the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 May 2026
Origin and evolution of satellite-observed cirrus clouds using Lagrangian microphysical modeling – Part 2: Evaluation and sensitivity analysis
Abstract. Linking cirrus cloud microphysical properties, their formation history, and their origin – whether liquid or in situ – to global satellite observations will represent a major step toward understanding these clouds, which are climatically important but not yet fully understood. This link is established by integrating Lagrangian microphysical modelling along air parcel trajectories ending at the points of satellite observations. In Part 1 of this study, detailed microphysical Lagrangian cirrus model CLaMS-Ice was coupled with DARDAR-Nice satellite retrievals to establish the DC-Ice (DARDAR → CLaMS-Ice) framework, enabling the derivation of origin-based metrics associated with satellite observations. Here, we evaluate the performance of the DC-Ice for revisited case studies and examine its sensitivity to key model parameters. Simulated ice crystal number concentration (Ni) and ice water content (IWC) are statistically compared with satellite retrievals and in situ observations. DC-Ice shows good agreement with observations for Ni and IWC, except for orographic cirrus, where the IWC from DC-Ice is lower than that from DARDAR-Nice. Sensitivity experiments indicate that variations in the parameterised sedimentation affect the simulated microphysical properties, especially IWC, while temperature fluctuations influence Ni. Nevertheless, it appears from the sensitivity analysis that origin-based metrics proved relatively robust except under most configurations. These findings highlight the viability of incorporating origin-based metrics into satellite observations, paving the way for improved global understanding of cirrus origins and their impact on the climate system. Future development includes the implementation of a three-dimensional sedimentation scheme in CLaMS-Ice and the extension of the framework to tropical regions.
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
(6946 KB) - Metadata XML
-
Supplement
(29328 KB) - BibTeX
- EndNote
Status: open (until 30 Jun 2026)
- RC1: 'Comment on egusphere-2026-2423', Anonymous Referee #1, 08 Jun 2026 reply
-
RC2: 'Comment on egusphere-2026-2423', Anonymous Referee #2, 11 Jun 2026
reply
General Comments:
It was difficult to evaluate Part 2 of this study without having access to Part 1 (which I was
unable to access), so Part 2 is evaluated here as a stand-alone paper. The paper appears
well-conceived and organized, although it was ambiguous to me how CLaMS-Ice was
initialized from the back-trajectories. Origin based metrics were often mentioned without
explaining what they refer to. The results make intuitive sense to me. I recommend
publication with minor revisions.
Specific Comments:
1. Figure 1: As noted on page 6 (1st full paragraph), the bimodality for Ni in Fig. 1 for CLaMS
Ice is due to the differing ice nucleation processes (hetero- and homogeneous freezing).
This bimodality does not appear for the satellite retrievals due to the assumption of a
monomodal PSD, but bimodality is apparent for the in-situ data. However, this Ni
bimodality for the in-situ data has a secondary pdf maximum at higher Ni relative to the
primary pdf maximum at lower Ni, whereas the opposite occurs for CLaMS-Ice (where the
primary pdf maximum occurs at higher Ni relative to the secondary maximum). Noting that
homogeneous freezing nucleation tends to produce higher Ni relative to heterogeneous
freezing, does this suggest that homogeneous freezing nucleation is more active in CLaMS
Ice relative to the in-situ observations? Moreover, these two pdf maxima in CLaMS-Ice
occur at lower Ni values relative to the in-situ Ni observations. Do the authors have an
explanation for this?
2. Lines 155-6: What was the average time difference between in situ sampling and the
satellite overpass time?
3. A simple formula for estimating the IWC associated with homogeneous freezing
nucleation (hom) is provided and validated in Mitchell and Garnier (2025, ACP, Eqn. 4).
Since it is merely the difference between water vapor density at the supersaturation
threshold for hom and the vapor density corresponding to ice saturation, it may be a lower
bound since stronger updrafts may produce vapor densities that exceed this threshold and
may provide more time for ice particle growth within a cloud layer. Nonetheless, it may be
an interesting benchmark that may impart greater physical understanding of the results
shown here in Fig. 1 and elsewhere. This is only a suggestion that the authors may use if
they feel this would improve their paper.
4. Lines 265-7: Regarding temperature fluctuations in the high updraft case in Fig. 3, these
results appear consistent with the Ni results in Fig. 1, with the primary maximum in CLaMS
Ice associated with hom (higher Ni) becoming more dominant with increasing temperature
fluctuations. However, the in-situ measurements indicate that hom produces a secondary
pdf maximum (i.e., not a primary maximum as with CLaMS-Ice). Again, does this suggest
that hom may be overactive in CLaMS-Ice?
5. Line 341: Was Δttraj discussed in Sect. 3 or elsewhere?
6. Fig. 5. Please provide a reason why the orographic case is almost pure in-situ cirrus,
even when fsed = 1 (i.e., no sedimentation). How representative do you think this case is
regarding orographic cirrus?
Technical Comments:
1. Abstract: “Nevertheless, it appears from the sensitivity analysis that origin-based
metrics proved relatively robust except under most configurations.” This sentence is hard
to understand since the last part (except under most configurations) appears to contradict
the earlier part.
2. Line 143: “interquartile range”; does this mean the 25 to 75 percentile range? If yes, is it
commonly understood?
3. Line 182: “observations” => satellite observations?
4. Line 223: “depletes” => enhances? (Fig. 4 and Spichtinger and Gierens, 2009a, ACP)
5. Line 308: “reducing” => increasing?
6. Line 322: “depletes” => increases? (ditto)
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 262 | 44 | 15 | 321 | 43 | 17 | 16 |
- HTML: 262
- PDF: 44
- XML: 15
- Total: 321
- Supplement: 43
- BibTeX: 17
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General Comments
This manuscript investigates the influence of random temperature fluctuations, sedimentation flux factors, and INP concentrations on IWC, Ni, cloud age, and ice origin (liquid-origin vs. in-situ and homogeneous vs. heterogeneous nucleation) across three case studies representing distinct updraft and cloud formation regimes. Sensitivity studies using ClaMS-Ice microphysical simulations are compared to DARDAR-Nice satellite retrievals.
The manuscript is well written and the scientific quality is consistent with the standard set by comparable studies. However, I have some minor comments to help further improve the manuscript.
Extensive variables
The microphysical properties are expressed as extensive variables (IWC in kg m-3 and Ni in m-3 ), which depend on the size of the system. In a Lagrangian framework, air parcels undergo thermodynamic expansion, which introduces a bias when comparing microphysical properties across trajectories or over the temporal evolution of a single trajectory. Is there a particular reason for this choice over intensive variables? Intensive variables are likely already used in the backend of CLaMS-Ice, and IWC is frequently expressed as such in observational studies (e.g., in ppmv as in Krämer et al. (2016)).
Influence of sedimentation
Section 5 notes that sedimentation into air parcels is not considered, identifying this as a source of uncertainty to be addressed in future work. However, this uncertainty warrants more detailed discussion in the context of the results presented in Section 4, particularly with regard to the ice origin classification (liquid-origin vs. in-situ and homogeneous vs. heterogeneous nucleation). It should be feasible to provide at least a qualitative assessment of which regions of the vertical distribution are most susceptible to ice sedimentation from above using the existing dataset. For instance, the ice origin in the orographic case is likely less affected by sedimentation than the other cases.
A further question concerns the utility of the ice origin classification, as employed here, for interpreting satellite retrievals. However, this falls beyond the scope of this manuscript.
Specific comments
Manuscript title & Line 4: The term 'Lagrangian microphysical modeling' may cause confusion, as it is also associated with the super-particle/super-droplet approach. I suggest clarifying that the DC-Ice framework employs a bulk microphysical scheme applied along Lagrangian air parcel trajectories. Additionally, in Line 4 the usage of 'Lagrangian microphysical modeling', and 'along air parcel trajectories' appear redundant in the same line.
Line 22: Availability of aerosols in general or only of aerosols that have ice-nucleating properties?
Line 57: Include definition of the concept ‘climate sensitivity’. Not all readers have a meteorology / climate science background and the term is not self-explanatory.
Section 2: While it is reasonable to defer the majority of the model description to Part 1, this manuscript would benefit from being somewhat more self-contained. I suggest briefly introducing ClaMS-Ice as a two-moment bulk microphysical scheme, and defining the principal output variables.
Line 132-133: Specify the IWC and Ni value ranges that define 'thick' cirrus and 'thick' mixed-phase clouds. Specify the range of updraft values.
Line 183-184: Is the assumption of coexistence between ice of different origin investigated in Part 1?
Line 260: Here the statement of Line 256 is repeated, consider reordering the relevant paragraphs to avoid this redundancy.
Line 268-270: While temperature fluctuations alter the saturation vapor pressure, the vapor mass density remains unchanged. Depositional growth is therefore still constrained, which could explain why IWC is relatively insensitive to random temperature fluctuations?
Line 278-279: Are these values of INP concentrations for standard temperature and pressure (STP) conditions? And if so are these concentrations recalculated to the cirrus level conditions?
Line 300: IWC in cirrus cloud is controlled by deposition of water vapor and thus thermodynamical constrained. Thus, with small sedimentation factors, we would expect IWC to be mostly insensitive to INP concentrations.
Line 329-330: Do the INP that initiate immersion or contact freezing of droplets, and thereby contribute to the pre-existing ice population, necessarily also act as INP for deposition nucleation? While this may generally hold, it warrants some consideration.
Section 4.1: Include a short summary of the algorithm to classify cloud origin and cloud age.
Line 355-357: The assumption is that each pre-existing ice crystal contains an INP? Therefore homogeneous freezing of cloud droplets and secondary ice processes are disregarded in the following interpretation? Is turbulent mixing and thus redistribution of INPs in ‘seeded’ air parcels considered?
Line 451-452: Why does an increase in ice crystal number concentration correspond to an increase in IWC, given that IWC is largely constrained by the thermodynamic environment? Is this related to reduced sedimentation loss due to the smaller mean crystal size associated with higher ice crystal number concentrations?
Technical corrections
Line 13: Superfluous ‘except‘