The MS is based on a really neat idea, and having a not too sophisticated instrument to detect / trace ion formation and its role in new particle formation events at a local level would be an important step forward. The MS therefore is certainly within the scope of Aerosol Research and could be a highly valuable contribution to the field. There are some issues with the MS itself, however, that should be fixed before it can be accepted for publication.

The MS presents data obtained with a novel instrument, but as this instrument (the Cluster Ion Counter, CIC) is not described in the MS, it is impossible to put the results into context. The reference describing the CIC (Mirme et al., 2024) is given as “to be submitted”, and this is definitely insufficient. Of course a “modified CIC” is used for this study, but without info on the original, the modifications, which are described in the MS, cannot be adequately appreciated. I therefore strongly suggest the authors add a dedicated section on the CIC, its operation and expected improvements of the modified CIC to the current MS.

The efficiency curves shown in Figure 1 should be explained in more detail. Channel 1 can be used to estimate the total ion concentration in the whole size range. But as the efficiency of CH 1 is way higher than the efficiency of CH 2 and CH 3 even at the sizes where their respective efficiencies peak, and there never is mention of subtracting counts from CH 2 and CH 3 from the CH 1 counts, the values given for “small ions” (i.e. < 2nm) seem to be much overestimated – or was there some extra data processing not mentioned in the MS? Clarification of this issue is definitely needed.

Lines 210 - 211: “… overall agreement between these two instruments is very good….” Looking at the figures, the good correlation coefficient does not really suggest a “very good” agreement. The correlation coefficients are ok, but the data deviate strongly from the 1:1 line with the concentrations in CIC CH 1 higher than those obtained by the NAIS (Figure 2). In both plots shown in Figure 3 a line “drawn by eye” also shows quite a large deviation from the 1:1 line. This issue is exacerbated by the fact that all these plots are log-log plots, so it should be discussed. The huge discrepancy between NAIS and CIC data in Figure 5, lower panel, should also be discussed in more detail.

In view of the lacking info about the CIC, the issue with the efficiency curves and the vast scatter of data in Figures 2 – 5, the conclusion statement in lines 286 – 288 seems to be over-optimistic (“are we able to utilise a simple ion counter to find out LIIF in a proper way. According to our results presented above, the answer is: yes”). The statement at the end of the conclusion section, however, can only be underlined: “if we want to investigate aerosol formation and growth rates for the nucleation mode (3 – 25 nm) ….. NAIS measurements are needed”

Other points:                                                      

The structure of the MS could be improved – it does not make sense mentioning Figures 6 – 9 first and discuss Figures 2 ff afterwards. The section on observation data might be shifted to a later position in the MS

Lines 200 – 205: no mention is made on the influence of the efficiency of the CIC Channel 1 – discuss the effect

Line 243 “NAIS are 0.237, 258 and 0.266 times ….” Missing “0.” In front of “258”….

Lines 273 – 275: discuss reason for the vastly different values for the formation rates of 2 nm particles at the different measurement locations

Statements lacking references:

line 46 “much less information is available” – unless the authors mean “no info”, references should be added here

line 130: basis for approximation “d_p,i … equal …. 1.2 nm for negative cluster ions….”

Review of “On the potential of Cluster Ion Counter (CIC) to observe local new particle formation, condensation sink and growth rate of newly formed particles” by Markku Kulmala, Santeri Tuovinen, Sander Mirme, Paap Koemets, Lauri Ahonen, Yongchun Liu, Heikki Junninen, Tuukka Petäjä and Veli-Matti Kerminen for consideration for publication in Aerosol Research

General Comments

In this manuscript, the performance of a modified Cluster Ion Counter (CIC) to accurately measure the small ion and intermediate ion concentration is evaluated. The Authors use a Neutral cluster and Air Ion Spectrometer (NAIS) as a benchmark instrument to validate the CIC, for a period of several weeks of measurements, in the winter and spring of 2024. The Authors also develop a set of equations that allows the derivation from CIC measurement, of important new particle formation (NPF) characteristics such as the condensation sink, the formation rate of intermediate ions, and their growth rate. They then go on to apply these equations on an example day with observed NPF: February 13^th, 2024.

This study fits well within the scope of Aerosol Research. The use of the CIC to characterize localized and regional NPF may well be important for NPF research, but the Authors fail to clearly articulate how, why or in which cases the CIC measurements are advantageous over NAIS measurements, and build a convincing narrative to support their conclusions. Some linkages between the results presented and their consequences are left unsaid, leaving the reader to fill in the blanks. Some sections and statements seem out-of-place and distract from the scientific narrative the reader is after. The different parts of the manuscript are often not logically connected and different vocabulary is used in different sections. While I do believe this research is deserving of eventual publication, it is also deserving of a better presentation and introduction to the Aerosol Research readership.

Specific comments

Abstract

The abstract should include a statement as to how this research is contributing to advancing the field. What can we do with these tools that couldn’t be done before? What is this research enabling?

1. Introduction

Last paragraph (lines 68-80): While the previous paragraph closes stating that the NAIS is essentially too good an instrument for the purposes, this last paragraph of the introduction does not state how or in which circumstance the CIC is a better choice than the NAIS. Is it because it is cheaper? Is it because less maintenance is required?

Lines 74 and 75 (as well as line 28 in the abstract): the Authors mention that they will derive an equation to estimate the “intensity” of local NPF or intensity of local new particle formation or local intermediate ion formation, LIIF. This is the last time the word “intensity” appears in the manuscript. Similarly, LIIF is not discussed anywhere in the core of the manuscript, until it comes back in the Conclusions section. It is not clear which equation represents the “intensity” of NPF or LIIF. Assumedly, the Authors mean J₂. Please harmonize the words used to describe different quantities throughout the manuscript.

2.1 Cluster Ion Counter (CIC)

Lines 91-98: In this section, we learn that the Authors are using a modified CIC, where the analyzer has been modified. First, please define what an analyzer is, or show it in a schematic of the modified CIC. Second, please explain how the analyzer was modified. If it is described elsewhere, please reference to this description, for example, at the end of the first sentence, on line 95.

Figure 1: Please name the x-axis “mobility equivalent diameter”, or specify which kind of diameter is being measured. Also consider making the legend more explicit by identifying which ion sizes each of these lines represent (except for channel 2).

Figure 1: In the main text, please explain the criteria for deciding the channels’ boundary diameters.

2.2 Conceptual Model

Please introduce what this is with a sentence or two at the beginning of the section. Also “conceptual model” is a rather vague name for a specific model or set of equations. Consider giving it a more specific name and adjusting the references in the text accordingly.

Line 111: please correct the units for the ion-ion recombination rate, α.

Equation 4: improve the presentation of this equation, as possible, so that it is easier to read.

Lines 129-131: How were these d_p,1 values chosen? Based on what curves, is it somewhere in the literature?

Equations 5a and 5b: indicate which is for the CIC and which is for the NAIS next to the equation rather than using parentheses.

Line 156: Perhaps it would be worth discussing the validity of this assumption, or referring to a previous discussion of this assumption.

2.3 Observations

This entire paragraph appears out of place and references to Figures 6, 8 and 9 are anachronic. I think the Authors are trying to describe the data available for each measurement station. If the Authors want to specify which data was used for each Figure, this could be presented in an orderly manner in a Table among Supplementary Materials.

With the sentences about Figs 6, 8, and 9 out of the way, this paragraph is still hard to follow and would benefit from being reorganized in a more logical manner.  

3.1 Instrument Comparison

It is clear that the NAIS is considered the instrument of reference here, with good reasons. However, the Authors should point to justificative materials in the literature, such as an evaluation of the NAIS’ accuracy, or it’s predominance in the concerned Research Communities.

Table 1 & Table 2 captions: “Statistics”, please replace by which statistics, assumedly “Percentiles”?

Table 1: “the negative concentrations […] are indicative of a noisy signal of the instrument” The implications of this should be discussed in the main text as well as the transformations applied (noting that there are no negative numbers in Fig. 3). Based on Table 1, it looks like a significant proportion of the concentration data is negative so how these data were handled and why are important. Plotting these numbers (including the negative ones) against the Virtual CIC channel 2-3 might be useful in telling where the CIC channel 2-3 becomes unreliable. 

Table 2 caption (lines 352-355): This explanation of the virtual channel, which in my opinion is name awkwardly, needs to be presented in the main text, named, and its name be used consistently across the manuscript. The description of how it was calculated does not belong in the caption. See my related comments under Figure 3.

Line 207: I think the word “detailed” can be taken out, or replaced by a more appropriate qualifier. Figures 2 and 3 rather appear to be bulk comparisons.

Line 209: Even though “larger” here refers to the concentration, the Authors will understand that the phrase “somewhat larger small ions” is not ideal. To be consistent with the second part of the sentence, which speaks of “lower concentrations”, I suggest the Authors use the word “higher small ions […] concentration”.

Figure captions in general: there are a lot of explanations in the captions that are being repeated in many captions. That information should be moved to the main text and does not need to be included in the captions. For example, once the measurement period has been defined in section 2.3, there is no need to point to the dates where measurement happened and did not happen at every figure caption. This information is unnecessarily repeated in Figs. 2, 3 and 4.

Figures 2 and 3: I would like to see the density of these points. A way to show that would be to box the points and show boxplots, or fit an unconstrained linear fit, or add transparency to the points. All we see at the moment is a mostly fully coloured area.

Lines 207-212: Nowhere in this paragraph is the fact that Figure 3 has two panes discussed. Nowhere is the difference between those two panes explained. This needs to be done in the main text, around this paragraph.

Figure 3: What is the difference between the calculation of NAIS Ch 2-3 and NAIS 2-2.3 nm? Which one of those two is equivalent to the “Virtual CIC channel 2-3” presented in Table 2? Since the CIC channel 2-3 is considered to cover the 2-2.3 nm range, how are these two concentrations different? This is not clear from the caption and these calculations should anyway be described in the main text. I also see that the upper pane is supposed to be covering a “wider size range of ions”. What is that range exactly?

Figure 3 caption (lines 392-396): This describes a calibration of the NAIS data, using the CIC as a reference. Firstly, this procedure should be described and defined in the main text (and be removed from the caption). Secondly, it is highly unusual to calibrate the reference instrument with the instrument under test. Unless there is a very good reason to present it this way, I strongly recommend flipping the roles so that the CIC concentration is calibrated to the NAIS concentration. This being said, the reasons behind this unusual manoeuvre may become clearer once the difference between the upper and the lower pane is clarified.

Figure 4 caption: same comments as for Figs. 2 and 3. Trim down the caption and have these descriptions of calculations once in the main text. Use their names consistently throughout the manuscript. Same confusion as to what the difference between NAIS Ch 2-3 and NAIS 2-2.3 nm is.

Figure 5: The legends should be inside the figure or the proportions be changed: they take way too much space at the moment.

Figure 5 caption: Much of the caption is repeated information. Please clean it up.

Figure 5 implications: In the text, please discuss why the CIC agrees better with NAIS Ch 2-3 at high concentration but better with NAIS 2-2.3 nm at lower concentrations. Was this behaviour expected?

Line 216: Why was March 10^th, 2024 selected? How does it compare to other days?

3.2 Application of CIC measurement in investigating condensation sink and local NPF

Figure 6 caption (line 437): Small ions are defined as I in the text, not N. I assume it should be I. Also, the last 3 lines should be removed as it is already explained in Section 2.3.

Line 225: Is this Eq. 5a or 5b? Which instrument are we assuming? Section 2.3 states that the CS was determined using a PSD system.

Line 243: Those numbers provided have 3 digits after the decimal. Are they all significant digits? Also, since the NAIS is the reference instruments, consider flipping the ratio to have the CIC CS be around 4 times those measured by the NAIS. Also, this, and Tables 1 and 2, sort of contradict your statement on lines 232-233, that the CIC has a higher detection efficiency, unless the NAIS data is inverted and the CIC data isn’t. How do you explain the difference? In section 2.3, the Authors state that a DMPS and PSD system were used to determine the “CS data”. Where is that shown or used? Which CS is closer to the truth? How applicable is this ratio between CIC-CS and NAIS-CS across days and environments? I think there should be a more in-depth discussion of how the CS differs for CIC and NAIS measurements. After all, it is one of the main products of this manuscript.

Figure 7 caption (lines 459-465): This discussion should be in the main text. Furthermore, the limitations of this technique to estimate the CS should be discussed.

Figure 7: The vertical lines should be thicker to distinguish them from spikes in the time series. The second vertical line should also be orange to match its associated time series.

Figure 7: Please add a subplot or a figure in the Supplementary Material where the appearance time method is used to determine the GR from NAIS data on the same day. Compare the GR obtained with the CIC to the one obtained with the NAIS.

Line 252: 2.1 nm is used here without explaining what it represents. Is 2.1 nm considered to be the peak of the Ch2-Ch3 detection efficiency curve? Eyeballing it, I would use a different number myself, hence the importance of being clear about where it comes from. A 0.1 nm error on a range of 0.8 nm could make a difference in the GR it gives.

Lines 250 and around: please refer to Eq. 10, or if it is too simple, remove it from Section 2.2.

Lines 259-260: This statement suggests that it is only possible to determine the GR under certain circumstances. What are these limitations, is it GR-related? Is it determined by the NPF’s intensity? What is the percentage of days where the GR is retrievable? Have you tried it for other days?

Lines 262-275: It is not clear whether any measurement data is used to generate the J2. Looks like they are look-up figures where a potential user would determine the 2.0-2.3nm concentration using the CIC and then plug that in the equation or look at Figure 8 or 9 to find the corresponding J2. Is that correct? Either way, this paragraph needs to be clearer.

Lines 273-275: while true and important, this statement appears unrelated to the rest of the contents in the paragraph. A simple sentence modification would probably explain why the is mentioned here. For example, “If one wanted to estimate the total formation rate…”, assuming this is the reason this sentence is here for.

Conclusions and summary

In general, I don’t feel that the conclusions are supported by the material in the manuscript. The conclusions feel detached from the manuscript, although connected to the introduction (only).

Line 287-288: “find out LIIF in a proper way”. I’m not sure what that means. Please rephrase. Also, since we haven’t seen any mention of LIIF since the introduction, linking LIIF to the contents of the manuscript would be essential.

Line 290: “a somewhat modified version of the CIC” Take out “somewhat” and add the reference to line 94. That said, Mirme et al., 2024 doesn’t seem to be available yet. Some short explanation of what this modification entails would be appropriate in Section 2.1.

Line 292: What is LIIF? What are its units? Where in the manuscript was it estimated?

Line 294: “ion sinks” is not mentioned anywhere in the manuscript. Did you mean Coagulation sinks or ion-ion recombination, or the sum of both? Avoid introducing new vocabulary and concept in the conclusions section.

Line 296-298: Is the portability of these equations and measurement principles to positive ions theoretical or have the Authors actually measured and evaluated it? The Authors suggest in the last sentence of this paragraph, that it might not actually work for positive ions. I suggest toning down the claims that the same is valid for positive ions.

Line 300: Please, make clear that the “estimated” values are based on CIC measurements. Also, I have seen ion concentrations compared, but I have not seen a direct comparison of CS between the CIC and any other instrument.

Line 302: “ion sinks” is mentioned again. It is not defined. Consider defining the term around Line 111 or mention its components instead.

Line 301: I have not seen evidence that CIC measurements were used in Beijing. Please clarify.

Line 305: I’m still not sure what LIIF is and how the CIC is effective to observe it. Which Figures show this?

Typos, etc.

Throughout the manuscript: Consider using 2.0-2.3 nm instead of 2-2.3 nm so both boundaries have the same assumed accuracy.

Table 1: Please fix the alignment of the titles so they align with their respective columns. Also add ‘small ions’ and ‘2-2.3 nm’ under the current titles for clarity.

Line 215: no “s” to ion.

Line 217: “pretty well”: that is a rather vague qualifier. Consider editing.

Lines 218-220: “captured consistently”? Please rephrase this. Probable meaning: peaks coincide?

Line 234: dependent on particle size (remove “a”)

Line 243: assuming 256 was meant to be 0.256?

Line 450: Suggested: Determining the ion growth rate (GR) based on CIC measurements.

Line 455: “similar to appearance time method” Is it similar to or is it the appearance method? Either remove “similar to” or describe how this method differs from the Lehtipalo et al. appearance time method.

Line 457: consider using 20% instead of 0.2.

Line 458: consider replacing “gained” with calculated or identified.

Line 253: “can be considered a very realistic one” à do you mean that the value fits well within the expected range?

Figures 8 & 9: consider merging these figures into one figure with two subplots.

Figure 8, y-axis: Formation rate “of negative ions” at 2 nm.

Line 300: The name of the internationally famous Hyytiälä station is misspelled!

Line 304: “pretty well” is pretty vague. Please quantify how well they agree.

This is an interesting analysis of the performance of the cluster ion counter (CIC). Overall, the paper is well written, and the results generally support the overall view that this instrument, while simple, can give some insights into important aspects relevant to atmospheric nucleation. I recommend acceptance but I would like the authors to consider the following comments, which I feel are needed to improve the paper:

Line 45: I am not sure how "averaged" applies to regional npf ... certainly we can average anything over large spatial scales. I recommend replacing with "that takes place"

Line 65: I think the reader would like to know what, specifically, is the unnecessary information that NAIS provides.

Line 127: Does the fact that ions are charged have any impact on their condensation sink rate? It appears that only the physical size of ions is considered here, but in other applications (such as ion-neutral reactions) charge does matter. Please discuss.

Lines 302 and 304: The phrases "relatively accurate" and "pretty well" are not helpful. Please be more specific based on your results.

Minor comment:

Line 288: If this is a question, the last phrase should end with a question mark. The rule: "If each point is a complete sentence, capitalize the first word and end the sentence with appropriate ending punctuation."