the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Drivers governing the seasonality of new particle formation in the Arctic
Abstract. New particle formation (NPF) is the phenomenon wherein gaseous precursors form critical clusters of barely a few nanometres in diameter, after which, under favourable conditions these particles can grow to climate-relevant sizes. Here we present measurements from 2022 to 2024 of particle and ion number size distributions from the Zeppelin Observatory (ZEP), an Arctic research station situated on the western edge of Svalbard. NPF events begin in April and continue occurring into November. The events at the start of the NPF season (i.e. April/May) are considerably stronger (i.e. a larger production of nucleation mode particles). The peaks in NPF strength coincide with peaks in the solar insolation experienced by arriving air masses. During the summer period NPF events occur on 20–40 % of days each month, however, there is a consistent decline in June. We show that the combined influence of solar radiation and the surface area of pre-existing aerosols (i.e. condensation sink, CS) are strong predictors for the likelihood of NPF. We develop a simplified predictive model which matches the frequency of NPF events identified via the classification schemes used in this study. We show that NPF events occur during the polar night (i.e. when the Sun does not pass above horizon), and speculate that these events are linked to high altitude air masses. Furthermore, we detail the likely geographic origins of nucleation within the Arctic, as measured at ZEP. We show that NPF events are considerably more likely to originate from the marine regions towards the west of Svalbard, particularly the Greenland Sea which presented the greatest likelihood that arriving air masses from this marine region would be linked to an NPF day. We also remark on the proportion of the Aitken mode particles within the Arctic that could originate from NPF; we show that NPF events lead to an increase in the number of Aitken mode particles. We measure over 50 NPF events where the nucleation mode particles grew beyond 25 nm, a diameter representing the minimum activation diameter for particles to act as cloud condensation nuclei. Overall, we present a concise picture of the lifecycle of nucleation mode particles in the Arctic, including the effect wet scavenging has in reducing the condensation sink, which in turn encourages NPF events to occur.
- Preprint
(1555 KB) - Metadata XML
-
Supplement
(1360 KB) - BibTeX
- EndNote
Status: open (until 30 Apr 2025)
-
RC1: 'Comment on ar-2025-11', Anonymous Referee #1, 13 Apr 2025
reply
The comment was uploaded in the form of a supplement: https://ar.copernicus.org/preprints/ar-2025-11/ar-2025-11-RC1-supplement.pdf
-
RC2: 'Comment on ar-2025-11', Anonymous Referee #2, 14 Apr 2025
reply
The study "Drivers governing the seasonality of new particle formation in the Arctic" by Heslin-Rees et al. presents an extensive dataset of particle and ion measurements in the Arctic atmosphere.
With statistical analysis of the dataset, the authors characterize the frequency of new particle formation and try to identify the circumstances of its appearance at the Zeppelin station on Svalbard. Further, they analysed the air mass origin and linked it to the likelihood of NPF events.
I would recommend the study to be published after addressing my minor concerns.
Is it possible to quantify how often ZEP was inside the boundary layer and how often it was above? Line 463 and following: Might the boundary layer height have an impact on the NPF timing and in general on the frequency as well?
The GR during July 2024 (if I read the month from the x-label correctly) is considerably higher than during other months. Do the authors have an idea why this might be the case?
I believe I do not quite understand the two values for the rolling mean - e.g. in Fig. 9 the authors show the 30 day rolling mean - what does the min. window 7 d mean exactly? Are periods with less than 7 days excluded?
I would recommend rounding the delta N boundary of 2.82 to 2.8 as the accuracy of the NAIS, to my knowledge, does not capture 1e-2 nm.
The authors use the vapor concentrations contributing to NPF in chapter 3. Those should also be mentioned in the introduction, so that the readers are on the same page and understand why they are being part of this discussion. Generally, the contributing vapors for NPF in the Arctic atmosphere should be discussed a little more in detail even though the vapors were not measured here.
line 337: DMS is not only a precursor to sulfuric acid, but also methanesulfonic acid.
line 416: This is more a question of curiosity: Do the authors think that dilution due to stronger insolation and thus elevated boundary layer heights during summer might play a role in the reduced NPF intensity?
line 500: I believe the sentence "HOM concentrations are very low during spring and increase in May, however it still shows their importance in contributing to Cv." needs a bit more elaboration. The abbreviation HOM is not yet mentioned or explained in the manuscript. Why do the authors think they are contributing to the Cv? Because the sum of IA, MSA and SA are not explaining the Cv alone?
line 540: The theory that during autumn months the nucleation pathway is rather dominated by organics which might show in the cluster ion relation sounds plausible. However, I would also keep in mind that during these months the NPF frequency was also considerably lower and I am not sure how much weight one can give the 5, or 3 events, respectively, in October / November.
line 577: What are the correlations of 2022 and 2023 and how do they compare to Park et al. (2018)?
line 696 and following: were there any CCN instruments on site as presented in Karlsson et al., 2021 which could be supporting the analysis and impact on CCN production via NPF in this paper?
Fig. 1: I recommend having more details in the figure caption to guide the reader to understand more quickly, which y-axis belongs to which parameters.
Is it possible that 31 days were included in June 2023? From the bar plot it reads like the total number of days included are 31 as during July and August
Fig. 2: I suggest using more distinct colors.
Fig. 4: I suggest adding a legend for the figure (mean, median) even though the box-whiskers plot makes them self explanatory. The x-axis label is a bit difficult to read, it is not fully clear which month belongs to which tick-mark.
Fig. 8: Do I understand this correctly, that the colorbar (panel a) is the count of negative ions? Which sizes are shown here? Do the authors have an explanation for the high concentrations of ions during low solar flux and low CS, e.g. is it linked to snow or cloud events, or are these days excluded here?
Fig. S16: The y-axis label seems to be wrong, I believe it should read nm, not meter.
Technical remarks:
Line 110: "both" is repetitive
Line 570: "at ZEP" is repetitive
Line 720: typo: "can contribute"
Line 751: typo: "rapid"
Citation: https://doi.org/10.5194/ar-2025-11-RC2
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
96 | 30 | 3 | 129 | 15 | 4 | 4 |
- HTML: 96
- PDF: 30
- XML: 3
- Total: 129
- Supplement: 15
- BibTeX: 4
- EndNote: 4
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1