Bishop Hill On the floods in Cumbria
Dec 7, 2015 Climate: Statistics Climate: WG2 Environment Agency ;)
The floods in Cumbria are obviously attracting a lot of attention this morning. A couple of things are exercising my mind.
Firstly, as readers are pointing out, the claim that 340mm of rain fell in 24 hours seems suspect. Nobody seems quite sure where it came from. I have seen it suggested that this was the amount that fell over two days and in Unthreaded, Mike Post wonders if it is actually the November rainfall total.
Secondly, how do we know what the 100-year flood is in any given river basin? The process looks less than straightforward and involves a wealth of assumptions.
The first assumption is often but not always valid and should be tested on a case by case basis. The second assumption is often valid if the extreme events are observed under similar climate conditions. For example, if the extreme events on record all come from late summer thunder storms (as is the case in the southwest U.S.), or from snow pack melting (as is the case in north-central U.S.), then this assumption should be valid. If, however, there are some extreme events taken from thunder storms, others from snow pack melting, and others from hurricanes, then this assumption is most likely not valid. The third assumption is only a problem when trying to forecast a low, but maximum flow event (for example, an event smaller than a 2-year flood). Since this is not typically a goal in extreme analysis, or in civil engineering design, then the situation rarely presents itself. The final assumption about stationarity is difficult to test from data for a single site because of the large uncertainties in even the longest flood records[3] (see next section). More broadly, substantial evidence of climate change strongly suggests that the probability distribution is also changing[7] and that managing flood risks in the future will become even more difficult.[8] The simplest implication of this is that not all of the historical data are, or can be, considered valid as input into the extreme event analysis.
Lots to dig into.
Bishop Hill
On Twitter, Derek Sorensen tracks it down to Honister.
Reader Comments (73)
Looking at the pictures, it looks like the flooding is taking place on flood plains. Its human stupidity to build or buy a house on natural geographic features such as these.
Anyway wasn't the CAGW meme at one drought and not floods ?
If only we didn't have such high demand for housing...................
Just check some of the amateur weather station data.
e.g. http://ehideaway.uk/weather/cockermouth/records.php has data for Cockermouth and they recorded 60.8mm of rainfall for Saturday 5th December. This figure is not the all time record for this station which is over 100mm.
The 340mm figure relates to Honister.
'Provisional figures from a rain gauge in Honister in Cumbria have shown that 341mm (over 13 inches) of rain fell in 24 hours - potentially a record breaking amount and more rain than seen during the devastating floods of 2009 and 2005 in the same area.'
http://www.itv.com/news/2015-12-06/record-breaking-rainfall-in-the-uk/
An X cm in 24 hours figure does not really convey the size of the problem, it depends on where nearby rain goes, towards you or away from you, and on the size and shape of your basin. Also, a dense concentration of rain may fall partly outside your basin one year, but the same concentration may fall mostly in your basin in another year. Finally, a monthly or yearly bin size may contain all of a heavy rain one year, but another year the heavy shower may straddle 2 adjacent bins, thereby producing lower figures.
another key assumption is stationarity of the watershed: if you build a dam or a bridge, let forest grow back, or harden the surface, then flood probabilities will change
The 2015 flood in Carlisle isn't as bad as the floods in 2005. I know as I have dry socks on this morning.
Carlisle is an ancient city at the confluence of the rivers Eden, Caldew, and Petteril. Up to 80 per cent of the traditional wetlands upriver on the Eden and its tributaries have been lost since the 1950s. These areas, use to absorb heavy rainfall before slowly releasing it into rivers, they were drained to improve grazing.
The city's magistrates court and the councils civic centre building were swamped by up to 8ft of water, well what a surprise. The main roundabout, 'Hardwick Circus' is actually built on the site of the 'old Priestbeck bridge' and the "Sands Centre' and it's car park are built upon the now filled in Priestbeck. Guess where the above buildings are located, YEP, in the old Preistbeck flood plain.
I don't know what the professionals are doing, but clearly the BBC News Channel has little experience or on-hand experts, who can explain a) risks of building and living in known flood plains, how civil engineers are *supposed* to design flood mitigation (but if funding does not follow then they might be under-designed), how urbanisation affects run-off, and how changes in run-off might affect flood mitigation plans built for another time, etc. etc. Basic Hydrology 101 and Hydraulic Engineering, taught in all good civil engineering universities.
Honister is quite close to Seathwaite, famously the wettest place in England, due to its geography. We used to holiday in Seatoller, at one end of Honister Pass, when we were kids. It felt like it hardly ever stopped raining.
Having worked for the Thamse Water Authority, as it was in the 1970s & 80s, mosy of the work undertaken by that august body was main river suverys & their tributaries, & the preparation of numerous flood alleviation schemes, whereby natural water-courses were modified so that water could be "controlled" & "guided" down certain pathways & chanels to avoid excess flooding, not necessarily to stop it altogether. My thoughts as an engineer, are tht avery little of this continued after prvatisation as the the water utilities were concerned only with water supply (the profiatable sector), not with maintenace, which was left in the hands of the "Environment Agency", which seemed to have little money for such capitol schemes & programmes. Clearly, under-investment in flood defences has been a major failing of successive guvments over the lasy 30 years or so, & needs to be addressed. Therefore, residents of these unfortunate locations, should be collaring & bending the ears of their local MPs, to spend less onwindmills & solar panels, & more upon flood prevention schemes, they'll probably find it cheaper in the long run!
Whilst my heart goes out to the poor people of Cumbria facing such devastation just before Christmas it appears to me that in this age of heavy rains, from whatever cause I know not, there is more that can and should be done. It is in the nature of the historical development of our towns and cities that the narrowest points of waterways - at the head of estuaries for example - were bridged. On almost every waterway I can recall or see on TV in the news broadcasts of floods there are old stone-arched bridges, usually partially blocked by debris, which are restricting the flow of floodwaters to the sea where they belong. In sensitive locations it is essential that these bridges regardless of their historical interest be removed and the flowpaths opened up and modern higher, wider, bridges be installed. Cheaper in the medium/long term than spending billions on making the towns' freeboard a few cms higher each time it rains heavily.
"Environment Agency’s rain gauge in Honister showed 341mm (13.4in) of rain fell in 24 hours from Friday evening at 6.30pm to the same time on Saturday." Guardian
also someone tweets 341.4mm with a graph
Yet the EA are retweeting this from one of their guys
"@TobyWillisonEA Dec 5
Provisional figures from a rain gauge in Honister #Cumbria shows 352mm of rain fell in 24hrs, if verified this would be a UK record" .....seems figs are murky
I don't find a convenient web interface to the EA guages , just an api..for programmers
BTW someone quotes 340mm as a 2009 record
BBC web "Storm Desmond looks likely to have set a record - provisional figure of 340mm of rainfall in 24hrs in Lake District. The 2009 high was 316mm"
.....see it's a convenient number if it's less than 316mm they can't call it a record
One source with extra high number is an anomally ..not news.
Well sympathies to the victims of the floods. It certainly looks like a 1 in 250 year flood from the helicopter video. Here overseas it is a silent video, I don't get the music. (naming storms does make for easy reference)
0-50s Lyth Valley
- Flooded farm is obviously much lower than adjacent dual carriageway, which cuts thru flooded land like a causeway. Looks like it was a floodplain in the past but then the river was narrowed and held in those high banks that the sheep are sitting on.
A591 - the river must have flowed thru massively causing broken road and landslides, which sometimes trapped water on the road. Vast hillsides are just low grass with no trees so runoff must have been fast.
Keswick Town - does look screwed. Now very dense housing ..too many holiday homes probably.
- Two rivers join together and feed into the north end of the lake, so anywhere around there is liable to flooding. Those white hotel/timeshare buildings look to be there.
2.26s Carlisle looks screwed with water about 3ft deep around a large area that football ground and normal housing with rescue boats ferrying people. Same place it flooded in 2005. Is it where the 3 rivers meet ?
..oh that link ends with a quote from the EA guy "The flood defences we have put in place would accommodate and defend against the flooding of 2005. The city would be safe from flooding."
and thereis the funny video "Dont Make Unnecessary Journeys" : Dance Remix of RTÉ's Teresa Mannion by Super Céilí
Action : You'd think when flood rain is forecast the EA would do everything in ADVANCE to almost zero the level of the rivers by opening sluices into designated flood runoff areas ?
..that's why in Europe you see those giant concrete rivers to channel floods to the sea.
Every few years Brisbane (Australia) experiences flooding.
Brisbane is built on the flood plain of the Brisbane River.
I wonder why Brisbane get these floods?
There's some truth in that but only for small scale catchments, especially in short duration summer thunderstorm floods, but over large catchments (in a UK context at least) changes in land use have very little if any bearing on the severity of extreme flooding events. The reason is simple, in extreme UK winter rainfall and snowmelt events, the ground is already saturated, and it does not matter how many trees you plant, or how many areas of sacrificial land you have, how many beavers George Monbiot personally re-introduces in the upper catchments, the water will find away around them pdq.
Large dams can and do have a significant effect in mitigating the severity of flood events, even in the UK. The 1990 and 1993 floods in Perth (Scotland) would be much more severe had it not been for the hydro schemes in Rannoch and on the Tummel impeding significant volumes of water for long enough, such that the peak flows from the un-regulated rivers in the south and east of the Tay catchment got through the gap between the Ochils and Sidlaws a few hours ahead.
My village has flooded (majorly) in 1998 and 2007. My house is about 20 ft above the normal river level (which runs through my back garden) and in 2007 the river rose about 18 ft! The old village was submerged.
The EA spent £350M clearing the river of fallen trees etc to make sure it was open in future and then added 9 INCHES of impacted soil to the top of the flood berm across the river from me! We asked why this (clearance) was not a regular occurrence and they claimed lack of funding. They also used the excuse of AGW!! When they finished they told us the flood defences had been upgraded for a 1 in a 100 yrs with a 50% increase for AGW (so now 1 in 150). Seems to me they like to use AGW as an excuse when what was required was proper flood-plain sluice-gate management to control the flow of water in the rivers. And now that that is followed we don't get flooded.....Hmmm.
BTW: Even with a letter from the EA explaining the work done and the 1/150 claim I am unable to get insured by anyone other than the company I'm already with (£7,500 excess now!)
"there are old stone-arched bridges, usually partially blocked by debris, which are restricting the flow of floodwaters to the sea where they belong"
But the water level on the downstream side of these old bridges seems no lower than the upstream side.
So you've be wasting your money. As Alan the Brit says we need proper maintenance upstream.
Can I suggest reading the report at: http://www.floodpreventionsociety.org.uk/FLOOD%20REPORT2015a.pdf
Oh, thanks Nial! We definitely need to regain some common sense & dispose asap of the AGW nonsense, it's is costing lots of people lots of money, & will ultimately cost Human lives, but then again, "they" couldn't give you know what about that! The EU's Water Directive & Habitats Directive need to be abolished, or leave the EU altogether & manage our own affairs, without dictats from on high from un-elected, un-accountable, un-sackable, hugely paid interfering bureaucrats!
Can't remember which news bulletin is was but one resident said he had lived there 15 years and had been flooded every 5 years. Wasn't the last bad one just a few years ago where the bridge collapsed taking away the Policeman who was redirecting traffic ?
Having had some professional experience of dealing with the consequences of flooding, my sympathies to all involved.
As Alan the Brit notes above, the responsibility for keeping water courses clear of obstructions, became an expensive liability that with privatisation, no one had a financial incentive to accept. It has also been convenient for some, to avoid expenditure, and blame the consequences on global warming. Somerset Levels from 2 years ago?
100 year floods. Waterside pubs, bridges etc, are a useful source of "benchmarks" carved into masonry indicating historic flood levels, and dates. These are described as 'anecdotal evidence' by those who find these 'records' do not match their computer models.
The concept of "100 year flood" is a bit deceptive. The term actually refers to a 1% likelihood flood. And that is by no means an accurately assessed likelihood.
Land use changes, waterway maintenance, drainage issues, subsidence, infrastructure maintenance all play large roles in making flooding more, or less, likely.
The likely response these days will be
'Global Warming mate, nothing to do with me, and nothing can be done'.
See you next year.
@ Breath of Fresh Air Dec 7, 2015 at 12:24 PM
"Wasn't the last bad one just a few years ago where the bridge collapsed taking away the Policeman who was redirecting traffic ?"
That was 2009; PC Bill Barker was killed was trying to stop people crossing Northside Bridge in Workington. You might also remember Cockermouth was also badly hit that year, they've got it bad this time too although not as widely reported, with most reports focussing on Keswick and Kendal, it seems. Apparently the 2009 floods were if not caused, at least not helped, by poor management of Thirlmere reservoir.
http://www.newsandstar.co.uk/overflowing-thirlmere-unable-to-help-alleviate-cumbria-flooding-1.643278?referrerPath=news/2.6280
Paul Homewood has an interesting initial comment on the "record". I gather that the Honister record may be a bit too spotty to claim a record.
https://notalotofpeopleknowthat.wordpress.com/2015/12/07/record-rainfall-at-honister/
I have just watched some of the videos and it does look like Cumbria and the Scottish Borders had a severe rainfall event. It rained quite heavily here over the entire Tay catchment (not a common occurrence) for a day or two but nothing very extreme, most of last week's snow had already melted, so the rivers coped as they should. So just about all the rivers in the Tay system were 1m or so below their record levels (most gauges having only been in place for 20-30 years).
Some observations:
- our upland landscape is still young - it is only 12,000 years since the end of the last glaciation, and we are always going to get landslides and slips when burns and becks get high. Nothing unprecedented or even unusual.
- I don't know Carlisle well but a lot of it seems very flat, and built on a flood plain per chance?
- there is ample historical evidence for severe flood events in the UK, back to the 13th Century at least.
- most old houses had earth floors downstairs, and no electricity or white goods. Flooding was just more of an inconvenience than a disaster.
- the flood return periods for most UK rivers are only based on 50 years worth of data. And the 60s 1970s and 80s were comparatively dry decades. When the Tay flooded in Perth in 1990 and 1993, the hydrologists in SEPA were in shock and quick to blame global warming. The older and more experienced hydro-electrical engineers who remembered the wet decades in the 1950s and 60s were not so surprised.
hunter 12:34, thanks! Probability theory is not my thing, but in an earthquake zone, every successive year without an earthquake, tends to increase the scale of damage, when an earthquake does happen. Heavy rainfall, or lack of, this month, has no consequence on rainfall in 12 months time.
What has changed, especially over the last 50-100 years, is the number of buildings on the historic flood plain that are liable to flood, AND prevent rainfall being absorbed by soil.
The Lake District is impermeable rock. It is only the soil which can act like a sponge, and absorb extremes of rainfall.
More here on the Daily Mash.
What a surprise, Dame Julia on R4 WATO saying Cumbrian floods 7x more likely due to climate change but "can't be definitive", of course and need more research (i.e. £money).
Also Tim Farron repeating the meme in the next interview.
Strewth...
There you go: Slingo has just claimed that these sorts of events are now seven times more likely all due to our emissions.
She probably has a study which calculates that the probability of these floods has change from 0.001% to 0.007%.
You've got to hand it to the Met Office for getting this very alarming figure to us so quickly.
I am staggered she didn't trot out the good old, "no specific weather event can be attributable to Climate Change, but yes this is the sort of thing we expect to see more of!". As if this kind of event has never happened before!
If Slingo has just said that the Cumbrian floods are 7x more likely due to climate change then she needs to explain her reasoning, and provide some evidence. That is an astonishing claim for a supposedly reputable scientist to make - and complete bollocks. Are there no scientists left in the Met Office who have the balls to challenge Julia's idiocy?
CO2, the miracle molecule of the 21st Century, which can make severe floods even more severe and also 7 times more likely. Not to mention droughts, heatwaves, and jet-stream blocking winter cold spells when necessary. I suppose 7x does have a Biblical sound to it. Does Julia not have any idea how stupid she looks and sounds?
http://snag.gy/LPUXq.jpg
The Met Office's reputation is not great but she is shredding what is left of it.
What most people have forgotten, 'the lakes' but more particularly the mountains that surround them are a domed outcrop, a granitic batholith responsible for the areas uplift, surface outcrops particularly famous of; Carrock fell, Skiddaw and Shap. Throw water on any sort of dome, natural domes and it runs off radially - it's a classic exemplar.
Then, the ground up there over the autumn and early winter has been long saturated, the very last thing the poor people of the greater Lakes were wishing for, was another prolonged rainfall event, trouble arrived in 'Desmond', a conveyor of a series of fronts interlocking and it poured and poured some more.
Anecdotally, I watched some [presumably] amateur footage of an upland stream flowing out of a [corrie?] valley, you could see the surrounding slopes where the water was cascading off the hillsides and the small stream was a torrent - hissing vapour and coloured black brown, flying down the stream bed. In all probability and during less wet times you could skip over this stream in a couple of bounds - but not on Saturday, the blumin' thing must have been twenty feet deep and looking like a small version of the Colorado.....I mused to myself, this [event] is going to be another bad one.
"More broadly, substantial evidence of climate change strongly suggests that the probability distribution is also changing"
Differentiating between probability distributions is difficult. Is it Gaussian? Is it something fatter tailed? Very difficult to tell from small samples, but the ramifications of having a fatter tail is huge.
The upshot is that when someone tells you something is a hundred year or thousand year event, be skeptical. The statistical problem of ascertaining that is difficult.
In relation to the Lyth Valley, the Environment Agency has been wanting to flood the valley for years. As happened in Devon they want to 'give the land back to nature'. The land has been farmed for a century or more but because it is only just above sea level requires pumps, channels and dykes. The EA had planned to switch the pumps off. See this report:
http://www.itv.com/news/border/2015-10-02/farmers-and-conservationists-row-over-flood-defences/
As with the Devon floods, the actions of the EA will have significantly contributed to the problems due to environmentalism and looking after the interests of wetland birds over farmers.
"Secondly, how do we know what the 100-year flood is in any given river basin? "
There's a fairly standard way of working out the hydrological formula for any river based on previous flows. This works on the basis that the log of flow is a Gaussian distribution (although it's actually a skewed Gaussian). From this it is possible to predict the likelihood of a Q100 flood.
So, this is using actual data from past flows and some well tried "rules of thumb" as to probability distribution.
Here we go. Just got in from the car where an excert on Radio 2's 3pm news, The Wet Office,s Dame Julia has jsut stated that the amount of rainfall in Cumbria was 400mm! Are we watching a numbers game wit ever increasing amounts? I do wish they'd make up their minds!
Historic data for 24hr rainfall totals is between 0900-0900GMT, and thus, the records from July 1955 still stand, as shown here.
http://www.metoffice.gov.uk/public/weather/climate-extremes/#?tab=climateExtremes
The 'any' 24hr record in 2009 Seathwaite appears as a note afterwards (now surpassed by 2015 apparently).
There is no way of knowing if the 1955 deluges, or other deluges, may have contained heavier rainfall in 'any' 24hr period that was split across 2 recording periods. With automated instrumentation, it is probably easy cherry pick the 'best' 24hr period.
Paul Homewood's article on the suggested Honister record drew this comment from me (a Cockermouth resident - thankfully unaffected, unlike many others):
I think there has been unseemly haste to declare a new record (which it may be, by the way, for what it’s worth – but as Paul’s article makes clear, it might not be worth very much).
I suspect there are reasons for this:
1. The Environment Agency needs to justify itself and the substantial sums of money it has spent, apparently with little or no effect. So they say on the one hand that the defences sort of did their job (by giving people more time than they otherwise would have had to prepare for the floods!) and on the other hand by going on about record rainfall, thereby implying that it’s not their fault there was so much rain – how could they anticipate such an extreme event?
2. The usual suspects will jump on any passing bus to claim it’s evidence of man-made climate change.
Whether it is or it isn’t evidence of man-made climate change (and I suspect it isn’t) doesn’t much matter. What really matters, as I said in a comment on Paul’s last post, is that we are wasting billions pretending to combat climate change by reducing our CO2 emissions, while the rest of the world happily invests in new coal power stations and increases their CO2 emissions (all the while pretending they’re not) thereby dwarfing and massively outweighing our puny efforts. If we really do have billions of pounds available to spend (despite a £1.56Tr national debt and ongoing annual deficits of c £70Bn) then my view is that we should be spending that money trying to adapt to climate change and protecting us from its consequences, not wasting it Canute-like on making no difference to something that is happening anyway – whether naturally or as a result of humankind’s activities.
@Alan 400mm ....over what time rime period did she say..24, 48 or 72 hours ?
Slingo's claim (see above) is a little worrying. I get a feeling of an organisation living for the next alarming prediction/observation rather than building a solid reputation. I have a gut feeling this will end in tears.
Pull the insurance company data
Each flood claim against its Post Code Is that Post Code in the flood plane
Finally the age of each property for each claimant
Its was always clear that following the floods of water would floods of BS claims that is all down to 'CAGW' , and this is so because the idea of weather not being climate has long been kicked into the grass and these days any extreme but no usual weather event is jumped on has 'proof '
At this stage it is worth remembering the reason we moved on from 'weather is no climate ' and that is because of the failure of the more CO2 = increased temperature link to be seen in reality . Instances of extreme weather have been jumped on because they otherwise failed , not because of the science.
The Cumbrians are extremely used to having bridges swept away and needing to rebuild them. In the past this was a straightforward cost-benefit analysis. Do we just put another one in the same place for £100,000 each time this happens, or do we fork out £1,000,000 for putting in other measures (flood surge diversion channels spring to mind) to help preserve the new bridge? Apparently, these days, one is expected to draw on a £1,500,000,000,000 per annum reparation fund raised from electricity consumers: that would buy a very large number of Braithwaite bridges.
[1] Braithwaite Bridge commemorative plaque, 2010.
People don't seem to realise that anything built on a flood plain, including flood management barriers, has the effect of reducing the flood plain area which increases the height of the floodwater.
Seathwaite is in the valley. Seathwaite records should be compared to Seathwaite records.
Every fell walker knows that the rainfall there will be less than at Styhead or the summit of Great Gable. I would guess that the Honister measurement is taken at the pass, a lot higher up, so could reasonably be expected to have higher rainfall and higher records.
The Met Office and Dame Julia Soupdragon know this, so why the attempt to make capital from it? Maybe Quentin Letts was asking a the right question.
So, are they crying "Global Warming" ? I just ran into a facebook post by a Green "public figure" named Andrew Durling, who, on the cop21 topic posted a graphic "The More We Burn The More We Flood.' I could not get anywhere with him when I sent him the relevant quotes and links from Roger Pielke, Jr., and the IPCC. http://rogerpielkejr.blogspot.com/2013/01/extreme-misrepresentation-usgcrp-and.html I don't think he can follow a coherent sentence. This Guradian link is what he posted to me. http://www.theguardian.com/environment/2012/jan/26/floods-worst-climate-change-uk?CMP=share_btn_tw
BBC 6 o'clock news report from Shukman (who else) attempting to attribute climate change as a contributor. Watch carefully and there's precisely zero legitimate evidence, but of course the association is enough to drive more hysteria.
Logically it's obvious the infrastructure for removing water from property built on a flood plain and in an area of exceptional rainfall is not up to the job. That has nothing to do with 'climate change' and everything to do with poor or inadequate planning.
Mark Hodgson, Dec 7, 2015 at 4:01 PM
A good post, I wholeheartedly concur - with every word of it.
My comment on unthreaded yesterday: IMHO, rainfall, whatever the volume, does not cause flooding.
Flooding is a result of the infrastructure being unable to clear the rain before it backs up.
If you allow new housing, hypermarkets, retail and industrial parks, and car parks to be built on flood-prone green-field sites; if you do not upgrade sewage and drainage systems to take into account new property builds; if you do not stop home-owners from concreting, tarmacing, or paving their gardens without providing adequate soak-aways; if you do not regularly dredge watercourses and remove trees, vegetation and other obstructions - that is what causes flooding.