Floods: 1 in a hundred year event within 4 years!

[Adam1]

http://www.guardian.co.uk/environment/2 ... -policeman

The environment secretary, Hilary Benn, who travelled to Cockermouth to assess the flooding, said defences built after the 2005 floods, which were designed to withstand a "one-in-100-years flood", were unable to cope with the volume of water.

[biffvernon]

He went on to say that it was more of a one in a thousand year event - which is a pretty silly statement. What he should have said was that as Anthropogenic Global Warming proceeds the probability of unusual weather events will increase. Yesterday's event is in line with model expectations and we will need to divert a significantly greater proportion of GDP towards adaptation and mitigation.

But he didn't. He just said once in a thousand years, which foolish people will interpret as meaning there was no way we could have predicted it and it won't happen again.

Mind you 372mm above Seathwaite is one hell of a lot of water. The greatest 24 hour rainfall total ever recorded in England.

[SILVERHARP2]

How much of this is down to building? if exactly the same amount of rain fell 100 year ago would these old town centres have been flooded as badly?

[clv101]

What he should have said was that as Anthropogenic Global Warming proceeds the probability of unusual weather events will increase....

Hilary Benn has the environment brief, not the climate change brief. That now sits next to energy with Ed Miliband.

I was due to go up to the Lakes tonight for a weekend's hiking. My bunkhouse is underwater now so I'm not.

[Mark]

I was due to go up to the Lakes tonight for a weekend's hiking.

Maybe we'll all need to take up canoeing to enjoy a weekend away in the Lakes.....

[biffvernon]

Hilary Benn has the environment brief, not the climate change brief. That now sits next to energy with Ed Miliband.

Yes, a subtle distinction. Ed's job is global warming prevention - as in not allowing coal burning. Hilary's job is mitigation and adaptation - as in building drains for the expected frequent event rather than not building them for the 1 in 1000 yr event.

Still, it was a hell of a lot of rain.

How much of this is down to building? if exactly the same amount of rain fell 100 year ago would these old town centres have been flooded as badly?

It's hard to be sure but my guess would be that it would have caused much more damage 100 years ago. Bridges and culverts and embankments are now in much better order than they used to be. Go back a thousand years though, to when a greater proportion of the fells were wooded, and it might have been a different story. Runoff tends to be slower from woodland than from grassland. More trees in the uplands could have a significant effect on reducing the suddeness of floods.

Talking of Ed Miliband,his last night's LSE lecture text here http://bit.ly/4CZ5uZ

Quite good, considering.

[2 As and a B]

A vision of the future.



Communication breakdown.

[fifthcolumn]

Runoff tends to be slower from woodland than from grassland. More trees in the uplands could have a significant effect on reducing the suddeness of floods.

Now there's a good reason for planting trees.

Slightly off topic but I read in some scientific journal or other than studies have indicated that it's the cutting down of trees on the coast that creates deserts as far inland as a couple of hundred miles.

[Catweazle]

Runoff tends to be slower from woodland than from grassland. More trees in the uplands could have a significant effect on reducing the suddeness of floods.

Now there's a good reason for planting trees.

Slightly off topic but I read in some scientific journal or other than studies have indicated that it's the cutting down of trees on the coast that creates deserts as far inland as a couple of hundred miles.

I think that some studies have blamed the drought in Australia on this.

[RenewableCandy]

Runoff tends to be slower from woodland than from grassland. More trees in the uplands could have a significant effect on reducing the suddeness of floods.

Now there's a good reason for planting trees.

Yes but people whinge about wanting to "conserve the moorland" or "preserve the meadows" both of which habitats would, if left to themselves, become forest.

[fifthcolumn]

Runoff tends to be slower from woodland than from grassland. More trees in the uplands could have a significant effect on reducing the suddeness of floods.

Now there's a good reason for planting trees.

Yes but people whinge about wanting to "conserve the moorland" or "preserve the meadows" both of which habitats would, if left to themselves, become forest.

Right then. Make me supreme dictator and I'll split the country in three.

The Nimby's and "nothing is wrong" can live in one part of the country.
The tech people can follow my plan for a glorious technological future and the hippies can have the other third, "back to the land" etc.

[Cycloloco]

He went on to say that it was more of a one in a thousand year event - which is a pretty silly statement. What he should have said was that as Anthropogenic Global Warming proceeds the probability of unusual weather events will increase. Yesterday's event is in line with model expectations and we will need to divert a significantly greater proportion of GDP towards adaptation and mitigation.

But he didn't. He just said once in a thousand years, which foolish people will interpret as meaning there was no way we could have predicted it and it won't happen again.
.....

The trouble with this sort of statement is that the event was predicted in a fairly specific way by date (3 months ahead) with no ref to CO2. See the recent news archive messages on:

http://www.weatheraction.com/pages/pv.a ... t3&fsize=0

[enso]

He went on to say that it was more of a one in a thousand year event - which is a pretty silly statement. What he should have said was that as Anthropogenic Global Warming proceeds the probability of unusual weather events will increase.

Yes. And to go into a bit more detail, the limitation with proclaimations of an event being a 1 in Xhundred years flood is that this is a statistical analysis based on the available river gauging data and/or rainfall data. Most river gauge records go back maybe a couple of decades. The standard UK flood estimation methodology from the Centre for Ecology and Hydrology builds up a flood-frequency relationship for any given location as follows in this rough and somewhat truncated summary:

- identify a group of gauges from similar catchments (i.e geology, soil type, topography, catchment area)
- sum the records for these gauges such that the total number of years (10 at this gauge plus 25 at this one plus 17 at another etc) is equal to 5 times the greatest return period of interest. So, to estimate the 1 in 100 year flood you need gauge records that add up to at least 500 years
- this dataset is then analysed to generate an average flood frequency relationship. i.e 30 m3/s for the 1 in 5 year event, 120 m3/s at 1 in 50 years, 180 m3/s at 1 in 100.

The point being that if you are tasked with assessing flood risk or analysing a particular flood event then you are limited to basing your assessment on the data available, which is largely restricted to being from the last 60 years or so. The problem with this is that it doesn't account for the fact that the climate is something of a moving target, human influence or otherwise. So what Hilary Benn should really have said was "it was a one in a thousand year event based on the flood frequency relationship of the last 60 years". Unfortunately this is pretty much the best we can do. I guess the Minister didn't quite manage to put the assessment of the flood event into this context.

Nevertheless, 372mm of rainfall in a single day is huge for the UK. Although, often as important as the actual rainfall is the degree of saturation of the soil and the height of river levels prior to the onset of rainfall. The prior conditions are really important as to how big a river flow gets for any given rainfall event.

How much of this is down to building? if exactly the same amount of rain fell 100 year ago would these old town centres have been flooded as badly?

My feeling is that in general, 100 years ago the impacts would have been less as we had less infrastructure and urban sprawl in areas of flood risk. I'm not sure on the specifics of this particular flood event though.

...my guess would be that it would have caused much more damage 100 years ago. Bridges and culverts and embankments are now in much better order than they used to be. Go back a thousand years though, to when a greater proportion of the fells were wooded, and it might have been a different story. Runoff tends to be slower from woodland than from grassland. More trees in the uplands could have a significant effect on reducing the suddeness of floods.

Biff, you probably have a point about the integrity of structures but I'm afraid that you are misinformed about the effects of woodland on flood generation. This is a widely held belief that forests 'soak up' rain water and reduce runoff. I thought this was just common sense until I got involved in flood hydrology work. However, decades of careful catchment monitoring pretty clearly shows that in the UK at least this just isn't the case. There appear to be a number of reasons for this (with a load of good scientific literature to back it up):

- the water storage provided by the surface areas of leaves are dramatically reduced in deciduous woodland outside of the growing season
- typically, the additional storage volume of the leaves combined with the effects of evapotranspiration can accommodate around about the first 5mm of rainfall, after which the canopy is effectively saturated and any additional rainfall then reaches the soil

Forests do appear to have some effects on reducing flooding but only in ways very specific to a few locations. For instance, in areas where flooding occurs from a rapid rise in groundwater such as happens in low-lying chalk areas, widespread tree cover can help reduce groundwater levels during dry periods thereby allowing aquifers to absorb more water during extreme rainall before they become full and can lead to surface flooding. Also, woodland on floodplains can slow the rate of flow of flood waters and as a result slightly lower peak water levels downstream, but only by a small amount.

So, I would conclude by saying that whilst there are many good reasons for wanting to increase the woodland cover of the UK, for most places this is largely irrelevant to managing the risk of flooding. Hope this was helpful.

[biffvernon]

Interesting. Are there perhaps forests and forests? It's common to see, in pine plantations, small ditches running straight downhill between each row of trees, specifically to speed up drainage. I was walking in Galloway a couple of weeks ago and thought this is not going to help slow runoff. On the otherhand, I crossed a new plantation of mixed deciduous trees that had been planted as a biodiversity and habitat improvement scheme where exactly the opposite effect seems to have been created.

I'm sure your statistical comments, Enso, are spot on. But the newspapers are all saying once in 1000 years. Without the backstory, the headlines are completely misleading. It's clearly politically convenient to make people think it won't happen for another 1000 years, but the increased flow of energy from the equator to the poles is bound to result, in this part of the world where tropical and polar airs meet, in a call for more umbrellas.

Anyone notice that last week saw China's heaviest November snowfall in living memory?

[adam2]

As enso posts above, the calculation of flood risk is far from accurate, due both to lack of enough data, and a changing climate.

However even if accurate figures are available, most people have a very poor understsanding of chance and statistics.
Many seem to assume that a "once in 100 years" flood event means it wont happen again for 100 years. In fact the risk is 1% a year after the flood or 99 years after the previous flood.