Good morning and welcome to this meeting of the Scottish Affairs Committee. Today we are looking at digital connectivity in Scotland, and we are very pleased to have with us today three people who know quite a lot about that issue and have personal experience of it, too. Could I start by asking you all to introduce yourselves and say what your role is, please?

Good morning, Chair. My name is Ann MacDonald. I am from Tiree, but I am here mainly as a representative of the Scottish Islands Federation, which is a network of community-led organisations that promote the interests of inhabited Scottish islands.

I am Rhoda Meek and I am also from the isle of Tiree. I am the Head of Communications and Gaelic at the Tiree Community Development Trust, so I am speaking on behalf of the community of Tiree today.

Hello, everybody. My name is Mhari McLeman and I am here today from the Shetland community. I am here as an outspoken resident who was severely impacted by broadband outages in 2025.

Excellent. Thank you all very much for making the journey down. I am glad you all got here safely. Rhoda, could I come to you first? I understand that following Storm Amy, Tiree was without any digital connection for up to seven weeks. Could you describe the impact that had on the island?

Absolutely. When Storm Amy landed on 3 October, we had a major power outage, which also simultaneously knocked out multiple services. As the power came back on, internet connectivity did not return. That included all forms of communication. It was not just home broadband, but also the 4G network, all voice services and, for a period of time, a substantial part of the 999 service as well. When the community realised somewhere around 8 October that our cable was broken, things started to get into gear in terms of figuring out where the impact was going to be most felt. To give you an example, households could not reliably call family or parts of the emergency services or medical services. Businesses could not take card payments or access their ordering systems. We could not top up electricity meters, because that relies on an internet connection in the shop. Things like paying your bills and accessing cash were very difficult for people. Buying fuel at a fuel station also requires all these internet connections, so everything that feels like a normal activity, day to day, was severely impacted. Everything comes in on that one cable, which is about the size and width of my fist. All forms, really, of communication were down until about 10 October. The BT Group sent out an emergency response team, and things started to gradually come back online, using Starlink as the main means of putting things like the medical practice back online and connecting some services back up. It was a couple of days later that the 4G masts came back online as well, again using satellite. It was EE with the emergency services network. It took until 21 October for the other network providers—O2 and Vodafone—to come back online. It was 18 November, a full seven weeks later, before the cable was repaired and Tiree was back to having full connectivity.

That sounds horrendous, I have to say. I understand from other conversations we have had that there was no 999 service for up to six days, as I think you mentioned. Was that a widespread loss or was it patchy?

It would be fair to describe it as patchy. At the height of the storm, there would have been no way of contacting the emergency services at all unless you had an analogue line still. Those are, of course, being phased out now, so a large part of the island no longer has analogue phone lines. That assumes that the person at the other end of that phone was on an analogue line, so it is very hard to extrapolate exactly what the consequences could have been, but they could have been catastrophic. One of the key points, I think, is that the 999 emergency services network on mobile defaults to the last mast of resort, which would go through all the providers if EE was not available. Unfortunately, the Vodafone dishes were also knocked out of alignment and that took down O2, so by mobile it was very difficult. At my end of the island I could get a weak signal from the Isle of Multinational, so in theory I could have got through, but I guess that theory and a weak signal while standing outside are not enough. One of the key things is that our medical practice is one of the buildings that had moved to digital and digital voice, which meant that the doctor’s surgery could not be contacted. That was probably, locally, one of the biggest impacts for us. For the first six or seven days, there was a handwritten note on the surgery door letting people know that if the doctor was not there and they could not reach him, this was the house he was staying at. On the one hand, it sounds quaint; on the other, it could be minutes that you are counting in an emergency, so I think we were very fortunate that there was not a catastrophe and that someone was not hurt—or worse. Yes, so it was patchy. But the big concern is that the emergency services network itself is reliant on that cable, which has a single point of failure. We also know that the battery back-up on that is very short. That was less of an issue for us because we simply did not have a connection, but the battery back-ups are low and there is no alternative route for that 999 service—there is no satellite back-up, for example. As soon as the emergency response team arrived, they were able to plug the masts into a satellite back-up. So, in future, one way of protecting that service would be to look at that as standard failover, with an expectation that that is done. My last point on the 999 issue is that there was no playbook. There was no clear incident plan. There was no rehearsed plan for what happens to emergency calling in that situation, who is in charge, where you go and all that. So, yes, it was widespread.

Is there work ongoing now to ensure a chain of command and that there is an alternative—that some system will kick in that will help people to navigate the problems they have?

I think that is one area where Committees like yourselves might be able to help. It remains unclear. There are local resilience partnerships. The communications providers are conspicuous by their absence in those. Power would be expected to be there; Scottish and Southern Electricity Networks would be expected to be there, and other critical infrastructure. However, digital, I believe, is not currently considered critical infrastructure. In a situation where you have medical alerts now on the internet, where you have children’s diabetes glucose monitors internet-enabled, I think there is a case for digital connectivity to be considered as critical infrastructure and for those providers to be included in local resilience partnerships. There is a conversation happening at the moment in Tiree, as an example, where the community development trust—which took a leading role in this incident because it had the skills within the organisation to do that—is being asked whether it could in future be part of that resilience planning. The theory is good, but the staff cost of that time and that implementation is large. Community development trusts are already carrying quite a burden and doing quite a lot of things that, as a rule, a statutory local authority would do. In our context, I think it is a conversation that needs to be had. It is also a role that needs to be resourced.

Thank you. I do not want to sound like a tabloid journalist here—not that I have that much against tabloid journalists, you understand—but were there near misses? Were there situations where there could have been a catastrophe, but it was averted either by people stepping in or by luck?

I think it was simply luck. I am not aware of a near miss, but I do know that some vulnerable people were unable to access care provision briefly. Our care services and our home care are fantastic. One of the beauties of an island, and one of the fortunate things about an island, is that we know each other and people go and check on each other. That is a good thing, but it should not be assumed that that is going to happen. Just prior to the power going out, one example was that somebody on Facebook asked whether anyone was available to help because water was coming in through their smoke detector, which was sparking, and the power went out shortly afterwards. The outcome in that situation was good, but if it had been a few minutes later or a different scenario, you could have seen a very different outcome there.

Beyond the care sector and emergency services, were there any other, I suppose, critical systems that failed as a result of these problems?

There were so many services that failed as a result of this problem. There was a real cascade effect, and it brought home to us all just how much we rely on the internet these days to get the basics done. The bank could not open. The post office could not open. The fuel station had no remote monitoring for its CCTV. Our fuel station is community owned and therefore not monitored 24/7 physically, so it relies on off-site mainland monitoring, which is really important when you are dealing with that amount of petrol. I guess the electricity top-ups are also important. In small businesses—I run my own—you rely on the internet for sales, bookings, doing payroll, crofters’ compliance—you have 28 days to register new calves, for example. This outage was double that, so people were in breach of legal responsibilities. The school was offline, but this situation coincided with half-term. For the school itself, had this continued or had it happened at a different time, Tiree is partnered with Oban and does a lot of video conference and a lot of online and remote learning. Young people were in the run-up to their prelims as well. There is a lot of online tutoring. I was in one meeting where someone said, when we were bringing this topic up and trying to avoid filtering on our community network, “Well, just they should just read a book.” Unfortunately, it is not as simple as that in 2026. This was not about streaming, and it was not about Netflix; it was about access to critical and important services.

That is very helpful. Thank you.

I would like to change to asking you specifically about the resilience that is or is not built in and how effective Starlink was in meeting the needs of local people. I am just wondering whether there was a cohort of people who were better supported when Starlink came in and whether other people were not as well provided for.

There was no resilience built in. With these single spur cables, you are almost building a single point of failure in as a design feature. So it was the cable or it was nothing. In Tiree we were fortunate. We have been running a community broadband network, due to the lack of fibre, for the last 21 years this year, and we were in the process of winding that network down. We still had half of it in place. When the emergency response team arrived from BT, they quickly deployed Starlink in a variety of key locations. We would like to thank that team. We cannot fault them; they did an incredible job. They arrived, they looked at the landscape, they looked at the thickness of our walls, they looked at the weather, and they quickly understood what we were dealing with. Starlink deployment was highly effective as an emergency stopgap—there are no two ways about it. It was deployed at the surgery, for example. Due to having our own community network, which was still covering about 160 premises, we were able to get a Starlink and plug that into our network, because of the number of vulnerable people and isolated people we were serving. Because of the way our network works, it took our network engineer an afternoon to plug it in and reroute it, and that put about a third of the island back online, certainly not at a fast speed—not at a speed you would want to stream on or watch TV on—but in a way that you could at least get the basics. We were able to provide a group of people in Tiree with a basic service, which allowed people to go to friends’ houses and try to get themselves online. Starlink was deployed to a number of community hotspots as well, so people could use the internet. We had a bench outside the telephone exchange, and people were parked outside the community hall in their cars, checking their emails and things like that. It was not ideal, but it was certainly far better than nothing.

Do you think that that solution is workable if a similar outage came again? I suppose, in addition to that, BT has taken the cost of that. Who do you think will ultimately bear the costs of these outages?

In the future, to us in Tiree, satellite seems like the sensible back-up option in almost all situations. We would want to see Starlink, or an equivalent, deployed as a failover mechanism to exchanges, so that if one route goes down, the satellite option kicks in, and the same for our mobile phone masts, particularly the emergency services one—there should always be a separate backhaul route for these. Tiree was in a very unique position because we had the ability to get internet to quite a large number of people. Had we been able to work more effectively, I guess, with BT, we could have got it to more people. Unfortunately, we ended up in a situation where we were pressured to restrict that service to only the vulnerable in the community, which was very hard for us to work out. We were being put in a very difficult position by being asked to work out who was vulnerable and to make decisions. At the same time, the home care and the council’s back-office services were offline. There was quite a lot of time spent negotiating and trying to keep that community connection online. A large part of the argument was directly cost related, because everything has a budget, and data costs money. I think the people who should pay for that data back-up are the cable owners and the communication providers, because if you are not building resilience in, the outcome could be catastrophic. If the resilience is not built in and you then have to provide a service, you should pay to be providing that service to people. We live four hours offshore. We are used to bad weather. We are used to making do and we do not expect perfection at all. When things go down, you know they are going to go down and you can plan for that, but there does have to be an expectation of a basic level of service. That should be provided not by community organisations working overtime, but by the communications providers themselves.

The contract for the service should include delivering the wider service when things go wrong?

Absolutely.

Thank you.

Rhoda, how adequate do the islanders consider BT’s response to the damage caused by Storm Amy?

I think that islanders were satisfied with the initial response. We were certainly very impressed by the response as soon as it became clear that the cable was broken, and I think that is an interesting point. On 3 October, the cable broke, and on 8 October, the community discovered that it was the cable that was broken. Prior to that, we thought there was an issue in the exchange in Mull. What BT knew and when, I do not know. It was a very confused time. There was a complete lack of communications. Fibre was being laid at the time in Tiree, and about 10 fibre vans were at my end of the island, with people waving a phone in the direction of Mull, trying to contact the office, so it was very confused. As soon as we were all aware there was an issue, BT scrambled an emergency response and they were fantastic on the ground. The response on the ground was excellent and it was what was required at the time. In terms of that longer-term piece about how you keep a community online and how you get the best service to people, between the critical failure and the fix, I think things were far less thought through. There is an expectation that this will not happen; there is an assumption it will not happen. I have sat in more than one meeting where, having asked—this was more than a few years ago—what would happen when the cable broke, I was assured that it would not break, that it was far more resilient than analogue and that that was not something we should be concerned about. It is not fun to be proved right in that situation. The questions have been asked for a long time. In the long term, we want to see more effort given to this question of resilience. It is not being taken as seriously as any community on a single cable expects it to be. I feel like I did not answer that question quite as well as I would have liked.

No, that is fine. Did they give any reason why it took seven weeks to repair the cable?

I think seven weeks was probably quite a short timeframe. It could have been an awful lot worse. The boat that fixes these cables was otherwise deployed. There are not many of them, and we had to wait for that to travel. You have to get a whole range of marine certificates and things signed off for work in these areas. Fortunately, the Scottish Government got involved somewhere around the middle to the end of October, when we put quite a lot of political pressure on, and things really did start to motor. A boat was put on standby. Seven weeks from start to finish—we probably did well. We got a good weather window. It should have been six weeks, but the weather blew up. But we were fortunate with the weather, because at one point we thought it could be well into the new year before we would get a boat out to do the fixes. So, yes, it was a long time, but it could have been a lot worse. After political pressure was applied, certainly, they pulled out all the stops to get it done, and we appreciate that.

It was resolved realistically. Do the islanders accept BT’s argument that two faults over 11 years demonstrates adequate levels of resilience?

No, they don’t. I would respectfully suggest to BT that frequency is not a relevant metric when the consequence could be catastrophic. Two faults that remove core communications, emergency calling, essential services and commerce for weeks at a time is not resilience at all. The fact is that this has happened twice, that it will happen again and that there is still no back-up plan that will kick in effectively to replace that feed at the exchange.

Yes. I hope they are listening in and that there is a back-up plan done soon.

We are now going to move on to look at the situation in Orkney and Shetland. Maureen, I think you have the next question.

Excuse me, I have a sore throat this morning. We have just heard about the severe impact Storm Amy had on Tiree. Turning to the disruption that Shetland faced in 2022, did the outage create similar challenges?

Yes, absolutely. The outage in 2022, unlike the situation in Tiree, was not related to storm damage; it was related to infrastructure and damage caused by an alternative mechanism. The outage in 2022 was, I would say, catastrophic in terms of impact on the community and making the community not feel safe. I would describe it as living in a war zone, when all your comms are cut off. Unlike Tiree, Shetland does not have a single point of failure. We are very lucky that Faroese Telecom has infrastructure that comes to Shetland—it comes down from Faroe, and then it goes to Orkney and on to mainland Scotland. The main cable is the one that goes south; that is the main cable that the connectivity is through. If there are any problems with that main cable, the idea is that everything reroutes to the north. When we had the catastrophic outage in 2022, the resilience cable to the north was currently broken. It had been severed, I believe, by a fishing vessel. The catastrophic outage happened when the cable that goes south also was severed by a vessel. This is not storm related at all. That cut off all the mobile phone masts and everybody’s connectivity, essentially. There was no certainty that 999 calls would work. The police declared a major incident. The police, within their means, put out more people into the community. The community was informed to flag down an emergency vehicle if you were in that situation. It was quite terrifying. Faroese Telecom acted as fast as it could, and I would say that it is always pretty efficient and gets on the case pretty quickly. What it did, I believe, was it put in a temporary solution. The cable was not fully severed. There were still some fibres that were operational, so it was able to do something to do with putting more power or signal strength through. By close of play that day, the BT Group had put in a similar temporary solution. So for one full day, it was catastrophic outage in terms of comms, I would say, and then for a period of about 10 days, while Faroese Telecom co-ordinated the repairs to that cable, there was ongoing disruption, including to members of the community who rely on telecare systems. It was probably during 2022 that we realised—although, as a community, probably did not properly understand—that some internet service providers have resilience built in and some do not. In terms of all the things that Rhoda highlighted for Tiree, it was a similar situation in Shetland. It was not nice to feel unsafe in your community.

Thanks for that, Mhari. You have actually covered my next question, which was about what impact that had. You were saying the police had it as a major incident, so I think you have covered how critical systems went down and how that affected the community. Thank you very much for that.

My questions are also to Shetland’s outspoken resident, as you described yourself. Can I turn to the damage that was caused in 2025? As I understand it, there were two incidents—one, I think, in July and one in October. The first was likely to have been caused by fishing boat damage, and the second perhaps by rough seas. It was reported that some residents remained online while others lost their service for days. Was there that disparity, and can you explain why that was the case?

Yes, I will try to explain. I was one of the customers who was impacted by having no connectivity in my home. My parents, who are in their 80s, were also impacted. In the July outage the cable break to the SHEFA-2 cable, which is owned by Faroese Telecom, happened between Orkney and mainland Scotland, so that impacted customers in Shetland and in Orkney. In October we had a 26-day outage, and the break was somewhere off Orkney, so it just affected Shetland customers in that particular instance. It was very confusing to be a member of the community whose connectivity has gone off but your neighbour’s was on. As an outspoken member of the community, I decided to do a little bit of digging. What we now understand is that some internet service providers—predominantly Faroese Telecom and Shetland Telecom, and the providers linked to the BT Group, which is EE, Plusnet and BT themselves—follow network guidance and have resilience built in. They have the ability. It is automatic; the traffic reroutes. If you are with Sky, Utility Warehouse, TalkTalk or Vodafone, what we realise now is that they do not pay for resilience. So when that cable broke, and as the repairs get organised, those customers just have no connectivity. People in the community were going, “Well, why do they not just reroute to the other cable?” There is no commercial agreement in place for those companies. There is capacity on the Faroese cable that goes north to Shetland. There is the possibility to put arrangements in place, but at some point those internet service providers have made a commercial decision: “Do you know what, a single point of failure is okay.” That is what the community are not happy about.

Can I ask if your own internet service provider had that alternative cable route? Did you experience that?

I was with Vodafone, and no, we were just off. We were off for 12 days in July and 26 days in October.

Without internet?

Without any internet, and with no landline. It is not a catastrophe like July or the situation in Tiree because I had friends; I could go and do my internet there. It was not a case of the 2022 outage, when cash machines are down and supermarkets are not operating. However, it did have a profound impact on those businesses and householders, including vulnerable householders, who found themselves saying, “Right, okay, I understand now I am with an ISP that does not have resilience.”

I take it that that was not widely known at the time. Is it widely known now which providers do and which do not?

Oh, it is widely known now. During the outages, particularly the second one, people tried to leave those ISPs in their masses. There is one-touch switch; customers should be able to change their internet service provider with ease. A lot of people had not moved after the July outage because they were waiting for the compensation to come through—the automatic compensation that never arrived. So we were a bit disgruntled. We were then facing another outage, but by that point people did not care about the compensation any more. They just wanted a digital connection. Many people tried to move providers—in their droves they were trying to move. Those internet service providers happily took the orders, and it did not work; in the vast majority of cases it did not work due to some failure with the local Openreach network that would not enable all switchover stuff to happen.

When you and other islanders took out the contracts originally, were you aware that those providers were failing to follow the network guidance you spoke about?

Absolutely not. There is nothing at the point of sale that tells the customer how resilient—

Is there now?

No, absolutely not.

Have you had communication with them about that?

The response of the—let’s call them the weak ISPs—to the community has been pretty woeful. When you reported the outage, they would claim things like, “Oh, we will give you unlimited data. We will send you a mini hub.” But those things do not work either, because the infrastructure is not adequate for the number of people needing to use them.

How satisfied have you been with their response since? Are the companies you mentioned—Vodafone, TalkTalk, Sky—taking steps now to build that resilience and comply with the guidance?

I have asked the question of Sky and Vodafone because I have been dealing with Sky with my parents. Neither of the operators would answer the question about whether they are going to use the existing resilience back-up route that is available. They speak about new cables coming a few years down the line, or about new satellite systems that are maybe going to be developed. There is no action that they are committing to take just now. Even though they made those commitments after the 2022 and July outages, there does not seem to be any evidence of any real progress having been made.

What message do you have for them this morning, then, if they are watching this?

Well, I am not sure if they have any customers left in Shetland, because I think most of them have moved on to the more reliable providers. They have absolutely failed island communities. If they are not prepared to follow network guidance, they just should not provide a service to island communities. There needs to be more openness and transparency about the level of service that they are able to provide to communities like Shetland. Perhaps more than that, Ofcom is the regulator. They made a very sobering point at a resilience summit that was organised in our community: they have the guidance, but it is guidance; they do not have any powers to force these ISPs to actually build resilience into the systems. It is similar with the compensation schemes: it is voluntary and non-enforceable. I think the community is currently feeling that Ofcom is a bit of a toothless regulator and does not have the legislative framework behind it to do what it needs to do to build resilience in service provision in communities like Shetland.

That is very interesting. Thank you very much, Mhari.

Thank you. We will now move on to a bit of an overview of the islands with Angus MacDonald.

I have a quick question one before I ask about that. Rhoda, with the Tiree outage after Storm Amy, I am surprised that a storm above the sea would damage the cable under the sea. Are we sure that it was a storm that caused it?

We as a community have not been given that information. I do believe that it was a break in the cable; to my knowledge it was a pretty clean break, but that is anecdotal. One of the things that often gets forgotten when it comes to things that sit under the sea is the extent to which the swell moves them around. The particular part where that cable comes in—it broke about 600 metres offshore, just behind the exchange—is a very rocky area of ground. It could have been cut. It could have been wearing for years. It could have been damaged and the storm affected it. It has been broken before, as BT has told us—with two faults in 11 years—and I believe that that was related to a fishing boat. I am also aware of anecdotal evidence of that cable moving under the seabed. It is plotted, but it does move under the seabed and has been lifted before now by fishing boats and put back down again.

Mhari, you mentioned two dragged anchors off Shetland? There were four instances off Shetland, weren’t there? Shetland experienced four instances in three years?

Yes, I believe so. Sometimes you are not so aware of the damage and sometimes you are.

Do we think it is always ships or anchors that are causing it in Shetland?

Hopefully, Faroese Telecom will be able to provide you with the answers. In many cases, I believe that the core infrastructure has been damaged by vessels. Occasionally, those vessels do not have their AIS systems switched on, which is mandatory in other countries in Europe. In the 2022 outage, I believe—I do not have 100% evidence—that the north cable was damaged, the south cable was damaged, and it was the same vessel.

Can we assume that the vessel was a local Scottish vessel, or is there any suggestion that there was a foreign vessel, perhaps a Russian vessel, in the area?

No, I think it was a fishing vessel. Faroese Telecom will have that information to hand for you. It might just tell me that it was a theory that it was the same vessel that damaged both cables.

The AIS tells you where the ship is, but it does not tell you where the cable is, does it? It can only be used to identify which boat did the damage—

In last summer’s outage, the vessel owner reported it very quickly. They recognised that they had caused damage to the cable between Orkney and the mainland. I do not think that happened in 2022.

Over the decades has it been a regular thing that cables are caught up? I had never heard of it until two or three years ago. Has it happened regularly over the last decade or two?

I think it is important to learn from the example of the power cables. Certainly on Tiree, we had a real spate, from about 2012 to 2016, of power cables deteriorating and going down. It speaks volumes that community-owned turbine projects now cannot get business interruption insurance for subsea cable damage because the power cables have been damaged so much that the insurers, the actuaries, know the risks of power cables. In terms of the fibre cables, the first lot in Scotland were laid in 2014 under Digital Scotland Superfast Broadband, so those are now 12 years old. There is another set that has been going in recently, with 16 cables to 15 islands. You have 35 cables down there, some of which are 12 years old, all rubbing on the seabed, so it is something waiting to happen. To me and to our members it is a surprise that cable damage has only happened twice in 11 years. With the power companies, they take preventive actions. They examine the cables. They check them. When the Tiree cable was mended, we said, “Why don’t you just check the rest of the cable back to Mull?” We were told, “No, no, no. We send a pulse of light down, and if the light goes down, it is fine.” There was no attempt to look at the outside armouring, the metal casing, to see how many filaments were still working. There could be other sections. It was a kilometre and a half that they replaced. There could be another section further down that equally is about to go. There is not that attempt to predict what is going to happen. It is, “No, no, it will not happen. It will be fine.” That is what we find worrying. That is certainly what our members find worrying. They are anxious and they are concerned about it.

A question to British Telecom would be: what are you doing to provide sustainability of the—

Yes, absolutely.

This is for you, Ann. What are the main threats to communities when damage occurs and partial or total connectivity is down?

We have been looking at this for quite some months, and we sent out a survey to our members and to contacts last month. We had replies from 34 islands, so we have some rich data. The impacts are, as Rhoda and Mhari said, widespread across the islands. As I said, members are concerned, they are anxious and they are frustrated. They are anxious about the danger to life—the things that Rhoda spoke about, and the telecare thing in particular. There were instances in Mull and Islay where people’s telecare did not work. You talked about near misses. There are reports of quite a few very near misses. During Storm Amy, the coastguard did not have any comms. There are examples of not being able to get 999 calls across the islands. In terms of day-to-day life, it just makes things so much harder when this happens. People are frustrated at the amount of time it takes. We talked earlier about getting the repair. It is interesting that the repair team for Tiree came up from Kent. You are thinking, “Why is the team in Kent?” There is one boat that the cable providers pay into a scheme for. That is based in Portland. The boat that came to the rescue in Shetland is in Dieppe. I understand that the MOD has a cable boat. There are two boats, so all the islands are thinking that if Shetland, say, goes down, what happens if Tiree goes down at the same time? We are going to have to wait for another island to be looked at first. They are frustrated about the time it takes. There are so many processes: the way that the cables were laid and funded, the jurisdiction of the UK Government, the fact that it is not devolved to Scotland, so you have the UK Government overseeing that. You have Scottish Government money, you have Openreach involved and you have local authorities involved. As far as the community is aware, it is not clear whose responsibility this is. Where do we go? How do we know what is happening? The agreements with Openreach, we do not know—we do not know what to expect. The communities cannot plan. They cannot think, “Is it worth us investing in a Starlink? Is it worth us investing in a radio satellite broadband system?” because things are completely opaque. We do not know what they are supposed to do. We do not know what “best endeavours” means.

You spoke about the key systems that can go down and the threat to life that can happen. Are there other systems that are non-life threatening that go down?

Absolutely. The thing that struck me when Rhoda was talking about the timing on Tiree—it was October into November—was that lots of school children were thinking about applying to colleges and universities with deadlines. If that had happened on a UCAS deadline, they could not get their applications in. If it had happened during prelims, they would not have been able to do their online tests. There was an example of somebody trying to complete a house purchase. The bank did not have any comms, so that person had to go to the mainland to complete their house purchase. It was widespread. People could not make their insurance claims, because they could not get online.

Threats to day-to-day life as well.

And this was not just Tiree. This was across all the different reports.

Overall, do you consider current back-up plans from internet service providers to be adequate when a cable is damaged?

Absolutely not. Given the evidence that colleagues have talked about, the problem is the uncertainty of what is going to be provided. We did not know what Openreach was going to provide for Tiree. Other communities have seen what has happened in Shetland and what has happened in Tiree, so they are wondering what would happen to them. There is no minimum resilience standard. We do not know what we are going to get. A team might be sent up. Rhoda talked about the team that came up from Kent. They were incredibly brilliant on Tiree, but that was probably down to the team. Once you get into higher management levels, it is a matter of, “Well, we can’t do this, we can’t do that,” because of the agreements with the Starlink, and I think that is something to be aware of. What can be done? What agreements can be put in place where there could be more Starlinks deployed? We are a small island in Tiree. If it happens on a bigger island, are four units enough? Can we only have them for a week? We need them for five weeks.

I think there are lots of barriers there causing the issues. What alternative measures are needed to ensure continued connectivity on the island when a cable is damaged?

Thirty-four islands have replied. They are starting to put in their own measures. Lots of development trusts in the islands are buying Starlinks as back-up. Lots of individuals are buying Starlinks. For islanders who can have generators, one of the biggest issues is power. When the power goes, the masts do not work. That is fine if you can afford it. Tiree was fortunate in having personnel experienced in connectivity, so they could step in. If it happened on the neighbouring island of Coll, I am not sure they would have the same expertise. They did not have the systems in place. I think we are in danger of an economic divide, as well, between those who can afford to have a generator, to have a back-up, and to have Starlink. I was able to get a different data plan very quickly. Not everybody can afford to have a full data plan to take its place. There is definitely a danger of a divide between those who can and those who cannot. Some community trusts have money, and have the money to pay for staff to help sort things out while these issues are going on. On Tiree, we were fortunate to have a large trust that could manage and oversee things. If you are in a small island with just a community association, you do not have that back-up.

You can see the barriers there again, financially for people.

Absolutely.

Thank you very much for that.

A supplementary from Elaine Stewart.

I am just looking at what the associated costs are and why there would be that disparity between being able to have Starlink and not?

Lots of people looked at getting Starlink during the outage in Tiree, and they have in other islands. You have the up-front cost of £299 for the rectangular dish, or you could have a free dish with a contract, so you are committed to paying £70 a month for a year. You could change your data plan, but then you would need a full data plan. If you have children doing schoolwork online, they each need access. The extra cost is significant.

It can be quite a lot of money for some families.

I would like to make the point that when Mhari talked about compensation, it is not that we are grasping for money. It is literally to compensate for those costs. So if you have had to take out an extra data package, you need to get that money back to cover it.

The last question comes back to me, and I would like to ask it to each of you: if there was one thing you could do to ensure that the impact of subsea cable damage, however it is caused, is not as severe for islanders, what would it be? I will start with you, Mhari. Hopefully you will all have different things. You can sneak in two.

Ann and Rhoda, don’t say the same one as me, okay! [Laughter.] I think there actually need to be more cables. If there were more subsea cables, if we are all serious about there being more subsea cables and about having routes and different fallback positions when cables do inevitably get damaged, that certainly would improve Shetland’s prospects going forward. There are so many, though.

Maybe some routine maintenance to go with it, by the sounds of it?

To be fair to Faroese Telecom, routine maintenance does happen. With my poor, weak ISP, there was also an incident in 2023 when it went off. That was Faroese Telecom doing some regular maintenance. They let all the ISPs know that that was going to be happening, but the ISPs failed to pass on the information to the customer.

Thank you. I think that has given us another one, about communication, so thanks for that. Rhoda?

Critically, for me, we have to try to eliminate single points of failure. Islands should not be on a single spur. We should be looking at loops to ensure that there is a back-up plan. The alternative is some sort of planned failover, so that if we do not have a loop system, it should be mandated that there is either a microwave or a satellite failover to these exchanges. My last one, if I may squeeze it in, is that emergency calling resilience has to be taken very, very seriously in our rural and island areas. Even if your exchange and the ISPs have a satellite failover, there should be a separate backhaul route, be it satellite or whatever else, and power to masts are run by the emergency services network.

Thank you. Ann?

I think my one thing is more general, and it is about asking for an assessment of the impact of failure on all these cables. Say there are 35, 36 cables out there, and we have no idea what might be happening. I would like an assessment of whether there is a back-up if things fail, what is needed and what would be put in place.

Thank you. I think there is a supplementary from Angus.

It is about this question of the telecoms, the 999 and the internet. You said that the electricity came back on again, but the 999 did not come back on again, and I assume from that that the 999 comes down a digital line rather than an analogue line. This comes up in casework to me the whole time. People are concerned that 999 will not work if the internet stops working. Can you just clarify that 999 requires internet?

There are probably people better qualified than me who can clarify the exact position on 999, but as I understand it, your 999 calls over an analogue line can be made without any difficulty at all, as long as your incoming is analogue. One thing that we have not discussed this afternoon, and it is probably part of a wider discussion, is the switch-off of analogue, and I would suspect that that is where a lot of people’s concerns are coming from. You can just pick up the phone, it is linked to the exchange and you can make your phone call. Your digital voice works on electricity. Everything requires power. It needs your router to have power. It needs the handset to have power. That is the big concern. So it is not even the cable. Assuming the cable is fine, a power cut could render you unable to make a phone call, particularly older people who may not instantly reach for their mobile.

So, with 4G and my mobile phone, we have a power cut and the power comes back on again. In terms of my mobile phone—it is not using the radio waves, because it is not using internet in order to function—the mobile phone system clearly stopped working in Tiree—

Your mobile phone works from 4G, and 4G is a form of internet.

Okay, thank you.

Thank you all very much for your evidence this morning. It has been extremely helpful, and has raised some points we had heard before, when we visited the islands, and some we had not heard before. It has been very helpful in reinforcing some ideas, but also in raising some new points to us. So thank you all for making the time to come down. I am glad you got here safely, and I hope you get back equally safely. I also I hope we do not hear about any problems on the islands with connectivity for a wee while, at least. Thank you all very much. Examination of witness Witness: Páll Højgaard Vesturbú.

Welcome, Páll, to this meeting of the Scottish Affairs Committee, where we are looking at digital connectivity in Scotland, but particularly in our islands. We are very grateful to you for being with us virtually today. Could you very briefly introduce yourself and your role, please?

Yes. Good morning. My name is Páll Vesturbú and I am the managing director of the infrastructure part of Faroese Telecom, which operates the local infrastructure here on the islands, and also the SHEFA-2 cable.

Thank you very much. Can I begin by asking you how many times SHEFA-2 has experienced damage since it entered service?

SHEFA-2 entered service in 2007, and unfortunately we have had many breaks. We have had over 20 breaks during the period.

Are those breaks categorised into severe and minor issues?

Yes. Usually, when we have these breaks, they are so severe that we need a repair ship to enter the site and do a repair.

Thank you very much for that. That is very helpful. I will pass you over now to Mr Angus MacDonald.

Páll, we have been hearing from people in Shetland, Tiree and the other islands in Scotland, which you probably listened in on. We were talking about fishing and anchoring being two of the main risks in terms of cable damage. Is this a new thing, and is it something you can see a solution for?

If I just can explain our experience with the faults we have, a large number of those faults have been caused by fishing vessels. During the first years, there was nothing really to do about that; we could not find the vessels that did it. But in the last years, we have been able to obtain some data from the Scottish Government, which helps us to identify the vessels. Usually the vessels that have caused the damage operate with the AIS system turned off, but all fishing vessels have another system, which is the VMS, the vessel monitoring system, which is used by the fishing authorities to keep track of the vessels. We have been able to get information from a freedom of information request to the authorities.

I might have missed this, but do you have the ability to fine them? Or what can you do to encourage them not to?

If I can just say, most fishing vessels operate with the AIS turned on, as they should. There are some very few vessels, actually, in our experience, that deliberately turn the AIS system off for some reason. Usually, those are the vessels that have caused damage to the cable. When there is no AIS system, it is very difficult to find out which vessel did it. One thing we would like is for there to be better enforcement of the AIS system regulation. Then, there is the other system, the VMS system, which the vessels usually, or I guess always, have turned on, because it is more serious if that is turned off. It is then possible to track the vessel. We do not really get information about what vessel did it, but from that information, we can usually, after some time, track what vessel did the damage.

Are they Faroese or Scottish boats? Do have we any way of identifying them? It is not done on purpose, I presume.

No, we do not presume that it is done on purpose. It is accidental. Usually it is Scottish boats.

Thank you.

It is been reported that part of the SHEFA-2 cable was not buried deep because the risk of damage was thought to be low at the time. Is SHEFA looking to rectify this?

When SHEFA was laid, the route was selected between the Faroe Islands and Shetland, and then Orkney and Scotland. The purpose of SHEFA was to serve the users here in the Faroe Islands. That was the main purpose of SHEFA, but then we could also connect the islands. The route seems to be quite challenging. The cable originally was buried in all the positions where it was possible to have it buried. Then there are challenges on the route regarding burial, and also the seabed conditions can change over time. We have experienced that.

Burying them deeper has challenges due to the seabed. Are we looking at burying them deeper?

Yes. It is very difficult to bury an existing cable deeper. When we have a repair, the vessel that does the repair has some jetting equipment, with which you can jet the cable into the seabed, but when there is hard, rock seabed, it is very difficult to do anything. The cable has to be in that location, and that is also the reason why the cables are very well marked on all the charts, and all the information is very well distributed to all the fishing environments.

And that would be the main challenge, do you think, to burying it deeper?

It would be the main challenge to burying it deeper all over the route where the SHEFA is today, yes.

Thank you.

Hi, Páll. Following the latest damage to SHEFA-2 in October 2025, it was reported that repairs could take up to three weeks. How long did it take in the end?

The last repair, which was caused by bad weather, was close to the shore, and it was complex because of that. So it was a more complex repair than usual. There had to be a lot of preparation work on the site, on the beach, to prepare a new landing. So the repair took about a month.

Okay, thank you for that. While Shetland and Orkney experienced significant disruption during these repairs, it was reported that those on the Faroe Islands did not lose internet connection. Why was this the case? Páll Højgaard Vesturbú: That was because there is another cable system also serving the islands here, the Faroe Islands. That was working during the disruption of the SHEFA-2 cable. The other cable served the local industry here.

Thank you for that.

Páll, in your written evidence to the Committee, you mentioned that repair timelines depend very much on how quickly vessels and equipment can be mobilised. Could you walk us through this process and explain why delays occur?

Usually a repair, if everything works well, can take about a week, or between a week and up to 10 days or maybe two weeks. The things that can influence that are typically if there is bad weather—of course, nobody can do anything about that, and that can cause a delay—and also if the vessel is busy with another repair at the time we have the fault. I think it was mentioned before that there are very few repair vessels, but usually repairs take a week to 10 days.

Does your company have its own repair vessels, or does it have to rely on others?

We have an agreement with a company that does this kind of service. We have a vessel allocated, so usually it can go out if it is not busy doing another repair.

Are there any steps that could be taken to ensure that repairs take place more quickly?

It is a bit difficult. You asked for a walk-through of the process. The vessel has to do some mobilisation. The vessel usually is quite ready to leave the harbour. Our vessel is located in France. It takes some days to sail to the repair site and then to pick up the cable, do the jointing of the cable and put in a new piece of undamaged cable. All this process typically takes around one week or 10 days.

Thank you very much. The next question is from Elaine Stewart.

You say it is operationally complex to conduct routine inspection of the cables. Why might this be?

It is a very large system going all the way from Faroe Islands down to Scotland, and it is a very costly affair to do a routine inspection. Sometimes we do more targeted inspections of areas that we think are critical. That is the reason.

You have also said that the risk is not equal everywhere. Are you confident that you have sufficiently identified current hot spots?

Of course, we know the hot spots. We know where we usually have, or have had, defaults, but unfortunately there can also be new locations. Seabed conditions can change, and fishing over the cable can also disturb the condition of the cable. Usually we know approximately the areas where the risk of faults is.

Given that hot spots can change over time, how are you monitoring and adapting to this change?

Regarding monitoring, we have a service. There is a company that provides a service for us, monitoring the cable. It is monitoring all the activity of vessels around the cable, and it contacts those that seem to be operating near it to warn them about risks. But, of course, they are using the AIS system for surveillance, so it is necessary that the vessel has the AIS turned on.

Thank you.

A supplementary from Angus MacDonald.

In our previous discussion, before you joined the call, we spoke about how long the cables last and about the fact that power cables are armoured and seem to be much more resilient. Witnesses were talking about digital cables being 12 years old, and they were worried about their resilience and the damage caused by, for example, rubbing against rocks. How long do you expect these underwater digital cables to last?

Usually we estimate that the cable can last 25 years. That is the historical lifetime of the cable. That has also been connected with the technology in the cable—the technology itself. In some cables, there is technology—there are repeaters—repeating the signal at certain distances in the cable. But in our case, the SHEFA-2 cable does not have repeaters; it is just a cable from end to end, with no active equipment. So, in principle, if the SHEFA-2 cable was intact, without too much damage, it could last much longer than 25 years.

Thank you. My last question is really a bit different from the discussion up to now. The Faroes have 5G everywhere, and we are very jealous of that. What did you do to make a decision to make that happen? Is it financially viable? Do you think it could be replicated in remote Scotland?

What I can contribute to that is this. We had two companies here, each having their own mobile network for some years and building out the network over the islands. It was very well built out all over the islands. But when the decision was made by the owners of the networks to install 5G, they could see that it would be very expensive to have two parallel networks delivering 5G services. Maybe one lesson, if we could contribute anything, is that for small societies with not so large populations, it is very important to focus on resilience for the networks, and maybe not so much on competition on the infrastructure, which can be done more in places where there are larger populations. I think that contributed. When they started to build out 5G, it was decided just to build one network. Then one network was expanded, and the other network was decommissioned. I think that improves situations in smaller communities.

And do you have any idea what cost to the Faroes was to become 5G?

It is difficult to put one number on the cost of 5G. Of course, it was an expensive build-out. When 5G was built out here, all the other technologies were built out as well—we had 2G, 3G, 4G and 5G. Now there are talks about decommissioning the older technologies. It is a significant cost. I think it is very sensible to focus on just building one network in small societies.

Thank you.

Thank you very much for being with us this morning. It has been very helpful to hear from you about your experiences of the cable and the issues that it has. We are visiting the Faroes in a couple of weeks, so we might bump into you on our travels. Thank you again. I now close the meeting.