Monday, April 01, 2013

Now Bjorn Lomberg Does It Too

Apparently Bjorn Lomberg can't read for comprehension either. On Facebook he writes:
The last hundred years is not only below the resolution of the reconstruction, but also not representative:

"the 20th century portion of our paleotemperature stack is not statistically robust, cannot be considered representative of global temperature changes, and therefore is not the basis of any of our conclusions."

It is worrying that they only tell us this *now*, after the story has been broadcast around the world.
Let's quote from the paper again:
Without filling data gaps, our Standard5×5 reconstruction (Fig. 1A) exhibits 0.6°C greater warming over the past ~60 yr B.P. (1890 to 1950 CE) than our equivalent infilled 5° × 5° area-weighted mean stack (Fig. 1, C and D). However, considering the temporal resolution of our data set and the small number of records that cover this interval (Fig. 1G), this difference is probably not robust.
The large red words are my emphasis, but perhaps they should have appeared that way in the paper too, so Lomborg would have seen them.


Anonymous said...

There is no known cure for wilful ignorance. If Lomborg is against it, it is probably true.

Louis Hooffstetter said...

"The large red words are my emphasis, but perhaps they should have appeared that way in the paper too...

I agree! They should have appeared that way in the press releases too!

Frank said...

Scientists are expected to include error bars or other methodology on their graphs so that readers can see at a glance which features are robust and which are not. Marcott failed to do so, misleading many about the significance of the 20th century spike. Their press release and public comments compounded this mistake.

Frank said...

David: You correctly complain that Steve McIntyre's blog articles are not peer-reviewed science. They are, however, "peer reviewed" by literally hundreds of climate scientists and highly educated readers who would love to catch Steve McIntyre making a serious mistake. He usually posts the computer code used to generate the figures in his posts, so his work is easy for others to check. Some of his followers use and build on the code he provides, thereby checking his work. He also usually posts links to the original work he criticizes and to prominent criticism of his work, so it's easy to get both sides of the story - something every good reporter and scientist should do. He doesn't censor his critics. For example, scientist Richard Telford has raised serious questions today that IMO cast doubt upon some aspects of what McIntyre has written about the re-dating controversy. If McIntyre doesn't reply with a follow up post or comment, I'll ignore the re-dating controversy. McIntyre responds to expert criticism of his science. This process provides SOME reassurance that posts at Climate Audit (and Lucia's BLackboard) have scientific validity. I seriously doubt that the peer review the average paper receives at a major journal is more rigorous than the review McIntyre receives at his web site every day. Both forms of peer reviewed science are far from perfect.

In contrast, "scientific" posts at WUWT and other skeptical sites are a totally different. The quality is extremely uneven. The comments section is dominated by politically motivated comments that drown out any serious questions or discussion. Expert commentators from ClimateAudit or the Blackboard rarely spend much time questioning posts and practicing climate scientists rarely say anything. Here it's buyer beware.

Political motivations and McIntyre's insolent style make it extremely difficult for McIntyre to get his original work or critical comments published in mainstream journals. If you don't believe this, read his story about the comment he failed to get published concerning Santer et al's successful rebuttal of Douglas and Christy's paper about the statistical significance of the absence of a hot-spot in the upper tropical troposphere. McIntyre and co-authors showed that Santer's OWN METHODS showed a statistically significant difference between observations and model predictions when data from 2000-2007 was added to the older data used by both Douglas and Santer. Peer review didn't challenge the correctness of McIntyre's work, but it made publication impractical. Under these circumstances - which were certainly partially created by McIntyre himself - you are grossly incorrect to conclude that McIntyre's work is not science because it isn't peer-reviewed science. In the future, open-access journals and crowd-sourced peer review may replace the obsolete, expensive, pay-walled, and non-transparent publication process controlled by elite members of some scientific societies. In the meantime, ClimateAudit provides some useful science that you can't find elsewhere. If you would stop kissing Michael Mann's posterior and smell some fresh air, perhaps you'd learn something new.

David Appell said...

Frank: Marcott has plenty of error bars and error bands. See Figures 1 and 2.

David Appell said...

Frank: Blog posts are not science: