---

Skewed science
A French scientist’s temperature data show results different from the official climate science. Why was he stonewalled? Climate Research Unit emails detail efforts to deny access to global temperature data

...Their first step was to assemble a database of temperature measurements and plot temperature charts. To do that, they needed raw temperature measurements that had not been averaged or adjusted in any way. Courtillot asked Phil Jones, the scientist who runs the CRU database, for his raw data, telling him (according to one of the ‘Climategate’ emails that surfaced following the recent hacking of CRU’s computer systems) “there may be some quite important information in the daily values which is likely lost on monthly averaging.” Jones refused Courtillot’s request for data, saying that CRU had “signed agreements with national meteorological services saying they would not pass the raw data onto third parties.” (Interestingly, in another of the CRU emails, Jones said something very different: “I took a decision not to release our [meteorological] station data, mainly because of McIntyre,” referring to Canadian Steve McIntyre, who helped uncover the flaws in the hockey stick graph.)

Courtillot and his colleagues were forced to turn to other sources of temperature measurements. They found 44 European weather stations that had long series of daily minimum temperatures that covered most of the 20th century, with few or no gaps. They removed annual seasonal trends for each series with a three-year running average of daily minimum temperatures. Finally they averaged all the European series for each day of the 20th century.

CRU, in contrast, calculates average temperatures by month — rather than daily — over individual grid boxes on the Earth’s surface that are 5 degrees of latitude by 5 degrees of longitude, from 1850 to the present. First it makes hundreds of adjustments to the raw data, which sometimes require educated guesses, to try to correct for such things as changes in the type and location of thermometers. It also combines air temperatures and water temperatures from the sea. It uses fancy statistical techniques to fill in gaps of missing data in grid boxes with few or no temperature measurements. CRU then adjusts the averages to show changes in temperature since 1961-1990. ...

...The decision to consider the 1961-1990 period as ‘normal’ was CRUs. Had CRU chosen a different period under consideration, the IPCC graph would have shown less warming, as discussed in one of the Climategate emails, from David Parker of the UK meteorological office. In it, Parker advised Jones not to select a different period, saying “anomalies will seem less positive than before if we change to newer normals, so the impression of global warming will be muted.” That’s hardly a compelling scientific justification!...

...In addition to calculating temperature averages for Europe, Courtillot and his colleagues calculated temperature averages for the United States. Once again, their method yielded more refined averages that were not a close match with the coarser CRU temperature averages. The warmest period was in 1930, slightly above the temperatures at the end of the 20th century. This was followed by 30 years of cooling, then another 30 years of warming....