Climate change must be examined over decades, not years
There has been a lot of nonsense written about the lack of much if any warming over the last few years. It’s not a new argument — in fact, I blogged about it here — but like an axe-wielding psycho from a cheap horror flick, it just keeps coming back.
At times like this, it is always useful to look at the data. The figure below shows the temperature anomalies (relative to the 1961-1991 average) from 1850 to 2007. The data are the Hadley HadCRUT3v analysis.
As you can see, 1998 was an extremely hot year, mainly due to the presence of the biggest El Niño of the 20th century. Depending on which data set you use, 2005 may or may not have been hotter. My guess is that 2005 was basically a statistical tie with 1998. Other years since 1998 have not reached the lofty warmths of 1998 and 2005.
A close look at the plot shows that this situation is not abnormal. In fact, global warming has stopped repeatedly over the last 150 years — meaning that there are many instances when the temperature reached a maximum that took many years to surpass.
The reason is clear. A warming rate of 3 degrees C per century corresponds to an annual average rate of warming of 0.03 degrees C per year. At the same time, interannual variability, such as El Niño events, are of the order of 1 degree C per year. Thus, over short time scales, the slow upward trend can be completely swamped by the large year-to-year variability.
Over the course of several decades, however, the slow warming trend dominates, and you end up with significant warming. Thus, to determine if global warming is occurring, you have to look at time periods of decades, not years.