Protection from Flooding

One of the aims of Researchers' Night all across Europe is to put people in touch with researchers whose work could directly affect their lives.  Dr Nick Odoni is a very good example of this as his research looks at ways to mitigate and avoid flooding and in particular to protect the town of Pickering in North Yorkshire from the flooding it's so often prone to.  Meera Senthilingam spoke to Nick to find out more.

Nick -   Pickering is at the end of a melt-water gorge that was formed probably during the last one or two Ice Ages.  The mean flow in Pickering beck is only a cubic metre per second or less over the year but occasionally you get really big storms over the North York Moors and the flow increases enormously, and there's a history there for flooding in Pickering when get these really big thunderstorms or really big slow moving depressions.  And we had a big event in June 2007 when this 1-metre average flow was suddenly increased to over 30 cubic meters per second and the town was quite seriously flooded.

Meera -   What is the first line of defence against this? Just to, kind of, barricade the town?

Nick -   Well, you've got spot on.  Normally, what would happen in this situation is we would think in terms of hard-engineered flood defences.  Now this could mean various things.  It could mean a bypass tunnel.  It could mean making the channel deeper and wider.  It could mean erecting hard concrete walls protecting the town say for two/two-and-a-half feet higher than the present banks of the river, extending some distance upstream or some distance downstream, or both.  But there are two particular problems with these defence mechanisms.  One, they're very expensive and two, they speed the flood wave downstream so you may protect Pickering but you may make it a lot worse for people further down the system.

Meera -   What are your possible solutions to avoid this?

Nick -   Well we're looking for something which might be cheaper and also more sustainable in the sense that it fits in more neatly with the natural environment.  We are looking at low impact upstream storage and structures which will slow the flow.  So as we farm the landscape, remove the trees, remove vegetation to a certain extent, the water speeds up very easily into the channels and that gives you a very sharp responsive system.  So what we are trying to do is by planting trees, by planting buffer strips by building ponds, and in this particular example we are seeing today, by building a small dam, small bunds; 1 or 2, 3 meters high at most, we can stretch out the flood wave again.  We can lower the peak and that's a much more sophisticated, but we hope sustainable way of preventing flooding in the town.

Meera -   Now, are each of these different measures, so the planting of trees and the creating of ponds all equally important?  So do you need the combination of all of these or some are going to make a bigger difference than others?

Nick -   Some are going to make a bigger difference than others in theory.  However, we have all the competing land uses and land management options to consider.  So we have a national park.  We have several SSSI's (Sites of Special Scientific Interest).  We have multiple land users.  We have the North York Moors railway. So although a scientific exercise might say, okay, bunds first then riparian buffer strips then ponds.  We may find that for land use reasons, with the competing pressures, we can't do it that way so therefore we have to consider all of them.

Meera -   Now you have a model set up here on the laptop as part of the exhibition in order to get a better insight into how these possible solutions will work.  So what do we have to do here?

Nick -   Okay.  What we've got here is a very simple bunding model.  This is just one aspect to the things that we are looking at and the idea is that people can have a go at selecting places in the river network along Pickering Beck or the other tributaries in the catchments and saying I'm going to try a little bund there, and a little bund here and a little bund somewhere else.  We can work out the storage volumes for bunds of different sizes.  And by combining these, we can see how much storage can we get and therefore can we take off a critical amount of flow from Pickering and prevent flooding?  Now, our target in this exercise is 10 cubic meters per second, 10 cumecs,  so we have a flow which is theoretically 30 and we're going to try and reduce it to 20 for as long as we can.  Can we get two hours protection?  Can we get five?  Can we get 20 hours of protection?

Meera -   A question I'm not quite to understanding about the whole prevention of flooding is you manage to keep this away for hours, and then what?

Nick -   Well, we know that probably seven hours isn't likely to be enough.  The extent the flood into 2007 was more like 16 hours, and in terms of a volume of water, it was something like 400,000 cubic meters of water.  We also know from our modelling here, that these two structures we've put in are giving us about 180,000 cubic meters of water.  So we know we need to try other things.  Now that might be more bunds or it might be some of these other measures like ponds or riparian buffer strips to further slow the flow.  We can also consider the re-meandering the channels because by putting the meanders back into the network, the water travels more slowly and therefore again it attenuates the top of the flood wave.  Finally, however, this is only an exploratory model.  We must test these results rigorously using the standard engineering models that the Environment Agency use.  But once they see our results they know where to test and they know what to test.

Meera -   How transferable then do you think these solutions and these various methods of preventing flooding are for other towns in other geographical regions that are prone to flooding?

Nick -   Generically, all of these ought to help to a certain extent but much will depend upon regional geology and land use and climate.  For example, the Pickering system doesn't have much in the way what we called ground water flow.  You might find on a chalk catchment or an area where there's a lot of limestone, we have to try different types of solution because there's so much more water coming from the ground.  So, it's not perfectly transferable.  But the idea is that you should use an exploratory model first such as this, and then go the more rigorous physically based models to work out the exact solution and the exact place.