The authors present data from two sites with long-term observations of small ion number size distributions of both polarities, focusing on very small clusters of <1 nm to 2 nm in diameter. Complementary observations of sulfuric acid and organic molecular clusters are used to examine their impacts on the dynamics of cluster growth toward new particles. The two sites, Hyytiälä (remote forest) and Beijing (polluted urban), contrast in their environments and these contrasts are reflected in different dependencies of cluster dynamics on the measured precursors. These differences are illustrated with case study examples.

The data sets are interesting and unique, particularly in that they span multiple seasons and years. The introductory discussions of charged particles in the atmosphere and the charged general dynamic equation are well written and helpful in providing background for non-specialists. The observations are presented clearly, but the discussion and interpretation seemed less in-depth than suggested by the title “Insight into cluster formation and growth”. Relationships are pointed to, but it’s not entirely clear whether and how they shed new light into what is known already about cluster formation and growth, or how they might be applied to predict processes in other locations.

With the presentation and discussion of equation (2), I expected the data would be interpreted in that context, considering the different terms’ influence on the evolution of the cluster sizes and particularly how the two sites support different magnitudes of some of the rates. I was therefore surprised that the analysis did not consider coagulation scavenging (CoagS). The discussion on lines 95-109 seems to indicate that this rate plays a key role, competing with the growth rate (GR) in shaping the small ion size distribution. Can the authors explain and justify why this term was not included in the analyses (Lines 108-109)? In particular, I anticipated that Beijing, as an urban region with higher particulate and precursor concentrations, would have lower overall ion concentrations due to competition for the ions by the coagulation sink. Indeed, Figure 2 shows that Beijing had lower number concentrations - but a rationale for this observation, supported by the data, is not provided here. On lines 342-344, the authors note that “Due to the high concentrations of both low volatility vapors and large particles, the dynamics of small ions in a megacity such as Beijing are different than in a rural site such as Hyytiälä.” This, and other observations that note that the sites are “different”, seems reasonable; but as an expected result, that isn’t further explored with the dataset, it doesn’t rise to the level of “insight”.

On line 421, the authors note: “In Beijing, CoagS is crucial in determining whether the growing clusters will survive to larger sizes or not.” The subsequent section goes on to use new particle formation (NPF) ranking as a means of sorting for conditions supporting growth of cluster into larger sizes. If CoagS could not be determined explicitly and used as a sorting variable, but NPF ranking is a suitable proxy, this argument could be motivated and explained in the introduction. That would help explain more explicitly the reasons for the use of NPF ranking and would provide a clearer link to equation 2.

I suggest the authors also address two other points that might be helpful to make explicit. First, what do the ion polarities depend on? Are some molecules preferentially charged with one polarity over another and does composition play a role? The role of polarity in cluster dynamics and growth is not addressed extensively, so it wasn’t clear why data were separated into positive and negative polarities for analysis and what should be gleaned from that, versus presenting the distributions of all charged ions together and how these totals trend with, e.g., sulfuric acid concentrations. Would different diurnal and seasonal trends in dominant polarity be expected, why, and what impacts on cluster dynamics are expected?

Second, the role of transport and meteorology is not included explicitly in the discussion. The assumption seems to be that the observations are very localized (e.g., line 457) and can be interpreted as such; if this is the case, it would be helpful to explicitly state this and provide some justification.  For example, is this approach justified because the lifetimes of these ions are so short? Some of the observations do seem to suggest that impacts from other influences are seen in the data (e.g., line 491 mentions discontinuous growth in a Hyytiälä case study; Lines 539-540 suggest changes in ambient aerosol and meteorology affected the Beijing case study). If the behaviors are highly localized, what is their broader relevance?

Overall, the data sets are very interesting and report characteristics of an understudied class of atmospheric particles. I suggest the manuscript and its utility for the community would be enhanced if the authors provide some additional interpretation (perhaps shortening the discussion that just points out trends, or move some to the Appendix), addressing the comments above.

Minor comments:

Line 328: “compared to results for Hyytiälä, the differences in the [Beijing] size distributions are small for either polarity.” This refers to Figure 6a, compared with 5a or 3a or 3. This confused me, because  in Figure 6a the shift in the peak diameter between positive and negative polarities looks larger than those in the other two figures.

Line 81: should be “the changes IN”

Line 120: should be “from which data WERE”

Line 131: should be “WERE also used”

Line 171: should refer to TABLE 1 (not Table 2)

Line 197: should refer to TABLE 2 (not Table 3)

Line 290: note to author left in “add percentiles here”

Line 582: remains to be IDENTIFIED

General Comments

The manuscript tackles a relevant question in atmospheric science: the dynamics of sub-2 nm ions and their interactions with precursor vapors and new particle formation (NPF). The study is well-motivated and supported by high-quality measurements from two contrasting environments (Hyytiälä and Beijing). The combination of long-term statistics with detailed case studies adds significant value. However, the discussion falls short in some key areas—for example, the role of the coagulation sink and its implications for ion lifetime—and several technical and formatting issues affect clarity. Overall, the paper presents unique and valuable data and offers a clear view of ion behavior in different atmospheric settings, but the interpretation does not fully deliver the level of insight suggested by the title. Substantial revisions are needed to strengthen the analysis and improve coherence before publication.

Specific Comments

A major weakness concerns the treatment of the General Dynamic Equation (Eq. 2). Although correctly introduced, the analysis does not explore the influence of key terms such as the Coagulation Sink (CoagS), which is acknowledged as important in the text but explicitly excluded from the scope of this study. While this choice is understandable, a brief discussion of its potential impact on ion lifetime and survival—especially in high-PM environments like Beijing—would help contextualize the observed differences between sites. If feasible, even an approximate estimate could strengthen the interpretation, but this could also be addressed in future work.

The discussion on polarity differences (negative vs. positive ions) is too brief. Since the data are presented separately for each polarity, a more detailed mechanistic explanation of why polarity matters—e.g., whether precursor composition influences charge preference—would improve clarity.

Growth rates (GR) are reported for selected case studies, which is consistent with the stated scope. However, the manuscript would benefit from clarifying why a systematic analysis across the full dataset was not attempted. Providing uncertainty estimates for the reported GR values, or at least acknowledging their variability, would enhance transparency.

Additional context is missing: meteorological factors such as temperature and relative humidity, which strongly affect seasonal variability and processes like HOM formation, are not considered. While this omission is consistent with the study’s focus, a short statement acknowledging these influences would help readers interpret the results and understand limitations.

Two coherence issues require attention.

The discrepancy between weak correlations in Beijing (Section 3.3) and clear growth observed in case studies should be explicitly clarified.

Implications for atmospheric modelling and climate feedbacks, mentioned in the introduction, are not revisited in the conclusions. Even a brief statement on how these findings could inform models would improve the manuscript’s impact.

Technical Comments

Pag.2, line 57 “…. complex climate-biosphere feedbacks (Kulmala et al., 2020; Kulmala et al., Kulmala et al., 2024b).” Reference might be missing or incorrectly formatted.

Pag. 3, line 81 “The changing in the small ion number concentration can be described…” change in: “The change in the small ion number concentration can be described…”

Pag. 5, line 171

The text refers to Table 2 for Hyytiälä data, but the correct table is Table 1.

Example suggestion: Replace “Table 2” with “Table 1”.

Pag. 5, line 180 “…behave similarly as the negative ones…” change in: “…behave similarly to the negative ones…”

Pag. 6, line 197

The text refers to Table 3, which does not exist; the correct reference is Table 2.

Pag. 11, line 296 “…organic compounds tend drive cluster growth…”change to: “…organic compounds tend to drive cluster growth…”

Pag. 14, line 362 “…we can see from the small ion size distribution for both polarities how this growth of small ions up to 2.0 nm is seen…” change in: “…the small ion size distribution for both polarities shows this growth of small ions up to 2.0 nm…”

Pag. 19, line 465 “…over a order of magnitude higher…” change in: “…over an order of magnitude higher…”

Fig. 1

X-axis scales differ between panels for Hyytiälä and Beijing, making comparison difficult.

Example suggestion: Use a common X-axis scale for both panels.

Fig. 3, Fig. 5, Fig. 6, Fig. 7, Fig. 9, Fig. 10–12

The text refers to panels (a, b, c, d), but the figures do not display these labels.

Example suggestion: Add panel labels (a, b, etc.) to all multi-panel figures for consistency.

Fig. 5

Missing panel labels and inconsistent panel sizes: the bottom-right scatter plot is smaller than the top-right panel.

Example suggestion: Add panel labels and make panel sizes uniform.

All figures with time ranges (e.g., Fig. 3, Fig. 6, Fig. 9, Fig. 10–12)

Time formatting is inconsistent: “8–16:00”, “8:00–16:00”, “08:00–16:00”.

Example suggestion: Use a consistent format such as “08:00–16:00”.