Background
Many coupled general circulation models (GCMs) of the atmosphere tend to underpredict the presence of subtropical marine stratocumulus clouds and fail to predict the seasonal cycle of such clouds.  These deficiencies are important because marine stratocumulus clouds have a major impact on sea surface temperatures below them; and they thereby impact many subsidiary phenomena with far-reaching climatic consequences.

What was done
The authors attempted to determine some of the consequences of these climate model deficiencies by examining the tropical response of a coupled GCM to quasi-realistic specified marine stratocumulus clouds and comparing the results to what they obtain from their model when it operates in its normal mode and fails to adequately express the presence of these clouds or their effects.

What was learned
In the words of the authors, "the annual mean and annual cycle of the low cloud fraction are both severely [our italics] underpredicted in the southeastern tropical Pacific in our control run."  Just how important this fact is may be appreciated by considering what happens in the western tropical Pacific when they remove these low clouds, as occurs in the normal application of their model: "the sea surface temperature warms [approximately] 5.5°C."

What it means
Until such cloud resolution problems as these - which tend to produce much warmer sea surface temperatures than what actually occur - can be resolved, current-generation GCMs will not be able to adequately model many phenomena of great importance to climate change.  Hence, their utilization at present cannot be construed to give much more than a crude qualitative picture of future possibilities.  To make decisions of global economic importance based on such an unfirm foundation is truly dangerous.