How to adapt the network to climate change

Network Rail’s Climate Change Task Force lead for Wales and Western Region Julie Gregory talks to Nick Brodrick about the challenges of adapting the network to shifts in temperatures and weather patterns.

“Where I sit right now, we have a lot of sites that aren’t resilient to today’s climate - let alone the climate in 30 years’ time.”

Julie Gregory is Network Rail’s Climate Change Task Force lead for Wales and Western Region. She is fully aware of the challenge ahead as sea levels rise - and with it the ever-increasing threat from storms.

The destruction of the Dawlish sea defence ten years ago (see RAIL 1008, 1010, 1012) led to the publication in 2016 of the South West Rail Resilience Programme’s study into the likely investment that would be necessary in future NR control periods. It embraces, for instance, the estuaries that follow the south Devon railway that ‘will need attention’ as far ahead as NR’s Control Period 10 (2039-44).

“We’ve had lots of learning from those events and we do need to take a more proactive approach,” Gregory explains.

“Now, what we’re doing is zooming out, looking at the whole of the region. Every region within NR is doing the same and saying ‘OK, what do we need to do? What are our highest priorities?’.

“We’re not just looking at a particular area - Exeter to Newton Abbot in that case - we’re now looking at everything and prioritising where we need to invest. We can let the Department for Transport plan the investment and not just land them with a massive bill to sign.”

In the five-year control period that started in April, NR has committed to climate adaptation pathways which will identify those ‘trigger points’ on a stretch of railway between A and B where it will become vulnerable.

Adds Gregory: “So, you might say that in the mid-2030s, it’s going to become flooded more often. Then, by the 2050s, it’s going to become inundated if we do nothing, so you’re identifying decision points.

“I think it’s fair to say that, historically, the railway has been quite reactive to things such as Dawlish and Carmont… getting through these adaptation pathways will help us learn where we need to invest without being led by knee-jerk reactions to specific events.”

She continues: “If we’re going to maintain the level of service we want, then we want to do something ahead of the 2030s.

“It could be a big infrastructure solution with big flood defences. Or if that’s not affordable, it could be to do something that gives us warning, keeps us safe, and then use diversionary routes, so we can at least inform passengers and be a bit more proactive in the way we deal with a flooding situation than we are now sometimes.

“Those are the two extremes and there might be loads of tunes you can play on it in between the two.”

Gregory says this kind of adaptation pathway has “never been undertaken before” on a national scale.

“We’ve just had a winter that’s been incredibly wet, with ten named storms by January 10. It’s been quite an exceptional year.”

Gregory, whose background is in meteorology, says it is “human nature to base your predictions of the future on recent experience”.

“It’s been a very wet winter, so flooding’s a big threat. We need to do lots of work to mitigate that.

“After summer 2022, it was all about getting very hot summers every year, so we need to do lots of work to mitigate that, too.

“Next winter, we could have an incredibly wet winter again or an incredibly dry winter, and have problems with drought in the summer. So, we have this huge variability that’s on top of the kind of gradual signal of climate change, and it’s difficult to not be too knee-jerk about it.”

Unsurprisingly, over the past two decades there has been an upward trend of delayed trains because of weather-related incidents. However, Gregory cautions that it is important to take wider context into account.

“We’re running a lot more trains now than we were ten years ago, and there have been various timetable changes, so that skews the figures, so when we have a delay somewhere, we get much more knock-on delay because of the number of trains we’re running and the frequency we’re running them at. There’s very little slack in the system at the moment and that muddies the water a bit.

“Where we’ve done asset renewals, we’ve really improved reliability, while there are other parts where a renewal is due because we have degradation of assets, so it’s not a clear signal because there are lots of things going on at the same time.”

However, the ongoing shoring up of the Devon sea wall and cliffs section has highlighted the tension between long-term resilience and taxpayer value for money, with some elements made ‘good’ for the next century and others only for the next few decades.

“I wrote the business cases for the Dawlish Sea Wall, sections 1-4. When you’re investing upfront in something that will prevent an interruption - possibly an event that will decimate your rail service for weeks at a time - you’re making an investment.

“But what you’re saving is the cost of repairs when those events happen in the future.

“You’re saving the cost of disruption to train services - replacement buses, compensation payments and so on.

“You’re saving the wider economic impact of people not getting to work, not getting to school or college, road congestion and safety impacts.”

Gregory says the event that NR seeks to prevent in particular areas “might happen once in the next ten years, but then in the following ten years it might happen twice, and then in the next ten years five times.

“It adds up massively as climate change takes off - it’s a non-stationary situation.

“These events will happen increasingly frequently as time goes on. Therefore, if you’re looking at a 50-year horizon, there’s often a very strong case to invest. Indeed, the investment cost over the period might not appear to be a great deal more than the projected repair cost.

“Investing in the sea wall at Dawlish cost £80 million for the two stretches and the bill for repairs in 2014 was in the region of £30m. That was some years earlier and a much shorter section. So, the cost of repairs, reactively, is of a similar magnitude to the actual investment in resilience.”

The two-month closure of the Riviera line was estimated by some to have cost the regional economy £1 billion.

“You’re saving all the disruption and the fact it might happen again and again, and the wider economic impacts as well. I don’t know how much investment it’s going to need in the future [on the Wales and Western Region]. That’s a massive question.”

The climate adaption pathways should provide the answer.

For the next round of spending (CP7, 2024-29) towards the adaptation pathways, that will be “some tens of millions”, which comes on top of drainage and other earthworks.

“On top of that, we’re looking at particular resilience schemes where a drainage renewal on its own wouldn’t solve it, or potentially looking at adding to a renewal scheme - just topping it up with a bit of resilience funding to make sure we’re taking account of future climate change a bit better.”

Does Gregory believe that the need to enhance or renew big pieces of infrastructure comes at not necessarily the worst time, because of the widely accepted notion that it’s a Victorian railway and lots of bridges are having to be replaced anyway because they’re old? Rather than just renewing that infrastructure, might NR take the opportunity to raise the height or make it more resilient to the crashing of waves?

“Yes, I think there are definitely opportunities there,” she replies.

“The timing with the structures is probably right . If they’d all been replaced 30 years ago, having not taken account of climate change, that would have been a missed opportunity.

“We need to look at all our major renewals and make sure we’re designing infrastructure fit for the future, and we’re trying to get all our standards updated to reflect that.

“I think there’s more awareness than there was. Some of our standards have been updated and we have new guidance notes on various aspects of climate adaptation.”

Yet, as we saw in RAIL 1012, plans to build a new railway formation along the Teignmouth Sea Wall have been postponed in favour of cliff containment. Evidently, it is also the case that NR is not going to throw cash at every conceivable infrastructure resilience scheme.

Says Gregory: “There will be a sea wall needed there at some point, but we held back on that because it’s not needed right now.

“That’s the other aspect to it. Yes, you can make a case quite easily for climate resilience, but there’s no way we would be able to afford to do everywhere like that. We have to try to hit that sweet spot where we don’t do it too early when there are better uses for the money, and we don’t do it so late that we’re going to get failures.

A one-size-fits-all approach is therefore not being pursued.

Gregory continues: “My view is that we need to be quite pragmatic and say that there are going to be some sections of railway where we’re going to say they simply can’t fail, can’t not be resilient, because they’re very highly used or maybe they’re the only link.

“Dawlish to Teignmouth is the only link for Devon and Cornwall to the rest of the network, so that’s a really vital route.

“You could say Paddington-Reading is the corridor that has had some performance issues lately, and we very much hit the headlines when that happens, so we can’t afford for that not to be resilient.

“You could have a debate about which category different parts of the network fall into - I don’t have all the answers, but I’m just saying that not everything can be bomb-proof.

“There will be some parts of the network where you say ‘Well, that one, we’re not going to be able to make a case for it’, perhaps for some of the branch lines that are susceptible to flooding. They just won’t stack up in terms of a strategic case - a ‘connecting people’ case and a monetary case - so in those cases, we might say that we’re going to bus passengers when it’s flooded.

“We have some parts of the network where it actually takes us the same amount of time on a bus as it does on a train, so it’s making sure we have that big picture.

“We’re working with local authorities and highway authorities to say ‘Right, what are we going to do for passengers to give them certainty when an area floods?’

“Because what we don’t want to happen is that we say ‘Well, we can’t afford to make that resilient’, and then we find there are no plans for buses, or that the roads are going to be inundated.

“There’s a lot around predictability, so can we be better about knowing when our infrastructure’s going to be impacted? With the best will in the world, even if DfT is very generous and well-funded, we’re not going to be able to make everything bomb-proof and resilient to all weather types as the climate changes.

“What we definitely need to do is be better at saying to passengers ‘Tomorrow, your train’s going to leave at this time instead of that time’, instead of them turning up at a station and saying ‘It’s delayed and that’s caused me uncertainty about when I get to my destination’.”

Tools in NR’s armoury include satellites to detect movement in earthworks, “so if we know in real time and can predict in advance when there’s going to be a landslip or slope failure, then we can be more proactive by routeing trains round it in advance”.

For Dawlish, the infrastructure operator has a protocol that predicts when overtopping of the wall is likely, based on onshore winds coinciding with high tides.

“Information is improving all the time, so if we’re a bit more confident in our forecasts, we can be a bit more proactive in our management decisions,” says Gregory.

Even so, she believes that maintaining every last yard of railway isn’t going to be practical: “There will probably be areas where we say it isn’t sustainable. There’s a section in Wales, at Fairbourne on the Cambrian Coast line. The town is performing managed retreat, so they’re saying they’re not going to be able to afford to put flood defences into the town. The town is vulnerable to sea-level rise; it’s a little way inland.

 

“So, what do we do there? Do we let the railway become the tidal defence? Do we say it’s not feasible and bus people across the section?

“There might be parts of the network where we say ‘We’re going to have a managed retreat here’. It’s important we keep a network going, so I’m not going to say that’s going to be a wide-scale practice, but just that we need to be open to that.”

Wales and Western Region’s Climate Change Task Force lead is refreshingly honest about the very real prospect that Britain’s landscape will change dramatically.

“Maybe for some of the coastal resorts where we have the railway line running right along the coast behind the sand dunes, as these places get pounded by increasing storms because of sea-level rise and more overtopping, it might be that it’s just not viable to have a rail network there.

“We know there are some parts of East Anglia that will become subject to sea-level rise, and there will be parts of the country where other infrastructure won’t be usable, so is that the case for railways? We just need to be open to that as a possibility.

“I do think that’s a difficult message, but I also think that’s where the adaptation pathways work will help, because then we’ll have an overall plan, we’ll know where our most vulnerable sites are to varying climate parameters.

“If we get an extreme snow event and we get criticised for investing lots of money in flooding instead, then we can actually say that we looked at the big picture, we’re looking at the next 30 years and providing the best resilience we can over a long term.

“We need to be driven by the data and making sure we take a long-term view, not a knee-jerk reaction.”

As a simple overview, Gregory foresees three categories for future responses to climate change:

• “The bomb-proof.”
• “The ‘We’re going to try to bounce back as quickly as we can from an event, so let’s make sure our equipment doesn’t get damaged. Let’s raise it up if needs be and make sure we don’t get a washout of all the ballast by using rock armour. We might be underwater for maybe a few days, but then can reopen seamlessly.’”
• “The areas where we might say ‘Sea-level rise is going to swamp this area and it’s not feasible’.”

On that third possible scenario, the forecast for a 0.5-metre sea-level rise in the next 30-50 years suggests that, inevitably, there will be places that are not going to be able to cope. Why, perhaps, would you keep the railway there if the conurbations it was serving were gone?

“I think we need to be open to that as an option, otherwise we’ll be putting a lot of public money into something that doesn’t stack up. Maybe the infrastructure it was connecting to won’t be there either.”

The railways don’t exist in isolation, of course, but there are certain ways in which they have previously been managed that might suggest otherwise.

“There’s definitely a need to talk to other sectors,” Gregory acknowledges.

Nor is that pure supposition. Since 2017, the South West Infrastructure Partnership has brought organisations and specialists together to “share information and connect relations between different sectors, in terms of planning and sharing experience on all aspects of infrastructure”.

It adds: “To achieve these objectives, it is essential that we speak with one voice and that we look broader than our own specialist interests.”

This is a point that Gregory is keen to pick up on: “Originally, it was about net zero, but now it’s expanded into climate adaptation.

“If I’m going to be developing plans to make the railway between Bristol and Exeter, across the Somerset Levels, resilient to flooding in the future, and if the Highways Agency is doing something similar for the M5, it’s important that we work together. That’s going to be more cost-effective than if we go off and do separate designs, separate plans, separate requirements.”

 

The partnership is “still at a bit of an embryonic stage at the moment”, but Gregory is hopeful that “as everyone begins to do things around the same time, we’ll get it to be joined-up and make some sensible decisions”.

Some of which are already being affected.

Hele and Bradninch level crossing (between Exeter and Cullompton) is a common flood site, succumbing seven times in the past decade when the River Culm has burst its banks. Part of the solution will be to demolish a nearby road bridge and replace it with a raised viaduct.

“As a result, we have a scheme that the DfT has funded, which will turn a once-in-two-year event into a once-in-five-to-seven-year event,” Gregory says with some relief.

“It will reduce the frequency of flooding, but it will still flood. And as climate change progresses, that once-in-seven-year event might become a once-in-three-or-four-year event anyway.

“What we’re now looking at is how we can improve that even more. There is nothing else to do with the infrastructure, but there is a nature-based solution.”

A new scheme is being developed that “looks at that whole catchment for the river”, of which Hele is downstream for most of its length.

The idea is to pay farmers to have attenuation ponds, to plant trees and to re-meander the river. Those solutions will prevent so much water running off the fields straight into the river (and town), which will in turn be slowed down.

“We need to be considering what is the most sympathetic way to do this for the environment. If we just did it in the railway corridor, we could make our drainage channels bigger and better and all of that. But actually, if we can stop all that run-off from the surrounding land before it gets to the railway by some of these natural flood management schemes, that’s going to be better for everybody.”

Who might fund such extensive works?

“I think it would be a multi-stakeholder funding model,” Gregory responds.

A similar project is already under way at the River Evenlode in Oxfordshire, with contributions from a number of sectors.

“It has NR contribution, and local authorities, other businesses and other people who could potentially be flooded all contribute to the initial capital cost.

“And there’s a payment that goes to the farmers year by year into the future, to pay them to keep that land as the natural flood management. It takes up more of that area of their fields if there are more meanders in the river, or there are more trees planted.”

For Hele and Bradninch, “that’s something that needs to be talked about a bit more”.

More widely, Gregory says she would “expect to certainly have information by the middle of this control period that would help us to set out where our main priorities are for proactive resilience”.

That would mean that by CP8, “we could start that journey to be more proactive, then in subsequent control periods, even more so”.

She concludes: “I’d hope within a couple of funding cycles we’re getting on top of our resilience issues and starting to hit them before they happen.”

A regional approach to track buckles

As well as torrential rain, waves and floods, there have been several instances of extremely high temperatures which can cause rails to buckle. The summer of 2022 was especially severe.

Does the trend towards higher temperatures and fewer extremely low ones have a bearing on how Network Rail manages its track?

“During that mid-July 40ºC peak, we put lots of speed restrictions on the whole network, which caused passengers lots of delays,” recounts NR’s Julie Gregory.

“But actually, the infrastructure was quite robust, with only one buckle on the whole of the Western route.”

“Continuously welded rail has a stress-free temperature where it’s neither under compression nor tension.

“Tension is when it’s cold and frosty, when the metal tries to shrink, so it’s being pulled and you get rail breaks sometimes.

“And equally, when it’s trying to expand in the heat, then it’s under compression and that can lead to buckles. 27ºC is [usually] the temperature where rail is stress-free, and it’s being neither pulled nor compressed.

“We need to be better at understanding the resilience of our network, and there’s lots of research going into that, trying to more accurately work out what the stress-free temperature of rails is, because they were installed many decades ago in some cases. And so, you don’t necessarily understand what they’re going to do under expansion.

“We need to know it accurately in order to be less risk-averse. We just put speed restrictions on everywhere [when the air temperature exceeds a certain threshold] because we just thought that we don’t know our stress-free temperatures accurately enough to be safe, but if we knew them more accurately, we’d be able to be more targeted with our speed restrictions and have less disruption.”

Gregory says there was a recent debate to decide whether rails should be stressed to a higher temperature, with the likely outcome that NR will likely shy away from a nationwide change, “because we’re still getting -5ºC happen, such as last winter.

“But there is perhaps an argument for having different stress-free temperatures in different parts of the UK because, for instance, Scotland does get more extreme minimum temperatures than we do in Western or Wales or Southern. And the same with the extreme heat, which tends to happen in the South East, so you could look at whether you should have regional differences.

“However, knowing the characteristics of the rail is the more important thing, so there’s lots of work going on into that at the moment, and how you measure it in a non-invasive way.”