Despite the cold winter and poor summer, 2008 temperatures worldwide were above the long-term average, the Met Office says in its report on last year’s weather.1
In the UK, 2008 was the 18th warmest year since 1914, with an average temperature of 9.03°C compared with an average of 8.36°C between 1961 and 1990. This is despite below average temperatures in October and December.
Worldwide, 2008 ranked higher in the temperature charts in tenth place. At 14.31°C it was 0.31°C above the mean for 1961-1990 and close to the 14.37°C predicted by the Met Office last January.
The organisation has predicted temperatures for the year ahead since 2000, with an average error of only 0.06°C (see figure). Its forecasts take account of natural and man-made climate forcing using data from previous years and projections for the year ahead.
Natural forcing includes widespread cyclical changes in sea-surface temperature in the tropical Pacific (the El Niño Southern Oscillation or ENSO), release of cooling dust clouds from volcanic activity and rhythmic variations in incoming solar radiation.
Year-to-year global temperature changes are linked most closely to these natural phenomena, whereas longer-term trends are dominated by human-induced forcing including greenhouse gas effects, ozone depletion and aerosols released by or formed as a result of industrial, combustion and transport emissions.
"The biggest cause of error [in these temperature forecasts] is predicting ENSO," says Andrew Colman of the Met Office. For instance ENSO’s warm El Niño phase ended earlier than expected in 2007, leading to an overestimated temperature forecast.
But the Met Office successfully predicted 2008 would be the 10th warmest year on record. Temperature rises last year were limited by a strong La Niña phase of ENSO, in which cold waters rise to the Pacific Ocean’s surface. This cools much of the southern hemisphere, where 2008 was only the 20th warmest or 0.11°C above average. In the northern hemisphere, 2008 was the 8th warmest or 0.51°C above average.
La Niña is set to weaken through 2009 so the Met Office predicts a return to the longer-term trend of strong temperature increases. They forecast a global average of 14.44°C, which would make 2009 the fifth warmest on record.
The 15 warmest years since 1880 have all occurred since 1990. Natural climatic variation alone cannot account for this warm cluster, according to researchers from Germany and Switzerland.2 If the temperature in one year were independent of the previous year, there is only a one in a million billion (1015) chance the cluster was due to natural variation.
But ocean circulation processes such as ENSO give global temperatures a year-to-year ‘memory’ where one warm year is likely to be followed by another. Eduardo Zorita at the Institute for Coastal Research in Geestacht, Germany and colleagues from the University of Berne modelled this memory effect to see if it could account for the exceptional series of warm years.
Even if memory effects are included, there is less than a one-in-a-thousand chance that recent global temperatures were caused by natural variation alone, the researchers found. In the language of the latest Intergovernmental Panel on Climate Change report, even a one-in-a-hundred probability is ‘exceptionally unlikely’.