Severn Tunnel alternatives/replacement options?

[Torfaen]

Moderator note: Split from

Severn Tunnel Closure Summer 2026

I noticed on GWR website that there'll be rail improvement work between Bristol to South Wales between 18-May and 15-June?! That is much longer than expected as I remembered only 2 weeks (or shorter) in recent years. I can't imagine they'll close it for 4 weeks! Maybe I am overreacting as it...

www.railforums.co.uk

Weren't there also some structural problems with the brickwork in the tunnel that caused it to be closed briefly about a year ago?

There must surely come a time when the existing Severn Tunnel is not viable to maintain any longer?

 

[ChilliSauce]

Weren't there also some structural problems with the brickwork in the tunnel that caused it to be closed briefly about a year ago? There must surely come a time when the existing Severn Tunnel is not viable to maintain any longer?

I think you are correct. The old seven tunnel has reached the end of its life ... Electrification in and now the maintenance of will compound the existing problems. Leading to longer and longer closures. And very large amounts of money having to be spent (wasted)

Rather than bit the bullet and build a new basically maintenance free tunnel.

Further I believe Single Line working is no longer allowed in the tunnel if the OLE is energised.

The cost of Electrification in the tunnel must of cost goodness knows how much, basically money wasted when bi-mode trains could of been deployed on the Paddington / Cardiff services using diesel traction in the tunnel.

 

[Torfaen]

The cost of Electrification in the tunnel must of cost goodness knows how much, basically money wasted when bi-mode trains could of been deployed on the Paddington / Cardiff services using diesel traction in the tunnel.

Then 387s wouldn't be able to run to Cardiff. It's the only GWR route where they can deputise for IETs.

 

[zwk500]

Rather than bit the bullet and build a new basically maintenance free tunnel.

As an off-line replacement? Because I don't think anybody could design a maintenance free tunnel by retrofitting into a 100+ year-old tunnel that runs directly through a major spring.

 

[The exile]

A causeway combined with a tidal power station? I’m sure there would be loads of environmental objections - not to mention economic ones. The one I don’t get is how harnessing tidal flows affects the littoral birds (other than where there is actual loss to structures) - the water is still going to rise and fall in the same way. Using barrages to hold back water to use at times of peak demand (like pump storage) would of course.

 

[zwk500]

A causeway combined with a tidal power station? I’m sure there would be loads of environmental objections - not to mention economic ones.

A causeway would be a horrifically damaging way to get tidal power. the Tidal Lagoons are a much more better proposal from that front (and they can act as pumped storage to some extent). If the railway does go over the river it'd want to have at least reasonable sections with through flow to avoid disrupting the sediment flow too badly.

The one I don’t get is how harnessing tidal flows affects the littoral birds (other than where there is actual loss to structures) - the water is still going to rise and fall in the same way.

Building a giant wall will stop the sediment flowing downstream, and the littoral birds eat the things in the sediment. If the sediment is no longer being replaced, it will erode very quickly. A quick look at Google maps will give an idea of just how much sediment the Severn Estuary receives from it's rivers.

 

[The exile]

A causeway would be a horrifically damaging way to get tidal power. the Tidal Lagoons are a much more better proposal from that front (and they can act as pumped storage to some extent). If the railway does go over the river it'd want to have at least reasonable sections with through flow to avoid disrupting the sediment flow too badly.

Building a giant wall will stop the sediment flowing downstream, and the littoral birds eat the things in the sediment. If the sediment is no longer being replaced, it will erode very quickly. A quick look at Google maps will give an idea of just how much sediment the Severn Estuary receives from it's rivers.

Get it - I’ve never heard it explained.

 

[ChilliSauce]

As an off-line replacement? Because I don't think anybody could design a maintenance free tunnel by retrofitting into a 100+ year-old tunnel that runs directly through a major spring.

That is why you choose a new alignment !


== Doublepost prevention - post automatically merged: 1 Apr 2026 ==

Then 387s wouldn't be able to run to Cardiff. It's the only GWR route where they can deputise for IETs.

If so much money hadn't been spent / wasted on electrification in the tunnel then they might of been money "left in the pot" to have completed electrification to Bristol T.M. by both routes . Then on those routes the 387s could of deputised for the IETs.

However, there is no hiding the fact that the GWEP Great Western Electrification Program was basically bankrupt, £millions over spent and years late.

The sad thing is the financial failure of that project ...by Great Western Projects has frightened governments away from committing to further main line electrification.

 

[Sealink]

A causeway combined with a tidal power station? I’m sure there would be loads of environmental objections - not to mention economic ones. The one I don’t get is how harnessing tidal flows affects the littoral birds (other than where there is actual loss to structures) - the water is still going to rise and fall in the same way. Using barrages to hold back water to use at times of peak demand (like pump storage) would of course.

Not sure if you were referring to this, but I know there is already something tidal power station on the Severn.
Whether it is still in use, I don't know.

 

[zwk500]

Not sure if you were referring to this, but I know there is already something tidal power station on the Severn.
Whether it is still in use, I don't know.

View attachment 201447View attachment 201448

I suspect that the structure there is for retaining sufficient cooling water for Oldbury Nuclear Power Station, visible in both images right on the coast. The wall being in the intertidal height means there's no pumping to manage, it just gets topped up twice a day!

A full Tidal barrage would dam the entire estuary (and has been proposed before, in a couple of variations). https://en.wikipedia.org/wiki/Severn_Barrage

The proposals for the tidal lagoons usually suggest something closer to the structure above, but with taller walls and turbines to generate electricity using the pressure differential caused by the tidal range.
This website examines some of the claims about tidal power and claims it isn't workable, but does explain the theory behind them quite well: https://euanmearns.com/swansea-bay-tidal-lagoon-and-baseload-tidal-generation-in-the-uk/

There are two generating opportunities on each tide. We begin with the lagoon empty and the flood tide coming in and the sluices closed (top panel). The lagoon wall holds back the sea. About 3 hours before high tide the sluices are opened through the generating sets and the basin is flooded and generation continues for about 3 hours until high tide is reached (middle panel). The sluices are then closed as the tide turns and ebbs and the lagoon wall holds the sea back. About three hours before low tide the sluices are once again opened for about three hours generation (lower panel). Hence there are two generating opportunities on each tide, 4 opportunities each day.

 

[Pigeon]

A causeway would be a horrifically damaging way to get tidal power. the Tidal Lagoons are a much more better proposal from that front (and they can act as pumped storage to some extent). If the railway does go over the river it'd want to have at least reasonable sections with through flow to avoid disrupting the sediment flow too badly.

A bridge. (Just so long as they don't let whoever designed that abominable motorway bridge anywhere near it.) Which is what they should have done in the first place, at the latest when they found they were never going to be able to turn the pumps off, if not in preference to restarting the derelict works. After all the Forth bridge was finished only 2 years after the tunnel eventually opened, and you don't need all that much more of a maximum span, so I don't think it would have been too much of a stretch.

 

[RichardAsh1981]

There was of course a Severn rail bridge, until a barge hit it in a fog and knocked a section down. Plus there was quite a tight weight limit on it, which isn't a problem for the tunnel.

 

[xdbjacx]

There was indeed a bridge at Sharpness, although had it survived the fog that promoted its demise, it would not be especially useful today. The only realistic option, and one that is becoming increasingly urgent, is to bore a new tunnel, or a pair of tunnels under the Severn. The existing arrangement was never entirely fit for purpose and has certainly now reached the end of its useful life, and should be abandoned.

 

[gingertom]

isn't it the case that a new tunnel will also encounter the same underground river?

 

[Class 170101]

isn't it the case that a new tunnel will also encounter the same underground river?

Unlikely with modern radar and measuring methods and modern construction methods it would be easier to avoid.

 

[ChilliSauce]

isn't it the case that a new tunnel will also encounter the same underground river?

Suggest not if you was clever with geological surveys ....

 

[Smwrff]

Build it as an immersed tube tunnel. Problem goes away.

 

[zwk500]

Build it as an immersed tube tunnel. Problem goes away.

If the gradients and navigation depth permit it on the 2nd largest tidal range in the world (I think?)

 

[Pigeon]

If the gradients and navigation depth permit it on the 2nd largest tidal range in the world (I think?)

Indeed (not to mention the currents). It certainly wouldn't be any good trying that where it is at the moment, which is basically an underwater gorge. For it to work at all it would have to be on an entirely new route some distance downstream and therefore with a much longer underwater section.

 

[Smwrff]

If the gradients and navigation depth permit it on the 2nd largest tidal range in the world (I think?)

The point is that an immersed tube tunnel does not need to be as deep as a bored tunnel - although both need to achieve navigation depth. And the shallower the depth the less the approach lengths.

 

[TrainTraveler]

It's quite expensive to replace the Severn tunnel.

The length would be of the same order as the HS2 tunnels between Old Oak Common and Euston and they will be quite a few billion pounds.

Mending each year will take a fair while to be a similar cost.

 

[Snow1964]

The point is that an immersed tube tunnel does not need to be as deep as a bored tunnel - although both need to achieve navigation depth. And the shallower the depth the less the approach lengths.

Is it even possible to build an immersed tunnel in a trench on such a fast moving tidal river, the mud and silt moves with every tide, so any trench dug for a tunnel will probably fill itself in with every tide change

 

[DelW]

From what I remember of articles in the engineering press about construction of the (then) Second Severn Crossing, the alignment and bridge type was chosen because of the geology and channel profile at that point. There is shallow water either side, under which there's relatively hard rock strata extending a long way out on the southern side ("English Stones"). There's a narrow but deep channel called "The Shoots", with the main cable-stay span of the bridge located exactly over it.

That's also almost exactly over the current tunnel alignment, and wouldn't be suitable for an immersed tube tunnel, which needs a dredgable bed and gentle slopes. So a new tunnel of that type would need a new alignment and approaches some way upstream or downstream.

 

[AdamWW]

That's also almost exactly over the current tunnel alignment

Indeed and if I recall correctly there is a sort of ground level bridge (I'm sure there's a proper technical term for this) where it passes over the tunnel to reduce the loading on the ground above the tunnel.

 

[Pigeon]

It seems to be unreasonably hard to find chart/map images that show the figures for the depths legibly, but you can sort of read some of the second image, and the first one does at least show the range it covers. Zero is chart datum / LAT so all the depths greater than zero indicate areas which can be uncovered by the tide (range 14m). The second image indicates this more clearly by colouring them in green. The rise from the low water mark to the shore is about the same as the drop to the middle of the channel.

Apart from the English Stones and a few similar bits most of the bottom is soft, but you can't rely on any of that staying the same depth because the currents are continually shoving it about - hence "changing depths" marked on the second image.

 

[DelW]

Indeed and if I recall correctly there is a sort of ground level bridge (I'm sure there's a proper technical term for this) where it passes over the tunnel to reduce the loading on the ground above the tunnel.

From https://severnbridges.org/category/the-second-bridge/

To avoid any additional loading on the Severn Railway Tunnel, the lengths of all viaduct spans were adjusted to ensure that a longer standard length of span would be available at the point where the viaduct would cross over the railway tunnel. Also bored piles were used to support the caissons on either side of the tunnel because no change in the loading condition on the tunnel could be tolerated. Monitoring devices were fitted to the tunnel lining to check for any movement in the structure of the tunnel. Where movement did occur, it was found to be related to the very different loads imposed by high and low tide, rather than by the new crossing.

The tunnel alignment is a bit further from the bridge main span than I thought, but still not far away.

Apart from the English Stones and a few similar bits most of the bottom is soft, but you can't rely on any of that staying the same depth because the currents are continually shoving it about - hence "changing depths" marked on the second image.

From the same source as above, much of the rock is strong enough to carry pier loads directly:

All the foundations were built using pre-cast concrete caissons, which are very large boxes without tops or bottoms. Most caissons were seated directly on the rock of the English Stones but some were on piled foundations.

 

[Pigeon]

From the same source as above, much of the rock is strong enough to carry pier loads directly:

Yes, where there is rock it tends to be pretty good, but most of the bottom is not rock but soft stuff which keeps moving about so you couldn't rely on it staying put if you dredged an immersed-tube trench into it.

 

[HSTEd]

I can't help but think a new bridge (as discussed in our other recent thread on the Severn railway arrangements!) would be better than a tunnel.

Road-rail tunnels bridges are very capable these days, with spans of 1100m or more being routine in the far east.

 

[DelW]

Road-rail tunnels are very capable these days, with spans of 1100m or more being routine in the far east.

Bridges I think

 

[HSTEd]

Bridges I think

Yes, I just independently noticed that! I will fix it