Paul Homewood provides a temperature roundup.
By Paul Homewood
A divergence this month between satellites (down) and surface (up), that has led to a bit of discussion. I don’t get too exercised by any of this, as on a month to month basis this sort of variability is quite common. Usually, the different datasets come back into line after a month or two.
There has, however, been a lot of hot air about it being the “hottest November on record”, so we need to get a few facts straight.
1) Leaving aside the satellite sets, where on RSS this November is the 3rd coldest of the century, it is not even the hottest November on HADCRUT.
2) There are 12 months in every year, so over a 12-year period, the odds of getting a “hottest month” is once a year.
3) Taking all months of the year, the…
View original post 362 more words






I’ve said this before and will no doubt have occasion to say it again.
Don’t use HadCRUT4!
HadCRUT4 is an apples-and-oranges average of surface air temperatures measured a nominal five feet above the land surface (CRUTEM4) and sea surface temperatures measured anywhere between a few inches and fifty feet below the surface of the ocean (HadSST3). The two data sets show different trends and must therefore be evaluated separately. (The reason they get averaged into the HadCRUT4 “surface temperature” record is to conceal as far as possible the fact that climate models can’t replicate observed SST trends worth a damn, and the oceans are of course where 99 point something percent of the available heat in the oceans and the atmosphere is stored.)
Here’s where November 2013 places in the “hottest November on record” stakes when we consider SATs and SSTs separately using the Hadley & CRU series (NCDC may be a little different):
SAT (CRUTEM4): Sixth warmest (November 2001, 2004, 2005, 2008 and 2010 were warmer)
SST (HadSST3): Second warmest (November 1997 was warmer).
And where it places in the “hottest month on record” stakes:
SAT (CRUTEM4): 37th warmest (warmest was December 1852, and if you don’t believe the 1852 data it was December 1997).
SST (HadSST3) 39th warmest (warmest was July 1998).
And remember that CRUTEM4 and HadSST3 are both heavily adjusted and not to be trusted either.
Merry Christmas to all. 🙂
Undocumented adjustments: Climate Science Crime.