The Australian's War on Science XXII

This column in the Australian from Frank Devine is mainly about how the latest Disney cartoon is "pernicious and propagandistic" and threatens our freedom, but he also includes some war-on-science stuff:

An alternative view of good coming out of the financial crash is that, with another global threat to worry about, we may be less attentive to assertions as axiomatic that we are guilty of causing catastrophic climate change.

It is a gross misrepresentation to pretend that anthropogenic global warming and its impacts are just axioms. See the various IPCC reports.

In fact, hundreds of reputable scientists do not accept this, although regrettably nearly all media unquestioningly does, with the exception in this country of The Australian, which frequently publishes dissenting opinion on its oped page, and Quadrant, whose editor Keith Windschuttle recently launched a monthly series of "denial" articles.

But strangely enough these "hundreds of reputable scientists" don't publish in peer-reviewed scientific journals. While Quadrant promotes pseudo-science: HIV/AIDS denial, evolution denial and the DDT ban myth.

In Quadrant's present issue, William Kininmonth, former head of Australia's National Climate Centre, argues in a dense 6000-word paper that computer models on which the UN's Intergovernmental Panel on Climate Change based its 2007 warnings of coming disaster are useless as predictive instruments. They are unplugged, he writes, by "gross under-specification" of the cooling effect of evaporation from the oceans as temperatures rise.

Well, that doesn't make a lot of sense. Evaporation does have a cooling effect, but it's precisely matched by the warming effect when the water vapour condenses. Lets look at Kininmonth's dense paper:

The magnitude of the down-welling long-wave radiation at the surface increases as the concentration of carbon dioxide increases. There is a corresponding rise in surface temperature that is constrained by the increase in surface energy losses (radiation emission, conduction and evaporation of latent heat). The earth's surface is predominantly water or well-vegetated land; increasing evaporation of latent heat is a dominant factor in the additional energy loss under carbon dioxide forcing. It is the additional evaporation of latent heat that will constrain surface temperature response to human-caused carbon dioxide emissions.

The mathematical formulation of surface temperature response to carbon dioxide forcing is straightforward, even considering water vapour feedback. For a doubling of carbon dioxide concentration the global average surface temperature increase from 280 ppm (before industrialisation) to 560 ppm (towards the end of the twenty-first century) will only be about 0.5°C.

But the no feedback sensitivity is 1K. Considering water vapour feedback will increase that, not decrease it. Soden and Held (2006) looked at the feedbacks of different factors in the models. While more evaporation will decrease the lapse rate and that produces a negative feedback, this is more than counterbalanced by the positive feedback from more water vapour in the atmosphere. Maybe Soden and Held are wrong, but Kininmonth provides no reason to believe so -- no equations or models or references to papers. Kininmonth continues:

Why then, it might be asked, do computer models give projected temperature increases that are nearly an order of magnitude larger; and why are there claims of dangerous "tipping points" and potentially runaway global warming?

Note that a climate sensitivity of 3K does not just come from computer models -- looking at climate change in the past gives a similar number.

The likely answer to these questions is the recent revelation that computer models grossly under-specify the rate of increase of evaporation with temperature, the factor that constrains surface temperature increase. In 2006, US researchers Isaac Held and Brian Soden reported that, on average, the rate of increase of evaporation with temperature in computer models used for the IPCC fourth assessment is only about one third of the expected value. In 2007, Frank Wentz and his US colleagues repeated the earlier finding and, on the basis of satellite analysis of rainfall, confirmed the expected rate of increase of evaporation with temperature as the appropriate value.

Oh wait, he does provide two cites. Held and Soden (2006) is a different paper from Soden and Held (2006) that I cited above, but they don't think that it contradicts their own paper on feedbacks. That's just Kininmonth's odd interpretation. And similarly, if you read Wentz et al, 2007, you'll discover that the authors don't share Kininmonth's belief that their results imply that there will be no significant warming.

All right, Back to Devine's piece:

Windschuttle's - and my - interest in fence-rattling about the "axiom" of carbon emissions causing climate calamity was aroused by the huge effort to stifle opposing opinion. It is scientific principle that dissent be engaged, not silenced.

Apparently Devine thinks that if he expresses his opinions at length in the Australian he is somehow still being stifled.

The axiomists' conduct can therefore be judged unprincipled.

Yeah, axiomists, how dare you silence brave Frank Devine!