Anthony Watts has been keeping a blog critical of NOAA/NASA climate monitoring due to external influences on weather stations (e.g., thermometers placed too close to paved parking lots or air conditioning units).
In a recent installment, he passes along news that recently popular stats on the record-setting heat of the past few years are potentially tainted by a a Y2K programming bug.:
What [Steve McIntyre] discovered was truly amazing. Since NASA does not fully publish the computer source code and formulae used to calculate the trends in the graph, nor the correction used to arrive at the “corrected” data. He had to reverse engineer the process by comparing the raw data and the processed data.
Here is one of his first posts where he begins to understand what is happening. “This imparts an upward discontinuity of a deg C in wintertime and 0.8 deg C annually. I checked the monthly data and determined that the discontinuity occurred on January 2000 - and, to that extent, appears to be a Y2K problem. I presume that this is a programming error.”[...]
This discontinuity was apparently pointed out to the powers-that-be, who have updated their stats.
According to the new data published by NASA, 1998 is no longer the hottest year ever. 1934 is.
Four of the top 10 years of US CONUS high temperature deviations are now from the 1930s: 1934, 1931, 1938 and 1939, while only 3 of the top 10 are from the last 10 years (1998, 2006, 1999). Several years (2000, 2002, 2003, 2004) fell well down the leaderboard, behind even 1900.
Strangely, I haven’t heard mainstream media pick up on this correction.