How GISS Temperatures Diverge From RSS
By Paul Homewood
As reader David points out, you cannot directly compare global temperature datasets with each other, as they all operate to different baselines. So, for instance, the current GISS temperature anomaly of 0.68C is based on a baseline of 1951-80, but the RSS anomaly of 0.29C is based against 1979-98.
But how do the trends of the two compare over recent years?
Let’s start by looking at the full comparison back to 1979, when satellite measurements from RSS first appeared.
At the end of 1979, the 12 month running averages were 0.11C and –0.09C, for GISS and RSS respectively. Figure 1, however, offsets the whole of the GISS set downwards by 0.20C, so both start at the same point, –0.09C.
Although RSS seem to have recorded lower temperatures during the 1984/5 and 1988/9 La Ninas, and the post Pinatubo period, for most of the 1980’s and 90’s it seems to have tracked GISS pretty well. Until 1998, that is.
Suddenly in 1998, RSS overtakes the GISS numbers by 0.16C. Since 1998, though, GISS temperatures have reversed that position and steadily increased in comparison to RSS. (With the exception of a brief period in 2010).
So what happened in 1998? A clue lies in two graphs produced by GISS. The first, in Figure 2, was part of Hansen et al 1999. Note how the 1998 annual mean is well above the 0.60C mark. Now compare this with Figure 3, which is the latest version on the GISS website.
Since 1999, GISS have adjusted their 1998 figures down.
Now let’s take a closer look at what has happened to two sets since 1998.
To put numbers to it:-
Annual Temperature Anomaly
So, since 1998, GISS have found an extra 0.33C of warming that RSS cannot find. This is an astonishing figure, which effectively doubles the amount of warming found in total since 1979.
Comparison of GISS with HADCRUT and UAH shows a similar divergence.
As Doctor McCoy might have said “It’s temperature, Jim, but not as we know it”.