Measurement report: A one-year study to estimate maritime contributions to PM<sub>10</sub> in a coastal area in Northern France

Ledoux, Frédéric; Roche, Cloé; Delmaire, Gilles; Roussel, Gilles; Favez, Olivier; Fadel, Marc; Courcot, Dominique

doi:https://doi.org/10.5194/egusphere-2023-169

Preprints

https://doi.org/10.5194/egusphere-2023-169

Preprints

24 Feb 2023

| 24 Feb 2023

Measurement report: A one-year study to estimate maritime contributions to PM₁₀ in a coastal area in Northern France

Frédéric Ledoux, Cloé Roche, Gilles Delmaire, Gilles Roussel, Olivier Favez, Marc Fadel, and Dominique Courcot

Abstract. This work is focused on filling the lack of knowledge associated with natural and anthropogenic marine emissions on PM₁₀ concentrations in Northern France. For this purpose, a one-year measurement and sampling campaign for PM₁₀ has been performed at a French coastal site situated in front of the Straits of Dover. The characterization of PM₁₀ samples was performed considering major and trace elements, water-soluble ions, organic carbon (OC), elemental carbon (EC), and organic markers of biomass burning and primary biogenic emissions. Furthermore, the source apportionment of PM₁₀ was achieved using the constrained weighted-non-negative matrix factorization (CW-NMF) model. The annual average PM₁₀ was 24.3 µg/m³ with six species contributing to 69 % of its mass (NO₃-, OC, SO₄^2-, Cl^-, Na⁺, and NH₄⁺). The source apportionment of PM₁₀ led to the identification of 10 sources. Fresh and aged sea-salts contributed to 37 % of PM₁₀, while secondary nitrate and sulfate contributed 41 %, biomass burning 10 %, and Heavy Fuel Oil (HFO) combustion from shipping emissions contributed 5 %, on yearly averages. Additionally, monthly evolution of the sources’ contribution evidenced different behaviors with high contributions of secondary nitrate and biomass burning during winter. In the summer season, 10 times higher concentrations for HFO combustion (July compared to January) and the predominance of aged sea-salts versus fresh sea-salts were observed. Constant weighted trajectories showed that the sources contributing to more than 80 % of PM₁₀ at Cape Gris-Nez are of regional and/or long-range origins with the North Sea and the English Channel as hotspots for natural and anthropogenic marine emissions and Belgium, the Netherlands, and the West of Germany as hotspots for secondary inorganic aerosols.

Received: 03 Feb 2023 – Discussion started: 24 Feb 2023

Download & links

Preprint (PDF, 1339 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (1339 KB)

Supplement (1163 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

04 Aug 2023

Measurement report: A 1-year study to estimate maritime contributions to PM₁₀ in a coastal area in northern France

Frédéric Ledoux, Cloé Roche, Gilles Delmaire, Gilles Roussel, Olivier Favez, Marc Fadel, and Dominique Courcot

Atmos. Chem. Phys., 23, 8607–8622, https://doi.org/10.5194/acp-23-8607-2023,https://doi.org/10.5194/acp-23-8607-2023, 2023

Short summary

Frédéric Ledoux et al.

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-169', Anonymous Referee #1, 19 Mar 2023

General comment

The paper reports a study of maritime contributions to PM10 in a rural coastal area of northern France. The topic is of interest and suitable for the Journal. The approach used is quite standard, however, the extensive dataset, the detailed statistical analysis, and the limited availability of data for this area make the paper useful to the scientific community. I would like to see it published after a revision step addressing my minor specific comments.

Specific comments

Please remove all instances of etc. in the paper. If something should be added, please do it explicitly.

Line 67. Please remove the “+”. In addition, this value seems to be very large compared to the measured ones and to the values published in other studies. What it represents Annual average or other short-term contribution?

Lines 77-80. It would be worth to mention here that the use of low-sulphur fuels actually reduce also primary PM emissions from ships how it has been demonstrated in several study and, of course, also secondary sulphate. So that the regulation was not done only for SO2.

Line 101. Are these quartz filters?

Line 172. Are you referring to secondary organic aerosol here?

Section 3.2. Considering that the receptor model used is still not widely applied, it should be useful to mention that it has comparable performances to the PMF model investigated in the recent work of Belis et al. (Atmospheric Environment: X, 5, 2020, 100053).

Lines 247-256. For the discussion on the V/Ni ratio, I suggest to have a look and mention the work of Gregoris et al (Environ Sci Pollut Res (2016) 23, 6951–6959) that shows ratios lower than expected in coastal areas with relevant impact of shipping as well as a strong spatial variability of this ratio.

Lines 293-295. I would not say that it is underestimated. The pint is that it was analysed here only the contribution to primary PM10 and that there would be also a contribution to secondary aerosol, mainly sulphate, that could be even larger than the primary one.

Lines 322-325. Please mention how and where these gases are measured.

Figure 4. Please use the same acronym as in the text (i.e. CBPF). To be honest, I do not understand why the CBPF of sea salt and aged sea salt are so different. A better interpretation of this aspect would be useful.

Figure S4. The use of three different level of blue is not a good choice because make the figure hardly readable generating confusions among the different sources. Please make a different choice for these colours.

Citation: https://doi.org/10.5194/egusphere-2023-169-RC1
- AC1: 'Reply on RC1', Marc Fadel, 28 Mar 2023
  
  We would like to thank the referee for his constructive comments that aim at improving the manuscript. In the attached document, we have replied to the different comments. Amendments in the manuscript will be done once the final version uploaded.
  
  Citation: https://doi.org/10.5194/egusphere-2023-169-AC1
RC2:
'Comment on egusphere-2023-169', Anonymous Referee #2, 24 Apr 2023

The manuscript reports a standard source apportionment study with a focus on the maritime source. Beyond the original results and the use of a specific receptor site to achieve the goals, the rest of the study presents only the trivial application of well-established source apportionment techniques (PMF-like, CWT, CPF). Another weakness of this study is the "age" of the data, which were collected 10 years ago. However, since the lack of papers focusing on maritime sources, I am in favour of publication. The authors are requested to improve the manuscript by adding more information about the outcomes of the CW-NMF model (in particular about the number of factors extracted).

Specific comments.
Lines 76-79 “Additionally, despite the IMO regulation for global sulfur limit of 0.5% from ship’s fuel oil applied starting January 2020, different countries are still adopting higher sulfur limits. It is worth noting that these limits were only set for sulfur content in marine fuels in order to reduce SO2 emissions but no regulations for PM components neither in the sea nor at ports were issued”. Not clear why this consideration is reported within the objectives of the study. Is this topic addressed in the current study? Please clarify.
Lines 66-80. Please comment on the choice of sampling PM2.5 to quantify the maritime contribution. Combustion emissions are expected to be finer than other sources. On the contrary, other sources in the coastal areas (e.g., sea salt) may emit large particles. Thus, it would be better to analyze PM2.5. Please comment to support your choice.
Line 90. “The sampling site is far from major continental pollution sources and can be considered as a background site.” Which background? Rural, urban, suburban? If rural, are you sure the site has the characteristics to be classified in this way? Please comment.
Section 2.2. The sampling campaign is 1 year long (1st of January to 31st of December 2013) and samples were collected at daily frequency, however only 122 samples were analyzed. Why? How the samples to be analyzed have been selected? Randomly? 1 day over 3? Please explain.
Line 100. Was the beta monitor tested against the gravimetric measurement? Which calibration audit? Please add info.
Line 168 “The yearly-mean PM10 concentration obtained for the set of sampling days was of 24.3 μg/m3”. Is this the real yearly average (365 days in a year possibly measured with the beta monitor) or the average over the 122 analyzed samples? It would be great to see both values to understand if the sample selection is consistent with (indicative of, or similar to) the year average.
Figure 2. Please report the error bars relative to the bootstrap results. This would clarify some source profiles.
Section 3.2. The description of the model setup lacks many details. In addition, it looks like you pushed the model beyond the limits of what it is capable of. It is hard to believe that you can see so many factors with just 122 samples and 28 variables without overfitting the data. For instance, the presence of a “metal-rich” factor may be due to the overfit. Thus, I would ask you to add the scaled residual plots to the SI material file along with all the diagnostics returned by the model. This is to better investigate the goodness of the selected model setup.
Line 227. “with an average Cl--to-Na+ ratio of 1.8 which is commonly observed for fresh sea salts (Seinfeld and Pandis, 2016).” Is the seawater composition also valid for other fresh sea salt ionic species such as K+, Mg2+, Ca2+, and SO42-?
Line 230. Is the cations/anions ratio balanced in this factor?
Line 240-255. The authors report some literature data in support of their findings. However, most of these studies refer to finer PM (PM2.5 or even PM1). For example, the paper by Khan et al (2021) reports the EC to levoglucosan ratio in PM1. The paper by Salameh et al (2018) refers to the OC-to-EC ratio of PM2.5. It is not reliable that the ratios between different variables (chemical species) remain unchanged on PM10, PM2.5 or PM1. At least, not all. Please comment or change the references.
Figure 3. Please provide the uncertainty associated with the results.
Section 3.5. It appears that most of the factors come from the same area of origin. This again pinpoints the possible data overfit. However, the results for HFO appear to come from a more "marine" source area than the two secondary factors. It would be better to zoom in on Figure 5 to better visualize the differences. After all, we are not interested in results where the model returns too low a contribution (thus, please cut the more distant areas).

Citation: https://doi.org/10.5194/egusphere-2023-169-RC2
- AC2: 'Reply on RC2', Marc Fadel, 10 Jun 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-169/egusphere-2023-169-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-169-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-169', Anonymous Referee #1, 19 Mar 2023

General comment

The paper reports a study of maritime contributions to PM10 in a rural coastal area of northern France. The topic is of interest and suitable for the Journal. The approach used is quite standard, however, the extensive dataset, the detailed statistical analysis, and the limited availability of data for this area make the paper useful to the scientific community. I would like to see it published after a revision step addressing my minor specific comments.

Specific comments

Please remove all instances of etc. in the paper. If something should be added, please do it explicitly.

Line 67. Please remove the “+”. In addition, this value seems to be very large compared to the measured ones and to the values published in other studies. What it represents Annual average or other short-term contribution?

Lines 77-80. It would be worth to mention here that the use of low-sulphur fuels actually reduce also primary PM emissions from ships how it has been demonstrated in several study and, of course, also secondary sulphate. So that the regulation was not done only for SO2.

Line 101. Are these quartz filters?

Line 172. Are you referring to secondary organic aerosol here?

Section 3.2. Considering that the receptor model used is still not widely applied, it should be useful to mention that it has comparable performances to the PMF model investigated in the recent work of Belis et al. (Atmospheric Environment: X, 5, 2020, 100053).

Lines 247-256. For the discussion on the V/Ni ratio, I suggest to have a look and mention the work of Gregoris et al (Environ Sci Pollut Res (2016) 23, 6951–6959) that shows ratios lower than expected in coastal areas with relevant impact of shipping as well as a strong spatial variability of this ratio.

Lines 293-295. I would not say that it is underestimated. The pint is that it was analysed here only the contribution to primary PM10 and that there would be also a contribution to secondary aerosol, mainly sulphate, that could be even larger than the primary one.

Lines 322-325. Please mention how and where these gases are measured.

Figure 4. Please use the same acronym as in the text (i.e. CBPF). To be honest, I do not understand why the CBPF of sea salt and aged sea salt are so different. A better interpretation of this aspect would be useful.

Figure S4. The use of three different level of blue is not a good choice because make the figure hardly readable generating confusions among the different sources. Please make a different choice for these colours.

Citation: https://doi.org/10.5194/egusphere-2023-169-RC1
- AC1: 'Reply on RC1', Marc Fadel, 28 Mar 2023
  
  We would like to thank the referee for his constructive comments that aim at improving the manuscript. In the attached document, we have replied to the different comments. Amendments in the manuscript will be done once the final version uploaded.
  
  Citation: https://doi.org/10.5194/egusphere-2023-169-AC1
RC2:
'Comment on egusphere-2023-169', Anonymous Referee #2, 24 Apr 2023

The manuscript reports a standard source apportionment study with a focus on the maritime source. Beyond the original results and the use of a specific receptor site to achieve the goals, the rest of the study presents only the trivial application of well-established source apportionment techniques (PMF-like, CWT, CPF). Another weakness of this study is the "age" of the data, which were collected 10 years ago. However, since the lack of papers focusing on maritime sources, I am in favour of publication. The authors are requested to improve the manuscript by adding more information about the outcomes of the CW-NMF model (in particular about the number of factors extracted).

Specific comments.
Lines 76-79 “Additionally, despite the IMO regulation for global sulfur limit of 0.5% from ship’s fuel oil applied starting January 2020, different countries are still adopting higher sulfur limits. It is worth noting that these limits were only set for sulfur content in marine fuels in order to reduce SO2 emissions but no regulations for PM components neither in the sea nor at ports were issued”. Not clear why this consideration is reported within the objectives of the study. Is this topic addressed in the current study? Please clarify.
Lines 66-80. Please comment on the choice of sampling PM2.5 to quantify the maritime contribution. Combustion emissions are expected to be finer than other sources. On the contrary, other sources in the coastal areas (e.g., sea salt) may emit large particles. Thus, it would be better to analyze PM2.5. Please comment to support your choice.
Line 90. “The sampling site is far from major continental pollution sources and can be considered as a background site.” Which background? Rural, urban, suburban? If rural, are you sure the site has the characteristics to be classified in this way? Please comment.
Section 2.2. The sampling campaign is 1 year long (1st of January to 31st of December 2013) and samples were collected at daily frequency, however only 122 samples were analyzed. Why? How the samples to be analyzed have been selected? Randomly? 1 day over 3? Please explain.
Line 100. Was the beta monitor tested against the gravimetric measurement? Which calibration audit? Please add info.
Line 168 “The yearly-mean PM10 concentration obtained for the set of sampling days was of 24.3 μg/m3”. Is this the real yearly average (365 days in a year possibly measured with the beta monitor) or the average over the 122 analyzed samples? It would be great to see both values to understand if the sample selection is consistent with (indicative of, or similar to) the year average.
Figure 2. Please report the error bars relative to the bootstrap results. This would clarify some source profiles.
Section 3.2. The description of the model setup lacks many details. In addition, it looks like you pushed the model beyond the limits of what it is capable of. It is hard to believe that you can see so many factors with just 122 samples and 28 variables without overfitting the data. For instance, the presence of a “metal-rich” factor may be due to the overfit. Thus, I would ask you to add the scaled residual plots to the SI material file along with all the diagnostics returned by the model. This is to better investigate the goodness of the selected model setup.
Line 227. “with an average Cl--to-Na+ ratio of 1.8 which is commonly observed for fresh sea salts (Seinfeld and Pandis, 2016).” Is the seawater composition also valid for other fresh sea salt ionic species such as K+, Mg2+, Ca2+, and SO42-?
Line 230. Is the cations/anions ratio balanced in this factor?
Line 240-255. The authors report some literature data in support of their findings. However, most of these studies refer to finer PM (PM2.5 or even PM1). For example, the paper by Khan et al (2021) reports the EC to levoglucosan ratio in PM1. The paper by Salameh et al (2018) refers to the OC-to-EC ratio of PM2.5. It is not reliable that the ratios between different variables (chemical species) remain unchanged on PM10, PM2.5 or PM1. At least, not all. Please comment or change the references.
Figure 3. Please provide the uncertainty associated with the results.
Section 3.5. It appears that most of the factors come from the same area of origin. This again pinpoints the possible data overfit. However, the results for HFO appear to come from a more "marine" source area than the two secondary factors. It would be better to zoom in on Figure 5 to better visualize the differences. After all, we are not interested in results where the model returns too low a contribution (thus, please cut the more distant areas).

Citation: https://doi.org/10.5194/egusphere-2023-169-RC2
- AC2: 'Reply on RC2', Marc Fadel, 10 Jun 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-169/egusphere-2023-169-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-169-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

AR by Marc Fadel on behalf of the Authors (10 Jun 2023) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (29 Jun 2023) by Eleanor Browne

Dear Authors:
I am happy to accept this for publication following attention to the following minor comments, many of which involve adding information present in the referee response document to the main manuscript.

1) Referee 1 comment 10 “…I do not understand why the CBPF of sea salt and aged sea salt are so different…”: Please add clarifying comments to the text along the lines of that supplied in the response document.
2) Referee 2 comment 2 “Please comment on the choice of sampling PM2.5…”: Please add clarifying comments to the text along the lines of that supplied in the response document.
3) Referee 2 comment 6 “The yearly-mean PM10 concentrations…”: Please clarify in the text that the average of the entire year is 22.8 ug/m3 and that this supports the 122 samples is representative of the entire year on average.
4) Referee 2 comment 8 “The description of the model setup…”: On line 169 (track changes version) it states that 122 samples were used as the input data, however in the response it suggests additional samples from 2013 and 2014 were also included. Please clarify the input data. If the data from the exceedance days in 2013 was included in the input data, then a discussion on if including exceedance days alters the model outputs is warranted. One can imagine that having a biased sampling (by including extra exceedance days, but not non-exceedance days) would alter the results. The 2014 data would also need explanation if that was included as input.

Additionally, the bootstrap analysis would benefit from a few additional sentences of explanation (SI is ok) and the results of Table S1 require a few sentences explanation of the meaning of the table (where is the number 100 coming from?).

5) Referee 2 #9 “Is the seawater composition…”: Please add a sentence to the text saying that the ratios were also valid for other ionic species.
6) Referee 2 #10 “Is the cations/anions ratio balanced in this factor?”: Please clarify in the text that it has a ratio of 1.14.
7) Please ensure that the SI material is cited in order. Currently Fig. S3 is cited before Fig. S2.
8) Fig. S6: Is the 75th percentile of the January value for secondary sulfates off-scale? It is hard to tell with the current axis.

Hide

AR by Marc Fadel on behalf of the Authors (03 Jul 2023) Author's response Author's tracked changes Manuscript

ED: Publish as is (06 Jul 2023) by Eleanor Browne

AR by Marc Fadel on behalf of the Authors (07 Jul 2023) Manuscript

Journal article(s) based on this preprint

04 Aug 2023

Measurement report: A 1-year study to estimate maritime contributions to PM₁₀ in a coastal area in northern France

Frédéric Ledoux, Cloé Roche, Gilles Delmaire, Gilles Roussel, Olivier Favez, Marc Fadel, and Dominique Courcot

Atmos. Chem. Phys., 23, 8607–8622, https://doi.org/10.5194/acp-23-8607-2023,https://doi.org/10.5194/acp-23-8607-2023, 2023

Short summary

Frédéric Ledoux et al.

Supplement

https://doi.org/10.5194/egusphere-2023-169-supplement

Data sets

Data for "Measurement report: A one-year study to estimate maritime contributions to PM10 in a coastal area in Northern France." Frédéric Ledoux, Cloé Roche, Gilles Delmaire, Gilles Roussel, Olivier Favez, Marc Fadel, and Dominique Courcot https://doi.org/10.5281/zenodo.7664528

Frédéric Ledoux et al.

Viewed

Total article views: 616 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
467	132	17	616	40	5	11

HTML: 467
PDF: 132
XML: 17
Total: 616
Supplement: 40
BibTeX: 5
EndNote: 11

Views and downloads (calculated since 24 Feb 2023)

Month	HTML	PDF	XML	Total
Feb 2023	112	35	4	151
Mar 2023	184	41	7	232
Apr 2023	55	16	2	73
May 2023	29	12	0	41
Jun 2023	48	13	2	63
Jul 2023	37	13	2	52
Aug 2023	2	2	0	4

Cumulative views and downloads (calculated since 24 Feb 2023)

Month	HTML	PDF	XML	Total
Feb 2023	112	35	4	151
Mar 2023	184	41	7	232
Apr 2023	55	16	2	73
May 2023	29	12	0	41
Jun 2023	48	13	2	63
Jul 2023	37	13	2	52
Aug 2023	2	2	0	4

Viewed (geographical distribution)

Total article views: 613 (including HTML, PDF, and XML) Thereof 613 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 04 Aug 2023

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (1339 KB)
Metadata XML

Short summary

The objective of this work is to quantify the emissions resulting from the marine compartment in Northern France whether from natural or human-made sources. Therefore, a one-year PM₁₀ sampling campaign was conducted at a coastal French site. Results showed that sea-salts contributed 37 %, while secondary nitrate and sulfate contributed 41 %, biomass burning 10 %, and HFO combustion from shipping emissions 5 %. Sources contributing to more than 80 % of PM₁₀ are of regional and/or long-range origins.


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