One of our most famous pieces of weather lore, right up there with "red sky at night...", is this rhyme related to St Swithin's Day.
St Swithin's day if thou dost rain
For forty days it will remain
St Swithin's day if thou be fair
For forty days 'twill rain nae mare
It's superstitious nonsense, obviously, because the atmosphere doesn't realise the date is July 15th and because an English summer is always somewhat changeable. A dead bishop interred in Winchester cannot affect our weather.
But there can be a nugget of truth in the saying. Some summers the jet stream settles to our north displacing all the wet weather and some years it settles over us bringing in an endless sweep of rain-bearing depressions.
I mention this because August 24th marks exactly 40 days after St Swithin's Day, which means today is the ideal day to look back and tot up this summer's weather to test St Swithin's hypothesis.
July 15th: dry
St Swithin's Day 2022, it may not surprise you to hear, was dry.
And it's only rained on nine of the subsequent 40 days, the other 31 being bone dry, which is exactly what you might expect in a drought-ridden summer.
n.b. my count is a personal count based on where I am. If it rains where I am, even slightly, then this counts as a wet day. If it doesn't rain where I am, not even a drop, then this counts as a dry day. It's an appallingly self-centred way of measuring rainfall, so meteorologically very suspect, but it does have the advantage of being consistent over time.
July 15th
wet days
dry days
2022
dry
9
31
Well done St Swithin... not spot on, but pretty close.
"'Twill rain nae mare" is a pretty much perfect summary of the last six weeks.
But one year tells us nothing. What we need is a lengthier series of data over several years and then we might be able to see some patterns. Fortunately I've been recording St Swithin's Day weather in my diary since 1980, because of course I have, and also counting how many of the subsequent 40 days were wet or dry.
July 15th
wet days
dry days
1980
wet
13
27
1981
dry
11
29
1982
wet
15
25
1983
dry
9
31
1984
wet
12
28
1985
wet
32
8
1986
dry
16
24
1987
wet
29
11
1987
dry
19
21
1989
dry
7
33
In 1980 St Swithin was very wrong - July 15th was wet but most of the subsequent days were dry. 1982 and 1984 were similarly off-key, with a slew of dry days following a wet July 15th. But there are three stonking successes here, namely 1983 and 1989 which were forecast to be dry and it hardly rained afterwards, and 1985 which was forecast to be wet and it carried on chucking it down. So maybe there is something in all this after all.
Let's continue.
July 15th
wet days
dry days
1990
dry
7
33
1991
wet
13
27
1992
dry
21
19
1993
wet
22
18
1994
wet
14
26
1995
wet
6
34
1996
dry
16
24
1997
dry
12
28
1998
dry
18
22
1990 was a proper success, a full-on sunny summer following a prediction of fair weather. But 1995 was a disaster, almost exactly the same split as 1990 but predicted to be wet throughout. Most of the other years were a bit more balanced. If you're counting, St Swithin has only swung the wrong way here seven years out of nineteen.
Sadly I stopped recording the summer's weather in my diary in 1999 because I was too focused on other events. It took until 2004 for me to pick up the baton again, and I've recorded it diligently ever since.
July 15th
wet days
dry days
2004
wet
24
16
2005
dry
20
20
2006
dry
20
20
2007
wet
21
19
2008
dry
24
16
2009
wet
20
20
2010
wet
16
24
2011
wet
21
19
2012
dry
18
22
These are much closer totals, with the number of wet and dry days never dropping below 16 or rising above 24. We also have the first three dead heats - 2005, 2006 and 2009 - making St Swithin's original prediction irrelevant. If you're counting, July 15th has been a dry day just as often as it's been wet.
July 15th
wet days
dry days
2013
dry
19
21
2014
dry
16
24
2015
wet
19
21
2016
dry
33
7
2017
dry
23
17
2018
dry
14
26
2019
dry
18
22
2020
wet
14
26
2021
dry
20
20
2022
dry
9
31
Again most of these years have fairly equal totals for wet and dry days, but there are two glaring outliers. 2016 was ridiculously wet, but followed a dry July 15th so St Swithin was wrong. And this year has been ridiculously sunny, and followed a dry July 15th so St Swithin was right. It's interesting that over the last decade July 15th has been dry a lot more often than it's been wet, whereas beforehand my data was running fifty-fifty.
Like I said a single year tells us nothing, but maybe we can draw some overall conclusions from the period 1980-2022.
• St Swithin's Day was dry 58% of the time and wet 42% of the time.
• The subsequent 40 days were mostly dry 63% of the time and mostly wet 29% of the time.
• St Swithin correctly predicted the general gist of the subsequent 40 days 55% of the time.
That's not much better than flipping a coin. It'd be more accurate if the rhyme just said...
Whatever St Swithin sees in the sky
For forty days 'twill likely be dry
• Only 5 summers out of 38 have been more than 75% dry (including this year).
• Only 2 summers out of 38 have been more than 75% wet (1985 and 2016).
• St Swithin only predicted five of these correctly.
More usefully...
• On average, the 40 days after St Swithin's Day include 18 days when rain falls and 22 dry days.
• i.e. an English summer is mostly dry.
