Archive
Thames Tideway Tunnel – The start
Thames Tideway Tunnel – The start
I have now started on Phase 3 with Arup and am working within the geotechnical team on the Thames Tideway Tunnel project. To avoid this blog getting too long I’ll present some of the background, my part in the plan and a brief discussion on sustainability.
Background. The London sewer network is predominantly a combined foul and surface water system. The sewers convey the waste to treatment works for processing. However, when it rains heavily the capacity of the sewers is often overwhelmed and the excess sewage is deposited into the Thames through combined sewer outfalls. To reduce the frequency of incidents of raw sewage discharge into the Thames each year the Thames Tideway Tunnel project aims to capture combined storm and foul water within a new 7.2m diameter tunnel running from Acton in the west to Abbey Mills in the east.
Overview of tunnel alignment through the London geology.
My part in the plan. Effectively the project is divided into three sections: west, central and east. Each section has a number of connections to capture the existing combined sewer outfalls and convey the flow into the new sewer. The process is generally to have an interceptor chamber which feeds into a further set of chambers to attenuate flow and then a drop shaft to connect to the new sewer. I have 2 sites to design the foundations for: Putney and Barn Elms. Arup are not responsible for the main sewer tunnel or drop shafts but are responsible for designing the other elements, including the chamber and connecting culvert foundations.
View of Putney Foreshore Embankment site
Putney has some interesting challenges: The connection to the existing sewer outfall is directly below Putney bridge. One of the proposals sees there needing to be a number of bored piles in the foundation design (questionable reasons: either as tension piles to avoid buoyancy issues or to minimise long term settlement between existing bridge outfall and new structure). Getting a piling rig into position will be a challenge due to headroom constraints. One method sees a temporary sheet pile wall planned to be installed around the whole site area (under the centre of one of Putney bridge arches (max headroom)), then dewatered, excavated and re-filled. The temporary works are out of my scope but understanding how the contractor will carry out the works is clearly important.
Existing outfalls (left), proposed interception chamber on right – this is 7m deep with a number of bored piles within the foundations proposed!

My responsibility is designing the interception chamber (left green), connecting tunnel (centre purple) and attenuation chambers (right brown/green/blue) foundations. Drop shaft and connecting tunnel to main sewer (out of picture) is not my responsibility.
Illustrative view of completed site.
So in terms of what I am doing – initial work is scoping further ground investigation for where there is risk – lack of confidence in certain parameters for example. Next comes a geotechnical design report and then will come developed design, detailed design and finally construction issue information. This aligns pretty well with the RIBA stages of work too.
Sustainability.
The aim of the project is to provide overflow capacity to the existing sewer infrastructure in order to reduce the frequency of sewage discharges into the river Thames.
Advantages and Disadvantages. This project will bring political and social benefits of a cleaner river and evidence of better waste management practices within the UK. The economic benefit is more questionable. Construction costs have multiplied by a factor of 3 from 2005 to £4bn raising the question of its cost worthiness. Moreover, other options of attempting to use land as attenuation (green field sites as soakaways, green roofs) through green infrastructure have been argued to be sufficient to meet the demands of future predicted storm levels. The legislation is in place to enforce greener infrastructure but it appears planning authorities do not follow this through as judicially as they might with greater resources.
The accuracy of the sewage over flow statistics has been questioned, with some groups stating that the level of pollution entering the Thames is lower than the Environmental Agency’s limits now and so the construction of the new sewer is a complete waste of money.
As part of the project, the Beckton Sewerage Treatment Works will also be upgraded to cope with additional demand. The disadvantage of increasing the amount of surface and foul water captured will be the need to pump all of the water through 60m of head at Abbey Mills which will result in higher maintenance costs and require significant energy to achieve. If some of the storm water had been captured through green infrastructure assets there would be less energy required to process it at the other end. Reducing the amount of storm water ever reaching the sewer therefore brings a double saving.
Clearly reducing the sewage discharge into the Thames is beneficial and the project will significantly improve the condition of the river. I suspect as a country likely to be at the forefront of developing environmentally friendly solutions the green infrastructure piece will come but for now at least a solution has been produced to an acceptable cost, timeframe and quality, assuming all goes to plan!
In summary. This project will be my sole focus. I have started to investigate wider strands within it – financial/commercial for a better background understanding but with a company like Arup, unless specifically requested, phase 3 will tend to be focussed on one major project.