| 18 Nov 2025
Biochemical Characteristics of the Sea Surface Microlayer in the Central Baltic Sea and Potential Signatures of Cyanobacterial Blooms
Abstract. The sea surface microlayer (SML) forms the <1mm thin ocean’s boundary with the atmosphere and plays a critical role in mediating air–sea gas exchange and biogeochemical cycling. However, the biological processes shaping its molecular composition remain insufficiently understood. During a research cruise in the central Baltic Sea (Eastern Gotland Basin), we investigated how phytoplankton, including cyanobacteria, influence the biomolecular composition of the SML. Although no major bloom was detected, distinct shifts in phytoplankton composition were observed, leading to pronounced differences in biomolecular characteristics between the SML and underlying water (ULW), and between conditions characterized by high and low cyanobacteria abundance. While SML enrichment patterns and carbohydrate concentrations were comparable to those previously reported for the Western Baltic Sea, concentrations of total amino acids (TAA) and surfactants were substantially higher, suggesting enhanced production by cyanobacteria. Distinct molecular signatures were associated with different phytoplankton size classes. During periods of high abundance of small phytoplankton (HPA<20 µm; Synechococcus-dominated), the SML was characterized by elevated surfactant and total combined carbohydrate (TCCHO) concentrations. Furthermore, Synechococcus sp. co-varied with the non-protein amino acid g-aminobutyric acid (GABA), particularly under HPA<20 µm conditions. This suggests that the production of surface-active organic matter may be linked to Synechococcus sp. In contrast, under high abundance of large phytoplankton (HPA>20µm; filamentous and colonial cyanobacteria), total amino acids (TAA), particulate amino acids (PAA>20µm), and particulate combined carbohydrates (PCCHO>20µm) were enhanced in the ULW, mirroring POC>20µm and cyanobacterial biomass patterns. The significant correlation between phytoplankton >20µm biomass and POC>20µm suggests that the particulate organic carbon pool was largely cyanobacteria-derived, even in the absence of a distinct bloom. Together, our results imply that phytoplankton size structure and taxonomy exert distinct biomolecular imprints on SML chemistry in the Central Baltic Sea. The contrasting roles of filamentous/colonial cyanobacteria (proteinaceous signatures) and Synechococcus (carbohydrate/surface-activity imprint) imply community-dependent modulation of surface activity and, indicate that changes in biodiversity may potentially impact air–sea gas exchange in the ocean.
This manuscript presents a comprehensive field study on the biomolecular composition of the sea surface microlayer (SML) in the Central Baltic Sea, with a particular focus on the role of cyanobacteria and phytoplankton size structure. The dataset is rich and combines multiple complementary approaches, including microscopy, flow cytometry, detailed biomolecular analyses, and surfactant measurements. The study has clear potential to advance our understanding of how phytoplankton community structure imprints on SML chemistry and potentially affects air–sea exchange processes.
However, several issues need to be addressed before the manuscript can be considered for publication. Most importantly, the classification scheme based on HPA/LPA and phytoplankton size classes is conceptually confusing and, in its current form, hampers the interpretation of results and discussion. The conflation of size range and abundance category, as well as the use of cross-definitions (e.g. HPA>20 μm corresponding to LPA<20 μm), should be clarified or simplified, ideally by adopting more explicit ecological descriptors.
The manuscript contains numerous punctuation and formatting inconsistencies throughout the text, which should be carefully checked and corrected by the authors.
Overall, this is a potentially impactful contribution, but substantial revisions are required to improve conceptual clarity.
Line100: The manuscript discusses surfactants, amino acids, and carbohydrates in the SML in relation to cyanobacteria. However, it remains unclear whether these compounds are produced directly by cyanobacteria or whether they primarily result from bacterial processing and degradation of cyanobacterial biomass and exudates. The relative roles of cyanobacteria versus heterotrophic bacteria in shaping the SML biomolecular composition require clearer clarification.
Line103-105: The SML was sampled using a glass plate or Garrett Screen, whereas the ULW was sampled using nets and/or discrete water samples at ~1 m depth. Given the use of different sampling methods, it is unclear to what extent the observed molecular differences between the SML and ULW may be influenced by methodological biases, particularly with respect to particulate versus dissolved fractions. In addition, Table A1 suggests that ULW samples were collected using a Niskin bottle, which raises some ambiguity as to whether both net- and bottle-based approaches were used for ULW sampling. The authors should clarify the sampling strategy and discuss potential methodological effects on the SML–ULW comparison.
In Section 2.2, net samples containing the >20 μm particulate fraction were diluted with filtered seawater at ratios ranging from 2:1 to 1:2 prior to subsampling. It is unclear whether this dilution procedure may have altered the structure of colloids or extracellular polymeric substances (EPS), potentially affecting aggregation state and surface activity measurements. The authors should briefly discuss whether dilution could influence EPS integrity and surfactant analyses.
In the microphytoplankton microscopy analysis, community composition and biomass estimates were restricted to the eight most dominant species. While these taxa likely dominate in abundance, it remains unclear whether less abundant species with potentially large cell volumes may still contribute disproportionately to biomass, EPS production, or surface-active compounds.
In Section 2.7, the manuscript states that triplicate 18 mL SML and ULW samples were prepared, but only one of the three replicate samples was analyzed. It is unclear why only a single replicate was measured and what the purpose of the remaining replicates was.
In Section 3.1, the classification of phytoplankton abundance is confusing. The low-abundance category of phytoplankton >20 μm is labeled as “LPA<20 μm”, which is identical to the low-abundance category of phytoplankton <20 μm. This notation conflates size class and abundance level and may mislead readers to interpret the low-abundance >20 μm group as belonging to a different size fraction. Low-abundance categories should retain the same size designation (i.e. >20 μm) and differ only in abundance. In addition, Aphanizomenon sp. is listed twice in the species list. Finally, the notation “300 x 10³ cells mL⁻¹” does not follow standard scientific formatting and should be written as “300 × 10³”.
In the second half of this section, the logic of the HPA/LPA classification becomes increasingly difficult to follow and begins to affect the interpretation of the results. Statements such as “under HPA>20 μm (which corresponds mostly with LPA<20 μm conditions)” and “LPA>20 μm conditions (which corresponds to HPA<20 μm conditions)” introduce cross-definitions that obscure which ecological states are actually being compared.
In Section 3.2, there is an inconsistency between the statistical significance and the descriptive language used to interpret the results. For TAA, terms such as “tendency,” “slightly higher,” and “more pronounced” are repeatedly used, while the corresponding statistical tests are either not significant (e.g. p = 0.076) or not explicitly reported.
In Section 4.2, the manuscript repeatedly emphasizes that enrichment factors (EFs) are close to 1, while at the same time using relatively strong wording such as “significantly higher,” “substantially higher,” or “highest average EF” to describe differences. For example, the mean TAA EF of 1.2 ± 0.4 largely overlaps with the previously reported range of 0.8–1.2, raising the question of whether this truly represents enhanced enrichment in a statistical or process-based sense. It may be more appropriate to describe these values as being at the upper end of the historical range or as consistent with, but slightly elevated relative to, previous studies, rather than implying a clearly enhanced enrichment.
4.3 Section, In the discussion of POC composition, the authors state that cellular biomass accounts for only ~13.5% of POC>20 μm, with the remainder attributed to heterotrophic plankton, detritus, and extracellular material. This interpretation is conceptually sound and well supported by the literature. However, the subsequent statement that “POC>20 μm was related to cyanobacterial cellular biomass >20 μm (HPA>20 μm: 1.6 ± 1.7; LPA>20 μm: 0.7 ± 0.7%)” is unclear. It is not evident whether these percentages refer to correlation strength, explained variance, or relative contribution to POC. The authors should clarify the meaning of these values and how they were derived. In addition, “Synecococcus sp.” appears to be a typographical error and should be corrected to “Synechococcus sp.”.
Line19: Please define the abbreviation “HPA” at its first occurrence.
Line36: Please add a space after the citation “(Engel et al., 2017)”.
Line53: The phrase “A large-scale oceanic in the Atlantic demonstrated” appears incomplete. A noun such as “study”, “survey”, or “investigation” seems to be missing.
Line 121: In “>20μm”, a space is missing before “μm”. Please check and correct similar formatting issues throughout the manuscript.
Line 123: “Garret” should be corrected to “Garrett”.
Line 130: The units “knt” and “knots” are used interchangeably. Please standardize the unit notation.
Line 131: In “~14min”, please add a space before “min”.
Line 146: “within ≤2 h” is recommended to be revised to “within 2 h”.
Line 153: “Bottle” should be “bottle”.
Line 185: Please provide a reference for the chlorophyll a measurement method.
Line 215: “sample” should be corrected to “samples”.
Line 223-224: Thirteen amino acids are stated, but only twelve are listed. In addition, please revise the punctuation in the references according to journal style.
Line 225: In “(Dittmar et al., 2009.; Engel & Händel, 2011)”, there is an extra punctuation mark after “2009”.
Line 231: “Sample were measured” should be corrected to “Samples were measured”.
Line 236: “the samples ionic strength was standardized” should be revised to “the sample’s ionic strength was standardized”.
Line 327, In the sentence “Cylindrotheca closterium and Chaetoceros sp. occurred at HPA>20 μm only with < 0.2 μg C L⁻¹)”, there is an extra closing parenthesis.
Line 407: The term “later amino acids” is incorrect and should be replaced with “latter amino acids”.
Line 419: “glucosamin” is missing the final “e” and should be corrected to “glucosamine”.
Line 435: The phrase “excluded … due to their low and abundance” is incomplete and should read “low abundance”.
Line 698: “Moring” should be corrected to “Morning”.