Friday
Dec112015
by 
Bishop Hill
Bishop Hill A little gem
Dec 11, 2015 Climate: WG2 ;)
The BBC's Inside Science strand had a brief section on the floods in Cumbria, with Nick Reynard of the Centre for Ecology and Hydrology interviewed on the subject.
During the interview he mentioned the startling fact that the average record length for river flow in the UK is 36 years.
Remember that next time you hear about the 100-year flood.
Reader Comments (17)
Here is a CEH paper from *15 years ago* that explains how to use river records with other, historical records to estimate flood history. Took me about 30 seconds to find on Google.
http://nora.nerc.ac.uk/8060/1/BaylissRepN008060CR.pdf
It's strange how at times it appears to be ok to use anecdotal evidence to infill gaps in the data, but if someone points to historical documents referring to Arctic ice melt then we just get fed the line that the record started in 1979.
This is where one needs some professional training and education in Hydrology. It's not clear those speaking in the media or in the Met Office have that education. There are established practices for taking "what we have" short durations of flood records and "extrapolate" to 100-year, or 250-year, or (whatever) even though there is no corresponding data for that duration. That's required *now* to define a design basis *now* since we can't wait for many years, e.g. 100 years, to collect the data to *pretend* we have the correct number. This used to be done using Gumbell Distribution methodology which is designed for "extreme events". I'm personally not sure what done now, but it can't be (or should not be) much different.
Even when one does this "extrapolation" to get the design basis, you still have to use that as your actual design basis. And the stake holders have to believe in that basis and provide the permissions and funding to design to that basis. I suspect that not done now either.
http://c3wales.org/wp-content/uploads/2012/10/MarkMacklin.pdf
this report from 2011 say's no and has Worldwide data on measurement of river flow
it also points to building on flood plains
River flow and flood information would have been held at a local level by Town Councils, Lock Keepers and others with a working interest.
On a more academic level, it is highly likely that a University such as Oxford, with a tradition of rowing on the Thames, might have rather better long term records than Thames Water and/or the Environment Agency. Cambridge and Durham Universities may have similar records.
I am sure there is a well known school near Slough with a rowing tradition on the Thames. Could David Cameron make some enquiries, whether he rowed at school or not?
And still we do not learn?
£200 million flagship scheme could see new homes built on floodplains
Parish records are usually a good source of (unbiased) information....
They would have documented (say) the 'Great Flood of 1796' - or whatever.....
Richard
A very useful link. Thank you.
The idea that you can pin down this sort event as the "worst ever" and thereby blame it (at least by implication — "may". "could", "consistent with") on global warming is puerile when the reliable records span less than a century and when, as was the case in Tewkesbury in 2007, the previous record had been circa 1484 it is quite impossible to identify whether the increased water level was due to increased rainfall or the inevitable environmental changes over the intervening 500 years!
This link gives a history of floods in the Severn catchment and provides fairly convincing evidence that leaping to convenient conclusions is to be avoided.
A similar argument applies in south-east Scotland (though over a matter of decades rather than centuries).
In August 1948 the River Tyne burst its banks in Haddington rising to a height of 10 feet above normal and topping the previous (8'9") record set in 1775. There was a repeat performance in September 2012 though I'm struggling to find a figure for the height of the river which suggests that it wasn't a record (knowing ny good friends at the East Lothian Courier!).
These are weather events and as long as the agencies (EA, SEPA) convince themselves that there is no need to dredge the Ulls Water to more than four-foot depth because climate change = no more floods when it suits the narrative and climate change = more floods when that suits the narrative then every time there is an El NIno and the jetstream shifts Glenridding is going to get wet and places like Tewkesbury are going to get wet because development further up the Avon and the Severn will combine with the same (unpredictable) climatic phenomena and Haddington is going to get wet because the Lammermuirs have precious little soil to retain rainfall anyway ...
.... and so on and so on!
It's w-e-a-t-h-e-r!! Stop pretending there is anything you can do about other than not build on flood plains and keep your defences in god nick!
Language is interesting.
The articles I see usually start with a claim of "unprecedented" - which implies that it has never happened before.
Then it is claimed to be a one in one thousand year event - which over 4.5 billion years means that it could have happened about 4.5 million times before.
Only in climate science can 4,500,000 = 0.
Our greenie friends have been complaining to the BBC about the LACK of coverage of climate change, especially when reporting floods, any guest or presenter who fails to mention CC gets complaints, the Feedback programme today on radio 4 had a long segment on these complaints. Some people want all floods blamed on CC!
The BBC totally fails to see that this is a political issue requiring balance, not a scientific one in which balance can be neglected, due to the overwhelming consensus thing that keeps getting trotted out.
It is interesting to note that many river level monitoring stations were commissioned after significant flood events in C20th because the local bureaucracy realised there was a need for real data for planning purposes...
As far as river level monitoring is concerned - the EA is in the frame there...
What is unsurprising and is actually causing real damage is the inability / unwillingness to join up all the monitoring resources. The EA has its stuff, the Met Office has its stuff - and never the two shall meet except it seems to massaged to justify failure related to some "unprecedented" , "unforseen", calamitous event for which both parties seek to avoid blame consequent from their Cnut -like predictions that have previously been presented as omniscience ...
Work has been done on integrating rainfall radar with associated water level monitoring - but these baboons seem to think that memos, PR releases, Twitter and cultivating tame politicians is the way to go when addressing public hazards consequent from poor weather - well - I've got news for them - it ain't.
@rms 11:46 am. The statistical extrapolation is frequently used to "predict" 100-year, 500-year, 1000-year extremes. Isaac Asimov has a beautiful story, "Franchise", on an electronic democracy. In it, statistical polling techniques have been perfected and it is no longer necessary to hold expensive elections.Only a small poll is needed, and statisticians have been able to reduce the required sample to 1 (one). Worth reading. There is a strong resemblance to today's climatology.
The Western Mail, a Welsh newspaper, has a rare sceptical article on the subject of flooding by Professor Mark Jacklin of Aberystwyth University.
Storm Desmond-style storms and floods are not so rare in Welsh weather history, warns scientist
http://www.walesonline.co.uk/news/wales-news/storm-desmond-style-storms-floods-10575917
"An expert at Aberystwyth University says there are many historic examples of such flooding."
The BBC is going to wear out the word 'unprecedented', and then there will be no more storms in their tea cups either.
"Parish records are usually a good source of (unbiased) information...."
Dec 11, 2015 at 1:22 PM | Unregistered Commentersherlock1
They can be even more unbiased than you may imagine. I had a look, a couple of months ago, at the facsimiles on film of the parish records from Salthouse, Norfolk. The original records suffered from severe water damage and what survived was rescued, quite a few decades back, by taking apart the books and conserving the pages separately. That makes them hard to follow, since, when a book is deconstructed like that, you no longer find page 83 opposite page 84, or the birth records from 1680-whatever opposite the burial records from the same year.
Well, when I got towards the bit that really interested me, it wasn't there, because it was in the part of the records that was beyond rescue. That, however, just showed how unbiased the floodwaters could be, because it wasn't the most recent pages that were impossible to save, but merely the ones in the centre of the book.
I have rowed a lot at Oxford and York, and spent quite a lot of time in rainy parts of the Lake District as well. My impression about these 'worst since whenever...' statements is that they are based on lines drawn on the wall of local buildings, to mark the flood high point - rather like pencil marks drawn on a wall to show how tall your children were once upon a time. I'm not sure these are 'flow' records.