Are Bank Holidays associated with bad weather?

[Weather Boy]

Are Bank Holidays associated with bad weather?

Introduction

Following the recent hot and sunny bank holiday (“BH”), I wanted to test the widely held belief that BHs are associated with poor weather.

Method and Limitations of Work

I used my own weather data to test the hypothesis.  This has the following limitations:

1.      1. It dates back only 10 years.

       2. It is for my location only.

h     3. The data may not be as accurate as Meto stations, especially prior to mid-2013 when I upgraded my weather station. This is particularly so for sunshine hours, as these are based on daily estimates.

Nevertheless, it should act as at least a guide for the hypothesis for my region of the country (south-east/ East Anglia).  I could have used Met Office local data, but it seems better to compare like with like, i.e. my data with my data.

I calculated the average for the relevant weeks of each BH, and then calculated the average for each BH for the same years.  I only looked at the two May holidays and August, because the winter holidays don’t seem to receive the same attention (as nobody is expecting warm, settled weather in any case) and Easter is difficult due to its variable date.

Note that I only looked at the day itself.  I did not look at BH weekends as a whole.  A further note is that these stats relate to the whole 24 hours, ending late evening.  This means that stats could mislead one if, say, one BH had heavy rain the previous night, clearing to a sunny day (although my notes do not suggest that this has often been the case, if at all).

Overall, therefore, this is not a very scientific study, but it is a reasonable guide to the issue.

Results and Comment

Early May Bank Holiday

I have not set out all of the data, only the averages:

	Max (C)	Min (C)	Rain (mm) per day	Sun hrs
May generally	18.92	7.95	1.48	5.88
Average 1-7 May	17.3	6.48	1.4	5.65
May BH	17.22	6.99	1.07	6.1

 

Late May Bank Holiday

	Max (C)	Min (C)	Rain (mm) per day	Sun hrs
May generally	18.92	7.95	1.48	5.88
Average 25-31  May	18.93	9.65	2.43	5.73
May BH	18.93	10.11	2.79	5.11

August Bank Holiday

	Max (C)	Min (C)	Rain (mm) per day	Sun hrs
Aug generally	22.7	12.97	1.9	5.27
Average 25-31  Aug	20.95	12.46	3.73	5.3
Aug BH	20.57	12.46	3.9	6

 

Conclusion

Summary

I am no statistician but it seems to me that deviation from average is not significant in any instance.  Indeed, in some instances, the exact correlation with average is almost spooky.

The following are the slight differences observed:

Early May BH

Slightly drier and sunnier than average, with slightly higher minima.

Rain-wise this BH appears to be less variable than the others and notably drier.  Perhaps not surprisingly, the range of temperatures is however variable with maxima varying by 12C and minima a huge 17C (2-15C).  There were no sunless days and it was significantly sunnier than the late May holiday despite the days being shorter.

Late May BH

This also has higher minima (at 0.7C, this is arguably significant) but is also cloudier and wetter than average.

As that rainfall is some 24% more than average, one might think that was significant.  However, it should be noted that the mean rainfall is hugely skewed by 2017 which received 16 mm of rain.  If we eliminate that year, the mean would be only 1.01 per day and thus very significantly drier than average.  Therefore, it seems unlikely that this is significant.

2017, along with just one other year, was also sunless, and it is this that makes the BH slightly cloudier than average.  Balancing this, 2012 and 2013 were both very sunny (and the former was the warmest).

Temperatures are the least variable for this BH, with a maxima range of 10.7C.

August BH

This has slightly cooler maxima and very slightly wetter but nevertheless sunnier than average (and this latter difference is more significant).

Rainfall wise, this was the most variable BH.  2014 and 2015 were two ‘stinkers’ in a row, with zero sun and substantial rainfall of 18 and 16 mm respectively.  Apart from those two, there was barely any rain at all.  These were the only sunless BHs, but generally, sunshine totals are not impressive.

Maxima was also very variable, with 2013 being almost as warm as 2017 but the two ‘stinkers’ mentioned above logging a pathetic 15.7C and 14.1C respectively, the latter being the joint lowest August maximum I have ever recorded.  This means that August BH maxima have a range of 13.1C, the highest of the three.

Brief Executive Summary

There is most likely no notable variation between average weather likely at the times of year and weather experienced over bank holidays.  This is not surprising given that the weather does not know or care when England &Wales has its bank holidays.

The limited variations that exist are unsurprising and probably not statistically significant.  Hence, they vary both ways: some suggest better BH weather, some worse.

Whilst not particularly unscientific, I think that when people next complain of BH weather, I shall say that on average it is no different from average for the time of year.

Follow Up Work

This begs a further question as to whether we have Bank holidays at the right time of year.  The average annual rainfall at my location based on my data is a fairly modest 601 mm which equates to almost 1.65 mm per day.  The early May BH is below this average but the other two days, and indeed in August, both the week and the BH itself, the figure is almost double that!

I propose to take a look at this issue in a follow up and it is for this reason that I have inserted the averages for the entire months in which the BH falls.

Meanwhile, it appears to me that more work could be done to investigate this hypothesis using CET and Met Office data over a longer period than the 10 years I have used.

I hope that this work provokes some discussion.

WB

[Blessed Weather]

23 hours ago, Weather Boy said:

Are Bank Holidays associated with bad weather?

(snipped)

An interesting topic WB and certainly one that fuels the British obsession of talking about the weather. Thank you for the time you've taken to compile your post. It looks as if your findings confirm another theory - the weather is likely to average itself out over the longer term. Just from a personal point of view, I dislike the fact that there's a flurry of BH's in Spring when despite strong sunshine temperatures are held back by lag from the yet-to-warm-up ground and surrounding seas. And then we have to wait until late August for the last holiday!

I've had a dig around for info and it appears that finding data that covers all Bank Holidays and embraces all of the UK is difficult. But I found a 2014 article by WeatherCast that had some interesting facts about the August Bank Holiday. Here's the main thrust and I've highlighted the key aspects:

Those of a more cynical nature may muse that it always rains on Bank Holidays, their views perhaps clouded by the recent six summers in a row (between 2007 and 2012) that were wetter than normal. Let’s take a look at the evidence for 30 years of Bank Holiday Mondays between 1984 and 2013.

It rained widely across the UK on approximately 10 of these years (1 in 3). In 1986, following a fine day on the Sunday, the Monday 25^th August was marred by the passage of ex-Hurricane Charley. This produced the wettest August Bank Holiday on record, with 135mm of rain falling in Aber, Wales, and more than 50mm of rain falling widely across Wales, Midlands and Northern England. It was almost as bad in 1992, when an unusually deep summer low pressure system swept heavy rain across much of the UK. There were westerly gales in southern England.

Nine of the last 30 years saw dry weather across approximately half of the UK, but with rain or showers either clearing away in the morning or arriving in the evening across remaining areas. For example, in 2011 the Bank Holiday saw fresh north-westerly winds and showers across northern and eastern areas, while most southern and western parts stayed dry.

Finally, 11 out of the last 30 years experienced a predominantly dry Bank Holiday Monday! As recently as last year, much of the UK enjoyed a fine day, with warm sunny spells and gentle winds. Just a few isolated showers developed in the far south, while rain clipped the far north-west of Scotland in the afternoon. Therefore, based on the last 30 year’s climate, a soaking is likely just slight more than one year in every three. North-western parts of the UK are most likely to experience rain.

Full article: http://www.weathercast.co.uk/nc/weather-news/news/article/august_bank_holiday_mondays_how_often_does_it_rain.html

[MP-R]

Very interesting research. Thanks for sharing! I've always thought there should be another bank holiday in June and July and one fewer in May to make the most of the better half of the year. Even one half way between the late summer one and Christmas would be nice.

Since I've kept records, the late May holiday has most often been the most unsettled one, and the August one the most settled (despite very mediocre Augusts overall). Some have indeed been horror shows like 2007 and 2008.

[Weather Boy]

Thanks, guys, for not lettting me skwak into the ether without any response from anyone!

I'll come back to the topic of weather, rather than bank holidays, in a second, but prior to the June election this year, Jeremy Corbyn proposed 4 new bank holidays for the saints day of each patron saint.  Fine in principle, I thought, but 3 of them are in the spring, when we have too many bank holidays anyway, and the other is in November, one of the dreariest months of the year (although that one at least does break up the long stretch from August until Christmas).

Interesting article, Malcolm.  I hadn't actually googled the issue at all.  Summarising, it sounds as if we have roughly 1/3 of BH's broadly fine across the country, 1/3 broadly wet, and 1/3 hit and miss, depending on where you are, presumably with the usual scenario that it's most likely to be the NW that gets the bad weather and SE getting the better weather.  I don't know how that compares with the weather in May and late August generally.

This leads me to the topic I want to explore again soon, which is, based purely on the weather, are the BH's in the right place?  I made the observation in my original post that the August BH is far wetter than average.  Note also the differential between average August max here and the average for the last week.  A fall of 1.85 C, suggesting that the first week of August, where the BH was prior to 1965 when it was moved, saw even more of a differential!

I'll post again when I research this further.

Finally, the mention of August BH 1986 made me go back to my childhood records, but unfortunately, they started in December 1986, so just missed that.