Referee comment - EGUsphere, egusphere-2023-1135

Influences of sources and weather dynamics on atmospheric deposition of Se species and other trace elements

Esther S. Breuninger, Julie Tolu, Iris Thurnherr, Franziska Aemisegger, Aryeh Feinberg, Sylvain Bouchet, Jeroen E. Sonke, Véronique Pont, Heini Wernli, Lenny H. E. Winkel

General comments    The relevant topic concerning sources and speciation of selenium in atmospheric deposition is addressed in this manuscript where a considerable amount of data is reported. Data were obtained by combining modelling and complementary analytical approaches that were applied to a set of precipitation/cloud water samples (respectively 18 and 12 events in a period of 2 months, that were sampled in sub-events) and a significant series of aerosol samples collected with weekly resolution during five years. The complexity of the topic is illustrated by the multiplicity of atmospheric processes identified to be involved in the cycling of elements, which is mainly discussed quantitatively and qualitatively for selenium but also applies to other elements. My opinion is that research reported in this manuscript, with impressive number of data, is well conducted and significant. However, and in spite of the size/length of the document, some important information is missing and some conclusions should be tempered (see below specific comments).

Specific comments

• Sample processing (lines 172-173, lines 182-184, line 192) and Chemical analysis (lines 215-218): from the methods description given, I understand that the conditions of sample treatments (extraction of water-soluble fraction of aerosol samples and pre-concentration of these fractions and of precipitation/cloud water samples) was the same before speciation analyses of Se and S species. However, details of method development given in Supplement S2 (in particular species stability during extraction and pre-concentration) only target on Se species and no information about the stability of S species is provided. This may be due to the fact that such extraction and/or pre-concentration methods were previously validated for S species? In that case corresponding reference(s) should be added. Otherwise, the whole discussion around identified S species may be somewhat more complicated.
• Lines 236-245: the quantification method used for Se and S species must be specified in this section.
• Line 351: as the indication of the range of values is given in previous sentence, the median value would be more meaningful here.
• Lines 519-533 and corresponding supplements: the comparison of example Se chromatograms illustrating the presence of a third Se-containing peak (fig S18a) to the standards one (supplement S3) seems to indicate a shift in retention times, although retention time units being different. It is not obvious to distinguish if the retention time of Org Se is different from the one of selenocystine standard? To facilitate readability, it would be helpful if chromatograms are plotted with same units (select min or s, and counts or counts per second for all layouts). The complementary use of cation exchange chromatography to try to understand the nature of Org Se peak is interesting. However, from Fig S19, the result of DMSeP addition is not quite clear as 2 peaks are increasing and one is decreasing. Then Se species recoveries are given (lines 524-525) but as indicated in previous comment, species quantification method was not indicated, which is important to know in particular in the case of Org Se peak, for which no commercial standard is available, to clarify recoveries and concentrations values (lines 530-533).
• Lines 539-541 and corresponding supplement: examples of S chromatograms of fig S18b indicate poor chromatographic separation of HMS (hydroxymethanesulfonate) and MSA (methanesulfonic acid) peaks for which resolution does not allow accurate compounds quantification thus calling into question the following discussion and correlations using proportions or concentrations of these S compounds.
• Lines 551-552: SeIV proportion in aerosol water extracts appears to be significantly lower in summer than in autumn samples, does this difference remain if proportion is calculated as % of total Se in aerosol instead of its water-soluble extract?
• Lines 558-562: as indicated in previous comments, without information about S species stability during extraction (and if so, preconcentration) and given the poor resolution between HMS and MSA peaks, the discussion involving HMS is very hypothetical.
• Lines 571-572: precipitation/cloud waters were collected in the period from end of August to October, comparison with aerosol water extracts of the same period (September to October) only indicates a difference between aerosol extracts and cloud waters, while SeIV proportion in precipitation waters was not significantly different neither from cloud waters nor aerosol extracts (from letters indicated in fig 5).
• Lines 622-624: same comment for discussion around MSA as previously indicated for discussion involving HMS.
• Line 662: Org Se was detected and not identified, the presence of single species in this chromatographic peak eluting at or close to the void volume of the column was not proven here.
• Supplement S2: the verification of Se species stability during extraction of water-soluble fraction of aerosols and pre-concentration step is well conducted using available commercial standards for validation of used methods. I have however a couple of points:

Line 59: partial transformation of SeMet is indicated to explain its lower recovery following extraction of water-soluble fraction of aerosols. It would be interesting to learn more (chromatographic detection of other Se-containing compound(s) and if so, which retention time(s)? ).

Lines 87-88: lyophilisation to dryness led to losses of organic Se species for which the authors indicated a possible "transformation to other organic Se species that are not retained by anion exchange". This is a very important point as compounds not retained by anion exchange should elute at or close to the void volume of the column, as it is the case for the unknown organic Se compound/pool later detected in samples.

Lines 97-104: the second set of tests compared different containers using lyophilisation to a residual volume < 1,5 mL. In this part, it is not clear in which aqueous medium (or media) the test was done? In the same way, tested media (lines 74-78) were ultrapure water, ammonium citrate solution and rainwater, but water-soluble fraction of aerosols does not seem to have been considered, is that right?

• Supplement S10: add in the table caption, the number of samples considered for calculated correlations. It is important information as although statistically significant with p-values < 0.01 or <0.05, the strength of correlations is weak to moderate.

Technical corrections

• Abstract line 19: check for duration of aerosol sampling 2015-2019 or 2015-2020?
• Line 167: "53.6 ± 2.8" round to the number of significant figures
• Line 205: revise "equipped with and SPS4…"
• Line 321: revise "in order to minimized…"
• Lines 352-354, 530-532 and 536-538: round to the number of significant figures
• Fig 5: aerosol time series is 2015-2019 in the figure, 2015-2020 in figure caption
• Supplement S3 line 127: revise "an anion exchanges…"
• Table S3: round to the number of significant figures (recoveries column)

Referee comments on the article: egusphere-2023-1135:

Influences of sources and weather dynamics on atmospheric deposition of Se species and other trace elements

By Esther S. Breuninger, Julie Tolu, Iris Thurnherr, Franziska Aemisegger, Aryeh Feinberg, Sylvain Bouchet, Jeroen E. Sonke, Véronique Pont, Heini Wernli, and Lenny H. E. Winkel

General comment

In the paper, the authors present a comprehensive study of the main factors describing the atmospheric deposition of Se. For this purpose, the authors present experimental measurements that are correlated with model results. It is important to highlight the large number of samples collected for the analysis (weekly samples during 5 years), as well as the precipitation samples obtained. Another interesting aspect of this work is the analysis performed at the high-altitude atmospheric observatory (Pic du Midi (French Pyrenees; 2877 m a.s.l.), which allows the analysis of the synoptic transport of Se. I consider that this article can be accepted in its present form because it is well-written and presents very interesting results.

Here are some minor details and suggestions.

Specific comments

• I recommend the authors write a shorter abstract. Please, introduce a more summarized writing where you highlight the main findings of the article. I think this paper presents a very complete analysis, but if the authors abbreviate this section, it makes it more appealing to the reader.
•  
• Line 37 The authors state that trace elements and isotopes in water are decoupled from clouds during precipitation. Can the authors indicate the specific result where this effect is observed?

• This is a very long article, as it presents different subsections. I recommend introducing a table of contents at the beginning of the manuscript.

• Methods section: the reviewer strongly recommends the authors extend the description of the study region, specifically to detail the Pic du Midi Observatory monitoring station site. It would be interesting to know the height of the mixing layer in the region. Since this study deals with a synoptic transport of Se, it would be interesting to know if the authors discarded the local influence (convective transport in the mixed layer).
• From your results: ¿it is possible to assess if Se concentrations decline with precipitation amount?
• What about the solubility of Se species: your study, only measures dilute Se species?
• In the conclusions, you state that: "for the first time we were able to identify an organic Se species as a biomarker of marine biogenic sources".

Could you describe the specific result that allows you to conclude this?

• Line 365: Se can be released to the atmosphere from a variety of sources (e.g., natural sources such as soil, water, and vegetation, as well as anthropogenic sources). Are the terrestrial contributions of Se at Pic du Midi only due to synoptic transport or to local sources?