The Remote Sensing System (RSS) has decided to switch to a version 3 much like the GHCN and UAH temperature sets have done in recent months. As I have discussed each in turn I felt it only fair to discuss the differences in the RSS temperature set. The GHCN change subtly enhanced the appearance of global warming. The UAH was the most straightforward as the adjustment was a month by month (and spatial) adjustment to a new average temperature.
The RSS adjustment is just confusing. There is no significant difference until January of 1999. After that there is a small difference for each month, but there is no pattern to the adjustment. 1999-2000 show a small shift upwards in temperature while 2001-2010 show a decrease in temperature. Nor is there a pattern by month. This version change is the most confusing of them all.
The monthly average adjustment since 1999 is shown here.
The summer months showed the largest decrease, but the adjustment is smaller than the UAH and only applies to the last 11 years which all other version changes have affected the entire range. This version is an unusual shift from the previous one.
Interestingly enough since the UAH was a more uniform shift in anomaly, there was almost no impact to the linear trend in temperature. Since the RSS decreased the recent temperature only it resulted in a decrease in the warming trend even though the anomaly change was small.
Version 2 trend: 0.163C/decade
Version 3 trend: 0.149C/decade
The change in appearance is small, but again because the difference was only applied to recent data, it will create the appearance of less warming than version 2.
Since there is no pattern to the change it is not possible to regress the data back to version 2 from version 3. For that reason I can only provide v3 updates from here forward. So I will be using v3 data in all future uses.