The presented study conducted two ship-based campaigns in the Baltic Sea in May and August 2021, and performed chamber experiments to investigate the emissions of biological particles with SSA with particular focus on bacteria. The experimental design was very comprehensive, and the results are useful to understand the air-sea exchange of bacteria. The manuscript is presented well, and I have only several comments to be clarified, mainly related to the methodology and the explanation of some results.

General comments:

DAPI staining also dyes fungal spores and other biological particles containing DNA blue under the UV excitation. How did the authors distinguish bacterial cells from other biological particles for FM enumeration? How large were the measured bacterial cells? The authors give some two reasons (Line 261-267) why the concentration of airborne bacteria 30-40 times higher than that of fPBAP (Figure 2, Line 249-257). For the first reason, were viruses and cell fragments enumerated using FM? For the second reason, did the author find some gels in FM images (e.g., Fig. 4 in Aller et al. 2005)? Because sonication was only 1 minute, some gels may still exist.

Figure 2: Why was the concentration of fPBAP likely quite stable? It is almost unchanged, especially when the chamber was operated in closed mode. What is the possible reason? Could it be a background signal? Are there any variations in the concentrations of CP and FP? Why not illustrate them in the figure?

Line 247: “This contrasts with previous studies that have reported higher bacterial concentrations in the SML”. Why was the result contrast to previous studies, e.g., Aller et al., 2005? The authors give two explanations in Line 282: What was the range of wind speed? Can SML form or keep at those wind speeds? How much was the thickness of sampled SML? Line 277-279, “This decrease could be attributed to the higher jet flow rate potentially preventing the formation of a SML, which typically contains higher concentrations of bacteria and organic material.” This explanation should also be based on the fact that bacteria were enriched during transport from subsurface waters to the SML. If there was no enrichment, the explanation is false.

Line 258-260: Did the authors measure the abundance of biological and fluorescent matter or any other proxies, for instances, Chl a, in surface water during August 2021 and June 2018? If measured, it is better to note them directly.

Line 366: Why did the author use “a factor of 0.00086 (mean, range 0.0003 - 0.0013)”? The authors mentioned “Notably, fPBAP represented only a minor fraction of all FP (1.4±2.7%) and an even smaller proportion of CP (0.15±0.38%)”. Was the fraction of bacteria cells determined from fluorescence microscopy to the total aerosol concentration larger than 0.15% in this study? Again, the measured size ranges of bacterial cells and total aerosol particles should be the same to compare.

Sect. 3.1.4: As mentioned in the abstract “16S rRNA sequencing identified the diversity of bacteria enriched in the nascent SSA compared to the underlying seawater.” But the description in this section does not mention it clearly, and the authors mentioned the difference/enrichment in the R/V Oceania campaign was not obvious. There are no statistic results for these comparisons, for instance, the values of richness index. Was the difference in the R/V Electra campaign significant?

Technical comments:

Line 19: cloud and ice condensation nuclei---> cloud condensation nuclei and ice nuclei

Table S1 and S2: what do the blanks, (X) and X mean? Does (X) mean that the analysis was not conducted?

S4: The genus and species should be in italic, and the name of each species should consist of two parts: the genus name and the specific epithet (species name).

Line 123: Bacterial enumeration

Line 194: WELAS--->OPSS

Line 199: MBS detects and sizes particles?

Line 231: 5.2±1.1 mg m−3

Line 413: add “richness index”?

Review for the manuscript “Quantification and characterization of primary biological aerosol particles and bacteria aerosolized from Baltic seawater”

Authors presented a study based on two campaigns performed on ships in Baltic Sea with aim to expand the knowledge on primary biological particle (PBAP) aerosolization with sea spray aerosols (SSAs). To achieve this goal, aerosols were generated in highly controlled plunging jet sea spray simulation chamber that circulates bulk sea water with ship’s flow-through system at 1.5m depth. The aerosol size distribution was measured with DMPS and OPSS, and fluorescent particles were measured online with an MBS. In addition to online measurements, aerosols were collected on filters for following bacterial enumeration with fluorescence microscopy, bacterial specie composition determination with 16S rRNA metabarcoding and ion chromatography, and sea water samples were collected at filter exchanges. The collected data was used to estimate bacterial emission fluxes, enrichment of bacteria in SSA compared to bulk sea water and further, enrichment of specific classes in SSA compared to sea water.

General comments

• There have been very few studies attempting to estimate the emission flux of bacteria in SSA and to determine the enrichment factors for different species to date. Authors presented estimations using two approaches and compared to the previously published estimations. However, the estimations rely heavily on parametrizations derived by Mårtensson et al (2003) restricting application of these estimations to the size range of the 0.02 < D_p < 2.8 µm.
• Authors encountered two controversial findings during the study: no difference in specie composition between sea microlayer samples and bulk sea water, and occurrence of ASVs in aerosol samples not observed in bulk sea water that these species were aerosolized from. These observations were discussed in the manuscript to an extent, but would require further clarifications (see specific comments).
• Figures, and supplementary material captions to the figures and tables require refinement. More details in specific and technical comments.

Specific comments

Lines 38-39: “Recent observations indicate that the contribution of the SML to SSA is relatively small compared to bubbles originating from subsurface water (Chingin et al., 2018; Frossard et al., 2019)”. Do the authors mean the contribution of SML to production of SSA from jet drops? Or from the point of view of only biological SSA, which are mainly concentrated in jet drops? As is, doesn’t this statement contradict with the findings of Wang et. al (2017) you mentioned in previous sentence, who indicated that the fraction of SSA produced with jet drops varies from 20% in natural sea water to 43% during algal bloom?

Lines 118-119: “However, due to poor staining, bacteria could not be confidently enumerated during the R/V Oceania campaign, so these samples will only be discussed in terms of bacterial community composition.” Were there any differences in sample handling that led to poor staining? Was the same staining method used for both campaigns or was it improved for the Electra campaign?

Section 2.2: Bacterial enumeration was performed using DAPI staining of particles larger than 0.2 µm in diameter and fluorescence microscopy. DAPI stains dsDNA in all PBAPs, did you differentiate bacteria from other species, for example fungi or diatoms, based on morphology of the observed cells? If yes, what were the criteria for classification of the cells as bacterial? What was the size range of the cells?

Figure 2: The bacterial concentration determined with fluorescence microscopy for dates 16/08-17/08 and 18/08 seem to have a slight decline though according to Figure, the chamber was in closed state. Is this observation significant in terms of sensitivity of the analysis? Would you contribute this decline in bacterial concentration to the cell disruption with plunging jet? If so, has the system been tested for disruption of microorganisms in closed state for prolonged periods of time?

Section 3.1.1: According to your results, sea microlayer (SML) samples did not differ from bulk sea water (SW) samples in bacterial species composition, during both ship campaigns. However, the R/V Electra campaign coincided with the bloom of filamentous cyanobacteria described in the results 3.1. and shown with calculated chlorophyll a concentration. This should show in high abundancy of 16S rRNA of cyanobacteria in the SML samples compared to SW samples, which was not the case in this study. This finding was explained by high windspeed and potential mixing of SML with underlying sea water during sampling. Unfortunately, the Figure S5 in Supplements is a copy of Figure 5, which does not indicate the relation of enrichment factor compared to the wind speed. Could the figure be updated? The wind speed was high during Electra campaign, was it also the case for Oceania campaign? If the mixing of SML with underlying water was due to sampling method, could you evaluate the reliability the results from Oceania campaign SML samples? Could the vicinity of ship affect the collection SML samples? In case if high windspeeds, could movement of the ship lead to mechanical mixing of surrounding waters?

Section 3.1.3: The bacterial emission fluxes were estimated using two approaches, both for bacteria in size range of 0.02 < D_p < 2.8 µm to which Mårtensson et al (2003) parametrization is valid, whereas the bacteria enumerated in campaign was sized larger than 0.2µm. Was it was accounted in the emission fluxes calculated here? What would be the estimate of the significance of emission of the bacteria and bacterial agglomerates larger than 2.8µm?

In the section 3.1.4, among the taxa observed in air samples, there were unique ASVs not found among SW and SML. Authors state that “One possible explanation for this could be selective aerosolization of certain taxa.”. If the selective aerosolization occurred, shouldn’t the same ASVs be also observed in SW and SML samples, but in lower abundancies? What is the limit of detection with chosen extraction and amplification method used? Authors suggest that that potentially some taxa may be “entirely removed from the seawater sample” due to selective aerosolization, but there are 357 ASVs from Oceania campaign and 102 ASVs from Electra campaign. How realistic is this this possibility? Another potential explanation was “the presence of free DNA in the head space of the chamber that can pass through the HEPA-filters”. HEPA filters are 99.97% efficient in filtering air and given intensive SSA production with plunging jet, shouldn’t the possible contamination through HEPA filters be much smaller than the observed number of unique ASVs in SSC samples? What were the controls and were there any indication of contaminants in them? Could this be results of over-amplification due to triple PCR? Could it be misclassification introduced by bioinformatic pipeline?

Lines 426-435: The enrichment of species in SSC samples compared to SW and SML requires more detailed look. In lines 429-431 “However, in contrast to these studies, we did not find a significant enrichment in Acidimicrobiia, Bacteroidia and Verrucomicrobia and decreased abundances of Actinobacteria and Alphaproteobacteria in SSC compared to the SW during both campaigns.” However, the Figure 5b indicates clear decrease in Actinobacteria and Alphaproteobacteria in SSC samples compared SW and SML samples, while the abundance changes in Gammaproteobacteria described in lines 426-428 is much more subtle. Was there statistical analysis performed to support the observations? Could you also discuss the enrichment of Cyanobacteria in SSC compared to SW and SML?

Figure 5: The results for different sample types from the same days are difficult to compare in the given format. Arranging the X-axis temporally would help comparing the results. Also, in panel (b), sample SW_210811 is presented twice, with different data. Which one is the correct one?

Figure S4: The x-axes are undescriptive; I would suggest changing them to be by date and below describes the sample type – bulk sea water and SML samples separately.

Technical corrections

Lines 29-32: Please add references to the text.

Line 31: Words “film drop” is written twice in a row, remove one repetition.

Lines 48-49: “Recent advancements in real-time single-particle analysis instruments using ultra-violet light-induced fluorescence allow continuous monitoring of fluorescent (fPBAP)”. I suggest changing “… fluorescent (fPBAP)” to “… fluorescent PBAP (fPBAP)” and to add few examples of aforementioned instruments.

Figure 1: “Dry air genator” -> “Dry air generator”. In caption “mulitparameter” -> "multiparameter"

Figure 4: Could the same classes be shown in one segment? For example, Gammaproteobacteria class is shown in two segments in panel (a).

Table S1: Insert description of the different abbreviations used in the table to the caption.

Figure S5: same as Figure 5, repeated by accident? Update the figure to match the caption.