Skip to content

Paper Shows Climate Models Underestimate Cooling Effect From Clouds By A Factor Of 4

Using satellite observations, the paper shows that the feedback from low clouds is indeed negative and is underestimated in climate models by a factor of four.
A paper published in the technical newsletter of the Global Energy and Water Cycle Experiment finds that climate models suppress the negative feedback from low clouds, which serve to cool the Earth by reflection of incoming sunlight. The paper notes that cloud feedbacks in computer models are not only uncertain in magnitude, but even in sign (positive or negative).
As climate scientist Dr. Roy Spencer has pointed out, a mere 1 to 2% natural variation in cloud cover can alone account for whether there is global warming or global cooling, despite any alleged effects of CO2.

Using satellite observations, the paper shows that the feedback from low clouds is indeed negative and is underestimated in climate models by a factor of four. This has the effect of the models greatly overestimating global warming from CO2 and underestimating the influence of variations of the Sun/cosmic rays via cloud formation.

Is There a Missing Low Cloud Feedback in Current Climate Models?

Graeme L. Stephens
Department of Atmospheric Science, Colorado State University, Boulder, Colorado, USA

Radiative feedbacks involving low level clouds are a primary cause of uncertainty in global climate model projections. The feedback in models is not only uncertain in magnitudebut even its sign varies across climate models (e.g., Bony and Dufresne, 2005). These low cloud feedbacks have been hypothesized in terms of the effects of two primary cloud variables—low cloud amount and cloud optical depth. The basis of these feedbacks relies on the connection between these variables and the solar radiation leaving the planet exemplified in the following simple expressions  (Stephens, 2005). …an increase in optical depth with an increase in temperature results in an increase in cloud albedo, suggesting a negative feedback.

The net consequence of these biases is that the optical depth of low clouds in GCMs (General Circulation Models) is more than a factor of two greater than observed, resulting in albedos of clouds that are too high. This model low-cloud albedo bias is not a new finding and is not a feature of just these two models. The study of Allan et al. (2007), for example, also noted how the reflection by low-level clouds in the unified model of the UK Meteorological Office is significantly larger than matched satellite observations of albedo, suggesting that this bias also exists in that model. The mean LWP (cloud liquid water path) of model clouds that contributed to this in the most recent Intergovernmental Panel on Climate Change assessment is close to 200 g/m2, which is also nearly a factor of two larger than observed.

The implication of this optical depth bias that owes its source to biases in both the LWP and particle sizes is that the solar radiation reflected by low clouds is significantly enhanced in models compared to real clouds. This reflected sunlight bias has significant implications for the cloud-climate feedback problem.  The  consequence is  that   this  bias  artificially suppresses the low cloud optical depth feedback in models by almost a factor of four and thus its potential role as a negative feedback. This bias explains why the optical depth feedback is practically negligible in most global models (e.g., Colman et al., 2003) and why it has received scant attention in low cloud feedback discussion. These results are also relevant to the model biases in absorbed solar radiation discussed recently by Trenberth and Fasullo (2010) and as explored in more detail in Stephens et al. (2010).

The Hockey Schtick, 7 July 2011