Authors correct error found in research article

by William Yeatman on August 31, 2004

in Science

Tim Lambert of the University of New South Wales, recently discovered (see his web log at science/mckitrick6.html) an error in calculations in a scientific article by Ross McKitrick of the University of Guelph and Patrick Michaels of the University of Virginia.  The article (see the June 9 issue of the newsletter), titled A Test of Corrections for Extraneous Signals in Gridded Surface Temperature Data and which appeared in Climate Research, found that recent temperature records were strongly influenced by socio-economic factors.

 Prof. McKitrick has now corrected the calculations upon which this conclusion was based.   In a draft erratum, available at, he writes, The principal effect of the correction is a reduced weight on the constant term and an increased weight on the COSABLAT variable itself. Indeed the correction improves the overall fit and removes the anomalously small cosine-latitude effect. The socioeconomic variables remain significant and the effects carry over from the station data to the gridded data as before.

 Because the main patterns of results persist across the revised tables, the original discussions as worded in our paper need only minor modification, and our overall conclusion, re-stated here, is unaffected:

 Overall, the results of this study support the hypothesis that published temperature data are contaminated with nonclimatic influences that add up to a net warming bias, and that efforts should be made to properly quantify these effects.

Comments on this entry are closed.

Previous post:

Next post: