Rural Economy and Connectivity Committee

Meeting date: Wednesday, February 19, 2020

Official Report 729KB pdf

---

Queensferry Crossing

Good morning, and welcome to the Rural Economy and Connectivity Committee’s sixth meeting in 2020. I ask everyone to please ensure that their mobile phones are on silent.

We come to agenda item 2, under which we will take evidence on the recent closure of the Queensferry crossing as a result of safety concerns. I welcome the Cabinet Secretary for Transport, Infrastructure and Connectivity, Michael Matheson, along with Dr Hazel McDonald, chief bridge engineer, and Lawrence Shackman, Queensferry crossing project manager for the design and construction phase, both from the Scottish Government, and Mark Arndt, who is representing Amey, the Forth bridges operating company. I invite the cabinet secretary to make a short statement of up to two minutes.

Good morning. I thank the committee for giving me an opportunity to provide an update on the closure last week of the Queensferry crossing. First, I thank those who were affected for their co-operation and patience, and I reassure them that public safety is of paramount importance in this area.

The decision to close the bridge was not taken lightly. On Monday 10 February, I was made aware that, despite monitoring arrangements being in place, ice had unexpectedly fallen from the bridge’s cables, causing damage to a vehicle travelling below. Our operating company recommended the closure of the bridge—a decision that Police Scotland fully supported. A road diversion was implemented and measures were taken to strengthen public transport until the bridge reopened on Wednesday 12 February. Although the closure of the bridge was unexpected and most unwanted, the engineers have used the period to better understand the circumstances that gave rise to the problem, which will help to improve the future management of the bridge.

Issues with ice were first noted in March 2019, and since then a range of actions have been taken. Ice sensors will be installed in the next few months. However, visual monitoring will remain the principal method of monitoring the bridge, as it is the most effective way to observe any development of ice.

In order to assist MSPs in understanding the nature of the challenges and the data that has been collected from the most recent incident, my officials have offered to provide a technical briefing, which will be held in Parliament tomorrow.

I am happy to respond to any questions that the committee may have.

Thank you. The first question is from Richard Lyle.

I note that the possibility of ice accumulating on the Queensferry crossing was considered at an early stage in the development of the bridge design. In view of the rarity of incidents of falling ice affecting bridges in the United Kingdom, it was considered that the risk would be best addressed through operational management of the completed bridge. Can you tell me who was on the original design committee, and who chaired that committee?

That was prior to my time in my current portfolio; I suspect that Lawrence Shackman is better placed to give you some of the history of the bridge design process. Nonetheless, I know that issues relating to ice build-up on cable-stayed and suspension bridges in the UK occur only very occasionally. There have been issues with ice on the Severn bridge and the Prince of Wales bridge, and on one occasion on the Forth road bridge, but such occurrences are rare. The assessment that was undertaken at the planning stage would have involved an acknowledgement that the risk of such an occurrence was very limited and should be dealt with through operational management.

On Richard Lyle’s wider point, the selection of a design for the bridge was part of the parliamentary process. Parliament considered the proposed designs and came to the decision that the current design was the most appropriate one. A parliamentary committee considered the legislation.

So there were members of political parties on that committee.

Yes, there were. It was convened by—

Can you confirm that Jackson Carlaw was the convener of that committee?

Yes—he was the convener of the committee that selected the design.

And he said at the time that the bridge was fantastic and “stunning”.

I think that he was quite right in stating that.

Lawrence Shackman can give the committee some background, from an engineering point of view, as to why icing was not viewed as a particular problem.

I ask Mr Shackman to be as brief as possible—seven other members want to come in on the back of Richard Lyle’s question, so a quick answer would be appreciated.

Certainly, convener.

The issue of icing was considered fully throughout the design process. As the cabinet secretary said, it was decided that a monitoring regime would be used rather than implementing any physical means such as heating chains, or any other heating mechanism, to alleviate issues with ice, because—based on knowledge at the time—such events were considered to be relatively rare. A lot of bridges around the world were managing their bridge structures through the observational method.

Throughout the design process, plenty of consideration was given to a whole range of factors. The committee will be fully aware of the wind regime, and the fact that we put windshields on the bridge to make it more resilient, which is one—

I took up your invite to walk over the bridge when it opened.

Yes—

Sorry—I will drag you back to the issue of ice, which is where we are at.

That aspect was considered in detail. However, in all the preparatory work and during the parliamentary bill stage, it was a very small aspect of the whole project, bearing in mind that we were talking about connecting roads, the type of bridge, hard shoulders, intelligent transport systems and all the other factors involved. I cannot recall that the Forth Crossing Bill Committee, during the bill process, went down to that level of detail to consider the build-up of ice, so I do not think that members would have been sighted on that particular issue at the time.

Last winter, falling ice was identified as a potential problem—indeed, a risk—and the committee was reassured that the matter would be looked into and addressed. I appreciate that the weather is outside the control of everyone in this room. However, last week, a car had its windscreen smashed in quite a serious incident that involved ice falling from the bridge, and the driver was very lucky not to have been injured. Do you regret the fact that an accident had to happen before we got to the stage of taking prompt action to resolve the issue of ice?

No, and I do not accept your characterisation of what has been done as involving very little work. That would be a gross misrepresentation of the work that has been undertaken since the incident occurred last March. My officials can give you some details of the nature of that work and the challenges around identifying the conditions in which issues with ice tend to occur.

Those who have suggested that, from now on, we will experience the same problem whenever it snows are factually wrong. It has happened on two occasions. During adverse weather periods such as the beast from the east, and during other periods of snow, there have been no such problems. Part of the challenge with what happened last March was that icing was not expected to take place. After the vehicles were damaged last year, the engineers faced a challenge in seeking to identify exactly where the ice had come from. Three possible areas were identified: the towers, the cables and the overhead gantries.

As the incident happened in a very short period of time, the engineers were unable to identify exactly where the ice had accumulated and fallen from. They thought that it had come largely from the towers as the principal source. However, their understanding of the weather at that time and the impact that it had on the bridge was limited—they did not have a data set to show that, if circumstances of a particular nature bring together a certain range of factors, the likelihood of icing is greater. Their level of knowledge on what had led to that particular incident was very limited.

The engineers sought technical expert advice on the matter, including international advice relating to other bridges that had experienced similar problems and the measures that had been taken. That work, which included hosting a conference to bring together experts in the field to explore what could be done to address the issue, was completed in October last year. The actions that were recommended as a result have already begun to be taken forward. The specifications for the weather-sensor equipment have been developed and tested, alongside the additional weather forecasting measures that have been put in place. In addition, this winter, we introduced a visual inspection regime that is to be put in place during periods of adverse weather. The bit that takes longer is the technical aspect, given the technical nature of the ice sensors and the way in which they have to be installed on the bridge.

A significant level of work has been undertaken. I refute any suggestion that nothing, or very little, has been done since March last year; that would be a gross misinterpretation of the work that has been undertaken. I highlight the challenging nature of that work, in which the engineers have had to try to pinpoint where the ice is coming from and the circumstances in which it occurs, as it develops only in certain circumstances. Mark Arndt may want to say a bit more about the work that has been done over the past year and the specific challenges associated with the very limited weather environment in which the issue starts to occur.

Before Mark Arndt enlightens us further, I have another point to make. Let us assume, given that we are talking about the Forth in winter in Scotland, that we may experience similar weather conditions. I would like to hear how we are going to ensure that no other vehicles will have their windscreens smashed by falling ice.

As has been explained, we saw a unique set of weather circumstances that resulted in ice formation on certain areas of the structure. That was the first time that we saw it happening on site; the engineers out there observed the ice falling on to the carriageway. We have analysed the meteorological data from those instances on site, and we now understand, and have a better insight into, the climatic conditions that resulted in the issue. We will be analysing that data along with the forecast information and the live data coming from the site.

We have now enhanced the monitoring regime, because the ice formation that we observed last Monday was very different from the slow build-up of ice that a lot of other big bridges have experienced. What we saw on Monday involved sleet and slush conditions that resulted in ice and slush and the like forming very quickly on the cables and parts of the towers. That was dissimilar to the experience of our counterparts in Sweden and elsewhere, with whom we have been liaising. We have taken that experience on board and adapted the monitoring regime. There will now be permanent monitoring patrols on the bridge throughout any inclement weather period, and all traffic management will be pre-mobilised in order to effect any closure, should that be necessary—

I am sorry—I want to narrow the discussion a little, as we have so many questions. Before Jamie Greene comes back to his last question, can you remind the committee exactly where the ice was forming? The cabinet secretary said that the options included the towers, the cables and the gantries.

We observed the majority of ice falling from the cables. The engineers also observed ice on the side of the tower, but most of it fell from the cables.

Jamie Greene has a final question, and then we will move on to questions from Colin Smyth.

Are the witnesses saying that, although issues with falling ice occur on other bridges, what happened last week on the Queensferry crossing was unexpected? Is it the case that it was not anticipated, and no one could have known that it was going to happen? Given what we now know about the way that the bridge reacts in certain weather conditions, how quickly will measures be put in place to improve its reliability and safety?

We have a three-phase plan to address that. As I said, enhanced monitoring will be in place from now on, throughout the winter period. We are also actively looking for permanent viable solutions to mitigate the formation of ice in the first place. That will take time, as it will involve research and development and the like. In the medium term, in anticipation of next winter, we will be looking at options to activate traffic management protocols more quickly, which could involve using the Forth road bridge in emergency circumstances. As I explained, the ice sensors will be installed over the next few months—

By next winter?

Yes—by next winter. The longer-term solution involves consolidating our research and development to enable us to introduce permanent viable solutions where they are appropriate to the climatic conditions and design activities of the Queensferry crossing.

09:45  

Cabinet secretary, when the icing problem first emerged last March, you said that it was extremely rare. However, the same problem has occurred this year. Based on the data that you have been able to analyse, how often is the problem likely to occur before you find any potential viable solutions?

Mark Arndt would be better placed to advise you on that, given that he has been involved in collecting the data.

We saw a unique set of circumstances that resulted in the formation of ice. That does not mean that those are the only such circumstances in which ice formation might occur—as I said, big bridges around the world experience various challenges in that regard. Without a doubt, I can say that all those whom we have consulted follow the same procedure, in which visual inspection is the primary monitoring method to see whether bridge closure is merited. I cannot quantify the number of occurrences that we can expect, but I can say that an enhanced monitoring regime is in place to safeguard the public, which is ultimately what we are trying to do.

So you are not able, based on the modelling, to provide a projection of how regularly it is likely to happen. Would it be once or twice a year? It is clear that there is a problem for which you have not yet found a solution, but you are not clear as to how often it might occur.

The historical analysis that we have done, in which we looked back over the couple of years’ worth of data that we have, indicates that the only two circumstances in which those unique characteristics arose were the incidents that took place in March last year and on Monday. On Tuesday last week, the climatic conditions were very similar, and on that basis we were proactive and the bridge remained closed while full-time monitoring took place, but there was no ice accretion on that day. The parameters are very sensitive to the variables on site. We are undertaking a data-gathering exercise, but monitoring is in place to safeguard the public.

I have a final point—

I will have to stop you there, unless it is very brief.

I want to make a point about viable solutions. We can monitor the situation and then close the bridge, but that is clearly not what people want to happen. We want to find a solution that will prevent the bridge from requiring closure. Based on the evidence that you have so far, how realistic is it that we will find a solution that will prevent the bridge from being closed while ice is cleared off?

We need to look at various options that are suited to the specific design characteristics of the Queensferry crossing. One option could involve coating the cable sheath in a particular hydrophobic material, and another could involve heating—there are a number of options.

We will gather intelligence on other bridges around the world, together with information on whatever modern technology and systems are available, and we will engage with a specialist research facility to carry out a laboratory simulation of the weather characteristics so that any solution that is taken forward to prevent ice from forming on the cables is suited to the specific characteristics of the Queensferry bridge.

Will that take months or years?

I am sorry, Colin, but I said that that was your last question, and I have to be hard. Peter Chapman is next.

There was a problem in March last year, and there is a problem again this year: the bridge has been closed for two days. The thinking was that ice formation would be a rare occurrence but, obviously, the evidence proves that it is not. Sensors are not enough—they simply tell us that there is a problem. What we need is a solution that enables the bridge to remain open. Mark Arndt illuminated what a solution might look like, but we need to find that solution. We need to find a way to prevent the ice from forming and allow the traffic to flow on a daily basis. That is the crux of my question. How do we do that, and when?

Let us be clear about the extent of the problem: it has occurred on two occasions. We should keep in mind that the cables were in place for several years during the construction of the bridge, and no problem was identified at that time.

Until March 2019, no problems had occurred. There have now been two occasions on which problems have occurred. Prior to the Queensferry crossing being built, a suspension bridge—I think that I am correct in saying that the Forth road bridge is a suspension bridge—was in place, and icing on the cables took place on only one occasion. Elsewhere in the UK, icing has occurred once on the Severn bridge and once on the Prince of Wales bridge; the problem has never happened on the Erskine bridge. If we look at the issue historically, we see that it has been a rare occurrence. However, it has now occurred on the Queensferry crossing on two occasions. The challenge has been to understand exactly why it is happening on the Queensferry crossing in a way that is different from the experience to date with any other cable-stayed or suspension bridge in Scotland.

It is also important to recognise—the international evidence supports this—that there is no off-the-shelf solution for bridges that experience such problems. The international experience shows that bespoke methods that are specific to the bridge and the particular circumstances that it experiences often have to be looked for.

The research to which Mark Arndt referred is partly about trying to identify the specific circumstances that have given rise to the issue with the Queensferry crossing and the potential mitigation options to address them. It is also about considering whether bespoke measures are required if they can be put in place within the bridge’s existing structural parameters.

Good morning. I have a quick question. You said that eight vehicles were damaged and that safety measures have been put in place. Can you clarify—I have my nurse hat on now—whether any person was injured or hurt as a result of the falling ice?

No persons were hurt by it. One of the challenges that was presented on the Monday involved the speed at which the ice formation took place. A squall of sleet and icy weather came in, and the ice built up very quickly and fell off very quickly. It happened over a fairly limited period of time. That was one of the major observations that the engineers were able to make on this occasion; they were not able to observe what happened last year. This time, they were able to observe how the ice behaved in a way that they had not previously been able to see.

No individuals were hurt. However, we need to look at ensuring that, where possible—it is always difficult to give guarantees on such matters—we can try to predict where there is a greater risk of ice forming. That is part of the work that is being taken forward. Observations are being made and, if there is a need to take action, a diversion can be put in place very quickly. That is part of the debriefing work and the further arrangements that have been taken forward based on what was found to have happened during the incident last week.

There are still a lot of questions to come, so I am afraid that each member will get to ask only one question. We will hear from Stewart Stevenson, followed by John Finnie and Maureen Watt.

I want to explore the statistical risk associated with the issue. On 9 February 2018, a street cleaner adjacent to Kew gardens station in London was narrowly missed by a 20kg block of ice that had fallen off an aircraft. There were 25 such incidents in the UK in 2013, 12 in 2014, 12 in 2015 and 10 in 2016. According to the Civil Aviation Authority’s figures, it is statistically more likely that aircraft flying into Edinburgh airport will drop ice on nearby inhabitants than that ice will drop from the Queensferry bridge on to people who are travelling below. Are you going to close Edinburgh airport next time there is a risk of icing?

That question is about Edinburgh airport, and we are discussing the Queensferry bridge. Stewart Stevenson has made his point, so I will bring in John Finnie.

There are a number of significant structures in the Highlands, such as the Skye bridge, the Kessock bridge and the Dornoch bridge—I thank all the staff who keep them open in very demanding circumstances. Can you reassure us that, if there are lessons from the Queensferry crossing that are of any relevance, they will be rolled out to those other structures, which are important for connectivity in the Highlands?

Of course, although some of those structures are not cable-stayed or suspension bridges, which have cables. Hazel McDonald, as the chief bridge engineer, is probably best placed to advise the committee on where there are any other lessons that need to be learned.

I ask Hazel McDonald to be brief, as we are drifting away from the Queensferry crossing.

The Kessock bridge is the only similar bridge, as it, too, is a cable-stayed structure. We have a number of estuarial crossings on which there is a possible risk of icing, but the Kessock bridge is the only cable-stayed structure. Anything that we learn from the Queensferry crossing will be applied to all our cable-stayed and suspension bridges.

I am not aware that there has been a problem with icing taking place on the Kessock bridge.

It has been suggested that the Forth road bridge should have been available as an alternative when the Queensferry crossing was closed because of ice. However, is it possible that the road bridge would not have been available either, given that the icing was accompanied by very high winds? During this winter, how often has the Forth road bridge been closed while the Queensferry crossing has remained open?

I will start with the latter part of your question, on whether there have been occasions on which the Forth road bridge has been fully closed while the Queensferry crossing has still been operating. In fact, that happened last weekend: on Sunday, the Forth road bridge was completely closed because of high winds, but the Queensferry crossing continued to operate. There have now been 55 occasions on which restrictions on the Forth road bridge would have applied when the Queensferry crossing continued to provide full access to vehicles. It is a much more resilient and reliable crossing than the road bridge, given the considerable disruption that was associated with the restrictions that often had to be applied to that bridge.

The Forth road bridge could not be used as a diversion route last week because of a number of different factors. First, the northbound carriageway is currently undergoing a major piece of refurbishment work that includes the removal of the joint spans, so there are gaps. Secondly, the southbound carriageway is currently in a contraflow, which is being used as a public transport corridor. There is no hard shoulder so, if a vehicle breaks down, the whole place comes to a halt—quite literally.

During the incident last week, the decision was made to increase the public transport offering. There was a 50 per cent increase in the buses coming across the bridge in the peak morning period and increased rail capacity to try to assist people in making journeys by public transport rather than using the bridge. We have also asked officials—

Michael, I think—

I think that it is important that I give a full response.

It is important that we try to get through as many questions as we can. The committee absolutely understands that there are a lot of administrative reasons why the other bridge could not be opened.

They are not administrative reasons—they are practical reasons, convener. Another reason why the road bridge could not have been opened was that we had a situation in which some works were taking place off the bridge that would need to have been completed to enable the bridge to be opened. For those reasons, it could not be opened. However, I have asked officials to look at whether, in future years, it would be possible for us to minimise the repair works that take place on the Forth road bridge during the winter months so that it can be used as a diversion should an incident of this nature happen again. That is part of the work that we are looking at taking forward.

I will bring in Mike Rumbles.

The cabinet secretary has answered my question—thank you.

That is perfect. I have one question to finish; the rest of the questions will have to be answered during the cabinet secretary’s question-and-answer session and briefing for MSPs at 5.15 tomorrow.

Cabinet secretary, the problem that you have described is not unique. As you have alluded to, there were two incidents on the Severn crossing in 2009, and there have been incidents on similar bridges in Canada, Norway, Sweden, the United States and Japan, which all date back to well before the Queensferry bridge was built. Why were sensors not included in the original specification, considering the problems that had previously been faced on those similarly constructed bridges?

10:00  

Lawrence Shackman is probably best placed to explain what happened at the time of the design process.

At the time of the design, leading up to the bill process, the team looked at other bridges around the world and how they dealt with those sorts of issues. We found that there was no common ideal solution to get rid of the problem.

To go back to what was said at the beginning of the meeting, the most tried-and-trusted method is to manage the risk by observation. Other mechanisms, which we have started to discuss, have been employed on other structures since we made the decision to go ahead with the observation method. A lot of those mechanisms have intrinsic problems. They are very costly to install, or they are unreliable in respect of the heating and the mesh on the cables—Mark Arndt alluded to that. In order to go down the path of any one solution, we need to do the research that Mark mentioned to find the most appropriate solution.

Using my own rules on myself, I guess that I will have to wait for an answer about why the sensors were not put on when the bridge was originally built, given that the problem was experienced worldwide before then. I will find that out tomorrow when I question the cabinet secretary.

I thank the cabinet secretary and the panel members, and I suspend the meeting briefly.

10:01 Meeting suspended.  

10:03 On resuming—