Recent Advances in Understanding Cloud Feedbacks

As the greater PCC community focused on climate feedbacks as the theme of the year for 2010, Mark Zelinka, one of UW's graduate students in Atmospheric Sciences, completed his Ph.D. dissertation on the topic. As we bid goodbye to Mark (see more on this in the winter 2011 PCC newsletter), we take the opportunity to highlight some of the results from the work of Mark Zelinka and Mark's thesis advisor, Dennis Hartmann, on cloud feedbacks.

Climate models suggest that in a warming climate clouds will act to further heat the planet. A recently published result of Mark Zelinka's thesis shows that clouds in models are behaving as one would expect from basic physics. Zelinka and Hartmann (2010) show that for one aspect of cloud feedback, that involving longwave or infrared radiation, the positive feedback that every model produces is simply due to the fact that clouds rise in a warming climate. Another recent study that got a good deal of press was by Andrew Dessler of Texas A&M University. Dessler used analyses of observations to lend credence to the positive feedback of clouds in climate models. Dessler's work shows that clouds in models do things that are consistent with what clouds do in nature.

A Ph.D. dissertation contains years of detailed, cutting edge study. Another chapter of Mark Zelinka's thesis explores 8 years of data from a suite of satellite instruments to investigate how tropical high clouds change as the Tropics warm and cool from year-to-year. The observational data suggest that clouds change as the Tropics warm in such a way as to allow further heating. One can think of this as a positive cloud feedback that operates on interannual timescales. Such inferences from observations are based on year-to-year natural variability, much of which is associated with El Nino - La Nina cycles. The cloud feedback for these natural variations are expected to be different from that associated with global warming which occurs over a much longer time frame. Studying these natural variations and their simulations in models can be used to test and even extend our understanding of the basic physics of cloud feedbacks.

Another chapter looks at energy transport from equator to pole. Mark found that the equator-to-pole gradient in the net (incoming minus outgoing) energy is altered on a warming planet due to the rich spatial structures of the various feedbacks. Specifically, more heat has to be fluxed poleward due to positive water vapor and cloud feedbacks, which are most strongly positive at low latitudes. Zelinka (and Hartmann) showed that the large uncertainties in cloud feedback directly translate into large uncertainties in poleward heat transport, and that the sluggish response of the underlying ocean results in the atmosphere having to carry more of the burden of transporting heat poleward as the planet warms.

The last section we'll highlight here was done in collaboration with Steve Klein, a former student of Dennis Hartmann, and Mark's present post-doctoral supervisor. They devised a new way to compute cloud feedback in climate models. Cloud feedback can be broken into components: (i) that due to changes in cloud fraction independent of changes in cloud optical properties and vertical distribution, (ii) that due to changes in cloud optical properties independent of changes in total cloud fraction or vertical distribution, and (iii) that due to changes in cloud vertical distribution independent of changes in total cloud fraction or optical properties. Such a breakdown allows for a much more detailed attribution of the types of cloud changes that give rise to the feedback. Using this technique they showed that rising clouds at all latitudes contribute substantially to positive longwave cloud feedback. They also demonstrated one aspect of the models that is fairly robust -- that cloud feedback at high latitudes are locally negative -- is due to clouds getting brighter rather than an increase in total cloud fraction. Because it is fairly easy to implement, the three are confident that this technique will be widely useful in identifying the processes responsible for causing cloud feedbacks, as well as their strength and uncertainty across models.

Look for additional publications from this work in the near future.

Acknowledgements: Many thanks to Mark Zelinka for much of the content above.

The scientific literature:

Zelinka, M.D., 2010: Towards an Improved Understanding of Cloud Feedbacks and Changes in Poleward Energy Transport Associated with Global Warming. Ph.D. Thesis University of Washington.

Zelinka, M.D. and D.L. Hartmann, 2010: Why is Longwave Cloud Feedback Positive? J. Geophys. Res., 115, D16117, doi:10.1029/2010JD013817. Selected as an "editors' highlight" by agu! Press Release:

A.E. Dessler. A determination of the cloud feedback from climate variations over the past decade. Science. Vol. 330, December 10, 2010, p. 1523

Additional background on cloud feedbacks:

Webcast "Clouds: The Wild Card of Climate Change" with David Randall, Colorado State University.

(Hartmann comments on Dessler's contribution)

Full Story