the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 May 2026
Isolating the influence of aerosols on Arctic cloud radiative effects during a polluted warm air mass intrusion
Abstract. Arctic warm air mass intrusions are key to the region's energy balance because they transport large amounts of heat and moisture from lower latitudes. The resulting changes in heat and moisture content influences cloud properties, the thermodynamic structure of the boundary layer, and surface radiation. Warm air mass intrusions can also carry significant amounts of aerosols into the Arctic, including aerosols from anthropogenic sources at lower latitudes, which could influence the radiative impacts of such events. In this study, we examine the role of aerosol transport during a warm air mass intrusion that occurred in the central Arctic in April 2020. We use a version of the regional chemistry-climate model WRF-Chem adapted for Arctic conditions (WRF-Chem-Polar), to investigate the radiative impact of aerosol-cloud interactions associated with the event. We isolate the effects of the high aerosol burden by running the model with and without anthropogenic emissions. Anthropogenic emissions increase cloud droplet number concentration by 117% and liquid water content by 52%. However, the net surface radiative impact of these aerosol-cloud interactions is limited over sea ice. The high albedo of the underlying sea ice limits shortwave cloud cooling, while the longwave effects of cloud perturbations are small. Over open ocean regions, the surface radiative impacts of the aerosols are stronger. Overall, these results show that the net effect of an extreme aerosol transport event is sensitive to the season in which the event takes place, due to strong dependence on the surface state and the background Arctic haze conditions.
- Preprint
(14038 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-2369', Anonymous Referee #1, 22 Jun 2026
-
RC2: 'Comment on egusphere-2026-2369', Anonymous Referee #2, 13 Jul 2026
The topic of this manuscript is the role of aerosols during a warm air mass intrusion (WAI) event over the Arctic and the associated impacts on clouds and surface radiation. It addresses an interesting and timely topic, and the study is well aligned with this journal’s scope.
Overall, I liked this paper, and think it will be a useful addition to the literature. However, I do think it has some important weaknesses that should be addressed prior to publication.
Most importantly, the current analysis puts a lot of emphasis on aerosol impacts on cloud and radiative impacts of the event, but the evaluation and validation in the study focuses mostly on aerosols, T, and Qv at the surface. Arctic clouds and boundary-layer structure are known to be very challenging to simulate, and matter a lot to cloud radiative effects. Thus, the paper would be strengthened by a more explicit assessment of whether the model captures the WAI event itself (timing, location, intensity) and whether key meteorological and cloud properties (e.g., cloud fraction, height) are realistically represented. If such evaluation has already been conducted in prior work for this event and model configuration, it should be clearly cited and summarized.
A few other points:
1) The manuscript notes the limitations of in situ observations for spatial context, which is why they say they need the model output. But as mentioned previously, there are known model challenges with correctly simulating clouds in the Arctic, and this will matter a lot to their cloud radiative effect results. That, plus the lack of cloud/WAI event model validation in the paper, pushes me to recommend that the authors add some remote sensing analysis, which I feel could substantially strengthen the paper. Even a qualitative comparison of event footprint, timing, etc. would add credibility (e.g., see https://go.nasa.gov/4f7JWb6). Similarly, cloud base height could be better constrained using lidar/cielometer from MOSAiC rather than relying only on surface T and Qv proxies.
2) I also felt that the aerosol validation was weaker than it should be. The manuscript validates aerosols primarily using surface measurements, but the aerosols most relevant for cloud interactions are those at cloud level. Over sea ice, strong temperature inversions can decouple surface and cloud-layer aerosol populations, meaning surface aerosol may not be representative of in-cloud aerosol. At minimum, the authors should discuss this limitation explicitly, but I feel just doing that alone would leave this a weaker paper. Thus, I recommend that the authors take advantage of the MOSAiC vertical aerosol constraints from balloon and lidar. I also recommend that they plot in Fig. 4 the values at cloud altitude from WRF-Chem-Polar. That will help demonstrate if the timing is similar at the altitudes that matter most.
3) The two-scheme approach is very commendable, but the manuscript’s current framing suggests that agreement in radiative conclusions across schemes implies robustness in their results that meteorology is more important than aerosols. I don't disagree with the meteorology being more important. However, Figure A1 appears to show near-zero aerosols in the Morrison configuration, which would naturally minimize aerosol–cloud effects. In that case, similarity in radiative response just reflects the fact that aerosol perturbations are effectively absent, rather than a meaningful confirmation of aerosol insensitivity. The authors may want to more explicitly state that the Morrison run may not provide an informative aerosol sensitivity test given that aerosols are unrealistically low. I'd honestly just take that extra analysis out, but I leave that decision up to the authors and editor.
4) The paper would benefit from more brief context (maybe a paragraph or so?) on how common WAIs are, their seasonality, and whether this event is representative of WAIs if they have that information. Additionally, the the conclusions should more clearly acknowledge that a) the study focuses on one extreme, moisture/heat-rich event, and does not explore cases where aerosols are transported without similarly strong thermodynamic forcing, and b) aerosol impacts on mixed-phase and ice-cloud processes are not fully explored, despite their potential importance in the Arctic.
More minor comments:
L. 16: “The Arctic is warming four times faster than the global average rate (Rantanen et al., 2022), a consequence of local climate feedbacks amplifying the effects of local radiative forcing and of warm air transported from the lower latitudes (Stuecker et al., 2018).”
I know it is probably a smaller factor, but isn’t greenhouse gas warming also a factor? That’s not necessarily a local feedback and it isn’t just from transported air.L. 24: “some studies have indicated a connection between springtime Arctic events” it would be helpful to be more clear about what kind of events is meant here (e.g., number or intensity of WAI events)
l. 35: “strong aerosol transport associated with WAIs can lead to extreme deposition events of light-absorbing impurities onto ice and snow (Thomas et al., 2017)” I think I’d feel more comfortable with a different word than “extreme” here, since the authors of that Thomas et al. paper did not use that word and my reading of that paper does not support that word either. Maybe “concentrated”, “considerable” or “significant” would be more appropriate?
l. 37: maybe mention that this is the same event that the authors will be describing in this paper, and how specifically it differs from those other papers, especially as these papers are brought up again in the conclusions.
Fig. A1: The caption says, "as in Fig. 1", but they use a different color scheme than Fig. 1. I recommend that they maybe use the same color scheme or mention that the blue is for the Morrison scheme.
Fig. 5 (g-i). Is this net (LW+SW) CRE? Please specify in the figure/caption.
l. 197: "Regions of increased cloud fraction do lead to surface longwave warming (not shown), however, these regions that experience longwave warming are not widespread and are typically offset by shortwave cooling."
It seems that Fig. 5 shows that there is some important regional variability, which should be mentioned here. It might also be helpful to show regions of increased cloud fraction in Fig. 5.L. 200: "Distributions of cloud droplet number concentration and cloud fraction are calculated for altitudes between 0 and 2 km, corresponding to the altitudes where clouds are simulated in the model" Is this consistent with what the cielometer/lidar indicated was the cloud base?
l. 225/ l. 249: The role of sea ice and season is important; however, lighting conditions (polar night vs daylight) also strongly modulate radiative impacts. I suggest the authors briefly discuss how results might differ during polar night and how often WAIs occur in darkness vs daylight.
L. 261: The statement appears too general given that the study focuses on a single extreme event with strong moisture and heat transport. I suggest that the authors rephrase to avoid overgeneralization and to clarify that the results may be very different if aerosols are introduced without the same moisture and heat content
L. 268: "Comprehensive modelling and observational approaches are needed to understand this evolving system, capable of capturing the interacting effects of sea ice loss, shifts in large-scale dynamics, and changing aerosol emissions." Aerosol loss processes and mixed and ice phase cloud processes and their interactions with aerosols are also important and can also change with climate.
Citation: https://doi.org/10.5194/egusphere-2026-2369-RC2
Model code and software
WRF-Chem-Polar Louis Marelle et al. https://doi.org/10.5281/zenodo.19736986
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 219 | 62 | 15 | 296 | 17 | 17 |
- HTML: 219
- PDF: 62
- XML: 15
- Total: 296
- BibTeX: 17
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review for: ‘Isolating the influence of aerosols on Arctic cloud radiative effects during a polluted warm air mass intrusion’ by Price et al. 2026
Price et al. have presented a modelling case study of a warm air mass intrusion observed in the Arctic by the MOSAiC campaign. They have performed simulations with and without anthropogenic pollution to investigate the role of aerosol transport within these warm air mass intrusions into the Arctic. They find that the aerosol do have an impact on the cloud droplet number and liquid water, but the effects on the radiative balance are limited, in part due to the sea ice state. The article is well written and provides an interesting insight into how human activity further impacts the Arctic, which provides important context for the region in a changing climate. All my comments are minor, and once addressed I recommend publication of this article.
Minor comments:
Line 12: I fully appreciate that the seasonality of the sea ice state makes this a reasonable conclusion, but I would rephrase this as a ‘suggestion’ or ‘implication’ rather than a proven result.
Line 24: Add ‘WAI’ between ‘Arctic events’ to be specific.
Line 25: Every year has a sea ice minima, so I would rephrase this to say ‘years in which the sea ice is below average’ or something along those lines.
Figure 1 Caption: Could you make the domain boundary lines a different colour/style to the inset boundary line, as when I first looked at the plot (and read the caption) I thought there might have been a nested model domain shown by the inset. Also could you make the inset smaller in area, as it is pretty hard to see the ship track/time because it is very small within the domain.
Line 94: Is it necessary to take the 9 nearest grid points when your resolution is 100km? I guess you are doing it to provide a level of uncertainty, but at that resolution you could easily be taking in other unrelated airmasses? Could this by why you did not see much impact from your aerosol changes? Did you compare the 9 grid mean to the nearest grid, or an interpolated point location?
Figure 2: Could you overlay the MSLP to give an idea of the synoptic settings that transported the pollution?
Line 105: Did Dada et al. do back trajectory analysis? Could you show it too? (adding the MSLP would give similar info though – so I understand if not).
Line 114: Do you have any ideas as to why the model overestimated in the earlier period? Is it a different airmass with different aerosol sources?
Line 167: I would clarify that only the Thompson results are described above/in Figure 4. I first read this to assume that the differences between the Thompson/Morrison were described above/in Figure 4.
Section 3.3: I was a little bit confused here. My understanding of the Morrison microphysics (and I assume the Thompson) is that it just handles the cloud microphysics and not aerosol microphysics (e.g. sources, growth, which is handled by MOSAIC). So why are there differences in the aerosol between the two microphysical schemes? Is it because of wet deposition? Do these schemes handle detrainment of aerosol when clouds evaporate? Or is the Morrison microphys not coupled to an aerosol scheme (if so you should state as that would not really be a fair comparison)?
Section 3.3/Figure 4/Figure A1: Could you not just put the Morrison results onto the same Figure so we can directly compare? I would make it a lot easier to see/reduce your figure count.
Line 219: add ‘during the warm air intrusion event’ at the end of sentence here to be specific.
Line 249: Where you say the ‘surface radiative impacts…. strongly dictated by seasonal conditions…’ I think you need to need to say ‘are likely’ as your study does not show that exactly. You also say ‘and background aerosols’. Do you mean the seasonality of background aerosol here too? I was thinking about that – if there is no sea ice, then there might be more sea spray/sulfate aerosol - so that would also change the background state and might also change your results?
My other thought here: Do warm air intrusions occur though the warmer months too (sorry – I am a bit ignorant about Arctic meteorology!)? If they don’t then the seasonality of sea ice doesn’t matter as much?
Figure A2: it seems the violin plots for the CDN background conditions are missing.