Investigation of Icelandic Dust Presence in the Aerosols Collected at Hornsund (Svalbard, Norwegian Arctic) in Spring 2019

Round 1

Reviewer 1 Report

The manuscript "Investigation of Icelandic dust presence in the aerosols collected at Hornsund (Svalbard, Norwegian Arctic) in Spring 2019" by Moroni et al, presents a morphologic and chemical analysis of Arctic aerosols obtained from SEM/EDS techniques. The method is well described, the study has been done carefully, and the interpretation makes sense. The method is also potentially interesting for other researchers.

The paper is adequate for the journal's scope; it is well-formatted and presents a valuable proposal, so it deserves publication in this Journal. I congratulate the authors for this work, done in environmental difficult conditions. However, before publication, I found some points that must be addressed.

- The order of the references: the first reference presented is [12].

- Tables 3a and 3b are referenced in the manuscript but not presented anywhere.

- abbreviations like i.e. or e.g. should be slanted.

- 2L/min should be 2 l/min.

- Figure 2, legend, 400 micro m-3 is separated into two lines; please include unbreakable space to avoid separation.

- Section 3.3: The third line ends before the end of the paragraph.

- Some double spaces are found throughout the manuscript.

Author Response

We would like to thank the Reviewer for the encouraging comments. We also apologize for the numerous defects in the text, due to formatting defects that occurred during our transcription of the manuscript into the required format. Below we explain how we corrected the errors.

- The order of the references: the first reference presented is [12].

The first thing that happened during the formatting (which we didn't realize) was the entire disappearance of the first paragraph of the Introduction. Now we have inserted it again (in yellow), and this has re-established the order of the references.

- Tables 3a and 3b are referenced in the manuscript but not presented anywhere.

Table 3a/3b reports representative individual particle analyses of different particle types in sediments and aerosols. It is very large in itself and for this reason at the time of submission we decided to include it as Supplementary Material. As such, however, it did not appear among the material attached to the manuscript and we regret this. We have realized that the table is necessary for the evaluation of individual particle analyses. For this reason we have tried to include it in the main text. The only way to do this was to insert it vertically. We hope this will be acceptable to the editorial staff.

- abbreviations like i.e. or e.g. should be slanted.

Yes indeed

- 2L/min should be 2 l/min.

Done

- Figure 2, legend, 400 micro m-3 is separated into two lines; please include unbreakable space to avoid separation.

Done

- Section 3.3: The third line ends before the end of the paragraph.

Done

- Some double spaces are found throughout the manuscript.

Done (hopefully all)

Reviewer 2 Report

Thank you for the opportunity to review this interesting work. Reading the Introduction, I got the impression that the work is to a large extent a continuation of the research conducted by the main author. In environmental research, besides originality or presenting novelty, it is also important to continue the research and extend it with new elements. In this work, the novelty presented in the context of the Hornsund results is a strength. I think it is worthwhile to highlight the strengths of the research presented and to highlight elements of originality compared to previous research of the author. The order of the figures should be reordered and their readability in the text should be improved - for this reason alone I consider the presentation of the results to be average. You should not write that differences in anything are significant when you have not carried out a statistical analysis of the results to confirm this. I suggest either removing these statements or adding the results of the statistical analysis.

In conclusion, I find the continuation of the study extended to the new research area of Hornsund very interesting. To the best of my knowledge, this type of study has not been conducted in Hornsund before, so the data presented are very important in terms of expanding our knowledge. However, it is the authors who should highlight this asset in the Introduction. The article requires a minor revision, after which I suggest its acceptance.

Below are some comments on the text, which I hope will help the authors to improve the quality of the manuscript.

Line 8: please check and correct the affiliations

Line 20: remove additional space between words

Lines 65-67: This is the only sentence where I see mentioning of Hornsund in the Introduction chapter. I see no literature or explanations why authors choose Hornsund for their research. No literature concerning what type of research were implemented in Hornsund till now. Were there any similar research implemented in Hornsund before by anyone else? Does research described in paper also fill in a gap in knowledge concerning integrated morphological and chemical particle analysis and meteorological approaches in Hornsund? These are the strengths of paper that should be highlighted in Introduction. I think that the authors may improve introduction and highlight the strengths of their research.

Line 100: remove additional space between sentences

Line 147: I would suggest to rewrite the sentence to avoid  starting a line with “Kavan et al.[34]….”

Line 155: 240h….isn’t it too long? With duration of propagation time also the error of such a modelling increase. Why the duration of propagation time was not limited by authors to 168h? If it was necessary than I suggest to highlight also the weakness of the long propagation time.

180: Could You please add information about the duration of mineral dust sampling in 2020? I see later in text information about air sampling duration in April/May 2019.

Line 189: Life 1 Mega System low-volume sampler- I do not see information about the volume of air flow through the filters as in case of  collection of mineral dust in line 180. Pleas add this technical information.

Figure 3: Please correct the red captions on the figures. Especially on Fig.3.a and b. it is impossible to read them.

Line 299: “…higher to much higher..”? This is no form of performance comparison. Correct.

Line 310: Why Figure 5 is shown in this chapter, while it is present in the text of the previous chapter?

Line 314: “ In the graph,…” which graph? Figure 5 or Figure 6? I suppose Figure 6, however presence of figure 5 in this chapter is confusing when description started with "In the graph..." comes straight below Figure 5.

Line 339: Where is Figure 7? Move it to the chapter where it is described.

Line 352-353: remove the space between lines

Line 369: I like the graph on Figure 8, however, the scale and its title needs to be improved. Poor quality of the scale caption.

Line 373: I love the idea of the graph on Figure 9. However, the quality of the scale captions is very poor. Please correct it.

Lines 386-396: What is the source of this information? There are given information that the dust plumes were visible in true color on MODIS, however I do not see any graph even in supplementary files. Does all of the information in paragraph were taken from the station in Akureyri? If yes, please consider highlighting it on the beginning of the paragraph in 1-2 sentences.

Line 397: Remove highlighting in colour the chapter title.

Line 438: Was it statistically confirmed that the change in the composition of aerosols between April and May was significant and in which cases?

Line 438-440: What does it mean “seems to be” or “rather well”. Please rewrite the sentence. You started that it is certain and later we have to deal with “seems to reflect” and “rather well”.

Line 462-463: This is unnecessary repetition in conclusions.

Author Response

We thank the Reviewer for the encouraging comments and appreciation. We also apologize for the numerous defects in the text, due to formatting defects that occurred during our transcription of the manuscript into the required format.

Line 8: please check and correct the affiliations

I actually found it correct (https://forscience.pl/language/en/). This one and the others were all provided by the co-authors themselves

Line 20: remove additional space between words

Done

Lines 65-67: This is the only sentence where I see mentioning of Hornsund in the Introduction chapter. I see no literature or explanations why authors choose Hornsund for their research. No literature concerning what type of research were implemented in Hornsund till now. Were there any similar research implemented in Hornsund before by anyone else? Does research described in paper also fill in a gap in knowledge concerning integrated morphological and chemical particle analysis and meteorological approaches in Hornsund? These are the strengths of paper that should be highlighted in Introduction. I think that the authors may improve introduction and highlight the strengths of their research.

The aim of this work was mostly methodological i.e., to develop and test a geochemical criterion for the characterization and comparison of the dust particles within aerosols in Svalbard. In Hornsund we had the opportunity to sample aerosols, snow and sediments in the same period which was also very favorable for the transport of dust from Iceland. This is the primary reason for the work and the present manuscript. Studies on aerosol and, especially, on dust composition in Hornsund are, actually, very rare and it is this reason that we have now tried to explain better in the Introduction.

Line <100: remove additional space between sentences

Done

Line 147: I would suggest to rewrite the sentence to avoid starting a line with “Kavan et al.[34]….”

Done

Line 155: 240h….isn’t it too long? With duration of propagation time also the error of such a modelling increase. Why the duration of propagation time was not limited by authors to 168h? If it was necessary than I suggest to highlight also the weakness of the long propagation time.

Accuracy of the trajectories depends upon the resolution of the meteorological data and propagation time can be increases accordingly. In this study the resolution is quite high, 0.5° x 0.5°, to follow for a sufficient long time trajectories with an erratic or circular pathways. However, the five areas of interest described in section 2.2 are located at a distance from the receptor site covered by most of them in less of five days, where the numerical results with this resolution are very stable. Moreover, the accuracy of the propagation and the presence of numerical artifacts were evaluated comparing those reaching the site at other three different heights, as stated in the section.

180: Could You please add information about the duration of mineral dust sampling in 2020? I see later in text information about air sampling duration in April/May 2019.

Duration of mineral dust sampling is reported in Table 2 as “Sampling time”. We’ve added a short sentence on this point in the main text.

Line 189: Life 1 Mega System low-volume sampler- I do not see information about the volume of air flow through the filters as in case of  collection of mineral dust in line 180. Pleas add this technical information.

The sampling flow rate has been reported in the text.

Figure 3: Please correct the red captions on the figures. Especially on Fig.3.a and b. it is impossible to read them.

We’ve changed them in white or black and added relief to improve readability.

Line 299: “…higher to much higher..”? This is no form of performance comparison. Correct.

We have changed the sentence.

Line 310: Why Figure 5 is shown in this chapter, while it is present in the text of the previous chapter?

It was just a matter of space (and formatting). We've fixed it now.

Line 314: “ In the graph,…” which graph? Figure 5 or Figure 6? I suppose Figure 6, however presence of figure 5 in this chapter is confusing when description started with "In the graph..." comes straight below Figure 5.

Done

Line 339: Where is Figure 7? Move it to the chapter where it is described.

Done.

Line 352-353: remove the space between lines

It’s gone while formatting.

Line 369: I like the graph on Figure 8, however, the scale and its title needs to be improved. Poor quality of the scale caption.

We’ve changed the font size.

Line 373: I love the idea of the graph on Figure 9. However, the quality of the scale captions is very poor. Please correct it.

We’ve changed the font size. Hope it is enough to be readable.

Lines 386-396: What is the source of this information? There are given information that the dust plumes were visible in true color on MODIS, however I do not see any graph even in supplementary files. Does all of the information in paragraph were taken from the station in Akureyri? If yes, please consider highlighting it on the beginning of the paragraph in 1-2 sentences.

MODIS images are public (https://worldview.earthdata.nasa.gov/) while PM₁₀ data from Reykjavik and Akureyri are from Icelandic Air Quality Information System (https://xn--loftgi-tua4f.is/). What we did was to consult the archives to check and validate infos. For this reason It did not seem necessary to us to include (in the manuscript/supplementary material) images and data simply extrapolated from archives without having processed them in some way. We limited ourselves to recalling the source of the data.

Line 397: Remove highlighting in colour the chapter title.

Done

Line 438: Was it statistically confirmed that the change in the composition of aerosols between April and May was significant and in which cases?

Indeed the term “significant” is inappropriate in this context, so we have replaced it with a non-statistical term.

Line 438-440: What does it mean “seems to be” or “rather well”. Please rewrite the sentence. You started that it is certain and later we have to deal with “seems to reflect” and “rather well”.

You’re perfectly right. Too much caution on these aspects is unnecessary and even useless.

Line 462-463: This is unnecessary repetition in conclusions.

We’ve removed it from the text.

Reviewer 3 Report

This paper offers an interesting way for determining sources of atmospheric dust by mineralogical anlaysis.  The paper is somewhar weak in methods used for sampling and complete characterization (by mass) of dust (aerosols).  Attention could be given to improving some elements of the study for the reader.  These include the following examples:

1.  There is ambiguity in sediment and aerosol size information.  Sediments are likely to be very coarse relative to aerosols sampled Is it assumed that the dust is uniform by size?).  It would help to provide perspective on particle size by adding a size scale in the micrographs of Fig. 3 and 4.  It would also help to provide information on the mineral dust mass fraction of the aerosol if available.  The grain size in sediments needs to be noted.  Also note that generally aerosol particle samplers  provide data in fine size ranges that are very different from coarse sediments. 

2.  The aerosol samplers need to be better described.  For example, the cyclonic unit is hand held with a size cut of 4 um.  The Life 1 samper and environmental chamber need better description and nominal size range sampled.

3.  Table 3 is missing

4.  The narrative should be clear about the PM10 sampling in Iceland--This is done by air pollution authorities--where did these data com from--archives??

5. Figure 9-define"regional" and"long range transport" in tems of spatial  scale and predominant direction??

6. Line 452 Iceland dust predominant source--assertion needs reference?

7. Lines 415-416 Extended day suspension hypothesis needs a better explanation including evidence of armospheric stability and size differention. vs fall out.  There may also  be some significance in the difference in April and May samples for NaCl (sea salt) and sulfate--a marine infuence on trajectories?

8. How were trajectories calculated--i.e. what wind data were available? Trajectories by hour don't make a lot of sense.  How would separation of trajectories by a measure of atmospheric stability or day or night--e.g. from Iceland or Svalbard weather stations?

Minor awkward sentences and overly long complex sentences in places like the first sentence in the abstract.

Author Response

This paper offers an interesting way for determining sources of atmospheric dust by mineralogical anlaysis.  The paper is somewhar weak in methods used for sampling and complete characterization (by mass) of dust (aerosols).  Attention could be given to improving some elements of the study for the reader.  These include the following examples:

1. There is ambiguity in sediment and aerosol size information.  Sediments are likely to be very coarse relative to aerosols sampled Is it assumed that the dust is uniform by size?).  It would help to provide perspective on particle size by adding a size scale in the micrographs of Fig. 3 and 4.  It would also help to provide information on the mineral dust mass fraction of the aerosol if available.  The grain size in sediments needs to be noted.  Also note that generally aerosol particle samplers  provide data in fine size ranges that are very different from coarse sediments.

Sediments are, undoubtedly, much coarser than aerosols. This is evident by comparing the scale bar of figures 3, 4 and 5 (in fact all the figures contain scale bars; to make them more visible we tried to enlarge the figures). This can make it difficult to compare one to the other. The best thing to do when you want to compare sediments with aerosols and dust is to resuspend the sediments in the laboratory. Which we do regularly using an apparatus specially made in our labs (1). However, large quantities of sediment are necessary to do this, and we did not have it in this case.

What we know, however, is that Icelandic dust is, like the sediments from which it comes, extremely rich in glass particles, therefore at least in this case dust composition does not depend much on particle size. The case of Hornsund is different, but due to the low sorting of the sediments it was possible to observe and identify and analyze the components of the finer fraction quite well. Moreover we have to consider that TSP (Total Suspended Particulate) was sampled without any kind of cut-off. We have no data of aerosol mass concentration nor reliable measurements of the mass fraction of mineral dust within the aerosol in Hornsund. In any case, and excluding the marine component (chlorides and sulphates), mineral dust is the predominant part of these aerosol samples. This observation (which has now been reported in the main text of the manuscript) is in agreement with what is reported in a paper currently under review (2), in which the authors analyzed the particulate matter of anthropogenic origin in aerosols in Hornsund in the same period.

(1) S. Bertinetti, E. Bolea-Fernandez, M. Malandrino, B. Moroni, D. Cappelletti, M. Grotti, F. Vanhaecke, Strontium isotopic analysis of environmental microsamples by inductively coupled plasma - tandem mass spectrometry. Journal of Analytical Atomic Spectrometry 37 (2022) 103-113.

(2) Pawlak, Filip and Koziol, Krystyna and Wilczyńska-Michalik, Wanda and Worosz, Mikołaj and Michalik, Marek and Lehmann-Konera, Sara and Polkowska, Żaneta, Characteristic Features of Anthropogenic Pollution in the Atmospheric Air at Hornsund in the Spring of 2019 (Svalbard, High-Arctic). Available at SSRN: https://ssrn.com/abstract=4570456 or http://dx.doi.org/10.2139/ssrn.4570456.

2. The aerosol samplers need to be better described.  For example, the cyclonic unit is hand held with a size cut of 4 um.  The Life 1 samper and environmental chamber need better description and nominal size range sampled.

We have added a few detail on the cyclone, since it is a known instrumentation, and some more information on the passive sampler, which was instead created in our labs.

3. Table 3 is missing

Table 3 reports representative individual particle analyses of different particle types in sediments and aerosols. It is very large in itself and for this reason at the time of submission we decided to include it as Supplementary Material. As such, however, it did not appear among the material attached to the manuscript and we regret this. We have realized that the table is necessary for the evaluation of individual particle analyses. For this reason we have tried to include it in the main text. The only way to do this was to insert it vertically. We hope this will be acceptable to the editorial staff.

4. The narrative should be clear about the PM10 sampling in Iceland--This is done by air pollution authorities--where did these data com from--archives??

Yes, they are from the archives of Icelandic Air Quality Information System (https://xn--loftgi-tua4f.is/en?zoomLevel=7&lat=64.894972&lng=-18.675028)

5. Figure 9-define"regional" and"long range transport" in tems of spatial  scale and predominant direction??

The figure shows how many of the 24 trajectories 240 hours long reaching each day the receptor passed above each area (Regiona, Russia, Greenland, GBR+IRL, Iceland) within an altitude of 800 m a.g.l. The predominant direction is not involved. In other words and referring to the description in par. 2.4, if a trajectory reaching the receptor in a defined day passed above one of the five areas described in the paragraph with an altitude below 800 m a.g.l, the counter for that day is increased by one. The total counter for each day is shown as a heatmap in Figure 9.

6. Line 452 Iceland dust predominant source--assertion needs reference?

We’ve changed the text and added the reference.

7. Lines 415-416 Extended day suspension hypothesis needs a better explanation including evidence of armospheric stability and size differention. vs fall out.  There may also  be some significance in the difference in April and May samples for NaCl (sea salt) and sulfate--a marine infuence on trajectories?

We cannot say anything sure on this point actually. In the High Arctic there can be a strong discrepancy between what happens above and within the boundary layer (in practice, above and below 300-400 m altitude). The slowing down of air masses is (partial) evidence of a condition of atmospheric stability. We also know that the period was characterized by temperature anomalies and intense precipitation (including rain) between 16 and 20 April, which led to the accumulation, at some point of the Hansbreen glacier, of nuclear contaminated particulate matter (Ref. 24 in the manuscript). As for the marine contribution, it is a leit-motif in Hornsund. All this to say that what happens to long-range transported aerosols reaching ground level is not easy to rationalize. For this reason, and since ours is not a synoptic or meteorological study of a special event, we have preferred to limit the discussion and rely on facts rather than hypotheses.

8. How were trajectories calculated--i.e. what wind data were available? Trajectories by hour don't make a lot of sense.  How would separation of trajectories by a measure of atmospheric stability or day or night--e.g. from Iceland or Svalbard weather stations?

Back trajectories where calculated, as described in par. 2.4, using "GDAS meteorological input fields with 0.5° x 0.5° resolution and a propagation time of 240". 24 trajectories per day, reaching the site at whole hours (i.e. at 00:00, 01:00, 02:00, etc) where computed.

Reviewer 4 Report

Dear Authors,

the manuscript “Investigation of Icelandic dust presence in the aerosols collected at Hornsund (Svalbard, Norwegian Arctic) in Spring 2019” by Moroni et al, shows the results of a combined approach for the study and characterization of Arctic aerosols in terms of chemical composition and morphology. The method involves the integrated use also of meteorological models as the rearranged back-trajectories (BT) by NOAA HYSPLIT. The results described tend to demonstrate a composition compatible with that of Icelandic events, however other sources and contributions cannot be totally excluded.

The manuscript is scientifically structured and detailed. I would like just some clarifications on the part relating to the meteorological data use:

-          if BTs starting from any atmospheric altitude up to 800 m are taken into consideration, only direction can be identified but there is no information on any aerosol load transported, as the approximation with what happens on the ground would be too forced

-          line 156 50 m is it correct ?

-          why, especially for the most significant days, not combine the statistical analysis of the BTs with single chosen and significant BTs?

-           wind data are acquired at 10 m a.g.l. while the BT at 500 m, the discrepancy in altitude is considerable and a correlation between the two datasets may not be indicative. Why you need wind data? Please describe me your idea.

The manuscript need also the following minor revisions.

MINOR REVISIONS:

Introduction

1)      It is unusual for the bibliography to start with reference no. 12 and then go back to previous numbers. I recommend rearranging it in ascending order.

2)      Reference 12 is cited many times. I understand it is a milestone but I would prefer the bibliography to be expanded especially in line 43 where year 2023 is mentioned but the reference is dated 2016

3)      Figure 1, although not necessary for research purposes, the names in figure b are unclear; remove them or improve readability

4)      Line 374 Great

5)      Figures 8 and 9, improve the readability of the graph labels

Author Response

Dear Authors,

the manuscript “Investigation of Icelandic dust presence in the aerosols collected at Hornsund (Svalbard, Norwegian Arctic) in Spring 2019” by Moroni et al, shows the results of a combined approach for the study and characterization of Arctic aerosols in terms of chemical composition and morphology. The method involves the integrated use also of meteorological models as the rearranged back-trajectories (BT) by NOAA HYSPLIT. The results described tend to demonstrate a composition compatible with that of Icelandic events, however other sources and contributions cannot be totally excluded.

The manuscript is scientifically structured and detailed. I would like just some clarifications on the part relating to the meteorological data use:

-          if BTs starting from any atmospheric altitude up to 800 m are taken into consideration, only direction can be identified but there is no information on any aerosol load transported, as the approximation with what happens on the ground would be too forced

The BT used for the production of Figures 8 and 9 and the associated analysis are those reaching the receptor at 500 m a.g.l., though trajectories reaching at different altitudes (50, 1000 and 3000) were computed too to better understand the circulation in the period of 2019 under investigation.

-          line 156 50 m is it correct ?

Yes it is

-          why, especially for the most significant days, not combine the statistical analysis of the BTs with single chosen and significant BTs?

If we understand correctly your question, we have 24 trajectories per day which, due to their quite long length (240 hours), might vary significantly in path and, to a larger extent, vertically. It would be subjective picking up only those with a very specific path. In our view, it only would make sense to the goal of describing a typical trajectory. In this case, a cluster analysis of the trajectories for the significant days might be more appropriate. However, due to the larger uncertainty of the vertical coordinate, endpoints height is not generally included in trajectories cluster analysis, decreasing the significance of the selection of a single BT.

-           wind data are acquired at 10 m a.g.l. while the BT at 500 m, the discrepancy in altitude is considerable and a correlation between the two datasets may not be indicative. Why you need wind data? Please describe me your idea.

Wind direction and speed have only local representativeness since they have strong dependence by the local orography. Final BT height, on the other end, should not be chosen too close to the ground due to numerical reasons. However, as described in par. 2.4, trajectories with starting point at 50, 1000 and 3000 m were computed as well and their paths were occasionally analysed to confirm the representativeness of the computed statistics.

The manuscript need also the following minor revisions.

MINOR REVISIONS:

Introduction

1)      It is unusual for the bibliography to start with reference no. 12 and then go back to previous numbers. I recommend rearranging it in ascending order.

This is an error due to formatting defects occurred during our transcription of the manuscript into the required format. The first thing that happened was the entire disappearance of the first paragraph of the Introduction. Now we have re-inserted it (in yellow), and re-established the order of the references.

2)      Reference 12 is cited many times. I understand it is a milestone but I would prefer the bibliography to be expanded especially in line 43 where year 2023 is mentioned but the reference is dated 2016

You are certainly right. The fact is that there are no geological descriptions. However, at least regarding line 43, it seems to us that the statement is of a such general validity that it does not need references. We have simply removed Ref. 12.

3)      Figure 1, although not necessary for research purposes, the names in figure b are unclear; remove them or improve readability

Done

4)      Line 374 Great

Done

5)      Figures 8 and 9, improve the readability of the graph labels

Done