the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 May 2026
Challenges in measuring sticky biogenic ice-nucleating macromolecules
Abstract. Ice-nucleating particles (INPs) are aerosol particles that influence mixed-phase clouds and in doing so impact weather and climate worldwide. To improve our understanding of ice production in mixed-phase clouds, we need techniques capable of accurately measuring atmospheric INP concentration spectra. However, there are sometimes discrepancies between the techniques commonly used to measure ambient INP concentrations, particularly when used in environments with abundant biogenic ice-nucleating material. Proteins and macromolecules adsorb to surfaces, such as filters, but the impact of this interaction on INP measurements is unknown. Here, we compare a widely used technique that involves washing collected aerosol particles off polycarbonate filters into aqueous suspension for subsequent INP droplet freezing assay (wash-off), with a technique where droplets are placed directly atop PTFE filters on a cold stage (drop-on) and also an online technique using an expansion chamber. Our results show that the wash-off technique underestimates the INP activity when free ice-nucleating proteins are present due to the poor recovery of the proteins from the filter into the wash-off suspension. However, there is much better agreement between techniques for INPs associated with coarse-mode mineral dust particles or cell fragments and for polysaccharide INPs from pollen. These findings indicate that some field-based INP measurements that use a wash-off technique may produce atmospheric INP concentrations that are biased low, particularly in regions with abundant proteinaceous INPs.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Aerosol Research.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(1752 KB) - Metadata XML
-
Supplement
(1408 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on ar-2026-17', Anonymous Referee #1, 22 Jun 2026
-
RC2: 'Comment on ar-2026-17', Anonymous Referee #2, 09 Jul 2026
This manuscript deals with an important methodological question in INP measurements: whether the commonly used filter wash-off technique may underestimate INP activity when proteinaceous or other sticky biogenic macromolecules are present. The comparison between wash-off, drop-on and some PINE measurements is valuable, and the main conclusion is interesting and relevant for the INP community.
Overall, the manuscript presents a valuable contribution and I think the main conclusion is plausible. I recommend publication after moderate revision. My main comments are mostly related to metrological aspects of the experiment and data treatment.
- For the wet samples, a medical nebuliser was used and after evaporation, dry aerosol particles remained in the chamber. However, it is not completely clear how the size distribution, concentration and composition of these particles were controlled during the experiments. Were the aerosol PSD and number concentration actively controlled, or only measured afterwards with APS/SMPS and used for normalisation? It would be useful to show or discuss the temporal stability of the aerosol during filter sampling and PINE measurements, including possible changes due to nebuliser output, reservoir depletion, wall losses or particle settling.
- For the dry Snomax experiments, the authors mention that electrostatic charge may have affected the size distributions. From a metrological point of view, it would be good to discuss if using an aerosol neutralizer, for example Kr-85 or X-ray, before filter sampling or PINE measurements could improve reproducibility in future experiments.
- The use of active site density, ns(T), is appropriate, but it depends directly on the aerosol surface area concentration. Therefore, the APS/SMPS merging and the assumptions used to convert aerodynamic and mobility diameters into volume-equivalent diameters are very important. The authors should better explain the choice of density and dynamic shape factor. In the Supplement, some values seem to be selected to obtain consistency between APS and SMPS data but the effect of this choice is not really quantified. In particular, assuming χ = 1 for all filtrates may be a simplification. A short sensitivity analysis showing how ns(T) would change for χ = 1.1 or 1.2 would make the results more robust.
- Equation 7 also needs clarification. If the distribution is dS/dlogDp, then the total surface area should be integrated over logDp, not directly over Dp. Maybe the calculation was done correctly in logarithmic bins, but the equation as written is confusing.
- The comparison between wash-off and drop-on also includes different filter materials, different pore sizes, different droplet geometry and different sample processing. I suggest adding a short paragraph acknowledging these metrological limitations. In addition, the two filter samplers were placed at different positions in the chamber, whichwas mixed with a fan, but it is not clear if the two sampling positions were tested to be equivalent. Were the filter types alternated between positions, or were duplicate samplers with the same filter tested? If not, this should be mentioned as a possible limitation.
- The manuscript states that, when the background-subtracted k(T) falls within the background, the sample value is set to zero. I think this treatment may bias cumulative spectra, especially for samples close to the blank. It would be clearer to report detection limits, upper limits or confidence intervals instead of forcing these values to zero. The authors should also clarify how many blanks were used for each technique, whether they were distributed across different experimental days, and how the blank uncertainty was propagated into the cumulative INP spectra.
Typos and grammar:
- Page 12, Sec 2.4: The sentence “To establish ….” appears incomplete or incorrectly joined. Please revise.
- Line 219: "Eq 4 to calculate the INP concentration for the wash of technique" should be "wash-off".
- Page 13, Sec 3.1.1: “contains derived from Pseudomonas syringae” I guess is meant “contains material derived from Pseudomonas syringae”.
- Page 15, Sec 4.1: “dry-disperse samples” should be “dry-dispersed samples”.
- Figure 6 caption: “activty” should be “activity”.
- Figure 7 title and caption: “Sea Surace Microlayer” should be “Sea Surface Microlayer”.
- Page 18, Sec 4.4: “particularly above 22 °C” perhaps is “above -22 °C”
- Ensure consistency in temperature degree symbols (e.g., "-10 °C" vs "-10 ◦C" used in different sections).
- Section 1: “Pue de Dôme” should be “Puy de Dôme”.
- line 245: Use “GmbH” rather than “Gmbh”.
- Use either “airflow” or “air flow” consistently.
- Use “litres per minute” or “liters per minute” consistently.
- Units in the supplementary tables show sampling volume as “L−1”. This should be “L” if the column is sampled volume.
- In the supplement, it is referred to experiments from “Table S2 to Table S8” but the dry Snomax table appears as Table S1. Is it left out for some reason?
Citation: https://doi.org/10.5194/ar-2026-17-RC2 -
AC1: 'Comment on ar-2026-17', Joseph Robinson, 23 Jul 2026
The comment was uploaded in the form of a supplement: https://ar.copernicus.org/preprints/ar-2026-17/ar-2026-17-AC1-supplement.pdf
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 239 | 88 | 13 | 340 | 49 | 15 | 14 |
- HTML: 239
- PDF: 88
- XML: 13
- Total: 340
- Supplement: 49
- BibTeX: 15
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Comments on the manuscript of "Challenges in measuring sticky biogenic ice-nucleating macromolecules" by J. Robinson et al.
General comments
The manuscript develops comparison between offline techniques commonly used to measure the ambient ice-nucleating particles (INP). Indeed, following a thorough review of the literature on the measurement of INP, the authors point out that there are sometimes large discrepancies between online and offline measurements, particularly when comparing the offline polycarbonate filter wash-off technique to others that require less sample processing.
The aim of this study is to conduct lab experiments under controlled conditions using well-defined types of aerosols, employing, in parallel, two off-line filter-based techniques and one on-line (ice nucleating chamber), in order to better characterize the situations already investigated during field campaigns and to evaluate the hypothesis that might explained the reported discrepancies.
The study is very well structured and was conducted in accordance with a rigorous scientific methodology. This experimental protocol was applied using a wide variety of samples (6 samples), mostly natural, with two possible methods of resuspension: dry-dispersed samples and wet-dispersed samples, in order to replicate realistic atmospheric conditions as closely as possible.
Overall, this study makes a significant contribution to the field of measuring INP concentrations, which remain a complex task due to biases associated with the sampling processes used which are clearly highlighted here. It offers recommendations on the use of the wash-off technique, which may underestimate INP levels, particularly when the air samples analysed are likely to contain biological material (notably proteins).
The manuscript is well written, the results are scientifically sound and clearly interpreted. The manuscript is generally suitable for publication in Aerosol Research after minor revisions listed below.
Specific comment.
The main questions concern the experimental methodology to calculate the active site density as a function of T, ns(T), and in particularly the role of the size distribution (SD). For a given type of aerosol, the characteristics of the SD, reported in the supplementary information, show some variability from one experiment to another. Can the authors comment on this situation? Did they have control over the variation in particle size and particle concentration or do these parameters vary more or less randomly ?
Could these variations in SD between experiments be the cause of the significant variations sometimes observed in the active site density measurements? Indeed, the nucleation process depends on particle size, yet this parameter is no longer taken into account in the calculation of ns(T) as this is expressed in terms of the total surface concentration of particles, which renders the size parameter irrelevant. Could the authors shed some light on this aspect?
Minor comments and technical corrections.
Section 2 Methods (page 4 to page 12) : In section 2.1, there is a subsection "2.1.1 Filter sampling" (line 163) Is it correct ? I suggest renumbering subsection '2.1.1' as section 2.2 and renumbering the following sections as 2.3, 2.4 ...
Line 199, page 7 and line 211, page 8 : If I am correct, units of k(T) should be cm-3 °C-1 and not cm-3 °C.
Line 226 : "ns(T) for each individual experiment plotted with its uncertainty can be found in SI Section S2".
- From my point of view, this sentence should be placed at the end of this section after introducing this ns(T) parameter.
- Can you specify how are calculated the incertainties on ns(T) (drop-on and wash-off techniques) showed in Figures SI section ? Do the uncertainties are coming from the µL-NIPI method?
Line 230: Eq(6) page 9, I suggest to specify also the unit of ns(T) , cm-2, even if it is evident and it can be also founded in Figure S3 of the Supplementary Informations.
Line 298 : "To establish". The sentence seems to be incompleted.
Line 321, Table 1. I propose listing the sample in the rows of Table1 in the order in which they appear in the main text: Dry Snomax, Copper River Dust, Agricultural.
Line 412, figure 6(a) : I suggest resizing the horizontal axis from -20 to -10°C in order to distinguish more clearly between the data relating to drop-on and wash-off techniques.