the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Jun 2026
Large-scale circulation and stratocumulus variability
Abstract. This study aims to understand the relationship between large-scale circulation and stratocumulus variability. We use reanalysis and satellite data to identify circulation patterns that couple with estimated inversion strength (EIS) and low-cloud cover (LCC) in stratocumulus areas. The results update the current understanding in two aspects: the limited direct influence of the tropical thermodynamic framework on stratocumulus, and the different responses of EIS and LCC to large-scale circulation.
Extratropical dynamics control EIS variability. From synoptic to interseasonal timescales (after deseasonalization), synoptic-scale Rossby ridges located directly over stratocumulus enhance stability throughout the tropospheric column and thereby increase EIS. On interannual timescales, planetary-scale Rossby waves coupled with a negative PDO-like (Pacific Decadal Oscillation-like) sea surface temperature pattern increase EIS. In contrast, LCC responds to circulation patterns similar to those associated with EIS, but with a systematic upstream (west and poleward) shift. This shift suggests a direct response of LCC to circulation through enhanced pressure gradients, which increase cold advection and offset the drying effect of Rossby ridges via stronger winds. The upstream Rossby ridges associated with increased LCC often overlap with the subtropical highs, which has led to the previous emphasis on thermodynamic processes that strengthen subtropical highs by enhancing the descending branch of the Hadley cell.
- Preprint
(35795 KB) - Metadata XML
-
Supplement
(17981 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-2787', Anonymous Referee #1, 02 Aug 2026
-
RC2: 'Comment on egusphere-2026-2787', Anonymous Referee #2, 03 Aug 2026
Review of Large-scale circulation and stratocumulus variability by Hairu Ding, Bjorn Stevens, Frank Lunkeit, and Nedjeljka Žagar
This study looks at the connections between large scale circulation of the atmosphere and the semi-permanent subtropical eastern ocean low cloud decks. To do this the authors examine the relationship between a measure of the static stability of the atmosphere (estimated inversion strength, EIS) and low cloud cover (LCC) over the major eastern subtropical low cloud cover regions. To do this, the authors use software that can track wave and wave-like features in meteorological reanalysis. Lagged compositing is used to evaluate the synoptic/large-scale patterns controlling EIS variability on timescales from synoptic through to interannual. The authors find that circulation patterns, particularly those due to extratropical Rossby waves, influence the LCC and EIS in different ways, suggesting that the EIS is not always a direct influence on LCC especially on synoptic to subseasonal timescales. Rossby wave ridges in the midlatitudes impact EIS largely through the free tropospheric temperature rather than the SST. These appear to be enhanced by local orography. SST changes play a large role in influencing EIS and LCC on interannual timescales.
The paper is interesting and attacks a longstanding question of why EIS is such a strong predictor of LCC on a range of timescales, particularly seasonal to interannual. The approach taken of grounding the meteorological influences on LCC as part of the extratropical and tropical wave patterns in the atmosphere is quite novel and will be of significant interest to those wishing to understand how the meteorological environment influences low clouds. That said, I don't really understand what the MODES software is doing exactly, or why it is really necessary. The use of it makes it difficult to interpret the findings in a more straightforward meteorological framework. Apart from some confusion I had with some of the grammatical expressions, I did find the manuscript to be interesting and compellingly structured, breaking the timescales of EIS and LCC variability into synoptic, subseasonal, interseasonal and interannual timescales. Compositing is used to evaluate the meteorological structures that are associated with anomalously high LCC and EIS conditions. The authors show that examining LCC-EIS correlation on the seasonal cycle hides an important difference between SST warming upstream and LCC. Removing the annual cycle makes this more apparent. The manuscript will be of interest to readers of ACP, and I suggest the authors consider my mostly minor comments before revising.
SPECIFIC COMMENTS:
Line 5: What does interseasonal mean when the record is deseasonalized?
Line 17: The STBL is not always well-mixed. Most of the Sc on Earth is probably not in a well-mixed PBL.
Line 22-24: These statements are not necessarily contradictory. What controls Sc in the current climate need not be the same as what controls climate changes.
Line 24/25: Provide evidence to support this.
Line 119: Is the "Sc area index" the same as the LCC?
Eqn. 4: What if 1000 hPa is below the surface?
Line 176: EIS has to be controlled by T700 on synoptic and subseasonal timescales given that SST (which controls the near surface temperature) does not vary strongly on these timescales.
Figure 1: Abscissae are not labeled. Should say "days". Also labeling on abscissa should be 10^3 for the rightmost value, not 10.
Line 206: I am unclear how the waveguide assertion is deduced from Fig. 3. The Rossby wave pattern does not seem to move along these jets - they often cross the jets.Line 221: Does Myers and Norris show that EIS and subsidence are not correlated? The authors should provide more specific info on this. Fig. 9 in Myers and Norris appears to show that EIS and subsidence are correlated positively.
Line 305: I don't see an 80% difference in Fig. 9b. Please explain what is meant here. Do the authors mean 8%?
Line 312: The decoupling induced by increased LHF involves increased entrainment, which then suppresses mixing through the entire PBL, but this would take time.
Grammatical issues:
Line 21: "contradictory" rather than "contradict"
Line 100: "type" rather than "types"
Line 122: "with deep convection", rather than "with a deep convection".
Line 133: "mass" not "mas"
Line 209: "Close to".
Line 263-264: Grammar. What does approaching Sc mean? Approaching the climatological region of peak Sc?
Line 308: What does "from the west poleward" mean? That would be from the southwest?
Line 336: This is not a sentence. Please adjust.
Line 337: I do not understand what is meant by "the preference of regions still shows". I can't follow much of this paragraph to be honest. Please consider rephrasing.
Line 359-360: I can't follow this. What does it mean to say a variable prefers winter conditions? It is high in winter? Low?
Citation: https://doi.org/10.5194/egusphere-2026-2787-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 107 | 38 | 13 | 158 | 24 | 14 | 16 |
- HTML: 107
- PDF: 38
- XML: 13
- Total: 158
- Supplement: 24
- BibTeX: 14
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
Overall recommendation: Not acceptable in its current form / requires major revision.
The study explores the large-scale dynamical processes that drive variability in subtropical stratocumulus. Stratocumulus clearly play a key role in Earth’s radiative budget and climate feedbacks. And most previous studies on the drivers of stratocumulus have focused on the climatological-mean or seasonal cycle. Ding et al. 2026 propose instead to explore the processes that govern stratocumulus on timescales ranging from the synoptic to interannual. In this sense, the paper is both needed and important.
However, I think the analysis design and treatment of the data are overly and unnecessarily complex and opaque. As a result, I think many of the results are difficult to interpret. I also worry about reproducibility. My own research is focused on the analyses of climate phenomena. But I nevertheless had a very hard time understanding how many of the results were generated and what is being shown in many of the figures. I think it would be effectively impossible to reproduce the results without relying on the MODES code. The methodology leaves the data very little ‘room to breathe' before it is analyzed.
I appreciate the analyses involved a lot of time and effort. But I think the paper requires extensive major revision before it can be considered for publication. I suggest the authors revise the paper with a focus on simplifying the analysis, reducing the reliance on existing code packages, and ensuring reproducibility. It would be OK to keep some of the results based on the MODES code. But those results should be supported by results based on more standard and straightforward analysis tools. I've provided an example below.
A few specifics.
- One of the main conclusions is that stratocumulus are primarily influenced by extratropical wave processes and that “instead of the tropical thermodynamic framework, we find an extratropical dynamics control EIS variability.” This is a strong conclusion and I worry it derives at least partially from the treatment of the data by the MODES package (since it reformats the data into normal modes of the equations). Reproducing the results based on simple linear methods would go a long way to addressing this concern.
For example, consider eddy activity at 500mb. The authors could 1) form Z*500 by removing the zonal mean from the daily data; 2) time filter Z*500 into the frequency bands of interest using a Butterworth (or similarly robust) filter; and 3) regress or correlate the resulting time series onto indices of Sc incidence as a function of lag. The results would be simple to reproduce and interpret, and if the conclusions of the study are robust, they should yield maps similar to those shown in Fig. 3.
- The study feels like it is motivated by a 'tool looking for an application'. If the authors keep the MODES results, I think they need to demonstrate what is gained from the MODES software that can not be inferred from more straightforward analysis techniques.
- Section 2.4 and Eq. 2 are overly terse and I don't think provide enough information to reproduce the results. Understanding Section 2.4 would require considerable familiarity with the source paper (Zagar et al. 2015). Likewise, why use the SciPy Python package (version 1.14.1) to compute coherence? Cross spectral analysis is straightforward to code using FFT. From the text, it is not clear how one might reproduce the results. How are degrees of freedom and significance assessed?
- Running-mean filters are generally not considered appropriate for precise time filtering. They are fine for display and simple low-pass filtering. But the sidelobes that result from the Fourier Transform of the boxcar window function render them unsuitable for the fine-scale filtering done in the analysis ( <15 days; 15-90 days; 90-365 days). I suggest using, say, a Butterworth filter or something along those lines. It is widely used, easy to apply, and has much smaller sidelobes.
- Some of the conclusions appear to derive from visual inspection of the results and lack quantitative support. For example, the PDO is listed in the abstract: “On interannual timescales, planetary-scale Rossby waves coupled with a negative PDO-like (Pacific Decadal Oscillation-like) sea surface temperature pattern increase EIS”. As far as I can tell, the conclusion is drawn from visual inspection of Fig. 10 and is never quantitatively tested. The PDO also exhibits most of its power on decadal timescales, which is beyond the scope of the data length in the paper. Perhaps it would be more robust to work with ENSO rather than the PDO?
- Why is the analysis period limited to an endpoint of 2019? Why not include data through 2025?
Some more detailed comments:
- Figure 1: What are the units on the x-axis of the coherence plots? Do the results show coh^2? How many degrees of freedom are assumed in the significance calculation? I recommend reproducing the coherence in Fig. 1c and 1d in subsets of the data to ensure reproducibility.
- Figure 2: What exactly is being shown here? Is this the amplitude of wave activity integrated over the entire globe associated with variations in LCC on the indicated timescale? If so, why the globe and not the regions of interest? The text and caption do not provide enough information to properly interpret the results, or to reproduce them.
- Figure 3: “We see that the circulation controlling synoptic EIS and LCC variability is an extratropical Rossby wave”. It is true that the results in Fig. 3 appear as Rossby waves. As noted above, I worry the wavelike nature of the results is mandated by the MODES software package. Do you recover similar features if you simply correlate, say, Z500* with variations in clouds in the respective regions? If not, why not?
- Figure 6: The text states “Hence, tropical perturbations are not the direct driver controlling the interannual variability of Sc.” The conclusion seems to be drawn from visual inspection of the results rather than a statistically significant relationship.
- Figure 7. The wave activity flux is a useful tool for diagnosing wave/mean flow interactions. But that is not the focus of this paper, and I don’t understand how the flux results add to the conclusions of the current study. The saturation of the shading in Fig. 7 is hard to follow.
Figure 10. I do not think the evidence in Fig. 10 is sufficient to warrant the conclusion “Figure 10 suggests that it might be due to some shared influence by low-frequency circulations”.
- The text would benefit from proofreading. There are a lot of typos. Here are a few examples but there are many more.
Line 21: “later studies with contradict results”
Line 28: “Regarding the large-scale mechanism”
Line 41: “Above discussion highlights”
Line 42: “there are several questions remain”
Line 59: “1987-01-01–2019-12-31” Why not just spell out Jan 1987-Dec 2019?
Line 103: “The tropical trapped”
Line 342: “Instead of the tropical thermodynamic framework, we find an extratropical dynamics control EIS variability.”
Line 189: “Figure 2 shows the zonal wavenumbers spectra”
Line 263: “When a Rossby ridge approaching Sc”