Cloud condensation nuclei at two Finnish sub-Arctic sites: long-term observations of aerosol activation, predictions based on a simple semi-empirical parameterization, and comparison to global climate model simulations

Stapleton, Charlotte G.; Sipilä, Mikko; Chandrasekharan, Arundathi; Merikanto, Joonas; Neitola, Kimmo; Nieminen, Tuomo; Makkonen, Risto; Joutsensaari, Jorma; Virkkula, Aki; Hyvärinen, Antti-Pekka

doi:10.5194/egusphere-2026-5001

Preprints

https://doi.org/10.5194/egusphere-2026-5001

Preprints

11 Sep 2026

| 11 Sep 2026

Status: this preprint is open for discussion and under review for Atmospheric Chemistry and Physics (ACP).

Cloud condensation nuclei at two Finnish sub-Arctic sites: long-term observations of aerosol activation, predictions based on a simple semi-empirical parameterization, and comparison to global climate model simulations

Charlotte G. Stapleton, Mikko Sipilä, Arundathi Chandrasekharan, Joonas Merikanto, Kimmo Neitola, Tuomo Nieminen, Risto Makkonen, Jorma Joutsensaari, Aki Virkkula, and Antti-Pekka Hyvärinen

Abstract. In this study, we present long-term observations of cloud condensation nuclei (CCN) at two locations in the Finnish sub-Arctic: Pallastunturi and Värriö. Both stations belong to the ACTRIS research infrastructure. At Pallastunturi, the interactions between aerosols and clouds have been measured in situ at a fjell top, 565 m a.s.l. since 2005. From Pallastunturi data, a simple parameterisation was developed to predict the activation diameter D₅₀ and subsequent CCN- concentrations, based on air mass origin and temperature. The parameterization was applied to aerosol size distribution measurements conducted at Värriö from 1998 to 2021. The average annual CCN concentrations were 140 ± 40 # /cm³ at Pallas and 170 ± 50 # /cm³ at Värriö. The CCN- concentrations exhibited statistically significant decreasing trends at both stations: for Pallas -2.5 CCN /cm³ / year, and for Värriö -4.2 CCN /cm³ / year. The decrease was most pronounced during February and March, i.e. during the Arctic Haze period. We also compared the results with EC-Earth3 global climate model simulations. Using two measurement sites separated by ~230 km reduces the uncertainty in the spatial representativeness of observations and provides a robust basis for climate model evaluation. EC-Earth3 predicted the measured CCN- concentrations fairly well at 0.2 % supersaturation, reproducing the observed long-term trend. However, the modelled seasonal variation was slightly different, predicting higher concentrations during winter time (November to April). The study demonstrates the value of long term in-cloud aerosol measurements in validating and improving climate predictions.

How to cite. Stapleton, C. G., Sipilä, M., Chandrasekharan, A., Merikanto, J., Neitola, K., Nieminen, T., Makkonen, R., Joutsensaari, J., Virkkula, A., and Hyvärinen, A.-P.: Cloud condensation nuclei at two Finnish sub-Arctic sites: long-term observations of aerosol activation, predictions based on a simple semi-empirical parameterization, and comparison to global climate model simulations, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2026-5001, 2026.

Received: 19 Aug 2026 – Discussion started: 11 Sep 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 2549 KB)

Supplement (1222 KB)

Download & links

Charlotte G. Stapleton, Mikko Sipilä, Arundathi Chandrasekharan, Joonas Merikanto, Kimmo Neitola, Tuomo Nieminen, Risto Makkonen, Jorma Joutsensaari, Aki Virkkula, and Antti-Pekka Hyvärinen

Status: open (until 21 Nov 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-5001', Anonymous Referee #1, 11 Oct 2026 reply
In this manuscript, authors studied long-term CCN concentrations at two sub-Arctic stations in Finland, Pallas and Värriö. The topic is important and it is suitable for ACP. The dataset is very rare and valuable. As is known, multi-year in-cloud size-resolved activation measurements at ground stations are very few in the world. So the 16-year twin-inlet record at Pallas is an important contribution to the community. The D₅₀ statistics (Figs. 6–8 and Table S3), the long-term size distribution trends and the open archived time series are main strengths of this work. I also appreciate that authors discussed some limitations of their method honestly (Sect. 3.5.2, L635–643). However, I have several major concerns about the interpretation, the robustness of trends and the model comparison. I think these problems should be solved before the conclusions can be accepted.
Major comments
The quantity calculated by Eq. 3 is not CCN in usual meaning, that is, particles activated at a given supersaturation. In fact, it is an estimation of the particle number which would activate in clouds like the clouds observed at Pallas, based on binned D₅₀. In the manuscript, this quantity is called CCN, N_CCN and N_d in different places (L143–144, L490, L598). It is also described as “observations” and “directly measured data” (abstract L19; L98–99). But except the in-cloud hours at Pallas, it is a parameterized product, not observation. This point is important for two reasons. Firstly, readers will compare these numbers with CCN counter dataset. Secondly, the model comparison is based on treating this quantity as same as modelled CCN at 0.2% SS. Therefore, it is suggested that authors: (1) use one clear and consistent term, such as “estimated activated particle number” or “N_CCN,eff”; (2) explain clearly what this quantity means for cloud-free hours and for Värriö; (3) revise the wording in title and abstract accordingly.

From Fig. 8a, D₅₀ increases very strongly with temperature, from about 90 nm in coldest class to about 130–190 nm in warmest class. At the same time, this region warmed significantly during study period, as authors also mentioned (Rantanen et al., 2022). So warming trend will put more hours into warmer classes. Then the assigned D₅₀ becomes bigger and derived CCN becomes lower, even the aerosol itself has no change. The change of air mass sector frequency may have similar effect. Therefore, the reported CCN trends are mixed by two effects: change of size distribution and change of classification variables. I strongly suggests authors to separate these two effects, for example: (1) recalculate the trends by using fixed D₅₀ (e.g. 100 nm, authors already showed it has similar performance); (2) keep the temperature and sector distributions at climatological values; (3) report the trends of temperature class frequency and sector frequency during study period.

In section 3.4, several parts need to be improved
The description of model configuration is not enough. More detailed information should be added about: (a)emission inventory and its base years; (b)how six ensemble members are different from each other; (c)which model level is used, because Pallas is at 565 m; (d)concentrations are at ambient condition or standard condition.

Authors chose 0.2% SS because it agrees with observations, and 1.0% SS is shown but not used in discussion. More physical way is to convert observed D₅₀ into equivalent critical supersaturation by κ-Köhler theory with reasonable κ range, then do comparison at this SS.

In Fig. 14a2 and b2, observed seasonal cycles seem to be monthly means, and peak is in February. But in Fig. 11 (medians), peak is in May–July. So the conclusion about model seasonal cycle depends on which statistic is used. Please compare them with same statistic and same period, or show both of them.

Minor comments:
“the longest CCN time series in the Arctic” (L109–110, L595): This claim is not consistent in manuscript, because L396 says “one of the longest … from sub-Arctic regions”.

Please check the cut-off of about 7 µm for the interstitial inlet, and discuss the possibility of sampling small droplets.

Equations lack explanation of each item. Please define each item and give its unit.

3. Upper limit of integration is upper bound of DMPS, not infinity. Please state it clearly and estimate contribution of particles above this limit.

L177–181. Please explain why such loose negative thresholds (−1 and −2) were chosen.

The air-mass sector geometry is confusing. In figure, Arctic sector seems to have latitude-type boundary, but in Table S2 sectors are defined by bearing angles (Arctic is 270–90° through north). Please make them consistent. Please also give sector definition for Värriö clearly, and explain why different boundaries are used for table derived with Pallas boundaries (L243–244).

Reply
Citation: https://doi.org/10.5194/egusphere-2026-5001-RC1

Charlotte G. Stapleton, Mikko Sipilä, Arundathi Chandrasekharan, Joonas Merikanto, Kimmo Neitola, Tuomo Nieminen, Risto Makkonen, Jorma Joutsensaari, Aki Virkkula, and Antti-Pekka Hyvärinen

Supplement

https://doi.org/10.5194/egusphere-2026-5001-supplement

Data sets

Data for manuscript:"Cloud condensation nuclei at two Finnish sub-Arctic sites: long-term observations of aerosol activation, predictions based on a simple semi-empirical parameterization, and comparison to global climate model simulations" Charlotte G. Stapleton, Mikko Sipilä, Arundathi Chandrasekharan, Joonas Merikanto, Kimmo Neitola, Tuomo Nieminen, Risto Makkonen, Jorma Joutsensaari, Aki Virkkula, and Antti-Pekka Hyvärinen https://doi.org/10.57707/fmi-b2share.kg4ej-9sy11

Charlotte G. Stapleton, Mikko Sipilä, Arundathi Chandrasekharan, Joonas Merikanto, Kimmo Neitola, Tuomo Nieminen, Risto Makkonen, Jorma Joutsensaari, Aki Virkkula, and Antti-Pekka Hyvärinen

Viewed

Total article views: 628 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
398	171	59	628	76	72	61

HTML: 398
PDF: 171
XML: 59
Total: 628
Supplement: 76
BibTeX: 72
EndNote: 61

Views and downloads (calculated since 11 Sep 2026)

Month	HTML	PDF	XML	Total
Sep 2026	300	138	35	473
Oct 2026	98	33	24	155

Cumulative views and downloads (calculated since 11 Sep 2026)

Month	HTML	PDF	XML	Total
Sep 2026	300	138	35	473
Oct 2026	98	33	24	155

Viewed (geographical distribution)

Total article views: 628 (including HTML, PDF, and XML) Thereof 628 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 22 Oct 2026

Short summary

Article presents observations of cloud condensation nuclei (CCN) in the Finnish sub-Arctic. At Pallastunturi, aerosol-cloud interactions have been measured since 2005. Based on these measurements, a simple model was developed to predict the CCN at Värriö for 1998–2021. The CCN concentrations have decreased about 50 % during the measurement period, especially during early spring, when polluted air from the south reaches the Arctic. A global climate model reproduced the observed long-term trend.


Total:	0
HTML:	0
PDF:	0
XML:	0