Prediction of source contributions to urban background PM<sub>10</sub> concentrations in European cities: a case study for an episode in December 2016 using EMEP/MSC-W rv4.15 and LOTOS-EUROS v2.0 – Part 1: The country contributions

Pommier, Matthieu; Fagerli, Hilde; Schulz, Michael; Valdebenito, Alvaro; Kranenburg, Richard; Schaap, Martijn

doi:https://doi.org/10.5194/gmd-13-1787-2020

Articles | Volume 13, issue 4

https://doi.org/10.5194/gmd-13-1787-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-13-1787-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 13, issue 4

Model evaluation paper

|

03 Apr 2020

Model evaluation paper |

| 03 Apr 2020

Prediction of source contributions to urban background PM₁₀ concentrations in European cities: a case study for an episode in December 2016 using EMEP/MSC-W rv4.15 and LOTOS-EUROS v2.0 – Part 1: The country contributions

Matthieu Pommier, Hilde Fagerli, Michael Schulz, Alvaro Valdebenito, Richard Kranenburg, and Martijn Schaap

Download

Final revised paper (published on 03 Apr 2020)
Supplement to the final revised paper
Preprint (discussion started on 28 May 2019)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of manuscript by Pommier et al.,', Anonymous Referee #1, 16 Jun 2019
- AC2: 'response to reviewer 1', Matthieu Pommier, 22 Sep 2019
SC1: 'Comment to gmd-2019-87', Philippe Thunis, 08 Jul 2019
- AC1: 'response to P. Thunis', Matthieu Pommier, 22 Sep 2019
RC2: 'Review of the paper “Prediction of source contributions to urban background PM10 concentrations in European cities: a case study for an episode in December 2016 - Part.1 The country contributions” by M. Pommier et al.', Anonymous Referee #2, 09 Jul 2019
- AC4: 'response to reviewer 2', Matthieu Pommier, 22 Sep 2019
SC2: 'comments about the article (Prediction of source contributions to urban background PM10 concentrations in European cities: a case study for an episode in December 2016 – Part.1 The country contributions)', Alain Clappier, 26 Jul 2019
- AC5: 'response to A. Clappier', Matthieu Pommier, 22 Sep 2019
SC3: 'Executive editor comment, code and data availability.', David Ham, 26 Jul 2019
- AC3: 'response to Editor', Matthieu Pommier, 22 Sep 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Matthieu Pommier on behalf of the Authors (23 Sep 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (05 Oct 2019) by Graham Mann

RR by Anonymous Referee #1 (09 Dec 2019)

RR by Anonymous Referee #2 (20 Dec 2019)

Suggestions for revision or reasons for rejection

The authors addressed many of the comments of the reviewers but in my opinion the most problematic ones are not answered sufficiently.

One of the issues I am still struggling with is the claim that there is good agreement between models. When I see the validation plots against the measurements with such biases and low least squares. Is it possible to plot the biases of both models against each other instead of comparing them both with the measurements? If both models consistently underestimate or overestimate at the same station locations I can assume that the different source contributions to the concentration are similar.

One could also wonder if the validation with measurements meaningful when comparing source contributions. Two models might disagree on the total concentration but agree on the share from a specific source region or sector. This would make them equally valuable for policy making. In the EEA air quality database some data are available about ‘Ammonium in PM2.5 (aerosol)’ (http://aidef.apps.eea.europa.eu/?source=%7B%22query%22%3A%7B%22bool%22%3A%7B%22must%22%3A%7B%22match_all%22%3A%7B%7D%7D%2C%22filter%22%3A%7B%22bool%22%3A%7B%22should%22%3A%5B%7B%22term%22%3A%7B%22Pollutant%22%3A%22Ammonium%20in%20PM2.5%20(aerosol)%22%7D%7D%5D%7D%7D%7D%7D%2C%22display_type%22%3A%22tabular%22%2C%22from%22%3A10%7D) Maybe it’s is more interesting to compare the ammonium share of PM2.5 for both models with the measurements. Unfortunately there aren’t many data for other PM fractions that can be compared with models.

The agreement metric is better explained in this version but I think there is a problem with the claims like “in average 68% of the hourly predicted PM10 concentrations have the same dominant country contributor.” As such this is meaningless. E.g. Imagine the case where in 90% of the cities the main contributor to PM10 are domestic emissions. If I randomly assign ‘domestic’ or ‘non-domestic’ to the main contributor I will have an agreement in 81% of the cases. To do significantly better than random the agreement should be 91%. However, I could not find the percentage of cities for which the domestic contribution is dominant I assume it is very high. Without comparison with this number the figures 9 and 11 have no meaning. When this percentage is known the probability on X or more % of agreements follows a cumulative binomial distribution. I think a p-value (probability to find this or a higher percentage) for each percentage of agreement should be calculated.

Hide

ED: Publish subject to minor revisions (review by editor) (17 Jan 2020) by Graham Mann

AR by Matthieu Pommier on behalf of the Authors (24 Jan 2020) Author's response Manuscript

ED: Publish subject to technical corrections (02 Mar 2020) by Graham Mann

In my decision-stage Topical Editor review I explained that one of the two reviewers (both of whom found major revisions were needed in their original reviews) found the revised manuscript would now be acceptable after minor revisions, the other findings major revisions still required.

As part of my decision-stage Topical Editor review I had reviewed the changes the authors had made to the manuscript, and the specifics of the major revisions the 2nd reviewer found were still required. I noted that that reviewer recommending major revisions were still required had acknowledged that "The agreement metric was better explained in the revised version" and the reviewers had addressed many of their comments.

Since that review indicated it was presentational matters, in the way the manuscript was written, that were the major revisions needed, my review essentially agreed with the other reviewer's recommendation of minor revisions, and although agreeing the text could be improved in some places, this was mostly the text in the Abstract, and the last part of the Conclusions, that needed sharpening up.

I provided a list of minor revisions to sentences in the Abstract and Conclusions, with once addressed, would make the manuscript then suitable for publication in Geoscientific Model Development.

The authors have revised the manuscript addressing each of the minor revisions I suggested, and the text in the Abstract and Conclusions is now much improved.

And I am now content to recommend the paper be published in its current form, with just two very minor further grammatical corrections to remedy in the Conclusions before the manuscript be published.

Additional (very minor) grammatical corrections:
------------------------------------------------------

1) Conclusions, 3rd-to-last paragraph

"...split into an influence from the others 30 countries used in the..."

Correct "others 30 countries" to "other 30 countries". (English grammar).

2) Conclusions, 2nd-to-last paragraph

"... methodologies agree in average by 68% in the determination of...."

Correct "agree in average" to "agree on average". (English grammar).

Hide

AR by Matthieu Pommier on behalf of the Authors (03 Mar 2020) Author's response Manuscript

Short summary

The EMEP and LOTOS-EUROS models comprise the operational source contribution prediction system for the European cities within the Copernicus Atmosphere Monitoring Service (CAMS). This study presents a first evaluation of this system, with hourly resolution, by focusing on one PM₁₀ episode in December 2016, dominated by the influence of domestic emissions. It shows that the system provides valuable information on the composition and contributions of different countries to PM₁₀.

Prediction of source contributions to urban background PM10 concentrations in European cities: a case study for an episode in December 2016 using EMEP/MSC-W rv4.15 and LOTOS-EUROS v2.0 – Part 1: The country contributions

Download

Interactive discussion

Peer-review completion

Prediction of source contributions to urban background PM₁₀ concentrations in European cities: a case study for an episode in December 2016 using EMEP/MSC-W rv4.15 and LOTOS-EUROS v2.0 – Part 1: The country contributions