Computers aren’t that clever when it comes to flooding

Emeritus Professor David Rhodes CBE FRS FREng

Emeritus Professor David Rhodes CBE FRS FREng

Professor David Rhodes told a packed house at the Kirklands Community Centre on Tuesday evening (more than 100 present) that Bradford MDC had given planning permission for housing developments on Derry Hill and Bingley Road on the basis of false information.

Lecture_4983_pc

The problem is that the computer model used by BMDC, and other local authorities, grossly underestimates the volume of water involved in the kind of weather events experienced in Menston recently. In the case of the January 2008 floods the volume of water was underestimated by an order of magnitude; in a later event the volume of water was about six times that predicted by the computer model. A member of the audience who seemed to have some knowledge of computer modelling seemed to confirm this, saying: “The computer model is inevitably wrong because it is working without sufficient data”.

WotWboxProfessor Rhodes said he had been able to estimate the approximate volume of water involved by estimating depth etc and speed of water coming down into Menston from video clips taken by members of the Menston Action Group. This made good sense from Prof Rhodes’s explanation.

100,000 cubic metres of water

Based on Prof Rhodes’s observation not only was diverting the water coming down the slopes into Menston impractical because of the volume and flow rates involved, but to store the water from the January 2008 event would have required a holding tank with a capacity of 100,000 cubic metres. Apart from cost, to have such a quantity stored above Menston would be extremely dangerous.

Prof Rhodes told the audience that the estimates used by flood risk authorities, like BMDC, were wrong because they accounted only for rain water falling on the ground, but the actual volume involved was far higher because of water coming out of the ground. In this situation, had the houses been built on the proposed sites in Menston, water would have been coming up under the houses causing a great deal of damage. It is no wonder that another expert, giving evidence to BMDC, when asked if he would buy a house on one of these sites, said absolutely not.

Victorian engineers had an answer

In earlier times, ground water had been pumped out into reservoirs. Victorian engineers did it to supply water to the former hospital at Highroyds; later a similar system was used to supply water to Menston residents. Both these systems were abandoned so now the ground water was available to flood Menston.

Prof Rhodes’s novel solution

Professor Rhodes said he had a solution. Basically this would involve horizontally boring deep in the ground where water accumulates and having permeable pipes there into which the water could drain, these flowing out into existing water courses which could handle the continuous flow involved. He suggested that this solution would cost only tens of thousands of pounds. What is more, he said that such a system of horizontal pipes could encircle Ilkley Moor to alleviate wider flooding problems, though this would cost tens of millions.

Advertisements

One response to “Computers aren’t that clever when it comes to flooding

  1. Andrew Fraser

    I attended the talk on 26/4/16. I am a Flood Defence Engineer of many years experience. I found the proposed use of `stripping back` the ground water base flow interesting. I feel it would be a niche application solution but could be effective at specific locations. It may be difficult to sell the idea and demonstrate it effectiveness if the boundary between` rapid runoff`situations and longer sustained floods are blurred and come into play. I feel that the rainfall actually falling in the periods leading to the flooding events described should be based on more local rainfall gauges than perhaps was used.
    There is misunderstanding with the flood risk maps issued by the EA. They give a good two dimensional outline of the flood risk area but have no relevance to flow depths or more importantly flow volumes which Prof Rhodes correctly eluded to. Hydraulic modelling can only take you so far. The more practical aspects have to be explored.

    Would the solution be used proactively for developers to push ahead on sites or retrospectively to resolve existing flooding problems? A good commercial opportunity or a mix up of poacher/gamekeeper roles?
    The Highroyds example should be a good test bed for the proposed method.
    I wish Prof Rhodes well with the idea.

    Andrew Fraser CEng MICE C.WEM MCIWEM

Comments are welcome on this or any post or page. They can be anonymous and are not edited unless obscene, illegal or personally abusive.

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