Skip to main content
U.S. flag

An official website of the United States government

Publication Date
17 March 2021

Investigating the Impact of Cloud-Radiative Feedbacks on Tropical Precipitation Extremes

Print / PDF
Powerpoint Slide

Extreme precipitation is an important aspect of climate variability and directly impacts society by causing flooding and other water management challenges. Yet climate models struggle to accurately capture extreme precipitation. This study shows one robust aspect of extreme precipitation across models: extreme precipitation over tropical oceans is strengthened through positive feedback with cloud radiative effects. This connection is shown for a multi-model ensemble with experiments that make clouds transparent to longwave radiation. In all cases, tropical extreme precipitation reduces without cloud radiative effects. Qualitatively similar results are presented for one model using the cloud-locking method to remove cloud feedbacks. The reduced extreme precipitation without cloud radiative feedbacks does not arise from changes in the mean climate. Rather, evidence is presented that cloud radiative feedbacks enhance the organization of convection.


Other studies have shown that cloud radiative effects impact convection and that organized convection produces extreme precipitation, but this is the first to show that in climate models the clouds impact extreme precipitation. By removing the radiative impact of clouds, extremes precipitation is reduced. Evidence is presented that the effect is not due to changes in the mean climate but through the impact that clouds have on organized precipitation. The clouds appear to enhance tropical variability in moisture modes like the Madden-Julian Oscillation and tropical depressions, and extreme precipitation events appear to be more common and more compact with cloud effects. This indicates that errors in representing clouds may contribute to errors in representing hydrologic extremes.


Clouds couple to the atmospheric circulation through latent heat release and by altering the radiative heating. This study investigates what happens when the radiative heating effects are removed. Across model configurations this results in a decrease in the fraction of precipitation that is extreme, showing that cloud radiative effects make extreme precipitation more extreme. The mechanism appears to be through cloud radiative feedbacks that promote the organization of convection. This study shows the importance of correctly representing clouds in climate models and also the importance of correctly capturing the emergent organization of convection. This result may be important as models are used with higher resolution and organized convective events become better resolved.

Point of Contact
Brian Medeiros
National Center for Atmospheric Research (NCAR)
Funding Program Area(s)