GISS Fiddle The Numbers Again – (Upwards Of Course)
By Paul Homewood
On running the last set of GISS temperature data through my spreadsheet, I noticed that they had added 0.03C, or more, of warming onto their numbers for January through to October 2012. For instance, January was 0.32C, and has been changed to 0.36C.
I cannot tell specifically when the numbers were changed, but the fact that October has been changed suggests that the changes were made in the last two months.
The net result is that the anomaly for the whole year is 0.56C, instead of 0.53C.
But it gets worse, as similar changes have been found back to 1998. I have not kept numbers for prior to 1979, so cannot be sure of changes to data earlier than then. But, of course, the anomaly is based on a baseline of 1951-80, so, relatively, we know those years cannot have been altered in total terms.
I obviously did not think to keep copies of the GISS dataset at the time, (but I am keeping this one!) However, I have found a link from AccuWeather, confirming the original figure for September of 0.60C. Now they are claiming 0.66C. You can check this for yourselves here.
It actually gets more complicated, because, at some stage since September, I must have updated the 0.60 to 0.62C. It is common for all four datasets to be marginally adjusted a month or two later.
There is no mention at all on the GISS website of any updates that could have caused this, and I have emailed them to see if they know why.
Maybe they have a rational explanation, and if they have, they have serious questions to answer as to why they did not announce it publically before altering the data.But, on the face of it, this is just another example of data tampering that they have been accused of in the past.
I’ll post up whatever answer I get back from them.