May 16, 2008

August 17, 2007

NASA Flacks for Global Warming and Skirts Scientific Ethics

By James Lewis
As pointed out in these pages, NASA has yet to own up fully to its historic error in misinterpreting US surface temperatures to conform to the Global Warming hypothesis, as discovered by Stephen McIntyre at ClimateAudit.org. This is not the first major error discovered by McIntyre and his coworker, Canadian economist Ross McKintrick, who previously uncovered the fatally flawed "hockey stick" climate curve, used to justify Global Warming alarmism by the 2001 Intergovernmental Panel on Climate Change.  

Here is the official ethics statement on scientific errors and the need for public correction from the American Physical Society, the national society of research physics:
"It should be recognized that honest error is an integral part of the scientific enterprise. It is not unethical to be wrong, provided that errors are promptly acknowledged and corrected when they are detected." [emphasis added]

(Ethics & Values - 02.2 APS GUIDELINES FOR PROFESSIONAL CONDUCT)
The more time that is allowed to pass before NASA makes a formal acknowledgement of error commensurate with the attention focused on the original announcement, the more it will be vulnerable to genuine charges of purposeful misconduct, as opposed to inadvertent error.

Michael Fumento pointed out that NASA has issued no less than five separate PR releases on the Global Warming hypothesis just this year, "each one alarming." But the major correction to internationally-broadcast claims about Global Warming have not received even a NASA press release, much less an official correction in a peer-reviewed journal like Science magazine. This is unacceptable. As Fumento wrote:
In pooh-poohing the revision, the GISS ignores the tremendous emotional impact it's had in practically claiming each year is hotter than the one before.
Honest scientists don't dismiss the significance of data errors; PR hacks do such things.

James Hansen of NASA is minimizing the importance of the changes by claiming that since the US is only a small potion of the total world, the overall global figures have not changed much. However, the US surface temperature record is considered to be the most complete, and longest, over the largest surface area, and also the most technically responsible available record. (It's very easy to screw up a temperature record, especially in many different locations, due to local heat and cooling sources.) It is not inferential (like ice cores), and it does not come from a scientifically backward country. No other continent-wide temperature surface record comes up to that standard, over such a long period of time.

So even though the US only covers 2 percent of earth surface, it's a crucial 2 percent in terms of the quality of the record.

In addition, raw data errors are very serious matters, even if there are excuses. In fact, the idea that NASA indulges in excuses is itself inculpatory. Good scientists just don't screw around with that. They just get out the correction, pronto, in a peer-reviewed, prominent journal, as soon as possible. That is because their entire reputation is at stake. There are numerous examples of scientists blowing their reputations if they were believed to have falsified their data. See the David Baltimore case. 

It is not even for the original erroneous author to decide on the significance of the error. That is up to the scientific community.

If NASA, a US government agency, will not own up fully to its own errors (which have now been corrected, quietly, on its GISS website), the American Physical Society must institute its own ethics inquiry to correct the record. The credibility of NASA and the entire scientific community are at stake.

If you want to see how common the practice of publishing errata and retractions is, take a look at PubMed (the National Library of Medicine public database of millions of biomedical abstracts).  Just type "errata" in, and you get more than 1,000 titles with the word "errata". The word "erratum" brings up another 3,000. The word "retraction" brings up almost 11,000 more.

A good model appears below. It just appeared in Science about an 8,200 year old low temperature event, which turns out to be possibly due to a technical artifact. Notice that the retraction was triggered by the fact that a re-analysis showed the original claim to be "uncertain". Since the burden of proof is on the scientist who published the finding, s/he must also publish the retraction.

Retraction of Baldini et al., Science 296 (5576) 2203-2206.
Science 10 August 2007:
Vol. 317. no. 5839, p. 748
DOI: 10.1126/science.317.5839.748b

LETTERS
Retraction of an Interpretation
In the Report "Structure of the 8200-year cold event revealed by a speleothem trace element record" (1), we presented a 7762-?m-long ion probe trace element traverse chosen to include the 8200-year event as detected in a previously published laser ablation oxygen isotope study from the same stalagmite (2). The oxygen isotope anomaly was distinct and dropped 8‰ below baseline values to a low value for the entire Holocene of -12‰ and was reproducible on a reverse track. However, recent reanalysis of the calcite believed to contain the oxygen isotope anomaly suggests that the anomaly was probably an analytical artifact possibly caused by laser ablation-induced fracturing during the original analysis (3). Consequently, without the original 18O "marker," the precise location in the stalagmite of calcite deposited during the 8200-year event is uncertain.
The trace element data in this Report, previously believed to correspond precisely with the entire 8200-year event, are now believed to represent the hydrological and bioproductivity response in western Ireland to a cold/dry event of uncertain provenance and intensity. The U-Th-derived dates of the event correspond approximately with the 8200-year event in Greenland ice cores, but without the additional guidance of the 18O anomaly, the precise timing in relation to the 8200-year event is now somewhat ambiguous. Unfortunately, it is now unlikely that the approximately 114-year duration ion probe track coincides with the entire 8200-year event (if at all); thus, the ~37-year estimate derived for its duration is probably no longer accurate. However, the trace element data remain robust and are interpreted as reflecting colder and drier conditions in western Ireland, followed by the return to more maritime conditions at the end of the first-order trace element anomaly. Additionally, the novel application of annual trace element cycles to build a high-resolution chronology and reconstruct paleoseasonality remains unchanged.
James U. L. Baldini
Department of Earth Sciences
Durham University
South Road
Durham DH1 3LE, UK
Frank McDermott
Department of Geology
University College Dublin
Dublin 4, Ireland
Ian J. Fairchild
School of Geography
Earth and Environmental Science
University of Birmingham
Birmingham B15 2TT, UK
References
     1.     J. U. L. Baldini, F. McDermott, I. J. Fairchild, Science 296, 2203 (2002).
     2.     F. McDermott, D. P. Mattey, C. Hawkesworth, Science 294, 1328 (2001).
     3.     I. J. Fairchild et al., Earth Sci. Rev. 75, 105 (2006).
James Lewis blogs at http://www.dangeroustimes.wordpress.com/
ATsupport2.jpg