Dr David Milledge and Prof Stuart Lane present the results of a recent study they authored on flood prevention in the Yorkshire Dales located in Northern England that investigates the role of controlling surface drains (grips) to manage flood risk.
The moorlands of large swathes of Upland Britain are covered by hundreds of kilometres of surface drains or ‘grips’. These grips were originally dug between the 1950s and the 1970s to dry out the peat soil with the expectation that this would improve the vegetation for grazing and game. Grips have dried out the peat, but ‘gripping’ has since been associated with a range of negative impacts both in the moors and in the rivers that flow from them. Drying the peat both changes the ecosystem that it sustains and makes it more susceptible to erosion. This eroded peat is then transported downstream and needs to be removed from the water before we can drink it.
Whereas some grip impacts have been clearly established, others, such as their impact on downstream flood risk, remain the subject of heated debate: in theory, the drains allow water to flow into rivers more quickly which should increase flood risk; but they also tend to dry the soil which may increase the capacity of the soil to store water at the start of a flood event.
Working with the UK Environment Agency and Natural England we developed a mathematical model to look at these two competing effects. We found that the effect of the drains in delivering the water more quickly was much smaller than we expected. As a result, grips generally reduced rather than increased flood risk at our test site: Oughtershaw in the Yorkshire Dales. This is because whilst an individual grip may deliver water more rapidly during a flood event, if all grips do this, then the combined effect on the flow peak is to cause it to be earlier, but not necessarily bigger.
As with almost all models our predictions are somewhat uncertain. We deal with this problem by considering many different scenarios so that we can say something about the range of possible outcomes. In this case we ran the model 3000 times to generate an ensemble of predictions (black) in comparison with our observations (red), shown in the figure below for a 39 day period from 13th December 2004 to 20th January 2005.
By undertaking this with and without grips, we can assess when the flows are significantly changed even given model uncertainty.
We found that the flow without a drained catchment grips blocked was significantly higher than that with a drained catchment suggesting that grips generally reduce rather than increase flood risk at our study site. This might seem a disappointing result for those that want to make a case for restoring our moorlands by blocking grips. It seems that grips do what they were designed to do — drain the soil — and that this can reduce flood risk but also can cause a range of unforeseen problems for water quality, carbon storage and wildlife habitats.
Returning the peat to its wet condition seems to be very important in addressing the problems of peat erosion — ecosystem change and water quality — but may not solve our flooding problems. Perhaps the take-home message from this is that conservation will not always have benefits in every aspect and arguably it shouldn’t need to. But with the models we have developed it may be possible to reduce flood risk in certain places if blocking grips is done strategically. Accounting for a number of factors that affect river peaks and flows within a particular location can assist government and local communities in developing a useful strategy for mitigating flood risk. Comparing the results from the models in this study with studies based on measurement on the ground can assess the degree of certainty we can have in the model and its ability to help us understand how land management practices, such as grips, can induce or alleviate river flooding.
Lane, S., & Milledge, D. (2012). Impacts of upland open drains upon runoff generation: a numerical assessment of catchment-scale impacts. Hydrological Processes DOI: 10.1002/hyp.9285