The GISS problem is merely a clear example of how long very basic errors sit around because despite claims of widespread peer-review, they've never really been examined...and hundreds of papers cite them, and other networks gain credibility by matching them...just as with the Mann MHB98/99 papers.
Mann claimed his papers were not just read, but actively peer-reviewed by THOUSANDS of scientists; the TAR heavily relied on his work (in part because the IPCC made him a chapter author, which he took full advantage of)...and yet it took 8 years and congressional pressure before the math was thoroughly checked and found "not skillful".
The GISS calculations are widely cited, and a great deal of faith is placed upon Hansen's team performing adjustments for which they are not required to really explain what they are doing in any examinable detail.
Mr. McIntyre, did not find the error by himself - he lacks the resources to set up an alternate climate reference network - but the adverserial process narrowed down the error which had been overlooked for 7-1/2 years.
In May, Mr. Watts noticed that his local USHCN station (labelled as "high quality" by the NOAA) was both severly outside of NOAA standards for environment and adjacent to several features that have been demonstrated to induce a measurement error in the 5C range; this was concurrent with his finding that just the change in paint type on Stevenson shelters was enough to alter station readings.
Mr. Watts then went to another local USHCN site, and found lesser, but substantial problems.
Mr. Watts then set up an internet site and began recruiting volunteers to systematically record other sites in an attempt to get a complete census of the USHCN network, and evaluate whether the network is consistant with the NOAA claims.
This survey is still ongoing (with approx 270 out of 1200 surveys completed), and results from it are suggestive in nature only, however while there are some sites which appear pristine (as the network is represented) - at least as regards a consistant environment - even if all further sites surveyed turn out to be proper, the numbers of sites outside of parameters seems enough to overwhelm the signal purported to be selected out of it.
One thing that appears to have happened systematically, is that sites were commonly moved adjacent to buildings apparently so as to allow the migration to MMTS sensors with a commonly used cable length. This occurred gradually in the 1980s and 1990s.
Generally the most extreme warming trends have been associated with the most absurd site placement, such as the site sitting IN the parking lot.
The claim has been made that the various semi-documented adjustment procedures account for the ill-placement. To date, it appears as though many of the most ill-sited sites have been adjusted down....and pristine sites adjusted upwards(or more properly, adjusted downwards in the past more than recently). Again, the survey is incomplete, and the adjustment methods are not clearly stated.
One of the sites photographed was the site in MN, which it was discovered had a large jump, and two large AC compressors placed absurdly close;
some associates of Mr. Watts assumed that the relocation of air conditioning compressors next to the sensors were responsible for a sudden jump in site records; this was countered that the jump did not exactly match (though there does seem to be a signal as well) the relocation by Mr. Rabbet and his AGW associates; Mr. McIntyre then dug about attempting to match other records, and discovered that for some reason that a semi-adjusted record which had been replaced some time ago, suddenly replaced the more adjusted records beginning in Jan 2000, without being normalized.
GISSs attempts to correct the programming error that caused that use further non-public adjustments, and though the programming error only directly affected data since Jan 2000, apparently the data was used to adjust data prior to that time as well. These source for these "corrections" are unknown as well...which is the real problem.
Many prominent studies which use the USHCN network don't identify exactly which sites they used - even upon explicit request - which again, blocks review. Others such as the "light1 = urban/light0 = rural" have been shown to have not merely flawed assumptions, but downright silly...and have been relied upon by other studies for use to derive controls.
even if all further sites surveyed turn out to be proper, the numbers of sites outside of parameters seems enough to overwhelm the signal purported to be selected out of it.
Or maybe not.
To make it scientific, they are going to have to work a bit harder than taking pictures and finding data problems at a few stations.
The bottom line is that this may cause the USHCN principals to examine their methods and data; and the ultimate bottom line will turn out to be the current scientific understanding of the issue -- the world is warming and the scientists know why.
Or why else is spring earlier and winter later?
Early Warning Signs: Spring Comes Earlier
Arctic spring arriving weeks earlier
” The GISS problem is merely a clear example of how long very basic errors sit around because despite claims of widespread peer-review, they’ve never really been examined...and hundreds of papers cite them, and other networks gain credibility by matching them...just as with the Mann MHB98/99 papers.
Mann claimed his papers were not just read, but actively peer-reviewed by THOUSANDS of scientists; the TAR heavily relied on his work (in part because the IPCC made him a chapter author, which he took full advantage of)...and yet it took 8 years and congressional pressure before the math was thoroughly checked and found “not skillful”.
The GISS calculations are widely cited, and a great deal of faith is placed upon Hansen’s team performing adjustments for which they are not required to really explain what they are doing in any examinable detail.
Mr. McIntyre, did not find the error by himself - he lacks the resources to set up an alternate climate reference network - but the adverserial process narrowed down the error which had been overlooked for 7-1/2 years. “
Meanwhile, from the AGW hypers we are told, ‘pay no attention to the man behind the curtain’.