Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Greater climate sensitivity implied by anvil cloud thinning


High clouds produced by tropical convection are expected to shrink in area as the climate warms, and the radiative feedback associated with this change has long been the subject of controversy. In a recent assessment of climate sensitivity, the World Climate Research Programme estimated that this feedback is substantially negative, albeit with substantial uncertainty. Here we examine the cloud response using an approach that treats high clouds as part of an optical continuum rather than entities with fixed opacity. We show that a substantial negative feedback is not supported by an ensemble of high-resolution atmospheric models. Rather, the models suggest that changes in cloud area and opacity together act as a weakly positive feedback. The positive opacity component arises from the disproportionate reduction in the area of thick, climate-cooling clouds relative to thin, climate-warming clouds. This suggests that thick cloud area is tightly coupled to the rate of convective overturning—which is expected to slow with warming—whereas thin cloud area is influenced by other, less certain processes. The positive feedback differs markedly from previous estimates and leads to a +0.3 °C shift in the median estimate of equilibrium climate sensitivity relative to a previous community assessment.

This is a preview of subscription content, access via your institution

Access options

Buy this article

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: The tropical ice cloud continuum.
Fig. 2: Model representations of the ice cloud continuum.
Fig. 3: The ice cloud response to warming and its radiative effects.
Fig. 4: Updating the PDF of ECS.

Similar content being viewed by others

Data availability

The DARDAR-Cloud satellite products are available at and the 2C-ICE products at RCEMIP model output is publicly available at, and output from the SAM-P3 model runs is available from the corresponding author on request. The derived statistics needed to reproduce the figures in this paper, as well as output from the SAM-P3 model runs, is available at (ref. 61).

Code availability

The code used for the climate sensitivity calculations is available from the WCRP at (ref. 62). The code needed to generate the figures in this paper is available at (ref. 61).


  1. Sherwood, S. C. et al. An assessment of Earth’s climate sensitivity using multiple lines of evidence. Rev. Geophys. (2020).

  2. Zelinka, M. D. & Hartmann, D. L. Why is longwave cloud feedback positive? J. Geophys. Res. (2010).

  3. Bony, S. et al. Thermodynamic control of anvil cloud amount. Proc. Natl Acad. Sci. USA 113, 8927–8932 (2016).

    Article  CAS  Google Scholar 

  4. Hartmann, D. L., Gasparini, B., Berry, S. E. & Blossey, P. N. The life cycle and net radiative effect of tropical anvil clouds. J. Adv. Model. Earth Syst. 10, 3012–3029 (2018).

    Article  Google Scholar 

  5. Gasparini, B., Blossey, P. N., Hartmann, D. L., Lin, G. & Fan, J. What drives the life cycle of tropical anvil clouds? J. Adv. Model. Earth Syst. 11, 2586–2605 (2019).

    Article  Google Scholar 

  6. Hartmann, D. L. & Berry, S. E. The balanced radiative effect of tropical anvil clouds. J. Geophys. Res. 122, 5003–5020 (2017).

    Article  Google Scholar 

  7. Ito, M. & Masunaga, H. Process-level assessment of the iris effect over tropical oceans. Geophys. Res. Lett. 49, e2022GL097997 (2022).

    Article  Google Scholar 

  8. Kubar, T. L. & Jiang, J. H. Net cloud thinning, low-level cloud diminishment, and Hadley circulation weakening of precipitating clouds with tropical West Pacific SST using MISR and other satellite and reanalysis data. Remote Sens. 11, 1250 (2019).

    Article  Google Scholar 

  9. Saint-Lu, M., Bony, S. & Dufresne, J.-L. Observational evidence for a stability iris effect in the tropics. Geophys. Res. Lett. (2020).

  10. Saint-Lu, M., Bony, S. & Dufresne, J.-L. Clear-sky control of anvils in response to increased CO2 or surface warming or volcanic eruptions. NPJ Clim. Atmos. Sci. 5, 78 (2022).

  11. Lindzen, R. S., Chou, M. D. & Hou, A. Y. Does the Earth have an adaptive infrared iris? Bull. Am. Meteorol. Soc. 82, 417–432 (2001).

    Article  Google Scholar 

  12. Su, H. et al. Variations of tropical upper tropospheric clouds with sea surface temperature and implications for radiative effects. J. Geophys. Res. (2008).

  13. Zelinka, M. D. & Hartmann, D. L. The observed sensitivity of high clouds to mean surface temperature anomalies in the tropics. J. Geophys. Res. (2011).

  14. Choi, Y.-S. et al. Revisiting the iris effect of tropical cirrus clouds with TRMM and A-Train satellite data. J. Geophys. Res. 122, 5917–5931 (2017).

  15. Igel, M. R., Drager, A. J. & van den Heever, S. C. A CloudSat cloud object partitioning technique and assessment and integration of deep convective anvil sensitivities to sea surface temperature. J. Geophys. Res. 119, 10515–10535 (2014).

  16. Liu, R. et al. High cloud variations with surface temperature from 2002 to 2015: contributions to atmospheric radiative cooling rate and precipitation changes. J. Geophys. Res. 122, 5457–5471 (2017).

  17. McKim, B., Bony, S. & Dufresne, J.-L. Physical and observational constraints on the anvil cloud feedback. Preprint at Authorea (2023).

  18. Mauritsen, T. & Stevens, B. Missing iris effect as a possible cause of muted hydrological change and high climate sensitivity in models. Nat. Geosci. 8, 346–351 (2015).

    Article  CAS  Google Scholar 

  19. Chambers, L. H., Lin, B. & Young, D. F. Examination of new CERES data for evidence of tropical iris feedback. J. Clim. 15, 3719–3726 (2002).

  20. Lin, B., Wielicki, B. A., Chambers, L. H., Hu, Y. & Xu, K.-M. The iris hypothesis: a negative or positive cloud feedback? J. Clim. 15, 3–7 (2002).

  21. Li, R. L., Storelvmo, T., Fedorov, A. V. & Choi, Y.-S. A positive iris feedback: insights from climate simulations with temperature-sensitive cloud-rain conversion. J. Clim. 32, 5305–5324 (2019).

  22. Williams, I. N. & Pierrehumbert, R. T. Observational evidence against strongly stabilizing tropical cloud feedbacks. Geophys. Res. Lett. 44, 1503–1510 (2017).

    Article  Google Scholar 

  23. Wing, A. A. et al. Radiative–convective equilibrium model intercomparison project. Geosci. Model Dev. 11, 793–813 (2018).

  24. Berry, E. & Mace, G. G. Cloud properties and radiative effects of the Asian summer monsoon derived from A-Train data. J. Geophys. Res. 119, 9492–9508 (2014).

    Article  Google Scholar 

  25. Chen, Y.-W. et al. High cloud responses to global warming simulated by two different cloud microphysics schemes implemented in the Nonhydrostatic Icosahedral Atmospheric Model (NICAM). J. Clim. 29, 5949–5964 (2016).

  26. Ramanathan, V. et al. Cloud-radiative forcing and climate: results from the Earth radiation budget experiment. Science 243, 57–63 (1989).

    Article  CAS  Google Scholar 

  27. Hartmann, D. L., Moy, L. A. & Fu, Q. Tropical convection and the energy balance at the top of the atmosphere. J. Clim. 14, 4495–4511 (2001).

  28. Stephens, G. et al. CloudSat and CALIPSO within the A-Train: ten years of actively observing the Earth system. Bull. Am. Meteorol. Soc. 99, 569–581 (2018).

    Article  Google Scholar 

  29. Knutson, T. R. & Manabe, S. Time-mean response over the tropical Pacific to increased CO2 in a coupled ocean–atmosphere model. J. Clim. 8, 2181–2199 (1995).

  30. Held, I. M. & Soden, B. J. Robust responses of the hydrological cycle to global warming. J. Clim. 19, 5686–5699 (2006).

  31. Jeevanjee, N. Three rules for the decrease of tropical convection with global warming. J. Adv. Model. Earth Syst. 14, e2022MS003285 (2022).

    Article  Google Scholar 

  32. Singh, M. S., Kuang, Z., Maloney, E. D., Hannah, W. M. & Wolding, B. O. Increasing potential for intense tropical and subtropical thunderstorms under global warming. Proc. Natl Acad. Sci. USA 114, 11657–11662 (2017).

    Article  CAS  Google Scholar 

  33. Romps, D. M. Clausius–Clapeyron scaling of CAPE from analytical solutions to RCE. J. Atmos. Sci. 73, 3719–3737 (2016).

    Article  Google Scholar 

  34. Lilly, D. K. Cirrus outflow dynamics. J. Atmos. Sci. 45, 1594–1605 (1988).

    Article  Google Scholar 

  35. Jensen, E. J., van den Heever, S. C. & Grant, L. D. The life cycles of ice crystals detrained from the tops of deep convection. J. Geophys. Res. 123, 9624–9634 (2018).

    Article  Google Scholar 

  36. Schmidt, C. T. & Garrett, T. J. A simple framework for the dynamic response of cirrus clouds to local diabatic radiative heating. J. Atmos. Sci. 70, 1409–1422 (2013).

    Article  Google Scholar 

  37. Wall, C. J. et al. Observational evidence that radiative heating modifies the life cycle of tropical anvil clouds. J. Clim. 33, 8621–8640 (2020).

  38. Dobbie, S. & Jonas, P. Radiative influences on the structure and lifetime of cirrus clouds. Q. J. R. Meteorol. Soc. 127, 2663–2682 (2001).

    Article  Google Scholar 

  39. Höjgård-Olsen, E., Chepfer, H. & Brogniez, H. Satellite observed sensitivity of tropical clouds and moisture to sea surface temperature on various time and space scales: 1. Focus on high level cloud situations over ocean. J. Geophys. Res. 127, e2021JD035438 (2022).

  40. Stubenrauch, C. J., Caria, G., Protopapadaki, S. E. & Hemmer, F. 3D radiative heating of tropical upper tropospheric cloud systems derived from synergistic A-Train observations and machine learning. Atmos. Chem. Phys. 21, 1015–1034 (2021).

    Article  CAS  Google Scholar 

  41. Pierrehumbert, R. T. Thermostats, radiator fins, and the local runaway greenhouse. J. Atmos. Sci. 52, 1784–1806 (1995).

    Article  Google Scholar 

  42. Roe, G. H. & Baker, M. B. Why is climate sensitivity so unpredictable? Science 318, 629–632 (2007).

    Article  CAS  Google Scholar 

  43. Wing, A. A. et al. Clouds and convective self-aggregation in a multi-model ensemble of radiative-convective equilibrium simulations. J. Adv. Model. Earth Syst. (2020).

  44. Stauffer, C. L. & Wing, A. A. Properties, changes, and controls of deep-convecting clouds in radiative-convective equilibrium. J. Adv. Model. Earth Syst. 14, e2021MS002917 (2022).

    Article  Google Scholar 

  45. Hartmann, D. L. Tropical anvil clouds and climate sensitivity. Proc. Natl Acad. Sci. USA 113, 8897–8899 (2016).

    Article  CAS  Google Scholar 

  46. Gasparini, B. et al. Opinion: tropical cirrus—from micro-scale processes to climate-scale impacts. Atmos. Chem. Phys. 23, 15413–15444 (2023).

  47. Delanoë, J. & Hogan, R. J. Combined CloudSat–CALIPSO–MODIS retrievals of the properties of ice clouds. J. Geophys. Res. (2010).

  48. Cazenave, Q. et al. Evolution of DARDAR-CLOUD ice cloud retrievals: new parameters and impacts on the retrieved microphysical properties. Atmos. Meas. Tech. 12, 2819–2835 (2019).

    Article  CAS  Google Scholar 

  49. Deng, M., Mace, G. G., Wang, Z. & Okamoto, H. Tropical Composition, Cloud and Climate Coupling Experiment validation for cirrus cloud profiling retrieval using CloudSat radar and CALIPSO lidar. J. Geophys. Res. (2010).

  50. Winker, D. M. et al. Overview of the CALIPSO mission and CALIOP data processing algorithms. J. Atmos. Ocean. Technol. 26, 2310–2323 (2009).

    Article  Google Scholar 

  51. Stephens, G. L. et al. The CloudSat mission and the A-Train: a new dimension of space-based observations of clouds and precipitation. Bull. Am. Meteorol. Soc. 83, 1771–1790 (2002).

  52. Sokol, A. B. & Hartmann, D. L. Congestus mode invigoration by convective aggregation in simulations of radiative-convective equilibrium. J. Adv. Model. Earth Syst. (2022).

  53. Khairoutdinov, M. F. & Randall, D. A. Cloud resolving modeling of the ARM summer 1997 IOP: model formulation, results, uncertainties, and sensitivities. J. Atmos. Sci. 60, 607–625 (2003).

    Article  Google Scholar 

  54. Morrison, H. et al. Parameterization of cloud microphysics based on the prediction of bulk ice particle properties. Part II: case study comparisons with observations and other schemes. J. Atmos. Sci. 72, 312–339 (2015).

  55. Kang, H., Choi, Y.-S., Hwang, J. & Kim, H.-S. On the cloud radiative effect for tropical high clouds overlying low clouds. Geosci. Lett. 7, 7 (2020).

    Article  Google Scholar 

  56. Hong, Y., Liu, G. & Li, J.-L. F. Assessing the radiative effects of global ice clouds based on CloudSat and CALIPSO measurements. J. Clim. 29, 7651–7674 (2016).

  57. Kubar, T. L., Hartmann, D. L. & Wood, R. Radiative and convective driving of tropical high clouds. J. Clim. 20, 5510–5526 (2007).

  58. Cess, R. D. & Potter, G. L. A methodology for understanding and intercomparing atmospheric climate feedback processes in general circulation models. J. Geophys. Res. 93, 8305–8314 (1988).

  59. Soden, B. J., Broccoli, A. J. & Hemler, R. S. On the use of cloud forcing to estimate cloud feedback. J. Clim. 17, 3661–3665 (2004).

  60. Lee, J.-Y. et al. in Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) 553–672 (IPCC, Cambridge Univ. Press, 2021).

  61. Sokol, A. B. Code & data to accompany “Greater climate sensitivity implied by anvil cloud thinning” by Sokol et al. (2024). Zenodo (2024).

  62. Webb, M. Code and data for WCRP climate sensitivity assessment (corrected version, December 2022). Zenodo (2022).

Download references


We thank C. Stauffer for processing and sharing RCEMIP data, L. Hahn for helpful feedback on this manuscript and J. Deutloff for helpful conversations about the treatment of cloud overlap. We acknowledge the many scientists who provided simulations for RCEMIP and the German Climate Computing Center (DKRZ) for hosting the standardized RCEMIP data. This work was supported by NASA FINESST grant 80NSSC20K1613 (A.B.S. and D.L.H.) and NSF grant AGS-2124496 (D.L.H.).

Author information

Authors and Affiliations



A.B.S. conceived the project, conducted the analysis, generated the figures and wrote the manuscript. C.J.W. ran the equilibrium climate sensitivity code and provided interpretation. D.L.H. interpreted results, contributed to manuscript revision and supervised all aspects of the project.

Corresponding author

Correspondence to Adam B. Sokol.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Geoscience thanks Brian Soden, Claudia Stubenrauch and Aiko Voigt for their contribution to the peer review of this work. Primary Handling Editor: Tom Richardson, in collaboration with the Nature Geoscience team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Relationship between IWP and optical depth.

Mean ice cloud optical depth τ as a function of IWP in the three combined radar-lidar satellite retrievals.

Extended Data Fig. 2 Observed cloud macrophysical properties in IWP space.

Cloud fraction composited by IWP and height in DARDAR v2.1.1. The red line shows the mean cloud top height at each IWP, with red shading between the 10th and 90th cloud top height percentiles. Following ref. 50, cloudy volumes are defined as those with nonzero ice water content and visible extinction coefficient exceeding 0.125 km−1. Data are for 150-180E and 12S-12N for the 2009 calendar year.

Extended Data Fig. 3 Modeled cloud macrophysical properties in IWP space.

Cloud fraction composited by pressure and IWP for the final 15 days of the RCEMIP simulations with Ts = 300 K. Grid boxes are considered cloudy if the total condensate mixing ratio exceeds 10−5 kg/kg. Red lines show the median cloud top pressure (CTP) of ice clouds, with red shading between the 10th and 90th CTP percentiles. The CTP statistics do not extend all the way down to IWP=1 g/m2 because such low IWPs can only result from ice mixing ratios below the cloudiness threshold.

Extended Data Fig. 4 Mean cloud radiative effects across the RCEMIP ensemble.

Box plots of Cice, Cthick, and Cthin in the CRM ensemble for each value of Ts. Boxes show Q1-Q3, the horizontal lines within each box show medians, red triangles show means, and each dot shows an individual model (n = 11). Outliers are defined as points that differ from Q1 or Q3 by more than 1.5x the interquartile range.

Extended Data Fig. 5 Results for individual RCEMIP models.

(a) Absolute and (c) fractional changes in f between 295 and 305 K, normalized by ΔTs. (b) C at 295 K and (d) ΔfC between 295 and 305 K. Thin lines show individual models. Heavy black lines show multimodel means. Heavy red lines show the ensemble standard deviation and are plotted on a different vertical axis shown in red on the right side of each plot.

Extended Data Fig. 6 Cloud fraction statistics for individual models.

Cloud fraction as a function of Ts for each model in the CRM ensemble. Panels are shown for different portions of the IWP continuum. The bottom-right shows the ratio of thin to thick ice clouds.

Extended Data Fig. 7 Correlation coefficients between IWP-resolved and domain-averaged quantities.

Correlation coefficients are shown for various quantities. Black: absolute changes in f (IWP) and fice. Green: fractional changes in f (IWP) and fice. Pink: ΔfC (IWP) and ΔfCice.

Extended Data Fig. 8 Relationships between changes in cloud fraction metrics and various radiative feedback components.

Changes in fice, fthick, and fthin versus ΔfCice and its area- and opacity-related components. ΔfCice plotted against a, Δfice; b, Δfthick; and c, Δfthin. The area component of ΔfCice plotted against d, Δfice; e, Δfthick; and f, Δfthin. The opacity component of ΔfCice plotted against g, Δfice; h, Δfthick; and i, Δfthin. Correlation coefficients are shown for each relationship and are marked with an asterisk if not statistically different from zero, that is if the 95% confidence interval includes zero (n = 11). All values are normalized by ΔTs.

Extended Data Fig. 9 Sensitivity of the ECS PDF to feedback uncertainty.

The test values for the anvil area and opacity feedback are displayed as N(x, y), which represents a Gaussian with mean x and standard deviation y. The three RCEMIP-informed feedback estimates use the same mean value but different standard deviations, which, in increasing order, correspond to the standard deviation of the RCEMIP models, the maximum absolute difference between a single model and multimodel mean, and the original standard deviation assessed by the WCRP.

Extended Data Table 1 Statistics for the posterior PDFs of equilibrium climate sensitivity

Supplementary information

Supplementary Information

Supplementary Discussions 1 and 2, Figs. 1–4, and Tables 1 and 2.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sokol, A.B., Wall, C.J. & Hartmann, D.L. Greater climate sensitivity implied by anvil cloud thinning. Nat. Geosci. 17, 398–403 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing