Zhang et al. present a comprehensive analysis of long-term in-situ air ion measurements in an urban environment, with comparisons to data from a boreal forest site. The study thoroughly investigates seasonal and diurnal trends and includes calculations of particle formation and growth rates. The results are consistent with existing literature and reflect the expected behavior of air ions in these contrasting environments. While the work is detailed and well-executed, it primarily reinforces previously established knowledge. One of the key findings is the observed decline in ion concentrations at the SORPES station compared to earlier measurements; however, the authors note that this decrease may be influenced by differences in instrumentation. Overall, the manuscript is best considered a detailed measurement report rather than a source of novel conceptual insights. I recommend the manuscript for acceptance after the authors address the following minor comments:

•    Page 1 line 22:

The reader might not directly understand what is meant by the “electrode effect”. Consider replacing by "Earth’s electrode effect" or "fair-weather electric field near the ground" 

•    Page 1 line 28: 

The authors report here ion formation rates at 3 nm which are otherwise not shown or referred to throughout the manuscript. Why did the authors chose to report ion formation rates at 2 nm and total formation rates at 2.5 nm in the main text? Why they did not report the ion and particle formation rates at the same size to make the calculation of the ion-induced fraction more reasonable?

•    Page 1 line 25:

The sentence needs rephrasing. Consider:  “likely due to the efficient scavenging of cluster ions by the high number density of neutral aerosol particles prevalent in a megacity environment”

•    Page 1 line 27:

Decisive factor for what? Be clearer: for reducing the cluster ion concentrations.

•    Page 2 line 43:

‘their role’ instead of ‘their roles’

•    Page 2 line 55-58:

Some references cited here pertain to ion-mediated nucleation, while others pertain to ion-induced nucleation. It is important to recognize that these two terms describe different pathways

•    Page 3 line 86:

Why the Wang et al. 2020, reference has the author’s first name?

•    Page 4 Figure 1

The authors should include credits or copyright statement for the map.

•    Page 4 line 108:

SMEAR II is situated

•    Page 5 line 133:

The NAIS data inversion is performed using the Spectops software and is not described by Wagner et al. The authors appear to conflate data inversion with the correction introduced by Wagner et al. Specifically, the Wagner et al. paper does not present inversion kernels; instead, it proposes a correction function to account for the detection efficiency of ions by the NAIS. Therefore, it cannot be claimed by the authors that the NAIS inversion kernel is calibrated based on Wagner et al., 2016 method.

•    Page 5 line 134:

This statement is also incorrect. The inversion performed by the Spectops software already accounts for internal losses within the NAIS. Corrections for diffusion losses in the sampling tube are typically applied after the data inversion.

•    Page 5 line 136:

The authors say here that the NAIS was unavailable from Sep 14 to October 15th. It is important to include this information in the captions of figure 3 and 4 as well. 

•    Page 5 line 139:

Are the authors certain that a twin DMPS system is used at the SORPES station? A measurement range of 6 to 800 nm seems unusual for a typical twin DMPS setup. This range is more typical for the Helsinki custom-made flow switching DMA.

•    Page 7 line 187:

Reference here should be to equations 3 and 4, not 4 and 5

•    Page 7 line 208-209:  “As cluster ions are continually produced in the atmosphere and have shorter lifetimes compared to larger ions, they have a higher probability of attaching to neutral particles”

The sentence needs to be rephrased. The short lifetime of the small ions is due to their higher coagulation probability with bigger neutral particles and not vice versa.

•    Page 8 line 211-212:

 Please also include the recent study published in PNAS on braking as a source of highly charged aerosol particles (https://doi.org/10.1073/pnas.2313897121) to the list of references about ions and traffic emissions.

•    Page 8 line 219-220: “This phenomenon could be caused by a fraction of very small negative ions being cut out due to the lower limit of the NAIS (Figure 2a-b)”.

If this is the case, then something similar should also be seen at the SMEAR II station. The distribution of negative ions there—although having a slightly higher peak diameter—is also not completely measured by the NAIS. 

•    Page 8 lines 223-225: “Additionally, the elevated positive cluster ion concentration at SORPES may indicate that SORPES hosts more air ions with the compounds containing the highest proton affinities, allowing them to capture positive charges.”

As this is the most likely explanation for the difference between the two sites, please consider presenting this argument first.

•    Page 8 lines 227-228: “As the mean size increases with air temperature, fewer negative ions fall below 0.8 nm, leading to a reduction in the concentration difference.”

This hypothesis can be easily tested by plotting the size distribution at different temperatures. Please consider adding such a plot to the supplement.

•    Pages 9-10 lines 246-247: “A possible explanation could be the elimination of large radiation sources near SORPES, particularly those from industrial activity”.

Please give a reference to this statement or explain more.

•    Page 12 line 308: Rephrase sentence. The concentration of cluster ions would be zero if the sink is higher than the production

•    Page 14 figure 5 and associated discussion

While the CS calculation method is consistent across both sites, the particle size distribution measured at SORPES covers a narrower size range compared to that at SMEAR. It is important that the same size range is considered in the CS calculation at both sites to ensure comparability. That said, the SORPES site is likely influenced by larger particles outside the detection range of standard DMPS systems, which could lead to an underestimation of the CS at this location. The authors are encouraged to include measurements of bigger particles from instruments such as OPCs or APS, in the calculation of CS if these are available.

•    Page 17 Lines 414- 416: “The peak in large ion concentration at SORPES (13:00 LT) appeared earlier than at SMEAR II (18:00 LT), which was partly caused by a higher growth rate of newly formed particles at SORPES”.

It is also worth mentioning here that the increase in intermediate ions also started earlier

•    Page 21 lines 485-493:

please cite here the important work of Gonser et. al (2014) providing a mechanistic explanation for the time difference between ion and total formation rates (https://doi.org/10.5194/acp-14-10547-2014)

•    Page 22 Line 504: “with only a slight increase when ranking values were higher than 80% (Fig.11d)”

The reference here should be to figure 11e.

•    Page 22 Lines 505-506: “at SORPES, ion formation rates showed little increase with rising ranking values, but particle formation rates increased by orders of magnitude (Fig.11b and 11d).”

The text is correct here, but the figures seem to have the wrong y-labels. Figure 11b should be the ion formation rate, while figure 11d should correspond to the particle formation rate as evidenced by the higher formation rates. 

•    Reference list: 

• Hirsikko, Yli-Juuti et al., 2007 : included twice in the reference list.
• Harrison & Carslaw, 2003: reference is not complete, needs page numbers or document number
•  Laakso et al., 2002: reference page number or document number is incorrect
• Iida et al., 2006: document number is not correct: J. Geophys. Res., 111, D23201
• Gagne et al., 2011: missing the journal information
• Leino et al., 2016: missing the journal information
• Hari et al., 2013: this is not the proper way to cite a chapter in a book
• Wagner et al., 2026: reference needs journal volume and page number
• Aliaga et al., 2023: must reference the final articles and not the discussion
• Komppula et., 2007: missing page numbers
• Mirme et., 2007: missing page numbers
• Hoppel 1986: could be a wrong reference as the mentioned publication have two other authors than Hoppel, R. V. Anderson, and John C. Willett. The right reference is  Hoppel, W. A. (1967). Theory of the electrode effect, J. Atmos. Terrest. Phys. 29 , 709
• Nieminen et al., 204: publication details missing
• Buenrostro et al., 2016: publication details missing

This study reported differential characterization of air ions in the boreal forest of Finland and the megacity of eastern China. The authors presented some interesting results. The manuscript is well-written. After the following questions are considered, it would be worthy of publication in ACP.:

1. Air temperature was identified as a crucial factor influencing ion polarity. Thus, the temperature of the instruments themselves might affect ion polarity. Consequently, this could introduce bias at the two sites. How can these biases be avoided?
2. What are the reasons that ion-induced nucleation could precede neutral nucleation in this polluted environment?
3. Lines 445–455: During new particle formation (NPF) events, both charged and neutral particles are present in the same atmospheric environment. What is the difference in their growth processes?
4. Lines 475–480: As PM2.5 concentrations decreased in China, the atmospheric oxidation capacity increased. This could lead to an increase in H2SO4 concentrations. Particulate sulfate primarily forms in the particle phase. A reduction in PM2.5 does not necessarily indicate a decrease in gas-phase H2SO4 concentrations.