the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Apr 2026
Biogeochemical regimes control marine aerosol emission of hydrogels in the Southwestern Pacific Ocean
Abstract. Ambient marine aerosols are frequently enriched in biogenic material. It was suggested that a critical fraction may be attributed to colloids and aggregates, which are composed of carbohydrates and proteins. Those hydrogels possess excellent cloud condensation and ice nucleation properties. Yet, most atmospheric measurements fail to detect marine hydrogels in aerosols directly, and the few studies which exist were conducted in the Northern hemisphere. Here, we present a comprehensive data set of carbohydrate and protein-enriched hydrogels in in sea spray aerosols (SSA) generated within representative regimes of the Southwestern Pacific Ocean. We relate the concentration of hydrogels in SSA to the occurrence of their precursors in surface seawater and other biogeochemical variables. The highest concentration (0.91 ± 0.72 × 105 m-3, corresponding to 4.1 ± 2.1 × 103 particles ng-1 Na+) and highest relative enrichment (5.3 ± 3.9 × 105) of hydrogels in SSA (size range: 0.5–30 µm) was observed within the subtropical front, which is biologically most active. This was contrasted by subtropical waters, in which SSA concentration and enrichment decreased by one order of magnitude. Interestingly, the carbohydrate-to-protein ratio shifted with size in SSA hydrogels, while no such size-shift existed for marine samples. In comparison to their marine precursors, supermicron hydrogels in SSA were primarily composed of carbohydrates. Our results suggest that hydrogels may complement a considerable fraction of ambient marine aerosols, and significantly contribute to the atmospheric pool of cloud condensation and ice nuclei, in particular above the remote oceans of the Southern hemisphere.
- Preprint
(1192 KB) - Metadata XML
-
Supplement
(234 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1873', Anonymous Referee #1, 15 Apr 2026
-
AC1: 'Reply on RC1', Theresa Barthelmeß, 27 Jul 2026
Review 1 for Barthelmeß et al., submitted to ACP
Reviewer The reviewers’ comments are highlighted in italic font.
Response Our response follows the reviewers’ comment and is written in regular font.
Reviewer The study “Biogeochemical regimes control marine aerosol emission of hydrogels in the Southwestern Pacific Ocean” by Barthelmeβ et al., submitted to ACP, presents measurements of hydrogels (TEP and CSP) using colorimetric staining assays and microscopic, size-resolved quantification in seawater from the Southwestern Pacific Ocean, as well as in nascent aerosols generated in a bubbling tank. The authors relate variations in gel properties to different biogeochemical regimes associated with distinct water masses and report enrichment factors during the transfer from seawater to aerosol phase. The study is motivated by the potential relevance of hydrogels for cloud formation and cloud properties.
Overall, I consider this study to be highly relevant for improving our understanding of aerosol–cloud interactions, particularly in the context of marine aerosols and their role in the climate system, with possible implications for future model parameterizations. I have no doubt that this manuscript has the potential to be suitable for publication after addressing the critics below.
However, in its current form, the manuscript suffers from substantial weaknesses in the presentation of results and discussion. These include considerable overselling of the findings, potential (mathematical or logical) errors in analysis and calculations, incomplete or misleading citation of certain information from the literature, a lack of clear structure, and, at times, a lack of rigor. These issues make it difficult to fully assess and appreciate the scientific value of this study. I strongly encourage the authors to invest additional effort into revising the manuscript to improve clarity, correctness, and overall quality of interpretation.
Response We thank the reviewer for the detailed review and his/her opinion that after addressing the main concerns listed below, our work will be of relevance for refining the knowledge on marine aerosol and cloud interactions. The critics raised are rather substantial and we will commit working towards improving the overall quality of the manuscript and developing its potential further.
Reviewer In the following, I outline my main concerns:
1.) Soot topic:
- a) Clarification and Focus of the “Soot” Discussion: Your discussion of soot particles is currently not fully convincing and, in its present form, feels somewhat distracting because it appears repeatedly throughout the manuscript. As a result, it becomes difficult to identify your main findings. In my opinion, it would help to clarify your central messages. For example: Is your key result that SAW water masses are generally contaminated with soot, but that the corresponding aerosols are not significantly affected? If yes, consider structuring the paper clearly around this finding. If not, I would recommend substantially reducing the emphasis on the soot discussion.
Response Our main conclusion with regard to the soot particles in the surface seawater (SSW) is that subantarctic waters (SAW) waters were not generally contaminated with soot. The potential soot particles were detected at two of the six SAW stations, and at those stations, they were confined to the smallest size bins (0.5–2.5 µm), limiting the scope of the analysis and interpretation of SAW hydrogels only marginally. As suggested by the reviewer, we will improve the manuscript by explaining this once in brief in the method section, while referring to the supplement for a further discussion. We will also restrain from selectively including (in relation to aerosol samples) or excluding (in relation to SSW samples) soot-associated samples. As outlined in d) we will remove the specific data points from the entire analysis. We are convinced that these measures will reduce the distraction and confusion of the reader.
Reviewer b) Scientific Concerns About the Interpretation: I also have some concerns regarding the plausibility of your interpretation: You describe soot as hydrophobic (Line 439), which would suggest it remains in the surface microlayer. How likely is it, then, to find significant amounts at depths of 6 m? If soot sinks as part of aggregates (Line 440), one would expect the particles at that depth to be embedded in hydrogels rather than appearing as isolated small particles?
Response We agree with the reviewer that the potential 'soot' particles in our examples (Figure 2) look isolated and are not concentrated within the stained hydrogels. We will integrate this observation into our argumentation, i.e. that we do not expect major interferences of the potential soot particles with the aerosol-related processes. The environmental behavior of soot particles in seawater remains insufficiently understood. Therefore, with regard to their occurrence at 6 m depth, we consider it also difficult to infer the absence or presence of surface microlayer enrichment solely from samples collected at 6 m depth. We will revise the manuscript to better acknowledge these uncertainties.
Reviewer c) Identification of Black Particles as Soot: I do not find the current presentation of the black particles as soot (Section 2.3) entirely convincing. The supporting arguments are currently scattered across the manuscript (e.g., Line 438 mentioning potential origin from Australian bushfires). I suggest: Collecting and presenting all arguments for identifying these particles as soot in a more focused way, ideally within the Methods section. Using more cautious wording, such as: “Based on the available indications, we suggest that these small particles are likely soot.”
Response As suggested by the reviewer, we will collect and present all arguments for identifying the black particles as potential 'soot' in the method section and also use a more cautious language to describe these particles, including the conclusive statement: 'Based on the available indications, we suggest that these small particles are likely soot.'
Reviewer d) Handling of Contaminated Samples: It is important to clearly state how these filters were treated in your analysis. For example: Were the affected samples consequently excluded from further analysis because they interfere with hydrogel quantification? Or were they retained and marked (e.g., with asterisks) in the results?
Response We agree with the reviewer that the inclusion or exclusion of these samples was not clear. For the statistical comparison of SSW samples, the filters which were potentially contaminated with 'soot' were excluded (e.g. as annotated in Table A3). We suggest to exclude the affected size bins of the two SAW afternoon samples from the entire analysis and representation in the figures. We will describe our approach in the method section.
Sea spray aerosol (SSA) samples are marked with asterisks (CSP area in SSA*) if the chemical composition was approximated by relying on the dual-staining approach.
Reviewer e) Data Presentation and Figures: The current visual presentation is somewhat confusing. For example, in Figure 3: The plots are labelled as CSP and TEP, yet include black particle samples that you explicitly state are neither CSP nor TEP. I recommend either: Adjusting the figure title and labels to reflect what is actually shown, or modifying the visualization to avoid mixing different particle types in a misleading way. Additionally, have you considered improving the microscopic images of the contaminated samples by applying color filters (removing black) with the image software to reduce the visual impact of these black particles?
Response In line with the above answer, we suggest to remove the associated size bins of the two potential 'soot' SSW samples from the analysis, figures, and tables. As the total abundance of hydrogels is affected by the smallest size bins, the representation of these categories will be also influenced, e.g., in Figure 4 or, as highlighted by the reviewer, in Figure 3, where their influence was previously described by adding the interquartile ranges (grey boxes). We hope this will further help to reduce any confusion while reading and any misleading interpretation.
We have applied the same method to enumerate the abundance of hydrogels in the SSA and in the SSW. The protocol is based on the methodology introduced by Engel (2009). This method has been widely applied during the last decades and is established as a standard protocol for SSW samples. To maintain comparability with the established body of literature and to support future extrapolation and model parameterization, we prefer to retain our current approach.
Reviewer f) My personal recommendation: Overall, my suggestion would be: Briefly describe the observation and treatment of (potentially) soot-contaminated samples in a concise paragraph within the Methods section. If you wish to explore this topic further, consider moving the extended discussion to the supplementary material.
Response Thank you for your suggestion and we happily follow this recommendation. We will briefly summarize our observation and the subsequent approach regarding the potential 'soot' particles in our method section and move the further discussion to the supplement.
Reviewer 2.) I found some of the phrasing somewhat strong and occasionally selective in how the literature is presented, which in places gives the impression of overstating the certainty or significance of the findings. For example, in Lines 13–14 you state that “these hydrogels possess excellent cloud condensation and ice nucleation properties,” whereas later (Line 37) this is phrased more cautiously as “hydrogels have been suggested to serve as cloud condensation…” This represents a substantial difference in level of certainty, and I would recommend aligning the wording more consistently with the actual state of the literature. Please check this for all statements throughout the manuscript.
Response We will check all statements carefully and adjust their consistency in the level of certainty throughout the whole manuscript. In general, we aim at a balanced literature review.
Reviewer 3.) You emphasize the differing properties of the four marine regions throughout the manuscript; however, their characteristics are not sufficiently introduced. Referring the reader to Sellegri et al. (2023) & Barthelmeß et al. (2025) in Lines 83/84 & 273/274 for this information is not helpful, as the manuscript should be understandable on its own without requiring prior reading of additional papers. I recommend summarizing and directly comparing the key features of these four regions within your manuscript, for example in an additional table. This could be combined with Section 3.3, potentially moved to the beginning of the Results section. Such a restructuring would help the reader better understand why these regimes are distinct and why differences in aerosol hydrogels are to be expected.
Response We reference our previous papers in L273 to clarify i) that these data have been partly introduced elsewhere and ii) that instead of including three samples per day (Barthelmeß et al., 2024) only two samples were included to align with the sampling frequency of CSP and TEP in the SSW.
We agree with the reviewer that referring to additional literature to characterize the difference between regimes alone, is inappropriate. Section 3.3 is dedicated to describe the biological and biogeochemical differences across regimes including overall thirteen distinct parameters such as e.g., total organic carbon (TOC) concentration or Chlorophyll-a size fractions. Differences including significances are summarized in Table A3. However, our description of regimes lacks the inclusion of salinity and nutrient profiles along the cruise track.
Salinity and nutrient data will be included and introduced (Table A3). We will further amend Figure 4 by including all important biogeochemical parameters (incl., salinity, nitrate and Chlorophyll a). The reader can then follow in parallel how biogeochemistry, SSW TEP and CSP as well as SSA hydrogels developed along the cruise track. Moreover, this figure will contain the information on which SSW samples were averaged to be compared to the SSA sample set.
Following the reviewer’s suggestion, we will move the description of regimes to the beginning of the result section before introducing hydrogel dynamics. We agree that this will improve clarity in whether differences in aerosolized hydrogels are to be expected.
Reviewer Enrichment calculations: Were your sodium measurements in aerosol particles size-resolved? Based on Table A2, this does not appear to be the case, as only a single sodium value per tank experiment is reported. If so, how were size-resolved enrichment factors for hydrogels (as presented in Table 1) calculated? Since sodium is known to be predominantly associated with the coarse mode and occurs at much lower concentrations in submicron particles, using size-resolved sodium data is important for obtaining accurate enrichment factors. If such size-resolved sodium measurements are not available, the only consistent approach would be to calculate enrichment factors for the entire size range as a bulk rather than for individual size classes. This also holds true for the hydrogel/Na+ ratios.
Response We thank the reviewer for raising this important point. We would like to clarify that the size-resolved analysis presented in Table 1 does not refer to commonly monitored aerosol size classes in atmospheric chemistry, but to hydrogel size bins based on microscopical approach (oceanographic method to enumerate TEP and CSP). That is, hydrogels were categorized according to their own size, independent of the size of the aerosols in which they were contained. The reported values therefore do not represent enrichment with respect to aerosol size classes, but rather describe how hydrogels of different intrinsic sizes are enriched or depleted relative to bulk sea salt.
Sodium (Na⁺) in SSA was indeed only considered as a bulk concentration integrated over all impactor stages. Consequently, the enrichment factors reported for different hydrogel size classes were calculated by normalizing the abundance of each hydrogel size class to the total Na⁺ concentration in SSA and SSW, respectively. We acknowledge that this differs from the conventional definition of size-resolved enrichment factors in atmospheric chemistry, where the numerator is resolved with respect to aerosol size and the denominator refers to a bulk seawater reference. This has been applied for TEP enrichment in aerosols before (Van Pinxteren et al., 2022).
As suggested by the reviewer and also in line with a comment of the third reviewer, we will provide bulk enrichment factors integrated over size-resolved SSA Na⁺ concentration which were covered by specific, pooled impactor stages. We are able to resolve hydrogel enrichment factors for submicron (stages 7-8: 500 nm to 1 µm) and supermicron SSA (stages 9-13: 1-10 µm).
We hope that we address the reviewers concerns appropriately by adopting these changes and that these efforts will result in a conceptually clear and interdisciplinary presentation of our data set.
Reviewer 5.) Please elaborate on the potential consequences for your tank SSA results arising from the use of 6 m deep water, without considering the SML.
Response We understand the SML as a dynamic interface, which is, in particular under turbulent forcing, reflecting on the conditions of the ULW (Silva et al., 2026; Engel et al., 2026). For example, it is known that CSP and TEP enrichment in the SML is resuspended once wind velocity rises above 5 m sec-1 (Sun et al., 2018). In equivalence to the formation of SSA, the SML forms via the upward transport of organic material by bubbles (Robinson et al., 2019). The particulate organic material is either ejected into the atmosphere or transiently retained at the air-sea interface. Within the chamber, we therewith also expect the formation of such a transient SML.
However, we acknowledge that dissolved, surface-active molecules become selectively enriched at the air-sea interface also under turbulent forcing (Sabbaghzadeh et al., 2017) and that the true SML is disproportionally affected by further atmospheric influences such as intense solar radiation or dust deposition (e.g., Rickards et al., 2022). Such influences on the organic matter pool are likely underrepresented in our SSW samples.
In general, we further assume that the SML is better represented by a multilayered model, which resolves the ‘nanolayer’ (~1 µm) and the microlayer (1-1000 µm). Recent evidence suggests that the ‘nanolayer’ is of importance to describe physicochemical and aerosol properties appropriately (Crocker et al., 2022; Asmussen-Schäfer et al., 2026). The origin of submicron and/or supermicron aerosols becomes therewith blurred and cannot be attributed to 'the SML'. A simple, potentially inaccurate attribution to ‘the SML’ was thus avoided. Notably, the ‘nanolayer’ cannot be sampled using conventional SML approaches such as the glass plate or the Garrett Screen.
In conclusion, we propose that the SSW in combination with constant bubbling in the plunging-jet system cues an inherent SML formation with certain caveats such as the lack of direct atmospheric forcings. We will elaborate on this theoretical background and the potential consequences in using 6 m deep waters in the discussion.
Reviewer 6.) You state that atmospheric measurements of TEP and CSP are scarce, and further describe TEP as polysaccharide-based and CSP as protein-based. However, to my knowledge there are several studies from different groups investigating polysaccharides and proteins in SSA. Integrating your findings more systematically, for example in terms of aerosol size distributions and the potential contribution of gel-like versus non-gel-like polysaccharides and proteins, could strengthen the contextualization of your study and better embed it within the existing literature.
Response We have elucidated on parts of the existing literature in the introduction (L40-L49), while extended references were used to better contextualize our results in the discussion (L400-L403; L455-L463). We will work on a more systematic approach in discussing our results in the light of the currently available literature and also work on integrating further relevant references.
Reviewer 7.) I was confused throughout the manuscript as to whether you are referring to results from seawater or aerosol samples. Please ensure consistent nomenclature (e.g. abbreviations, brackets, indices) or a clearer overall structure to make this distinction unambiguous. I may have misunderstood some aspects due to this lack of clarity. What exactly do you mean by “marine samples” (e.g. L22)? Aren’t all samples marine in this study?
Response We will implement both suggestions: We will ensure a consistent nomenclature and improve the overall structure of the manuscript. With regard to L22 'marine samples' should be replaced by SSW samples.
Reviewer 8.) Abbreviations are introduced repeatedly throughout the manuscript, which makes the text harder to follow. It would greatly improve readability if all abbreviations were defined once (ideally at first occurrence) and then used consistently thereafter, at least within the main text. I recommend carefully reviewing the manuscript from beginning to end to ensure consistent usage. For example, the term “subtropical front (STF)” is introduced multiple times in slightly different forms (e.g., L85, L186, several figure captions, Table A4, L507, L516). A similar pattern occurs for “SSW” (e.g., Lines 97, 104, 185, 246) and other abbreviations such as Chla and STW.
Response We will introduce all abbreviations only once in the main text at their first occurrence. We will continue in introducing all necessary abbreviations in the abstract, in figure, and table captions (e.g. L246) to ensure that these manuscript elements can be understood without referring to the main text. We are convinced that this will increase the overall clarity of the manuscript.
Reviewer 9.) Since you measured size-resolved hydrogels, did you consider visualizing their relative size distribution in an additional figure and comparing it with the few existing studies, either from field observations or laboratory experiments? Also comparing them between seawater and aerosol could be interesting.
Response This is a good idea and we will highlight the size distribution by implementing additional figures and amend our discussion accordingly.
Reviewer More specific comments:
L14–15 (&L67/68): “most atmospheric measurements fail to detect marine hydrogels” appears to be a very strong statement. Please rephrase it in a slightly more cautious way, e.g. in the sense that atmospheric measurements of marine hydrogels are still scarce and only limited studies exist.
Response As suggested, we will rephrase this sentence more carefully:
L14: Yet, studies on atmospheric measurements detecting marine hydrogels are still limited.
L67: Although hydrogels seem to be important components of the biogenic aerosol pool, data on hydrogels in aerosols are still scarce.
Reviewer Abstract in general: Please introduce the terms TEP and CSP in the abstract, as these are central to the manuscript. Since you use staining-based approaches, these are also the appropriate and standard terms. Replacing them with “carbohydrate- and protein-enriched hydrogels” without proper introduction to their chemical relationship feels potentially misleading.
Response We will introduce the terminology transparent exopolymer particles (TEP) and Coomassie stainable particles (CSP) while shortening the abstract elsewhere. We agree with the reviewer that the correct terminology should be included in the abstract.
Reviewer L16-17: Please be more cautious with the use of the term “representative regimes.” First, as far as I understand, every region has its own specific characteristics; therefore, please specify for which parameters or aspects you consider them representative. Second, your sample size is quite limited (which is of course common for this type of field measurements), but it should be acknowledged that the dataset may not fully capture seasonal and meteorological variability required for a complete sense of representativeness. In this context, the term “representative” appears potentially misleading, as it could imply that these regions are comprehensively characterized. Furthermore, it feels somewhat controversial that within this relatively small sampling area you already define four distinct regimes with apparently very different characteristics, this also raises the question of what exactly is meant by “representative” in this context.
Response We agree with the reviewer that the term 'representative' is misleading given the few data points at hand.
It is well established that subantarctic water (SAW) is enriched in macronutrients, particularly nitrate, while the subtropical water (STW) is enriched in micronutrients, particularly iron (Moore et al., 2013). The SAW and STW water masses are characterized by opposing nutrient limitations (Moore et al., 2013). The Southern Ocean (including the SAW) is commonly referred to as a high-nutrient low-chlorophyll (HNLC) area, in which iron addition alleviates phytoplankton growth limitation (Moore et al., 2013; Basterretxea et al., 2023). In the subtropical front (STF), on the other hand, where STW and SAW mix, phytoplankton blooms occur year around due to ideal nutrient conditions (Murphy et al., 2001). Due to the underlying Chatham Rise east of New Zealand, the STF is fixed in its location (Murphy et al., 2001). Waters east of New Zealand therewith represent a region in which major biogeochemical regimes can be encountered within small spatial scales. This clear spatial separation became also apparent during our campaign as e.g., the SAW was marked by elevated nitrate concentrations (Saint-Macary et al., 2023; Sellegri et al., 2023).
In this context, we can associate the three main identified regimes (STF, SAW, STW) to major biogeochemical water types, however, acknowledge that i) this was not well enough explained, ii) also seasonal, interannual, and decadal variability shape these major biogeochemical regimes on local to regional scales (Murphy et al., 2001; Basterretxea et al., 2023), and iii) short-term fluctuations may introduce considerable variance within each specific region. Therefore, given the small data set at hand, the dataset should be regarded as only cautiously 'representative'. We will rephrase related sentences and sections, including the abstract, accordingly.
Reviewer L24–25: “…and significantly contribute to the atmospheric pool of cloud condensation and ice nuclei.” This is a very strong statement for the abstract, given that neither cloud condensation nuclei nor ice nuclei were measured in this study, nor was any related modelling performed. Please consider rephrasing this statement more cautiously or removing it altogether.
Response We agree and will formulate a more cautious statement:
L24: Our results suggest that hydrogels may complement a considerable fraction of ambient marine aerosols with potential contributions to the atmospheric pool of cloud condensation and ice nuclei.
Reviewer Line 41, Line 457, Line 485: In these sections, enzymes and proteins appear to be used interchangeably, which appears confusing. While enzymes are indeed a specific class of proteins, they are typically discussed in the context of their catalytic function to initiate certain chemical reaction. However, their catalytic role does not seem to have any relevance for your train of thoughts. It would be helpful to clarify which term is intended in each case. Please chose for one of the two.
Response Thank you for highlighting this source of potential confusion. Since our measurements do not distinguish enzymes from other protein fractions, the appropriate term and reference in the context of our dataset is 'protein'. However, some literature reports explicitly on enzymes (e.g., RuBisCo, which is one of the most abundant proteins in phytoplankton, Alsante et al., 2023). In this context, we would like to further refer to 'enzymes'. We will revise the introduction to appropriately introduce the term protein i.e., by highlighting the different classes which contribute.
Reviewer L44/45: ‘Proteins […] represent persistent ice nuclei’…. this statement is surprising in its current form, as protein-based ice nuclei are generally considered among the most fragile types of ice nuclei rather than persistent ones. Furthermore, could you rephrase this sentence to clearly specify which marine proteins found in the SSA are meant, and in what sense they act as particularly efficient or persistent ice nuclei?
Response We agree with the reviewer, and will clarify the statement by removing the misleading term 'persistent' while extending our explanation. What we intended to convey was that denaturation by heating does not always remove IN activity of proteins, while activation sites of certain proteins are maintained over several thawing and freezing cycles (Alsante et al., 2023; Cascajo-Castresana et al., 2020).
Clarification L44-45: ..., where they contribute to ice nucleation activity, although they are constantly exposed to denaturation or repeated thaw and freeze cycles (Alsante et al., 2023; Cascajo-Castresana et al., 2020).
Reviewer L46: “enabling ice cloud formation”… consider rephrasing to “enabling ice formation in clouds.”
Response We will follow this suggestion.
Reviewer L48/49: Please add for which region and which temperature range this information holds true
Response We agree that this information is crucial and will clarify this accordingly. The authors of the respective paper analyzed whether a global parameterization found a good agreement with a data set on marine ice nuclei collected during several campaigns in the Southern and Northern Hemisphere (Hartmann et al., 2025). Within a temperature range of -15°C to -20°C, marine polysaccharides potentially contribute 44% of INPs.
Reviewer L71/72: ‘in SSA or ambient aerosols’…. please choose one term only.
Response We will introduce and clarify the terms. We defined the terms accordingly: 'Sea spray aerosols (SSA)' can be directly linked to sea spray emissions while the term 'ambient aerosols' acknowledge all contributions to the ambient pool, including secondary and primary aerosols from local as well as remote sources. Our chamber study addresses SSA, however, some of the cited literature addresses hydrogels from ambient aerosol samples, which were for example collected at the Cape Verde Atmospheric Observatory (CVAO, Van Pinxteren et al., 2022). Please also refer to the subsequent comment.
Reviewer L72: “…and none over the Southern Hemisphere…” This statement is certainly correct. However, as I understand it, your study did not include ambient atmospheric measurements of hydrogels in the Southern Hemisphere, but rather tank experiments using water collected from the Southern Hemisphere. Tank studies are certainly valuable; however, it is important to clearly distinguish between laboratory experiments and ambient atmospheric observations so that readers are not misled. In its current form, this statement appears potentially misleading and should be rephrased accordingly.
Response We agree that it is important to clearly distinguish between artificially generated SSA and the ambient aerosol pool. While the referred sentence is still part of the introduction and represents a conclusive, general statement with reference to the already introduced literature including tank studies and ambient aerosol samples, we specified our approach in a subsequent sentence in L75. We would therefore like to keep the conclusive sentence while we will clarify our subsequent statement.
L75: Based on a comparison of four regimes in the Southwestern Pacific Ocean, which were distinct in their biological activity, we here investigate biological factors influencing the occurrence of hydrogels in artificially generated SSA.
We would like to add at this point that our approach has certain advantages in comparison to assessing only ambient aerosols i.e., we were able to directly compare hydrogels in SSA to local, biogeochemical conditions and do not have to account for potential contamination with hydrogels from close-by locations or remote areas, which could have already undergone atmospheric processing.
Reviewer L82: “Surface seawater samples”?
Response Sample descriptions will be unified throughout the whole manuscript and introduced abbreviations will be consistently used, e.g. surface seawater samples (SSW).
Reviewer L89: The tank was constructed from “stainless steel.” In Lines 347 and 363, you argue that waters in the region of your study are depleted in iron, which influences (inhibits) microbial growth. Would corrosion under saline conditions potentially lead to an increase in dissolved iron within the tank water? If so, could this have altered the biogeochemical and microbial conditions during the experiments through the release of iron ions?
Response We agree with the reviewer that we should briefly assess the potential influence of the chamber on SSW biogeochemistry and will include a brief statement along the following line of argument: Iron availability is enhanced by complexation with carbohydrates serving as organic ligands (Hassler et al., 2011). However, as the residence time of the water in the tank was very short (~4 min), phytoplankton and bacterial cell metabolic rates (including iron uptake rates, Hudson and Morel, 1990) limit the ability of cells to influence water biogeochemistry within this short time window. Subsequently to iron uptake, anabolic rates in phytoplankton cells also relate to the availability of sun light. Our SSA chamber was setup in the lab and not exposed to ambient solar light. In summary, we do not expect that the set-up will interfere with the biogeochemistry and metabolism of cells.
Reviewer L90 ‘….,which is equivalent to a water depth of 10 cm”. This statement is not clear to me. Does this refer to a water level of 10 cm in the tank (i.e., the filling height)?
Response Yes, 10cm is the filling height in the tank. We will clarify this sentence.
Reviewer L91: ‘The residence time of the water in the chamber was ~4 min.’ If the water remains in the tank for such a short time while the ship continues along its cruise track, it seems likely that a single tank aerosol sample could be influenced by water masses with varying biogeochemical properties over the ~24-hour sampling period. Did you collect multiple seawater samples to account for this variability? If not, how did you ensure that the single corresponding seawater sample is representative of the water that actually generated the aerosols?
Response We collected two discrete SSW samples within the SSA sampling period to determine TEP and CSP (please also refer to L180 to L183). SSA collection started at ~11:00 AM and lasted until the next day ~9:00 AM. TEP and CSP SSW samples as well as all further biogeochemical data were collected at around 8:00 AM and 4:00 PM. Afternoon and next-morning samples were averaged to be compared to the respective SSA sample. As mentioned above, we will elucidate in more detail the biogeochemical differences encountered across regimes. We will also add a clearer figure, so the reader is able to follow and judge upon the pooled sample sets. Biogeochemical, SSW and SSA hydrogel data will be simultaneously displayed in a new figure, replacing the current Figure 4.
Reviewer Section 2.2: Please add a suitable reference for each measurement parameter, where the full measurement protocol is described in detail. Furthermore, please clarify where the sodium measurements are described in the methods section.
Response We will add additional information to the methods used, explicitly with regard to the Na⁺ measurements.
Sodium concentrations in SSA were determined using a 13-stage DEKATI cascade impactor operated at a flow rate of 7 L min⁻¹. Quartz filters served as impaction substrates. To increase aerosol mass for the later on chemical analysis, filters from adjacent impactor stages were pooled into three size fractions: stages 1-6 (30-500 nm), stages 7-8 (500 nm to 1 µm) and stages 9-13 (1-10 µm). Samples were extracted in Milli-Q water by 30 min of sonication. Extracts were analyzed by ion chromatography for the quantification of the main inorganic ions (Sandrini et al., 2016). An IonPac CS16 3 × 250 mm Dionex separation column with gradient MSA elution and an IonPac AS11 2 × 250 mm Dionex separation column with gradient KOH elution were deployed for cations and anions, respectively.
Reviewer L101: Did you measure free or combined amino acids?
Response We measured total dissolved and total amino acids (DAA/TAA), i.e. the dissolved fraction included hydrolysable (combined) and free amino acids. This is described in more details in Barthelmeß et al., 2025 and the references within, however, we will add the relevant references with regard to the applied methods also in this manuscript.
Reviewer L107: Could you clarify why 10 mL of Milli-Q water was filtered for the blank, while 30-80 mL were filtered for the field samples? This difference raises the question of whether the blank and samples are directly comparable.
Response The purpose of the Milli-Q blank is to account for background contamination originating from the filters, handling and dyes rather than to replicate the volume of the field samples. Milli-Q water has a very low organic carbon concentration and is filtered to exclude any particles (filtered through 0.2 µm; ~8 µM TOC). In contrast, the dyes consist of complex organic molecules. Furthermore, Coomassie Blue is diluted with 0.2 µm-filtered seawater, as dilution with pure Milli-Q water can promote colloid formation of the dye. Both dye solutions are filtered prior to use, and both the blank and field samples receive the same volume of dye (1 mL). Therefore, filtering 10 mL of Milli-Q water is sufficient for the blank because its purpose is to wet the filter, to assess contamination introduced by the dye and the filter and address any contamination due to handling and staining procedures. The larger sample volumes (30–80 mL) are filtered to collect adequate amounts of environmental particulate material for analysis. We hope this sufficiently addresses the reviewers concern and will add a small explanation in the method section.
Reviewer L114-120 This section might be more appropriately placed in Section 2.1 (Sampling)
Response We agree with the reviewer and will adjust the method section accordingly.
Reviewer L121–124: For aerosol filters, a “dual-staining approach” was used due to the limited availability of filters. Please elaborate on how this methodological difference between seawater and aerosol filter processing may have introduced artefacts and potentially affected the comparability of the resulting data.
Response We compared the results of the dual-staining and single-staining approaches previously. We validated whether dual-stained model substances and environmental samples reproduced single-stained filters (Scheidemann et al., in prep.). Applied in the right order (first Coomassie Blue than Alcian Blue), the dyes do not interfere with each other. With respect to the SSA hydrogel data set, we found no indicators that the dual-staining procedure introduced systematic artefacts (supplementary information). We would like to highlight, that the enumeration and size measurements with respect to hydrogels in SSW and SSA were conducted by applying the RGB-channel based analysis (Engel, 2009), while the distinction of the SSA chemical composition was based on separating different shades of blue by hue values, saturation and lightness. We are aware that this approach represents a compromise. To clarify this methodological difference and its potential implications for data comparability, we will expand the description of the dual-staining approach in the main manuscript. We acknowledge and will highlight that our dual-staining approach and the subsequent enumeration and composition estimate of hydrogels in SSA may entail an inherent bias, which we cannot fully resolve.
Reviewer L141: ‘In comparison to hydrogels in SSW,” …. reads strange in combination with the rest of the sentence. Is it possible to eliminate this part?
Response We will rephrase this sentence. The information with regards to the SSW comparison will be described elsewhere:
L141: The SSA collection on filters was conducted over a time period of 18.5 to 24 hours (Table A2), and only one replicate could be derived.
L180: For the comparison of SSW data with hydrogels in SSA, consecutive SSW data points were pooled to derive a mean over the time of SSA generation, including an afternoon (4:00 PM) and next-morning SSW sample (8:00 AM).
Reviewer L147: Consider rephrasing the section title “Data processing and Statistics” or similar.
Response We will rephrase the title and consider the suggested title a good idea.
Reviewer L149–155: I appreciate that statistical tests were performed to assess the suitability of the microscopic data. However, it remains unclear what conclusions can be drawn from the results regarding normality and homogeneity. A brief clarification (ideally in one sentence) would be helpful. In addition, since the data appear to violate these assumptions in several cases, it would be important to explain what consequences were drawn from this (e.g., data interpretation, change in measurement protocol,…). How does this affect the reliability of your results?
Response We thank the reviewer for this comment and acknowledge that the information provided in this paragraph is incomplete in its current form. A brief explanation is only provided in L176 to L178. These tests are the basis for selecting the subsequent non-parametric statistical tests with rank-transformed data. The subsequent statistical approaches are more robust against violations of normality and homogeneity assumptions and therefore provide an appropriate framework for evaluating the observed patterns. In general, small, biological sample sets often violate both assumptions. In this case, the violation was consolidated by CSP distribution which was influenced by potential soot particles in the SSW samples. We will add a brief clarification in L149–155.
L149-155: For TEP and CSP, normality was not always given but for CSP even the assumption of homogeneity was violated. Therefore, we have subsequently applied non-parametric statistical tests with rank-transformed data, which are more robust against these violations. Furthermore, the SSW samples which were potentially contaminated by soot were excluded from downstream analysis.
Reviewer L155: Please specify what is meant by “certain parameters”.
Response We will specify all relevant parameters and whether they were normally and heterogeneously distributed.
Reviewer L161: “If the EF is smaller than 1,…..”I think this statement typically applies to enrichment in the SML. For aerosol enrichment, this interpretation does not appear directly appropriate.
Response This is certainly true once we have adjusted our approach to comply with the commonly applied method in atmospheric chemistry to calculate SSA enrichment, we will also adjust the description of the enrichment factor and as outlined above.
Reviewer L191: How do you expect heavy rainfall to have impacted seawater at 6 m depth? Was the upper water column strongly diluted? Do you still consider the assumption of approximately 10 g Na⁺/L (Line 163) to be valid for your subsequent enrichment calculations? Please clarify and streamline how relevant this observation is for the discussion.
Response We thank the reviewer for this comment. We considered whether accounting for variations in seawater salinity would improve the enrichment calculations. However, we prefer to retain a generic seawater Na⁺ concentration (approximately 10 g Na⁺ L⁻¹) for the enrichment calculations.
The expected SSA enrichment factors in our study are on the order of 10³, whereas the observed variability in SSW salinity was comparatively small. Salinity differed by approximately 1 PSU between the investigated oceanic regimes (Sellegri et al., 2023). After the documented rainfall event, we observed a slight salinity drop in this area at 6 m depth by around 0.2 PSU (SAW, 22.03.2020). However, these salinity changes correspond to only minor variations in Na⁺ concentration and would therefore have a negligible effect on the calculated SSA enrichment factors.
To avoid introducing unnecessary complexity without affecting the results, we therefore retain the generic seawater Na+concentration for all enrichment calculations. We will clarify this rationale in the revised manuscript and streamline the discussion of the rainfall event to emphasize that its influence on the enrichment calculations is expected to be negligible.
Reviewer Table 1: Do these enrichment factors refer specifically to CSP, TEP, or to hydrogels in general?
Response The total abundance of SSW particles (including CSP and TEP) were calculated and related to the total abundance of hydrogels in SSA. Conclusively, the enrichment factors represent hydrogels in general. We will clarify this in the caption.
Reviewer L229: “SSA hydrogel abundance and area varied greatly along the cruise track”… this statement is potentially misleading, as the study is based on tank experiments rather than in situ atmospheric measurements along the cruise track.
Response We only partly agree with the reviewer. We have described in detail how we assessed SSA generated from SSW along the cruise track and via the application of the plunging-jet system, so it can be assumed that the reader is aware of the sampling method. However, we will introduce our results in this section with a reminder i.e., by using the term 'artificially generated SSA'.
L229: Hydrogel abundance and area of artificially generated SSA varied greatly along the cruise track...
Reviewer Figure 6: Add significance to correlation plots. Could you make it visually more obvious if the parameters were measured in the seawater or in the aerosols?
Response We will adjust the correlation based heatmaps by including significances and improving the visualization by clearly distinguishing between SSA and SSW data.
Reviewer L340: “…high amino acid degradation indices…” sounds interesting but in the current form not entirely clear. Could you briefly elaborate and clarify what this implies?
Response Thank you for your interest. Degradation indices are based on dissolved amino acid concentration either integrating differences in the relative DAA composition (specific, relatively enriched DAA imply fresh phytoplankton production, while elevated fractions of other DAA are characteristic for organic matter degradation) or the molar fraction of DAA contributing to DOC. This is described in greater detail elsewhere (Barthelmeß et al., 2025 and literature within), however, we will provide a more comprehensive explanation and the relevant original literature to the interested reader within the discussion section. To comply with the recommendations of the second reviewer, we will remove these results and references from the result section.
Background: For example, the degradation index based on amino acid composition can be determined by a principal component analysis (Kaiser and Benner, 2009; Dauwe et al., 1999). Phytoplankton-associated bacteria entail three major groups, of which one is specialized in the consumption of GlX, Arg, Leu, Iso, and Val (Ferrer-González et al., 2021). Moreover, certain amino acids are preferred over others in bacterial degradation experiments (Amon et al., 2001). Elevated relative abundances of the non-proteinogenic amino acid GABA have been associated with enhanced bacterially driven organic matter degradation. Higher proportions of Ser, Gly, and Ala are characteristic of refractory organic matter (e.g., Kaiser and Benner, 2009; Dauwe et al., 1999).
Reviewer L380-385: Please reconsider whether the extended repetition of results from Lines 253–255 is necessary here, or if this section could be condensed.
Response We will condense these sentences.
L380-385: To the best of our knowledge, we are the first to report hydrogel concentrations in SSA across biogeochemical regimes of the Southern Hemisphere Ocean, showing pronounced regional variability consistent with differences in biological activity. Although our data set is small, our findings align well with previous findings in which aerosol emission was pronounced above productive oceanic regions based on satellite-derived Chl a concentration (Long et al., 2014; O’Dowd et al., 2015; O’Dowd et al., 2004).
Reviewer L386-394 This section appears rather long and partially contradictory. It could potentially be shortened to something like:’ Aller et al. (2017) reported TEP and CSP as 1–100% of SSA particle mass using a colorimetric approach, and although methods slightly differ, their SSA concentrations (2–14 μg XG and BSA equivalents m-3) are broadly consistent with our hydrogel area in SSA when translated into equivalent concentrations using the SSW-derived relationship.’?
Response We thank the reviewer for this suggestion and will shorten the relevant sentences.
L386-394: Aller et al. (2017) reported TEP and CSP as 1–100% of artificially generated SSA and ambient aerosol particle mass using a colorimetric approach, and although methods differ, their SSA and ambient air concentrations (2 and 14 μg XG and BSA equivalents m-3, respectively) are broadly consistent with our hydrogel area in SSA when translated into equivalent concentrations. In seawater, colorimetrically determined concentrations correspond to the microscopically determined area, which were shown to linearly correlate (Dreshchinskii and Engel, 2017; Cisternas-Novoa et al., 2014; Passow and Alldredge, 1995).
Reviewer L403-404: For which size range were the enrichment factors calculated?
Response Agreed, this is an important detail, we will report the size range for which Van Pinxteren et al., (2022) calculated the mentioned enrichment factors (TEP: 5-10µm).
Reviewer L420 “bulk seawater” or “underlying seawater”? The SML is also seawater, so the terminology should be used more precisely.😉
Response Thank you for spotting this inconsistency: We referred to the underlying surface seawater, which is commonly used to reference the SML.
Reviewer L445–446: Did you account for the conversion between (spherical) radius and diameter? As I understand, hydrogels were characterized in your study using diameters, whereas the discussion appears to refer to radii from the cited reference. Please check this for consistency. Additionally, consider specifying a size range for the “film fragments”.
Response We will clarify this in the manuscript and add the missing information, including our assumption.
We are aware of the difference in terms of metrics. The underlying assumption is that droplets with a diameter smaller than 4µm (r=2µm) can only carry hydrogels with an area limited by its droplet size. Only above a SSA diameter of 4µm, jet droplets start to slightly dominate the SSA population, i.e. hydrogels ranging from 5 to 10µm in diameter are more likely associated to the collapse of the bubbles cavity center (jet droplet ejection) than film fragmentation. Film fragmentation starts to be the dominating process of SSA production at an approximate r=0.05µm (d=0.1µm) and thus fall below the smallest sizes of hydrogels (d>0.5µm) assessed in this study. Film droplet production exhibits a peak around a diameter of ~1µm (r=0.5µm), i.e. the underlying mechanism responsible for transporting SSA hydrogels between 0.5 and 2.5µm in diameter is most likely dominated by film fragmentation.
Clarification L447-449: We assume that SSA droplets <4 µm in diameter can only carry hydrogels smaller than the droplet itself. Hydrogels 5–10 µm in diameter are therefore more likely transported by jet droplets, which dominate above ~4 µm SSA diameter. In contrast, smaller hydrogels (0.5–2.5 µm) are likely transported predominantly by film droplets, as film fragmentation produces SSA with a peak diameter of ~1 µm.
Reviewer L471: This section is titled “4.3.1 ….”, but there is no section 4.3.2. Should it be 4.4 instead?
Response We considered this a sub-section of 4.3 focusing on hydrogel dynamics in the SSW potentially explaining features with relevance for aerosolized hydrogels. However, we agree with the reviewer that it should be either representing section 4.4, in which the outlined relevance is clarified or be integrated into section 4.3. As we need to considerably restructure the discussion also based on the recommendation of the second reviewer, we will find suitable solution.
Reviewer Figure 7 is intended to summarize the hypothesized hydrogel formation and aerosol composition. While I find the visual design appealing, I am currently unable to extract clear mechanistic information on the underlying processes from this figure. It is not clear how the different regions (STF, Mix, SAW, STW) are represented and how they relate to biogeochemical seawater processes and aerosols. Furthermore, it is difficult to distinguish whether these shapes represent bacteria, algae, or gel structures. In addition, the color scheme is somewhat confusing, as bright blue appears to represent both TEP and the air–sea interface (SML?), while dark blue represents CSP, the nanolayer (?) and underlying water. What do we learn from the representation of the sun and the sun light in the water column? I recommend revising the figure to improve clarity and ensure that it can be understood more intuitively, ideally without requiring extensive explanation in the caption. Furthermore, I suggest considering the revised version of Figure 7 as a potential Table of Content figure rather than a main figure, as it appears more conceptual than data-driven.
Response We thank the reviewer for this feedback and will work on simplifying the figure, including a more intuitive structure and color scheme. This point was also raised by the third reviewer. We also like the reviewer's suggestion, and will use it as a figure of content instead.
Reviewer I am aware of a recent preprint by Freney et al. (https://doi.org/10.5194/egusphere-2026-87), which appears to be based on the same campaign(?) and the same tank experiments(?). While the two studies clearly address different aspects, there seems to be some overlap in the underlying data and context. It may be beneficial for the authors to consider cross-referencing this work where appropriate, as it could help to strengthen their own conclusions.
Response We will cross-reference this study. Indeed, the data stem from the same campaign.
Reviewer Tables A1 and A2 (Longitude, Latitude, Air Volume, etc.): replace commas “,” with “.”
Response Thank you for noticing this, we will correct it.
Reviewer Table A2: specify the start and end times of aerosol sampling (including exact hours). Add “+” to “Na+”. Replace “Filtration time” with “Sampling time”.
Response We will add the missing details.
Reviewer Table A4 (where is Table A3?): ensure the entire table is fully visible on the page.
Response The overview table containing information on mean concentrations etc. across regimes was supposed to be labeled as Table A3, the following table including information on complementary F-statistics is Tab. A4. We will ensure that everything is fully displayed.
We would again like to thank the reviewer for his/her eye for detail and his/her genuine interest. We acknowledge the invested work to improving this manuscript!
Literature
Alsante, A. N., Thornton, D. C. O., & Brooks, S. D. (2023). Ice nucleation catalyzed by the photosynthesis enzyme RuBisCO and other abundant biomolecules. Communications Earth and Environment, 4(1), 1–9. https://doi.org/10.1038/s43247-023-00707-7
Amon, R.M.W., Fitznar, H. P., Benner, R. (2001). Linkages among the bioreactivity, chemical composition, and diagenetic state of marine dissolved organic matter. Limnology and Oceanography 2001, 46 (2), 287-297. DOI: 10.4319/lo.2001.46.2.0287.
Asmussen-Schäfer, F., Ribas-Ribas, M., Wurl, O., Friedrichs, G. (2026). Linking surface coverage with surfactant activity to refine the role of surfactants for air-sea gas exchange. Biogeosciences, 23(9), 3159–3178. https://doi.org/10.5194/bg-23-3159-2026.
Barthelmeß, T., Cristi, A., Deppeler S., Safi, K., Sellegri, K., Law, C. S. and Engel, A. (2025). Pronounced Diel Cycling of Dissolved Carbohydrates and Amino Acids in the Surface Ocean and across Diverse Regimes. Environmental Science and Technology, 1(59), 419–429. https://pubs.acs.org/doi/10.1021/acs.est.4c00491.
Basterretxea G., Font-Muñoz, J.S., Hernández-Carrasco, I., Sañudo-Wilhelmy, S.A. (2023). Global variability of high-nutrient low-chlorophyll regions using neural networks and wavelet coherence analysis. Ocean Science, 19(4), 973-990. https://doi.org/10.5194/os-19-973-2023.
Cascajo-Castresana, M., David, R. O., Iriarte-Alonso, M. A., Bittner, A. M., & Marcolli, C. (2020). Protein aggregates nucleate ice: The example of apoferritin. Atmospheric Chemistry and Physics, 20(6), 3291–3315. https://doi.org/10.5194/acp-20-3291-2020.
Cisternas-Novoa, C., Lee, C., Engel., A. (2014). A semi-quantitative spectrophotometric, dye-binding assay for determination of Coomassie Blue stainable particles. Limnology and Oceanography: Methods, 12, 604-616. https://doi.org/10.4319/lom.2014.12.604.
Crocker, D. R., Kaluarachchi, C. P., Cao, R., Dinasquet, J., Franklin, E. B., Morris, C. K., Amiri, S., Petras, D., Nguyen, T., Torres, R. R., Martz, T. R., Malfatti, F., Goldstein, A. H., Tivanski, A. V., Prather, K. A., & Thiemens, M. H. (2022). Isotopic Insights into Organic Composition Differences between Supermicron and Submicron Sea Spray Aerosol. Environmental Science and Technology, 56(14), 9947–9958. https://doi.org/10.1021/acs.est.2c02154.
Dauwe, B., Middelburg, J. J., Herman, P. M. J., and Heip, C.H.R. (1999). Linking diagenetic alteration of amino acids and bulk organic matter reactivity, Limnology and Oceanography, 44, 1809–1814. https://doi.org/10.4319/lo.1999.44.7.1809.
Dreshchinskii, A. und Engel, A. (2017). Seasonal variations of the sea surface microlayer at the Boknis Eck Times Series Station (Baltic Sea). Journal of Plankton Research, 39 (6). pp. 943-961. https://doi.org/10.1093/plankt/fbx055.
Engel, A. (2009). Determination of Marine Gel Particles, Practical guidelines for the analysis of seawater (O. Wurl, B. Raton, & [u.a.] (eds.)). CRC Press.
Engel, A., Friedrichs, G., Krall, K., Jähne, B. (2026). Wind-induced collapse of the biopolymeric surface microlayer induces sudden changes in sea surface roughness. Biogeosciences, 23(6), 2101-2117. https://doi.org/10.5194/bg-23-2101-2026.
Ferrer-González, F. X., Widner, B., Holderman, N. R., Glushka, J., Edison, A. S., Kujawinski, E. B., Moran, M. A. (2021). Resource partitioning of phytoplankton metabolites that support bacterial heterotrophy. ISME Journal, 15 (3), 762-773. DOI: 10.1038/s41396-020-00811-y.
Hartmann, S., Schrödner, R., Hassett, B.T., Hartmann, M., van Pinxteren, M., Fomba, K.W., Stratmann, F., Herrmann, H., Pöhlker, M., and Zeppendeld, S. (2025). Polysaccharides–Important Constitutes of Ice-Nucleating Particles of Marine Origin. Environmental Science and Technology, 59, 5098-5108. https://doi.org/10.1021/acs.est.4c08014.
Hassler, C. S., Schoemann, V., Nichols, C. M., Butler, E. C. V., & Boyd, P. W. (2011). Saccharides enhance iron bioavailability to southern ocean phytoplankton. Proceedings of the National Academy of Sciences of the United States of America, 108(3), 1076–1081. https://doi.org/10.1073/pnas.1010963108
Hudson, R.J.M., Morel, F.M.M. (1990). lron transport in marine phytoplankton: Kinetics of cellular and medium coordination reactions. Limnology and Oceanography, 35(5), 1002–1020. https://doi.org/10.4319/lo.1990.35.5.1002.
Kaiser K., Benner, R. (2009). Biochemical composition and size distribution of organic matter at the Pacific and Atlantic time-series stations. Marine Chemistry, 113(1-2), 63-77. https://doi.org/10.1016/j.marchem.2008.12.004.
Long, M. S., Keene, W. C., Quinn, P. K., & Bates, T. S. (2014). Light-enhanced primary marine aerosol production from biologically productive seawater. Geophysical Prospecting, April, 6413–6419. https://doi.org/10.1002/2014GL061184.Received
Moore, C. M., Mills, M. M., Arrigo, K. R., Berman-Frank, I., Bopp, L., Boyd, P. W., Galbraith, E. D., Geider, R. J., Guieu, C., Jaccard, S. L., Jickells, T. D., La Roche, J., Lenton, T. M., Mahowald, N. M., Marañón, E., Marinov, I., Moore, J. K., Nakatsuka, T., Oschlies, A., … Ulloa, O. (2013). Processes and patterns of oceanic nutrient limitation. Nature Geoscience, 6(9), 701–710. https://doi.org/10.1038/ngeo1765
Murphy, R. J., Pinkerton, M. H., Richardson, K. M., Bradford‐Grieve, J. M., & Boyd, P. W. (2001). Phytoplankton distributions around New Zealand derived from SeaWiFS remotely-sensed ocean colour data. New Zealand Journal of Marine and Freshwater Research, 35(2), 343–362. https://doi.org/10.1080/00288330.2001.9517005
O’Dowd, C., Ceburnis, D., Ovadnevaite, J., Bialek, J., Stengel, D. B., Zacharias, M., Nitschke, U., Connan, S., Rinaldi, M., Fuzzi, S., Decesari, S., Cristina Facchini, M., Marullo, S., Santoleri, R., Dell’anno, A., Corinaldesi, C., Tangherlini, M., & Danovaro, R. (2015). Connecting marine productivity to sea-spray via nanoscale biological processes: Phytoplankton Dance or Death Disco? Scientific Reports, 5(May), 1–11. https://doi.org/10.1038/srep14883
Passow, U., Alldredge, A. (1995). A dye‐binding assay for the spectrophotometric measurement of transparent exopolymer particles (TEP). Limnology and Oceanography, 40(7), 1326–1335. https://doi.org/10.4319/lo.1995.40.7.1326.
Rickard P., Uher, G., Upstill-Goddard, R., (2022). Photo-Reactivity of Surfactants in the Sea-Surface Microlayer and Subsurface Water of the Tyne Estuary, UK. Geophysical Research Letters, 49(4). https://doi.org/10.1029/2021GL095469
Robinson, T. B., Wurl, O., Bahlmann, E., Jürgens, K., & Stolle, C. (2019). Rising bubbles enhance the gelatinous nature of the air–sea interface. Limnology and Oceanography, 64(6), 2358–2372. https://doi.org/10.1002/lno.11188
Sabbaghzadeh, B., Upstill-Goddard, R. C., Beale, R., Pereira, R., Nightingale, P. D. (2017). The Atlantic Ocean surface microlayer from 50◦ N to 50◦ S is ubiquitously enriched in surfactants atwind speeds up to 13ms−1, Geophys. Res. Lett., 44, 2852–2858. https://doi.org/10.1002/2017GL072988.
Saint-Macary, A. D., Marriner, A., Barthelmeß, T., Deppeler, S., Safi, K., Santana, R. C., Harvey, M., & Law, C. S. (2023). DMS cycling in the Sea Surface Microlayer in the South West Pacific: 1. Enrichment potential determined using a novel sampler. Ocean Science, 19(1), 1–15. https://doi.org/10.5194/egusphere-2022-499
Sandrini, S., Van Pinxteren, D., Giulianelli, L., Herrmann, H., Poulain, L., Facchini, C.M., Gilardoni, S., Rinaldi, M., Paglione, M., Turpin, B.J., Pollini, F., Bucci, S., Zanca, N., Decesari, S. (2016). Size-resolved aerosol composition at an urban and a rural site in the Po Valley in summertime: Implications for secondary aerosol formation. Biogeosciences, 16(17), 10879-10897. https://doi.org/10.5194/acp-16-10879-2016
Sellegri, K., Harvey, M., Peltola, M., Saint-Macary, A., Barthelmeß, T., Rocco, M., Moore, K. A., Cristi, A., Peyrin, F., Barr, N., Labonnote, L., Marriner, A., McGregor, J., Safi, K., Deppeler, S., Archer, S., Dunne, E., Harnwell, J., Delanoe, J., … Law, C. S. (2023). Sea2Cloud: from biogenic emission fluxes to cloud properties in the South West Pacific. Bulletin of the American Meteorological Society, 1017–1043. https://doi.org/10.1175/bams-d-21-0063.1.
Silva, A., Nikzad, S., Barthelmeß, T., Engel., A., Herrmann, H., van Pinxteren, M., Wirtz, K., Wurl, O., and Schartau, M. (2025). Meta-analytical insights into organic matter enrichment in the surface microlayer. EGUsphere [preprint], https://doi.org/10.5194/egusphere-2025-4050.
Sun, C. C., Sperling, M., & Engel, A. (2018). Effect of wind speed on the size distribution of gel particles in the sea surface microlayer: Insights from a wind-wave channel experiment. Biogeosciences, 15(11), 3577–3589. https://doi.org/10.5194/bg-15-3577-2018
Van Pinxteren, M., Robinson, T.-B., Zeppenfeld, S., Gong, X., Bahlmann, E., Fomba, K.W., Triesch, N., Stratmann, F., Wurl, O., Engel, A. , Wex, H. & Herrmann, H. (2022). High number concentrations of transparent exopolymer particles in ambient aerosol particles and cloud water - a case study at the tropical Atlantic Ocean. Atmospheric Chemistry and Physics, 22 (8). pp. 5725-5742. DOI 10.5194/acp-22-5725-2022.
Citation: https://doi.org/10.5194/egusphere-2026-1873-AC1
-
AC1: 'Reply on RC1', Theresa Barthelmeß, 27 Jul 2026
-
RC2: 'Comment on egusphere-2026-1873', Anonymous Referee #2, 23 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-1873/egusphere-2026-1873-RC2-supplement.pdf
-
AC2: 'Reply on RC2', Theresa Barthelmeß, 27 Jul 2026
Responses addressing Review 2 for Barthelmeß et al., submitted to ACP
Reviewer The reviewers’ comments are highlighted in italic font.
Response Our response follows the reviewers’ comment and is written in regular font.
Reviewer This manuscript by Barthelmeß et al extends the classic research into marine gel particles with this report on hydrogels (TEP and CSP) from the Southwestern Pacific Ocean, as well as in nascent aerosols generated in a bubbling tank. Using colorimetric staining assays and microscopic, size- resolved quantification they compare data on hydrogels in the spray generated from samples collected in the Southwestern Pacific Ocean. The motivation for this study is to extend our understanding of the relationship between polysaccharidic and proteinaceous gels in the sea surface microlayer to aerosolized particles and atmospheric processes including the formation of cloud condensation nuclei (CCN) and ice nucleating particles (INP).
Response We thank the reviewer for his/her contribution to improving this manuscript.
Reviewer As part of the justification, in LINE 15 why should it matter that “the few studies which exist were conducted in the Northern hemisphere” (and Hemisphere should be capitalized! Why would you expect a major difference? This should be part of the motivation for this study. Just collecting more samples by itself is not a justification.
Response We agree with the reviewer that in its current form the justification is incomplete and will adjust our abstract accordingly. We extended our explanation in the introduction by highlighting that, in contrast to the Northern Hemisphere, the oceans are responsible in balancing the cloud albedo in the Southern Hemisphere especially via the emission of marine biogenic aerosols (L72). As hydrogels are a product of phytoplankton production and bacterial degradation processes, it is important to link their occurrence in sea spray aerosols (SSA) to the specificity of the underlying marine ecosystem. In the Southern Hemisphere oceans, biogeochemical characteristics are quite unique, e.g. the sub-Antarctic waters are recognized as an extended high-nutrient low-chlorophyll (HNLC) region. We will restructure our introduction so that this justification is extended and more prominently presented.
L14-15: Yet, studies on atmospheric measurements detecting marine hydrogels are still limited, and few focus on the Southern Hemisphere oceans, despite their key role in cloud formation.
Reviewer In general, this study is relevant for improving our understanding of ocean-atmosphere interactions in the context of aerosols and although with very limited data, worth publishing.
Response Thank you for highlighting the potential of our study and we will work on improving the manuscript further.
Reviewer However, there is room for improvement. The limited number of samples is unfortunate, but still allows for some conclusions. The authors do oversell the study with the title. CONTROL is a very strong word. More accurately Biogeochemical regimes influence the chemical characteristics of aerosolized hydrogels.
Response We agree with the reviewer that a more appropriate title would be: 'Biogeochemical regimes influence the chemical characteristics of aerosolized hydrogels'. We also agree that we should discuss more prominently the limitation of our study, which arise from a very limited data set.
Reviewer In general they show this (influence of biogeochemistry on aerosolized hydrogels), but not convincingly given the limited data set. The authors are encouraged to address not only the specific issues highlighted, but overall make the manuscript more rigorous in presentation and interpretation of the data.
Response We will follow this recommendation to restructure and edit the manuscript. We will present the data in a more comprehensive way while paying attention to not overinterpreting the limited data set at hand. We agree that for certain underrepresented categories such as the subtropical regime (STW) or SSA data, the statistical power and the extent to which data can be interpreted independently is limited.
This will be achieved by implementing the following steps:
1) Methods: We will restructure the method section by consolidating descriptions of the bubble chamber, adding a dedicated subsection describing the determination of sodium (Na⁺) concentrations in SSA, and providing a clearer rationale for the statistical approaches and their underlying assumptions.
2) Data presentation: We will replace the current Figure 4 with a new figure that presents the temporal evolution of the key biogeochemical parameters (chlorophyll a, nitrate concentration, and salinity) alongside hydrogel concentrations in surface seawater (SSW) and sea spray aerosol (SSA). This chronological presentation will clarify the assignment of biogeochemical regimes, the pooling of SSW data for comparison with SSA samples, and the temporal and spatial variability of the dataset.
3) Data reporting: We will improve the representation of results by implementing a framework, in which it is clearer which data were excluded or are not available (relevant for soot associated SSW samples and missing sodium concentration in SSA). We will adjust our data format (mean ±SD) to better encompass limited sample sizes.
4) Discussion and interpretation: We will include a dedicated section discussing the limitation of this study. We will also strengthen the comparison with previous studies to better contextualize the limited data coverage, e.g., with regard to the STW. We will ensure that our conclusions appropriately reflect the scope and limitations of the available data set.
Reviewer For example, the introduction provides a good summary of the current literature, an overview of sources and chemical composition of gels/ hydrogels in the SML and relevance to sea spray aerosols (SSA). However, it is dense, often redundant, and sometimes ambiguous which detracts from the flow. Several sentences are overly long exceeding 30 words, making it difficult to follow the authors intentions. Frequent parenthetical citations interrupt flow too. Some terms are introduced without immediate definition (e.g., “hydrogels” appear before TEP/CSP are defined). And, there are often a repetition of ideas (e.g., hydrogels as CCN/INP is mentioned multiple times). It would be much more helpful to the reader if the authors would make sure there are clear transitions between themes (particle type and composition, atmospheric relevance, oceanic production, and knowledge gaps), break long sentences into two or more, move citations to the end of sentences to complete a thought, introduce TEP/CSP earlier or define hydrogels more explicitly at first mention, and consolidate repeated statements about the potential role of these gels in atmosphere processes. For example: The term “hydrogels” is used broadly. It is not clear if the authors are referring to polymer networks, colloidal gels, or operationally defined TEP/CSP.
Response We will restructure the introduction by also paying attention to better transitioning between topics and following a clearer outline, shorten overlong sentences, and introduce terms such as hydrogels and TEP/CSP in a more systematic way.
After briefly sketching the overarching aim and context, the relevant Southern Ocean biogeochemistry will be introduced, followed by hydrogel particle types and composition, their emission into sea spray aerosol, their atmospheric relevance, the remaining knowledge gaps, and finally a brief overview of the study design.
We will move citations, where appropriate, to the ends of sentences to improve readability. This will help as well to reduce repetitions. We agree that the use of the term "hydrogels" was insufficiently defined. In the revised manuscript, we will explicitly define the term at its first occurrence and clarify that, throughout this study, it is used operationally to refer to the combined fraction of TEP and CSP.
Reviewer The hypothesis (LINE 495) illustrated in Figure 7 that bacterial processing in the SML favors the accumulation of surface-active, proteinaceous rather than carbohydrate-rich material with TEP rapidly degrading leading to larger carbohydrate rich particles which are aerosolized and smaller protein-rich particles is consistent with finding in Aller et al., 2017 that CSP in the surface waters and the SML was concentrated in larger size particles while TEP was concentrated in smaller particles with both aerosolized.
Response Thank you for pointing this out, we will include this reference to further underline this hypothesis in the discussion. As suggested by the first and third reviewer, we will also improve the visual presentation of this schematic overview figure.
Reviewer In the abstract, LINE 22 the statement about carbohydrate-to-protein ratio talks about “marine samples”. Weren’t all the samples from marine waters?
Response We will adjust this to 'surface seawater samples'. As pointed out by the reviewer, all samples were marine.
Reviewer While the methods section contains all essential methodological details, the narrative is dense, occasionally repetitive, and sometimes unclear in organization. Several sentences are overly long, and some information is not in logical order. The narrative jumps between topics (e.g., chamber design to aerosol sampling to filter storage to blanks to staining), which makes it more difficult to follow. The description of oceanic regimes is followed by aerosol chamber details but is disconnected from the SSA sampling workflow. The plunging-jet system description is split across multiple sentences and paragraphs. The transition from SSA sampling to filter staining is abrupt.
Response We will restructure and consolidate the method section to improve the logical order, reduce redundant descriptions, and clarify the workflow. Related methodological information will be grouped into dedicated subsections. While the revised overall structure is outlined below, we exemplarily present the revised subsection describing SSA generation and sampling.
2.1 General approach (study area, study design, sampling scheme)
2.2 Plunging-jet System
Sea spray aerosols (SSA) were generated using a plunging-jet system (chamber volume: 10 L) as schematically presented by Sellegri, Barthelmeß, et al. (2023) and originally introduced by Fuentes et al. (2010). The chamber walls and lid were constructed of glass and stainless steel, respectively. The chamber was continuously supplied with seawater from the ship's underway system (depth ~6 m) and filled to 3.6 L, corresponding to a water depth of 10 cm. The residence time of seawater in the chamber was approximately 4 min. SSA was produced by splashing seawater through plunging jets onto the chamber water surface at a flow rate of 1.2 L min⁻¹ (Sellegri, Barthelmeß, et al., 2023). SSA were sampled from the chamber headspace after drying with a silica gel diffusion dryer (relative humidity <40%). The aerosol size distribution generated by the plunging-jet system closely resembles the natural SSA size spectrum, including coarse-mode particles that are likely associated with jet droplets (Fuentes et al., 2010; Wang et al., 2017; Sellegri et al., 2023).
Hydrogels in SSA were collected on 25 mm polycarbonate filters (0.2 µm pore size, Nucleopore, Whatman) mounted in 47 mm polycarbonate filter holders using a silicone ring to accommodate the filter size difference. SSA hydrogel collection started around 11:00 AM local time and finished the next day around 9:00 AM (sampling duration between 18.5 and 24 hours, Table A2). Filters sampled the chamber headspace air for in average 22 h at a flow rate of ~2.2 L min⁻¹, corresponding to an average sampled air volume of 2.9 m³. The mean flow rate per filter was calculated by averaging the flow rates measured at the start and end of the collection period. Single replicate samples were collected on eight consecutive days (19–26 March 2020). On the final day of the campaign, the bubble chamber was supplied with Milli-Q water to obtain an operational blank. After collection, filters were stored in petridishes at -20 °C until staining in the home laboratory.
Sodium (Na⁺) concentrations in SSA were determined using a 13-stage DEKATI cascade impactor operated at a flow rate of 7 L min⁻¹. Quartz filters served as impaction substrates. To increase aerosol mass for the later on chemical analysis, filters from adjacent impactor stages were pooled into three size fractions: stages 1-6 (30-500 nm), stages 7-8 (500 nm to 1 µm) and stages 9-13 (1-10 µm). Samples were extracted in Milli-Q water by 30 min of sonication. Extracts were analyzed by ion chromatography for the quantification of the main inorganic ions (Sandrini et al., 2016). An IonPac CS16 3 × 250 mm Dionex separation column with gradient MSA elution and an IonPac AS11 2 × 250 mm Dionex separation column with gradient KOH elution were deployed for cations and anions, respectively.
2.3 TEP and CSP
2.3.1 Hydrogels in SSW
2.3.2 Hydrogels in SSA
2.4 POM and DOM
2.5 Data processing and Statistics
Reviewer LINE 140. It is not clear why the SSA collection onto filters was conducted over a ~6-hour period with no replication. Regarding the dual staining. Why were the number of filters limited? Not clear why an apology is necessary. If Engels method is followed why the apology? Rather than use the same filter for dual staining was cutting the one filter in half and using one half for the TEP and the other for CSP considered? Did it matter whether you stained first with the CBB then Alcian Blue or vice versa? Would it make any difference? If they have been divided, the number of samples could have been increased from 1 or 2. The authors defer all discussion of deviation of methods from what one would expect to the supplemental section.
Response We would like to clarify that the SSA filters were collected over a time period of 18.5 to 24 hours (as previously described in L128, L141 and Table A2). However, we are not quite certain which apology the reviewer is referring to.
i) We acknowledge that the explanation why we had only a single replicate fell short and will include a brief statement in the method section. The set-up of the plunging-jet chamber and installed filer holders and lines, as well as the days of our cruise, were limited. As the goal of the campaign was to address the interdisciplinary links between the marine biosphere and atmosphere, only one filter holder was available for the collection of aerosolized hydrogels, the other lines served e.g., for sampling DNA. Initially, our cruise was supposed to last three weeks, with sufficient time scheduled for each biogeochemical regime. Instead, the cruise was abruptly shortened to 11 days and we had to return to Wellington harbor due to the Covid-19 pandemic outbreak.
ii) The method does not allow to cut the filter in half prior to the staining procedure as we operate with a vacuum filtration system on which we need to mount the circular filters and, subsequently, dyes. We applied the Coomassie Blue, then the Alcian Blue solution to the filters. The order is relevant as Coomassie Blue partly stains Alcian Blue. Our results of the dual-staining approach were validated by comparing it to single-stained filters. Comparative tests included samples from model substances and a Baltic Sea station (three depths). The model substances and the environmental samples exhibited consistent values in abundance and area regardless of the staining approach (Scheidemann et al., in prep.). We acknowledge and will highlight that our dual-staining approach and the subsequent enumeration and composition estimate of hydrogels in SSA represents a compromise and therewith may entail an inherent bias, which we cannot fully resolve.
iii) We further acknowledge that we should summarize the main conclusion from our supplementary material in the method and a dedicated discussion section, however, would prefer to keep the detailed description how the hydrogel composition in aerosols was estimated in the supplementary material.
Reviewer One of my main problems with this paper concerns the first method that Engel proposed to contrast carbohydrate and proteinaceous particles, but subsequent studies took a more quantitative analytical approach with xanthum gum and bovine albumin serum calibration. LINES 124 and 386-390. Some comparisons to literature are not fully parallel (e.g., comparing XG equivalents to colorimetric -area-based measurements). The use of the colorimetric technique vs XG and BSA equivalents are not explained and no direct comparison is made before adopting this technique. Methodological differences when comparing to other studies should be clarified in the main text and not buried in a supplemental section. While they state that their color technique relies on a classical approach (Engle, 2009) it only provides “approximate the contribution of CSP and TEP”. Nevertheless, their results are quantitative. Any discussion about comparison of a microscopic method with the widely accepted quantitative colorimetric technique (XG and BSA equivalents) needs to be made in the text.
Response We would like to clarify a point that may not have been sufficiently explained in the manuscript. The approximation mentioned in the text (L125) refers to the chemical composition of the hydrogels inferred from the dual-staining approach (i.e., distinguishing protein-rich from carbohydrate-rich particles), rather than to the quantification of particle abundance and area (as stated in L126). We will revise the method section to clarify this in a more comprehensive way. The quantification of particle abundance, size distribution and area was performed following the well-established microscopic image analysis approach of Engel (2009). Specifically, ImageJ was used to separate the individual RGB-color channels and identify all Alcian Blue and Coomassie Blue stained particles, irrespective of whether they were stained by the classical (Engel, 2009) or dual-staining approach. Thus, total particle abundance, size distribution, and area in SSA hydrogels are derived from the RGB-color analysis and are independent of the compositional approximation based on the dual-staining approach. Quantification of SSA and SSW abundance, size distribution, and area of hydrogels are thus based on the same analytical approach.
Regarding the comparison with previous studies, we agree that studies using microscopic image analysis are not directly equivalent to those employing colorimetric quantification expressed as xanthan gum (XG) or bovine serum albumin (BSA) equivalents (L390-393). Because published observations of aerosolized TEP and CSP are very scarce, we still considered it useful to include this comparison despite the methodological differences. We will revise the discussion to explicitly acknowledge methodological differences in the main text. We would also like to point to the studies of Passow and Alldredge (1995) and Cisternas-Novoa et al., (2014) in which it was shown that the microscopically determined area and the colorimetric determined concentration expressed in BSA eq. are linearly correlated including culture and environmental samples. We will include these references. By also incorporating the comments of the first reviewer, we suggest to slightly shorten the paragraph, while explicitly pointing out the methodological differences.
L386-394: Aller et al. (2017) reported TEP and CSP as 1–100% of artificially generated SSA and ambient aerosol particle mass using a colorimetric approach, and although methods differ, their SSA and ambient air concentrations (2 and 14 μg XG and BSA equivalents m-3, respectively) are broadly consistent with our hydrogel area in SSA when translated into equivalent concentrations. In seawater, colorimetrically determined concentrations correspond to the microscopically determined area, which were shown to linearly correlate (Dreshchinskii and Engel, 2017; Cisternas-Novoa et al., 2014; Passow and Alldredge, 1995).
Finally, we chose the microscopic approach because it provides particle abundance, size distributions, and area which were central objectives of this study. In contrast, the colorimetric method yields bulk concentrations but does not provide information on the size distribution or area of individual particles.
We hope, we could address the concern of the reviewer appropriately.
Reviewer Also, regarding methods, the chamber description is mixed with sampling volumes and residence times, then the description reverts back to chamber operation. The section would benefit from clearer structuring, more concise phrasing, and improved flow between procedural steps.
Response As stated above, we agree with the reviewer and will restructure the method section to improve readability and clarity.
Reviewer The results give a good introduction to the 4 oceanic regions and their biological characteristics; however, this section often jumps between regimes, soot events, size bins, and Na+ normalization without clear transitions which makes it difficult to follow the progression from raw observations to the patterns and then to the interpretation. For example, soot contamination is introduced abruptly in the middle of the results, interrupting the flow of hydrogel abundance patterns. The anomalous soot events and heavy rain events are identified and contextualize helping to justify the exclusion of some samples.
Response This point was also raised by the first reviewer. We agree and will restructure the results accordingly. We will collect and present all arguments for identifying the black particles as potential soot in the method section and remove it from the result section. We further suggest to exclude the affected size bins of the two SAW afternoon samples from the entire analysis and representation in the figures and therewith further clarify the data representation and interpretation.
Reviewer That being said, listing daily values (e.g., “on the 23rd of March...”) LINE 232 stops the flow of the narrative. Given that the date is important for the pattern of results perhaps the information could be reorganized by date followed by the results.
Response We agree and will implement this suggestion.
L232: The fraction of CSP number in aerosols was largest in the STW (53%) and smallest in the SAW (23rd of March: 15%).
We will implement this format as well in L235 and L259.
Reviewer The authors need to clarify the implications of limited SSA replication. The quantitative detail provided obscures the fact that the results are VERY LIMITED. There is no explicit statistical framing, only the statement that “no statistical tests were performed” for SSA (LINEv229) due to limited sample size”. This is misleading. Presumably, the lack of replication was not intentional but, it still leaves the possibility that the degree of variability is much greater or much less.
Response We understand the reviewers concern and will add an explicit statement that the lack of replicates precludes an estimate of the variability.
L229: Hydrogel abundance and area of artificially generated SSA varied greatly along the cruise track. However, the limited number of filters available for each regime precludes a reliable estimate of the variability, and therefore the SSA results should be interpreted as descriptive rather than statistically representative. Due to the limited sample size, no statistical tests were performed.
Reviewer So while there are definitive values for abundance, area, CSP/TEP ratios, and enrichment factors, stating that the SSA hydrogel concentrations “ranged from 0.9 ± 0.7 × 105 m−3... to 0.2 × 105 m−3...” gives the impression that the number of samples (n) is greater than 2. In this case, just giving the 2 values is better given the VERY LIMITED data.
Response We agree, we will provide both values instead of the mean.
Reviewer Interpretation creeps into the results section in many instances. For example, the authors should just present the fact that the CSP fraction decreased with particle size and not include interpretation that proteinaceous particles were the predominate type.
Response We kindly disagree, however, will carefully check the results again to exclude any potential interpretation. We belief the reviewer referrers to L237, where we wrote: "Proteinaceous particles were thus predominantly present in the smaller size bins." This is a direct consequence of numerous CSP particles (15-35%) and small CSP area (4-15%) in relation to overall detected SSA hydrogels. This is shown by the presented data as illustrated in Figure 5 and reported in greater detail in the subsequent sentences. To acknowledge the reviewers concern, we will delete this sentence.
Reviewer The contrast between SSW and SSA size-bin patterns is well articulated. LINES 237-243) Results make no sense as written “CSP contributed especially to the smallest size bin reaching its minimum in the largest size bin”. Totally needs clarification.
Response We will rephrase and clarify this sentence.
L241: CSP contributed most to the smallest hydrogels (size bin 0.5–1 µm; 45 ± 12%). Its contribution decreased progressively with increasing hydrogel size, reaching a minimum in the largest hydrogels (size bin 5–10 µm; 9 ± 11%). Largest hydrogels were predominantly composed of TEP.
Reviewer Some key findings are buried and should be highlighted more explicitly. The highest concentration of hydrogels and highest relative enrichment of hydrogels in the Subtropical Front is a major finding and deserves a more prominent discussion, perhaps in the conclusion.
Response We thank the reviewer for this suggestion and will highlight this key finding.
Reviewer The figures and table are well aligned with the narrative. The use of pie diagrams to illustrate proportional abundance in Figure 5 is effective. Figure 5 and Table 1 directly support the text and clearly illustrating the size-resolved patterns. However, size-resolved abundance appears in the text, in Figure 5, and in Table 1. sometimes with overlapping phrasing. Consolidating these into a single, coherent subsection would improve readability.
Response We will follow this suggestion and consolidate the results on size-resolved abundance within one subsection.
Reviewer This section would benefit from reorganization into subsections: (1) SSW patterns, (2) SSA patterns, (3) size-resolved patterns, (4) enrichment to help improve transitions and maintain the flow of the results. There is a certain amount of redundancy and repetition. For example, hydrogels as CCN/INP is mentioned in two separate places. Bubble bursting mechanisms are described in detail, but similar information appears earlier in the literature review.
Response We will reorganize the structure of the results and take the reviewers suggested succession of subsections. We will also work on streamlining our discussion accordingly. We agree, this will help to reduce redundancy.
Reviewer The authors are encouraged to consolidate discussion of the roles of hydrogels in atmospheric processes into a single, well-structured paragraph and to keep bubble-bursting details in the description of the study design.
Response We agree, the discussion could be shortened and clearer in its structure. While reducing the length of the paragraphs in the discussion (e.g., L408 to L443), which focus on bubble-bursting related SSW literature, we would still like to keep some of the references as they illustrate how SSW properties, bubble bursting mechanisms, and SSA composition are interlinked (e.g., L445 to L465).
Reviewer Sections 3.3 Biogeochemical characteristics mixes descriptive statistics with previously published findings...
Response The studies we referenced are from the same campaign (e.g., L274). They are included to inform the reader about the previous work, which already describe the encountered biogeochemical regimes. Following the reviewers concern, we will move this information to the method section. We will further remove the references related to the use of degradation indices (L294) and outline them in the first subsection of the discussion in relation to biogeochemical regimes.
Reviewer and 3.4 Biogeochemical factors influencing hydrogel occurrence in seawater and aerosols contains detailed results and comparison with previous studies, but the narrative jumps between particle size classes, phytoplankton groups, CSP and TEP in the water and in SSA, to ecological interpretation, and then to mechanistic explanations. This loses the reader who can no longer follow the main points.
Response We think the reviewer is referring to section 4.1 of the discussion. In section 3.4 (L299 to L315), we only briefly describe the results of the correlation matrix presented in Figure 6a and Figure 6b. In section 4.1, we will work on clarifying the discussion including better transitions between topics. We will work on restructuring the discussion's subsection to better separate between i) contextualizing our results with the light of previous literature, ii) ecological interpretation, and iii) potential mechanistic explanations.
Reviewer The connection between phytoplankton community composition and hydrogel types are interesting but the claim for example that Synechococcus driving CSP enrichment are speculative given that they present no data or referenced discussion of blooms of these cyanobacteria, or mucus or exudate production which would provide hydrogel material.
Response We have enumerated Synechococcus spp. abundances (section 3.3 and Table A3). Synechococcus spp. exhibited highest abundances in the MIX in concert with an elevated fraction of SSW CSP area. However, both results were insignificant. We also included a reference which shows CSP production by Synechococcus spp. cultures (L330, L341). We agree that this is only a tentative explanation in the context of describing hydrogel dynamics in the MIX regime (L342) and will remove this statement.
Reviewer The fact that the result section extends beyond results and incorporates discussion leads to some repetition in the discussion section. In keeping with the journal’s structure, the results should be kept in the results section and move all mechanistic interpretations to the discussion.
Response We will screen our result section to exclude any interpretation and clearly separate the literature comparison from mechanistic interpretations.
Reviewer There are repeated references to Chl a patterns (i.e. LINES 302-305, 326, 333-335, 368, 435) repeated statements about nutrient regimes (STF vs MIX vs SAW vs STW), and multiple mentions of nanophytoplankton abundance patterns. Consolidate repeated descriptions of regime characteristics. Just present each biogeochemical variable once, then refer back to it when needed.
Response We will work on reducing redundancy in the result and discussion sections. Results and discussion fragments dealing with Chl a pattern and regime dependencies will be combined to two condensed subsections, which will be dedicated to i) describing general biogeochemical regimes and ii) associated hydrogels dynamics.
Reviewer There are occasional grammatical errors (“22rd”, “24st”) in the Fig. 4 legend
Response Thank you for noticing, we will correct this.
Reviewer Some phrases are vague (LINE 273) regimes were characterized by specific changes...
Response We will remove this sentence and, as described above, move the associated references to the method section. We reference our previous papers in L273 to clarify i) that these data have been partly introduced elsewhere and ii) that instead of including three samples per day (Barthelmeß et al., 2025) only two samples were included to align with the sampling frequency of CSP and TEP in the SSW.
Reviewer ... and the use of the future tense (LINE 272). Occasional awkward phrasing LINE 476 (“aggregation time of TEP are extended”).
Response We will remove the future tense in L272 and correct the phrasing in L476.
L272 ...the focus of the next paragraph is set on describing the abundance of bacteria- and phytoplankton...
L476 When exposed to solar radiation, TEP aggregation is inhibited, TEP size decreases, or may cause TEP aggregates to disperse completely.
Reviewer Some anthropomorphic phrasing LINE 431 (“bubbling scavenges, up concentrates and aggregates”). LINE 481 (“bacteria benefit from concentrated substrate”).
Response We will replace the critical phrases. We will further screen the text for other anthromorphisms.
L431 Conclusively, rising bubbles scavenge, concentrate, and aggregate precursors more efficiently into aerosols in regimes in which freshly produced, polymeric carbohydrates occur.
L481 Conversely, the bacterial community can considerably enhance aggregation kinetics, possibly because concentrated substrate promotes bacterial activity.
Reviewer Some claims are speculative without explicit qualifiers and should be changed. (e.g., soot interfering with bubble bursting). LINES 440. The authors have no evidence for soot particles on the aerosol filters which they presumably confirmed with microscopic examination, they speculate that soot may have interfered with bubble scavenging and bursting.
Response We agree with the reviewer. A short statement summarizing the influence of soot particles on surface properties will be moved to the supplementary material and we will retain from speculating about their influence on SSA production. This would comply as well with the suggestions of the first reviewer.
Reviewer The occasional over-interpretation of patterns (e.g., attributing enrichment differences solely to polymer size) I attribute to the paucity of data. Without data to support the interpretation, implying causation should be avoided.
Response We agree with the reviewer that our discussion should avoid implying causal relationship where it is not directly supported by our data. While the SSA dataset is very limited, the SSW dataset comprises 20 duplicate measurements of TEP and CSP from locations along the cruise track and was compared with a diverse set of independent biogeochemical variables. The correlation analysis (Figure 6a) shows that the occurrence of large aggregates is associated with elevated biological activity (indicated by parameters such as TON, TOC, the abundance of nanophytoplankton, Chl a > 2µm and >20µm) and is therefore consistent with literature-based hypotheses linking aggregate formation to biological activity. The proposed role of polymer length in the formation of stable versus unstable and large versus small aggregates is based on previous studies rather than direct measurements in this work. Our data do not demonstrate a causal relationship. We will revise the discussion accordingly, clearly distinguishing observations from literature-based interpretations and restricting these mechanistic considerations to the SSW dataset.
We thank the reviewer again and hope we could address his/her concerns appropriately.
Literature
Aller, J. Y., Radway, J. A. C., Kilthau, W. P., Bothe, D. W., Wilson, T. W., Vaillancourt, R. D., Quinn, P. K., Coffman, D. J., Murray, B. J., & Knopf, D. A. (2017). Size-resolved characterization of the polysaccharidic and proteinaceous components of sea spray aerosol. Atmospheric Environment, 154(February), 331–347. https://doi.org/10.1016/j.atmosenv.2017.01.053
Barthelmeß, T., Cristi, A., Deppeler S., Safi, K., Sellegri, K., Law, C. S. and Engel, A. (2025). Pronounced Diel Cycling of Dissolved Carbohydrates and Amino Acids in the Surface Ocean and across Diverse Regimes. Environmental Science and Technology, 1(59), 419–429. https://pubs.acs.org/doi/10.1021/acs.est.4c00491.
Cisternas-Novoa, C., Lee, C., Engel., A. (2014). A semi-quantitative spectrophotometric, dye-binding assay for determination of Coomassie Blue stainable particles. Limnology and Oceanography: Methods, 12, 604-616. https://doi.org/10.4319/lom.2014.12.604.
Dreshchinskii, A. und Engel, A. (2017). Seasonal variations of the sea surface microlayer at the Boknis Eck Times Series Station (Baltic Sea). Journal of Plankton Research, 39 (6). pp. 943-961. https://doi.org/10.1093/plankt/fbx055.
Engel, A. (2009). Determination of Marine Gel Particles, Practical guidelines for the analysis of seawater (O. Wurl, B. Raton, & [u.a.] (eds.)). CRC Press.
Fuentes, E., Coe, H., Green, D., de Leeuw, G., and McFiggans, G. (2010). Laboratory-generated primary marine aerosol via bubble-bursting and atomization, Atmos. Meas. Tech., 3, 141–162,https://doi.org/10.5194/amt-3-141-2010
Passow, U., Alldredge, A. (1995). A dye‐binding assay for the spectrophotometric measurement of transparent exopolymer particles (TEP). Limnology and Oceanography, 40(7), 1326–1335. https://doi.org/10.4319/lo.1995.40.7.1326.
Sandrini, S., Van Pinxteren, D., Giulianelli, L., Herrmann, H., Poulain, L., Facchini, C.M., Gilardoni, S., Rinaldi, M., Paglione, M., Turpin, B.J., Pollini, F., Bucci, S., Zanca, N., Decesari, S. (2016). Size-resolved aerosol composition at an urban and a rural site in the Po Valley in summertime: Implications for secondary aerosol formation. Biogeosciences, 16(17), 10879-10897. https://doi.org/10.5194/acp-16-10879-201
Sellegri, K., Harvey, M., Peltola, M., Saint-Macary, A., Barthelmeß, T., Rocco, M., Moore, K. A., Cristi, A., Peyrin, F., Barr, N., Labonnote, L., Marriner, A., McGregor, J., Safi, K., Deppeler, S., Archer, S., Dunne, E., Harnwell, J., Delanoe, J., … Law, C. S. (2023). Sea2Cloud: from biogenic emission fluxes to cloud properties in the South West Pacific. Bulletin of the American Meteorological Society, 1017–1043. https://doi.org/10.1175/bams-d-21-0063.1
Sellegri, K., Barthelmeß, T., Trueblood, J., Cristi, A., Freney, E., Rose, C., Barr, N., Harvey, M., Safi, K., Deppeler, S., Thomspson, K., Dillon, W., Engel., A., Law, C. (2023). Quantified effect of seawater biogeochemistry on the temperature dependence of sea spray aerosol fluxes. Atmospheric Chemistry and Physics, 23(20), 12949–12964. https://doi.org/10.5194/acp-23-12949-2023.
Wang, X., Deane, G. B., Moore, K. A., Ryder, O. S., Stokes, M. D., Beall, C. M., Collins, D. B., Santander, M. V., Burrows, S. M., Sultana, C. M., & Prather, K. A. (2017). The role of jet and film drops in controlling the mixing state of submicron sea spray aerosol particles. Proceedings of the National Academy of Sciences of the United States of America, 114(27), 6978–6983. https://doi.org/10.1073/pnas.1702420114
Citation: https://doi.org/10.5194/egusphere-2026-1873-AC2
-
AC2: 'Reply on RC2', Theresa Barthelmeß, 27 Jul 2026
-
RC3: 'Comment on egusphere-2026-1873', Anonymous Referee #3, 01 May 2026
Barthelmeß et al. presents a novel, comprehensive investigation into the emission of hydrogels in sea spray aerosols (SSA) across four distinct biogeochemical regimes in the Southwestern Pacific Ocean. The study used a plunging-jet bubble chamber to simulate natural SSA generation, and quantified hydrogels in both surface seawater (SSW) and the resulting SSA. Overall, the manuscript is well presented and the finding is important for better understanding of hydrogel contents in aerosols and their impacts on cloud condensation and ice nuclei formation. However, I think the manuscript needs further revision before considering for publication.
- I am not sure if that is my pdf problem, I saw that most of the figures in the manuscript are not clear. Make sure all the figures are clear enough to read.
- Abstract, Line 16: Typo for “in in”.
- Line 83: What are the physicochemical and biological properties of the seawater taken form the four marine regions? Do the properties change when introducing to the bubble chamber? The information should be introduced in the text.
- Line 92: It is better to briefly include the information of aerosol size distribution in the text or in the SI.
- Line 141: Why the sampling time varies? Will this have effects on the chemical analysis? As can be seen in the Table A2, the Na in SSA is n/a when the filtering time is less than 20h.
- L160: What is the particle size of aerosol used for the enrichment factors calculation?
- Line 186: There are only two samples for the subtropical waters (STW), with such a few sample, how will it affects the statistics or reliability of the results with these samples?
- Line 439: The authors suggest soot may have interfered with bubble scavenging in the SAW but was not visible on aerosol filters. Could this have contributed to the observed variability in SSA concentrations?
- Figure 7: The figure is quite confusing and I have problem of understanding the logical flow of the key message without going through the figure caption and main text. The coloring is hard to follow, for examples, what does yellow circle and in the seawater represent? While for the size of pie charts, does it mean anything here? It needs a thorough revision.
- There is a missing Table A3 in the manuscript. Please double check.
Citation: https://doi.org/10.5194/egusphere-2026-1873-RC3 -
AC3: 'Reply on RC3', Theresa Barthelmeß, 27 Jul 2026
Responses addressing Review 3 for Barthelmeß et al., submitted to ACP
Reviewer The reviewers’ comments are highlighted in italic font.
Response Our response follows the reviewers’ comment and is written in regular font.
Reviewer Barthelmeß et al. presents a novel, comprehensive investigation into the emission of hydrogels in sea spray aerosols (SSA) across four distinct biogeochemical regimes in the Southwestern Pacific Ocean. The study used a plunging-jet bubble chamber to simulate natural SSA generation, and quantified hydrogels in both surface seawater (SSW) and the resulting SSA. Overall, the manuscript is well presented and the finding is important for better understanding of hydrogel contents in aerosols and their impacts on cloud condensation and ice nuclei formation. However, I think the manuscript needs further revision before considering for publication.
Response We thank the reviewer for this supportive feedback and will work on improving the manuscript further.
Reviewer I am not sure if that is my pdf problem, I saw that most of the figures in the manuscript are not clear. Make sure all the figures are clear enough to read.
Response Thank you for letting us know, we will check the quality of the figures before uploading the revised pdf version.
Reviewer Abstract, Line 16: Typo for “in in”.
Response We corrected the error.
Reviewer Line 83: What are the physicochemical and biological properties of the seawater taken form the four marine regions? Do the properties change when introducing to the bubble chamber? The information should be introduced in the text.
Response 1) Biogeochemical parameters are introduced in section 3.3. We will further integrate salinity and nitrate concentrations, which were used to assign the SSW and SSA samples to biogeochemically contrasting regimes. We will include a comprehensive figure, in which we will present the most important biogeochemical properties of the seawater (incl. salinity, nitrate, Chl a). This will also comply with the recommendation of the first reviewer.
Surface tension measurements in dependence of regimes and temperature as well as their combined effect on SSA fluxes are presented elsewhere (Sellegri et al., 2023). We will include a brief section in which we will discuss our results in the light of the results presented by Sellegri et al. (2023).
2) We will further address briefly whether we expect changes in the SSW properties due to the usage of the bubble chamber. We will add information to the manuscript on how the bubbling chamber might affect biogeochemistry and in particular TEP and CSP aggregation dynamics. This will again comply with the recommendation of the first reviewer. Two concerns may be raised:
i) Does the bubble chamber influence the metabolism of the organisms by e.g., introducing iron into a micronutrient limited system, e.g., the sub-Antarctic water (SAW)?
Iron availability is enhanced by complexation with carbohydrates serving as organic ligands (Hassler et al., 2011). However, as the residence time of the water in the tank was very short (~4 min), phytoplankton and bacterial cell metabolic rates (including iron uptake rates, Hudson and Morel, 1990) limit the ability of cells to influence water biogeochemistry within this short time window. Subsequently to iron uptake, anabolic rates in phytoplankton cells also relate to the availability of sun light. Our SSA chamber was set up in the lab and not exposed to ambient solar light. In summary, we do not expect that the set-up will interfere with overall biogeochemistry and the metabolism of cells.
ii) Does bubbling affect aggregation dynamics of TEP and CSP?
Bubbling does affect the aggregation dynamics of TEP and likely also CSP (Robinson et al., 2019). In the plunging-jet chamber, the persistent bubble flux promotes the scavenging and upward transport of organic material, leading to its transient accumulation in the sea surface microlayer (SML). Consequently, the chamber environment inherently supports aggregation dynamics of TEP and CSP. Although parts of the particulate organic material remain temporarily enriched in the SML before being mixed into the bulk water, another fraction is transferred into SSA. We therefore consider the plunging-jet system to reproduce the key physical processes that influence TEP and CSP aggregation dynamics as well as SSA fluxes under natural conditions at the air-sea interface, making the chamber a suitable experimental analogue for studying these processes.
Reviewer Line 92: It is better to briefly include the information of aerosol size distribution in the text or in the SI.
Response We will add this information to the method section.
Sodium (Na⁺) concentrations in SSA were determined using a 13-stage DEKATI cascade impactor operated at a flow rate of 7 L min⁻¹. Quartz filters served as impaction substrates. To increase aerosol mass for the later on chemical analysis, filters from adjacent impactor stages were pooled into three size fractions: stages 1-6 (30-500 nm), stages 7-8 (500 nm to 1 µm) and stages 9-13 (1-10 µm).
Reviewer Line 141: Why the sampling time varies? Will this have effects on the chemical analysis? As can be seen in the Table A2, the Na in SSA is n/a when the filtering time is less than 20h.
Response In general, we aimed at changing the filters approximately around 9:00. In between 09:00 and 12:00 the chamber was used for another experiment (Sellegri et al., 2023). Samples to assess the ion composition in SSA were also generated during the last days of the campaign. Sodium (Na+) concentration in SSA is simply not available because of an instrumental problem in the home laboratory, it is not because the temporal reduction in SSA filtration affected the chemical analysis. We will implement this information in the method section.
The sampling time of SSA collection was reduced during the last days of the campaign in response to the Covid-19 pandemic outbreak. The New Zealand authorities instructed the vessel to return to the New Zealand EEZ and, then, harbor. Our cruise was therefore limited to 11 days instead of the planned three weeks. These changes in schedule came along with considerable logistical constraints in completing the main aims of the campaign, including the above-mentioned experiments. SSA Sampling time was therefore adjusted.
Reviewer L160: What is the particle size of aerosol used for the enrichment factors calculation?
Response Our previous calculation was based on the total concentration of Na+ collected irrespective of SSA sizes. Our previous enrichment factors referred to the size range of the microscopical analysis of the hydrogels in SSA and SSW only. We are now aware that this does not comply with the established practice in the atmospheric chemistry community. We will recalculate the enrichment factors taking only the stages of the impactor into account (Na+ measurements), which are relevant for the hydrogel particle size distribution (stages 7-8: 500 nm to 1 µm and stages 9-13: 1-10 µm). This also follows the suggestion of the first reviewer.
Reviewer Line 186: There are only two samples for the subtropical waters (STW), with such a few sample, how will it affects the statistics or reliability of the results with these samples?
Response We acknowledge that the subtropical waters (STW) are represented by only two SSW samples, which limits the statistical power and the extent to which this regime can be interpreted independently. To investigate the relationships between TEP, CSP, and biogeochemical conditions, such as the analyses presented in the heatmap (Figure 6), all SSW samples were pooled and thus our conclusions are not driven by the STW subset alone. We have selected rank-based non-parametric test because they better address small and unequal sample sizes. Moreover, the variability between the two STW samples is relatively low in comparison to the other regimes for most of the measured parameters. Relatively higher variability was only observed for DCCHO concentrations and picophytoplankton abundance. The larger difference in DCCHO concentrations likely reflects diel phytoplankton metabolism and associated degradation processes (Barthelmeß et al., 2025). However, we recognize that these measures and observations do not overcome limited sampling within the STW. We will explicitly acknowledge the scarcity of STW data in the revised manuscript, address the limited statistical power and independent interpretation (section 2.3), and better embed our STW results in the context of previous studies.
Reviewer Line 439: The authors suggest soot may have interfered with bubble scavenging in the SAW but was not visible on aerosol filters. Could this have contributed to the observed variability in SSA concentrations?
Response As discussed in L441–443, SSA hydrogel concentrations were more variable in the SAW than in the mixed regime (MIX), despite comparable TEP and CSP concentrations (area). We thus argued along the same train of thought as the reviewer. However, soot particles were identified solely based on optical indices and appeared as isolated particles rather than being embedded within stained hydrogels (as observed by the first reviewer). We therefore have no direct evidence that soot accumulated at the air–water interface or interfered with TEP/CSP aggregation dynamics. Nevertheless, given the limited literature at hand, we cannot rule out a potential influence of soot on bubble-mediated processes. The observed variability might be thus best explained by the limited sample size (as suggested by the second reviewer).
To better account for the listed uncertainties and consolidate all arguments, we will revise the manuscript: We will exclude the contaminated SSW samples from the analysis, elaborate on our approach in the methods, and move the associated further discussion of possible soot effects on bubble bursting to the supplementary material.
Reviewer Figure 7: The figure is quite confusing and I have problem of understanding the logical flow of the key message without going through the figure caption and main text. The coloring is hard to follow, for examples, what does yellow circle and in the seawater represent? While for the size of pie charts, does it mean anything here? It needs a thorough revision.
Response We thank the reviewer for this feedback and will work on simplifying the figure, including a more intuitive structure and color scheme. Instead of using the figure as a main element, we will adjust it to a 'table of content' figure which can be displayed along with the abstract. To better serve that purpose, we will also include boxplots next to the figure to visualize our averaged data on SSW and SAA hydrogel area.
Reviewer There is a missing Table A3 in the manuscript. Please double check.
Response Thank you for pointing this out, instead there were two Tables named 'A4'. We will correct this.
Literature
Barthelmeß, T., Cristi, A., Deppeler S., Safi, K., Sellegri, K., Law, C. S. and Engel, A. (2025). Pronounced Diel Cycling of Dissolved Carbohydrates and Amino Acids in the Surface Ocean and across Diverse Regimes. Environmental Science and Technology, 1(59), 419–429. https://pubs.acs.org/doi/10.1021/acs.est.4c00491.
Sandrini, S., Van Pinxteren, D., Giulianelli, L., Herrmann, H., Poulain, L., Facchini, C.M., Gilardoni, S., Rinaldi, M., Paglione, M., Turpin, B.J., Pollini, F., Bucci, S., Zanca, N., Decesari, S. (2016). Size-resolved aerosol composition at an urban and a rural site in the Po Valley in summertime: Implications for secondary aerosol formation. Biogeosciences, 16(17), 10879-10897. https://doi.org/10.5194/acp-16-10879-2016
Sellegri, K., Barthelmeß, T., Trueblood, J., Cristi, A., Freney, E., Rose, C., Barr, N., Harvey, M., Safi, K., Deppeler, S., Thomspson, K., Dillon, W., Engel., A., Law, C. (2023). Quantified effect of seawater biogeochemistry on the temperature dependence of sea spray aerosol fluxes. Atmospheric Chemistry and Physics, 23(20), 12949–12964. https://doi.org/10.5194/acp-23-12949-2023.
Robinson, T. B., Wurl, O., Bahlmann, E., Jürgens, K., & Stolle, C. (2019). Rising bubbles enhance the gelatinous nature of the air–sea interface. Limnology and Oceanography, 64(6), 2358–2372. https://doi.org/10.1002/lno.11188
Citation: https://doi.org/10.5194/egusphere-2026-1873-AC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 311 | 123 | 39 | 473 | 43 | 27 | 23 |
- HTML: 311
- PDF: 123
- XML: 39
- Total: 473
- Supplement: 43
- BibTeX: 27
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review for Barthelmeβ et al., submitted to ACP
The study “Biogeochemical regimes control marine aerosol emission of hydrogels in the Southwestern Pacific Ocean” by Barthelmeβ et al., submitted to ACP, presents measurements of hydrogels (TEP and CSP) using colorimetric staining assays and microscopic, size-resolved quantification in seawater from the Southwestern Pacific Ocean, as well as in nascent aerosols generated in a bubbling tank. The authors relate variations in gel properties to different biogeochemical regimes associated with distinct water masses and report enrichment factors during the transfer from seawater to aerosol phase. The study is motivated by the potential relevance of hydrogels for cloud formation and cloud properties.
Overall, I consider this study to be highly relevant for improving our understanding of aerosol–cloud interactions, particularly in the context of marine aerosols and their role in the climate system, with possible implications for future model parameterizations. I have no doubt that this manuscript has the potential to be suitable for publication after addressing the critics below.
However, in its current form, the manuscript suffers from substantial weaknesses in the presentation of results and discussion. These include considerable overselling of the findings, potential (mathematical or logical) errors in analysis and calculations, incomplete or misleading citation of certain information from the literature, a lack of clear structure, and, at times, a lack of rigor. These issues make it difficult to fully assess and appreciate the scientific value of this study. I strongly encourage the authors to invest additional effort into revising the manuscript to improve clarity, correctness, and overall quality of interpretation.
In the following, I outline my main concerns:
1.) Soot topic:
a) Clarification and Focus of the “Soot” Discussion: Your discussion of soot particles is currently not fully convincing and, in its present form, feels somewhat distracting because it appears repeatedly throughout the manuscript. As a result, it becomes difficult to identify your main findings. In my opinion, it would help to clarify your central messages. For example: Is your key result that SAW water masses are generally contaminated with soot, but that the corresponding aerosols are not significantly affected? If yes, consider structuring the paper clearly around this finding. If not, I would recommend substantially reducing the emphasis on the soot discussion.
b) Scientific Concerns About the Interpretation: I also have some concerns regarding the plausibility of your interpretation: You describe soot as hydrophobic (Line 439), which would suggest it remains in the surface microlayer. How likely is it, then, to find significant amounts at depths of 6 m? If soot sinks as part of aggregates (Line 440), one would expect the particles at that depth to be embedded in hydrogels rather than appearing as isolated small particles?
c) Identification of Black Particles as Soot: I do not find the current presentation of the black particles as soot (Section 2.3) entirely convincing. The supporting arguments are currently scattered across the manuscript (e.g., Line 438 mentioning potential origin from Australian bushfires). I suggest: Collecting and presenting all arguments for identifying these particles as soot in a more focused way, ideally within the Methods section. Using more cautious wording, such as: “Based on the available indications, we suggest that these small particles are likely soot.”
d) Handling of Contaminated Samples: It is important to clearly state how these filters were treated in your analysis. For example: Were the affected samples consequently excluded from further analysis because they interfere with hydrogel quantification? Or were they retained and marked (e.g., with asterisks) in the results?
e) Data Presentation and Figures: The current visual presentation is somewhat confusing. For example, in Figure 3: The plots are labelled as CSP and TEP, yet include black particle samples that you explicitly state are neither CSP nor TEP. I recommend either: Adjusting the figure title and labels to reflect what is actually shown, or modifying the visualization to avoid mixing different particle types in a misleading way. Additionally, have you considered improving the microscopic images of the contaminated samples by applying color filters (removing black) with the image software to reduce the visual impact of these black particles?
f) My personal recommendation: Overall, my suggestion would be: Briefly describe the observation and treatment of (potentially) soot-contaminated samples in a concise paragraph within the Methods section. If you wish to explore this topic further, consider moving the extended discussion to the supplementary material.
2. ) I found some of the phrasing somewhat strong and occasionally selective in how the literature is presented, which in places gives the impression of overstating the certainty or significance of the findings. For example, in Lines 13–14 you state that “these hydrogels possess excellent cloud condensation and ice nucleation properties,” whereas later (Line 37) this is phrased more cautiously as “hydrogels have been suggested to serve as cloud condensation…” This represents a substantial difference in level of certainty, and I would recommend aligning the wording more consistently with the actual state of the literature. Please check this for all statements throughout the manuscript.
3.) You emphasize the differing properties of the four marine regions throughout the manuscript; however, their characteristics are not sufficiently introduced. Referring the reader to Sellegri et al. (2023) & Barthelmeß et al. (2025) in Lines 83/84 & 273/274 for this information is not helpful, as the manuscript should be understandable on its own without requiring prior reading of additional papers. I recommend summarizing and directly comparing the key features of these four regions within your manuscript, for example in an additional table. This could be combined with Section 3.3, potentially moved to the beginning of the Results section. Such a restructuring would help the reader better understand why these regimes are distinct and why differences in aerosol hydrogels are to be expected.
4.) Enrichment calculations: Were your sodium measurements in aerosol particles size-resolved? Based on Table A2, this does not appear to be the case, as only a single sodium value per tank experiment is reported. If so, how were size-resolved enrichment factors for hydrogels (as presented in Table 1) calculated? Since sodium is known to be predominantly associated with the coarse mode and occurs at much lower concentrations in submicron particles, using size-resolved sodium data is important for obtaining accurate enrichment factors. If such size-resolved sodium measurements are not available, the only consistent approach would be to calculate enrichment factors for the entire size range as a bulk rather than for individual size classes. This also holds true for the hydrogel/Na+ ratios.
5.) Please elaborate on the potential consequences for your tank SSA results arising from the use of 6 m deep water, without considering the SML.
6.)You state that atmospheric measurements of TEP and CSP are scarce, and further describe TEP as polysaccharide-based and CSP as protein-based. However, to my knowledge there are several studies from different groups investigating polysaccharides and proteins in SSA. Integrating your findings more systematically, for example in terms of aerosol size distributions and the potential contribution of gel-like versus non-gel-like polysaccharides and proteins, could strengthen the contextualization of your study and better embed it within the existing literature.
7.) I was confused throughout the manuscript as to whether you are referring to results from seawater or aerosol samples. Please ensure consistent nomenclature (e.g. abbreviations, brackets, indices) or a clearer overall structure to make this distinction unambiguous. I may have misunderstood some aspects due to this lack of clarity. What exactly do you mean by “marine samples” (e.g. L22)? Aren’t all samples marine in this study?
8.) Abbreviations are introduced repeatedly throughout the manuscript, which makes the text harder to follow. It would greatly improve readability if all abbreviations were defined once (ideally at first occurrence) and then used consistently thereafter, at least within the main text. I recommend carefully reviewing the manuscript from beginning to end to ensure consistent usage. For example, the term “subtropical front (STF)” is introduced multiple times in slightly different forms (e.g., L85, L186, several figure captions, Table A4, L507, L516). A similar pattern occurs for “SSW” (e.g., Lines 97, 104, 185, 246) and other abbreviations such as Chla and STW.
9.) Since you measured size-resolved hydrogels, did you consider visualizing their relative size distribution in an additional figure and comparing it with the few existing studies, either from field observations or laboratory experiments? Also comparing them between seawater and aerosol could be interesting.
More specific comments: