The paper named Secondary aerosol formation in marine Arctic environments: A model measurement comparison at Ny-Ålesund

made for an interesting read including a detailed size resolved aerosol model and and good dataset. My impression is however that sometimes the authors have a tendency to  explain everything based on the novelty of the apporach compared to more traditional works.

The most serious concern to me possibly undermining some of the conclusions is that although the aerosol model is very complex with respect to MSA formation and aqueousformation as well as NH3 related new particle formation (NPF) it seems to lack a number of other basic reactions. It is possible that this information can be found in the referenced material, but if so I think this information can be included easily without.

This includes: Aqueous phase production of SO4 from SO2 , e.g. Seinfeld and Pandis (1997) (H2O2 and O3). E.g The last lines of the abstract point to MSA only for the particle growth.

Gas phase production of SO2 from DMS (Possibly unclear description in the text)

NPF from other mechanisms, binary or in combinations with organic low volatile compounds, e.g. Vehkamäki et al. (2002), Paasonen et al. (2010). Are these processes found or assumed to be unimportant.

As these are the "traditional" explanations for the reaction pathways and growth I think it is needed to show the relative fraction of these processes compared to new. SO I would like to see the fraction of DMS going to as oppsed to SO2, and aqueous phase SO2-SO4 as opposed to DMS-MSA(aq)

I have also a general question about the supersaturation experiments.

"Increasing S to 0.8% increases

accumulation mode particles, since more particles with Sc < S are activated to cloud droplets

(Aitken mode concentration decreases with respect to BaseCase simulations, since more smaller particles are activated into cloud droplets)."

Even for S of 0.8 % the critical diameter is more than 50 nm,  actually I think the observed local number minimum may work as a proxy for activation size.

Figure 8 however show an impact already at 20-30 nm. Is it possible that the reason for the reduction is not from direct activation of the particles but rather an increased cloud droplet surface area, i.e the inverse of the explanation given for cloudoff.

Figure S8 shows as far as I  can see the same lines for NPFOff and no disolution. Does NPF in the model depend on cloud processes?

One last question question with respect to assumptions about the relatively high number of 3-12 nm shown in observations figure S8, compared to the apparently? lower number found in figure 8. It is hard to compare visually with log-normal distributions, but still figure 8 does seem to give a much lower number of for the 3-12 nm size.

General comment:

It is easy for the reader to mix up PM and PN. I think it is common to use only N for the number but that is more of a suggestion and I leacve that to the authors.

Last section of abstract, as discussed above. Are all processes included?

Details:

--"size < 12nm" Please define size as radius or diameter the first time you define it. Later on size is fine as long as it retains the same definition.

--Measurement period. Did you experiment with classifying the trajectories when discussing the results. Eg. a western airflow is expected to have both marine and more clouds than a easterly flow so the trajectories with marine caracteristics may have experienced lower emissions than continental pathways so the size distribution may also be caused by different emissions, not only the cloud processing.

Section 2.2 Sea surface temperature above 0. --> No trajectories from areas with sea-ice so no negative SST?

Table 1: Please make the table smaller and more readable. Also I think it is useful for the reader if you also refer to the table in the results section.

Table 3. For readability please consider using the same order of species in the text as in the table

line 678. Sea-spray aerosols are not scavenged --> Why?

721: Typo tin-cloud --> in-cloud

The manuscript describes a modelling study using the ADCHEM model updated with a complex MSA-halogen mechanism in simulating new particle formation in the Arctic, and the comparison using observation from Arctic sites and campaigns. The manuscript is very-well written and easy to follow, and the results nicely support the conclusions. I find the manuscript suitable for publication in ACP, after addressing the comments and corrections I have listed below.

Introduction

What is the aim of the study and the hypothesis? A paragraph at the end of the introduction section on this could be useful.

Line 57: Recent studies (e.g. Lenssen et al., 2019) suggest the warming rate is up to a factor of three).

Lenssen, N. J. L., Schmidt, G. A., Hansen, J. E., Menne, M. J., Persin, A., Ruedy, R., and Zyss, D.: Improvements in the GIS- TEMP Uncertainty Model, J. Geophys. Res.-Atmos., 124, 6307– 6326 https://doi.org/10.1029/2018jd029522, 2019.

Line 72: Correct as Arrigo and van Dijken (2015), and throughout the manuscript.

Methods

Line 148: Why is this period chosen? Is it a period of observed NPF or is it the ALANDIA campaign? How does the model behave in non-NPF periods?

Line 159: Are there also PM2.5 chemical composition measurements available?

Are biomass burning emissions not taken into account?

Results

Figures 1 and 2. Is it possible to show a third panel where obs-model is shown? It would be useful for the reader to compare visually what is written in the text.

Line 369: Is it Figure 1a or Figure 2a?

Line 449: Does the model take into account the sulfate production via in-cloud oxidation of SO2? The model discrepancy highlighted in 466-471 can be a result of this process not taken into account.

Figure 4 could be considered to be removed as the text does not provide any discussion on the temporal variation or magnitude compared to earlier measurements. I would rather provide a figure (a bar blot or box whisker) comparing the simulations with the measurements.

Figure 5b and Table 2 shows the same information in different ways, which is well described in the text. I would move one of them to the supplement.

Line 573: Correct the sentence.

Line 575: Are the NMB values representing under 200 m asl or the whole vertical extent?

Can the good agreement in the 200-600 m compared to the first 200 meter imply that the model is doing poorly close the local sources, which can be attributed to the uncertainty in the emissions, while 200-600 meters represent more transported particles and the model captures this transport?