British Isles weather diary

Climatological averages for the Reading University climatological station

While the weather is what we observe over a short period of time, the climate is how the atmosphere "behaves" over relatively long periods of time.

Meteorologists typically calculate the climate, by means of climatological averages, over a period of about 30-years for individual weather stations. In recent years these 30-year periods have included 1961-1990 and 1971-2000, and now include 1981-2010.

A 30-year period is used in order that the contribution from one or two extreme months does not impact too much on the results - thus after a cold month in the UK like December 2010 any 10-year average would be changed by over 0.5C in places by the inclusion of data from such a month, although some of this would be cancelled out by warmer Decembers.

Why might we be interested in different 30-year periods? Well, the climate is always changing - no two 30-year periods are the same - and as meteorologists are interested in the departures from the normal (i.e. how hot or cold is it compared to the average) then a recent long-term average should be used. This is particularly important if we believe that the climate is beginning to change relatively quickly.

In this note I look at the 30-year data for Reading Universioty and relate it to the 30-year period 1971-2000 to look for signs of warming/wetting, etc.

The following table lists the climatological averages for the 30-year period 1981-2010.

1981-2010 climatological averages.
Month Mean max temp Mean min temp Mean temp Mean precipitation Mean sunshine
degC degC degC mm hours
Jan 7.7 1.9 4.8 60.5 56.5
Feb 8.0 1.7 4.8 41.0 75.9
Mar 10.8 3.5 7.1 44.4 109.0
Apr 13.5 4.7 9.1 48.0 160.3
May 17.0 7.7 12.3 46.3 188.1
June 20.0 10.5 15.3 44.6 189.4
July 22.4 12.7 17.5 45.9 197.5
Aug 22.1 12.5 17.3 52.3 191.4
Sep 19.0 10.3 14.7 50.2 138.1
Oct 14.9 7.6 11.3 72.2 106.6
Nov 10.7 4.4 7.5 66.3 63.3
Dec 7.9 2.2 5.1 63.0 46.1
Year 14.5 6.6 10.5 634.7 1522.3

Here, the maximum temperature is the highest air temperature observed in a Stevenson (thermometer) screen that typically (but not always) occurs around 2pm, while the minimum temperature is the lowest air temperature, observed in the same screen, typically (but not always) just after dawn. The mean temperature is the average of these two.

Precipitation includes any fall of water in the form of rain/drizzle, sleet/snow, hail, fog and dew/hoar frost - with any solid precipitation being melted to give a water-equivalent depth.

Note that the values of air temperature are average daily values - daily values can differ considerably from the averages.

Likewise, the precipitation and sunshine amounts are average monthly values and it is not unusual for precipitation amounts to be in the range 200% to 25% of the values shown, which sunshine can sometimes vary between 150% to 50% of the amounts shown.

A table showing extreme values on individual days is shown here that illustrates how variable daily values can be when compared to the monthly mean.

1981-2010 extreme daily values and year of occurrence. Note that when the same sunshine value occurs in two years then the earlier is listed.
Month Highest daily temp Lowest daily temp Wettest day Sunniest day
degC degC mm hours
Jan 14.7 (1998) -14.5 (1982) 21.6 (1999) 8.1 (1987)
Feb 17.0 (1990) -11.6 (1986) 26.8 (2009) 9.3 (1995)
Mar 20.4 (1990) -6.4 (2001) 23.8 (1984) 12.1 (1997)
Apr 25.7 (2003) -3.5 (2010) 30.6 (1991) 13.7 (1999)
May 28.5 (2005) -1.0 (2010) 31.8 (1985) 15.5 (1985)
June 31.4 (1995) 1.5 (1991) 42.6 (1998) 15.3 (1994)
July 35.3 (2006) 5.2 (1993) 42.5 (2007) 15.3 (1986)
Aug 36.4 (2003) 5.1 (1983) 44.2 (2004) 14.5 (1981)
Sep 29.6 (2006) 1.2 (1987) 76.3 (1992) 12.6 (1982)
Oct 25.5 (1985) -4.4 (1997) 49.3 (2000) 10.2 (1997)
Nov 18.1 (2010) -8.3 (1983) 28.7 (2003) 8.2 (1986)
Dec 15.8 (1985) -13.4 (1981) 31.3 (1995) 6.9 91986)
Year 36.4 (2003) -14.5 (1982) 76.3 (1992) 15.5 (1985)

Monthly extremes

One indication as to the way in which climate is changing is to look at the distibution of occurrences of monthly extremes. Thus in an unchanging climate, if we were to examine when the highest temperature of each of the calendar months (12) occurred in a 30-year period we might expect to find 4 per decade. Clearly, as with the example of tossing a coin, this distribution might need more than 12 samples to approach the statistically-expected result.

With this caveat in mind, consider the following table:

The distribution of the 12 values for the individual calendar months shown in the left-most column - temperature.
Decade 1981-1990 1991-2000 2001-2010
Highest temperature 2 4 6
Highest mean max. temp. 2 4 6
Highest mean min. temp. 3 2 7
Total of above 3 numbers 7 (19%) 10 (28%) 19 (53%)
Lowest temperature 7 3 2
Lowest mean max. temp. 7 4 1
Lowest mean min. temp. 8 4 0
Total of above 3 numbers 22 (61%) 11 (31%) 3 (8%)

This shows a tendency for 'warm records' to be set in the 1990s and 2000s, while cold records were set mainly in the 1980s and 1990s.

The distribution of the 12 values for the individual calendar months shown in the left-most column - precipitation.
Decade 1981-1990 1991-2000 2001-2010
Wettest month 2 5 5
Driest month 6 4 2

A similar table for precipitation shows a tendency for 'dry records' to be set in the 1980s and 1990s, while wet records were set mainly in the 1990s and 2000s. However, the caveat about the number of values (24 in all) needs to be born in mind here, and the departure from a 4-4-4 decadal-split is small.

How is the climatology evolving?

Finally, how do the 1981-2010 averages compare to those for 1971-2000?

1981-2010 climatological averages minus the 1971-2000 averages. A positive number implies a warming/wetting/getting sunnier in recent years.
Month Mean max temp Mean min temp Mean precipitation Mean sunshine
degC degC mm hours
Jan +0.4 +0.3 +0.9 +2.3
Feb +0.3 +0.2 +0.5 +5.4
Mar +0.5 +0.4 +2.3 +2.2
Apr +0.8 +0.3 +0.7 +8.6
May +0.5 +0.4 -0.3 -3.8
June +0.7 +0.4 -5.5 +2.5
July +0.4 +0.3 +4.7 -4.7
Aug +0.3 +0.4 -0.2 -3.7
Sep +0.5 +0.3 -7.3 -1.4
Oct +0.4 +0.5 +7.7 -1.1
Nov +0.4 +0.4 +7.4 -4.6
Dec -0.2 -0.4 -1.4 -1.3
Year +0.4 +0.3 +5.1 +0.7

What stands out in this table is the warming that has occurred in each month, except for December. During the past 30 years we have had the two coldest Decembers of the past 100 years (in 1981 and 2010) - so the 30-year period started and ended with a very cold December!

Changes in sunshine amounts must be judged with caution as the location of the sunshine recorder has moved during the 30-year period and there is a suggestion that tree growth may be reducing the record a little during winter months.

Rainfall has increased by about 1% between the two periods. In particular the late autumn appears to have become wetter, although some of this is offset by a driest September.

Last updated 15 February 2011.