| 24 Oct 2025
Inferring processes governing cloud transition during mid-latitude marine cold-air outbreaks from satellite
Abstract. Cloud morphological transitions strongly influence radiative effects and the regional radiation budget. Marine cold-air outbreaks (MCAOs) over the northwestern Atlantic feature such transitions, from overcast stratiform to broken cumuliform cloud fields downwind. Characterizing these transitions requires an understanding of the thermodynamic and dynamical evolution of the marine boundary layer and the interplay between warm- and cold-phase processes. Using a novel 'space–time exchange' approach, we construct instantaneous trajectories using reanalysis winds and extract geophysical variable traces along these trajectories from GOES-16 satellite snapshots for five MCAO events sampled during the NASA ACTIVATE campaign (2020–2022). Clear directionality of traces in liquid water path (LWP)–droplet number (Nd) space reveals sequential dominance of drop activation, condensational growth, and collision–coalescence during cloud thickening. Patterns of traces in domain-LWP versus domain-IWP (ice water path) suggest fingerprints of two distinct mixed-phase processes: (i) gradual liquid depletion via vapor deposition and (ii) rapid depletion via riming, preceded by co-growth of liquid and ice. Elevated Nd suppresses peak LWP and delays cloud breakup. A large spread in shortwave albedo is found during cloud transition, reflecting mixed-phase processes. Metrics denoting cloud organization converge towards the end of the transition, despite differences in cloud micro- and macro-physical properties among cases. These results underscore the central role of frozen hydrometeors in governing cloud transitions and demonstrate a powerful framework for process inference based on satellite snapshots using the 'space-time exchange' approach. This framework offers a new pathway to benchmarking model representations of mixed-phase microphysics and advancing model-observation synergy.
Competing interests: Two of the authors are members of the editorial board of Atmospheric Chemistry and Physics. Other than this, the authors declare that they have no conflict of interests.
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.
For five marine cold-air outbreaks (MCAOs) that occurred during the ACTIVATE campaign, the authors generate Lagrangian trajectories and extract geostationary satellite retrievals along them, in particular liquid and ice water paths (LWP and IWP) and cloud droplet number concentration (Nd). Using stereotypical process signatures, the authors infer process occurrence during overcast-to-broken cloud transitions. Lastly, the authors examine cloud organizational metrics.
While the approach is highly innovative and the paper is well written, it rests on a series of assumptions. In my opinion, the authors should verify these assumptions. Given the scope of the proposed revisions, I recommend returning the manuscript for major revisions.
Major concerns
Satellite retrievals and derived products: The authors should explain SatCORPS retrievals performance under MCAO conditions where publications exist or else express the lack of such performance analysis. In addition, the authors should explain in more detail the pixel-based liquid-ice phase categorization. The latter issue is particularly relevant where condensate becomes increasingly mixed and retrievals may confuse condensate mass (e.g., in updrafts that are typically both high in LWP and IWP) and how it would affect the shown analysis. Furthermore, the authors should at least briefly demonstrate the veracity of subadiabaticity assumptions needed for Nd retrievals (e.g., via ACTIVATE dropsonde data), especially where clouds are increasingly convective natured. Lastly, the authors should clarify whether LWP includes cloud and rain condensate.
Process signatures: The authors show anticipated LWP-Nd process signatures in Fig. 3a. While the authors cite previous work, they should in more detail explain which synoptic conditions were previously targeted (e.g., is the cited work investigating subtropical Sc?) and weather any differences are expected under MCAO conditions. For example, currently entrainment is shown to have either no impacts on Nd when homogeneous or subtle impacts when heterogeneous; previous work (e.g., Tornow et al., 2022) has demonstrated strong cloud condensation nucleus (CCN) dilution effects as cleaner free-tropospheric (FT) air is entrained into the marine boundary layer. Furthermore, given the non-negligible role of secondary ice production in these cases – what type of process signature is expected?
Steady conditions: The authors show that horizontal winds remain approximately steady during daytime hours. Given the important role of large-scale vertical winds in shaping MCAO cloud evolution (e.g., Tornow et al., 2023), were vertical wind speeds along the trajectory truly steady? The authors should at least briefly explore this question for a single layer close to cloud tops (e.g., 700 hPa).
Test on Lagrangian simulations: In their discussion (l. 358-360) the authors suggest applying this framework to Lagrangian simulations. LES and SCM simulations now exist for four out of the five cases (https://github.com/NASA-GISS/LES-SCM). While LES simulations could serve as an additional proxy for field data. (e.g., to assess subadiabaticity assumptions), it also offers microphysical source terms that can directly connect to the process signatures. Lastly, observational constraints from MAC-LWP (Elsaesser et al., 2017, located at https://github.com/NASA-GISS/LES-SCM/tree/main/data_files) may help to further corroborate SatCORPS LWP retrievals.
Minor concerns
l. 164 It would be good to show the range of meteorological conditions across trajectories for each case as shading behind lines. It would also be good to show large-scale subsidence (see above major concerns).
ll. 173-174 Precipitation appears to set in at re << 15 um; please modify or else explain.
ll. 182-184 Nd appears to decrease once LWP ~ 100 g m^-2 is reached; could this be explained by early collision-coalescence or alternatively by FT CCN dilution (see above major concerns)?
ll. 192-193 Is this corroborated by ACTIVATE measurements?
ll. 193-195 Is a reduced LWP consistent with the existing literature?
ll. 201-205 Since most IWP retrievals are notoriously uncertain, it is important to explain any strengths and weakness of the SatCORPS retrieval (see above major concerns). Can we even trust the order of magnitude here?
Fig. 4 Along with the above concern, please add error bars to data points.
ll. 207-209 How many pixels are there in a 1x1 degree domain and is one ice pixel is sufficient to render the domain “mixed”?
ll. 214-216 Could rain also cause liquid depletion?
ll. 207-220 To what degree can this LWP-IWP evolution be affected by the binary condensate classes? I wonder if retrieval samples (e.g., spatially resolved IWP and LWP values in progressive domains) could be informative.
ll. 225-228 It is unclear how representative these 2DS samples are. Perhaps other metrics may be more informative (e.g., how many flight seconds of co-existing liquid and frozen particles from in-situ probes) or there is a way to quickly determine sample representation?
l. 232 Specific thresholds from the retrieval would be quite important here (see above major concerns).
ll. 232-234 CTTs from ACTIVATE’s HSRL seem to disagree here, showing 2022-01-29 at -10 degC (Fig. 5 in Tornow et al., 2025). Could this stem from surface contamination in optically thinner or broken clouds within GOES pixels?
ll. 247-256 A lot of the earlier findings (that were initially “intended to conceptually indicate the dominant characteristics of a given processes”) are relied on here without any uncertainty. This very much reads like a discussion, and I suggest moving it there.
ll. 252-253 Please check this sentence.
ll. 254-255 (and also l. 11 and l. 348) It is unclear what exactly the “spread” is. Is it a large range in albedo at any given cloud fraction?
ll. 257-266 Given the general importance of meteorological boundary conditions, I wonder if this paragraph should be moved to the beginning of Section 3?
ll. 271-275 Could the diurnal evolution of MBL aerosol upwind of cloud formation (e.g., Tornow et al., 2025b) explain some of this behavior?
ll. 289-290 I suggest also looking into changing subsidence patterns here (see earlier comment).
ll. 337-340 I would soften “evident” here, assuming that a combination of other processes (e.g., entrainment plus collision-coalescence) could also lead to a precipitation signature.
ll. 358-360 Output from observationally constrained Lagrangian simulations of four of these cases is now available (see major concerns). Application of the authors’ approach to simulations would make the paper (and the “line of evidence” for model evaluation) stronger by (1) bypassing potential satellite retrieval issues and (2) applying it to coherent output with known process rates in it. Please contact me (email: ft2544@columbia.edu) if needed.
References
Elsaesser, G.S., C.W. O'Dell, M.D. Lebsock, R. Bennartz, and T.J. Greenwald, 2017: The Multi-Sensor Advanced Climatology of Liquid Water Path (MAC-LWP). J. Climate, 30, no. 24, 10193-10210, doi:10.1175/JCLI-D-16-0902.1.
Tornow, F., A.S. Ackerman, A.M. Fridlind, B. Cairns, E.C. Crosbie, S. Kirschler, R.H. Moore, D. Painemal, C.E. Robinson, C. Seethala, M.A. Shook, C. Voigt, E.L. Winstead, L.D. Ziemba, P. Zuidema, and A. Sorooshian, 2022: Dilution of boundary layer cloud condensation nucleus concentrations by free tropospheric entrainment during marine cold air outbreaks. Geophys. Res. Lett., 49, no. 11, e2022GL098444, doi:10.1029/2022GL098444.
Tornow, F., A.S. Ackerman, A.M. Fridlind, G. Tselioudis, B. Cairns, D. Painemal, and G. Elsaesser, 2023: On the impact of a dry intrusion driving cloud-regime transitions in a mid-latitude cold-air outbreak. J. Atmos. Sci., 80, no. 12, 2881-2896, doi:10.1175/JAS-D-23-0040.1.
Tornow, F., A. Fridlind, G. Tselioudis, B. Cairns, A. Ackerman, S. Chellappan, D. Painemal, P. Zuidema, C. Voigt, S. Kirschler, and A. Sorooshian, 2025: Measurement report: A survey of meteorological and cloud properties during ACTIVATE's postfrontal flights and their suitability for Lagrangian case studies. Atmos. Chem. Phys., 25, no. 9, 5053-5074, doi:10.5194/acp-25-5053-2025.
Tornow, F., E. Crosbie, A. Fridlind, A.S. Ackerman, L.D. Ziemba, G. Elsaesser, B. Cairns, D. Painemal, S. Chellappan, P. Zuidema, C. Voigt, S. Kirschler, and A. Sorooshian, 2025b: High accumulation mode aerosol concentration and moderate aerosol hygroscopicity limit impacts of recent particle formation on Northwest Atlantic post-frontal clouds. Geophys. Res. Lett., 52, no. 18, e2025GL116020, doi:10.1029/2025GL116020.