Hydrological Engineer David Ramsbottom on the Oxford Flood Alleviation Scheme

(Source:https://www.oxfordshirefloodtoolkit.com/contacts/alleviation-scheme/ ) 

We have a proposed flood channel, the yellow area on the map. And we also have some flood defences. You can see on the illustrations that the area that is occupies is very wideIn addition, including on a cross section. We have these areas of flood banks that have been put into the scheme. This slide shows two cross sections. All of this information, incidentally, has been provided by the Environment Agency and others.

The top cross section shows part of the flood scheme where there's an existing channel. And the general idea is to lower the ground and allow the floodwater to flow in that lowered ground adjacent to the existing channel. The bottom cross section shows where there isn't an existing channel. And so, they're putting in a small channel here. 

But again, they're relying on lowering the ground to get a lot of water through. But this actually is a very inefficient way of passing flood water. And the reason for that is that you're dealing with shallow depths of flow, which means that the velocities are fairly low, and they're taking up a very large width to get a relatively small amount of water through.

In a more conventional flood alleviation scheme you have a much bigger channel, which is much more hydraulically efficient, and you get far more flow. 

And that's what it looks like. And it is very similar to the previous diagram. It's a trapezoidal channel and it looks very much like the River Thames. It carries a similar flow and about 40% of the total flood flow would go down the Jubilee River.

The difficulty with the design that we have for Oxford is that because of the way it's been designed, the actual flow capacity is quite small. As I say, the Jubilee river carries about 40% of the total flow at Maidenhead whereas the flow here in Oxford is less than 15%. So, it's really quite a small proportion of the total flow. The general idea of a flood relief channel is good on a catchment like the Thames which has so much water that storing it upstream is  not really practical, simply because of the vast volume. 

So, the general idea is to get the water through the flood risk area as efficiently as you can. And that's the basic concept of a flood relief channel. What happens is water goes into the flood relief channel, and the water level drops. So, there’s less water on the floodplains and less flooding. And the graph here shows the water levels against the flow. 

The blue line represents water levels with no scheme in place, and the red line represents water levels with the scheme in place. Basically, for a for a smallish flood [a flood with a 5% chance of occurrence per year] the scheme will reduce water levels by up to 30 centimetres (point three of a metre), which is about one foot, and that's the maximum. 

For a large flood [a flood with a 1% chance of occurrence per year], then the reduction in level will be point two metres or less, about eight inches. So that's about the length of my pen. So, it's not a huge amount.

​

And I think there's a fundamental flaw in the scheme that they've designed it in such a way that the capacity is actually quite small, and the reduction in flood level is quite small, and therefore the overall flood benefit is quite small. And when the scheme first set out, there were no flood defences. That was all reliant on the flood channel. But now we have such a small capacity of flood channel, we have to have flood defences as well. The practical problem with flood defences is that we have a gravel floodplain and there is a lot of gravel in the floodplain. And when the river is in flood, it's in front of the defences for several days. So, the risk is the water will go underneath the flood defences and cause flooding. Now, I understand there has been some attention paid for that by driving sheet piles into the ground beneath the defences by not a very large amount to try to reduce the amount of groundwater flow through the ground. But as far as I know, it's not proven that these flood defences will avoid seepage underneath, causing flooding during a flood event. So, I've noted that concern regarding groundwater impacts. And also, the fact we need the flood defences in the first place because they were ruled out in the early days. 

When I've looked at the information on the internet about the scheme, there are hardly any numbers of either flows of water or water levels. This is the one diagram I could find which show roughly the flows during a flood with a 1% chance per year, which we sometimes call the 100-year flood. 

And you can see there there's a flow of 218 cubic metres per second coming down the river Thames, that's about 218 tonnes per second. So that's a vast amount of water, another 73 coming from the Cherwell giving a total of 291. I think it's slightly different, but I estimate a flow of about 30 cubic metres per second in the flood channel, which is in flood flow terms not very big. In the documentation for the flood risk assessment for the scheme, the consultants have used a figure of 35% increase in flow to take account of climate change. And the flow we've got there in the flood channel is about 15% of the total in the River Thames. And if the flow of the river Thames goes up by 35%, it's quite clear that during the course of the 21st century, our river channels will have more water in them during a flood than they do at the moment. So, it means that we are not future proofed.

​

​

The response of the Environment Agency at the moment is the Thames Valley floods scheme. And the idea here is to implement measures throughout the Thames catchment in order to reduce flows not just for Oxford, but throughout the catchment. And there are lots of different measures under consideration. But in their documentation, they recognise the volumes of water here are vast. So, it's not a question of putting some small bunds across farmer's fields, you're looking really at huge amounts of water. And when I looked at this more than 10 years ago, I managed to flood about 20 square kilometres of farmland, trying to find enough land to store water to protect Oxford from flooding. So, this is just coming to the end of its consultation period. It's a nice idea but there are two technical issues associated with this. The first thing is whether it's feasible, whether you could actually find enough volume in the catchment, because this is all about flood volume. And volumes of water in floods are vast if you imagine there's 200 tonnes per second, and that's running for maybe 2, 3 or 4 days, plus the run up to that plus the recession afterwards, you're talking about 10s of millions of tonnes or cubic metres of water. So, the feasibility of this is questionable. And then the viability. If you did find enough space to put all these storage areas, or whether you could change the land use on sufficient scale to reduce the flood runoff, would it be viable? Would it be acceptable? Would it be beneficial? Would it be cost-beneficial? I can imagine the farmers would not be pleased. But it's their land and they may not join in. So that's my summary. I just have concerns about the scheme. I think it has serious flaws. And I must admit I'm very encouraged by some of the comments made already pointing out some of the difficulties with the scheme."

Response to Questions

What about approach to whole catchments and to natural flood management solutions? One of the issues is the Oxford alleviation scheme has been done pretty much in isolation at the moment, and should we be looking at doing any upstream work?

​

So far as I understand it, this scheme has been undertaken in isolation. But if you ask the Environment Agency, the plan is to deal with climate change is to do things in the upstream catchment. And I've been involved quite a bit in what they call natural flood management, which is where you try to hold water back in an upstream catchment, either by storing it in various places around the landscape, or by changing land uses, and infiltrating more water into the soil and reducing runoff using vegetation and changing land management. And I've been involved in research on this since the mid 1990s. And then at one stage, we calculated that if we had trees all over the Thames catchment, the flow at Kingston would go down by about 40% but that's not a very realistic position. 

What I have been doing is doing looking at schemes in some detail, so they did some work at Milton-Under-Wychwood, which is a small village and we looked at completely protecting that village from flooding by doing upstream measures that involved holding water back and what we found was that to achieve, really to achieve flood mitigation, you need huge volumes of storage. And there is, as I've mentioned before, there is a big question mark about whether that's possible, practical or feasible. And in the Environment Agency’s online information about the Thames catchment scheme, they list something like 17 measures. And in each one, they quite rightly say you need vast volumes or lots of these. You need to do a lot to get the water flow down sufficiently to avoid flooding. So, in other words, it's on the agenda, it's being looked at. But at the moment, there is no proof anywhere in this country or anywhere else, that this kind of approach can succeed in reducing flood risk in large river catchments. There are around the world many reservoirs, which are multi-purpose, and they include flood control, famously things like the Three Gorges dam and so on. And these do have a huge impact on flooding, and obviously have vast structures and vast impacts. And so that's one extreme of holding water back. So, it is possible, but you end up with something vast, taking up lots of land and I'm sure would never work in in the Thames catchment. But I think the jury's out on exactly how much can be achieved by doing works upstream. I think that's got to be part of the debate. 

So, whilst the present scheme will deal with more or less the present situation, I think the Environmental Agency is relying on these upstream works. And my only word of warning, as I say, is the amount of work that you have to do. The amount of land that has to be involved is very large. And I completely agree with the gentleman representing the landowners because the landowners will have the final say on such things. 

​

The potential of Regenerative farming for flood alleviation 

On a brighter note, there are farmers in the Thames catchment looking at different ways of managing their land, including changing ploughing practices having much more ground cover, they use herbal leys and various other techniques. I think  what they're trying to do is rebuild the topsoil, reduce runoff and reduce soil loss, and there is research going on looking at the benefits of that on water runoff. And because it's farming, if something like that took off, you could imagine large areas being farmed in that way, with the possibility of significant reductions in runoff. And from all I've seen, that's the greatest possible hope, simply because we need to go in that direction to restore our topsoils. But also, there's the potential to cover very large areas of land whereas I can't imagine covering the whole attachment in trees, for example. So, as I say it's work in progress, but we're in a very early stage. And at the moment, the next barrier is whether it is viable from the point of view of costs and benefits, etc. That's further down the line, the world is changing, it may be viable in 10 years’ time.

Questions on Channel proposal

Was a larger, more conventional channel ever considered?

What I can tell you is the original concepts way back, and this is going back to about 2005, was to have a large, parallel River to the River Thames. I know there's lots of challenges in Oxford, so it's not as straightforward as the Maidenhead scheme, which is very clearly a 12-kilometre-long river that runs parallel to the River Thames. So, the flood water is shared between those two channels. That means the amount of water on the floodplains goes right down. That was the original concept for Oxford. And generally speaking, because of the gravels in the floodplains along the Thames, it's been the preferred approach for major schemes on the Thames upstream of Teddington in their thinking. So, for example, the lower Thames scheme, which is downstream of Windsor has three flood relief channels. And as far as I'm aware, they are similar in concept to the Jubilee River, this large parallel River. 

What's happened in Oxford is, it seems to me, that there's a compromise that's occurred. There's environmental pressures and pressures from other groups. I stopped working on this in 2008 and I don't know what's happened between then and now. But they did start off with a concept of a large channel, taking a large proportion of the flood flow, which would produce a significant reduction in flood level, which would be in my opinion, at least a reasonable flood scheme. But obviously, that's a long time ago. And yeah, we've moved on. 

CHAIR COMMENT So are you arguing that the this enormous channel, which nobody here likes, is actually not big enough? 

DAVID - Well, the point really is you're dealing with, as I've demonstrated, first of all, floods have a vast amount of water in them, and that's why they were not looking at upstream storage because the amount of water is too great. However, the best way to deal with a situation where you can't store water upstream is to get it through the flood risk area as efficiently as possible and draw the water levels down. And I have to say internationally, this is just what's going on in the Netherlands in a programme called room for the river, where they're actually moving the great dikes on the river Rhine back to create more space to get water through. And that's what this is all about. 

But the difficulty in Oxford is that there are obstructions. And there's been a lot of debate about what you should do to help the flow through these obstructions. And as I say, I lost contact with a debate a long time ago. In my view the scheme has to allow something like 80 or 100 cubic metres per second down the channel. We're getting about 30 at the moment. That's just my opinion from the past. I'm sure there are lots of other opinions. But yes, that would be the way at the time.