Official Report: Minutes of Evidence

The Chairperson (Mrs Erskine): Hello, Suzie. Thank you for coming to the Committee. This has been a huge piece of work. I can see that from the comprehensive research paper that you have given us. Thank you for all your work on that, in advance of what you will tell us. Suzie, if you do not mind, will you briefly run through the paper? Then I will come to members' questions, as usual.

Ms Suzie Cave (Research and Information Service): Thank you. Members asked previously for the number of residential dwellings that have been given planning permission within an area of flood risk in Northern Ireland. In the paper, we refer to an area of flood risk as meaning an area in Northern Ireland at medium risk of either river flooding in a one-in-100-year chance or tidal flooding in a one-in-200-year chance.

The purpose of the paper is not for it to be an assessment of whether correct decisions were made in these instances, but it will, hopefully, help members when considering the issue of flood risk for potential new owners and new builds. It is not conclusive in any way. The limitations to use of the data are set out in the paper, and, unfortunately, the section on limitations and caveats is probably one of the largest sections in the whole paper.

I will briefly go through the methodology, which is in section 2. This is the first time that these two sets of data have been used in this way. I would nearly say that it was a bit of an experiment to see whether anything meaningful could be produced. I sought advice from DFI Rivers on how best to use the data, because a lot of the flood data came from it and from planning data. We used two sets of published data: one was from DFI Planning for planning applications, which were filtered for residential and approved applications only; and the other was from flood map data for both river and tidal flooding, and those were merged. We used the most up-to-date forms available and published. Those are intersected and presented on the map in figure 1 in section 3.

If you look at the map, you see that there are two colours of dots. The orange dots show approved residential applications from 2023-24. The red dots show the approved residential applications that are located in areas of flood risk. Each dot — orange and red — represents an approved residential application, which may be for one dwelling, a number of dwellings or an extension, depending on the purpose of the application. Out of the 5,848 approved residential applications throughout Northern Ireland, 61 were seen as being located in an area of flood risk. That represents 1% of all approved applications from 2023-24.

Looking at the map, you see that most of the applications seem to be in the Mid Ulster region. Due to the granular nature of the flood data, it can be quite hard to see at the Northern Ireland level. For illustrative purposes, we have provided a breakdown by council in the appendix.

As I said, there are a number of caveats to the data, which are set out in section 3.1 of the paper. One of the limitations is due to the differences in the flooding data that is used: it is "detailed" or "strategic". Strategic modelling is coarser and tends not to be as specific as detailed modelling. Planning authorities use strategic modelling; in fact, the planning policy does not state whether strategic or detailed modelling should be used. If DFI Rivers is conducting a consultation or if a flood risk assessment (FRA) is required, it will refer to the detailed modelling. Given the purposes of this paper, however, we decided to keep it consistent with what the planning authorities use when they make their initial judgement, so this data is strategic.

Another point relates to the planning data that is used. The application locations have been recorded by planning authorities using coordinates within the boundary of the whole site. That does not consider the coordinates to the centre of the buildings themselves. As a result, that analysis could flag applications as a flood risk where the planning application coordinates are in a flood risk area but the proposed building footprint is not. DFI Rivers uses coordinates to the footprint of the building itself. There is a bit of a difference in the coordinates; it is a bit more specific if they are going to the building itself as opposed to the whole site.

Another factor is the change in flooding data over time. DFI Rivers does not update flood mapping in a piecemeal fashion. Its modelling may be challenged on an individual basis as a result of an FRA that has been conducted by a consultant using more complex modelling. The data has not been updated yet and is to be updated in future. So, on that basis, using the current data, it could seem that the application is in a flood risk area, but, essentially, it is not, once whatever is published at the moment is updated. Furthermore, consultations may have been carried out using archived flood data, as I explained. So, there is the potential that some of the planning applications that are identified in figure 1 may not have been considered to be in a floodplain at the time that they were consulted on.

I know that that is quite a lot to get through. However, I have provided a few considerations, even on the basis of those limitations. The paper is purely a snapshot, using the most recent data. It does not provide for a trend over time, and it may be of use to ask the Department and planning authorities whether, in their experience, it reflects a change and whether there has been improvement.

In figure 1, 1% of applications were flagged as an issue using strategic floodplain data. However, due to the caveats that I have explained, such as the use of detailed modelling, building-specific coordinates and potential changes in flood map data over time, some of those applications may not be or have been considered a flood risk issue. This exercise was difficult, because of those caveats. Therefore, should there be a more consistent approach in the use and update of data to help streamline the process and to help with analysis? Would that require more resources, particularly for planning authorities and DFI Rivers? What do insurers and mortgage lenders use to base their assessments on? Can it cause problems for owners if a planning decision is finally granted using detailed mapping, but insurers use strategic mapping and consider it a flood risk?

I move now to some reasons why the applications were perhaps given approval. Due to the time frame, we decided to pick, at random, one application from each council area to look at the reason behind the approval. Essentially, we have looked at 10 out of the 61. Belfast was not included because it did not have any applications within a flood risk area. I have provided a summary on why they may have been given approval. An application may have been considered by the planning authority as an exception. That is explained in my paper with reference to the strategic planning policy statement (SPPS) and planning policy statement (PPS) 15. An exception could be due to the replacement of an existing building, for example. In those cases, DFI Rivers advice may have been taken as a material consideration in its overall decision-making process, with appropriate mitigation provided. In some cases where an application was considered contrary to policy, it was balanced against other material considerations and factors, such as the restoration of an historic listed building, as it was in one case. It was provided with protections under the Planning Act (Northern Ireland) 2011.

It is also the case that part of the site may lie within a flood risk area, but the dwelling itself does not. In some cases, conditions and informatives were added to the approval in relation to the part of the site within the floodplain — a garden or access lanes, for example. In some cases, the strategic flood map indicated that the site was within a flood risk area. However, more detailed and accurate modelling that was carried out under, say, the flood risk assessment or by DFI Rivers showed that the site did not in fact lie within the floodplain boundary.

An application may have been granted approval at outline planning stage. If a request for an FRA or drainage assessment is not made at that stage, later requests to revisit it at reserved matters could not be granted. According to planning authorities, the principle of the house could not be revisited at reserved matters. As such, an application considered as being in a strategic flood risk area may be approved.

I will look now at some of the considerations. Given the sample size used in the paper, it may be of interest to ask the Department and planning authorities whether the reasons identified in the examples are representative of their overall experience or whether there are other reasons that may be of interest.

As mentioned in some examples, it appears that implementation of the flood risk assessment and mitigation measures are the responsibility of the developer. If specific measures from the FRA are not written into conditions attached to an approval, who ensures that those are carried out and carried out to a satisfactory standard?

The planning authority has the final say, and DFI Rivers provides advice as a statutory consultee, but what happens in cases in which DFI Rivers might not be fully satisfied with an application or the information given, yet approval might still be granted?

If an application is considered to be contrary to planning policy, such as the SPPS or PPS 15, what is the decision process? Where and by whom is the final decision made? Do the requirements of the Reservoirs Act (Northern Ireland) 2015 have an impact on applications? For example, in some cases, DFI Rivers was not satisfied with the information provided on reservoir safety, yet permission was granted. If an FRA or drainage assessment was not requested at outline planning stage, can that be revisited at reserved matters stage? If not, why not? If it was missed at outline planning stage, could that present problems down the line? Does that need to be considered further?

A lot of those questions relate to process and planning policy, so members may want more detail, especially on the process to approve such applications and on whether DFI Rivers and planning authorities find the process and policy efficient and effective.

That is a quick run through the paper. I am happy to take questions.

The Chairperson (Mrs Erskine): Thank you very much, Suzie. We appreciate it. Correlating the data must have been a real headache. I can just about imagine what it would have been like to correlate it all and put it together, particularly when you are dealing with the planning authorities and DFI Rivers using different types of data. That might point to some of the issues in the planning system arising from that.

I will ask a stupid question. Has that always been the case? Have they historically used different data for strategic and detailed modelling?

Ms Cave: I do not know how long that has been the case. You probably need to ask DFI Rivers how long it has been going on. I do not even want to guess. It is done that way in the current process.

The Chairperson (Mrs Erskine): OK.

Mr Boylan: Thanks, Suzie, and welcome back. I do not know how many papers you have done for us over the years.

I have a couple of questions. I am still a bit surprised. Mark was around when we did PPS 15, back in the day. How has the process not been coordinated properly? The Department of Agriculture and Rural Development (DARD), I think, did a flood mapping exercise a number of years ago, and we then did PPS 15. You may not have looked at this, in which case perhaps you can come back to us, but how does coordination happen in other jurisdictions? Planning clearly needs to be coordinated.

You touched on insurance. House insurance protection, especially, has to be a major problem for people. We keep having a conversation about there being no coordination with Northern Ireland Electricity (NIE) when people apply to build a house, and this is another prime example. Surely it is up to Departments to have that information. It is about mapping and coordination between the two. It was one Department at one point, I think, and it separated. Have you found better practice in other jurisdictions that we could learn from? Can you comment on the insurance issue?

Ms Cave: Certainly. For this piece of work, I did not look at the process and policy in other jurisdictions. Some of the reports referred to Northern Ireland Environment Agency (NIEA) guidance. Local development plans are still in progress, so it is about seeing how PPS 15 is carried through in those and in local policy plans. Local policy plans will be more detailed and site-specific, so they may use more detailed flooding data. That is not the case at the moment, and no planning policy states whether, initially, they should look at it at a strategic or a detailed level. It depends on how readily available the modelling, resources and expertise are that are required to do that on an ongoing basis. At the moment, if an issue is flagged at a strategic level, they will look at it in more detail, using the more detailed modelling. Yes, I can certainly look to other jurisdictions to see whether something similar is being done.

Mr Boylan: What about the insurance issue?

Ms Cave: Yes. I will flag again that it depends on what information and the type of data that they are getting access to. Flood Maps (NI) refers to strategic data. Again, it could be flagged as a flood risk issue, but then, when looked at in greater detail, it is not.

Mr Boylan: Thank you very much. That was interesting.

The Chairperson (Mrs Erskine): Super, thank you. Does anybody else want to ask a question? Keith, I knew that you would want in, given that Mid Ulster is right up there.

Mr Boylan: There are 4,900 houses out of the 5,000 over in Mid Ulster. [Laughter.]

Mr K Buchanan: I am not going to go into detail. It was 13. I could, but that is for another day. [Laughter.]

My question is about liability. Let us say that the planning authority — for example, Mid Ulster District Council — gives a green light to a planning application in a floodplain under strategic mapping, and that person builds a house and cannot get insurance. Is there a liability in council?

Ms Cave: That is a very good question that I do not have the answer to.

Mr K Buchanan: I do not expect you to know, to be fair. I have seen it before, and I watched it one night on YouTube. Great fella — lovely man — needs a house, but that man will be standing in water one day. That is exactly why that man has passed the house — lovely guy. They are all nice people looking for planning applications, but everybody is standing in water.

Mr Durkan: They are family.

Mr K Buchanan: I appreciate that you may not be able to answer that question. My concern is that, if such applications are being passed when DFI experts are saying that it is a floodplain, regardless of whether it is strategic or the other terminology that is used, it is still a flood risk, and insurance says no. The person could point to the planning authority and say, "You passed my house, but I cannot get insurance." That is more of a comment than a question.

The Chairperson (Mrs Erskine): It is also about the timescale of flood maps changing and applications being in the planning system at that time and things like that. We know about the planning system being elongated and how that can change things in plans as well. That works both ways, I suppose.

Mr Boylan: I am trying to understand whose responsibility it is to update it. Keith made the point about the council giving permission, and I support that point, but who coordinates? There should be one system, should it be used under PPS 15 or others. As part of the stakeholder engagement, those people should have a say in the matter. I have seen it happen time and again. There are more questions than answers. To be fair, Suzie has provided a good paper.

The Chairperson (Mrs Erskine): Absolutely, because it allows us to go back. It will be worthwhile for us to take all the caveats and issues that Suzie has raised in the report to the Department's door to find out what has gone on. We can ask about the liability issue, for example, and that type of thing as well. It might be worth checking up, Cathal, on the Flood Re scheme that is referred to. I have a feeling that that runs out after a period; it may run out in 2036. Do not quote me on that. Flooding is not going away. It would also be worth checking out, again, which flood maps they are using as a result of that. If they are using a different flood map from the one that planning is looking at for applications, that can change things for them. It might be worth scoping that out as well.

Mr Durkan: Thanks a million for this, Suzie. There is a lot in it. I am sorry that I am going to be a pain and ask about what is not in it. I do not want to give you more work: you might have this somewhere else but just not in the paper. I know the level of work that has gone into it.

We see the approval rate of applications in 2023-24 in flood risk areas. Do we have any idea of the refusal rate in similar areas? If you track it, year-on-year — I have nothing to base this on other than a hunch — I would say that you will see a decrease in the approval rate, as councillors and decision makers become more aware of the risk of flooding, which, I suppose, is down to the fact that we are seeing more of it.

When it was introduced, PPS 15 was upheld as being very strong policy — the best planning policy related to flooding on these islands. However, to date, it has not been implemented consistently or properly across council areas. On the approvals, there is an engineering solution to everything, so there are applications that will come in — they could, more often than not, be commercial ones — that can just throw money at a solution to mitigate the risk for not just that application but existing surrounding properties.

Ms Cave: You talked about the refusal rate. The data is there; it shows whether an application has been approved or refused. That is something that I could provide. I did not provide it in the paper, but we now have a bit of a methodology. This was nearly like a pilot that we could use for different types of the data.

Mr Durkan: There are probably fewer applications. There have been no approvals in Belfast. Is that down to fewer applications, or just the fact that those flood risk areas are already saturated, for want of a better word?

Mr Boylan: Suzie mentioned in the paper that the way forward is going to be the local area plans. They are going to be key. They are identifying areas at the minute. Mark is right about PPS 15. Councils have autonomy to make decisions, but, if we do not get any level of consistency, you will continue to have this issue.

The Chairperson (Mrs Erskine): That also concerns me. DFI Rivers, which is a consultee, uses different maps and data from what the planning is going to decide on.

Mr Boylan: Eventually, those plans go to DFI; the final plan will be assessed by DFI. However, at least we have started that conversation. When local area plans are ready to go, they will be overviewed by DFI. One of the questions that we need to ask is about the relevant data and how it is updated. That is why I asked about how it is updated in other areas. I do not know.

Mr Durkan: There is also a difficulty with accessing and interpretating — interpreting, sorry. [Laughter.]

Mr Boylan: That is a new word, Mark. We will use that.

Mr Durkan: It is difficult for applicants and objectors to interpret those ever-changing maps.

Ms Cave: Yes. The level of expertise that is probably needed to conduct the detailed modelling means that it is something that could not be run against every application that is made; it would be only when it is flagged strategically, and then it can be passed to DFI Rivers to investigate further through its consultation.

Mr Boylan: There is a good bit of information out there already, to be fair. It is quite interesting, Chair.

The Chairperson (Mrs Erskine): Yes, it is. Suzie, I think that that is all for now. Thank you so much. Thanks for using us as a pilot. We will probably see you again. Thank you for your time and your work on this. We really appreciate it.