You might have heard recently that Network Rail has been denied planning permission to demolish and construct a new overbridge at Steventon in Oxfordshire. The village of Steventon is approximately 4.5 miles west of Didcot Station on the Great Western main line (GWML), which cuts right through the village. The Grade II listed Steventon Bridge, built by Brunel, is an elliptical three-arch brick overbridge which carries the B4017 over the GWML, linking the two halves of the village. Approximately 440 and 800 metres west of the bridge are two level crossings, named Stocks Lane and The Causeway.
Network Rail has identified 29 bridges in Oxfordshire, of which 19 need to be modified and 10 demolished, in order to comply with Overhead Line Electrification (OLE) standards as part of the £2.8 billion electrification of the GWML.
Back in 2012, Steventon Bridge was one of the bridges Network Rail identified for demolition and rebuilding as the bridge itself does not have adequate clearance for OLE. Also, the two adjacent level crossings are positioned such that the 5.8-metre minimum OLE wire safety clearance required for level crossings could not be achieved given their close proximity to the bridge.
Concern about collective impact
To ensure compliance with OLE standards, Network Rail calculated that a permanent speed restriction (PSR) of 60 mph on both lines would be necessary. This would introduce a delay of 90 seconds per train, which equates to 9.4km distance travelled. Network Rail expects 128 trains to run in each direction under the bridge, when the route is fully electrified. They also expect the collective impact of the PSR to be potentially significant.
Following further preparatory work, Network Rail officially submitted proposals for the demolition of the Grade II bridge to Historic England and The Vale of White Horse District Council four years ago.
What Network Rail hadn’t expected was the significant resistance from the local community in Steventon village to the proposed closure of the B4107 for 10 months.
By November 2017, Network Rail was expressing frustration with the continuing delay in obtaining approval to demolish the Brunel bridge – the original application had been submitted in April 2014.
A letter to the Vale of White Horse Council from Network Rail’s route sponsorship director, Simon Maple, stated that they were concerned that the application had already been deferred six times, emphasising that Network Rail had employed significant resources into considering alternatives, none of which were viable.
The letter also highlighted that, in response to heritage concerns, Network Rail had revised plans for the replacement bridge to include a brick finish rather than change the design. However, these efforts did not alter the fact that the work would cause a 10-month road closure, thus cutting off a main route into the village.
In accordance with planning application rules, Network Rail was invited to the council’s planning committee meeting on 1 August but was not allowed to discuss the options and impact of the bridge reconstruction. The Network Rail team was frequently reminded that the committee was only being asked to consider, on balance, whether the bridge/heritage loss is or is not outweighed by the public benefit at a national level for an infrastructure project, making a decision to accept or reject.
Meanwhile, objectors were challenging the demolition on the grounds of its economic impacts on the village, which actually had no bearing on the planning committee discussion at this stage in the process. The 10-month road closure was considered unacceptable by the local community while Network Rail considered it to be the only way to introduce OLE to the route.
In hindsight, perhaps Network Rail was being too cautious about the timescales for reconstruction but, equally, it does also appear to be a shortcoming in the planning process that this integral aspect of the project was not considered to be appropriate for discussion, even though it was uppermost in the local communities’ concerns.
Support from Council officials
Initially, Historic England raised doubts about the application in light of the parish council’s concerns, saying it raised “serious questions” about the necessity of demolishing the Victorian bridge. However, a final report from Historic England’s buildings area inspector, Richard Peats, explained that, following clarification from Network Rail, all other potential options were either impossible, would pose a risk to the running of the railway, or would involve even lengthier closures to the other routes into Steventon.
As a consequence, both Historic England and the district council’s conservation officer supported Network Rail’s latest plans. Earlier this year, at a meeting of the Vale of White Horse Council’s planning committee, councillors recommended approval of the demolition of Steventon Bridge.
Strong local concern
However, at the meeting on 1 August, the chairman of Steventon Parish Council stated that the parish council had produced its own report which outlined other ways of resolving the problem. The report stated that the Network Rail proposal would have a ‘devastating’ impact on local business and that Network Rail had ignored other viable options including lowering the track or jacking up the bridge.
The chairman added that a lot of businesses would close as they relied on passing trade from that route into the village. The view was that the timescale should be significantly reduced and alternative methods to reconstruction should be sought. He added: “With regret we conclude that Network Rail have not made a clear enough case for the bridge to be demolished in order for electrification to be delivered on time.”
After considering this information, the district council planning committee decided that the disruption would be unacceptable and denied Network Rail planning permission to reconstruct the bridge – a decision that has made headline news.
Lowering the track
Network Rail standards state that, in order to create an acceptable gradient for trains, 100 metres of track either side of the bridge has to be re-laid for every 100mm of track lowering. At Steventon, this would require the works to extend to, and beyond, Stocks Lane level crossing located to the west of Steventon Bridge.
Among other discounted options, both a full track lower, and a partial track lower were discussed further within Network Rail’s submission documents. These two options took into account other factors such as the level crossings in the area, the necessary OLE wire gradient, and, of course, the flood risk.
To accommodate a full track lower and to achieve sufficient clearance at both crossings for OLE wires, the track lower at the bridge would need to be about 770mm. This would require approximately 800 metres of track and ballast to be renewed both sides of the bridge along with the lowering of both level crossings and would include lowering a four track section to the east of the bridge. In addition, an excavation of approximately 1,500 mm depth at the bridge would be needed to form the new track bed.
A dig of this depth would be likely to destabilise the toe of the cutting and so significant earth retaining structures would need to be installed around the bridge. This would probably be a reinforced concrete wall with ground anchors tying back into the cutting. The structural strengthening of the bridge to allow for the excavation works would be significant and require a long closure of the railway to achieve this.
A technical report written by CH2M Hill (now Jacobs) concluded that a significant lowering of the track level would introduce a very high risk to the railway of flood events. Considering this risk of flooding to the rail network, and to the local community, Network Rail could not consider this option as a viable solution.
Partial track lowering, based on site survey information, was then considered and Network Rail determined that the minimum depth calculated should be 444mm in order to provide both the minimum overhead line clearances and pantograph-to-bridge clearances. A track lower of 444mm would still incur many of the issues associated with a full track lower as well as other matters. The extent of track needing to be rebuilt would be a minimum of 444 metres either side of the bridge, and require a total depth of excavation of around 1,100mm to form the track bed.
Network Rail did consider reducing the 5.8-metre OLE wire clearance at the two level crossings and imposing a restriction on vehicle height but this option was discounted by the parish council as it would further restrict trade.
Issue 145 (November 2016) of Rail Engineer included a report on the successful jacking trial of a 220-tonne brick arch bridge using the ElevArch® process, developed by Freyssinet and Bill Harvey Associates. After preparatory work to strengthen the brick-built structure and consolidate it into one liftable mass, it is cut free from its abutments and raised to a new position using hydraulic jacks. Additional brick courses are then inserted to support the bridge in its new position. The advantage of this system is that the original listed brick structure is retained, albeit with a few additional courses, and the cost of demolishing and removing the entire bridge and manufacturing a new one is eliminated.
Level crossings are considered to be a safety hazard anyway, so a more radical option would be the possibility of completely removing both level crossings, thereby releasing the project from the constraints that the level crossings impose. Closure of the level crossings to traffic and retaining pedestrian access was also considered and proposed by Network Rail. but this was rejected by the council due to the impact this would have on local residents and businesses.
It has been suggested by the local council that the new Hitachi trains could still function when tracks are flooded as they are bimodal. Well, technically, yes this is an option but, as Paul Pryke, Hitachi’s deputy head of engineering in the UK, explained, and I quote: “Our Intercity Express Programme trains are designed to maintain full functionality during and after running through floodwater up to a depth of 100mm above rail level, although it’s important to note that these conditions may cause speed restrictions to be applied. We will work with route partners GWR and Network Rail if speed restrictions are necessary, and these speeds will be dependent on the specific circumstances at the time.”
Acknowledging that speed restrictions would probably be necessary in heavy flooding conditions as an acceptable aspect of the design is surely not an element that one would want to consider for a new high-speed route.
When pressed for more information on the possible implications of a speed restriction, a Hitachi spokesman checked and responded: “Speed restrictions may be required, depending on how vast the flood water is…their precise words were ‘this is a train not a ship!’. Always reassuring to hear that clarification from our engineering team.”
It doesn’t sound as though the train manufacturer views the prospect of its product routinely running through 100mm (four inches) of water as a serious proposition.
Also, one has to consider all the other problems that would emerge from allowing track to flood. There could be signalling problems, wet spots could form and ballast washed away. The arguments against allowing a design to include flooding must be very persuasive.
It is interesting to note that the bridge has undergone several phases of alteration in its history, including underpinning the foundations, the replacement of its copings, insertion of tie plates and re-facing in engineering brick. However, the most unsympathetic alteration was the insertion of concrete tie bracing to the side arches, work carried out in 1963. It is clearly not a bridge that is in good condition.
It appears that everyone is trying to do the right thing and it is a credit to our democratic process that all voices, including those of local people, are heard. It might be frustrating for engineers who are trying hard to deliver a very demanding and complex project. Nevertheless, such an approach, frustrating though it might be, ensures that engineers stretch their skills to the limit to find an acceptable solution and, in the long run, that has got to add value to the process.
In this particular case, it is hard to see an alternative to reconstruction, but perhaps there is scope to reduce the 10-month timeframe to accommodate some of the concerns that the local community has. This could well depend on the cooperation of the utilities, which would need to disconnect their buried services across the bridge and reinstate them afterwards – a process that can take a lot of time and might well have contributed to Network Rail’s unpopular ten month estimate.
As a colleague pointed out to me recently, with the help of Bailey Bridges, we used to build bridges one half at a time. It’s an opportunity for engineers to be ingenious!
Network Rail registered an appeal to the Planning Inspector in September 2018.
There are three possibilities:
- A full enquiry, which would take 12 to 18 months;
- A hearing, which would take approximately nine months;
- A written representation, with a response in three to six months.
Read more: Looking forward to InnoTrans 2018