Borders Railway Extension: suggestions on how this should progress

[47271]

Alex Hynes on a radio phone in https://twitter.com/i/status/1188471698636587009
He is now promising six car 170s in the peak once HST roll out is complete.

You'll be waiting a while for six cars then...

 

[NotATrainspott]

Given the over projection of Midlothian passenger numbers a railhead at Gorebridge may well have been a pretty bad decision. I think Midlothian passengers will significantly increase over the next decade as more housing, particularly Shawfair, is built and the roads in and around Edinburgh get even more congested.

I can’t speak re Leven and Renfrew but I am very surprised that re-opening Peterhead had a better business case then Tweedbank.

Lothian Buses is pretty strong in Midlothian so it's not that surprising that the local services are struggling slightly. The slow route into Waverley will do quite a bit to undo any speed savings from not being a bus.

Going up to 4tph to Gorebridge is already passively provided for on the line as built (e.g. how Newtongrange station is built to have a second platform added when most others were centered on the trackbed). Couple that to the planned junction and signalling improvements around the east of Edinburgh and it'll make the train a much more desirable option for commuters into the city centre.

Alex Hynes on a radio phone in https://twitter.com/i/status/1188471698636587009
He says passenger numbers are growing faster than they can add more carriages. He is now promising six car 170s in the peak once HST roll out is complete.
Six cars are the maximum that most of the stations can handle so if they prove inadequate some sort of infrastructure improvements will be necessary.

Six car at 2tph in the peak is pretty good going for any commuter route.

 

[cle]

That's all pretty impressive, very cool to see. How is off peak, weekend and counter/tourism use? As it sounds very Edinburgh commuter in nature.

 

[railjock]

That's all pretty impressive, very cool to see. How is off peak, weekend and counter/tourism use? As it sounds very Edinburgh commuter in nature.

It is very Edinburgh focussed both for commuter and leisure.

 

[railjock]

Lothian Buses is pretty strong in Midlothian so it's not that surprising that the local services are struggling slightly. The slow route into Waverley will do quite a bit to undo any speed savings from not being a bus.

Going up to 4tph to Gorebridge is already passively provided for on the line as built (e.g. how Newtongrange station is built to have a second platform added when most others were centered on the trackbed). Couple that to the planned junction and signalling improvements around the east of Edinburgh and it'll make the train a much more desirable option for commuters into the city centre.



Six car at 2tph in the peak is pretty good going for any commuter route.

I read somewhere that turning services back at Gorebridge would be an issue.

 

[Paul Kerr]

Good question. There’s also the not insignificant issue of getting out of Tweedbank.

Joining this discussion very late, but I don't think getting out of Tweedbank will be that difficult if Network Rail agrees to a level crossing for the station access or brings the access road in from a different point. The line of the route east of Tweedbank is not taken up by the Melrose bypass until immediately east of the old station building at Melrose, and even then only for 2 short sections which could be addressed by building the line alongside the road. The more problematic issues are crossing the A68/A6091 and then figuring out how to get the line through Newtown St. Boswells.

South of the old Kelso Junction the trackbed is largely unobstructed and most of the major earthworks and structures are intact. I know reinstatement will be expensive but 3-400 million to reinstate 17 miles as a single track with 1 or 2 short dynamic loops on a double track width roadbed seems wildly pessimistic to me, especially when you consider the obstacles the Borders line deviated around at Millerhill and Galashiels (about 2.5-3 miles had to be built on new alignments as part of the reopening project). If you take the Tweedbank line costs as a benchmark (about GBP 10M per mile), you're looking at a cost of GBP 170M. That seems to be a better ballpark estimate in my mind.

According to the book David Spaven wrote on the Borders Railway, Gala to Hawick was actually the busiest section of the original line in its heyday. I think the same could be true if the Hawick section reopens. The settlements on the line are not that large but there is a significant hinterland around the route which could be tapped into if the line reopens and railheads are established at St Boswells and Hawick (e.g. Kelso, Jedburgh, Earlston).

 

[Paul Kerr]

As an aside passenger numbers have not far exceeded expectations. They are just about on target. It is true that there are more coming from Tweedbank / Gala than expected, but this is offset by fewer coming from the Lothian stations.

Actually what you state there is not true. The business case to reopen the line assumed 450,000 (+/-) passengers per year. Actual passenger numbers are close to 2.5 times that number. It is true that Alex Salmond who was FM at the time set a target for the line to reach 1 million passengers; that was achieved and exceeded in the 1st 12 months after the line opened.

 

[Paul Kerr]

I don't think the case for keeping the line was particularly brilliant, given the nature of the area and traffic patterns at the time, but there was probably a good case for a singled line to Gala as a commuter route, which could have been operated quite cheaply by DMUs.

But yes, you make a very good point there that a lot of people don't seem to grasp. It's all well and good railing against a decision made fifty years ago, but if it was a mistake, it's a very expensive one to correct, and I doubt it's worth it.

It is true that the official case to retain the northern section was not good, but interestingly, the transport minister who approved closure (Richard Marsh) admitted afterwards off the record that closing the line north of Hawick was a mistake. David Spaven and his team ran an analysis of the costs and subsidies BR's accountants used as justification to close the route and he estimated that the real costs of a DMU operated basic railway from Edinburgh to Hawick would be about half of what BR and Marsh's advisers at the DoT estimated. I suppose we need to set that against the backdrop of a very anti-railway attitude at that time, and even BR managers deemed closing routes as a promotion opportunity. Hindsight is 20-20, I suppose.

I think extending to Hawick has a case which merits investigation at least; it won't be cheap, I agree, but it won't be as expensive as some of the figures I've seen quoted on this thread. I do agree with your comments on extending south of Hawick though; it makes little strategic or economic sense given that it will never be competitive as a through route unless they built the line on a straighter alignment with easier gradients, shadowing the A7 through Teviothead and Langholm. Doing that would come at a huge expense...

 

[Bald Rick]

I know reinstatement will be expensive but 3-400 million to reinstate 17 miles as a single track with 1 or 2 short dynamic loops on a double track width roadbed seems wildly pessimistic to me, especially when you consider the obstacles the Borders line deviated around at Millerhill and Galashiels (about 2.5-3 miles had to be built on new alignments as part of the reopening project). If you take the Tweedbank line costs as a benchmark (about GBP 10M per mile), you're looking at a cost of GBP 170M. That seems to be a better ballpark estimate in my mind.

As you’ve come to this late, you might want to read this and other threads about the cost of new railways.

The Borders line wasn’t £10m/mile, it was about £15m/mile when all costs were accounted for. And that was costs incurred between 5 and 10 years ago. It was also very cheap.

The going rate for future reinstatement (ie costs being incurred in the near future) is around £30m/mile. Worth noting that the East West Rail Western section is above that, both for the bit already done (Oxford - Bicester) and the section under construction now (Bicester / Aylesbury to Bletchley).

 

[tbtc]

Actually what you state there is not true. The business case to reopen the line assumed 450,000 (+/-) passengers per year

...but the line wasn't built because of an outstanding business case - it was built at the price of pork-barrel politics at Holyrood.

Maybe, if we were assessing several Scottish projects as part of a "beauty parade" we'd have carried out more rigorous assessment into the predicted passenger numbers (but you're not going to waste much time modelling hypothetical numbers when the Labour Party have agreed to build it to entire the LibDems into coalition.

Same deal at Alloa (re-opened as the price of pacifying locals there who may have been upset at the increased numbers of freight trains - i.e. re-opening the line was a small price to pay to free up paths over the Forth Bridge by diverting them through Stirling.

That said, some projects are bound to do better than predicted passenger numbers, some fail to meet them (even in a rising tide, where general passenger numbers across the network may be 10% or 20% higher than when a project was first mooted) - people like to focus on the outliers rather than look at the trends.

 

[Tobbes]

Where in Hawick would you build the station?

 

[Journeyman]

Where in Hawick would you build the station?

I think that very much depends on whether extending further is a realistic prospect.

 

[Tobbes]

I think that very much depends on whether extending further is a realistic prospect.

Well, quite. Presumably the case for going further south is made (much) easier if the Hawick station is on a viaduct next to Morrisons' petrol station, with the bridges paid for from this scheme. More likely, the carpark east of the leisure centre.

 

[Paul Kerr]

As you’ve come to this late, you might want to read this and other threads about the cost of new railways.

The Borders line wasn’t £10m/mile, it was about £15m/mile when all costs were accounted for. And that was costs incurred between 5 and 10 years ago. It was also very cheap.

The going rate for future reinstatement (ie costs being incurred in the near future) is around £30m/mile. Worth noting that the East West Rail Western section is above that, both for the bit already done (Oxford - Bicester) and the section under construction now (Bicester / Aylesbury to Bletchley).

Thanks for the info on the East-West route; I’ll do some further digging on that. However, unless I’m missing something fundamental, I don’t think we’re looking at an apples to apples comparison here. The line to Hawick would essentially be the same “rail-light” spec used for the Borders Railway:

1. RA3 multiple unit spec
2. Single track with at most 2 dynamic loops (one in the St Boswells area and potentially another one closer to Hawick if pathing dictates)
3. Basic station facilities
4. 2 new stations only (St Boswells and Hawick); there were only 2 intermediate stops in between on the Waverley Route and I doubt either would reopen.

The East West route will be a full double track route, correct? It also requires integration with multiple main line routes on its length, so the cost and complexity will inevitably be much higher. On the abandoned section east of Bedford there are multiple obstacles that need to be dealt with, and the intersection at Bedford is going to add huge cost. For the Hawick extension, the only existing infrastructure integration that needs to be dealt with is the current terminus at Tweedbank, and at least 15 of the 17 miles east and south is largely unobstructed. With the basic railway spec, £30M per mile seems excessive.

On the existing Borders Railway Cost, construction cost was £294M. For 30 miles (+/-) of new line that works out at £9.8M per mile. There were additional costs up front of £59M for land purchase and preparatory works (moving utilities in the Galashiels area and remediating subsidence in Midlothian, for example), but much of that up front cost was eaten up by the abortive DBFO model that Transport Scotland was pursuing before the tender was let to Network Rail. Letting that slide, your total cost was £353M or £11.8M per mile. The section south to Hawick does not have to deal with old mine workings; there will inevitably be some challenges but this section does not have many of the complex issues the Borders Railway project had to deal with.

I recognize that costs have gone up since this line was built, but a jump from £11.8M to £30M per mile in a few years seems excessive. Nevertheless, thanks for the info and I’ll investigate further.

 

[Paul Kerr]

Well, quite. Presumably the case for going further south is made (much) easier if the Hawick station is on a viaduct next to Morrisons' petrol station, with the bridges paid for from this scheme. More likely, the carpark east of the leisure centre.

I honestly don’t think extending south of Hawick is a realistic prospect as has been pointed out, as the line will never really be a viable alternative to Carstairs as a through route to the south. Assuming Hawick is the terminus, the original terminus site is practically untouched and the old platform still exists; that seems the most logical spot for a new station to me.

 

[swt_passenger]

...The East West route will be a full double track route, correct? It also requires integration with multiple main line routes on its length, so the cost and complexity will inevitably be much higher. On the abandoned section east of Bedford there are multiple obstacles that need to be dealt with, and the intersection at Bedford is going to add huge cost.

I think when the “costs of EWR” are being discussed, it’s purely in the context of Bicester to Bletchley (Western section phase 2), because that’s all the recently approved TWAO covers. But let’s not exaggerate the difficulties of integration with the existing railways at either end of this section, because the track layout already exists. For example the double junction at Bicester is already fully in place, provided during the Chiltern to Oxford project, as are the existing junctions with the WCML and Marston Vale routes. They’ll presumably need renewals at the Bletchley end but it’s not actually being built from scratch.

But east of Bedford will effectively be a brand new railway, its costs will inevitably be far higher than the Bicester to Bletchley.

 

[Bald Rick]

Thanks for the info on the East-West route; I’ll do some further digging on that. However, unless I’m missing something fundamental, I don’t think we’re looking at an apples to apples comparison here. The line to Hawick would essentially be the same “rail-light” spec used for the Borders Railway:

1. RA3 multiple unit spec
2. Single track with at most 2 dynamic loops (one in the St Boswells area and potentially another one closer to Hawick if pathing dictates)
3. Basic station facilities
4. 2 new stations only (St Boswells and Hawick); there were only 2 intermediate stops in between on the Waverley Route and I doubt either would reopen.

The East West route will be a full double track route, correct? It also requires integration with multiple main line routes on its length, so the cost and complexity will inevitably be much higher. On the abandoned section east of Bedford there are multiple obstacles that need to be dealt with, and the intersection at Bedford is going to add huge cost. For the Hawick extension, the only existing infrastructure integration that needs to be dealt with is the current terminus at Tweedbank, and at least 15 of the 17 miles east and south is largely unobstructed. With the basic railway spec, £30M per mile seems excessive.

On the existing Borders Railway Cost, construction cost was £294M. For 30 miles (+/-) of new line that works out at £9.8M per mile. There were additional costs up front of £59M for land purchase and preparatory works (moving utilities in the Galashiels area and remediating subsidence in Midlothian, for example), but much of that up front cost was eaten up by the abortive DBFO model that Transport Scotland was pursuing before the tender was let to Network Rail. Letting that slide, your total cost was £353M or £11.8M per mile. The section south to Hawick does not have to deal with old mine workings; there will inevitably be some challenges but this section does not have many of the complex issues the Borders Railway project had to deal with.

I recognize that costs have gone up since this line was built, but a jump from £11.8M to £30M per mile in a few years seems excessive. Nevertheless, thanks for the info and I’ll investigate further.

I can tell you that the final actual construction cost for Borders was rather higher than £294m. There’s also a whole load of costs not accounted for in the construction or ‘prep’ costs, not least all the work that went into the parliamentary bill.

Yes, the scope vs East West Rail is different, but not that much. Whilst it surprises many, the difference in cost of a new double track railway is only about 20-40% more than a single track - much of the activity is the same. The costs I was referring to for EWR are for the Western section only, ie Oxford - Bicester, and Bicester / Aylesbury - Bletchley (the latter part of which is going to be single track). Yes there is an interface at each end, although the one at the Bicester end is already built, and at the Bletchley end it only needs a relatively simple change. As new double track railways go, it is about as simple as it can be. Its 28 miles, of which a third is single track, and the cost is well over £30m/mile. Note that the land for the line is already in railway ownership, although some land needs to be taken permanently for new bridges and the one new station at Winslow.

 

[385001]

With the decarbonisation strategy coming soon would the Scottish Government be keen for an expensive non-electrified extension of a (mostly) non-electrified line?

 

[Bald Rick]

With the decarbonisation strategy coming soon would the Scottish Government be keen for an expensive non-electrified extension of a (mostly) non-electrified line?

Expensive and carbon intensive (to build) extension...

To meet Scottish Govt Carbon objectives, it may well need to demonstrate net carbon neutrality over the appraisal period, which I believe it would struggle to do even if it was electrified.

 

[Paul Kerr]

I can tell you that the final actual construction cost for Borders was rather higher than £294m. There’s also a whole load of costs not accounted for in the construction or ‘prep’ costs, not least all the work that went into the parliamentary bill.

Yes, the scope vs East West Rail is different, but not that much. Whilst it surprises many, the difference in cost of a new double track railway is only about 20-40% more than a single track - much of the activity is the same. The costs I was referring to for EWR are for the Western section only, ie Oxford - Bicester, and Bicester / Aylesbury - Bletchley (the latter part of which is going to be single track). Yes there is an interface at each end, although the one at the Bicester end is already built, and at the Bletchley end it only needs a relatively simple change. As new double track railways go, it is about as simple as it can be. Its 28 miles, of which a third is single track, and the cost is well over £30m/mile. Note that the land for the line is already in railway ownership, although some land needs to be taken permanently for new bridges and the one new station at Winslow.

Thanks for the additional info and clarifying the spec on the EWR. What is the proposed RA spec for the new line? Is it being built to carry freight as well or will it be a "railway-light" multiple unit only spec as was done on the Borders line? From what I understand, the Borders project realized major savings by building to RA3.

You mention the final cost for Borders was higher than the publicly stated figure of £294m. What was the final construction cost (+/-)? Everything I can find in the public domain states the cost to build was £294m, with prep costs ranging between £53m and £59m.

 

[paul1609]

I have friends at St Boswells they are already keen users of the borders railway. If the railway was extended I think they would just use a closer railhead. i get the impression that would be the case for much of the borders. The real attraction of rail is the ease of getting in to Edinburgh itself. Not sure about Hawick itself but i reckon that any increase in passengers would be very marginal. They would of course be doing more miles by rail.

 

[Hagbard Celine]

Joining this discussion very late, but I don't think getting out of Tweedbank will be that difficult if Network Rail agrees to a level crossing for the station access or brings the access road in from a different point.

There is a business park currently under construction to the east of Tweedbank station. The access road has passive provision for a bridge over any railway extension which would also become the new station access.

 

[Bald Rick]

Thanks for the additional info and clarifying the spec on the EWR. What is the proposed RA spec for the new line? Is it being built to carry freight as well or will it be a "railway-light" multiple unit only spec as was done on the Borders line? From what I understand, the Borders project realized major savings by building to RA3.

You mention the final cost for Borders was higher than the publicly stated figure of £294m. What was the final construction cost (+/-)? Everything I can find in the public domain states the cost to build was £294m, with prep costs ranging between £53m and £59m.

Ah, not being in the public domain, I cannae say.

I don’t think the savings of RA10 (or RA8) to RA3 are that big. It’s basically for underbridges. Everything else is the same spec.

 

[Tobbes]

Ah, not being in the public domain, I cannae say.

I don’t think the savings of RA10 (or RA8) to RA3 are that big. It’s basically for underbridges. Everything else is the same spec.

Thanks @Bald Rick , that seems sensible. Are all of the strucutres built to W10 / electrification clearances?

 

[Paul Kerr]

Thanks @Bald Rick , that seems sensible. Are all of the strucutres built to W10 / electrification clearances?

Yes, all structures are built to electrification clearances. I believe the gauge used is W12 but I could be mistaken.

 

[Paul Kerr]

Ah, not being in the public domain, I cannae say.

I don’t think the savings of RA10 (or RA8) to RA3 are that big. It’s basically for underbridges. Everything else is the same spec.

Does it not include the trackbed as well? The Borders line has a ruling speed limit of 55mph for locomotives with speed limits of 20mph over most of the refurbished wrought iron girder bridges over the Gala Water; multiple units can theoretically do 75 to 90mph on most sections of the line.

 

[433N]

Whenever the Hawick extension and the sums involved are bandied around, I feel compelled to put the case for prioritising a case for public transport improvement to the south west of Edinburgh which suffers chronic congestion (a few of the top ten hotspots for traffic congestion in the UK) and is always ignored - yet more house building is continuing apace. The Borders line did something to improve the chronic congestion at the Sheriffhall roundabout which lies on the bypass which connects the A1 to the north of Scotland as well as coping with local commuting demand into and around Edinburgh but little for points to the west.

As someone who has to cope with the nightmare of south west Edinburgh (which has no rail) at rush hour, extension of the Borders Railway to Hawick falls into vanity project territory - which, of course, is not to say that it won't go ahead.

 

[Bald Rick]

Does it not include the trackbed as well? The Borders line has a ruling speed limit of 55mph for locomotives with speed limits of 20mph over most of the refurbished wrought iron girder bridges over the Gala Water; multiple units can theoretically do 75 to 90mph on most sections of the line.

I don’t know about the trackbed, but the difference in spec would be minimal - perhaps an extra 50mm of ballast and an extra 2 sleepers per length. Perhaps £10k a mile extra. Negligible really.

 

[Journeyman]

What's interesting about the Borders Railway is how much has changed in the last fifty years. Back then, very few people in Galashiels would have regularly commuted to Edinburgh, as most people would have worked locally, and the local services on the route were pretty sparse - it was always more of a through route, although now of course it's primarily an outer-suburban commuter railway.

Extending to Hawick would probably boost local traffic at the south end, and maybe lure more tourists into Melrose, but going on further seems like a very unviable project. I know the area around Riccarton Junction and Steele Road, and a huge chunk of the landscape south of Hawick is similar - there's simply nothing there, and unless there's huge demand for journeys like Gala/Hawick - Carlisle - which I doubt there is - it won't happen.

The idea of a diversionary route doesn't really stack up either, because the whole thing would need to be a double track electric railway to be any use on that front. Also, in the event of extreme weather, it's more likely to be blocked than the WCML - I think people forget how challenging it was to operate.

 

[paul1609]

What's interesting about the Borders Railway is how much has changed in the last fifty years. Back then, very few people in Galashiels would have regularly commuted to Edinburgh, as most people would have worked locally, and the local services on the route were pretty sparse - it was always more of a through route, although now of course it's primarily an outer-suburban commuter railway.

Extending to Hawick would probably boost local traffic at the south end, and maybe lure more tourists into Melrose, but going on further seems like a very unviable project. I know the area around Riccarton Junction and Steele Road, and a huge chunk of the landscape south of Hawick is similar - there's simply nothing there, and unless there's huge demand for journeys like Gala/Hawick - Carlisle - which I doubt there is - it won't happen.

The idea of a diversionary route doesn't really stack up either, because the whole thing would need to be a double track electric railway to be any use on that front. Also, in the event of extreme weather, it's more likely to be blocked than the WCML - I think people forget how challenging it was to operate.

Is a new station actually planned for Melrose? The abbey is a little over a mile along the river from Tweedbank. Unless the main road is going to be re-routed there is very little room at the Old Melrose Station. I thought that the only intermediate station was going to be St Boswells?