It is time to stop quivering in our boots in pointless fear of the future and just roll up our sleeves and build it.
- Ray Pierrehumbert

Thursday, November 12, 2009

Deficit of Record Cold, Surfeit of Inexpertise

Dot Earth reports
Scientists sifting for trends in record high and low temperatures across the United States have found more evidence of long-term warming of the climate, with the biggest shift coming through a reduction in record low nighttime temperatures. That is a pattern long predicted by climate scientists using computer simulations.
Nevertheless, comment #2 reads, in full:
re: "with the biggest shift coming through a 'reduction' in record low nighttime temperatures."

That is evidence for land change usage, not an increase in CO2 concentrations.
I replied:
The first place to look for observational confirmation among extreme events would indeed be a smaller incidence of nighttime record lows without an accompanying increase of daytime record highs.

Record lows occur on clear nights when the surface radiates to space and the surface cools quickly. Greenhouse gases directly interfere with that process. Therefore all else equal less severe low temperatures are to be expected.

Climate models of every stripe, as far as I know, generally agree that this is the case.

On the contrary, direct heating equally would increase both record highs and record lows, not substantially affecting the proportion.

It is because the first order effect of CO2 is to suppress surface cooling rather than to enhance surface warming that this is indeed a reasonable place to look for observational confirmation.

Comment #2, being among the earliest comments, is likely to be widely read. Already it has been recommended by 3 readers. This site does not allow for a thumbs down vote, nor for extra weight on actually informed opinion. Thus, Dot Earth, despite the best of intentions, contributes to the vast array of misinformation on the subject.

We need ways to filter information based on the competence of the speaker to address the subject at hand. I am sure "Mac" knows things about which I am ignorant, would be ill at ease to see me making wild guesses about such topics, and very uncomfortable seeing them gain prominence by the approval of people who share my ill-informed prejudices.

It would be appropriate if "Mac" expressed himself with an appropriate level of humility for his knowledge on the subject, and the "recommenders" who must be equally ill-informed did the same. Since people lack such restraint, and since the supply of inexpertise is vastly larger than the supply of expertise, the public is making very poorly informed judgments about this issue and others where complex information is at issue and genuine expertise should matter.


Paul said...


I was sucked into this Dot Earth post earlier today.

I seldom read Dot Earth because Revkin still strains too hard for "balance" at the expense of science. The comments are generally a travesty. Once in a while a scientist involved in the particular issue posts and that is of value. The rest is mostly crap. The vote for posts is particularly galling.

If it's any consolation, I returned to Dot Earth and waded through the gibberish to your post and voted for it. Unfortunately Mac is still leading you 4 to 2 but wmar is way in the lead.


Anna Haynes said...

> We need ways to filter information based on the competence of the speaker to address the subject at hand.

That's what the "editor's selections" tab is supposed to be for; only the editor hasn't selected any comments.

> The comments are generally a travesty....The vote for posts is particularly galling.

Yes. Dot Earth was built to be wide open to parasitism, and lo, it got overrun.

David B. Benson said...

At last count its

MT --- 15
Mac --- 6

Might be some hope after all.

Michael Tobis said...

I just brazenly recommended my own article, putting it firmly in first place on the recommended list at 16.

David B. Benson said...

Michael Tobis --- I do that all the time on DotEarth and don't feel the slightest twing of guilt.