GENERAL COMMENTS

This is a very interesting and well-written manuscript that defends the idea that ice nucleation activity in pollen is probably the result of exaptation of another trait related to the polysaccharides that englobe pollen grains. The methods are clear and the conclusions are justified.  It would improve the manuscript if the authors considered making an analysis and discussion about the amount of soluble ice nucleation active material per pollen grain per plant species.  Is this a uniform trait across plant species and is it stable in face of environmental conditions?  Those considerations have implications for the comparisons made here and the impact on cloud processes and plant biology.

SPECIFIC COMMENTS

Title: The title is not clear.  Firstly, one might ask what variability.  Secondly, it would be more appropriate to say “suggests” rather than “indicates”.  A better title could be “High interspecific variability in ice nucleation activity suggests pollen ice nucleators are incidental”

L 30. This is an outdated taxonomy.  There are 3 domains of life: Eukaryotes, Prokaryotes and Archaea.  Within the Eurkayrotes there are 4 kingdoms: plants, animals, eumycota (true fungi) and the Stramenopila (formerly Chromista) that contain flagellated organisms. And, there are also the “non living” organisms, i.e. viruses, for which some have wondered about their ice nucleation activity. Taxonomy is always changing (see wikipedia for a summary of this change: https://en.wikipedia.org/wiki/Kingdom_(biology)).  The number and names of the taxonomic divisions is not really pertinent to this work. I recommend simply stating that ice nucleation is wide spread across many different types of organisms and indicate examples.

L 39. This section describes pollen in terms of its chemistry.  However, the authors suggest that knowledge about INA of pollen would have implications for atmospheric science. Hence, the authors should also tell the reader about the capacity for pollen to become airborne.  Indeed, is all pollen meant to fly? Overall, the authors should introduce information about the biology and “life cycle” of pollen.

L 39. Furthermore, in this section, here the authors indicate that ice nucleation activity is assessed from solutions of material that is washed from the pollen grains.  This information needs to be emphasized to help readers understand that ice nucleation activity will NOT be assessed strictly on a per pollen grain basis via tests of suspensions of pollen grains.  I had not paid attention to this detail in my first round of reading the manuscript and it threw me off – especially given that the Materials and Methods section comes after the Results section and that I have the unfortunate habit of reading a manuscript from beginning to end in the order that it is presented.  Here in the introduction section it would be very useful to present information on the amount of INA material released per grain and if this is constant among plant species.  This is critical information because, for pollen INA to be pertinent for the life history of a plant, it needs to function on a whole grain.  In addition, even if the material in the washed-off form is active in cloud processes, it is unlikely that its distribution in the atmosphere is homogenous – it is more likely to follow the distribution of the pollen grains themselves.  If the amount of INA material per g (or per grain) of pollen is very different among species, how can this be accounted for in the analyses and can the results be expressed on a per grain basis to allow for pertinent comparisons?

L 84. The authors have written “measurements ...... has....”.  Please correct the grammar.

L 85. Here the authors mention pollination by animals.  Some readers might be lost if there is not a presentation of the different types of pollen, as suggested above.

L 216. Here the authors explain that their results do not support that the “underlying biological function” of pollen nucleating ice is a bioprecipitation mechanism.  It might be more precise to state that their results do not support that “there is a positive selection pressure - for pollen in general - of rainfall on ice nucleation capacity.  Ice nucleation might indeed be involved in depositing pollen from the atmosphere in some cases, but their results show that there is no signal of positive selection pressure for this across all of the different types of pollen and not for the wind-disseminated pollen in particular. 

L 291-296. In this paragraph the authors indicate that the microbiology of the pollen could be a factor in the ice nucleation activity.  For pollen collected from outdoor sources, this is a real possibility.  The authors could mention how they could have accounted for this factor i.e. microbiological analyses). Furthermore, the comparable results for filtered and raw suspension is a strong suggestion that the INA they measure is not due primarily to the microflora of the pollen.

L 333. The authors state that droplet arrays were cooled at a rate of 2 °C/min.  Why this rate of cooling?  Does it correspond to the rate that the pollen INA materials would encounter in nature?

L 356. The authors calculated the number of nucleation sites per gram pollen.  Is it possible to calculate per pollen grain? Is the weight of pollen homogenous across plant species?  Furthermore, is the weight of pollen stable for any given plant species or is it variable depending on environmental conditions? I would expect these topics to be addressed somewhere in the manuscript.

L 387 and onward. In the Conclusion, could the authors discuss the plasticity of plants in terms of the amount of INA polysaccharide that it produces on pollen grains?  Is this is stable trait?  Is it susceptible to environmental conditions and which ones?  This could give insight into the differences observed between species even if the role of the trait is incidental concerning ice nucleation per se

This is a well-written and important paper which fits well into the journal Biogeosciences. This manuscript should be published after some changes:

Main comments

When Pummer 2012 recognized that soluble macromolecules can trigger heterogeneous ice nucleation, they assumed polysaccharides being responsible. The same conclusion was later also drawn by other authors, e.g. Dreischmeier 2017 and Gute 2020. The reason is that in the FTIR spectra the bands of polysaccharides are so intense that they overlay all other signals. However, Pummer 2013 already detected protein signals in their detailed study by Raman and FTIR spectroscopy. Only recently Burkart 2021 found evidence that proteins are present and are the responsible INMs. The fact that proteins and polysaccharides are present in the same solution might account for inherent mixtures of both or even for glycoproteins as the important INMs. In contrast to FTIR spectroscopy, fluorescence spectroscopy can clearly differentiate the proteins. Therefore, I strongly recommend to add fluorescence-excitation-emission-maps to figure 5, in order to have more meaningful results.

In figure 3 the authors have correlated the representative nucleation temperature with biological and geographical parameters and show that the impacts of these are not significant. However, when comparing fig 4 with the results in figure 3f, it becomes obvious that the growth elevation is only categorized in three classes, the selection of which is not clear. In literature, at least 4 categories are known, i.e. mountain zones 0-1800m, 1800-2300m, 2300-3000m and 3000-xm. When applying these mountain zone categories then a difference might become visible showing the change from alpine to snow zone being related to a significant increase of the nucleation temperature. In general, I appreciate such correlation boxplots. However, I wonder that the authors did only correlate representative nucleation temperature but did not also correlate other important parameters such as extractable amount of INMs, average mass of the INMs, size of the INMs, sizes of the aggregates of the INMs or even the intensity of the fluorescence signal (related to the protein concentration). This would significantly enhance the information value of the paper.

Minor comments

The term “INMs” has been coined by Pummer 2012 as “ice nucleating macromolecules”. Please add “macro“ when referring to this definition.

INM mg-1 extracted pollen was the general value. Unfortunately, this is not a very precise value since pollen have different amounts of extractable material on their surface (see Burkart 2021). More precise would be to determine the amount of soluble material in the solution subsequently to the extraction process by evaporating the water (or at least to show that the difference between mg-1 Pollen and mg-1 solute is neglectable).

INMs have not only be found on pollen but also on leaf, bark, stem and branches (see Felgitsch 2018).

References

Burkart, J., Gratzl, J., Seifried, T., Bieber, P., and Grothe, H.: Subpollen particles (SPP) of birch as carriers of ice nucleating macromolecules, Biogeosciences Discuss., 1–15, 2021.

Dreischmeier, K., Budke, C., Wiehemeier, L., Kottke, T., and Koop, T.: Boreal pollen contain ice-nucleating as well as ice binding “antifreeze” polysaccharides, Sci. Rep., 7, 1–13, https://doi.org/10.1038/srep41890, 2017.

Felgitsch, L., Baloh, P., Burkart, J., Mayr, M., Momken, M. E., Seifried, T. M., Winkler, P., Schmale, D. G., and Grothe, H.: Birch leaves and branches as a source of ice-nucleating macromolecules, Atmos. Chem. Phys., 18, 16063–16079, https://doi.org/10.5194/ACP-18-16063-2018, 2018.

Gute, E. and Abbatt, J. P. D.: Ice nucleating behavior of different tree pollen in the immersion mode, Atmos. Environ., 231, https://doi.org/10.1016/j.atmosenv.2020.117488, 2020.

Pummer, B. G., Bauer, H., Bernardi, J., Bleicher, S., and Grothe, H.: Suspendable macromolecules are responsible for ice nucleation activity of birch and conifer pollen, Atmos. Chem. Phys., 12, 2541–2550, https://doi.org/10.5194/acp-12-2541 2012, 2012.

Pummer, B. G., Bauer, H., Bernardi, J., Chazallon, B., Facq, S., Lendl, B., Whitmore, K., and Grothe, H.: Chemistry and morphology of dried-up pollen suspension residues, J. Raman Spectrosc., 44, 1654–1658, https://doi.org/10.1002/jrs.4395, 2013.