This study presents a comprehensive analysis to show that locally emitted fungal spores can serve as ice nucleation particles at high temperatures at a site in the European sub-Arctic. The analysis of the fluorescent aerosols is in-depth and convincing. This study provides diverse methods to consolidate the statements, including chemical, biological, and meteorological techniques. I believe that it is worth publishing in this journal when the following issues are resolved. Overall, I suggest a major revision based on the comments below.

Major comments:

1. 3.3: When the station is in-cloud, how can cloud-processing can influence the concentration and type of the HFAPs, especially AB and ABC particles?
2. 3.4 and Figure 6: please explain why in different seasons, the patterns are different for the same meteorological factor.
3. Section 3.7: in the last two paragraphs of this section, the authors discuss the observations of INPs and fluorescent aerosols in other studies. However, most of discussions do not differentiate highly fluorescent aerosols and all fluorescent aerosols. This might be partly the reason that some observations do not find relationships. Please refine the discussions by explicitly considering the fluorescence intensity.

Specific comments:

1. L27-29: currently, the link between the first part of this sentence and the “their activation temperature” is not clear. Please rephrase this sentence.
2. L162: “identifiable spore if both are meant”: not clear. Do you mean impossible to distinguish between these two types?
3. Figure 2a: I suggest modifying the axis label as “Fraction of HFAPs in TAPs”.
4. Figure 2 legend: I suggest adding one sentence to tell the reason why AC is not shown in the plot.
5. Figure 3: the y axis for temperature is absent. “Horizontal transparent red lines” should be “Vertical …”.
6. Figure 4a: what does “global” mean in “global radiation”?
7. Figure 4b: Do NON-HFAP also include FAPs other than the HFAPs?
8. L347: please add the data for AB, e.g., in SI.
9. Figure 6: in (b) the ABC concentration can > 5 *10-3 cm-3, however, in (c) the same overall data become <2 when binned by wind speed. Please check the processing of the data.
10. L480: not only specific to “this temperature”, but also to the characterized freezing mode.
11. Figure 10c: why are the parts for red and black solid lines at the lowest T range absent while those of the dashed lines are there?
12. Figure S3: (a) and (b) are the same as those in Figure 4. Just keep (c).
13. SI text 1: both WIBS and APS can characterize the size distribution of the aerosols. A direct comparison between the size distributions of total aerosol particles is much better than the overall comparison of the total concentration in reflecting the characterizing ability of the two instruments.

Technical comments:

1. L101: to me, “LIF” is more commonly used than “LiF”
2. L175-176: “AE”, “LGC”, and “ITS” are not defined.
3. L189: “LMP” is not defined.
4. L255: “does no” should be “does not”.
5. “3.1 The contributions of different HFAPs changes seasonally”: “changes” should be “change”
6. L549: Should be “of which none has been tested for INA”.
7. L552: “two” should be capitalized for the first letter.
8. L559: should be “one of the most abundant”.

Gratzl et al. present findings from a long-term monitoring campaign using a plethora of aerosol, bioaerosol and INP measurement instrumentation to determine the identity of high-temperature ice-nucleating particles as being local fungal spores in the European sub-Arctic. The analysis and correlations from the multiple datasets are robust and in-depth, honing in on the identification of fungal spore populations as defining the INP population and the influence that this finding could have in a warming climate. The work represents a step forward in better understanding the nature of biological INPs in the atmosphere, which is notoriously difficult given the complexity of such samples, and highlights the need for long-term monitoring campaigns with appropriate instrumentation and design of experiments for elucidation of atmospheric INP populations. This article demonstrates very useful findings and I believe it should be published following minor revisions, in particular one major comment surrounding bacterial INPs.

Major comments:

1. Fungal spores and pollen are discussed in detail, but I am surprised that there is no discussion on the potential influence of bacteria. 16S analysis is often performed alongside ITS. It is surprising to not see the influence of bacteria (and their proteins) investigated despite their mention in the Introduction and a brief discussion at the end of the article. The authors give the impression that bacterial INPs are active at much warmer temperatures (> -10 °C), but this is not necessarily always the case. Table 2 makes clear that fungal spores can cover a range of activities from -3.5 °C to -35 °C (or have no INP activity), but the same argument is not made here for bacteria, and so bacterial influences are dismissed despite being in a rich biome that is likely home to a wide range of bacterial diversity.
2. Line 455: Make it clear that Cladosporium is an Ascomycota. This is somewhat confusing, however, as on one hand Cladosporium is stated as being the most abundant identifiable fungal spore type, but then Ascomycota are shown in Fig 9 as being low in relative abundance. Were identifiable spore types in very low abundance compared to the other classifications?

Minor comments:

1. Line 53: Some additional fungal INPs may be found in the recent list compiled by Tarn et al.: https://doi.org/10.1063/5.0236911
2. Line 60: The authors should cite recent papers from the group of Santl-Temkiv related to biological sources of INPs in the Arctic; while these are related to INPs in marine waters, they may represent an atmospheric source: https://doi.org/10.5194/ar-3-81-2025; https://doi.org/10.5194/acp-25-3327-2025
3. Line 61: What about the Arctic Haze?
4. Line 64: “Attributed” instead of “contributed”.
5. Line 85: Does this campaign (particularly the second half in 2023) have a name?
6. Line 88: Give the size range of the APS.
7. Line 92: Did fog vs. no fog have an impact on ns?
8. Line 101: “Laser-induced”, not “light-induced”.
9. Line 113: Please put these (A, B, AB etc.) into a table. It will be much easier to follow throughout the paper.
10. Line 128: What kind of filter hold and filter system were used? What inlet was used (TSP? PM10?)? What was the pump and how was the orifice fabricated (unless it was commercial)?
11. Line 128: What were the typical sampling times and volume collected?
12. Line 130: What volume of water were the filters washed in?
13. Line131: How many droplets? 96? 384?
14. Line 134: Please provide the INP equation, and in particular a brief description of the uncertainties that are mentioned and their calculation.
15. Line 139: In particular, such heat-labile INPs are believed to be “proteinaceous”, since the heat is expected to denature proteins.
16. Line 146: In which mode was PINE operated here? What was the temperature range? Was it operated at one temperature or was it thermocycled?
17. Line 158: So there is data from every day that the trap was used?
18. Line 170: Use “Environmental DNA (eDNA)…” the first time the technique is discussed (or in the sub-heading).
19. Line 174: What is AIT?
20. Section 2.6 (eDNA): Do any of these abbreviations mean anything or are they all proprietary? E.g. API, P3 AE, AIT? Define ITS, but also, what is meant by ITS2 region?
21. Section 2.7: For how long was the impactor usually run, and what volume of air was usually sampled? Was an after-filter installed to capture the smaller particles that passed through the main impactor cut-offs? What was the method for analysis under brightfield and fluorescence microscopy, what was being looked for?
22. Line 256: “since their high contribution during warmer months suggests an biological influence on their concentration.” – But are these not already HFAPs? Unless I am missing something here, then there must naturally be a biological influence on the concentration of biological particles.
23. Section 3.2: Could there be an influence of blowing snow on INP concentrations? (see for example, https://doi.org/10.1525/elementa.2024.00047). Are there any correlations with windspeed?
24. Figure 2 caption: I would consider adding definitions of HFAPs and TAPs to this first figure caption where they are mentioned.
25. Figure 3: Is there meant to be no temperature scale? This seems like an oversight.
26. Figure 3: Even under maximum snow cover, HFAPs are higher than in early winter. Were there any potential sources of HFAPs in the area? Lichen on trees? Moss on rocks?
27. eDNA sequencing: It is unfortunate that this was not performed while the WIBS, Burkard or APS were not running. Can the authors comment on the inability to correlate this data with any of the other bioanalytical techniques and how that may impact the interpretation of the results?
28. Line 287: I am surprised to see so little discussion of the HyIce INP results from the 2018 campaign, other than a very brief mention of Vogel et al. (2024) much later in the article.
29. The English is generally good throughout, but there are some notable spelling and grammatical errors throughout. The article should be proofread thoroughly prior to final submission.
30. Line 362: What are the sizes of these grains and how do they compare to the WIBS cut-off? I see this is discussed a little later and the pine pollen sizes were generally very large, but what about the other pollen grains? What is the detection limit of grain size on the Burkard trap analysis method? How do these other pollen grains, particularly those in the appropriate size range (if possible depending on the method’s limitations) correlate with ABC_3?
31. Line 387: “These are starch granules which can be expelled from pollen grains…” – I do not think it is correct to definitively say that they are starch granules, unless I am mistaken pollen can expel various SPPs, including starch molecules, but also presumably pollen grains can expel a number of sup-pollen particles including various polysaccharides (of which starch is one type), as well as proteins and glycoproteins.
32. Figure 7: What happens when ABC_3 are correlated against spores, while ABC* are correlated against pine pollen?
33. Figure 7b: Although there’s an apparent correlation, the number of data points above ~500 m-3 for pollen are very small.
34. Section 3.6: How does these results correlate with the ABC* discussion in the previous section? ABC is discussed here but not ABC*.
35. Line 455: This part is confusing; if 0.44 % of fungal reads are from Cladosporium then why is it being scaled to 100 %?
36. Line 489: Bacteria may exhibit a range of activation temperatures, how can you be so sure that there is no influence of bacterial populations in the range of interest here (i.e. around -10 to -15 oC) and that it is nearly all fungal spores?
37. Line 496: Do dust particles show up in any WIBS channels?
38. Line 505: Qualitatively speaking, how do the heat treatments compare to the WIBS data in terms of HFAP loading?
39. Line 540: Be a little careful here, several biological species defined as not having ice-nucleating properties have only been tested at warmer temperatures due to the type of instrument used or the nature of the method. Some may therefore exhibit activity at colder temperatures. I cannot speak to all of the examples provided here - some were clearly tested throughout a wide temperature range down to homogeneous freezing, so the statement of “Not all species within a genus exhibit ice nucleation” still holds true - but it is worth bearing in mind in general.
40. I am surprised to see so little discussion of the PINE data in Section 3.7 or 3.8 given that it was concurrent to and with similar resolution as the WIBS. It is certainly used (e.g. Figure 10b), but one might expect time series of PINE vs. WIBS for different groups (A, ABC) to help track the correlation. Is there are reason that such data is not shown?
41. There are several similarities between this article and the authors’ recent one in ESSD given that it was from the same campaign and discusses FBAPs and TAPs from the WIBS and other instruments. The authors should make clear in this article what is new compared to the ESSD publication.

This manuscript presents measurements in Northern Finland of INPs, HFAPs, spore and pollen counts, and DNA sequencing. The authors suggest that local fungal spores may dominate the high temperature INP population in the region, with possible implications for Arctic cloud processes under future climate scenarios. I believe this article is in scope with ACP after minor revisions based on the following comments.

Abstract:

L.10 It is great to use multiple data types (INP, HFAP and DNA), however please consider rephrasing “for the first time” to something like “for the first time to the best of our knowledge…” unless a thorough literature review confirms this claim. 

L.12 “Findings” is incorrectly capitalized, and the phrasing “Our findings indicate that” is used multiple times in close proximity (L.7 and L.12). Please revise for stylistic clarity and to avoid redundancy.

L.13-14 it’s unclear whether this study tested fungi for ice nucleation activity (INA), or whether this refers to results from previous studies. Please clarify and specify.

Introduction:

L.20 Rather than citing only a review, please consider referencing original studies that include field-based INP concentration measurements to strengthen this statement.

L.33-34 The statement that biological INPs were discovered in the 1970s overlooks earlier work. Please revise or reformulate.

L.51 “Less is known about the INA of fungal spores” could be more clearly phrased. Also, recent work from the group of Konrad Meister (e.g., Schwidetzky et al., 2023) should be referenced here to reflect the current state of knowledge.

L.55 It may strengthen this section to reference Sanchez-Marroquin et al. (2021), who combined INP measurements with SEM imaging and found mostly Ascomycota-type spores, with only occasional detection of basidiospores.

L.60 Ideally define the cutoff of "high-temperature INPs" clearly (e.g., “active above –15 °C”).

L.60 The current formulation implies that INPs active above –15 °C are specific to terrestrial sources in Arctic and sub-Arctic regions. Please rephrase or reflect the broader geographic relevance and support the statement with appropriate references e.g. Mason et al. 2015.

L.68 The connection between INP measurements and improved climate predictions is made too abruptly. Consider adding a more gradual buildup that explains how INPs influence cloud microphysics, which in turn affects cloud feedbacks and radiative forcing. This would help avoid overstatement.

L.70 The hypothesis at the end comes across somewhat surprising. Please provide a rationale for this hypothesis earlier in the introduction.

Methods

L.93 Could the authors please specify how much the inlet is heated? Additionally, a brief discussion on how this may affect the collected DNA and INP concentrations would be appreciated.

L.95 Are the WIBS concentrations reported per L or standard L?

L.120–121 The phrasing in this sentence could be refined for better readability.

L.159–162 Could the authors please clarify how plant spores were distinguished from fungal spores? Including typical examples or representative images would be helpful.

Section 2.6 (DNA Sequencing) For clarity, it would be helpful to define the following abbreviations: AIT, AP1, P3, AE, LGC, ITS, LMP, V3, bp, OTU.

Results and Discussion

L.257 Please clarify the meaning of “a biological influence on their concentration.”

L.329 Did the analysis focus only on the effect of rain on HFAP, or were INP, DNA, or spore counts also evaluated? 

Supplement L.95 How frequently were rain events observed, and what was the approximate rainfall amount during those events? Could the authors also comment on how representative the two examples shown in Fig. S9 are?

L.347–351 A brief interpretation of these results would be helpful to guide the reader through their implications.

L.358 Is there a rationale for the selection of the FL1 threshold? Providing a brief justification would improve the transparency of the method.

Fig. 7 (ABC vs identifiable spores) Maybe including logarithmic tick marks on the x-axis.

L.395–396 Please clarify whether the following statements refer to ABC or ABC* particles. Additionally, rather than stating “marked orange on Fig. 7a,” it may be clearer to say “marked as orange vertical lines in the lower subplot of Fig. 7a.”

Table 1: The phrase “The concentration differences between identifiable spores and ABC particles are 1, 2 and 3 orders of magnitude” could be clarified.

L.384 Please briefly explain the reasoning for ruling out ABC_3 particles as pollen grains here.

Section 3.8 It would be helpful to clarify whether you really mean ABC here (or rather ABC* or ABC_3 as differentiated in previous section).

L.481 The statement that all INPs at –13.5 °C are ABC particles may be too strong. As correlation does not imply causation, a more cautious interpretation would be advisable.

L.492 Should the phrase “for temperatures > –8.5 °C” be revised to “for INPs active at temperatures above –8.5 °C”? Please clarify.

L.519 This finding is consistent with Mignani et al. (2021), who observed that aerosol particles >2 µm were better predictors for INPs active at –15 °C than particles >0.5 µm at Weissfluhjoch, Switzerland.

References

Mason, R. et al. (2015). Ice nucleating particles at a coastal marine boundary layer site: correlations with aerosol type and meteorological conditions, Atmos. Chem. Phys., 15, 12547–12566, https://doi.org/10.5194/acp-15-12547-2015

Mignani, C. et al. (2021). Towards parameterising atmospheric concentrations of ice-nucleating particles active at moderate supercooling, Atmos. Chem. Phys., 21, 657–664, https://doi.org/10.5194/acp-21-657-2021

Sanchez-Marroquin et al. (2021). Mineral and biological ice-nucleating particles above the South East of the British Isles, Environ. Sci.: Atmos., 1, 176-191. https://doi.org/10.1039/D1EA00003A

Schwidetzky, R. et al. (2023). Functional aggregation of cell-free proteins enables fungal ice nucleation, Proc. Natl. Acad. Sci. U.S.A. 120 (46) e2303243120, https://doi.org/10.1073/pnas.2303243120

I agree with the overall assessment by the other three Referees that this is a well conceived and performed investigation worth publishing. My only comment in addition to what has already been said, relates to an issue triggered by Figure 3b in combination with Table 2.  Figure 3b suggest that temperature alone can explain much of variation in ABC concentration, while snow cover does not seem to have an additional effect. Otherwise, there would be a step change between pink (snow covered period) and blue (no snow cover) dots in this graph. Yet, both clouds of dots nicely merge, which contradicts the statement in line 271 that “Snow cover significantly affects AB and ABC particle concentrations..” This observation makes me wonder whether the source of ABC particles is not to be sought above ground? Likely candidates are, like Referee #2 points out in their comment 26, lichen or moss on trees or rocks, which brings me to Table 2 from which the ice-nucleation active lichen mycobionts reported by Kieft and Ahmadjian (1989;  https://doi.org/10.1017/S0024282989000599 ) are missing. It would be regrettable, if this potential source of ABC particles and INPs was not considered in a revised version of this paper.

Sorry for the mistake: In my comment posted earlier today I meant Figure 5b, not 3b.