the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Jun 2023
Measurement report: Ion clusters as indicator for localnew particle formation
Abstract. Atmospheric aerosol particles have considerable impact on climate, both directly by scattering radiation and indirectly by acting as cloud condensation nuclei. A major fraction of global aerosol number is formed in atmospheric new particle formation (NPF). These atmospheric particles consist of both neutral particles and charged atmospheric ions, and atmospheric ion number concentrations have been observed to indicate NPF. In this work, atmospheric ion concentrations were studied with the aim of finding the best suited size range of ions to characterize local NPF. Both negative and positive ion number size distributions measured by Neutral cluster and Air Ion Spectrometer (NAIS) at the SMEAR II measurement station in Hyytiälä, Finland were used. Ion sizes between 1.6 and 3 nm were considered. We found that the negative ions between 2.0 and 2.3 nm are well suited for representing NPF. In addition, the influence of transport on the observed 2.0–2.3 nm ion concentrations is considerably smaller than for larger ions. Therefore, we recommend the negative ions with diameters 2.0–2.3 nm as the best choice for characterization of local NPF.
- Preprint
(971 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-1108', Anonymous Referee #1, 04 Jul 2023
The manuscript by Tuovinen et al. presents analysis of Neutral cluster and Air Ion Spectrometer (NAIS) ion data taken at the SMEAR II measurement station in Hyytiälä between 2016 and 2020. From their analysis the authors claim that negative ions between 2.0 and 2.3 nm are best for the characterization of local new particle formation (NPF) events. However, there is a lack of in-depth analysis of the data that supports the claims, an actual applied characterization of a local NPF event is absent and, finally, the writing is missing details and the figure references are incorrect. Therefore, I could not recommend this manuscript for publication in Atmospheric Chemistry and Physics (ACP). Here are the main points that led to the decision.
The main analysis is based on the median, 25%, and 75% quantiles of the ion’s diurnal cycles (Sec. 3.1). Four different size bins between 1.6 and 3 nm of positive and negative ions were considered. I see several issues with this analysis. The authors just compare seasons where NPF events are observed more regularly (Mar. - May, Fig. 2) to seasons where they are less common (Sep. – Nov., Fig. 3). However, they wrote: “These data were used from all the available days, and no distinction was made based on whether the days had been classified as NPF days or not.” I think this is defeating the whole purpose of the analysis. It would be interesting if the ion signal differs during an NPF event independent of the seasons. There is no discussion/analysis of whether the ion signal is influenced by seasons in general, e.g., analyzing only non-NPF days for all the seasons.
For the actual analysis, the authors only compare peak heights of the median or 75% quantile for the different seasons to conclude that the negative ions in the size bin 2.16 nm are the best for representing NPF. First, the authors do not explain why they use the mainly the 75% quantile for their peak height analysis. Second, the authors did not discuss the spread between the 25% to 75% quantile, which represents the variability of the ion concentration. Third, most of the features of Fig. 2 (Mar. – May) are also visible in Fig. 1 (whole time period). There is no discussion of what this means for distinguishing a NPF signal from a non-NPF signal.
The argument about excluding the positive ions from further analysis is also very weak in my opinion. In most cases the positive ions have a stronger signal, which might be a plus in this size range below 3 nm, and the spread of the 25% to 75% quantile is most cases smaller. There might be some arguments for the exclusion, but I don’t think it is obvious only from the presented data.
To my understanding, the transport model of Sec. 3.2 consists only of very simple linear equations, which are not mentioned in the text but are easy to derive (distance = wind speed * time; particle size = GR * time + initial size; substitute time; the outcome matches with the presented data). However, the equations and a table would be more illustrative than plotting linear equations in a y-log plot. The initial conditions, taken from other publications, mainly define the outcome, e.g., the initial size favors the choices to the next larger size bin and excludes the first size bin. I think this model is too simplistic for a main finding (defining “local” NPF) in an ACP publication. The authors did not even try to validate the model by any other means.
In Sec. 3.3 (Impact of data amount on ion diurnal cycle) a lot of details are missing. The authors do not explain how they reduced the data to 50%, 10% and 1%. For the reduction, did they include all days whether they were NPF events or not? How does this affect the data quality? If the ion size bin should be a good indicator for local NPF events, why not show it on a day where there is local NPF event? Is the ion signal only a result of the statistical analysis or could it be really observed during one NPF event? These and many more unanswered questions arise. In addition, two “Figure Error: Reference source not found” show a certain carelessness, which could be noticed throughout the manuscript.
In general, I see the analysis and model as too simplistic to be published in ACP. I would encourage the authors to see their analysis as a starting point to explore the NAIS ion signal in more depth for (local) NPF events and attach this analysis as supporting information to the resulting publication.
Citation: https://doi.org/10.5194/egusphere-2023-1108-RC1 -
RC2: 'Comment on egusphere-2023-1108', Anonymous Referee #2, 09 Jul 2023
Tuovinen et al. utilized a Neutral cluster and Air Ion Spectrometer (NAIS) at the SMEAR II measurement station in Hyytiälä, Finland, to find the best ion size to represent the local new particle formation (NPF). They found that negative ions between 2.0 and 2.3 nm are the best range to represent local NPF. The study can help improve our knowledge of new particle formation. However, this study is not well designed. The analysis is not profound and lacks the necessary statistical analysis to support their conclusion. Moreover, the paper is poorly written, the missing details and figure numbers are incorrect, and some figures are lost. This stud is also not within Atmospheric Chemistry and Physics (ACP) scope. Thus, I suggest rejecting this paper. Please see my comment below to support my decision.
- In the Methods section, the authors did not include all the necessary details:
- It is not clear how did they restrict their analysis to atmospheric ions
- It is not clear why did you choose 1.6 and 3 nm. The authors should explain this in the Method section. They explained that ions larger than 3 nm might not be local ions. I do not understand this very well. Do you mean you do not have any ion larger than 3 nm generated at the local? How do you know that? How can you prove this?
- How do you define the background?
- You should add statistical analysis in this section and show results later.
- Did you use any chemical transport models for section 3.2? How do you calculate GR? Did you include chemical reactions? What is your gas precursors and what is the NPF mechanism? This part is not clear at all. You need to add a description of the model in the method section.
- Do you have any meteorological data? Moreover, do you have any gas phase measurements?
- Your results do not convince me.
- Why do you only show March, May, September, and November data?
- Your figure numbers are wrong, so I had difficulty referring to the figures. Also, you did not include the figures you discussed in section 3.3.
- Your figure 1-4 should be violin plots, not line plots. Violin plots can help audiences visualize the distribution and variation.
- I do not understand how you could use size to indicate new particle formation. I understand that the ion concentration threshold in a size bin should be a good indicator for local NPFm (e.g., NPF happens when the ion concentration is higher than a value).
Citation: https://doi.org/10.5194/egusphere-2023-1108-RC2 - In the Methods section, the authors did not include all the necessary details:
- AC1: 'Comment on egusphere-2023-1108', Santeri Tuovinen, 26 Sep 2023
Status: closed
-
RC1: 'Comment on egusphere-2023-1108', Anonymous Referee #1, 04 Jul 2023
The manuscript by Tuovinen et al. presents analysis of Neutral cluster and Air Ion Spectrometer (NAIS) ion data taken at the SMEAR II measurement station in Hyytiälä between 2016 and 2020. From their analysis the authors claim that negative ions between 2.0 and 2.3 nm are best for the characterization of local new particle formation (NPF) events. However, there is a lack of in-depth analysis of the data that supports the claims, an actual applied characterization of a local NPF event is absent and, finally, the writing is missing details and the figure references are incorrect. Therefore, I could not recommend this manuscript for publication in Atmospheric Chemistry and Physics (ACP). Here are the main points that led to the decision.
The main analysis is based on the median, 25%, and 75% quantiles of the ion’s diurnal cycles (Sec. 3.1). Four different size bins between 1.6 and 3 nm of positive and negative ions were considered. I see several issues with this analysis. The authors just compare seasons where NPF events are observed more regularly (Mar. - May, Fig. 2) to seasons where they are less common (Sep. – Nov., Fig. 3). However, they wrote: “These data were used from all the available days, and no distinction was made based on whether the days had been classified as NPF days or not.” I think this is defeating the whole purpose of the analysis. It would be interesting if the ion signal differs during an NPF event independent of the seasons. There is no discussion/analysis of whether the ion signal is influenced by seasons in general, e.g., analyzing only non-NPF days for all the seasons.
For the actual analysis, the authors only compare peak heights of the median or 75% quantile for the different seasons to conclude that the negative ions in the size bin 2.16 nm are the best for representing NPF. First, the authors do not explain why they use the mainly the 75% quantile for their peak height analysis. Second, the authors did not discuss the spread between the 25% to 75% quantile, which represents the variability of the ion concentration. Third, most of the features of Fig. 2 (Mar. – May) are also visible in Fig. 1 (whole time period). There is no discussion of what this means for distinguishing a NPF signal from a non-NPF signal.
The argument about excluding the positive ions from further analysis is also very weak in my opinion. In most cases the positive ions have a stronger signal, which might be a plus in this size range below 3 nm, and the spread of the 25% to 75% quantile is most cases smaller. There might be some arguments for the exclusion, but I don’t think it is obvious only from the presented data.
To my understanding, the transport model of Sec. 3.2 consists only of very simple linear equations, which are not mentioned in the text but are easy to derive (distance = wind speed * time; particle size = GR * time + initial size; substitute time; the outcome matches with the presented data). However, the equations and a table would be more illustrative than plotting linear equations in a y-log plot. The initial conditions, taken from other publications, mainly define the outcome, e.g., the initial size favors the choices to the next larger size bin and excludes the first size bin. I think this model is too simplistic for a main finding (defining “local” NPF) in an ACP publication. The authors did not even try to validate the model by any other means.
In Sec. 3.3 (Impact of data amount on ion diurnal cycle) a lot of details are missing. The authors do not explain how they reduced the data to 50%, 10% and 1%. For the reduction, did they include all days whether they were NPF events or not? How does this affect the data quality? If the ion size bin should be a good indicator for local NPF events, why not show it on a day where there is local NPF event? Is the ion signal only a result of the statistical analysis or could it be really observed during one NPF event? These and many more unanswered questions arise. In addition, two “Figure Error: Reference source not found” show a certain carelessness, which could be noticed throughout the manuscript.
In general, I see the analysis and model as too simplistic to be published in ACP. I would encourage the authors to see their analysis as a starting point to explore the NAIS ion signal in more depth for (local) NPF events and attach this analysis as supporting information to the resulting publication.
Citation: https://doi.org/10.5194/egusphere-2023-1108-RC1 -
RC2: 'Comment on egusphere-2023-1108', Anonymous Referee #2, 09 Jul 2023
Tuovinen et al. utilized a Neutral cluster and Air Ion Spectrometer (NAIS) at the SMEAR II measurement station in Hyytiälä, Finland, to find the best ion size to represent the local new particle formation (NPF). They found that negative ions between 2.0 and 2.3 nm are the best range to represent local NPF. The study can help improve our knowledge of new particle formation. However, this study is not well designed. The analysis is not profound and lacks the necessary statistical analysis to support their conclusion. Moreover, the paper is poorly written, the missing details and figure numbers are incorrect, and some figures are lost. This stud is also not within Atmospheric Chemistry and Physics (ACP) scope. Thus, I suggest rejecting this paper. Please see my comment below to support my decision.
- In the Methods section, the authors did not include all the necessary details:
- It is not clear how did they restrict their analysis to atmospheric ions
- It is not clear why did you choose 1.6 and 3 nm. The authors should explain this in the Method section. They explained that ions larger than 3 nm might not be local ions. I do not understand this very well. Do you mean you do not have any ion larger than 3 nm generated at the local? How do you know that? How can you prove this?
- How do you define the background?
- You should add statistical analysis in this section and show results later.
- Did you use any chemical transport models for section 3.2? How do you calculate GR? Did you include chemical reactions? What is your gas precursors and what is the NPF mechanism? This part is not clear at all. You need to add a description of the model in the method section.
- Do you have any meteorological data? Moreover, do you have any gas phase measurements?
- Your results do not convince me.
- Why do you only show March, May, September, and November data?
- Your figure numbers are wrong, so I had difficulty referring to the figures. Also, you did not include the figures you discussed in section 3.3.
- Your figure 1-4 should be violin plots, not line plots. Violin plots can help audiences visualize the distribution and variation.
- I do not understand how you could use size to indicate new particle formation. I understand that the ion concentration threshold in a size bin should be a good indicator for local NPFm (e.g., NPF happens when the ion concentration is higher than a value).
Citation: https://doi.org/10.5194/egusphere-2023-1108-RC2 - In the Methods section, the authors did not include all the necessary details:
- AC1: 'Comment on egusphere-2023-1108', Santeri Tuovinen, 26 Sep 2023
Data sets
Dataset for Measurement report: Ion clusters as indicator for local new particle formation Santeri Tuovinen, Janne Lampilahti, Veli-Matti Kerminen, Markku Kulmala https://doi.org/10.5281/zenodo.8059335
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 353 | 162 | 24 | 539 | 18 | 21 |
- HTML: 353
- PDF: 162
- XML: 24
- Total: 539
- BibTeX: 18
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1