the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Aug 2026
Contrasting Arctic Cirrus Clouds: Properties and Processes from Observations and Simulations
Abstract. Cirrus clouds in the Arctic play a central role in the regional radiation budget, yet their formation pathways and properties remain poorly constrained, especially during Warm Air Intrusions (WAI). In this study, we analyse and contrast cirrus observed during WAI events (WAI cirrus) with those formed under undisturbed Arctic Conditions (AC cirrus) using remote sensing measurements from the HALO–(AC)³ campaign, combined with backward trajectory modelling with CLaMS-Ice and LAGRANTO. This allows us to connect the optical, macro- and microphysical properties measured via lidar and radar to the cloud origin and dynamics during their evolution. We find that WAI cirrus encountered stronger vertical transport along their trajectories compared to AC cirrus. At the time of detection, WAI cirrus were geometrically and optically thicker, exhibited higher depolarization ratios and ice supersaturation, and contained more numerous but smaller ice crystals, consistent with homogeneous ice nucleation at colder temperatures. AC cirrus, in contrast, showed larger ice crystals and a comparably greater contribution from heterogeneous ice nucleation. As WAI events are projected to become more frequent and long-lasting, a resulting increase in the formation of WAI cirrus may enhance their influence on the Arctic radiation budget leading to cooling. Our results provide observational constraints on cirrus variability in the Arctic and emphasize the need for their improved representation in weather and climate models.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.
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.- Preprint
(2165 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 14 Oct 2026)
- RC1: 'Comment on egusphere-2026-3788', Anonymous Referee #1, 11 Sep 2026 reply
-
RC2: 'Comment on egusphere-2026-3788', Anonymous Referee #2, 29 Sep 2026
reply
This manuscript combines HALO-(AC)3 airborne observations from WALES and MIRA-35 with VarCloud retrievals, CLaMS-Ice microphysical simulations, and LAGRANTO back trajectories to compare cirrus sampled during warm-air intrusions (WAI) and background Arctic conditions (AC). The dataset is valuable, and the descriptive differences in cloud depth, optical depth, humidity, and retrieved microphysics could form a publishable contribution. As indicated in the title, the cloud properties and processes are investigated using both observations and simulations. Therefore, appropriate and consistent statistical treatment of the observational and simulated samples is essential for drawing robust conclusions. Several improvements are recommended in this regard, as detailed below.
Major comments
A central finding of this manuscript is that WAI events span a broader temperature range than AC cirrus events. However, this contrast may partly arise from systematic differences in their latitude and altitude distributions rather than from intrinsic differences between the two cloud types. Clouds occurring at lower latitudes or altitudes generally have higher temperatures. Although the authors report the sample sizes in Section 2.2, they provide insufficient information on the latitudinal and altitude distributions of these samples. To strengthen the robustness of the results, the authors should quantify and control for the effects of factors, such as latitude and altitude, before attributing the observed temperature difference to cloud type. These results would help us determine whether the observed changes in WAI are primarily driven by environmental differences, or whether WAI still differs from AC under comparable environmental conditions.
The classification criteria for WAI and AC events are unclear. Section 2.2 appears to imply that the events were classified according to the flights during which they were observed. However, Section 2.3 states that cirrus clouds were classified through a qualitative assessment of environmental parameters, including ambient temperature and water-vapor conditions. The authors should clarify the specific classification criteria, explain how these parameters were evaluated, and describe how the two approaches are related. To improve the transparency and reproducibility of the analysis, detailed information on each research flight should also be provided in the Supplementary Information.
Measurement and retrieval uncertainty is missing from the key contrasts. The parameters analyzed in this paper are derived from various methods, including observations and simulation. RHi combines WALES water vapour with model temperature, PLDR and OD are lidar products, and IWC and effective radius come from VarCloud. These quantities have different random errors, systematic biases and detection limits. The median WAI-AC difference is only 2.88 percentage points for RHi and 0.61 percentage points for PLDR. Without a quantitative assessment of uncertainty and confidence intervals, it is difficult to determine whether the reported differences are statistically robust.
A clear description of how the two models were configured and run is essential for interpreting their outputs. However, important details of the model setups are missing. For example, in line 130, the authors state that the starting point can be any arbitrary location but do not explain how the starting points were selected in this study. Were all cloud-containing bins observed by WALES used as starting points? How many starting points were used in total? Meanwhile, The very large percentage differences in Table 1 also arise from near-zero AC baselines (IWC and ICN), making the percentages misleading. This naturally raises concerns about whether the configuration and parameter settings of model C are appropriate.
Why are different trajectory durations used for the two models, namely 24 h for one model and 100 h for the other? This discrepancy may affect their comparability and should be carefully justified. For example, the pressure change since cloud formation in Model L is approximately two to three times that in Model C. Please clarify whether this difference is partly attributable to the different trajectory durations. In addition, Figure 3 presents the cloud age and pressure change only for Model C. Please provide the corresponding results for Model L to enable a consistent comparison.
Minor comments
Section 2.2. A brief introduction to the specifications of WALES and HAMP is suggested, including, for example, their range resolution and spatial resolution at nadir.
Figure 1. The figure caption is unclear. Please clarify what “counts” refers to. In addition, the meaning of “count” is not sufficiently explained in the main text.
Figure 2. Do PLDR and OD represent the mean or median values within each bin? Please clarify this explicitly.
Figure 2d. High OD values are observed in the regime with temperatures around 220 K and RHi of approximately 130%. This hotspot is somewhat unusual, as OD appears to increase with temperature in this region. Is there any physical explanation for this OD hotspot? A similar feature can also be seen in Figure 3b.
Figure 6 and 7, line 208, The unit of IWC are different.
Figure 4, 5 and Figure 6. Are the distributions significantly different? A statistical test is recommended to assess whether the observed differences are statistically significant.
Figure 2, 3, and 7, It would be informative to quantify and report the difference between WAI and AC within each bin. In particular, are the WAI and AC values significantly different under the same temperature and RHi conditions?
Table 1. The pressure change and mean vertical velocity have an error of two-order-of-magnitude. Please double check the data.
Table 1. The IWC values obtained from the three data sources (C, L, and V) differ substantially. Please discuss the possible reasons for these discrepancies. Should the unit of CLaMS-Ice be kg/kg? Please verify this and keep consistent across these model comparison.
Figure 10. How do you determine the cloud origin and ice nucleation process in each grid bin? Please make a explanation.
Line 303, “In the previous sections”
Line 314. bracket missed.
Line 345 infer that WAI cirrus contain more complex and larger crystals from their higher PLDR, whereas Table 1 and lines 356-358 report a smaller mean effective radius for WAI. Please explain these different conclusion.
Citation: https://doi.org/10.5194/egusphere-2026-3788-RC2 -
RC3: 'Comment on egusphere-2026-3788', Anonymous Referee #3, 05 Oct 2026
reply
General comments
This article presents a novel aircraft observation dataset of cirrus clouds over the Arctic region during warm air intrusions and evaluate air mass origins using Lagrangian backward trajectory models. The study determines the cloud age and ice nucleation process through the macro- and microphysical properties of the clouds, and presents physical evidence of distinct processes of Arctic cirrus clouds during Warm Air Intrusion.
The dataset is clearly described, and the two models complement each other. However, the results section is underdeveloped and lacks in-depth interpretation. Detailed comments are listed below:
Specific comments
- Figure 2, is temperature measurement available during flights? Why is ECMWF model temperature used?
- Section 2, it would be informative to describe the flight strategy, area of operation, weather conditions during the campaign. It only says the campaign was carried out in the spring, but what are the days of flights? what are the intervals between flights? what time of the day were the measurements taken?
- Section 2.3, please provide the quantitative criteria as how WAI events are classified using the parameters.
- As I understood, the LAGRANTO is better for tracing the location where clouds were formed, and CLaMS-Ice is better for deducing what type of ice nucleation process during the cloud formation. How do we interpret different results predicted by the two models? They IWC from all three data sources are very different in Table 1.
- Line 210, what’s the difference between the lead times of the two models?
- Line 215, Should it be “transition from younger to older parts while the temperatures decrease”?
- Figure 3 shows very nicely the evolution of cloud particles along trajectory, but the interpretation in section 3.1 does not tell a very comprehensive story. The interpretation in section 4 line 314-319 should be part of section 3.1.
- Figure 8 and 9, these two figures are not necessary; the statistics can be better summarized in tables.
- Figure 10, what does the color gradient represent in the figure? Does darker red/blue mean higher possibility of HOM/HET? Why is the darkest red (HOM) along the 100% RHi but not at higher RHi?
- Discussion section should be about scientific implications beyond the results, but section 4 is mainly the interpretation of results in section 3, thus, most of section 4 should be part of section 3.
- Section 5, it’s ok to list future work, but it’s more important to discuss the application of the results from this study. For example, how can your observation help improve the prediction of Arctic cirrus during WAI events?
Technical corrections
- Line 64-65, add “,” before “and even more so”; change “even more so WAI cirrus” to “even more so for WAI cirrus”
- Line 78, should be “in the spring of 2022”
- Please add commas properly in long sentences throughout the manuscript. For example, line 115 add comma before “we use”, line 122, add comma before “we used”.
- Line 144, no need for “:” after “the”
- Line 154, remove the “ ‘ ” after “liquid-origin cirrus”
- Line 303, do you mean “In the previous sections”?
- Line 314, should be “(Fig. 3 (a) and (b))”
Citation: https://doi.org/10.5194/egusphere-2026-3788-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 160 | 94 | 36 | 290 | 35 | 37 |
- HTML: 160
- PDF: 94
- XML: 36
- Total: 290
- BibTeX: 35
- EndNote: 37
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of 'Contrasting Arctic Cirrus Clouds: Properties and Processes from Observations and Simulations' by Georgios Dekoutsidis et al.
This work contains a detailed analysis of the optical, macro- and microphysical properties of Arctic cirrus clouds. The remote sensing data was recorded during the HALO–(AC)³ airborne measurement campaign using Varcloud retrievals from a combined lidar and radar system. In addition, backward trajectories are analysed using two different models, CLaMS-Ice and LAGRANTO. The analysis is based on separating the cirrus clouds into two groups: The first group are cirrus clouds that formed during warm air intrusions and the second group are cirrus clouds that formed during arctic conditions. For these two groups the formation of the clouds, conditions along their trajectories, the optical, macro- and microphysical properties at the time of detection and their origin and ice nucleation processes are compared. Differences were found for instance in the ice particle size and concentration, the cloud geometrical and optical depth. Possible reasons for and consequences of these differences are thoroughly discussed. The results important for the accurate modeling of Arctic cirrus cloud radiative properties. Therefore I recommend this article for publication in Atmospheric Chemistry and Physics with minor revisions (see comments below).
General comments:
-Please add quantitative numbers (which you have very details in your results section) to the two or three most important findings that you give in the abstract or conclusion. This will provides more details in comparison to only quantitative stating more/higher/stronger/smaller etc.
-Provide the threshold parameters for T and available water vapour or backward trajectories inside Arctic circle to be considered as AC or WAI.
This will provide more transparency on the separation. Are there any mixed/ unclear cases that were excluded in the analysis?
-The description of Varcloud is rather short in comparison to CLAMS-Ice and LAGRANTO. Please add a few sentences on how the combined lidar and radar
retrieval works and which parameters can be retrieved.
-Please add a short comparison how the found micro- and macrophysical properties compare to other studies of Arctic cirrus clouds. E.g. De La Torre Castro et al. (2023) and Järvinen et al. (2025) for N, r_eff, IWC), similar for the macrophysical properties. Are the findings consistent?
-L366-377: During warm air intrusion the airmasses originate from non-polar regions where different INP concentrations are expected in comparion to Arctic cirrus where the airmasses originate from Arctic regions. This should also affect on the prevalence of homogeneous vs. heterogeneous ice nucleation. Please add a few sentences on this as well to the discussion.
Specific comments:
L39 & L44: Reference missing
L150: Please rephrase, simplify and elaborate. I don't understand this sentence.
L154: Please elaborate why liquid origin clouds are assumed to be heterogeneously formed. Why can they not be formed homogeneously without INP when the temperature drops below 235 K?
L178: Grammar incorrect. Also two 'however' in a row.
L185: Grammar incorrect [...] microphysics are [...]
L242 & L344/345: Reference missing
Technical corrections:
-Please double check the first paragraph, e.g. L27,31,35 points and commas are not formatted properly.
-Same goes for paragraph 2.3.2 (e.g. L131,L143,L144)
-Make physical variables like T, RHi, IWC and t0 cursive.
-Write out acronyms the first time they appear (e.g. WALES, Varcloud, DARDAR, HAMP, LAGRANTO, CALIOP, CPR,...)
-Please consistently use or not use Oxford comma (L424)
-L433: 3 should be upper case (L433)