U.K. Floods: If Only They Had Dredged?

Over the last couple of months, we have seen a pack of welly-booted U.K. journalists from News at 10, Newsnight and Channel 4 News interviewing a series of rather depressed and bedraggled flood-victims of Somerset and the Thames Valley. A good proportion of these have been parroting The Daily Mail‘s claim that it is all the Environment Agency’s fault through a lack of dredging.

Is it true?

Let’s look at an Environment Agency presentation called “River Dredging and Flood Defence: To Dredge or Not to Dredge“. To start with, we have a river channel and an adjacent floodplain.

EA Floodplain jpeg

When we have extreme precipitation, the channel overflows into the floodplain.

EA What Happens jpeg

The critical question for the Environment Agency relates to the relationship between the flow of water that can be accommodated in the channel to the flow of water that can be contained within the adjacent floodplain during an extreme weather event.

EA Change to River jpeg

And in the case of the January floods, we do have data to suggest that river flow has been setting records and so could not have been accommodated in most river channels unless extraordinary aggressive and expensive dredging had taken place. The chart below is from the Centre for Ecology and Hydrology’s January 2014 Hydrological Summary (click for larger image):

River Flow jpeg

The black circles indicate exceptionally high flow, and the arrows a new record. So the Thames, for example, recorded a flow 263% above the long-term average in January (you may need to follow the link to the monthly report to see this clearly).

The Environment Agency then goes on to point out that if you want to keep water off the floodplain then you would need to both deepen and widen river channels and repeat dredging at regular intervals to prevent silting. This all costs money, but it can be done.

In my former home of Japan, flood defence has to account for deluges following typhoons. Accordingly, the system must cope with irregular floods that are many multiples of the long-term average river flow. As a result, the country is covered in massive concreted river channels through which just a trickle of water flows for many years—until a big typhoon hits. It is not pretty flood defence. You could do this to the Thames, but it certainly would no longer look like the Thames of Jerome K. Jerome or Kenneth Grahame by the time such work had finished.

Moreover, you can’t modify the river channel at one point and not at others. If you did, water flow would hit pinch points, which are frequently structures like bridges.

Pinch Points jpeg

The agency also directly contradicts an assertion made by Christopher Booker of the Daily Telegraph. Booker claimed that if only dredgers were allowed to dump silt on the river banks costs could be kept down. The dreaded EU is supposed to prevent such a common sense action for environmental reasons. As with all Booker’s writings, anything that doesn’t fit the story is, of course, left out; in particular, the common sense result of dumping silt on river banks is to raise the floodplain and reduce the volume of water that can be held therein.

Raising the Floodplain jpeg

Against this background, let’s return to The Daily Mail accusation; “non-dredging “was what did it”. Below is an example of the reportage. As an aside, almost all the U.K. climate skeptic web sites have been carrying these photos as well. The Daily Telegraph also ran with them in an article here. They specifically relate to the Somerset Levels.

Daily Mail Dredging jpeg

What I don’t like about these ‘before and after’ photos is that they are taken from a different angle. In the second photo, we can’t see the span of the river channel since the left bank blocks our view. However, there does appear to be some build-up of the river bank on the right bank as evidenced by the circular hole being partially obscured. Nonetheless, we can’t calculate the change in flow capacity from these ‘before and after’ pictures alone.

We also don’t know exactly when the original photo was taken: only in the ‘1960s’. We do know, however, that the Somerset Levels had a massive flood in 1960. You can see this in a County Gazette 50th anniversary article here, with a series of photos here. There were also, incidentally, major floods in the area in 1951, and these led to comprehensive flood defence works—which obviously weren’t sufficient to cope with the rainfall in 1960.

The Daily Telegraph story with the ‘before and after’ photos also includes this assertion from a local pro-dredging group Flooding on the Somerset Levels Action Group (FLAG).

A spokesman for the FLAG group has got hold of meticulous rainfall records for the area around the Parrett and Tone for the last 20 years.

They reveal between December 1993 and February 1994 around 20 inches of rain fell – five inches less than during the same time this year.

A spokesman for the group said: “So roughly the same rainfall but far more flooding now.

“What has changed? Dredging seems to be the biggest obvious difference between then and now.”

Based on FLAG’s figures, 20 inches fell in 1993/4 against 25 inches on this occasion. This is not really “roughly the same rainfall”; 2013/4 is 25% higher than 2013/14 on their numbers. Given we experience flooding when a river channel reaches capacity, this 25% by itself looks significant.

Moreover, we don’t have access to FLAG’s data, but we do have access to rainfall records via the Met Office for the nearby Yeovilton weather station. Yeovilton is approximately 18 miles away from Burrowbridge, which is in turn in the heart of the Parrett/Tone area and the site of the bridge photoed above. Yeovilton is also 5 miles north of Yeovil in the Environment Agency map below (although not marked) and firmly inside the Parrett Catchment Flood Management Plan (CFMP).

Parrett CFMP jpeg

In December 1993 rainfall was 117.9 mm at Yeovilton, 99.8mm in January 1994 and 81.1 mm in February 1994. That comes out at 28.9 cm or 11.4 inches, although I admit it is very difficult to tell if we are comparing apples and pears here. We don’t yet have Yeovilton data for the current flooding episode (is released with a lag), but when it comes out we will have a direct comparison of 1993/4 and 2013/4 rainfall.

Interestingly, while the Somerset Levels have been badly affected by the current floods, centres of population that are deemed at risk have mostly come through unscathed. A June 2013 report on the Parrett CFMP pinpointed the larger towns susceptible to flooding, none of which have been hit hard this time around:

One Percent Change of Flooding jpeg

So Environment Agency flood defence work to date can be seen as quite successful in terms of protecting built-up areas in the Parrett CFMP. It has been less successful in protecting isolated villages and rural areas, but to a large degree this was due to choice rather than neglect.

At this point, I will stress that the Environment Agency’s own work does show that dredging would have a substantial impact on reducing the number of days villages and agricultural land remain under water. The charts below have been going around all the usual climate skeptic blogs and are taken from an Environment Agency hydraulic modelling study that can be found here:

Drainage Model Parrett jpeg

The charts show a far quicker recovery from flooding events after theoretical dredging, although in the case of Curry Moor the peak flood level is very similar before and after the dredging.

So is that game, set and match to the pro-dredgers? Not quite. For a start, we should always keep in mind that the Somerset Levels are quite unique in terms of their hydrological challenges and don’t provide any real lessons for major rivers such as the Thames. But far more important is the fact that the same climate skeptic blogs who printed the above chart rarely print this one (source here):

Parrett and Tone Cost Benefit jpeg

The above charts show benefits and costs under different assumptions So the choice to dredge could have been made, but the benefit-to-cost ratio (BCR) would be somewhere around one to one.

Is such a project in line with government guidelines? In May 2011, a new approach to the funding of projects was unveiled by the Department for Environment, Food and Rural Affairs (DEFRA) with respect to flood and coastal erosion risks . You can find details here. The underlying methodology underpinning all government benefit-cost-ratio calculations can be found in the Green Book, the Treasury’s bible for expenditure decisions.

An evaluation of Defra’s flood defence project decision-making was undertaken by the hydrology specialist JBA Consulting in March 2012 (here); they looked at a series of case studies to see which ones would have received government grants. You can see these here (click for larger image):

JBA Consulting BCR jpeg

The media frequently reports a benefit-to-cost ratio of eight as the hurdle rate for flood projects to go ahead, but, as the table above shows, the methodology is more complex than that. Nonetheless, only three of the six projects that would have been eligible for flood defence grant in aid—the ones which had partnership funding scores above 100%. These three had benefit-to-cost ratios of 7.23, 8.14 and 4.68, respectively. The Parrett and Tone dredging scheme can’t get anywhere close to these numbers.

In short, David Cameron, who has just announced the recommencement of dredging on the Somerset Levels, has thrown Treasury benefit-to-cost ratio guidance out the window due to the political fallout from the floods. I doubt, however, that this ad hoc response will translate into the longer term promotion of flood defence projects with low BCR ratios. Why should flood defences be treated differently from cancer wards, kindergarden places, pedestrian crossings or pollution control? And given that the Treasury has a hard budget restraint, my belief is that they won’t be once the political hullabaloo has subsided.

Meanwhile, climate change has barely begun to change the risk equation. The Environment Agency has this to say about the Parrett catchment in a time of climate change:

Climate Change Parrett jpeg

The 500 mm sea level rise figure for 2100 is possible, but the scientific consensus in now for an upper level of possible sea level rise of one metre or more by century end. Nonetheless, even using the agency’s current, rather conservative assumptions, flood risk for  towns in the Parrett catchment jumps significantly:

Current and Future Flood jpeg

My own personal opinion is this: since climate change is radically altering flood risk, individuals should proactively protect themselves. Expecting the government to always bail oneself out through dredging, insurance or whatever is, in itself, a high risk strategy. There is a limit to what the government can do—or afford to do. And when the government reaches such a limit, you are on your own.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s