Assessing the Sources of Submicron Airborne Elements at two sites in the Fos-Marseille Basin through Rolling Positive Matrix Factorization

Brezins, Mathilde; Chazeau, Benjamin; Marchand, Nicolas; Durand, Amandine; Gille, Grégory; Bourjot, Romain; Prévôt, Andre S. H.; Jaffrezo, Jean-Luc; Uzu, Gaëlle; D'Anna, Barbara

doi:10.5194/ar-2026-5

Preprints

https://doi.org/10.5194/ar-2026-5

Preprints

11 Feb 2026

| 11 Feb 2026

Status: this preprint is currently under review for the journal AR.

Assessing the Sources of Submicron Airborne Elements at two sites in the Fos-Marseille Basin through Rolling Positive Matrix Factorization

Mathilde Brezins, Benjamin Chazeau, Nicolas Marchand, Amandine Durand, Grégory Gille, Romain Bourjot, Andre S. H. Prévôt, Jean-Luc Jaffrezo, Gaëlle Uzu, and Barbara D'Anna

Abstract. Contributions and evolution of fine elemental particulate matter (PM) sources were investigated in the Marseille-Fos basin (South of France) based on a 1 year-long (January–December 2023) study using on-line x-ray fluorescence (Xact) PM₁ measurements. The region's intense anthropogenic activity and complex meteorological conditions make it an ideal case study for fine aerosol characterization. Given the limited information available on fine elemental sources in the area, a dual-site approach was implemented, combining an urban background site (MRS-LCP) and an industrial site (FOS) to distinguish between regional and local emission influences. Source apportionment was conducted using a rolling Positive Matrix Factorization (PMF) method, implemented via the Source Finder Professional (SoFi) toolkit. Several tests were carried out to determine optimal rolling PMF parameters. Eventually, a 21-day rolling window configuration was selected, resolving nine factors at FOS and eight at MRS-LCP, with seven similar factors detected at both sites. Among them, three were attributed to secondary aerosols, including sulfur photooxidation leading to sulfate-rich aerosols (S-rich factor) and the formation of halogenated reactive particulate species (Cl-rich and Br-rich factors). Additionally, biomass burning, shipping, and dust related factors were identified at both locations. In contrast, three industrial factors (Steel Industry, Zn-Industrial, Pb-Industrial) were detected at FOS, while only the Steel Industry factor appeared at MRS-LCP, suggesting downwind transport of industrial plumes from Fos-sur-Mer to Marseille under Mistral and thermal breeze regimes. Furthermore, the comparison of dynamic rolling PMF approach to static PMF analysis, demonstrated higher dissimilarities across factors profiles, reflecting enhanced ability of Rolling PMF to capture seasonal variability in aerosol sources. Overall, this study highlights the dominant anthropogenic imprint on submicron PM elements and the effectiveness of dynamic source apportionment in complex coastal-industrial environments.

Received: 30 Jan 2026 – Discussion started: 11 Feb 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 1273 KB)

Supplement (4716 KB)

Download & links

Mathilde Brezins, Benjamin Chazeau, Nicolas Marchand, Amandine Durand, Grégory Gille, Romain Bourjot, Andre S. H. Prévôt, Jean-Luc Jaffrezo, Gaëlle Uzu, and Barbara D'Anna

Status: final response (author comments only)

RC1:
'Comment on ar-2026-5', Anonymous Referee #1, 11 Mar 2026
Comments on the manuscript of “Assessing the Sources of Submicron Airborne Elements at two sites in the Fos-Marseille Basin through Rolling Positive Matrix Factorization” by Mathilde et al.

General comments
This manuscript investigates the sources of submicron airborne elements at an urban background site and an industrial site in the Fos-Marseille Basin using one-year high-time-resolution Xact measurements and rolling PMF. The study is well-designed, data-rich, and methodologically sound, with a clear novelty in applying the rolling PMF approach to long-term elemental XRF data. The dual-site setting enables the identification of regional and industrial contributions, and the major sources (secondary aerosols, shipping, biomass burning, steel industry, etc.) are reasonably interpreted. The manuscript is generally suitable for publication in Aerosol Research after minor revision, as several key scientific clarifications, data validation, and presentation improvements are required to strengthen the robustness of source apportionment and readability.

Specific comments
Line 195: Are there any references indicating the shortcomings of static PMF in capturing the seasonal variations in long-term data? If so, please suggest the relevant literature.

When comparing the static and Rolling PMF, only the variability of the factor profiles was analyzed, without comparing the differences in the quantitative source contribution concentrations between the two methods (such as the annual average contribution and seasonal contribution differences of each source). It is suggested to supplement the relevant quantitative comparisons to more comprehensively demonstrate the advantages of Rolling PMF.

Does the selection method for high BDL (below MDLs) element data affect the construction of the error matrix, thereby influencing the source apportionment results of PMF? Is it necessary to compare the PMF factor results after removing or retaining the high BDL elements?

In Section 4.2, the analysis of each PMF factor reveals that the Zn-rich factor (MRS-LCP site) has excluded traffic and industrial sources, and it is hypothesized that solid fuel combustion is the main source. However, there is a lack of quantitative coupling analysis with biomass burning organic aerosols (BBOA), such as concentration correlation and ratio characteristics. It is recommended to supplement relevant statistical analysis to strengthen the basis for attributing the source of this factor.

It is suggested to compare the correlation between the HOA factors of TOF-ACSM OA PMF and the Steel Industry and Zn-rich factors at MRS-LCP, in order to further determine whether road transport may contribute to the Steel Industry and Zn-rich factors at MRS-LCP.

The Cl‑rich factor from Xact and Cl^- from ACSM represent the same submicron particulate chlorine. A correlation analysis is suggested to validate the factor.

The extremely high contribution of the S-rich factor (>50%) is reasonable given strong photochemical sulfate formation in this photochemically active, industrial, and coastal Mediterranean basin, but the authors should further quantify SO₂/sulfate sources (shipping, industry, etc.) to improve interpretation.

Technical corrections
Line 114 and 119: Please ensure the consistency of punctuation and formatting in the writing of latitude and longitude values (43°27’32.092’’N;120 4°56’4.412’’E).

Is the “X” in Table 1 indicating that the instrument was equipped at this site?

Line 157-158: Excessive commas in the “The lack of quantitative assessment … on marine aerosols,” and the citation format of the literature needs to be adjusted in “(e.g. fresh and aged sea salt, Pey et al., 2013a)”.

Line 214: Fig. S3 & S4 should be corrected to Figs. S3 & S4. All similar situations in the text should be uniformly revised.

Line 311: “eg.” should be “e.g.”.

In the sentence “PM₂.₅ in Sylvestre et al., 2017 versus PM₁ in the present study”, “Sylvestre et al., 2017” should be “Sylvestre et al. (2017)”.
Citation: https://doi.org/10.5194/ar-2026-5-RC1
RC2: 'Comment on ar-2026-5', Anonymous Referee #2, 13 Apr 2026

Regarding the manuscript entitled “Assessing the Sources of Submicron Airborne Elements at two sites in the Fos-Marseille Basin through Rolling Positive Matrix Factorization,” I find the study to be very well written and scientifically interesting. The manuscript focuses on a highly relevant and complex region, and the dual-site design provides valuable insight into the spatial variability of submicron elemental sources. In particular, the work is noteworthy because it applies a methodological approach that has not been implemented before on a year-long, high-time-resolution elemental dataset. Overall, the study makes a valuable contribution to the understanding of airborne particulate sources in this important coastal-industrial region and can be published after minor revisions.
Lines 214–223: The selection of elements retained in the PMF input matrix deserves a more detailed discussion in relation to factor identification. Since several elements were kept despite very high fractions below detection limit, the authors should clarify how their inclusion affected the robustness and interpretability of the resolved factors, especially for industrial factors driven by low-abundance tracers.
Lines 230–233: The exclusion of short-lived events may have important consequences for factor identification. Since episodic events such as Sirocco outbreaks, fireworks, or local construction emissions can carry distinct elemental fingerprints, removing them may simplify the factor structure but may also suppress potentially meaningful sources or alter mixed factors. This effect should be discussed more explicitly. Why have the authors not reintroduced these points at the end of the analysis and add extra factors?
Lines 254–264: The fact that the Shipping and Cl-rich factors could not always be identified in unconstrained seasonal runs raises an important question about factor stability. The manuscript would benefit from a clearer explanation of how confidently these factors can be considered physically distinct sources rather than partially resolved or constraint-dependent solutions.
Lines 265–269: The decision to retain a Biomass Burning factor throughout the year is understandable, but the manuscript should discuss more carefully whether this reflects a true year-round source or whether, during summer, potassium may partly be redistributed from other sources into this factor. This is especially important for factor identification in periods when independent biomass-burning tracers are weaker.
Lines 355–356: The observation that the Zn-rich factor may reflect multiple sources is important, but the discussion should go further. If this factor is not chemically homogeneous, the authors should clarify whether it should really be interpreted as a single source, or rather as a mixed factor grouping several Zn-containing emissions.

Citation: https://doi.org/10.5194/ar-2026-5-RC2

Mathilde Brezins, Benjamin Chazeau, Nicolas Marchand, Amandine Durand, Grégory Gille, Romain Bourjot, Andre S. H. Prévôt, Jean-Luc Jaffrezo, Gaëlle Uzu, and Barbara D'Anna

Supplement

https://doi.org/10.5194/ar-2026-5-supplement

Mathilde Brezins, Benjamin Chazeau, Nicolas Marchand, Amandine Durand, Grégory Gille, Romain Bourjot, Andre S. H. Prévôt, Jean-Luc Jaffrezo, Gaëlle Uzu, and Barbara D'Anna

Viewed

Total article views: 301 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
194	88	19	301	45	11	21

HTML: 194
PDF: 88
XML: 19
Total: 301
Supplement: 45
BibTeX: 11
EndNote: 21

Views and downloads (calculated since 11 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	98	35	11	144
Mar 2026	86	41	7	134
Apr 2026	10	12	1	23

Cumulative views and downloads (calculated since 11 Feb 2026)

Month	HTML	PDF	XML	Total
Feb 2026	98	35	11	144
Mar 2026	86	41	7	134
Apr 2026	10	12	1	23

Viewed (geographical distribution)

Total article views: 306 (including HTML, PDF, and XML) Thereof 306 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 15 Apr 2026

Download

Preprint (1273 KB)
Metadata XML

Short summary

The Marseille-Fos basin faces high anthropogenic pressure from industry, maritime and road transport, combined with specific weather conditions that further degrade air quality. Our study focuses on fine metallic pollution, which can penetrate deep into the lungs and cause harmful effects. Using one year of measurements at two sites, we identified ten main pollution sources, half directly linked to human activities, highlighting clear risks for the environment and public health.


Total:	0
HTML:	0
PDF:	0
XML:	0