I’ve just read an old Weather Magazine from 1984, and read a letter in it that listed the warmest 30 and 61 day periods from the Daily Central England Temperature series. I thought that I would try to update the original table with the latest values from the daily CET series, courtesy of the Met Office.
Because the warmest temperatures don’t always align with the start and end of any one month, the application looks at all possible overlapping periods from 1772 till now, and calculates an average mean, maximum and minimum temperature and anomaly. At one time this would have taken several minutes of processing on a PC, but with today’s multi-core processors this takes just a few seconds.
Surprisingly, the results I get aren’t exactly the same as the table in the letter, all I can think is that some of the values may have been changed in the last 30 odd years. I’ve done a couple of spot checks with values in a spreadsheet, and as far as I can see the answers I get are correct (fig 4), please let me know if you find they are wrong.
The first table (fig 1) ranks the warmest 30 day periods since 1772. The period between the 22nd of June and the 21st of July 1976 still heads the list as it did in the original table with a mean of 20.37°C, but only fractionally from 1995 and only for mean temperatures. The original table had a mean of only 19.8°C for this period for some reason, I double checked as correct with a spreadsheet in Google Sheets (fig 4). The hot month of July 1911 appears at #31 and not #7 as in the original but with the same mean. Most of the top twenty entries are from either 1976 or 1995, with a couple of periods from 2006. So for now 1976 lives on as the hottest 30 day period since 1772, even if it’s by just 0.007 of a degree Celsius.
When it comes to 61-day periods, it’s more or less a clean sweep for 1995, which takes the first 18 places. The 2nd of July to the 31st of August 1983 which was the warmest 61 day period in the original table, comes in much lower at #27, with a number of other 61 day periods in 1976 higher than it, although with a temperature that’s slightly higher (18.49°C) than in the original table.
The beauty of the application is that I can also find the warmest week, or any other period come to that, since 1772. This time 1976 beats 1995 to the first six places in the hottest weeks since 1772 table (fig 3), with the first week of July 1976 the hottest week with a mean maximum of 31.49°C, which is +11.25°C above the long-term average (1961-1990).
It is clear, however, that accurate comparisons of past hot summers can only be made when, first, statistical inconsistencies are eliminated and, second, the choice of time-period is kept consistent or at the very least clearly specified. welcome further comment on this subject.
- Beresford. A.K. (1984) Hot summers. Weather, 39(6), pp.193-195
- Ann Storey, C. K. Folland (1984) Hot Summers in Central England and Central England Temperature, 84, pp.287–289