Aerosol radiative effects in the ultraviolet, visible, and near-infrared spectral ranges using long-term aerosol data series over the Iberian Peninsula

Mateos, D.; Antón, M.; Toledano, C.; Cachorro, V. E.; Alados-Arboledas, L.; Sorribas, M.; Costa, M. J.; Baldasano, J. M.

doi:https://doi.org/10.5194/acp-14-13497-2014

Articles | Volume 14, issue 24

https://doi.org/10.5194/acp-14-13497-2014

© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-14-13497-2014

© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 14, issue 24

Research article

|

19 Dec 2014

Research article |

| 19 Dec 2014

Aerosol radiative effects in the ultraviolet, visible, and near-infrared spectral ranges using long-term aerosol data series over the Iberian Peninsula

D. Mateos, M. Antón, C. Toledano, V. E. Cachorro, L. Alados-Arboledas, M. Sorribas, M. J. Costa, and J. M. Baldasano

Download

Final revised paper (published on 19 Dec 2014)
Preprint (discussion started on 02 Apr 2014)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC C1922: 'Revision of the paper by Mateos et al.', Anonymous Referee #2, 29 Apr 2014
- AC C3819: 'reply to Referee1', David Mateos, 19 Jun 2014
  - AC C3826: 'new version of the article with marked changes', David Mateos, 19 Jun 2014
RC C1956: 'Revision of the paper Mateos et al.,2014', Anonymous Referee #1, 30 Apr 2014
- AC C3831: 'reply to Referee2', David Mateos, 19 Jun 2014
  - AC C3838: 'new version of the article with marked changes', David Mateos, 19 Jun 2014

Peer-review completion

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

AR by David Mateos on behalf of the Authors (19 Jun 2014) Author's response Manuscript

ED: Referee Nomination & Report Request started (27 Jun 2014) by William Lahoz (deceased)

RR by Anonymous Referee #1 (12 Jul 2014)

RR by Anonymous Referee #2 (22 Jul 2014)

Suggestions for revision or reasons for rejection

General comments
The revised manuscript by Mateos et al. has only in minor part improved compared to its first version. The authors have not satisfactorily addressed all my previous comments. It remains several open questions and doubts concerning the analysis and the results, which are detailed in the following.

At first, some point-to-point comments to author answers (author answers are indicated with ++).

REVIEWER OLD COMMENT: My second main concern regards the fact that part of the ARE analysis and discussion does not provide neither new methods nor results. The obtained ARE values and the dependence on SSA, as also discussed by the authors, are in good agreement with several other studies performed in the Mediterranean basin. So basically the results of this paper confirm things that we already know. In my opinion, the most original and interesting part concerning ARE estimates, is the discussion of the different spectral contributions (UV, SW, Vis, NIR), which unfortunately I have found only at the end of Section 6. I suggest the authors to consider reorganize the discussion around ARE estimates, especially in Section 6, in order to better highlight their findings.

++ In order to compare our results with the previous studies, new Table 4 summarizes the findings about AFE values. As the reviewer can see, the time periods of previous studies are shorter, being in most cases only referred to case studies.

REVIEWER NEW COMMENT: I do not agree with this since many of the other studies refer to multiyear data (3 to 9 years), so covering a similar time period.

++ The method followed in this study with the aerosol properties simplifies the spectral behavior and the evaluation of ARE can be carried out with high accuracy.

REVIEWER NEW COMMENT: This is not so evident, since the accuracy of the ARE results depends on the accuracy of the used aerosol properties. As discussed in Appendix A your assumptions introduce an important level of uncertainty.

++ The used long-term databases have produced a detailed evaluation of ARE depending on aerosol properties.

REVIEWER NEW COMMENT: Not sure, see the previous comment.

OLD COMMENT: Third point, the analysis of uncertainties is not completely developed. For instance, can you provide error bars in Fig 4.? Also, it is not clear how the ARE uncertainty has been estimated. A certain number of assumptions have been performed to implement model calculations (Sect. 3), however the possible effect of these assumptions on the calculated ARE is not investigated (see for instance the specific comments below). To assess these uncertainties is however necessary to better constraining your results.

++ We included the new Appendix A in the manuscript, where the reviewer can find the justification of the choices. The results shown in the Appendix support our methods.

REVIEWER NEW COMMENT: The sensitivity study provided in Appendix A is correct and allows quantifying the impact of the assumed optical properties on model calculations. Concerning this, I have 2 comments.
1. Line 498: add an appropriate reference
2. The results of the sensitivity study indicate that there between 2 and 10% difference in the simulations of the net spectral radiative fluxes as a function of the SSA and g values used in the calculations (deviations from the 0.90 and 0.75 values fixed for AOD<0.15). I consider these deviations absolutely NOT negligible, because comparable to the absolute values of the forcing obtained by model calculations. Which is the ARE results (value +/- uncertainty) if you take into account for this 2-10% error (the paper misses from a rigorous ARE error analysis in Section 3, lines 197-214)? These uncertainties on the ARE retrieval at AOD<0.15 (which represent a very large fraction of your data) largely impact the significance of the results shown in Figures 5 to 9). As a consequence, the results both in term of absolute values, seasonality, and temporal trends are largely affected by that. A possible solution would be to distinguish data with AOD>0.15 and AOD<0.15 and to look at what you obtain in this case.

REVIEWER OLD COMMENT: Section 4 and Figure 4: can you specify the number of datapoints or measurement days for each year? Are they uniformly distributed throughout the different seasons for the different years? What about cloud cover? It is possible that some differences in the annual values reported in Fig 4 are related in part to specific episodes, such as for example an enhanced cloud cover during specific periods which has affected CIMEL measurements?

++ We added a comment about this topic to the manuscript. The aerosol measurements of CIMEL are only performed under cloud-free conditions, and this fact can produce that some events are only visible in some stations.

REVIEWER NEW COMMENT: you did not completely answer to the raised questions which instead are critical to understand the differences observed in Figure 4.

REVIEWER OLD COMMENT: Section 5, page 8792, lines 26-27: I guess the larger contribution in the visible is due to the fact that the max of the solar spectrum is found around 700 nm.

++ As the ARE is obtained as a difference between two SW radiation values, the maximum of VIS range in each SW radiation could not be true in the ARE. This is the reason why we emphasized the maximum in the VIS range.

REVIEWER NEW COMMENT: what I meant is that since the max of shortwave radiation is found in the VIS, the aerosol-radiation interaction at these wavelengths is generally more effective than in the other spectral ranges.

Other specific comments

Introduction : There are still some problems in the Introduction. At first I would eliminate Lines 52-54, since this sentence is not necessary and imprecise. I would suggest replacing Forster et al. (2007) with Boucher et al. (2013) (new IPCC report). Also, I find that the second paragraph is still not clear and imprecise. For instance, in lines 67-68 you should specify that you refer only to dust particles for which the LW effect is relevant. Lines 72-73: not clear, and in any case the solar zenith angle plays a role not only in the UV region.

Section 4: lines 233-245 and 277-290, is there some effect due to the different altitude of the 6 stations? Why in Figure 4 only the Barcelona data present error bars (the same for Figure 6)? Paragraph lines 308-318: it is very difficult to thrust in the AOD temporal trend from the data in Figure 4. Some reasons are:
- There are not error bars in the plots;
- We do not know how many measurements days correspond to each year data at each station; also, how they are distributed along the year? You did not answer to this in the revised version of the paper.
- By looking at your results and considering the results of statistical anlyses, I would conclude that the trend is NOT observed, except at one or two stations;

Sections 5 and 6: In general, the accuracy of the obtained ARE is highly impacted by the uncertainties associated to model calculations for AOD<0.15 data (see one of the previous comments regarding Appendix A). This raises many questions on the obtained results and associated discussion.

Line 326, please add references.

Lines 332-333: it is not clear, I suggest rewriting the sentence.

References: There is an excessive number of self-citations, also with several works not related to the present study. I have noticed that in the revised version some others have been added (like for exemple the Mateos et al. 2014 JGR paper on cloud properties, which I find out of the main topic of the paper). In addition, I feel that there are some sentences which are not necessary but that have been added with the intention of including some citations (for instance, lines 158-160, 194-196).

In conclusion, I think that the paper still needs substantial revisions and that in the present form it does not deserve publication. One of my major concerns regards the accuracy of ARE estimates, which are highly impacted by the performed assumption on aerosol optical properties. The main conclusions are not supported by the data, in particular the temporal trends which are very uncertain and difficult to trust. I would encourage the authors to review their analysis and then to consider resubmission to ACP or to another journal.

Hide

ED: Reconsider after major revisions (27 Aug 2014) by William Lahoz (deceased)

AR by David Mateos on behalf of the Authors (02 Oct 2014) Author's response Manuscript

ED: Referee Nomination & Report Request started (10 Oct 2014) by William Lahoz (deceased)

RR by Anonymous Referee #2 (21 Oct 2014)

Suggestions for revision or reasons for rejection

General comments
Compared to its first submitted version, the paper by Mateos et al. has largely improved. Now I find the discussion in Appendix A and Section 4 more consistent and complete. The reader has now all the elements to evaluate the accuracy of the results. The paper benefits of this. However I have to say that I find the discussion of the trends still quite weak and, in my point of view, difficult to trust. Personally, by looking at Figure 4 I am not at all persuaded that there is a real trend in the data (despite the results of your statistical analysis suggest this). I observe a large variability in the AOD, which however can be linked to the different conditions and data availability at the various sites, as well as “climate” parameters, such as the QBO, as also discussed in the text. So, I suggest the authors not to stress too much this part in the paper (I would reconsider point 2 in the conclusions, for instance). I would suggest analysing the results of the AOD, ARE and forcing efficiency mostly with the point of view of a “multi-year dataset”, so discussing the spatial and interannual variability. As stated by the authors, in fact, this work analyses an extended dataset and provides a large view of the Iberian Peninsula, so I think the authors should concentrate mostly on this aspect. I would avoid looking at these data as a database for trend analysis over the Penisula.
Based on this, I would encourage the authors to adapt the discussion accordingly.

Specific comments
Lines 65-66: please rewrite or split in two sentences
Lines 514-516: I do not understand what you mean, please rewrite
Figure 4: please gives an indication of a scale for the symbols (a big symbol corresponds to more than 100 cases? 200 cases?)
Figure A1: it is not clear how to read the Figure, please clarify this in the caption

Hide

ED: Reconsider after major revisions (26 Oct 2014) by William Lahoz (deceased)

AR by David Mateos on behalf of the Authors (10 Nov 2014) Manuscript

ED: Publish subject to technical corrections (14 Nov 2014) by William Lahoz (deceased)

Please address the points below, taking special care of the points about the trends.

L. 44: Do you need “in fact”?

L. 48: You mean the diffuse component? If so, specify.

L. 72: You mean the aerosol climatology for the Iberian peninsula? If so, specify.

L. 81: “…on the radiative…”.

L. 83: Remove “followed”.

L. 98: Better to use “…is provided…”.

L. 108: “…protocols by the…”.

L. 109: “…approximately every 12 months of operation.”.

L. 111: I suggest replacing “about” with “of”.

L. 120: Do you mean “produce” when you say “elaborate”?

L. 126: “…e.g., …”. Apply elsewhere.

L. 127: “…for analyzing…”.

L. 135: “…same order of the…”.

L. 140: “…effect is unrealistically…”.

L. 210, 212: “…the Sen method…”.

L. 261: Do you need “actually”?

L. 302: indexes -> indices. Checked in the web, and “indices” is the appropriate plural form here.

L. 311-312: Rephrase this, as it does not indicate that the trend is not significant. Also, reword the next two sentences to be consistent. Here and elsewhere, would it be better to talk about “changes” rather than “trends”?

L. 316: mixed -> mix

L. 384: I suggest “The AFE…”, to avoid starting a sentence with an acronym. Apply elsewhere (e.g., L. 428).

L. 466: The previous sentences mention a positive trend; here you mention a decrease. Please check.

Figs. 1, 2, 3 caption: Indicate that the plots are for the six stations studied.

Fig. 4 caption: “…The larger the symbol, the larger the…”.

Fig. 9 caption: define the particle size ranges (using information from legend).

Fig. A1 caption: indicate which panel is which.

Hide

AR by David Mateos on behalf of the Authors (19 Nov 2014) Author's response Manuscript

Short summary

A long-term analysis of aerosol radiative effects over the Iberian Peninsula is carried out. A reduction of aerosol effects on solar radiation at the surface is observed in the 2000s. Aerosol forcing efficiency is stronger for small and absorbing particles. The contributions of the ultraviolet, visible, and near-infrared spectral intervals to the total shortwave efficiency vary with the aerosol types, producing the visible range the dominant contribution for all aerosol types.