Head in a Cloud

troposphere and stratosphere meet blogosphere

Head in a Cloud random header image

Comments on “An important constraint on tropical cloud – climate feedback” by Hartmann and Larson

February 9th, 2007 by Jamison A. Smith, Ph. D. · 1 Comment

I think this paper is very interesting, but I have to admit that I don’t understand it very well. Hopefully, this post will spur some useful discussion.

Here’s the citation info:

Hartmann, D. L. and K. Larson, “An important constraint on tropical cloud – climate feedback,” Geophys. Res. Lett. 29, doi:10.1029/2002GL015835, 2002.

The authors’ primary conclusion is “…that the temperatures at the detrainment level [of deep tropical convection], and consequently the emission temperature of tropical anvils, will remain unchanged during climate change.”

This conclusion surprised me.

As the oceans warm, the absolute humidity of the overlying atmosphere should increase. This increase in moisture provides more latent heat to drive convection. Thus, the detrainment level of convection should rise in a global warming scenario.

The authors argue, however, that the detrainment level of convection is controlled by radiative cooling in the upper troposphere. Radiative cooling (primarily by water vapor) generates the atmospheric instability that produces convection. Above 200 hPa, the concentration of water vapor is so low that it becomes ineffective at cooling the atmosphere through radiative emission. The lack of cooling above 200 hPa prevents convection from continuing upward. The authors use an atmospheric model to show that the clear-sky radiative heating profile is insensitive to changes in sea surface temperature.

Some comments:

- The radiative heating profile (Fig. 2) should have been plotted with a logarithmic y-axis. Contrary to the text, the profiles are sensitive to the sea surface temperature and show a systematic response in the upper troposphere (150-400 hPa) and the lower troposphere (700-1000 hPa).

- Cloud top temperatures also show a systematic decrease with an increase of sea surface temperature (Fig. 3). A reduction of cloud top temperature from 201 K to 198 K (resulting from an increase of sea surface temperature from 297 K to 303 K) would produce a decrease in outgoing longwave radiation of 6% which amounts to a forcing of tens of Watts per square meter over the convection. I disagree with the authors’ conclusion that “…longwave emission from anvil clouds does not change much with increasing SST.”

- The horizontal resolution of the model is 160 km, so convection must be parameterized. Will real convection behave this way? Maybe someone has done a similar simulation with resolved convection. The authors state that there are upward vertical motions in the convective regions of the model and downward subsidence in the clear regions. I don’t understand how this model can have appreciable vertical motions at this resolution. I thought that convective parameterizations merely shuffled constituents within a column with zero net vertical motion.

Tags: climate · modeling

1 response so far ↓

  • 1 seand // Feb 14, 2007 at 11:26 am

    Hey Jamie,

    Have you looked to see what works cite this work? I’m wondering if there have been updates on this type of work, and if they still agree with the main conclusions given here. (?) s

Leave a Comment