| 29 Jan 2026
Opposing entrainment effects of cloud droplet sedimentation during the pre-breakup stage of the stratocumulus to cumulus transition
Abstract. Cloud droplet sedimentation is known to influence the evolution of the stratocumulus-topped boundary layer by reducing entrainment. Although this mechanism is well studied regarding the early evolution of stratocumuli, its sustained effects over longer timescales remain largely unexplored. Here, we use large-eddy simulations to investigate how sedimentation influences stratocumulus development in the context of the stratocumulus to cumulus transition. We conduct 48 h long simulations of 10 transects in the Northeast Pacific, covering the full deepening stage before cloud breakup. All sedimentation cases show the previously reported initial reduction in entrainment, whereas the later stages reveal different effects depending on the cloud's liquid water path (LWP). While the more frequent precipitating, high-LWP (LWP > 50 g m−2) cases continue to exhibit weaker entrainment, the non-precipitating, low-LWP (LWP ≤ 50 g m−2) cases reverse the initial effect and show stronger entrainment. In those radiatively unsaturated low-LWP clouds, the increase in LWP due to the initial entrainment reduction initiates a feedback chain that amplifies LWP, longwave cooling, and turbulent circulations in the boundary layer, ultimately leading to increased entrainment. Initial studies showed that droplet sedimentation reduces entrainment in short (≤ 6 h) simulations of low-LWP clouds, which has been extrapolated in the literature to all stratocumuli on much longer timescales. Our results suggest that this extrapolation is indeed correct in common high-LWP clouds, although it had previously been inferred from the rare low-LWP regime, where the opposite is found. Meanwhile, we find that cloud breakup remains largely unaffected across the transition.
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.
General Comment
This paper addresses a less explored issue of the impact of cloud water sedimentation on the stratocumulus to cumulus transition at its early stages. The topic is relevant, the goal of the paper is well-defined, the methodology used is mostly valid, and the results are interesting. The writing is already of good quality. I recommend publishing the paper after major revision.
Specific comments
Please clarify that the saturated and unsaturated cases relate to optically thick and thin clouds, as the current terms can easily be confused with saturated and unsaturated air. I suggest considering a change in terminology to “optically thick” and “optically thin” clouds
Introduction:
The process-level explanation of what happens during the SCT, including the impact of the diurnal cycle, is also provided in van der Dussen (2014) and Kurowski et al. (2025). An important recent observational study of Pugsley et al. (2025) highlights the role of nighttime processes on aerosol-cloud interactions, whereas Lebsock et al. (2024) show the response of LWP to aerosol perturbations.
Section 2.1
It suggests a 1-D approach (SCM); Please clarify.
McGibbon and Bretherton showed that LES can be used to mimic observations on a moving platform (ship), with important changes in the mean thermodynamic state accounted for via advective tendencies. However, your approach assumes continuity of cloud-related processes as if it were a Lagrangian mass-conserving perspective. What is the relevance of your approach to a Lagrangian one if you do not adjust for the impact of microphysical changes? Furthermore, depending on the ship's speed, the impact of cloud water sedimentation will be different in this setup, which is not physical.
What is the configuration of the control run? Is it simply without cloud water sedimentation?
When you run your simulations for 6 hours to spin up the model, do you account for any diurnal cycle during that phase? It seems to be a high-impact approach during which the model already develops significant differences between the control and other simulated states. What exactly happens during that phase in terms of the mean cloud state and droplet number concentrations? What is the purpose of it? Does it lead to significantly different initial states? Does it change the response of the WTG correction? It needs to be clarified, because all the differences you get at later times may strongly depend on that initial state.
Furthermore, your Fig. A4 shows that you indeed start with significantly different initial states: 100 g/m2 vs 200 g/m2, which immediately impacts further cloud development.
Section 2.2
L113: “low” relative to what?
L114: What do you want to say in this sentence? It is unclear.
L119-125: Explain all the symbols used. What are N, L, m here? This paragraph needs more clarification.
This section lacks comparison of cloud droplet sedimentation to Bretherton et al. (2007). What is similar? What is different? Are the values of terminal velocity comparable between their study and yours?
Section 2.3
It seems that the WTG corrections can strongly alter the simulated state and can also be simulation-dependent. Is that the case? If yes, wouldn’t it be more reasonable to turn them off completely and focus on studying the subtle effects of cloud water sedimentation without them? When comparing two different simulation results, does it mean you are looking at both cloud water sedimentation and different WTG effects combined? If yes, then how do you attribute the differences to cloud water sedimentation only?
Section 3.1
Do your calculations of pseudo-albedo account for solar zenith angle? Kurowski et al. (2025) show how to include it.
When comparing cloud albedo, did you use GOES-15 effective radius? How does it compare to the LES? Running LES on an 8 km × 8 km domain has strong limitations on the maximum size of cloud structures represented in the domain. Such small domains are not feasible for simulating changes in closed vs open cell dynamics, which needs to be clarified. Can you provide any numbers regarding cloud albedo values from LES and GOES? I assume they are compared at the same local times.
L48: This comparison is rather qualitative given all the differences mentioned above.
Figs 4 and 5: How does this comparison change for different times of day and night? Do the differences depend on it?
Fig. 6: The two evolutions shown are almost the same but shifted in the vertical because of the differences in the initial state. Consequently, the differences between the two oscillate around that initial difference. That figure shows that the 48 h evolution is not much different, but the initial spin-up leads to the main differences. How big of a contribution comes from WTG corrections? Are they assumed to be the same in both cases shown?
Do surface fluxes change between the control and other simulations? Does cloud water sedimentation have any impact on it?
Have you looked at the differences in decoupling caused by sedimentation?
Most statistics are reported using domain-averaged LWP, but radiative impacts and entrainment feedbacks are more directly tied to in-cloud properties (LWP_c weighted by cloud fraction, optical thickness τ_c). Please consider adding parallel in-cloud/conditional analyses (or at least a decomposition into cloud-fraction and in-cloud components) to better connect the results to satellite observables and recent ACI studies.
The effects of cloud water sedimentation are also discussed in Kurowski et al. (2025), in addition to the diurnal cycle analysis of cloud susceptibility to aerosols. How do your results compare to theirs? Part of the conclusions seem to be consistent with their sensitivity study. How does your process-level understanding relate to theirs?
References:
Kurowski, M. J., Lebsock, M. D., and Smalley, K. M.: The diurnal susceptibility of subtropical clouds to aerosols, Atmos. Chem. Phys., 25, 15329–15342, https://doi.org/10.5194/acp-25-15329-2025, 2025.
van der Dussen, J. J. and Co-authors (2013), The GASS/EUCLIPSE model intercomparison of the stratocumulus transition as observed during ASTEX: LES results, J. Adv. Model. Earth Syst., 5, 483–499, doi:10.1002/jame.20033.
Pugsley,E. Gryspeerdt, & V. Nair, Cloud fraction response to aerosol driven by nighttime processes, Proc. Natl. Acad. Sci. U.S.A. 122 (47) e2509949122, https://doi.org/10.1073/pnas.2509949122 (2025).
Smalley, K. M., Lebsock, M. D., & Eastman, R. (2024). Diurnal patterns in the observed cloud liquid water path response to droplet number perturbations. Geophysical Research Letters, 51, e2023GL107323. https://doi.org/10.1029/2023GL107323