## Monday, December 06, 2010

### RSS: 1998 will remain the warmest year

Update, RSS: The RSS AMSU November data are out, 0.312 °C, coolest in 2010 so far. The average temperature anomaly 0.551 °C recorded in 1998 will only be beaten by 0.489 + Dec/12 in 2010 if the December 2010 anomaly, Dec, will exceed 0.738 °C which is virtually impossible.

With the likely December value around 0.3 °C, 2010 will stay a marginally significant 0.035 °C cooler than 1998 but safely (by 0.14 °C) beating the bronze year, 2005. A near-record 2010 seems to be a purely UAH AMSU result. RSS AMSU will conclude that the 1998 leadership will stay unchallenged by 12th following year in a row.

UAH: December anomaly above 0.42 °C would make 2010 hottest

Roy Spencer has released the November 2010 UAH AMSU temperature anomalies. The global temperature anomaly is 0.38 °C, the coolest month of 2010, which is followed by the previous month, October 2010. The tropics have significantly cooled down during 2010 - by 0.7 °C or so - but the two hemispheres dropped by 0.2 or 0.3 °C per year only.

This El-Nina-related cooling is here but it is slower than some of us expected and slower than the post-El-Nino cooling at the end of 1998. So the battle about the warmest year is still open. Clearly enough, it should be declared a statistical tie already at this moment because the difference can no longer become statistically significant.

If you still care which year will be warmer assuming an accurate calculation of the centidegrees Centigrade :-), the answer is that the average UAH AMSU 2010 temperature anomaly will be
0.482 °C + Dec/12
where Dec is the currently unknown December 2010 anomaly. Please tolerate a possible 2-millikelvin error caused by different lengths of different months which I neglect. Meanwhile, the average 1998 anomaly was
0.518 °C.
If you equate the the anomalies and solve the linear equation, you obtain
Dec = 0.42 °C.
Inconveniently, because this solution gets divided by 12, the error of the figure above is 10 times bigger than the previous one, about 0.02 °C.

So if the December 2010 UAH AMSU anomaly is going to be above 0.42 °C or so - which also means above the October 0.426 °C anomaly and the lower November anomaly - 2010 is going to become the warmest year on the UAH AMSU record. If it won't, 1998 will defend its leadership once again.

Similar approximate ties will occur in the other three major datasets, too. NASA's GISS differs because they will have the same tie between 2010 and 2005, their hottest year so far. Meanwhile, 2011 is pretty much guaranteed not to become a warmest year.