Skip to main content
U.S. flag

An official website of the United States government

Publication Date
6 May 2016

Compensation between Cloud Phase Partitioning and Cloud Cover in GCMs and Its Implications for Cloud Feedback

Print / PDF
Powerpoint Slide
Science

Global climate models struggle to properly simulate the relative amounts of liquid and ice in “supercooled” clouds below freezing. The authors investigated the implications of how phase is partitioned on mean state clouds and cloud feedback across 26 global climate models.

Impact

The results suggest that the across-model compensation between subtropical amount feedback and extra-tropical optical depth feedbacks may be an artifact of model tuning. This may be artificially suppressing inter-model spread in climate sensitivity.

Summary

Models whose supercooled clouds contain less liquid and more ice tend to have greater cloud coverage. This is surprising given that ice crystals should precipitate more readily and decrease cloud cover. Compensation between increased ice concentrations (which decreases reflectivity) and increased cloud fraction (which increases reflectivity) leads to roughly the same amount of sunlight being reflected by all the different GCMs, and points toward model tuning as a cause. This apparent tuning also affects cloud feedbacks in the models. In "icier" models, the cloud optical depth feedback in regions with abundant mixed-phase clouds is more negative (which makes sense physically), but the cloud amount feedback in the subtropics is also more positive owing to larger decreases in low cloud coverage.  The latter effect appears to be due to the aforementioned tuning between cloud cover and mixed-phase partitioning in the current climate.  Importantly, it suggests that the across-model compensation between subtropical amount feedback and extra-tropical optical depth feedbacks -- which acts to reduce inter-model spread in climate sensitivity – may be an artifact of model tuning.

Point of Contact
Mark Zelinka
Institution(s)
Lawrence Livermore National Laboratory (LLNL)
Funding Program Area(s)
Publication