main Results:

• Seven distinct PM sources were identified: steel and pigment manufacture, road dust, traffic exhaust, regional transport, crustal contributions, and sea spray.
• Wind direction data helped attribute crustal material to regional drainage flows.
• Motorway traffic was found to contribute 4-5 µg m⁻³ more PM10 than local traffic.

Conclusions:

• The combination of PCA, MLRA, and wind direction data is a powerful tool for detailed PM source identification and characterization.
• Understanding source contributions and locations is essential for designing targeted air pollution abatement strategies.