3,- 1University of Patras, Chemical Engineering, Patras, Greece
- 2Institute of Chemical Engineering Sciences, Foundation for Research and Technology– Hellas (FORTH/ICE-HT), Patras, Greece
- 3Institute of Environmental Assessment and Water Research (IDAEA-CSIC), Barcelona, Spain
- 4Department of Applied Physics-Meteorology, University of Barcelona, Barcelona, Spain
Atmospheric aerosols play a critical role in air quality, human health, and climate. Ultrafine particles (UFPs) are of particular concern due to their strong health impact and their dominant contribution to particle number concentrations. While receptor models such as Positive Matrix Factorization (PMF) are widely used to estimate the contribution of various sources to aerosol mass, their application to aerosol number remains challenging. In this work, we apply PMF to the number size distributions predicted by a three-dimensional chemical transport model. The approach used is the same as that used for the PMF analysis of measurements, but in this case the true source contributions are known and the PMF method results can be evaluated.
PMCAMx-UF is a three-dimensional chemical transport model that simulates aerosol number and mass distributions by explicitly resolving key atmospheric processes, including advection, deposition, gas-phase chemistry, nucleation and coagulation. The model is applied over Europe at 36 x 36 km resolution, with increasing resolution over Athens where a 1 x 1 km grid is used. The aerosol number distribution is described using 42 sections. The contribution of the various sources to aerosol number according to PMCAMx-UF is quantified using the approach of Posner and Pandis (2015) for a summer and a winter month. Positive Matrix Factorization (PMF) was also used to apportion sources of particle number size by decomposing the PMCAMx-UF simulated size distributions into factors and calculating their time-resolved contributions.
PMF seriously underestimated the contribution of new particle formation to particles larger than 10 nm in Athens during the summer, N10. PMF estimated that 25% of N10 was due to new particle formation, while this process was actually responsible for 62% of the N10. At the same time, PMF overestimates the contribution of traffic-related sources (57% compared to 13%). During winter, PMF does a reasonable job quantifying the role of new particle formation (17% versus the correct 22%) but still overestimates the role of traffic (71% compared to 34%).
Posner, L. N., & Pandis, S. N. (2015). Sources of ultrafine particles in the Eastern United States. Atmospheric Environment, 111, 103–112. https://doi.org/10.1016/j.atmosenv.2015.03.033
How to cite: Poulikidi, E., Garcia-Marles, M., Siouti, E., Patoulias, D., Querol, X., and Pandis, S.: Evaluation of Positive Matrix Factorization for the Source Apportionment of Particle Number, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11000, https://doi.org/10.5194/egusphere-egu26-11000, 2026.
