Skip to content

Another Look At UK Rainfall Variability

March 28, 2013

By Paul Homewood


A couple of days ago, I looked at Sir John Beddington’s claim that UK rainfall is becoming more variable. My analysis of year on year variation suggested that it was not.

Reader HK suggested an excellent alternative way of measuring this variability. Again using the England & Wales Precipitation Series, dating back to 1766, he took the lowest annual rainfall and the highest one occurring in the previous 10 year period, and then calculated the difference.

This calculation was then done for each year in the series, using a 10-year rolling cycle.

So, for example, in 2012 the highest and lowest years in the previous 10 years were :-


2012 – 1244mm

2003 – 761mm

Difference – 483mm


While in 1861 we get:-


1852 – 1213mm

1854 – 673mm

Difference – 540mm


So if we plot the differences , we get:-




There are peaks and troughs, but the peak we are on at the moment is no different to the peaks of the 1960’s and 20’s, and much less than some seen in the 19thC. Furthermore, the trend line is clearly on the decrease.

  1. Andy DC permalink
    March 28, 2013 1:26 pm

    Despite what alarmists claim, there is very little new under the sun.

  2. Ray permalink
    March 28, 2013 1:42 pm

    Another way of looking at this might be the rolling standard deviation over various periods.
    Both the 10 year and 30 year rolling figures are higher than the late 1980’s/1990’s but much lower than in the late 19th and late 18th centuries. Both measures are trending downwards over the entire series.

    I wish someone had asked Sir John Beddington on which data he was basing his conclusions. Like a lot of alarmists, he seems to be concentrating on only the most recent figures, without looking much further back.

Comments are closed.

%d bloggers like this: