What does ‘average’ mean?

  • What does ‘average’ mean?
    20.01.2011 16:14


    December was obviously a remarkable month in the UK and there was no need to refer to facts and figures to guess that temperatures were well below average. 

    When the calculations were made at the end of the month they showed that the Central England Temperature (CET), which acts as a proxy for the country as a whole, was 5.7 degrees below December’s average, in a temperature series dating back to 1659.

    While the UK, much of Europe and parts of the USA shivered, however, the figures were being compiled for the annual global temperature, and that told a very different story.

    The cold parts of the globe, although heavily populated, were far outweighed by the extent of the warm regions through 2010. These included the Arctic, particularly Greenland and northeast Canada, much of Africa, the Middle East and Russia, which had record-breaking heat during the summer.

    According to NASA’s Goddard Institute for Space Studies (GISS), 2005 was the hottest year globally in their 131 year instrumental record. The Hadley Centre in the UK says 1998 but the difference is negligible and it is essentially a tie.

    Those two years have now been joined by 2010, which was equally warm at 0.74 degrees Celsius (1.34 degrees Fahrenheit) above the global average.

    What average is that, though? Thirty-year averages are the accepted norm in climate and weather studies, and the GISS statistics use the period 1951 to 1980 as a base line.

    However, the next warmest years are 2002, 2003, 2006 and 2007, and they are not far behind the top three. It is clear, therefore, that a more recent thirty year period will move that line – the average – upwards, and the anomaly will appear to shrink.

    In fact, the University of Alabama in Huntsville (UAH) has already progressed with its averages, and has switched to using the 1981 to 2010 period for comparison.

    Even so, the comparisons and trends remain the same, and this still shows significantly above-average global temperatures during 2010, with an anomaly of +0.41 degrees Celsius – a shade lower than 1998 according to UAH records but again a statistical tie.

    Given that this is a comparison with figures partly comprising the very warm last decade, it is quite remarkable, and suggests how much cooler, for whatever reason, it was in the early 1980s.

    This will be true whatever recent base line used. The UK Met Office and the World Meteorological Organisation (WMO) currently use 1961 to 1990. On that basis, the WMO has just declared 2010 to have been 0.53 degrees above average, and according to them just about the warmest on record.

    The Japanese Meteorological Agency takes 1971 to 2000. In the USA, NOAA’s National Climatic Data Center (NCDC) stands on its own by looking at the whole of the twentieth century.

    The numerical values of temperature anomalies will differ from agency to agency, therefore, because of the different base lines they take for their averages, but the absolute figures and the trends remain the same.

    By: Stephen Davenport