UK Daily Rainfall Extremes
By Paul Homewood
Following last year’s wet weather in the UK, Julia Slingo had this to say:-
The trend towards more extreme rainfall events is one we are seeing around the world, in countries such as India and China, and now potentially here in the UK. The long-term trend towards wetter weather is likely to continue as global air temperatures rise.
But what have been the trends in the UK?
The Met Office provide daily rainfall data back to 1931 for England & Wales, from which I have extracted the top 50 rainfall days. (This equates to days of 21.92mm or more). The distribution of these events is shown in Figure 1.
Figure 1
The wettest day was 25th August 1986, when 43.23mm fell on average across the country. The wettest day last year recorded 28.50mm, yet there have been seven days in earlier years which recorded higher amounts, all prior to 2001.
It is unusual for there to be three such days in a single year, as we saw last year. But it is not unprecedented, as both 1971 and 2000 also recorded three each.
We can also look at the decadal trends in Figure 2.
Figure 2
Although the last decade has been seen more heavy rain days than the 1990’s, it has been pretty much on a par with the 1961-1990 period, and significantly down on the 1970’s, even when 2012 is taken into account.
The Met figures are, of course, nationally averaged, and there are bound to be regional variations. Nevertheless, there is clearly nothing to back up Slingo’s assertion that extreme rainfall in the UK is on the increase, either in frequency or severity.
I have counted the number of days with over 10mm of rain in the UK since 1931, based on MO HadUKP figures.
The number of days with over 10mm of rain in 2012 was 31, which is the highest on record, the next highest being 27 such days, in 1993 and 2002.
The 10 year moving average for daily rainfall over 10mm was 18.5, which is slightly lower than the figure of 18.8, in 1960 and compares with a peak of 20.1 in 2009.
This could be seen as positive evidence of “climate change”, but I don’t believe that the records are sufficiently long to say that.
In fact there is a 76% correlation between annual rainfall figures and daily rainfall of over 10mm, which may make it possible to estimate figures for earlier years. Using the available data since 1931, the relationship between annual rainfall and daily rainfall of over 10mm is as follows: y = 0.0314x -11.973, where y = days with over 10mm rainfall and x = annual rainfall. If anything, this formula underestimates the frequency of rainfall over 10mm, but based on it, it seems quite likely that the frequency of such rainfall days was similar 1768 and 1872 to 2012.
Using this formula
Paul:
Slingo’s statement is ambiguous beyond belief. First, the term “rainfall events” is undefined, allowing her, if needed, to claim that she was speaking of regional or localized events if the national stats don’t back her up; or using different parameters than those you have chosen.
Second, what she says about the UK is that there is the “potential” for such a trend to emerge here — suggesting, actually, that such a trend is not manifest in the existing data. It’s more of a “the sky might fall at some undefined point in the future” kind of warning.
It’s probably better to view this as an attempt to help buttress the groundwork for claiming that any bad weather can be defined as an “extreme event”, one which is further evidence of global warming, or climate change, or climate disruption, or climate weirding…or whatever the term du jour.
She’d certainly make a good politician!
Paul, are my trackbacks getting to you? I see the odd one gets through so wondered if there was a WordPress delay or is it just that they have to wait to clear your moderation?
They do have to wait for moderation, but there all seem upto date now.
Thanks Paul. I’m a newbie at WordPress so wasn’t sure if I had it set up correctly.
Good to see you back.
A good rebuke of a feeble “correlation = causation” effort.
But please proofread at least once. “But have been the trends in the UK?” Say what?
Thanks, Brian. Sorted now.