Currying confusion

Dr. Judith Curry is chair of the School of Earth and Atmospheric Sciences at Georgia Tech, and has an extensive background in studying Earth's climate, particularly regarding changes in storms, hurricanes, and the like under changing climate conditions. She recently coauthored what seems a very interesting paper on the growth of Antarctic sea ice - apparently the effect of a moderate degree of warming such as we've seen so far is to actually increase sea ice extent in the southern ocean, thanks to increased precipitation in the form of snow. Higher sea temperatures mean more evaporation of water (mostly closer to the equator) which in turn leads to higher levels of precipitation (mostly further south), and if it's cold enough to snow, then paradoxically the result is actually more ice on the water surface, not less. There is also a salinity reduction effect that helps maintain the ice. This result indicates there are some conditions under which the ice-albedo feedback effect is actually negative: warming producing more ice which results in more reflection of sunlight from Earth's surface, and a cooling response. Of course, if things warm too much the snow will turn to rain and the ice will go away, so the effect is only temporary.

So that's the sort of good scientific work Dr. Curry has continued to be involved in. However, she has also been making appearances around the "blogosphere" and in public lectures that have conveyed a very different message - essentially that there are huge uncertainties in our understanding of Earth's climate, casting doubt on essentially every result presented by the IPCC in their reports. In particular, Dr. Curry has strongly criticized IPCC discussion of paleoclimate (a field quite remote from her own expertise) and promoted the work of various contrarians - the long back and forth in this thread at RealClimate is strongly illustrative, though part of that debate bounced around to other blogs along the way. In particular, in this thread Gavin Schmidt and Judith Curry had a long exchange on the uncertainty questions she was raising, trying to get a clear detailed statement of what she saw as the problem. I'm not an expert in most of that, but as far as I could tell she was not able to come up with any specifics that were actually valid and relevant to the question, and retreated to vague epistemology (or "ontic") issues in the end.

But in the middle of that there was a claim she raised on a subject that I have read up on and thought I understood very well myself - the no-feedbacks climate sensitivity, #76 of Monckton's errors in his Physics and Society article. The quantity is well defined and determined strictly by radiative considerations, similar to the forcing from changes in greenhouse gas concentrations. And it is actually known to considerably greater precision than the forcing or other factors - to within 2 percent or so (forcings are only known to within about 10%). But Curry seemed to get this quite backwards in her claim (comment #205 on the above thread):

Carbon dioxide is increasing in the atmosphere, fossil fuels are contributing in a dominant way, but there are many uncertainties in the global carbon budget in terms of sources and sinks. But overall, these are two “pillars.” After that, the uncertainties become greater. How much greenhouse gas warming (in the absence of feedbacks) actually warms the earth has been insufficiently accurately calculated IMO; whereas the radiative forcing is well understood, the translation of this into surface temperature change requires including the complexity of the land surface, ocean and ice/snow surface processes, in the context of everything else going on (even without feedbacks) is much more complicated than the simple calculations that give 1C sensitivity to doubled CO2. Is the 1C reasonable? Sure. But we cannot have a high confidence in this until we’ve done a better job of actually calculating it. And from there, things become more uncertain. The climate system is very complex, it is not easily understood.

Somehow she seemed to think the warming response (in the absence of feedbacks) depended on changes in "land surface, ocean and ice/snow", whereas I thought it was a purely radiative issue. Was there some difference in our definitions? I asked, and she responded (eventually) that she was referring to the same standard definition of no-feedbacks response. So what could she mean?

It is true that under conditions of "radiative forcing", when the energy received by our planet is out of balance with the energy it sends back out into space, that means there is a process of absorption of energy by the planet. That process of energy increase may not even involve any warming - when ice melts, it stays at the same temperature, just absorbing energy to change from solid to liquid state. And there could be various processes (such as the increasing ice area in the negative ice-albedo feedback in Curry's recent paper, or changes in cloud behavior) that reduce the energy imbalance without strictly requiring surface warming. These factors mean that the rate at which the surface warms up may be slower or faster depending on complex response processes. Most importantly, the ocean warms slowly because of its huge heat capacity - we expect the full warming response to any radiative imbalance to take at least several decades, just due to the way the ocean stores energy.

Nevertheless, if the question is what the final "equilibrium" response of the planet is "without feedbacks", then we need to be clear on the partitioning between various different "feedbacks" and a "bare" response, without feedbacks. The feedbacks represent different responses of Earth's climate system to energy flow changes - like changing ice, changing water vapor levels in the atmosphere, changing cloud patterns, etc. But what does "no feedbacks" mean?

There is a standard definition of the response "without feedbacks", referenced by the IPCC which is provided by Bony et al (Journal of Climate 19:3445 (2006)), appendix A. It is a physical quantity defined by the atmospheric temperature profile and radiative properties:

The most fundamental feedback in the climate system is the temperature dependence of LW emission through the Stefan-Boltzmann law of blackbody emission (Planck response). For this reason, the surface temperature response of the climate system is often compared to the response that would be obtained [...] if the temperature was the only variable to respond to the radiative forcing, and if the temperature change was horizontally and vertically uniform [...]

I.e. the response without feedbacks is quite strictly defined as the response under which "temperature [is] the only variable" that changes, and it changes by the same amount throughout the surface and troposphere (changes in temperature gradient, by contrast, are referred to as the lapse rate feedback).

This is a theoretical construct because there are always feedbacks that differ from this bare response in a real climate system. But note in particular it doesn't care *how* the temperature changed, or how long it took. It could be instantaneous, or it could take 1000 years. It could change just the temperature of just a few meters of the ocean or it could raise the entire ocean depth's temperature. What this response does is tell us is the quantity by which you need to raise temperatures (while changing nothing else) to restore balance at the top of the atmosphere, in response to a radiative forcing, so the planet is no longer gaining or losing energy.

And this response is relatively straightforward to calculate for a given collection of atmospheric temperature profiles. The strict calculation is easily done in models, as in the 2006 paper by Soden and Held referenced by Bony et al, where Table 1 provides feedback parameters including the "no-feedback" "Planck" response for 14 different coupled ocean-atmosphere models, and Table 2 provides similar numbers from the coupled model intercomparison project (CMIP II). The entire range of the numbers is from -3.13 to -3.28 W/Km2, with a mean of -3.216 and standard deviation of 0.04. See also the discussion of this parameter, referred to as λ0, in the paper.

What that means is that doubling CO2, which adds a radiative forcing of 3.7 W/m^2 (plus or minus about 10%) would cause a bare "without feedbacks" warming of (3.7/3.216 K) or about 1.15 K (plus or minus about 10%). The uncertainty in the no-feedbacks response is essentially all determined by the uncertainty in the forcing itself, not in the value of the (theoretical) no-feedbacks response parameter.

Where all the uncertainties in climate science lie is in the feedbacks, and the complex land and ocean processes Curry refers to play a critical role in those uncertainties. But on the "without feedbacks" number, as far as I can tell, she was just plain wrong.

Is it a wonder that so many people are confused about this stuff?


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

I read somewhere that an

I read somewhere that an early (like 1994 or so) GCM predicted the exact effect we are now seeing in the southern ocean: warming->more snow->more ice. Does anyone have a cite to that? I'd hate to think I'm delusional :-)
Presumably this recent paper from Curry references the one I think I remember - but I only have the Discover report, not the paper itself.



has a mention of the 1992 Manabe work that indicated that Antarctic sea ice barely changed even under 2XCO2.

She also believes that

She also believes that overall sensitivity is probably low, and specifically assert that a Charney sensitivity well below 2C is plausible.

In a recent thread at Kloor's I pointed out that recent deep-time paleo work strongly argued against this, and her response was to point me to a recent review paper (Hegerl co-auth) she said agreed with her. I looked and... no, it didn't. This kind of sloppiness is very hard to excuse.

She doesn't understand that

She doesn't understand that in a large group of people who read blogs a few will actually follow the links and report back.