the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Mar 2023
Investigating the contribution of grown new particles to cloud condensation nuclei with largely varying pre-existing particles – Part 1: Observational data analysis
Abstract. This study employed multiple techniques to investigate the contribution of grown new particles to the number concentration of cloud condensation nuclei (CCN) at various supersaturation (SS) levels at a rural mountain site in North China Plain from 29 June to 14 July 2019. On eight new particle formation (NPF) days, the total particle number concentrations (Ncn) were 8.4±6.1 ×103 cm-3, which were substantially higher compared to 4.7±2.6 ×103 cm-3 on non-NPF days. However, the Nccn at 0.2 %SS and 0.4 %SS on the NPF days were significantly lower than those observed on non-NPF days (P<0.05). This was due to the lower cloud activation efficiency of pre-existing particles resulting from organic vapor condensation and smaller number concentrations of pre-existing particles on NPF days. A case-by-case examination showed that the grown new particles only yielded a detectable contribution to Nccn at 0.4 % SS and 1.0 % SS during the NPF event on 1 July 2019, accounting for 12±11 % and 23±12 % of Nccn, respectively. The increased Nccn during two other NPF events and at 0.2 % SS on 1 July 2019 were detectable, but determined mainly by varying pre-existing particles rather than grown new particles. In addition, the hygroscopicity parameter values, concentrations of inorganic and organic particulate components, and surface chemical composition of different sized particles were analyzed in terms of chemical drivers to grow new particles. The results showed that the grown new particles via organic vapor condensation generally had no detectable contribution to Nccn, but incidentally did. However, this conclusion was drawn from a small size of observational data, leaving more observations, particularly for long-term observations and the growth of pre-existing particles to the CCN required size, needed for further investigation.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1648 KB)
-
Supplement
(695 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1648 KB) - Metadata XML
-
Supplement
(695 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-539', Anonymous Referee #1, 13 Apr 2023
The manuscript discusses the impact of new particle formation on CCN based on a measurement campaign conducted on a mountain site situated in the North China Plain. The subject matter is significant, and the dataset is valuable, making me eager to see a comprehensive study that is worthy of publication. However, I have concerns about the manuscript's quality and the inadequate discussion of the findings. At its present state, I cannot recommend it for publication in ACP. Major revisions are necessary before it can be considered for publication. Please refer to my comments and suggestions below.
General comments:
- In this study, the kappa parameter is used as a crucial link between CCN and aerosol chemical composition in the analysis of the impact of NPF on CCN. However, the computed kappa values in this study appear unusual. For instance, Fig. 3-5 often displays a consistent low value below 0.1. Moreover, the equation (1) for calculating kappa cannot be applied if the value obtained is below 0.2 (Petters and Kreidenweis 2007). Additionally, for the situation examined in this study, the aerosol particles are likely to be externally mixed due to the presence of newly formed and pre-existing particles. As a result, the relationship between CCN and kappa, assuming an internally mixed state, may exhibit significant deviations (Wex et al. 2010). It is crucial to discuss how this will affect the findings in this study.
- The quality control of the CCN and PNSD dataset in this study appears to be unclear, which raises doubts about the reliability of the findings. Given that bulk CCN measurements were conducted on a mountain site, calibration of the CCN counter at this altitude and correction for water depletion are necessary (Lance et al., 2006; Rose et al., 2008; Lathem and Nenes, 2011). Furthermore, because the critical diameter and kappa values are computed based on the comparison between PNSD and CCN number concentration, it is essential to verify the consistency between the CCN counter and CPC. For example, it is crucial to check how they perform when measuring particles that are large enough for CCN activation under specific SS conditions. Unfortunately, I could not find any information about these crucial aspects in this study.
- In this study, the impact of both newly grown and pre-existing particles on CCN number concentration during NPF events was analyzed. It was found that in two NPF cases, pre-existing particles, rather than newly grown particles, were responsible for the enhancement of CCN number concentration. This phenomenon is significant and may have occurred in other measurement activities, but was identified as NPF events that did not affect CCN number concentration. Since this phenomenon was observed in two out of ten NPF cases in this study, it is worth investigating whether it has been observed in previous measurement studies. I recommend that the authors conduct a more thorough survey, including a comprehensive comparison with previous results, especially those obtained from mountain sites. This can significantly increase the novelty and significance of this study since if the enhancement of CCN number concentration is incorrectly attributed to newly grown particles, the estimated contribution of NPF to CCN number concentration may be significantly skewed.
Specific comments:
- L30-40 in P2: It would be beneficial to first introduce the global average kappa before evaluating the kappa value of different chemical compositions as higher or lower.
- L31-33 in P7: Which rural site is being referred to? Is it the SMEAR II station? How does the comparison of PNSD and kappa between the SMEAR II station and the site in this study differ? Will these differences affect the comparison of the contribution of different sources to CCN in this study?
- L20-21 in P10: Since the PNSD width of newly formed particles at a specific time can range up to tens of nanometers and the number concentration of newly formed particles that are much larger than the median mode diameter can still be higher than the number concentration of pre-existing particles, it is not appropriate to use the median mode diameter alone to determine whether the grown new particles were too small.
- L2-3 in P13: As you have identified a more reasonable start time of estimating the net contribution of the grown new particles, please present your findings based on the analysis of the net contribution of the grown new particles.
- L21-22 in P13: The references you have cited refer to measurements taken in polluted urban areas, which is not at all applicable to this study.
- L31-37 in P13: The authors speculate that the formation of NH4NO3 and the condensation of hygroscopic organics are major drivers in increasing the CCN number concentration during NPF events, based on the measurement of NO3 and SOC. However, it is important to consider the potential contribution of organonitrates, which are an important secondary aerosol composition consisting of both NO3 and organics (Rollins et al., 2012; Ehn et al., 2014). This should be addressed in the discussion, as it may have an impact on the conclusions drawn in this study.
- L11-15 in P14: The authors compared the calculated CDNC to the measured CCN number concentration and concluded that only a small portion of CCN can form cloud droplets. While this conclusion may be correct, the authors' analysis is flawed. This is because CDNC is determined not only by the number of CCN but also by the actual SS present in the atmosphere (Pruppacher and Klett, 2012). Under varying meteorological conditions, the SS can differ significantly and can significantly impact CDNC even if the number of CCN remains constant. Therefore, this aspect needs to be taken into account in the corresponding discussions.
- L7-8 in P16: This speculation is difficult to understand. In addition, please provide more information on the implications or prospects on the findings of this study. In particular, given that you have a companion paper, presenting it and discussing its relevance will highlight the importance of this research.
Technical corrections:
L20 in P4: It should be "splitter" rather than "spitter".
L36 in P4: What are the start and end times of the TSP sampling?
L16-19 in P8: This sentence is unclear. Please revise for clarity.
L39-40 in P8: This sentence is unclear. Please revise for clarity.
L25 in P10: It's odd to say "organic vapor was growing on the pre-existing particles". Use "condensing" instead.
L3 in P15: There is no definition of PC1 or PC2. Please provide a definition.
Fig. 2: The range of the Y-axis in panel (c) is too large, and it appears that there is no variation of CDNC with time. Please adjust the range of the Y-axis.
Fig. 3-5: Please include a time series of the critical diameter (Dc) in the panel of PNSD.
Fig. 7: The markers in panel (a) are too small to identify. Please increase their size.
Reference:
Ehn M, Thornton J A, Kleist E, et al. A large source of low-volatility secondary organic aerosol [J]. Nature, 2014,506(7489):476–479.
Lance, S., Nenes, A., Medina, J., and Smith, J. N.: Mapping the operation of the DMT continuous flow CCN counter, Aerosol science and technology, 40, 242–254, 2006.
Lathem, T. L. and Nenes, A.: Water vapor depletion in the DMT continuous-flow CCN chamber: Effects on supersaturation and droplet growth, Aerosol Science and Technology, 45, 604–615, 2011.
Petters, M. D. and Kreidenweis, S. M.: A single parameter representation of hygroscopic growth and cloud condensation nucleus activity, Atmos. Chem. Phys., 7, 1961–1971, 2007.
Pruppacher, H. R. and Klett, J. D.: Microphysics of Clouds and Precipitation: Reprinted 1980, Springer Science & Business Media, Berlin, Germany, 2012.
Rollins A W, Browne E C, Min K E, et al. Evidence for NO control over nighttime SOA formation [J]. Science, 2012,337(6099):1210– 1212.
Rose, D., Gunthe, S. S., Mikhailov, E., et al.: Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment, Atmos. Chem. Phys., 8, 1153–1179, 2008.
Wex, H., McFiggans, G., Henning, S., and Stratmann, F.: Influence of the external mixing state of atmospheric aerosol on derived CCN number concentrations, Geophys. Res. Lett., 37, 10805, doi:10.1029/2010GL043337, 2010.
Citation: https://doi.org/10.5194/egusphere-2023-539-RC1 -
AC1: 'Reply on RC1', Xiaohong Yao, 19 Jun 2023
The manuscript discusses the impact of new particle formation on CCN based on a measurement campaign conducted on a mountain site situated in the North China Plain. The subject matter is significant, and the dataset is valuable, making me eager to see a comprehensive study that is worthy of publication. However, I have concerns about the manuscript's quality and the inadequate discussion of the findings. At its present state, I cannot recommend it for publication in ACP. Major revisions are necessary before it can be considered for publication. Please refer to my comments and suggestions below.
Response: The authors express gratitude for the constructive comments and revise the manuscript accordingly. We acknowledge the importance of conducting a comprehensive comparison with previously reported mountain observations and have incorporated this aspect to further enhance the depth of our analysis. Moreover, the weakness of this study is also clarified, particularly for their influences on our analysis.
General comments:
1.In this study, the kappa parameter is used as a crucial link between CCN and aerosol chemical composition in the analysis of the impact of NPF on CCN. However, the computed kappa values in this study appear unusual. For instance, Fig. 3-5 often displays a consistent low value below 0.1. Moreover, the equation (1) for calculating kappa cannot be applied if the value obtained is below 0.2 (Petters and Kreidenweis 2007). Additionally, for the situation examined in this study, the aerosol particles are likely to be externally mixed due to the presence of newly formed and pre-existing particles. As a result, the relationship between CCN and kappa, assuming an internally mixed state, may exhibit significant deviations (Wex et al. 2010). It is crucial to discuss how this will affect the findings in this study.
Response: The authors would like to express their appreciation for the comments received. Based on long-term Hygroscopic Tandem Differential Mobility Analyzer measurements at a rural site (Wangdu, 38∘40′ N, 115∘08′ E, 51 m a.s.l.) in NCP (private talk with Prof. Nan Ma), the annual averages of κ values in the rural site indeed decreased from approximately 0.3 to around 0.15 during the last decade because of large decreases in SO2 and NOx emissions in the region. Miao et al. (2015) also used a Hygroscopic Tandem Differential Mobility Analyzer to measure κ values of aerosols in the clean rural atmosphere at the top of Mt. Huang in China. They also found two distinct modes of κ values, i.e., at approximately 0.3 and less than 0.1, respectively. A lot of κ values in this study below 0.1 might reflect the real situation after substantial decrease in SO2 and NOx emission in NCP. We agree that the calculated κ values using equation (1) might suffer from the large errors. However, to compare with those widely reported in the literature, we calculated the κ values larger than 0.1 and assign a constant value of 0.05 to those below 0.1. Moreover, we adopted the approach used by various groups in research community to estimate κ values. The approach assumes that the atmospheric aerosols are internally mixed. The approach is invalid when the grown newly particles are large enough to be activated as CCN in competing with pre-existing particles. The weakness and related references have been added in the revision.
2. The quality control of the CCN and PNSD dataset in this study appears to be unclear, which raises doubts about the reliability of the findings. Given that bulk CCN measurements were conducted on a mountain site, calibration of the CCN counter at this altitude and correction for water depletion are necessary (Lance et al., 2006; Rose et al., 2008; Lathem and Nenes, 2011). Furthermore, because the critical diameter and kappa values are computed based on the comparison between PNSD and CCN number concentration, it is essential to verify the consistency between the CCN counter and CPC. For example, it is crucial to check how they perform when measuring particles that are large enough for CCN activation under specific SS conditions. Unfortunately, I could not find any information about these crucial aspects in this study.
Response: To address the comments. A correlation analysis between Nccn at 0.2% SS and Ncn>100 was conducted and added in the revision. The results of this analysis demonstrate these two variables were reasonably consistent. The particle number size distributions (PNSD) measured by the FMPS have been corrected using the well-established approach in the literature. The measured PNSD is reasonably consistent with those measured by a SMPS, although only a limited amount of data was available. The comparison has been added in revised Supporting Information. Moreover, a comparison between the FMPS data and WPS data was conducted after the campaign and the consistent results were also provided in revised Supporting Information. We bought a commercial service provided a DMT vendor in China for the calibration of the CCN counter before the campaign. We didn’t make the on-site calibration for the CCN counter. Based on the references, the absence of on-site calibration may lead to as large as ±5% uncertainty on the supersaturation (SS) and consequently cause up to 18% analytic errors on the measured Nccn. The information has been added in the revision. The analytic errors, however, less likely affect our comparison of Nccn at different times within a single day, when it can be reasonably assumed that analytic errors in percentage are invariant. This has clarified in the revision.
3. In this study, the impact of both newly grown and pre-existing particles on CCN number concentration during NPF events was analyzed. It was found that in two NPF cases, pre-existing particles, rather than newly grown particles, were responsible for the enhancement of CCN number concentration. This phenomenon is significant and may have occurred in other measurement activities, but was identified as NPF events that did not affect CCN number concentration. Since this phenomenon was observed in two out of ten NPF cases in this study, it is worth investigating whether it has been observed in previous measurement studies. I recommend that the authors conduct a more thorough survey, including a comprehensive comparison with previous results, especially those obtained from mountain sites. This can significantly increase the novelty and significance of this study since if the enhancement of CCN number concentration is incorrectly attributed to newly grown particles, the estimated contribution of NPF to CCN number concentration may be significantly skewed.
Response: Agree. The comparison has been added in the revision, although the observations of CCN during the NPF events at the mountain sites are very limited. Moreover, Ren et al. (2021) recently reviewed NPF effect on CCN from 35 sites worldwide. The updated information has also been included in the revision.
Specific comments:
1. L30-40 in P2: It would be beneficial to first introduce the global average kappa before evaluating the kappa value of different chemical compositions as higher or lower.
Response: Agree. It has been revised.
2. L31-33 in P7: Which rural site is being referred to? Is it the SMEAR II station? How does the comparison of PNSD and kappa between the SMEAR II station and the site in this study differ? Will these differences affect the comparison of the contribution of different sources to CCN in this study?
Response: In the revision, it has been clarified to compare with the observations made on Mt. Huang in China during the summer of 2014. Mt. Huang is surrounded by developing areas and less affected by anthropogenic air pollutants. In addition, their Hygroscopic Tandem Differential Mobility Analyzer observations on Mt. Huang showed the κ values had two distinct modes distributions at ~0.3 and <0.1, which are reasonably consistent with those estimated in this study.
3. L20-21 in P10: Since the PNSD width of newly formed particles at a specific time can range up to tens of nanometers and the number concentration of newly formed particles that are much larger than the median mode diameter can still be higher than the number concentration of pre-existing particles, it is not appropriate to use the median mode diameter alone to determine whether the grown new particles were too small.
Response: Agree. For the maximum median mode diameter of the grown new particles at 46±2 nm, the width of PNSD should be considered. In the revision, it reads as “However, the grown new particles were still too small to be activated as CCN at 13:00–15:00 with the median mode diameter plus three times of the corresponding standard deviation (3×2 nm) to be considered. From 18:00 to 24:00, the maximum median mode diameter of the grown new particles stopped at 46±2 nm and three times of the corresponding standard deviation at 3×2 nm.”
4. L2-3 in P13: As you have identified a more reasonable start time of estimating the net contribution of the grown new particles, please present your findings based on the analysis of the net contribution of the grown new particles.
Response: The part has been revised as “However, using Nccn,diffat 11:00 appeared to be more reasonable than using Nccn,diff at 14:00 to estimate the net contribution of the grown new particles to Nccn at 1.0 % SS. The test results are presented below. We assumed the Nccn,diff value at 11:00 (864 cm-3) to represent the Nccn,diff of pre-existing particles after 12:00, and assumed that the Nccn,diffwas invariant after 12:00. It can obtain that the net contribution of the grown new particles was 769±514 cm-3 from 12:00 on 1 July to 05:00 on 2 July, accounting for only 23±12 % of Nccn at 1.0 % SS. The maximum net contribution was 1.9 ×103 cm-3 at 18:00 on 1 July, which accounted for 42 % of Nccn at 1.0 % SS. We also observed a minimum contribution of 4 % at 03:00 on 2 July, which was consistent with the disappearance of new particle signals. Alternatively, we tried that the Nccn,diff at 14:00 (1533 cm-3) represented the substrate constant Nccn,diff of the pre-existing particles after 15:00. We observed negative net contributions of the grown new particles to Nccn at 1.0 % SS after 22:00 on July 1, suggesting that the Nccn,diff of pre-existing particles was overestimated.”
5. L21-22 in P13: The references you have cited refer to measurements taken in polluted urban areas, which is not at all applicable to this study.
Response: Thanks. The references have been updated in the revision.
6. L31-37 in P13: The authors speculate that the formation of NH4NO3 and the condensation of hygroscopic organics are major drivers in increasing the CCN number concentration during NPF events, based on the measurement of NO3 and SOC. However, it is important to consider the potential contribution of organonitrates, which are an important secondary aerosol composition consisting of both NO3 and organics (Rollins et al., 2012; Ehn et al., 2014). This should be addressed in the discussion, as it may have an impact on the conclusions drawn in this study.
Response: We agree that organonitrates are an important secondary aerosol composition, especially during nighttime. However, the hygroscopic properties of organonitrates are poorly characterized in the existing literature. We doubt whether they are comparable with NH4NO3 on affecting κ values of atmospheric particles. Thus, we added the related discussion with secondary organic tracers in the revision. It reads as “In the literature, organonitrates were reported as an important secondary aerosol composition at nighttime. However, the species were not measured in this study. Thus, the influence of organonitrates on κ values of the observed atmospheric particles is unknown.”
7. L11-15 in P14: The authors compared the calculated CDNC to the measured CCN number concentration and concluded that only a small portion of CCN can form cloud droplets. While this conclusion may be correct, the authors' analysis is flawed. This is because CDNC is determined not only by the number of CCN but also by the actual SS present in the atmosphere (Pruppacher and Klett, 2012). Under varying meteorological conditions, the SS can differ significantly and can significantly impact CDNC even if the number of CCN remains constant. Therefore, this aspect needs to be taken into account in the corresponding discussions.
Response: Agree. The part has been revised as “The CDNC is a strong function of the actual SS present in the atmosphere in addition to Nccn (Pruppacher and Klett, 2012). This large difference between the observed Nccn and satellite-derived CDNC implies that the actual SS in the atmosphere might be substantially smaller than 0.2%. Shen et al. (2018) recently reported the actual SS varying from 0.01% to 0.05% during fog events observed in the NCP. Thus, it is not surprising to find that only a small fraction of CCN could competitively capture water vapor to form cloud droplets during the study period (Shen et al., 2018; Jiang et al., 2021; Gong et al., 2023). Moreover, the CDNC during the period from 29 June to 14 July was 120±86 cm-3. The satellite-derived CDNC on 2–3 July were even lower than the average, suggesting that the NPF event was unlikely to have any influence on CNDC at such low actual SS.”
8. L7-8 in P16: This speculation is difficult to understand. In addition, please provide more information on the implications or prospects on the findings of this study. In particular, given that you have a companion paper, presenting it and discussing its relevance will highlight the importance of this research.
Response: It revised as “This suggests that the SS to form cloud droplets in the NCP atmospheres may be substantially smaller than 0.2%. Again, the grown new particles didn’t yield a detectable contribution to Nccn at 0.2% SS during all eight NPF events. Thus, it is reasonably argued that the grown new particles might not act as Nccn in NCP atmospheres with actual SS largely smaller than 0.2%.”
Technical corrections:
1. L20 in P4: It should be "splitter" rather than "spitter".
Response: Corrected.
2. L36 in P4: What are the start and end times of the TSP sampling?
Response: The information has been added in the revision. It started from 06:00 and ended at 06:00 in next day.
3. L16-19 in P8: This sentence is unclear. Please revise for clarity.
Response: In the revision, it reads as “However, the occurrence of pre-existing particle growth seemed infrequent and was observed only on 4 July (1 of 16 days). On that day, larger and smaller pre-existing particle growth was observed from 88 nm to 116 nm and from 24 nm to 32 nm, respectively (See Fig. S2).”
4. L39-40 in P8: This sentence is unclear. Please revise for clarity.
Response: For clarification, the last two sentences are combined in the revision. It reads as “It is important to note that the observational data alone cannot provide evidence of any additional evolution of new particles and their additional contribution to Nccn after the new particle signal disappears, particularly considering the infrequent occurrence of the pre-existing particle growth.”
5. L25 in P10: It's odd to say "organic vapor was growing on the pre-existing particles". Use "condensing" instead.
Response: Corrected.
6. L3 in P15: There is no definition of PC1 or PC2. Please provide a definition.
Response: PC1 or PC2 represent the obtained two major principal component factors, which have been defined in the revision.
7. Fig. 2: The range of the Y-axis in panel (c) is too large, and it appears that there is no variation of CDNC with time. Please adjust the range of the Y-axis.
Response: Corrected.
8. Fig. 3-5: Please include a time series of the critical diameter (Dc) in the panel of PNSD.
Response: Added.
9. Fig. 7: The markers in panel (a) are too small to identify. Please increase their size.
Response: Corrected.
References:
[1]Ehn M, Thornton J A, Kleist E, et al. A large source of low-volatility secondary organic aerosol [J]. Nature, 2014,506(7489):476–479.
[2]Lance, S., Nenes, A., Medina, J., and Smith, J. N.: Mapping the operation of the DMT continuous flow CCN counter, Aerosol science and technology, 40, 242–254, 2006.
[3]Lathem, T. L. and Nenes, A.: Water vapor depletion in the DMT continuous-flow CCN chamber: Effects on supersaturation and droplet growth, Aerosol Science and Technology, 45, 604–615, 2011.
[4]Petters, M. D. and Kreidenweis, S. M.: A single parameter representation of hygroscopic growth and cloud condensation nucleus activity, Atmos. Chem. Phys., 7, 1961–1971, 2007.
[5]Pruppacher, H. R. and Klett, J. D.: Microphysics of Clouds and Precipitation: Reprinted 1980, Springer Science & Business Media, Berlin, Germany, 2012.
[6]Rollins A W, Browne E C, Min K E, et al. Evidence for NO control over nighttime SOA formation [J]. Science, 2012,337(6099):1210– 1212.
[7]Rose, D., Gunthe, S. S., Mikhailov, E., et al.: Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment, Atmos. Chem. Phys., 8, 1153–1179, 2008.
[8]Wex, H., McFiggans, G., Henning, S., and Stratmann, F.: Influence of the external mixing state of atmospheric aerosol on derived CCN number concentrations, Geophys. Res. Lett., 37, 10805, doi:10.1029/2010GL043337, 2010.
Response: The refences have been included in the revision to improve the quality of discussion.
Citation: https://doi.org/10.5194/egusphere-2023-539-AC1
-
RC2: 'Comment on egusphere-2023-539', Anonymous Referee #2, 20 Apr 2023
General comments
This paper is a case study discussing whether new particles generated by new particle formation (NPF) events grow and contribute to CCN based on two weeks of atmospheric observations. It is an important topic involving aerosol-cloud interactions, and the accumulation of such case studies is meaningful given the large spatiotemporal variability of aerosol properties. However, because of the following problems, a major revision of manuscript is needed before it can be accepted for publication.
The authors explain the case in several categories as to whether NPF events contribute to CCN number concentrations, but it feels illogical that the method of categorization is ambiguous and that even on the same day it is discussed in separate sections by supersaturation. In addition, cloud particle number concentrations were not measured in the observation site, so caution should be exercised when comparing calculated values with observed CCN number concentrations. The authors raise three research questions for introduction, so in conclusion part, it is necessary to answer these clearly. In particular, there is no description of the answer to the third research question: What implications do our findings have on knowledge gaps for CCN sources in NCP?
Other detailed points are listed below.
Specific comments
P2 L16
The lower limit of the specific particle size in this study should be mentioned in the method section.
P3 L28
Can the authors identify or estimate the period of cloud coverage at the observation site? Are data on relative humidity in the atmosphere, for example, available? Because cloud formation contributes to the deposition of pre-existing particles, it is considered an important data to deepen the discussion of NPF case studies.
P4 L12
It would be helpful if the authors include the piping diagrams for the four instruments in the supplement material.
P4 L22
How often switch the SS setting for CCNC? i.e., How many minutes each supersaturation setting lasted?
P6 L6
Yao et al. (2005) is missing in the reference list. It should be Yao et al. (2007) or Yao et al. (2010)?
P6 L26
It is questionable whether these satellite products reflect clouds that form at observation sites. In the introduction, the authors write that observation sites are often covered in morning mist, but can satellite products capture this phenomenon?
P6 L29
How many days back in the backward trajectory analysis?
P7 L18
As commented on the methodology, it is questionable whether the CDNC calculated based on the satellite products reflects the CDNC of the lower clouds at the observation site. As the CDNC strongly depends on the water vapor supersaturation in the cloud and, for that matter, the ascent velocity of the air mass, so it is also advisable to avoid comparing CCN concentration with CDNC without these discussions.
P7 L24
It’s not mandatory but there is a paper that summarizes CCN number concentrations (at 0.2% SS) at sites around the world (including mountain sites).
Schmale, J.et al.: Long-term cloud condensation nuclei number concentration, particle number size distribution and chemical composition measurements at regionally representative observatories, Atmos. Chem. Phys., 18, 2853–2881, https://doi.org/10.5194/acp-18-2853-2018, 2018.
P7 L31
This reviewer feels there is a weak basis for estimating the contribution rates of natural and anthropogenic sources. Since the observation site is in the vicinity of Beijing, it may depend strongly on the wind directions.
P8 L30
This paragraph barely discusses particle size. Whether or not a particle acts as a CCN depends strongly on the particle size, so it is natural that a particle in the newly formed nucleation mode does not contribute to NCN>100 orNCCN.
P9 L5
Wang Y., et al., 2020 seems to be missing in the reference list.
P9 L11
This reviewer doesn't quite understand the intent of discussing the events of July 1 in subsections by supersaturation. Shouldn't 3.4 and 3.5 be in the same section?
P9 L24
Because of the wide range of kappa value of organics, this is likely to depend largely on the type of organic matter.
P10 L36
The 3 July case also confirmed an increase in NCCN, but concluded that this was due to pre-existing particles. this reviewer is not sure what the difference is between this case (3 July) and the section 3.3 cases. Couldn't the case of 3 July be included in 3.3?
P14 L8
Same comment as P7-L18.
Section 3.7
This part is unique and interesting but somewhat speculative. Can't SIMS get information about inorganic materials on the particle surface? If possible, discussing it in conjunction with organic matter would make the discussion more robust.
P14 L34
This reviewer does not understand why the authors concluded that condensation of inorganic vapors is dominant in larger size particles (> 60 nm).
Figure 1
Airmass on the day of the NPF event is being transported from the north or northwest direction. Was there a difference in the airmass pathway on days with and without NPF events?
Figure 2a
It is very hard to see and misleading if separate axes for NCN and NCCN were used. Can it be represented by a single axis, with the vertical axis being the logarithmic axis? Also, making a separate graph of the activation ratio (NCCN/NCN) would be helpful to see how the cloud activation potential is.
Figure 2c
Please zoom in on the vertical scale to see the changes more easily. What is the meaning of the inset figure (CDNC in June and July)?
Figure 3b and 3e
Same comment as Figure 2a. Better to have NCN >100 andNCCN on the same axis.
Figure 4b and 4e
Same comment as above. Also, the captions of the figures should provide sufficient explanation to understand them. (without looking at the text or other figures)
Matters related notations
- “P” for p-values should be italicized in lower case letter "p".
- All variables (NCN, NCCN, κ....) in the manuscript and supplementary material should be italicized.
Citation: https://doi.org/10.5194/egusphere-2023-539-RC2 -
AC2: 'Reply on RC2', Xiaohong Yao, 19 Jun 2023
The authors explain the case in several categories as to whether NPF events contribute to CCN number concentrations, but it feels illogical that the method of categorization is ambiguous and that even on the same day it is discussed in separate sections by supersaturation. In addition, cloud particle number concentrations were not measured in the observation site, so caution should be exercised when comparing calculated values with observed CCN number concentrations. The authors raise three research questions for introduction, so in conclusion part, it is necessary to answer these clearly. In particular, there is no description of the answer to the third research question: What implications do our findings have on knowledge gaps for CCN sources in NCP?
Response: The authors carefully considered the comments and revised accordingly to enhance the overall quality of the presentation.
Specific comments
1. P2 L16 The lower limit of the specific particle size in this study should be mentioned in the method section.
Response: It is 5.6 nm and this has been clarified in the Experimental, i.e., “The FMPS was used to measure particle concentrations in the size range of 5.6 nm to 560 nm in 32 channels at a frequency of 1 Hz.”
2. P3 L28 Can the authors identify or estimate the period of cloud coverage at the observation site? Are data on relative humidity in the atmosphere, for example, available? Because cloud formation contributes to the deposition of pre-existing particles, it is considered an important data to deepen the discussion of NPF case studies.
Response: It is difficult to accurate estimate the period of cloud coverage at the mountain area since the cloud widely separated. In the revision, satellite cloud coverage and the ground-level relative humidity were presented in the SI to facilitate the analysis.
3. P4 L12 It would be helpful if the authors include the piping diagrams for the four instruments in the supplement material.
Response: Added.
4. P4 L22 How often switch the SS setting for CCNC? i.e., How many minutes each supersaturation setting lasted?
Response: As clarified in the revision, each supersaturation (SS) setting for the CCNC lasted for a duration of 5 minutes. However, an additional 5 minutes were used for switching SS at 1.0% SS to 0.2% SS due to the necessity of establishing supersaturation equilibrium.
5. P6 L6 Yao et al. (2005) is missing in the reference list. It should be Yao et al. (2007) or Yao et al. (2010)?
Response: Sorry for this. It has been corrected in the revision.
6. P6 L26 It is questionable whether these satellite products reflect clouds that form at observation sites. In the introduction, the authors write that observation sites are often covered in morning mist, but can satellite products capture this phenomenon?
Response: Satellite data reflect observations between 12:30-14:30. This has been clarified in the revision.
7. P6 L29 How many days back in the backward trajectory analysis?
Response: 24-hrs air mass back trajectories were used for this study. This has been added in the revision.
8. P7 L18 As commented on the methodology, it is questionable whether the CDNC calculated based on the satellite products reflects the CDNC of the lower clouds at the observation site. As the CDNC strongly depends on the water vapor supersaturation in the cloud and, for that matter, the ascent velocity of the air mass, so it is also advisable to avoid comparing CCN concentration with CDNC without these discussions.
Response: We agree the actual SS mainly determined by ascent velocity of the air mass is critical in continental atmospheres affected by anthrophonic air pollution to some extent. In the revision, we added related discussion accordingly. Moreover, the Nccn observed in the rural mountain atmosphere was considerably higher than the cloud droplet number concentrations derived from satellites. This suggests that the SS to form cloud droplets in the NCP atmospheres may be substantially smaller than 0.2%. Again, the grown new particles didn’t yield a detectable contribution to Nccn at 0.2% SS during all eight NPF events in this study. Thus, it is reasonably argued that the grown new particles might not act as Nccn in NCP atmospheres with actual SS largely smaller than 0.2%.
9. P7 L24 It’s not mandatory but there is a paper that summarizes CCN number concentrations (at 0.2% SS) at sites around the world (including mountain sites). Schmale, J.et al.: Long-term cloud condensation nuclei number concentration, particle number size distribution and chemical composition measurements at regionally representative observatories, Atmos. Chem. Phys., 18, 2853–2881, https://doi.org/10.5194/acp-18-2853-2018, 2018.
Response: Agree. We have added a comparison of the observations of CCN during the NPF events at the mountain sites in the revision.
10. P7 L31 This reviewer feels there is a weak basis for estimating the contribution rates of natural and anthropogenic sources. Since the observation site is in the vicinity of Beijing, it may depend strongly on the wind directions.
Response: This is a rough estimation. When the wind directions from the south and southwest, the contribution of Nccn from primary and secondary anthropogenic aerosols should be even larger. In the revision, we added “In Beijing, NPF events were also observed in polluted atmospheres with air masses originating from the south and southwest (Wu et al., 2007). In those cases, anthropogenic aerosols expectedly yield an even larger contribution to Nccn.”
11. P8 L30 This paragraph barely discusses particle size. Whether or not a particle acts as a CCN depends strongly on the particle size, so it is natural that a particle in the newly formed nucleation mode does not contribute to NCN>100 orNCCN.
Response: Agree. This is why we presented the case-by-case examination of the growth of new particles and the potential contribution to Nccn. To better service the reader, we added “ Moreover, the case-by-case examination of the growth of new particles and potential contribution to Nccnwill be presented in Section 3.2-3.3.”
12. P9 L5 Wang Y., et al., 2020 seems to be missing in the reference list.
Response: It is “Wan et al., 2020”. Corrected.
13. P9 L11 This reviewer doesn't quite understand the intent of discussing the events of July 1 in subsections by supersaturation. Shouldn't 3.4 and 3.5 be in the same section?
Response: The two subsections have been combined.
14. P9 L24 Because of the wide range of kappa value of organics, this is likely to depend largely on the type of organic matter.
Response: In the revision, it reads as “Assuming that the activated aerosols at 0.2 % SS were internally mixed and mainly composed of inorganic ammonium salts and organics (Petters and Kreidensohler, 2007; Rose et al., 2010, 2011), both of them likely yielded an appreciable contribution to the total mass concentration of the associated aerosols. However, the exact percentages relied on the κ values of various organics.”
15. P10 L36 The 3 July case also confirmed an increase in NCCN, but concluded that this was due to pre-existing particles. this reviewer is not sure what the difference is between this case (3 July) and the section 3.3 cases. Couldn't the case of 3 July be included in 3.3?
Response: We combined the two sections together in the revision.
16. P14 L8 Same comment as P7-L18.
Response: The paragraph has been revised as “We examined the satellite-derived CDNC over the mountain area during and after the NPF day. The values were 169 cm-3, 89 cm-3, and 101 cm-3 on 1, 2, and 3 July, respectively (See Fig. 2c). These values were approximately one order of magnitude smaller than the observed Nccn at 0.2 % SS on those 3 days. The CDNC is a strong function of the actual SS present in the atmosphere in addition to Nccn (Pruppacher and Klett, 2012). This large difference between the observed Nccn and satellite-derived CDNC implies that the actual SS in the atmosphere might be substantially smaller than 0.2%. Shen et al. (2018) recently reported the actual SS varying from 0.01% to 0.05% during fog events observed in the NCP. Thus, it is not surprising to find that only a small fraction of CCN could competitively capture water vapor to form cloud droplets during the study period (Shen et al., 2018; Jiang et al., 2021; Gong et al., 2023). Moreover, the CDNC during the period from 29 June to 14 July was 120±86 cm-3. The satellite-derived CDNC on 2–3 July were even lower than the average, suggesting that the NPF event was unlikely to have any influence on CNDC at such low actual SS.”
17. Section 3.7 This part is unique and interesting but somewhat speculative. Can't SIMS get information about inorganic materials on the particle surface? If possible, discussing it in conjunction with organic matter would make the discussion more robust.
Response: SIMS can detect inorganic ions, but interference is large because of the low ratios of signal and noise. This has been clarified in the revision.
18. P14 L34 This reviewer does not understand why the authors concluded that condensation of inorganic vapors is dominant in larger size particles (> 60 nm).
Response: The sentence has been revised as “when semi-volatile organic and inorganic vapors may have overwhelmingly condensed on the sized particle surfaces and covered up the high-molecular-weight organic fragment signals.”
19. Figure 1 Airmass on the day of the NPF event is being transported from the north or northwest direction. Was there a difference in the airmass pathway on days with and without NPF events?
Response: Wu et al. (2007) reported statistical analysis of NPF events using a year-long measurement in Beijing, NPF events were associated with low RH and sunny conditions. The air masses from the north and northwest usually carry dry and clean air, favoring the occurrence of NPF events. Non-NPF events were usually associated with strong condensational sink or absence of sunny conditions. The reference has been added and summarized to support the analysis in the revision.
20. Figure 2a It is very hard to see and misleading if separate axes for NCN and NCCN were used. Can it be represented by a single axis, with the vertical axis being the logarithmic axis? Also, making a separate graph of the activation ratio (NCCN/NCN) would be helpful to see how the cloud activation potential is.
Response: Revised.
21. Figure 2c Please zoom in on the vertical scale to see the changes more easily. What is the meaning of the inset figure (CDNC in June and July)?
Response: Revised. The observational period is too short to be representative. The two month-long data of CDNC can better reflect its range. This has been clarified in the revision.
22. Figure 3b and 3e Same comment as Figure 2a. Better to have NCN >100 and NCCN on the same axis.
Response: Revised accordingly.
23. Figure 4b and 4e Same comment as above. Also, the captions of the figures should provide sufficient explanation to understand them. (without looking at the text or other figures)
Response: Revised accordingly.
Matters related notations
1. "p" for p-values should be italicized in lower case letter "p".
Response: Corrected.
2. All variables (NCN, NCCN, κ....) in the manuscript and supplementary material should be italicized.
Response: Corrected.
Citation: https://doi.org/10.5194/egusphere-2023-539-AC2
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-539', Anonymous Referee #1, 13 Apr 2023
The manuscript discusses the impact of new particle formation on CCN based on a measurement campaign conducted on a mountain site situated in the North China Plain. The subject matter is significant, and the dataset is valuable, making me eager to see a comprehensive study that is worthy of publication. However, I have concerns about the manuscript's quality and the inadequate discussion of the findings. At its present state, I cannot recommend it for publication in ACP. Major revisions are necessary before it can be considered for publication. Please refer to my comments and suggestions below.
General comments:
- In this study, the kappa parameter is used as a crucial link between CCN and aerosol chemical composition in the analysis of the impact of NPF on CCN. However, the computed kappa values in this study appear unusual. For instance, Fig. 3-5 often displays a consistent low value below 0.1. Moreover, the equation (1) for calculating kappa cannot be applied if the value obtained is below 0.2 (Petters and Kreidenweis 2007). Additionally, for the situation examined in this study, the aerosol particles are likely to be externally mixed due to the presence of newly formed and pre-existing particles. As a result, the relationship between CCN and kappa, assuming an internally mixed state, may exhibit significant deviations (Wex et al. 2010). It is crucial to discuss how this will affect the findings in this study.
- The quality control of the CCN and PNSD dataset in this study appears to be unclear, which raises doubts about the reliability of the findings. Given that bulk CCN measurements were conducted on a mountain site, calibration of the CCN counter at this altitude and correction for water depletion are necessary (Lance et al., 2006; Rose et al., 2008; Lathem and Nenes, 2011). Furthermore, because the critical diameter and kappa values are computed based on the comparison between PNSD and CCN number concentration, it is essential to verify the consistency between the CCN counter and CPC. For example, it is crucial to check how they perform when measuring particles that are large enough for CCN activation under specific SS conditions. Unfortunately, I could not find any information about these crucial aspects in this study.
- In this study, the impact of both newly grown and pre-existing particles on CCN number concentration during NPF events was analyzed. It was found that in two NPF cases, pre-existing particles, rather than newly grown particles, were responsible for the enhancement of CCN number concentration. This phenomenon is significant and may have occurred in other measurement activities, but was identified as NPF events that did not affect CCN number concentration. Since this phenomenon was observed in two out of ten NPF cases in this study, it is worth investigating whether it has been observed in previous measurement studies. I recommend that the authors conduct a more thorough survey, including a comprehensive comparison with previous results, especially those obtained from mountain sites. This can significantly increase the novelty and significance of this study since if the enhancement of CCN number concentration is incorrectly attributed to newly grown particles, the estimated contribution of NPF to CCN number concentration may be significantly skewed.
Specific comments:
- L30-40 in P2: It would be beneficial to first introduce the global average kappa before evaluating the kappa value of different chemical compositions as higher or lower.
- L31-33 in P7: Which rural site is being referred to? Is it the SMEAR II station? How does the comparison of PNSD and kappa between the SMEAR II station and the site in this study differ? Will these differences affect the comparison of the contribution of different sources to CCN in this study?
- L20-21 in P10: Since the PNSD width of newly formed particles at a specific time can range up to tens of nanometers and the number concentration of newly formed particles that are much larger than the median mode diameter can still be higher than the number concentration of pre-existing particles, it is not appropriate to use the median mode diameter alone to determine whether the grown new particles were too small.
- L2-3 in P13: As you have identified a more reasonable start time of estimating the net contribution of the grown new particles, please present your findings based on the analysis of the net contribution of the grown new particles.
- L21-22 in P13: The references you have cited refer to measurements taken in polluted urban areas, which is not at all applicable to this study.
- L31-37 in P13: The authors speculate that the formation of NH4NO3 and the condensation of hygroscopic organics are major drivers in increasing the CCN number concentration during NPF events, based on the measurement of NO3 and SOC. However, it is important to consider the potential contribution of organonitrates, which are an important secondary aerosol composition consisting of both NO3 and organics (Rollins et al., 2012; Ehn et al., 2014). This should be addressed in the discussion, as it may have an impact on the conclusions drawn in this study.
- L11-15 in P14: The authors compared the calculated CDNC to the measured CCN number concentration and concluded that only a small portion of CCN can form cloud droplets. While this conclusion may be correct, the authors' analysis is flawed. This is because CDNC is determined not only by the number of CCN but also by the actual SS present in the atmosphere (Pruppacher and Klett, 2012). Under varying meteorological conditions, the SS can differ significantly and can significantly impact CDNC even if the number of CCN remains constant. Therefore, this aspect needs to be taken into account in the corresponding discussions.
- L7-8 in P16: This speculation is difficult to understand. In addition, please provide more information on the implications or prospects on the findings of this study. In particular, given that you have a companion paper, presenting it and discussing its relevance will highlight the importance of this research.
Technical corrections:
L20 in P4: It should be "splitter" rather than "spitter".
L36 in P4: What are the start and end times of the TSP sampling?
L16-19 in P8: This sentence is unclear. Please revise for clarity.
L39-40 in P8: This sentence is unclear. Please revise for clarity.
L25 in P10: It's odd to say "organic vapor was growing on the pre-existing particles". Use "condensing" instead.
L3 in P15: There is no definition of PC1 or PC2. Please provide a definition.
Fig. 2: The range of the Y-axis in panel (c) is too large, and it appears that there is no variation of CDNC with time. Please adjust the range of the Y-axis.
Fig. 3-5: Please include a time series of the critical diameter (Dc) in the panel of PNSD.
Fig. 7: The markers in panel (a) are too small to identify. Please increase their size.
Reference:
Ehn M, Thornton J A, Kleist E, et al. A large source of low-volatility secondary organic aerosol [J]. Nature, 2014,506(7489):476–479.
Lance, S., Nenes, A., Medina, J., and Smith, J. N.: Mapping the operation of the DMT continuous flow CCN counter, Aerosol science and technology, 40, 242–254, 2006.
Lathem, T. L. and Nenes, A.: Water vapor depletion in the DMT continuous-flow CCN chamber: Effects on supersaturation and droplet growth, Aerosol Science and Technology, 45, 604–615, 2011.
Petters, M. D. and Kreidenweis, S. M.: A single parameter representation of hygroscopic growth and cloud condensation nucleus activity, Atmos. Chem. Phys., 7, 1961–1971, 2007.
Pruppacher, H. R. and Klett, J. D.: Microphysics of Clouds and Precipitation: Reprinted 1980, Springer Science & Business Media, Berlin, Germany, 2012.
Rollins A W, Browne E C, Min K E, et al. Evidence for NO control over nighttime SOA formation [J]. Science, 2012,337(6099):1210– 1212.
Rose, D., Gunthe, S. S., Mikhailov, E., et al.: Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment, Atmos. Chem. Phys., 8, 1153–1179, 2008.
Wex, H., McFiggans, G., Henning, S., and Stratmann, F.: Influence of the external mixing state of atmospheric aerosol on derived CCN number concentrations, Geophys. Res. Lett., 37, 10805, doi:10.1029/2010GL043337, 2010.
Citation: https://doi.org/10.5194/egusphere-2023-539-RC1 -
AC1: 'Reply on RC1', Xiaohong Yao, 19 Jun 2023
The manuscript discusses the impact of new particle formation on CCN based on a measurement campaign conducted on a mountain site situated in the North China Plain. The subject matter is significant, and the dataset is valuable, making me eager to see a comprehensive study that is worthy of publication. However, I have concerns about the manuscript's quality and the inadequate discussion of the findings. At its present state, I cannot recommend it for publication in ACP. Major revisions are necessary before it can be considered for publication. Please refer to my comments and suggestions below.
Response: The authors express gratitude for the constructive comments and revise the manuscript accordingly. We acknowledge the importance of conducting a comprehensive comparison with previously reported mountain observations and have incorporated this aspect to further enhance the depth of our analysis. Moreover, the weakness of this study is also clarified, particularly for their influences on our analysis.
General comments:
1.In this study, the kappa parameter is used as a crucial link between CCN and aerosol chemical composition in the analysis of the impact of NPF on CCN. However, the computed kappa values in this study appear unusual. For instance, Fig. 3-5 often displays a consistent low value below 0.1. Moreover, the equation (1) for calculating kappa cannot be applied if the value obtained is below 0.2 (Petters and Kreidenweis 2007). Additionally, for the situation examined in this study, the aerosol particles are likely to be externally mixed due to the presence of newly formed and pre-existing particles. As a result, the relationship between CCN and kappa, assuming an internally mixed state, may exhibit significant deviations (Wex et al. 2010). It is crucial to discuss how this will affect the findings in this study.
Response: The authors would like to express their appreciation for the comments received. Based on long-term Hygroscopic Tandem Differential Mobility Analyzer measurements at a rural site (Wangdu, 38∘40′ N, 115∘08′ E, 51 m a.s.l.) in NCP (private talk with Prof. Nan Ma), the annual averages of κ values in the rural site indeed decreased from approximately 0.3 to around 0.15 during the last decade because of large decreases in SO2 and NOx emissions in the region. Miao et al. (2015) also used a Hygroscopic Tandem Differential Mobility Analyzer to measure κ values of aerosols in the clean rural atmosphere at the top of Mt. Huang in China. They also found two distinct modes of κ values, i.e., at approximately 0.3 and less than 0.1, respectively. A lot of κ values in this study below 0.1 might reflect the real situation after substantial decrease in SO2 and NOx emission in NCP. We agree that the calculated κ values using equation (1) might suffer from the large errors. However, to compare with those widely reported in the literature, we calculated the κ values larger than 0.1 and assign a constant value of 0.05 to those below 0.1. Moreover, we adopted the approach used by various groups in research community to estimate κ values. The approach assumes that the atmospheric aerosols are internally mixed. The approach is invalid when the grown newly particles are large enough to be activated as CCN in competing with pre-existing particles. The weakness and related references have been added in the revision.
2. The quality control of the CCN and PNSD dataset in this study appears to be unclear, which raises doubts about the reliability of the findings. Given that bulk CCN measurements were conducted on a mountain site, calibration of the CCN counter at this altitude and correction for water depletion are necessary (Lance et al., 2006; Rose et al., 2008; Lathem and Nenes, 2011). Furthermore, because the critical diameter and kappa values are computed based on the comparison between PNSD and CCN number concentration, it is essential to verify the consistency between the CCN counter and CPC. For example, it is crucial to check how they perform when measuring particles that are large enough for CCN activation under specific SS conditions. Unfortunately, I could not find any information about these crucial aspects in this study.
Response: To address the comments. A correlation analysis between Nccn at 0.2% SS and Ncn>100 was conducted and added in the revision. The results of this analysis demonstrate these two variables were reasonably consistent. The particle number size distributions (PNSD) measured by the FMPS have been corrected using the well-established approach in the literature. The measured PNSD is reasonably consistent with those measured by a SMPS, although only a limited amount of data was available. The comparison has been added in revised Supporting Information. Moreover, a comparison between the FMPS data and WPS data was conducted after the campaign and the consistent results were also provided in revised Supporting Information. We bought a commercial service provided a DMT vendor in China for the calibration of the CCN counter before the campaign. We didn’t make the on-site calibration for the CCN counter. Based on the references, the absence of on-site calibration may lead to as large as ±5% uncertainty on the supersaturation (SS) and consequently cause up to 18% analytic errors on the measured Nccn. The information has been added in the revision. The analytic errors, however, less likely affect our comparison of Nccn at different times within a single day, when it can be reasonably assumed that analytic errors in percentage are invariant. This has clarified in the revision.
3. In this study, the impact of both newly grown and pre-existing particles on CCN number concentration during NPF events was analyzed. It was found that in two NPF cases, pre-existing particles, rather than newly grown particles, were responsible for the enhancement of CCN number concentration. This phenomenon is significant and may have occurred in other measurement activities, but was identified as NPF events that did not affect CCN number concentration. Since this phenomenon was observed in two out of ten NPF cases in this study, it is worth investigating whether it has been observed in previous measurement studies. I recommend that the authors conduct a more thorough survey, including a comprehensive comparison with previous results, especially those obtained from mountain sites. This can significantly increase the novelty and significance of this study since if the enhancement of CCN number concentration is incorrectly attributed to newly grown particles, the estimated contribution of NPF to CCN number concentration may be significantly skewed.
Response: Agree. The comparison has been added in the revision, although the observations of CCN during the NPF events at the mountain sites are very limited. Moreover, Ren et al. (2021) recently reviewed NPF effect on CCN from 35 sites worldwide. The updated information has also been included in the revision.
Specific comments:
1. L30-40 in P2: It would be beneficial to first introduce the global average kappa before evaluating the kappa value of different chemical compositions as higher or lower.
Response: Agree. It has been revised.
2. L31-33 in P7: Which rural site is being referred to? Is it the SMEAR II station? How does the comparison of PNSD and kappa between the SMEAR II station and the site in this study differ? Will these differences affect the comparison of the contribution of different sources to CCN in this study?
Response: In the revision, it has been clarified to compare with the observations made on Mt. Huang in China during the summer of 2014. Mt. Huang is surrounded by developing areas and less affected by anthropogenic air pollutants. In addition, their Hygroscopic Tandem Differential Mobility Analyzer observations on Mt. Huang showed the κ values had two distinct modes distributions at ~0.3 and <0.1, which are reasonably consistent with those estimated in this study.
3. L20-21 in P10: Since the PNSD width of newly formed particles at a specific time can range up to tens of nanometers and the number concentration of newly formed particles that are much larger than the median mode diameter can still be higher than the number concentration of pre-existing particles, it is not appropriate to use the median mode diameter alone to determine whether the grown new particles were too small.
Response: Agree. For the maximum median mode diameter of the grown new particles at 46±2 nm, the width of PNSD should be considered. In the revision, it reads as “However, the grown new particles were still too small to be activated as CCN at 13:00–15:00 with the median mode diameter plus three times of the corresponding standard deviation (3×2 nm) to be considered. From 18:00 to 24:00, the maximum median mode diameter of the grown new particles stopped at 46±2 nm and three times of the corresponding standard deviation at 3×2 nm.”
4. L2-3 in P13: As you have identified a more reasonable start time of estimating the net contribution of the grown new particles, please present your findings based on the analysis of the net contribution of the grown new particles.
Response: The part has been revised as “However, using Nccn,diffat 11:00 appeared to be more reasonable than using Nccn,diff at 14:00 to estimate the net contribution of the grown new particles to Nccn at 1.0 % SS. The test results are presented below. We assumed the Nccn,diff value at 11:00 (864 cm-3) to represent the Nccn,diff of pre-existing particles after 12:00, and assumed that the Nccn,diffwas invariant after 12:00. It can obtain that the net contribution of the grown new particles was 769±514 cm-3 from 12:00 on 1 July to 05:00 on 2 July, accounting for only 23±12 % of Nccn at 1.0 % SS. The maximum net contribution was 1.9 ×103 cm-3 at 18:00 on 1 July, which accounted for 42 % of Nccn at 1.0 % SS. We also observed a minimum contribution of 4 % at 03:00 on 2 July, which was consistent with the disappearance of new particle signals. Alternatively, we tried that the Nccn,diff at 14:00 (1533 cm-3) represented the substrate constant Nccn,diff of the pre-existing particles after 15:00. We observed negative net contributions of the grown new particles to Nccn at 1.0 % SS after 22:00 on July 1, suggesting that the Nccn,diff of pre-existing particles was overestimated.”
5. L21-22 in P13: The references you have cited refer to measurements taken in polluted urban areas, which is not at all applicable to this study.
Response: Thanks. The references have been updated in the revision.
6. L31-37 in P13: The authors speculate that the formation of NH4NO3 and the condensation of hygroscopic organics are major drivers in increasing the CCN number concentration during NPF events, based on the measurement of NO3 and SOC. However, it is important to consider the potential contribution of organonitrates, which are an important secondary aerosol composition consisting of both NO3 and organics (Rollins et al., 2012; Ehn et al., 2014). This should be addressed in the discussion, as it may have an impact on the conclusions drawn in this study.
Response: We agree that organonitrates are an important secondary aerosol composition, especially during nighttime. However, the hygroscopic properties of organonitrates are poorly characterized in the existing literature. We doubt whether they are comparable with NH4NO3 on affecting κ values of atmospheric particles. Thus, we added the related discussion with secondary organic tracers in the revision. It reads as “In the literature, organonitrates were reported as an important secondary aerosol composition at nighttime. However, the species were not measured in this study. Thus, the influence of organonitrates on κ values of the observed atmospheric particles is unknown.”
7. L11-15 in P14: The authors compared the calculated CDNC to the measured CCN number concentration and concluded that only a small portion of CCN can form cloud droplets. While this conclusion may be correct, the authors' analysis is flawed. This is because CDNC is determined not only by the number of CCN but also by the actual SS present in the atmosphere (Pruppacher and Klett, 2012). Under varying meteorological conditions, the SS can differ significantly and can significantly impact CDNC even if the number of CCN remains constant. Therefore, this aspect needs to be taken into account in the corresponding discussions.
Response: Agree. The part has been revised as “The CDNC is a strong function of the actual SS present in the atmosphere in addition to Nccn (Pruppacher and Klett, 2012). This large difference between the observed Nccn and satellite-derived CDNC implies that the actual SS in the atmosphere might be substantially smaller than 0.2%. Shen et al. (2018) recently reported the actual SS varying from 0.01% to 0.05% during fog events observed in the NCP. Thus, it is not surprising to find that only a small fraction of CCN could competitively capture water vapor to form cloud droplets during the study period (Shen et al., 2018; Jiang et al., 2021; Gong et al., 2023). Moreover, the CDNC during the period from 29 June to 14 July was 120±86 cm-3. The satellite-derived CDNC on 2–3 July were even lower than the average, suggesting that the NPF event was unlikely to have any influence on CNDC at such low actual SS.”
8. L7-8 in P16: This speculation is difficult to understand. In addition, please provide more information on the implications or prospects on the findings of this study. In particular, given that you have a companion paper, presenting it and discussing its relevance will highlight the importance of this research.
Response: It revised as “This suggests that the SS to form cloud droplets in the NCP atmospheres may be substantially smaller than 0.2%. Again, the grown new particles didn’t yield a detectable contribution to Nccn at 0.2% SS during all eight NPF events. Thus, it is reasonably argued that the grown new particles might not act as Nccn in NCP atmospheres with actual SS largely smaller than 0.2%.”
Technical corrections:
1. L20 in P4: It should be "splitter" rather than "spitter".
Response: Corrected.
2. L36 in P4: What are the start and end times of the TSP sampling?
Response: The information has been added in the revision. It started from 06:00 and ended at 06:00 in next day.
3. L16-19 in P8: This sentence is unclear. Please revise for clarity.
Response: In the revision, it reads as “However, the occurrence of pre-existing particle growth seemed infrequent and was observed only on 4 July (1 of 16 days). On that day, larger and smaller pre-existing particle growth was observed from 88 nm to 116 nm and from 24 nm to 32 nm, respectively (See Fig. S2).”
4. L39-40 in P8: This sentence is unclear. Please revise for clarity.
Response: For clarification, the last two sentences are combined in the revision. It reads as “It is important to note that the observational data alone cannot provide evidence of any additional evolution of new particles and their additional contribution to Nccn after the new particle signal disappears, particularly considering the infrequent occurrence of the pre-existing particle growth.”
5. L25 in P10: It's odd to say "organic vapor was growing on the pre-existing particles". Use "condensing" instead.
Response: Corrected.
6. L3 in P15: There is no definition of PC1 or PC2. Please provide a definition.
Response: PC1 or PC2 represent the obtained two major principal component factors, which have been defined in the revision.
7. Fig. 2: The range of the Y-axis in panel (c) is too large, and it appears that there is no variation of CDNC with time. Please adjust the range of the Y-axis.
Response: Corrected.
8. Fig. 3-5: Please include a time series of the critical diameter (Dc) in the panel of PNSD.
Response: Added.
9. Fig. 7: The markers in panel (a) are too small to identify. Please increase their size.
Response: Corrected.
References:
[1]Ehn M, Thornton J A, Kleist E, et al. A large source of low-volatility secondary organic aerosol [J]. Nature, 2014,506(7489):476–479.
[2]Lance, S., Nenes, A., Medina, J., and Smith, J. N.: Mapping the operation of the DMT continuous flow CCN counter, Aerosol science and technology, 40, 242–254, 2006.
[3]Lathem, T. L. and Nenes, A.: Water vapor depletion in the DMT continuous-flow CCN chamber: Effects on supersaturation and droplet growth, Aerosol Science and Technology, 45, 604–615, 2011.
[4]Petters, M. D. and Kreidenweis, S. M.: A single parameter representation of hygroscopic growth and cloud condensation nucleus activity, Atmos. Chem. Phys., 7, 1961–1971, 2007.
[5]Pruppacher, H. R. and Klett, J. D.: Microphysics of Clouds and Precipitation: Reprinted 1980, Springer Science & Business Media, Berlin, Germany, 2012.
[6]Rollins A W, Browne E C, Min K E, et al. Evidence for NO control over nighttime SOA formation [J]. Science, 2012,337(6099):1210– 1212.
[7]Rose, D., Gunthe, S. S., Mikhailov, E., et al.: Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment, Atmos. Chem. Phys., 8, 1153–1179, 2008.
[8]Wex, H., McFiggans, G., Henning, S., and Stratmann, F.: Influence of the external mixing state of atmospheric aerosol on derived CCN number concentrations, Geophys. Res. Lett., 37, 10805, doi:10.1029/2010GL043337, 2010.
Response: The refences have been included in the revision to improve the quality of discussion.
Citation: https://doi.org/10.5194/egusphere-2023-539-AC1
-
RC2: 'Comment on egusphere-2023-539', Anonymous Referee #2, 20 Apr 2023
General comments
This paper is a case study discussing whether new particles generated by new particle formation (NPF) events grow and contribute to CCN based on two weeks of atmospheric observations. It is an important topic involving aerosol-cloud interactions, and the accumulation of such case studies is meaningful given the large spatiotemporal variability of aerosol properties. However, because of the following problems, a major revision of manuscript is needed before it can be accepted for publication.
The authors explain the case in several categories as to whether NPF events contribute to CCN number concentrations, but it feels illogical that the method of categorization is ambiguous and that even on the same day it is discussed in separate sections by supersaturation. In addition, cloud particle number concentrations were not measured in the observation site, so caution should be exercised when comparing calculated values with observed CCN number concentrations. The authors raise three research questions for introduction, so in conclusion part, it is necessary to answer these clearly. In particular, there is no description of the answer to the third research question: What implications do our findings have on knowledge gaps for CCN sources in NCP?
Other detailed points are listed below.
Specific comments
P2 L16
The lower limit of the specific particle size in this study should be mentioned in the method section.
P3 L28
Can the authors identify or estimate the period of cloud coverage at the observation site? Are data on relative humidity in the atmosphere, for example, available? Because cloud formation contributes to the deposition of pre-existing particles, it is considered an important data to deepen the discussion of NPF case studies.
P4 L12
It would be helpful if the authors include the piping diagrams for the four instruments in the supplement material.
P4 L22
How often switch the SS setting for CCNC? i.e., How many minutes each supersaturation setting lasted?
P6 L6
Yao et al. (2005) is missing in the reference list. It should be Yao et al. (2007) or Yao et al. (2010)?
P6 L26
It is questionable whether these satellite products reflect clouds that form at observation sites. In the introduction, the authors write that observation sites are often covered in morning mist, but can satellite products capture this phenomenon?
P6 L29
How many days back in the backward trajectory analysis?
P7 L18
As commented on the methodology, it is questionable whether the CDNC calculated based on the satellite products reflects the CDNC of the lower clouds at the observation site. As the CDNC strongly depends on the water vapor supersaturation in the cloud and, for that matter, the ascent velocity of the air mass, so it is also advisable to avoid comparing CCN concentration with CDNC without these discussions.
P7 L24
It’s not mandatory but there is a paper that summarizes CCN number concentrations (at 0.2% SS) at sites around the world (including mountain sites).
Schmale, J.et al.: Long-term cloud condensation nuclei number concentration, particle number size distribution and chemical composition measurements at regionally representative observatories, Atmos. Chem. Phys., 18, 2853–2881, https://doi.org/10.5194/acp-18-2853-2018, 2018.
P7 L31
This reviewer feels there is a weak basis for estimating the contribution rates of natural and anthropogenic sources. Since the observation site is in the vicinity of Beijing, it may depend strongly on the wind directions.
P8 L30
This paragraph barely discusses particle size. Whether or not a particle acts as a CCN depends strongly on the particle size, so it is natural that a particle in the newly formed nucleation mode does not contribute to NCN>100 orNCCN.
P9 L5
Wang Y., et al., 2020 seems to be missing in the reference list.
P9 L11
This reviewer doesn't quite understand the intent of discussing the events of July 1 in subsections by supersaturation. Shouldn't 3.4 and 3.5 be in the same section?
P9 L24
Because of the wide range of kappa value of organics, this is likely to depend largely on the type of organic matter.
P10 L36
The 3 July case also confirmed an increase in NCCN, but concluded that this was due to pre-existing particles. this reviewer is not sure what the difference is between this case (3 July) and the section 3.3 cases. Couldn't the case of 3 July be included in 3.3?
P14 L8
Same comment as P7-L18.
Section 3.7
This part is unique and interesting but somewhat speculative. Can't SIMS get information about inorganic materials on the particle surface? If possible, discussing it in conjunction with organic matter would make the discussion more robust.
P14 L34
This reviewer does not understand why the authors concluded that condensation of inorganic vapors is dominant in larger size particles (> 60 nm).
Figure 1
Airmass on the day of the NPF event is being transported from the north or northwest direction. Was there a difference in the airmass pathway on days with and without NPF events?
Figure 2a
It is very hard to see and misleading if separate axes for NCN and NCCN were used. Can it be represented by a single axis, with the vertical axis being the logarithmic axis? Also, making a separate graph of the activation ratio (NCCN/NCN) would be helpful to see how the cloud activation potential is.
Figure 2c
Please zoom in on the vertical scale to see the changes more easily. What is the meaning of the inset figure (CDNC in June and July)?
Figure 3b and 3e
Same comment as Figure 2a. Better to have NCN >100 andNCCN on the same axis.
Figure 4b and 4e
Same comment as above. Also, the captions of the figures should provide sufficient explanation to understand them. (without looking at the text or other figures)
Matters related notations
- “P” for p-values should be italicized in lower case letter "p".
- All variables (NCN, NCCN, κ....) in the manuscript and supplementary material should be italicized.
Citation: https://doi.org/10.5194/egusphere-2023-539-RC2 -
AC2: 'Reply on RC2', Xiaohong Yao, 19 Jun 2023
The authors explain the case in several categories as to whether NPF events contribute to CCN number concentrations, but it feels illogical that the method of categorization is ambiguous and that even on the same day it is discussed in separate sections by supersaturation. In addition, cloud particle number concentrations were not measured in the observation site, so caution should be exercised when comparing calculated values with observed CCN number concentrations. The authors raise three research questions for introduction, so in conclusion part, it is necessary to answer these clearly. In particular, there is no description of the answer to the third research question: What implications do our findings have on knowledge gaps for CCN sources in NCP?
Response: The authors carefully considered the comments and revised accordingly to enhance the overall quality of the presentation.
Specific comments
1. P2 L16 The lower limit of the specific particle size in this study should be mentioned in the method section.
Response: It is 5.6 nm and this has been clarified in the Experimental, i.e., “The FMPS was used to measure particle concentrations in the size range of 5.6 nm to 560 nm in 32 channels at a frequency of 1 Hz.”
2. P3 L28 Can the authors identify or estimate the period of cloud coverage at the observation site? Are data on relative humidity in the atmosphere, for example, available? Because cloud formation contributes to the deposition of pre-existing particles, it is considered an important data to deepen the discussion of NPF case studies.
Response: It is difficult to accurate estimate the period of cloud coverage at the mountain area since the cloud widely separated. In the revision, satellite cloud coverage and the ground-level relative humidity were presented in the SI to facilitate the analysis.
3. P4 L12 It would be helpful if the authors include the piping diagrams for the four instruments in the supplement material.
Response: Added.
4. P4 L22 How often switch the SS setting for CCNC? i.e., How many minutes each supersaturation setting lasted?
Response: As clarified in the revision, each supersaturation (SS) setting for the CCNC lasted for a duration of 5 minutes. However, an additional 5 minutes were used for switching SS at 1.0% SS to 0.2% SS due to the necessity of establishing supersaturation equilibrium.
5. P6 L6 Yao et al. (2005) is missing in the reference list. It should be Yao et al. (2007) or Yao et al. (2010)?
Response: Sorry for this. It has been corrected in the revision.
6. P6 L26 It is questionable whether these satellite products reflect clouds that form at observation sites. In the introduction, the authors write that observation sites are often covered in morning mist, but can satellite products capture this phenomenon?
Response: Satellite data reflect observations between 12:30-14:30. This has been clarified in the revision.
7. P6 L29 How many days back in the backward trajectory analysis?
Response: 24-hrs air mass back trajectories were used for this study. This has been added in the revision.
8. P7 L18 As commented on the methodology, it is questionable whether the CDNC calculated based on the satellite products reflects the CDNC of the lower clouds at the observation site. As the CDNC strongly depends on the water vapor supersaturation in the cloud and, for that matter, the ascent velocity of the air mass, so it is also advisable to avoid comparing CCN concentration with CDNC without these discussions.
Response: We agree the actual SS mainly determined by ascent velocity of the air mass is critical in continental atmospheres affected by anthrophonic air pollution to some extent. In the revision, we added related discussion accordingly. Moreover, the Nccn observed in the rural mountain atmosphere was considerably higher than the cloud droplet number concentrations derived from satellites. This suggests that the SS to form cloud droplets in the NCP atmospheres may be substantially smaller than 0.2%. Again, the grown new particles didn’t yield a detectable contribution to Nccn at 0.2% SS during all eight NPF events in this study. Thus, it is reasonably argued that the grown new particles might not act as Nccn in NCP atmospheres with actual SS largely smaller than 0.2%.
9. P7 L24 It’s not mandatory but there is a paper that summarizes CCN number concentrations (at 0.2% SS) at sites around the world (including mountain sites). Schmale, J.et al.: Long-term cloud condensation nuclei number concentration, particle number size distribution and chemical composition measurements at regionally representative observatories, Atmos. Chem. Phys., 18, 2853–2881, https://doi.org/10.5194/acp-18-2853-2018, 2018.
Response: Agree. We have added a comparison of the observations of CCN during the NPF events at the mountain sites in the revision.
10. P7 L31 This reviewer feels there is a weak basis for estimating the contribution rates of natural and anthropogenic sources. Since the observation site is in the vicinity of Beijing, it may depend strongly on the wind directions.
Response: This is a rough estimation. When the wind directions from the south and southwest, the contribution of Nccn from primary and secondary anthropogenic aerosols should be even larger. In the revision, we added “In Beijing, NPF events were also observed in polluted atmospheres with air masses originating from the south and southwest (Wu et al., 2007). In those cases, anthropogenic aerosols expectedly yield an even larger contribution to Nccn.”
11. P8 L30 This paragraph barely discusses particle size. Whether or not a particle acts as a CCN depends strongly on the particle size, so it is natural that a particle in the newly formed nucleation mode does not contribute to NCN>100 orNCCN.
Response: Agree. This is why we presented the case-by-case examination of the growth of new particles and the potential contribution to Nccn. To better service the reader, we added “ Moreover, the case-by-case examination of the growth of new particles and potential contribution to Nccnwill be presented in Section 3.2-3.3.”
12. P9 L5 Wang Y., et al., 2020 seems to be missing in the reference list.
Response: It is “Wan et al., 2020”. Corrected.
13. P9 L11 This reviewer doesn't quite understand the intent of discussing the events of July 1 in subsections by supersaturation. Shouldn't 3.4 and 3.5 be in the same section?
Response: The two subsections have been combined.
14. P9 L24 Because of the wide range of kappa value of organics, this is likely to depend largely on the type of organic matter.
Response: In the revision, it reads as “Assuming that the activated aerosols at 0.2 % SS were internally mixed and mainly composed of inorganic ammonium salts and organics (Petters and Kreidensohler, 2007; Rose et al., 2010, 2011), both of them likely yielded an appreciable contribution to the total mass concentration of the associated aerosols. However, the exact percentages relied on the κ values of various organics.”
15. P10 L36 The 3 July case also confirmed an increase in NCCN, but concluded that this was due to pre-existing particles. this reviewer is not sure what the difference is between this case (3 July) and the section 3.3 cases. Couldn't the case of 3 July be included in 3.3?
Response: We combined the two sections together in the revision.
16. P14 L8 Same comment as P7-L18.
Response: The paragraph has been revised as “We examined the satellite-derived CDNC over the mountain area during and after the NPF day. The values were 169 cm-3, 89 cm-3, and 101 cm-3 on 1, 2, and 3 July, respectively (See Fig. 2c). These values were approximately one order of magnitude smaller than the observed Nccn at 0.2 % SS on those 3 days. The CDNC is a strong function of the actual SS present in the atmosphere in addition to Nccn (Pruppacher and Klett, 2012). This large difference between the observed Nccn and satellite-derived CDNC implies that the actual SS in the atmosphere might be substantially smaller than 0.2%. Shen et al. (2018) recently reported the actual SS varying from 0.01% to 0.05% during fog events observed in the NCP. Thus, it is not surprising to find that only a small fraction of CCN could competitively capture water vapor to form cloud droplets during the study period (Shen et al., 2018; Jiang et al., 2021; Gong et al., 2023). Moreover, the CDNC during the period from 29 June to 14 July was 120±86 cm-3. The satellite-derived CDNC on 2–3 July were even lower than the average, suggesting that the NPF event was unlikely to have any influence on CNDC at such low actual SS.”
17. Section 3.7 This part is unique and interesting but somewhat speculative. Can't SIMS get information about inorganic materials on the particle surface? If possible, discussing it in conjunction with organic matter would make the discussion more robust.
Response: SIMS can detect inorganic ions, but interference is large because of the low ratios of signal and noise. This has been clarified in the revision.
18. P14 L34 This reviewer does not understand why the authors concluded that condensation of inorganic vapors is dominant in larger size particles (> 60 nm).
Response: The sentence has been revised as “when semi-volatile organic and inorganic vapors may have overwhelmingly condensed on the sized particle surfaces and covered up the high-molecular-weight organic fragment signals.”
19. Figure 1 Airmass on the day of the NPF event is being transported from the north or northwest direction. Was there a difference in the airmass pathway on days with and without NPF events?
Response: Wu et al. (2007) reported statistical analysis of NPF events using a year-long measurement in Beijing, NPF events were associated with low RH and sunny conditions. The air masses from the north and northwest usually carry dry and clean air, favoring the occurrence of NPF events. Non-NPF events were usually associated with strong condensational sink or absence of sunny conditions. The reference has been added and summarized to support the analysis in the revision.
20. Figure 2a It is very hard to see and misleading if separate axes for NCN and NCCN were used. Can it be represented by a single axis, with the vertical axis being the logarithmic axis? Also, making a separate graph of the activation ratio (NCCN/NCN) would be helpful to see how the cloud activation potential is.
Response: Revised.
21. Figure 2c Please zoom in on the vertical scale to see the changes more easily. What is the meaning of the inset figure (CDNC in June and July)?
Response: Revised. The observational period is too short to be representative. The two month-long data of CDNC can better reflect its range. This has been clarified in the revision.
22. Figure 3b and 3e Same comment as Figure 2a. Better to have NCN >100 and NCCN on the same axis.
Response: Revised accordingly.
23. Figure 4b and 4e Same comment as above. Also, the captions of the figures should provide sufficient explanation to understand them. (without looking at the text or other figures)
Response: Revised accordingly.
Matters related notations
1. "p" for p-values should be italicized in lower case letter "p".
Response: Corrected.
2. All variables (NCN, NCCN, κ....) in the manuscript and supplementary material should be italicized.
Response: Corrected.
Citation: https://doi.org/10.5194/egusphere-2023-539-AC2
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 429 | 158 | 24 | 611 | 47 | 10 | 12 |
- HTML: 429
- PDF: 158
- XML: 24
- Total: 611
- Supplement: 47
- BibTeX: 10
- EndNote: 12
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Xing Wei
Yanjie Shen
Xiao-Ying Yu
Huiwang Gao
Ming Chu
Yujiao Zhu
Xiaohong Yao
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1648 KB) - Metadata XML
-
Supplement
(695 KB) - BibTeX
- EndNote
- Final revised paper