Three Key Processes Drive Marine Low-Cloud Cover Feedback

Abstract:

The tropical marine low-cloud cover (LCC) feedback is a key component of global cloud feedbacks. Differences in simulations of the feedback (ranging from a considerably negative to large, positive) are sources of significant spread in the temperature responses of climate models to anthropogenic forcing. Despite its importance, the underlying processes driving the feedback are not well-understood. Here we present evidence that in models it is mainly driven by three warming-induced changes---(1) strengthening tropical inversion, (2) increasing surface latent heat flux, and (3) an increasing moisture gradient in the lower-troposphere. The sensitivities of LCC to these changes, diagnosed through 20th-century climate variability, vary a great deal across models, contributing to the spread in the projected LCC changes over 21st-century. A methodology is devised to observationally constrain these sensitivities. We find that averaged over five main tropical low cloud regions, LCC decreases by 2% or more in models where the simulated sensitivities are consistent with the observed. This suggests that a positive LCC feedback is more consistent with observation than a negative one. Correcting biases in the simulated sensitivities will clearly be an important step towards a better simulation of cloud feedbacks in climate models.