Roger Brookhouse & Anor v The Environment Agency

This is the Tribunal’s final decision in Mr and Mrs Brookhouse’s reference under the Water Resources Act 1991, and in the arbitration of a contractual claim under a deed dated 6 November 2014. In a decision dated 30 November 2023, Brookhouse v Environment Agency [2023] UKUT 282 (LC), the Tribunal found that the claimants’ property has been damaged, and will continue to be damaged, by raised groundwater levels caused by works carried out by the respondent, the Environment Agency. That decision (“the 2023 decision”) was only about liability; no determination was made at that stage about the quantum of compensation payable to the claimants. A further hearing took place over eight days in March 2026, and this decision about quantum is made in light of the evidence we heard at that hearing.

The claimants were represented at the quantum hearing (not at the liability hearing in 2023) by Mr James Pereira KC and Ms Caroline Daly; the respondent continues to be represented by Ms Galina Ward KC and (at the quantum hearing) Ms Natasha Jackson.

This decision is not self-contained; it assumes knowledge of the 2023 decision, and cannot be understood without first reading that decision. To summarise briefly: in 2014 the respondent built a flood defence wall (“the new wall”) along the river frontage of the claimants’ home, The King’s Lodging (“the property”) in Sandwich, Kent. It was not a continuous wall across the whole frontage because the property has a historic dock, whose wall meets the new wall at a right angle. On a number of occasions in the next few years the garden of the property was subject to flooding, with water entering from below ground. The respondent made valiant attempts to prevent further flooding, but at the hearing in 2023 we heard evidence that the respondent’s own consulting engineers had agreed with the claimants’ expert that a “flow path” had been established beneath the wall of the claimants’ dock, and that remediation work by the respondent had addressed the symptoms of the problem but not the cause. In the 2023 decision we made the following findings:

This decision about quantum is made on the basis of those findings; the 2023 decision remains unappealed. We have now heard new expert evidence, some of which called those findings into question. We discussed the implications of that with counsel at the hearing, and it is uncontroversial that the experts who gave evidence to us in March 2026 were not only entitled but required to give us their true opinion, and cannot be criticised for disagreeing with anything in the 2023 decision; as witnesses they are not bound by those findings. If the evidence now showed on the balance of probabilities that we were wrong, the claimants are nevertheless entitled to a decision on the basis of what was decided in the 2023 decision and the Tribunal could not go behind that decision unless both parties agreed. However, as will be seen, that is not the case, and our decision proceeds straightforwardly on the basis of what was decided in 2023, supplemented by our findings on the extensive fresh evidence we have now heard.

This decision falls into three parts. In the first part we provide some context by summarising what has happened since the 2023 decision, setting out the parties’ positions about quantum, and deciding what would have been the current value of the property if the respondent’s works had not been carried out.

In the second part of the decision we determine the compensation due under the Water Resources Act 1991 for the rest of the damage to the claimants’ property.

In the third part of the decision we determine the compensation due under the terms of the Deed, relating to the garden and the swimming pool.

Two and a half years have passed since the 2023 decision. One reason for that delay will be obvious in reading this decision, namely the extent of the expert evidence called by the parties. Altogether seven pairs of experts provided reports for the 2026 hearing, in the disciplines of water and drainage, architecture and heritage, structural engineering, horticulture, swimming pools, valuation, and quantity surveying. Another reason for the delay is that very little progress was made between the handing down of the 2023 decision and the early summer of 2024, because the claimants changed their legal representation.

Meanwhile nothing has been done to restore the house and garden, inevitably, because the groundwater problems caused by the work cannot be remedied until the parties have this decision. There have been no more extra-high tidal inflows. The respondent helpfully provided data from their tidal records, which give readings every 15 minutes, and in the time since the 2023 hearing only 20 days where the tide exceeded 2.9 metres AOD have been recorded, for a maximum of 6 readings and mostly only for two or three readings; only on 6 days did it exceed 3 metres, never for more than 4 readings, and only three instances, on two days of tides over 3.1 metres. So the absence of excessive inflows is unsurprising, and the respondent does not suggest that the conditions that have previously caused those inflows have occurred since 2019. At the 2023 hearing the hydrology experts, Mr Evans for the claimants and Mr Groves for the respondent, agreed that extra-high tidal inflows will recur in the future at very high tides, and it is of course likely that tides will generally get higher as time goes on (paragraphs 34 and 164 of the 2023 decision).

Some progress was made at a case management hearing in June 2024, and after that two significant developments took place.

When they made their reference to the Tribunal in 2020 the claimants’ position was that the proper measure of compensation was the diminution in the value of the property, on the basis that the cost of remedial works might exceed the value of the property, that they could not reasonably be expected to carry out the extensive remedial works themselves, and that that measure would remove the need to quantify every single defect, saving the Tribunal and the parties time and cost. The sum claimed was not quantified but was expected to exceed £500,000.

Following the case management hearing in June 2024 the Tribunal gave the claimants permission to amend their Statement of Case; paragraph 62 of the amended pleading says this:

“Full compensation for their injury is claimed on the basis of (a) the cost of repair to the Property and its equivalent reinstatement to its condition prior to the damage caused by the Respondent’s works; (b) the cost of measures needed to protect the Property from further damage; (c) any diminution in the value of the Property consequent upon the Respondent’s works; (d) other consequential losses (including fees and costs) caused by the Respondent’s works and the pursuit of compensation, where these are foreseeable, not too remote, and not unreasonable.”

The Tribunal raised an eyebrow at the pleading of a claim for both the cost of remediation works and the diminution in value of the property, but in the end it has not been necessary to explore that curiosity. What the claimants sought in the quantum hearing was the cost of remediation, including both the reinstatement of the property to its pre-works condition and its protection from future damage, together with other consequential losses such as fees and costs, including pre-reference costs. The respondent has for the most part accepted that that will be the measure of compensation, with the proviso that compensation for the cost of restoring the pre-works groundwater levels should not exceed what would have been the current unblighted value of the property (i.e. what it would have been worth absent the respondent’s works). As will be seen, the cost of restoring the water levels does not exceed that value and so, again, a difficult legal issue does not arise.

In the 2023 decision we explained that prior to commencement of the works the parties entered into a deed dated 6 November 2014 (“the Deed”). Its terms addressed a number of preparatory matters – for example, the respondent covenanted to carry out a drains survey before the work started – and also set out obligations in relation to the swimming pool and the garden.

On 18 June 2025 the parties made a reference by consent under section 1(5) of the Lands Tribunal Act 1949 for the Tribunal to act as arbitrator in respect of the respondent’s liabilities under the Deed. The Deed included obligations to restore the garden after the works were completed, and to replace the swimming pool if it was damaged beyond repair, and the claimants ask for the cost of restoring the garden and replacing the pool under the Deed rather than the statute. The respondent makes no objection to these two items being claimed in this way, although it takes the view that its liability in respect of the garden and the pool would have been the same whether assessed under the Deed or under the statute.

As we said above, we have heard evidence from experts both in physical disciplines – engineering, horticulture etc – and in quantity surveying. The point of this decision is to generate figures for the cost of compensating the claimants; but in order to reach those determinations we have had to make assessments, and hear evidence, about causation (in particular the raised water levels at the property, but also for example the cause of various elements of damage to the house). To stay with the most complex issue, the water levels, that means that we have heard two sets of evidence: first from the drainage engineers in order to determine the appropriate solution for the raised water levels, and then from the quantity surveyors who have given evidence about what each solution will cost.

The evidence has been complex, and the experts’ positions have been something of a moving target as discussions and compromises have taken place. Only on 6 March 2026 did the respondents’ solicitors, at the claimants’ request, set out in correspondence the amounts that the respondent accepted it was liable to pay. The parties’ positions both moved to some extent during the hearing itself. The following paragraphs set out the parties’ eventual positions on liability under all the heads claimed by the claimants under both the statute and the Deed. The figures all represent costs that the claimants will have to incur and on which they will have to pay VAT, so that VAT will have to be added as part of the compensation, and in the table at paragraph 24 below we have added VAT as necessary. All figures are set out to the nearest pound.

Looking first at the statutory claim, first, the parties agree that £36,742 is payable in respect of work to the house. The parties also agree that further costs are payable for preparatory work and other costs associated with the agreed work to the house, which the claimants put at £186,232 and the respondent at £20,775.

The largest item in issue is the cost of work to lower the groundwater levels at the property. The claimants seek the cost of the scheme put forward by their drainage expert (involving the installation of a new dock wall) in the sum of £1,810,803. The respondent argues for the scheme put forward by its own expert (which it says will cost £145,797 but the claimants say will cost £247,662). If the Tribunal is against it on that, then it says that the claimants’ scheme for lowering the water levels will cost £360,410.

The parties agree that the garage will need to be demolished and rebuilt, and that the cost will be £104,884; they also agree that a garden wall needs to be restored at a cost of £11,542. The claimants seek a further £22,095 to reinstate the drainage system, and that is not agreed. The claimants say that £115,030 is payable in respect of investigative costs, together with miscellaneous other costs (which we address in detail below) of £20,500; the respondent says that £94,344 is payable in respect of investigative costs only.

Turning then to liability under the Deed in respect of the garden, the parties agree that £728,828 is payable; the claimants seek an additional £139,169 for some further agreed items, which the respondent costs at £69,274. The claimants seek an additional £302,500 in the event that the Tribunal finds in favour of the drainage solution proposed by the respondent’s expert; as will be seen below, this does not arise.

Under the Deed again, the claimants say the respondent is liable under the Deed to reinstate the swimming pool, at a cost of £213,988; the respondent denies that it is liable to replace the pool, but if it is it says that the cost will be £150,000.

The parties agree that the claimants’ pre-reference costs are payable in the sum of £275,000, which if we have understood correctly includes VAT.

For convenience, here are the figures in tabular form:

Both parties adduced expert evidence about the value of the property as it stands, and as it would have been absent the respondent’s works. It was always anticipated that such evidence would be needed because the claimants initially claimed the loss in value of their property. Following their amendment to their statement of case the two potential bases for a claim under the 1991 Act – cost of remediation and loss in value – stood as alternatives and so a determination about loss in value was still required. We revert later to loss in value; for now, we think that it is important to set the scene for this decision about the amount due to Mr and Mrs Brookhouse by summarising the evidence for what they would have had in the absence of the respondent’s works, i.e. the current unblighted value of The King’s Lodging.

Expert valuation evidence for the claimants was given by Mr Colin Cottage BSc (Hons) MRICS, Managing Director of Valuation and Compensation at Ardent Management Limited, based in London. Expert valuation evidence for the respondent was given by Mr Christopher Francis MRICS, a director in the Valuation Department of Fletcher King, based in Central London. When first instructed Mr Francis was working in Canterbury for Strutt and Parker.

In Mr Cottage’s opinion the current market value of the property disregarding the effect of the respondent’s works would be £1,900,000 and in Mr Francis’s opinion it would be £1,400,000. Their opinions had been expressed in their first reports dated 27 October 2025 and 23 October 2025 respectively, and were restated in their joint statement dated 28 January 2026. The values were assumed to remain relevant at the date of the hearing and we continue to adopt that assumption at the date of our determination.

The valuers agreed a description of the property as a 15^th Century, Grade II listed, timber framed detached three storey dwelling with a two storey annex occupying grounds of around 0.4 acres. The property is bounded by a listed brick wall to its frontage to Strand Street and a river wall to its River Stour frontage. It has the benefit of riparian rights and a river frontage of around 60m, including the width of the 8m long private dock. Within the grounds there is a single storey garage, accessible from Strand Street, and at the other end of the garden there is a swimming pool. The property was altered in the 1950s and the 1970s but retains an abundance of period features.

The total gross internal area was agreed at 5,790 sq ft, including loft/attic space and the boiler room but not the garage. The ground floor accommodation provides a hall, dining room, drawing room, kitchen and annex extension with study. The first floor has four bedrooms with an en-suite shower room to the master bedroom and a family bathroom. The second floor has a single bedroom, with bathroom, a long gallery and attic store. The annex first floor contains a further bedroom and bathroom. In summary the property has six bedrooms and four bathrooms.

It was agreed that the property has an attractive outlook across the River Stour and is not blighted by traffic noise or proximity to other buildings. The boundary wall to Strand Street means that the property is private and not overlooked. The kitchen is described as relatively small compared to other properties of similar size, but the drawing room and dining room are large. Some of the bathrooms at the property require updating and a purchaser would be likely to consider general redecoration necessary.

The experts agreed that the property should be valued using the comparable method and that amongst the range of evidence considered the most useful comparable evidence was provided by three sales of period properties in Sandwich town centre: Haven House, Pellicane House and Stuart House. The sale prices were adjusted to October 2025 using house price indices. It is regrettable that the experts could not agree on the most appropriate index but, in the context of rounded market values, the scale of difference is small and we use the adjusted figures provided by each expert for their preferred evidence.

The table below sets out the key features of the three preferred comparable properties, and analysis of their sale prices, together with the respective opinions of value provided by the experts. The evidence is presented in descending order of value or adjusted sale price.

Mr Francis maintained doggedly that the value of the property should be less than the sale price of Stuart House. Originally the figure of £1,550,000 was the amount for which Stuart House was known to be under offer in May 2025, but the sale had completed in December 2025 at that figure, which could therefore be used without adjustment for October 2025. Mr Francis could not explain why a purchaser would pay £150,000 less for the property, which is 42% larger and has a riverside location. His main argument was that the condition of Stuart House was better, but this did not stand up to scrutiny.

It is apparent from the table above that Mr Francis’s figure of £1,400,000 is an outlier in terms of both capital value and value per sq ft and we conclude that we can give it no weight.

Mr Cottage used an analysis of sale price per sq ft to derive his value of £1,900,000 for the property. He relied mainly on the evidence of Haven House and Pellicane House, which he considered to be directly comparable listed houses of similar size and type to the property, although both lacking a riverside location. Pellicane House had sold after recent restoration so would justify a higher value per sq ft than the property. Haven House was similar to the property in most respects, including plot size and condition, but less private having ground floor windows facing directly onto the street. In Mr Cottage’s opinion the value per sq ft of the property would fall between that of the two comparables at £328/sq ft, giving a value of £1,899,120 or £1,900,000 on rounding.

When asked to comment on how a purchaser would justify paying over £250,000 more for the property than for Haven House (an uplift of 16% on the adjusted sale price) Mr Cottage said that the current Land Registry House Price Index would increase the adjusted sale price of Haven House so that the difference today would only be in the order of 13%. This seems to miss the point that no request was made to update the valuation of the property, to account for movement in the market between October 2025 and the hearing, so it is not appropriate to make a comparison with an updated adjusted sale price. Mr Cottage emphasised that the additional 667 sq ft of floor area provided by the property was the equivalent size of a modest two-bedroom terrace house. This comment had not appeared in Mr Cottage’s reports and seemed to be more a defence of his valuation based on £ per sq ft than a comment which would have relevance in the market for large historic houses.

Mr Cottage acknowledged that the kitchen at the property is small in comparison with the other two properties, and agreed that a purchaser would be likely to undertake works to extend it through into the annex. He also acknowledged that the 2014 Ridge report, made before the respondent’s works, suggested that there were maintenance issues to address at the property beyond general redecoration, which a purchaser would need to factor into their bid for the property. The condition of the pool was one of those issues, but he had not made a specific allowance for that. Mr Cottage considered that overall the maintenance burden would be similar for any listed property of this size and type.

Looking at Mr Cottage’s valuation in the context of the most relevant comparable evidence, set out in the table above, we conclude that he has been generous in adopting a value of £328/sq ft. We consider that a value of £320/sq ft, equivalent to that achieved at Haven House, is more appropriate. This would give a market value of £1,850,000, which is £210,000 more than the adjusted sale price of Haven House to reflect greater size and a better location, while taking into account the need for expenditure to upgrade a small kitchen, and a pool near the end of its life

We therefore determine that the market value of the property today, ignoring the respondent’s works, would be £1,850,000.

It will be apparent from what we have said so far that this decision may be felt to be disappointingly lacking in legal interest. It is a decision made on the basis of findings of fact and judgments of what, in the context of this unique property, has to be done to give the claimants the “full compensation” to which they are entitled under the 1991 Act (see paragraph 42 below) and to give them a sum of money that will fund the performance of the respondent’s liability under the Deed. It creates no new law. Nevertheless it is of crucial importance to the claimants, for whom the property is their retirement home and whose lives have been disrupted for over 12 years now by the works and their consequences. We acknowledge their distress, as does the respondent, and we reiterate the view we expressed in the 2023 decision that the respondent has done its best and has acted fairly throughout. It is the Tribunal’s task to ascertain, without emotion, the compensation due to the claimants.

Section 165 of the Water Resources Act 1991 gives the respondent power to carry out works relating to flood defence and drainage. Section 177 gives effect to Schedule 21 to the Act, of which paragraph 5 provides:

“(1)

Where injury is sustained by any person by reason of the exercise by the appropriate agency of any powers under section 165(1) to (3) of this Act, the appropriate agency shall be liable to make full compensation to the injured party.

(2)

In case of dispute, the amount of any compensation under sub-paragraph (1) above shall be determined by the Upper Tribunal

The parties agree that there is no magic in the term “full compensation”, and that the Tribunal’s task is to award a sum that will “put the [claimants] fairly and reasonably in the position in which [they were] before the damage occurred, so far as a pecuniary award can do so” (Scott J in Ward v Cannock Chase District Council [1986] 1 Ch 546, at 576, quoting Henchy J in the Irish Supreme Court in Munelly v Calcon Limited [1978] IR 387. There was some disagreement between the parties as to the state of the authorities in the event that the Tribunal has to decide whether to award the cost of remediation or the loss in value of the property; but the respondent agrees that the Tribunal is to award the cost of remediation in any event under all heads but one. On one point, namely the work required to restore the groundwater levels in the garden, Ms Ward KC explained that the respondent’s position is that if remediation costs of that item alone exceeds the loss in value to the property then compensation should be limited to the latter. We emphasise that the respondent made this argument solely about the cost of restoring the groundwater levels in the garden; as to the rest of the work the respondent accepts in any event that it must pay the cost of remediation.

There was some discussion at the hearing as to whether loss in value equates simply to the current unblighted value of the property minus the cost of the works needed to restore the house and garden as determined in this decision – which as we shall see brings the current value down to zero so that the loss in value is equal to the current unblighted value – or whether the property might retain some residual value even in its unremediated state. In closing Ms Ward KC confirmed that the respondent’s position is that the loss in value of the property is the cost of the works of restoration required. As we shall see, the compensation awarded in respect of the restoration of groundwater levels, including VAT, does not exceed the current unblighted value of the property, and therefore the respondent’s argument that loss in value should be paid instead does not arise.

The compensation to be awarded under the statute falls under three heads. The first is the cost of remediation of the damage to the house.

The second is the cost of restoring the groundwater levels in the garden to their pre-works status. Our discussion is set out in two sections because we have to go into some detail first to decide which of two competing schemes of work is the reasonable course to adopt, and second to decide how much that scheme will cost.

Under the third heading we group together a number of miscellaneous costs, which are rather less complex.

First we make two preliminary points about the sums to be awarded, and we comment on our assessment of the two quantum experts – all of which is equally applicable to the statutory claim and to the claim under the Deed.

The first point is about inflation. Generally where damages are awarded to represent costs that are not to be incurred immediately, no allowance is made for the fact that those costs will be higher when eventually incurred, because the recipient will have been able to invest the money in the meantime and earn interest on it. Despite that the quantum experts have added an allowance for inflation for some costs, presumably because they were asked to calculate the present cost of future work; we would have expected the respondent to discount the resulting figure for early receipt. But Ms Ward KC confirmed in closing that the respondent agrees that an allowance for inflation can be added, subject to issues about how it is calculated. That is a generous concession on the respondent’s part; we have not added all the inflation figures together, but it is easy to see for example that the respondent’s allowance for inflation amounts to nearly £50,000 for the garden costs (paragraph 211 below). We have therefore included an allowance for inflation where the two experts agree a particular figure that includes inflation, or have agreed that a sum is to be added but disagree about its calculation. Where one expert adds a figure for inflation and the other does not we have not done so.

The second point is about rounding. The expert witnesses have predicted costs to the last penny, for various reasons. In recording the experts’ evidence we have ignored the pennies. Where we have had to take a view about a particular cost we have given a round figure, and we have applied some rounding to our totals.

Turning then to our assessment of the quantum experts, who have given their opinion about the costs of almost all aspects of the work to be done (the exception being the garden and, on Mr Steel’s part, the swimming pool). Both are highly qualified quantity surveyors. They appended to their final joint statement a Scott Schedule setting out, first in summary and then in detail, the costs agreed and in dispute, together with Mr Dunne’s and Mr Steel’s workbook where each sets out their own thinking.

There are some important differences between the two experts. Mr Steel, unlike Mr Dunne, is a specialist in dispute resolution and does far less actual project work than does Mr Dunne; his most recent project was 12 months ago and was an aviation project. He used price books for his source of prices, mostly the Building Cost Information Service (“BCIS”) and occasionally Spon’s Architects’ and Builders’ Price Book (“Spon’s”), whereas Mr Dunne used a combination of quotations, price books, and his experience of previous projects.

Mr Steel tended to view the works as less complex than did Mr Dunne, and he generally did not make allowance for things to cost more than usual because of the age and consequent fragility of the property; we think that was the result of his having less practical experience than Mr Dunne and of his not having visited the property, nor even been to Sandwich. It led to many instances where Mr Dunne identified and costed specific items which Mr Steel regarded as wrapped up in another figure such as professional fees or contingency. In some cases he omitted things for which special provision was needed, and in others he had no regard for the fact that if certain contingencies were to happen he clearly had not made enough provision; for example, he allowed for a jack-up barge to be used to re-build the dock wall, but because he did not know how many weeks it would be needed for he only allowed for one week. For all those reasons, while we do not reject Mr Steel’s evidence, we view it with some caution.

Mr Dunne was exceptionally well-informed and took far more trouble than did Mr Steel to find out about matters of which he did not have experience. He was keenly aware of the heritage considerations in all aspects of the work to be done, indoors and outdoors. Nevertheless we view his evidence too with some caution because whereas Mr Steel tended to under-egg the pudding, Mr Dunne generally went too far in the opposite direction. He was conspicuously the voice of prudence, making provision for every possible eventuality and then adding contingency on top. To use a metaphor that came up a number of times in cross-examination, he advocated Rolls Royce provision.

With those comments in mind we look first at remediation for the house.

Remediation for the house

The house is hundreds of years old. In the 2023 decision we heard evidence that damp, and some cracking, were present before the respondent’s works were carried out. But we accepted evidence that the raised groundwater has caused damp to rise within the wooden framework of the house and has affected the fabric. For the 2026 hearing the parties instructed experts in architecture and conservation, Mr Franics Maude of Donald Insall Associates and Dr Chris Miele of Montagu Evans LLP to give evidence about the extent of the damage caused to the house by the respondent’s works and what needs to be done to fix it. The evidence of Mr Maude and Dr Miele was agreed and so they were not called at the hearing.

The work that it is agreed needs to be done includes the replacement of some external stonework and brick, and the tying back of the east wall of the house where it has bowed outwards; the restoration of timber posts in the hall which have been damaged by water at their base, together with the replacement of some panelling and plaster; replacement of Yorkstone paving and some panelling in the dining room; repair of several fireplaces; replacement of panelling, plaster and tiling in the drawing room and bedrooms.

The quantity surveyors, Mr David Dunne of Baqus Construction and Property Consultancy for the claimants and Mr Duncan Steel of Turner & Townsend for the respondents, gave evidence on the cost of that work. They have broken down the costs in some detail by reference to Mr Maude’s and Dr Miele’s specific recommendations, so for example they have costed each fireplace, plastering for each of the rooms concerned, tiling where necessary and so on. The differences between them are not all in one direction; for some items Mr Dunne assesses a higher cost, for some Mr Steel’s costing is higher, and in the end there was very little between them as to the work itself. In light of that evidence the parties have agreed a sum of £36,742 for the work itself. We refer to that sum as the “core cost”. It is surprisingly low. It is not difficult to spend that much on an ordinary urban property needing a new roof or kitchen.

However, the experts disagree about the associated costs of that work – preliminaries, surveys, professional fees, contingencies and so on; the claimants put these costs in total at £186,232, nearly six times the core costs, and the respondent at £20,775.

The parties also instructed structural engineers, partly to assess the damage to the house but also to give evidence as to the effect on the house if water levels in the garden were not restored to their pre-2014 levels. The claimants instructed Mr Peter Corcoran of The Morton Partnership, who has had a long involvement with the property which he first inspected in August 2014. The respondent instructed Mr Stuart Hookham of William J Marshall & Partners. Because of the parties’ agreement about what needs to be done to the house we do not need to say anything about their evidence about the damage to the house. And at the end of the hearing the respondent abandoned its position that nothing needs to be done to return the water levels to their pre-works level, and therefore we do not need to say anything about the structural engineers’ evidence.

Accordingly it remains only to consider the evidence of the quantum experts about the associated cost of the £37,000-worth of work on the house. The difference between them is striking, and it arises from some differences in overall approach.

The first is methodological. Mr Steel started from the agreed £36,742 cost of the work and added four sums calculated as percentages of the cost of the work: 15% for preliminaries, 10% for overheads and profit, 10% for contingency and 12.5% for professional fees to arrive at his additional £20,775. Mr Dunne provided for eight additional sums: the four items provided for by Mr Steel, and four additional sums. Only two of those eight further sums were calculated as percentages; the rest were individually itemised.

The reason for that difference in methodology is, we think, one we have already highlighted above: the experts’ differing assessment of the nature of the work. Mr Dunne was very conscious of the age and fragility of the house. But in paragraph 89 of Mr Steel’s report he said “I see nothing unusual about the house works.” He took his prices from the BCIS Minor Works price book, appropriate to work on a two- to three-storey building costing less than £1.25 million. He made no allowance for example for special requirements for a listed building in a conservation area. He took the view that preliminaries and professional fees, for example, could be calculated as a simple percentage of overall cost, whereas Mr Dunne made provision for specific items under preliminaries and specific professional fees. As we shall see, we took the view that Mr Steel’s approach was inadequate, whereas Mr Dunne was in some cases over-providing.

Another difference is that Mr Dunne took the view that the work on the house would itself take 6 to 8 weeks but that a further 12 or 13 weeks was needed for drying time for plaster etc, and thought that the whole programme would be carried out over a single period of 21 weeks, with the house emptied of furniture – which would be stored – and the claimants themselves accommodated elsewhere in Sandwich at the respondent’s expense. Mr Steel made no assumptions about timescale, but saw no need for the house to be emptied. On this point we prefer Mr Steel’s approach. We take on board Mr Dunne’s point that it would be uneconomic for a specialised professional to do one room at a time, because they might not be fully occupied on each visit. But this is a very big house. It would be perfectly possible to devise a programme of work doing a few rooms at a time. It is not realistic to empty a house of its residents and furniture for 21 weeks in order to do £40,000-worth of work, but Mr and Mrs Brookhouse have a second home in London which they occupy, as Mrs Brookhouse said in evidence, up to 30% of their time. It is not suitable for long stays because there is no garden for the dog. But clearly they make short stays there, and could design their time away to fit with periods when the house is less than convenient for them, for example while the kitchen is being re-tiled or the east front is being pinned.

With those points in mind we can look at the eight specific items about which the experts disagreed.

Protection, security and cleaning

Mr Dunne provided for £16,057 under this head, Mr Steel provided nothing. Mr Dunne’s figure is made up as follows:

In total we award £2,800 under this heading.

Surveys

Mr Dunne provided for £3,192 under this heading, made up of £2,000 for a timber decay survey as recommended by Mr Maude in case there is as yet undiscovered decay, and £1,000 for a dry rot survey, plus inflation. Mr Steel thought these two items would be included in the investigation costs which we consider at paragraph 184 and following in section 3 below. Neither of these items is mentioned in what either expert says about those costs. Dry rot is nothing to do with the respondent, but we allow £2,000 under this head for a further survey of the timber. As discussed above we make no allowance for inflation since this is not an agreed figure. In any event we think this survey would be carried out quite soon, before the final scope of the works is determined.

Preliminaries

Mr Dunne allows £33,028 for preliminaries, on the basis of work being done over a 21-week period by a specialist contractor with a “small management and supervision team” comprising a project manager, quantity surveyor, site manager, and site foreman, together with “welfare”, skip hire, and plant and equipment. Mr Steel allows 15% of the agreed £36,742, at £5,511, which he says is the BCIS recommendation.

We are not assisted by the fact that the two experts disagree about what the BCIS actually says on this point. We suspect that 15% is usual for ordinary projects on modern houses, and would be inadequate for this one. Equally we do not believe that preliminaries would amount to nearly 100% of core costs. Mr Dunne’s management team is too big to be realistic for this amount of work. In cross-examination Mr Dunne referred to projects where preliminaries had amounted to 30% or more. His explanation for his higher proportion was that this is not the total refurbishment of a house but a lot of little jobs. We do not understand why that would raise the preliminaries.

Doing the best that we can with inadequate information we allow 30% of £36,742, which is £11,023.

Overheads and profit

Mr Dunne said that he would normally allow 10% overheads and profit but raised it to 20% because this is a listed building in a conservation area and specialist contractors would charge more. Mr Steel maintained that 10% was appropriate, However, Mr Steel’s figure was not half of Mr Dunne’s, because Mr Dunne took 20% of £89,019, being the sum of the core costs and his figures for items (i) to (iii) above, making £17,804, while Mr Steel took 10% of £42,253, which was his total for the core costs plus his figures for items (i) to (iii), making £4,225.

We have nothing to tell us which of the two is correct, save for the considerations that as usual Mr Steel has not made any allowance for the nature of the building and that Mr Dunne tends to over-provide. We allow 15% of £36,742 plus our figures for items (i) to (iii), namely (£2,800 + £2,000 + £11,023) = £52,565, making £7,885.

Contingency

Both experts allowed 10% for contingencies. Mr Dunne calculated 10% of his total so far (£36,742, + items (i) to (iv) above) = £10,682, and Mr Steel did the same, making £4,648.

Adopting the same method, we calculate a 10% contingency allowance as 10% of £36,742 plus our figures for items (i) to (iv), namely (£2,800 + £2,000 + £11,023 + £7,885) giving a total of £60,450 of which 10% is £6,045.

Professional fees

Mr Dunne allowed £33,774 on the basis that the project would require a manager (£1,000), an architect/contract administrator (£14,000), a structural engineer (£10,000), a quantity surveyor (£7,000) and a principal designer (£1,000) (plus inflation). Mr Steel allowed 12.5% of his total so far which came to £6,391.

Again we think that Mr Dunne’s team is obviously top-heavy and that this level of management and expert involvement is excessive for a project of this value, even on a listed building. We accept Mr Steel’s allowance of 12.5% of the total so far, at £66,495, which is £8,312.

Consents

Mr Dunne allows £11,670 for obtaining listed building consent (£10,000) and building control compliance (£1,000). Mr Steel takes the view that these consents are included in professional fees or investigation costs. We cannot see that listed building consent falls within investigation costs. We think that obtaining building regulations consent would fall within professional fees. We therefore allow the former but not the latter, £10,000.

Alternative accommodation and the storage of furniture

Mr Dunne has qualifications in health and safety. In his view the house would not be a safe environment for Mr and Mrs Brookhouse and their dog during his 21-week programme of work. We have rejected that plan and so we do not allow his £16,200 for accommodation nor his £43,825 for the removal and storage of furniture, the grand piano, art and antiques.

The claimants will have choices about the way the work is done, and if there is a time when they would rather be out of the way they have accommodation in London which they use regularly in any event. Equally there is no need for the house to be emptied of furniture. For the most part it can be dust-sheeted or – for example where a floor is being worked on – moved to another room.

That said, the quantum experts agree that a sum should be awarded for moving furniture. Mr Steel included £500 in his costing of the core work – and so in his view it is a component of the agreed sum – whereas Mr Dunne included nothing within the core costs because he thought that £43,825 was required as a separate item. So the experts agree that a cost will be incurred in moving furniture. If there are antiques that need to be moved around, or perhaps the grand piano, we think that £500 is inadequate. We therefore award a sum of £5,000 for the moving around of the piano, antiques and artwork.

Compensation under this heading therefore amounts to £89,802 and we round this to £90,000, to which VAT will be added.

In the 2023 decision we found that the groundwater levels in the garden of the property had been raised as a result of the respondent’s works, by between 0.5 and 1 metre, probably nearer 1 metre. It is worth being aware that the controversy between the parties related not to the then current groundwater levels but to what those levels had been before the new wall was built. As to the current levels there was very good quality evidence available in the form of the “Eldred data”, as we explained in paragraphs 98 and following:

“98.

In January 2018, at the claimants’ instigation and with the respondent’s agreement, Eldred Geotechnics were instructed to carry out a study of the water levels in the garden; it will be recalled that by that date there had been a series of extra-high tidal inflows and a number of phases of remedial work. In February 2018 Eldred installed boreholes which were continuously monitored from 28 February to 9 May 2018. That exercise produced a rich dataset (“the Eldred data”) showing the movement of water in the ground in response to the tide and to rainfall at various points in the garden. …

99.

… Because the Eldred data are of such good quality they have given rise to rather more agreement than disagreement, and so we do not need to rehearse all the detail studied by the experts. They agree that the boreholes were correctly placed and that (with some caveats about a couple of the boreholes) the Eldred data are an accurate measure of water levels during that period. …

100.

Most importantly, they agree that the Eldred data showed groundwater levels near the house - in the area bounded by boreholes 1, 2 and 9 and by pipes 2, 3, 4 and 5 to be around 2m AOD during the period of monitoring. …

101.

That level of around 2m AOD is an important point of agreement; the central issue between the hydrologists is whether water levels near the property (where tidal response is weak) were at or near that level before the respondent’s works commenced. The claimants’ case is that levels were 0.5m to 0.9m lower, and the respondent says they were not.”

On the question of the pre-works water levels we examined a number of items of evidence, none of which by itself was conclusive but which together gave us confidence that the pre-works water levels were between 0.5m and 1m lower than they were when the Eldred data were gathered (and there was no suggestion from the hydrologists that water levels had fallen since that time).

What we did not decide in the 2023 decision was the mechanism by which groundwater levels had been raised. We acknowledged the suggestion by Mr Evans, the claimants’ hydrology expert, that water flowing in with the tides could be trapped behind underground structures in the garden; there had been more than one river wall in former years and it seems that the garden may have been gradually built out toward the river, leaving old structures behind under the ground. In giving directions for expert evidence to be adduced at the quantum hearing we pointed out that the experts might wish to give further consideration to how water levels had been raised, as a step towards devising a solution – as indeed they have done.

At the start of the hearing in March 2026 the respondent’s position was that there was no need for anything to be done to restore the groundwater to pre-works levels, because the damage already done to the house can now be remedied and any ongoing damage can be addressed in the normal course of maintenance. That was the evidence of Mr Stuart Hookham of William J Marshall & Partners, the structural engineering expert instructed by the respondent; the claimants’ expert Mr Peter Corcoran of The Morton Partnership disagreed. During the course of the hearing it became clear that the difficulty with the respondent’s position was that the cost of that ongoing maintenance, insofar as it was increased by the raised groundwater, was unquantified; and for that reason on the final day of the hearing the respondent conceded that it would be necessary to carry out work to restore groundwater levels to their pre-works status. As a result we do not need to say anything else about the evidence of the structural engineers.

Whilst it is therefore agreed that something has to be done, the parties are a very long way apart about what has to be done and about the cost of whichever solution is adopted. Both parties instructed water and drainage experts. Mr Mohammad Yadollahi of Alvarez & Marsal Disputes and Investigations LLP gave evidence for the claimants and Mr Richard Brinkworth of RSK Global Experts for the respondent. Mr Yadollahi advocated the demolition of the dock wall and the installation of sheet steel piles (“SSP”) at a depth matching that of the new wall, so as to continue the new wall round the corner and along the side of the dock. Mr Brinkworth believed that groundwater levels could be controlled without doing anything to the wall, by the installation of a drain in the garden.

In the paragraphs that follow we first describe the solution offered by each expert and then explain why we determine that Mr Yadollahi’s solution is the appropriate one, and is to be funded by the compensation paid by the respondent.

Mr Yadollahi’s first report, which is a model of clarity, began by explaining and illustrating the works carried out by the respondent. It started from the Tribunal’s determination that water levels at the property have been raised, examined how that has happened, and discussed options required to rectify the groundwater levels. Mr Yadollahi described the remedial works undertaken by the respondent in 2015, 2016 and 2017 and expressed the view that it was unsurprising they had not been successful. The remedial works in the dock corner extended only down to the riverbed, whereas leakage was lower down. He went through correspondence between Mr Evans of Evans & Langford – the claimants’ hydrology expert at the first hearing – and Mr Holland of CH2M, the engineers employed by the respondent, and highlighted Mr Holland’s agreement that the excessive inflows could only come from the dock corner, and that it was “apparent that a flow path has established, possibly beneath the construction.” Mr Holland acknowledged that remedial works so far were addressing the symptom – by adding layers of grouting and other infill behind the new wall and the dock wall – rather than the cause of the problem. We refer to the “dock wall” to mean the dock return wall, at right angles to the new wall, between the garden and the dock.

The dock wall itself is a complex structure. The original concrete wall has been faced on the water side by SSP installed by the claimants in 2002. Mr Yadollahi explained that that SSP does not go deep enough to cut off water ingress. It stands on concrete below river level. Further layers have been added in the course of the remedial works, including grouting and infill above river-bed level on the garden side and an “inverted T” concrete wall added to the structure by the respondent.

In the course of the remedial works the old wall fronting the garden and the dock wall were significantly damaged, as can be seen in photographs showing chunks of concrete missing, or (in the case of the dock wall) lying on the ground after falling off. Mr Yadollahi pointed out that it is very risky to damage 1940s concrete, of which these two walls were made, because of the poor quality of materials available at that time. Both walls had been subjected to impact and vibration; the old wall had its height reduced with a percussive breaker, and the dock wall was demolished at the corner using a jackhammer and perhaps an excavator. He thought it was likely that the concrete at the bottom of the dock wall, below the riverbed, had been damaged and that water ingress was happening there, where no investigations have been carried out. Any solution had to address the water ingress at lower as well as higher levels, and had to be independent of the compromised dock wall. Dye testing carried out in 2016 had failed to identify the source of the problem because the pit was not deep enough, and hence the failure of remedial work carried out to date, because it had only added material above river-bed level.

Turning to the situation as it now stands, Mr Yadollahi observed that all but one of the flap valves in the new wall have been sealed – he did not know why, since Mr Holland of CH2M believed that the excessive leakage into the property was not related to the flap valves. That meant that the free draining material between the old and new walls is now providing the inbound water with a “freeway” between the south and north of the site.

Mr Yadollahi explained that he and Mr Brinkworth had carried out a “joint investigation” in July2025. Employing the services of Evans and Langford and Thanet Drainage they had had trial pits dug in two locations, one near the dock corner and one halfway along the new wall. Piezometers were installed in boreholes beside each trial pit and another near the swimming pool (as well as a further trial pit and a borehole near the house, to which we refer later). Water seeped into both pits as the tide rose, significantly more at TP1 near the dock corner than in TP2. Mr Yadollahi’s opinion was that the seepage observed was more than could be expected through the clutches (the connections between sheets of steel piling) or the flap valves in the new wall, and that the source – as agreed in 2017 – must be the dock corner. That, he concluded, was the source of seepage into both pits, due to the permeability of the ground between the pits and in particular the free draining materials as mentioned above.

Interestingly, the remains of old timber structures were found in the trial pits. Mr Yadollahi regarded that as significant. He agreed with Mr Evans’ hypothesis that inbound water gets trapped behind them, thus raising overall water levels over time.

As to remedial options, Mr Yadollahi recommended addressing the problem at source by demolishing the current dock wall and replacing it with SSP driven deep below the river-bed, to match the new wall itself, so as to create a cut-off and eliminate water ingress. The new SSP would be tied in to the dock head wall, and connected with clutches to the new wall at the dock corner. Mr Yadollahi presented a design for the proposed work, prepared by Tony Gee & Partners using the respondent’s own diagrams. Tony Gee & Partners have been appointed by the claimants as their consultants for future remedial work, on the recommendation of the College of Experts.

Mr Yadollahi set out a methodology for the work; there would be a staged demolition, removing the inverted T wall and installing the new piling before demolishing the existing shallow SSP. This solution would address the problem at source, at the requisite depth, and provide a once-and-for-all solution; it achieved what, in Mr Yadollahi’s opinion, the respondent should have done originally rather than terminating the new wall at the dock corner.

Crucial to Mr Yadollahi’s solution is his view that whilst the new wall was of course intended to be permeable, through both clutch joints and flap valves, the problem to be solved is not the regular and intended movement of water but the overall raised water levels established by the Eldred data, and the excessive inflow experienced at times of extra high tides. Equally crucial is his agreement with the consensus of Mr Evans and Mr Holland that a flow path has been established beneath the dock wall and is the source of the problem. Nothing that Mr Yadollahi observed in the course of the joint investigation cast doubt, in his mind, on the correctness of that view. The excessive inflows could not be accounted for by the presence of the flap valves, and the raised water levels could not be lowered or prevented by opening the flap valves. He also firmly rejected the idea, put to him in cross-examination, that water levels at the property will be lowered once the garden is re-planted, on the basis that while plants take up water their effect does not go deep enough to make a difference here.

Mr Yadollahi explained that if the flow path beneath the dock wall is not cut off there is a danger that material will gradually be washed away and in the future – perhaps in 40 or 50 years’ time - the ground will collapse and a sinkhole could be created.

Mr Yadollahi suggested the addition of a “French drain” – comprising a perforated pipe submerged 1 metre deep along the length of the garden – as a way to drain water trapped on the site so as to lower water levels immediately after the new dock wall was installed; he stressed that this was a precaution, and might not be needed.

Mr Brinkworth’s first report described briefly the property and the respondent’s works, and mentioned but did not comment on CH2M’s acknowledgement of a “flow path” beneath the dock corner. He provided a photograph of the outside face of the corner and observed that there was no evidence of any outflow around it.

Turning to the joint investigation in the summer of 2025, Mr Brinkworth explained that he saw nothing untoward in the flow of water into the trial pits. Outflow was slow, which he thought was consistent with the fact that all but one of the flap valves are closed.

Mr Brinkworth stressed that the ground at the property is complex, the result of thousands of years of interaction between ground and water. He commented on the paucity of data available to the Tribunal in the earlier hearing, and expressed the view that there had been “an oversimplification of the previous assumed ground model in order to deliver a straightforward persuasive message for the purposes of litigation.” He pointed out that the ground should not be assumed to be homogenous nor to have a single consistent phreatic surface.

We understand the “phreatic surface” of the ground to be the level at which the water pressure upwards does not exceed atmospheric pressure downwards; accordingly, water is not forced above it by pressure from the saturated ground below, but may travel upwards by capillary action. Therefore, in Mr Brinkworth’s view, the piezometers used to produce the Eldred data “may not necessarily be indicative of the phreatic surface in the soils at that location, particularly where there is a confining overlying layer of clayey soils.” He felt that the Eldred data had been misunderstood; piezometers may not provide an accurate measurement of water levels in an enclosed aquifer, because water can flow freely upwards into the piezometer whilst being impeded in the ground itself by impermeable layers. Essentially Mr Brinkworth was not convinced that groundwater levels at the property had been raised. He said that the level observed on the day of the joint investigation in 2025 was at the pre-works level observed by GL Martin in 2007.

Mr Brinkworth dismissed Mr Evans’ theory that underground structures on the property were retaining water, saying that there was no evidence of a continuous intact structure that could have that effect.

Mr Brinkworth agreed with Mr Yadollahi that the flap valves in the new wall should not have been sealed, as all but one of them is. His recommendation was that the flap valves should be opened, that further monitoring should take place, and that care should be taken not to disturb the natural relationship between the ground and the river, nor to create significant new drainage pathways. In addition to that he recommended the installation of a 2.2 to 2.8m deep ground drainage trench on the landward site of the wall, falling at 1 in 100 in a southerly direction and with a pumped drain outfall over the wall, “in order that local groundwater levels can be kept in line with conditions prior to construction of the wall”.

It is important to appreciate that this diagram is not to scale. Pause for a moment to contemplate the width of the chamber, 30 cm, and to contrast its depth of between 2 and 3 metres. Mr Brinkworth thought that despite its depth it could be installed using a drag box, i.e. a moveable barrier preventing saturation of the channel as it was dug.

The diagram above is the only specification of the drain provided. In cross-examination Mr Brinkworth explained that the channel would sit between the old and new walls and would travel the length of the site, parallel to the river, save at the north-west corner where the swimming pool is. It comprises a perforated pipe within a sand-filled chamber, covered at the top to prevent rainwater getting in, draining towards a sump beside the dock. The pump at that point could be controlled and could maintain groundwater at whatever level was desired.

Questions asked by the two experts of each other reinforced the impression given in Mr Brinkworth’s first report that he did not really think there was a problem. He took the view that the problem of excessive inflows was fixed by the grouting works done in 2016. He did not think there was any abnormal inflow from the dock corner. He expressed concern that sealing the dock corner would mean that rainfall would be trapped on site and unable to discharge into the river. Mr Brinkworth had no concerns about predicted higher river levels in the long-term future.

He was sceptical of a solution that sealed the dock corner, and took the view that if Mr Yadollahi’s solution were adopted there would have to be a pumping system to prevent the build-up of water.

Mr Brinkworth’s solution would be difficult to instal; serious engineering is needed to dig a channel 30cm wide by (say) 2.5 metres deep, and the effect of that work on the ground surrounding it would be profound. That is perhaps not too much of a problem since the garden has to be renewed in any event. It would require ongoing maintenance and indeed a constant power source for the pump; but in terms of overall cost Mr Brinkworth’s solution is a great deal less expensive than Mr Yadollahi’s, whether one takes the claimants’ costings (£1.8 million for Mr Yadollahi’s scheme. £247,662 for Mr Brinkworth’s) or the respondents’ (£360,409 for Mr Yadollahi’s scheme, £145,796 for Mr Brinkworth’s). And there is an attractive simplicity to it: water will always find the easiest way out, and Mr Brinkworth’s drain allows it to do just that.

If the problem to be addressed were as Mr Brinkworth sees it, that normal inflow to the property is not getting out efficiently, then the appropriate solution would no doubt be to help the water out of the property. Mr Brinkworth said in re-examination: “This is a drainage problem, so a drain seems a reasonable solution.”

But the Tribunal has already found that this is not a drainage problem. It is an inflow problem, causing raised water levels which are damaging the house, and causing inundation of the garden at times of particularly high tides.

As we said at paragraph 4 above the expert witnesses for the 2026 hearing were entitled to disagree with the Tribunal. They were obliged to give us their true opinions, and if they thought that the Tribunal got it wrong in 2023 they were obliged to say so and could not be criticised for doing so.

We could perhaps ignore such disagreement on the basis that the claimants are entitled to a judgment on the quantum of liability that proceeds from the unappealed findings of the 2023 decision. We think that it would be most unconstructive simply to ignore disagreement. If Mr Brinkworth’s evidence led us to conclude, on the balance of probabilities, that we were wrong in the 2023 decision we would say so, and would then address the legal arguments as to what we should do as a result of our finding. But if Mr Brinkworth is wrong then again it is important that we say so, both in fairness to Mr Yadollahi and to the hydrologists who gave evidence in 2023, and to assist the claimants who have to live with the physical consequences of the respondent’s work.

We are wholly unconvinced by Mr Brinkworth’s evidence. His solution cannot be adopted because it proceeds from a failure to understand the evidence given in 2023 both about tidal inflows and about groundwater levels, and as a result, it is designed to address the wrong problem.

So far as tidal inflows are concerned, Mr Brinkworth expressed the view that the 2016 remedial works were effective in preventing these for the future. We cannot begin to understand how he formed that view given the fact that further inflows occurred in 2017 and again in 2019 (see the 2023 decision). Nor can we understand what he makes of the agreement of Mr Evans and Mr Holland that a flow path had been established beneath the dock wall, nor of the agreement of the hydrology experts (Mr Evans and Mr Groves) in 2023 that the extra-high tidal inflows would recur. In cross-examination, although not in his report, he said a number of times that the hydrology experts had agreed that the source of the ingress of water was the new wall, and specifically the clutch connections. He could not readily point to any source for that assertion, and when he did eventually identify what he regarded as a supporting reference, it turned out to be one of many instances in the 2023 hydrology evidence where the experts agreed that the new wall was not impermeable and was not intended to be (see for example paragraph 21 of the 2023 decision).

We asked Mr Brinkworth at the hearing if he felt the experts were wrong in 2023 when they agreed that the source of the excessive ingress was the dock corner. He said he did not know because he did not have the evidence to make that assessment. Yet a wealth of information was available to him. It was open to him to do as Mr Yadollahi did and go through the factual accounts of what occurred between 2014 and 2019 so as to evaluate the experts’ conclusions, but he did not do so. The strong impression Mr Brinkworth gave us was that he was interested only in what he could see in 2025. Hence his concern with the equilibrium of land and water, ignoring the fact that that equilibrium has already been disturbed by the respondent’s works and the evidence that that disturbance had caused damage and will cause more in the future. It was put to Mr Brinkworth that along the whole extent of the new wall in Sandwich there is no evidence of excessive inflow or raised groundwater levels except at the claimants’ property, which points to the conclusion that something unique to this property, rather than the new wall itself, is the source of the problem. Mr Brinkworth had no answer to that.

Nothing in Mr Brinkworth’s evidence casts doubt upon the evidence we heard in 2023 that the reason why too much water is getting in to the property is an inflow path below the dock wall.

Turning then to the groundwater levels, Mr Brinkworth was sceptical of the Eldred data and in particular of the fact that this includes extensive piezometer readings. He is not a hydrologist. He made much of the possibility of misunderstanding of piezometer readings, yet the hydrologists in 2023 were agreed about the groundwater levels measured by the Eldred data. What they disagreed about was not the current groundwater levels, but the pre-works levels. Indeed, Mr Brinkworth admitted in cross-examination that the hydrologists in 2023 were aware of the possibility of misinterpreting piezometer reading for the reasons he described, had discussed it, and had eliminated some readings as a result – although he had not thought to say that in his report.

As to Mr Brinkworth’s view that the groundwater in summer 2025, at the time of the joint investigation, was around the level seen in 2007, as Mr Pereira pointed out in cross-examination the 2007 reading was taken during a period of exceptionally wet weather (over twice the average rainfall was recorded in June and July 2007) whereas in June 2025 rainfall was well below average; we do not think that low water levels in June 2025 cast doubt on the Eldred data. What was observed on a single day in a hot dry summer, some years after the very high tides and excessive inflows seen from 2014 to 2019, cannot negate the rich data from a survey over months. And as Mr Pereira pointed out in closing there was also a reading of 1.75 metres AOD from the Albury borehole near the house (see paragraph 92 above), and in January 2026 a reading of 1.995 AOD.

Nothing Mr Brinkworth said casts doubt upon the expert consensus in 2023 that the groundwater levels after the respondents’ works were at around 2 metres. The word “around” is important; neither the experts nor the Tribunal supposed that there was a single phreatic surface across the property. It was clear that levels varied; but the range of levels was agreed.

We have to add that Mr Brinkworth did give consideration to the capacity of his drain to deal with extra-high tidal inflows, (despite his view that that problem had been solved in 2016). He carried out a “Darcy calculation”, which calculates flow by reference to the permeability of the ground, the hydraulic gradient, and the area. Mr Brinkworth calculated that the additional flow from a tide of 3m AOD could readily be absorbed by his drainage arrangement. Mr Yadollahi pointed out that Mr Brinkworth’s calculation was wrong because he has calculated hydraulic gradient on the basis that the ground level is at 0 metres AOD, which means that the additional inflow at a very high tide, as a proportion of normal inflow, is much too low. Mr Brinkworth did not have an answer to that.

In any event, whether or not Mr Brinkworth’s drain is able to get rid of water after an extra-high tidal inflow, that is not really the point. The evidence in 2023 was that at times of very high tides water bubbles up into the garden – saline water, with obvious detrimental effects upon vegetation – and forms pools. The assurance that that volume of water will drain away again is not much comfort; the remedial objective is to stop it getting in.

We conclude that Mr Brinkworth’s proposed solution would probably be ineffective. Mr Yadollahi’s solution, by contrast, is a targeted solution, addressing the problem that has been identified, in the absence of any evidence whatsoever of any other problems. It has the advantage of matching the design of the new wall itself, simply continuing it around to the dock return. It incorporates the respondent’s own mechanisms for inflow and outflow using flap valves, and needs no ongoing management but relies upon the natural relationship between land and river. Essentially it completes what the respondent started. We conclude that Mr Yadollahi’s solution should be adopted.

The parties’ positions as to what Mr Yadollahi’s solution would cost are very far apart indeed. Mr Dunne’s total cost amounted to £1,810,803 by comparison with Mr Steel’s cost of £360,410, both excluding VAT. The experts structured their costings under six headings as summarised below:

Mr Dunne priced individual items and stages, using the design drawings provided by Tony Gee (included in Mr Yadollahi’s report, see paragraph 94 above) and the construction sequence described by Mr Yadollahi in his report. He used cautious assumptions and made allowance for all possible risks, within a generous timeframe of 21 weeks. Mr Dunne had discussed the proposed works with JT Mackley & Co Ltd (“Mackleys”), a specialist contractor with experience of projects along the same stretch of river, and used indicative costs received from them in his assessment. These took account of difficulties relating to reaching the property (by road or river), working on and adjacent to a tidal river, and the proximity of adjacent buildings. Mr Dunne’s costs allowed for inflation at 4.91% using the BCIS figure on the assumption that investigations, design and obtaining consents would take place in the second to fourth quarters of 2026, and that dock wall works would take place in summer 2027.

In Mr Steel’s opinion the scope of remedial works proposed by Mr Yadollahi was at a very high level making it difficult to ascertain, with confidence, what a programme would look like. He had not attempted to obtain market information from a contractor, nor had he asked to see that which Mr Dunne had obtained from Mackleys. As with other aspects of the works, he had utilised “all in rates” from industry sources, primarily BCIS, to provide an indicative estimate of costs based on measured areas rather than looking at the specific needs of this unusual site. Where he did not have information he tended to assume a minimum provision rather than thinking through what might be needed, as we said above.

On the other hand, the difficulty with Mr Dunne’s costings is that they are ultra-risk-averse. Mackleys have, understandably, priced on the basis of a very generous time estimate which, in our judgment, must include a considerable element of provision for what could go wrong. It is beyond belief, for example, that the work will take 21 weeks when the respondent’s work along the river frontage of the property took eight weeks, and equally beyond belief that it will take four weeks for the actual installation of 8 metres of SSP along the dock wall (see below paragraph 138), when the 54 metres of the river wall took the respondent 6 days in 2013. Mackleys cannot be questioned about their costings. We have not seen their original figures. They have not been exposed to any commercial testing; they were produced in a context where there were no competitors. And on top of Mackleys’ costings Mr Dunne then added nearly £150,000 for contingencies. The result is, again, Rolls Royce provision.

Whilst in general we prefer Mr Dunne’s approach because he has given much more thought to the unique characteristics of this site, in our judgment it is very likely that he has over-provided. Taking a cautious approach in the interests of both parties, where Mr Dunne’s costings depend upon a time estimate we have reduced that estimate by 25%; in some cases we have made a further reduction on the basis that the provision for equipment or personnel seems manifestly over-elaborate.

As we said above, Mr Dunne added a figure for inflation to his costings, but Mr Steel did not. In line with our comments at paragraph 48 above we have therefore not allowed any addition for inflation, which we regard as the correct approach.

With those points in mind we review the competing costings under the six headings.

Mr Dunne’s total under this heading was £540,308 and Mr Steel’s total was £132,922. Works costs are grouped under 10 sub-headings, which we deal with in turn.

The core cost of supply and installation was agreed at £13,799 but Mr Dunne allowed a further £3,000 plus inflation for design by a specialist contractor. Mr Steel considered that this would already be allowed for in the general heading of professional fees. We agree with Mr Steel and include the agreed cost of £13,799 but make no addition for design fees.

Mr Dunne allowed £36,120 (adding inflation to make £37,894), which covered £1,760 for removal of timber cladding, steel panels and access ladders, a 10 day process of hydro-demolition of the T gravity wall (recommended by Mackleys to minimise vibration) costed at £18,800 and excavation of structures below with a 13.5 tonne excavator at £8,000, together with disposal of the excavated material (£7,560).

Mr Steel allowed £9,101, largely calculated from a price book but including £1,902 for disposal. His best defence of this sum, which included no allowance for hydro-demolition, was that he would normally expect a more developed design to be available for costing; but he did not accept that the demolition would take 10 days. Originally Mr Steel had only allowed for partial demolition of the wall along the piling line using standard rates but, during the hearing, he acknowledged that he should have allowed for full demolition and in that case his figure would have been closer to Mr Dunne’s. Essentially we felt that he had not taken on board the complexity of what is there at present and of the methodology set out by Mr Yadollahi in his report.

In principle we prefer Mr Dunne’s more detailed approach. However we are sceptical of the idea that it will take 10 days to demolish the existing structures. In line with what we said above we make a 25% reduction to his £18,800 for work and operatives and to his £8,000 for the excavator and operator. In line with what we have said above we do not add a further sum for inflation. Accordingly from £36,120 we deduct 25% of £26,800, = £6,700 to make £29,420.

The work included under this heading is removal of the existing sheet piling along the dock wall, and the installation of a new sheet piled wall and its connection to the main river wall and dock head wall. Mr Dunne allowed £213,036 after inflation. Before inflation his allowance was £203,058, of which £28,818 was for materials (sheet piling, corners, connection and clutches) and £174,240 for plant and labour, costed over a four week period. The cost of plant included hire of a Giken vibrationless piling rig, with a Supercrush attachment to deal with any unusual obstruction from the existing concrete buttress below river level on which the existing piling rests – as discussed in Mr Yadollahi’s report (see paragraph 90 above) - at £25,000 per week, to which was added £12,000 for delivery and collection. A reaction stand was costed at £500 per week, together with £2,000 for delivery and collection. Labour was costed at £7,500 per week for a piling gang, plus £1,560 per week for their subsistence and hotel, and £4,400 per week for a 4-man gang of “skilled operatives”. A welder was costed at £2,200 per week for two weeks. Excluding the cost of the welder and the costs of delivery and collection of plant the weekly cost is £38,960, so the overall cost is very sensitive to the time over which these costs are budgeted.

Mr Steel allowed £75,144 for the wall, based on BCIS costings with an uplift for specialist connector piles at the corners. He said that his costing rate was based on use of a silent and vibrationless rig, and agreed that a Giken piling rig was the only silent and vibrationless system, but he had not considered that a Supercrush attachment would be necessary and had not appreciated that it might be needed to demolish the existing underwater concrete.

Mr Steel seemed to dispute Mr Dunne’s proposed working time of four weeks, but could not give any opinion on an alternative shorter period. We said above that we are in general sceptical of Mr Dunne’s time periods; in this case Mr Dunne has produced a programme of work in which the Giken is scheduled to be mobilised for three weeks not four, which is a further indication of the caution in his costings. Accordingly in line with the indication we gave above we reduce the costings for plant and labour by one week. In doing so we take Mr Dunne’s £174,240, subtract £18,400 for the cost of the welder and the costs of delivery and collection of plant, to give £155,840, of which 25% is £38,960. Deducting £38,960 from Mr Dunne’s figure reduces the overall cost for the sheet piled wall to £164,098.

The work involved under this heading is for a reinforced concrete wall and foundation, with reinstatement to the capping beam. Mr Dunne costed the construction of the RC wall itself at £30,293 including inflation and Mr Steel at £13,040 excluding inflation. Mr Steel’s lower figure resulted mainly from his assumption that less demolition would be required (see (b) above) and therefore less reinstatement. Following his concession on the need for full demolition, Mr Steel accepted that his figure under this heading would be close to Mr Dunne’s. We therefore also adopt Mr Dunne’s pre-inflation figure of £28,875. The costs of brickwork facing, timber cladding and refixing access ladders were agreed at £10,548, so the total payable under this head is £39,423.

This item was agreed at £2,869.

Under this heading fall two different items.

One is the replacement of the existing 150mm diameter perforated drain on the garden side of the dock wall, which Mr Dunne priced at £440, or £461 after inflation, and Mr Steel made no allowance for.

The other item is the 100mm French drain (see paragraph 98 above). Mr Dunne priced it at £20,041, composed of detailed costs for materials and labour, adding inflation to make £21,027. Mr Steel’s cost of £6,233 was for a less complex design without additional drains back to the river, and no inspection chambers or pumps.

Given the shortfalls in Mr Steel’s provision we accept Mr Dunne’s costings for these two drains at £20,481.

A cost of £2,185 was agreed for provision of 200mm of topsoil over the working area, ahead of the garden works.

We were presented with an extraordinary divergence of opinion under this heading with Mr Dunne allowing £191,730 (£182,750 before inflation) and Mr Steel nothing at all, because he said that his “all in” rates included plant and labour for the relevant activities. We did not accept Mr Steel’s “all in” rates and preferred Mr Dunne’s approach, so we have to accept that there may be additional costs under this head. But Mr Dunne’s costings for the above items included some labour and some plant and he made adjustments to avoid duplication.

So we accept that there will be costs under this heading, but with some caution. A contentious element of Mr Dunne’s allowance is his adoption of a 21-week period for the project, and his costing of various items for the whole or most of that period. The weekly cost of £2,750 for a dumper, fuel, compressor, pump and hoses, small tools and a pick-up van amounted to £57,750 over a 21 week period, plus a further £6,400 for a forklift for 8 weeks and £10,000 for delivery and collection; a 13-tonne excavator, breaker and operative were costed at £4,000 per week for 18 weeks to give £59,500 (after some allowance for inclusion elsewhere) plus £4,000 for delivery and collection. Labour comprised a four-man gang of skilled operatives at £4,400 per week; but Mr Dunne deducted the labour already costed under the headings above and so allowed for their being employed for just over 10 weeks under this head, rather than the full 21 weeks, making £45,100.

Mr Dunne confirmed that his programme for the project was put together by Mackleys, although he had proposed a number of adjustments to get it down to 21 weeks, which retained the minimum of two to four weeks required as a “float”. He was of course challenged on his estimate of 21 weeks to complete a stretch of dock wall only 8m in length when the respondent had completed all the works associated with a 54m length in eight weeks). He took us through the various stages and tasks provided for in his programme and explained that many of the complexities involved in demolishing and replacing the dock wall would not have been encountered in the respondent’s flood wall project. The larger project would also have benefited from more efficient deployment of plant and equipment, and use of the Giken piler would have been quicker because there had been no need to use a supercrush attachment.

We cannot safely rely on Mr Steel’s assumption that his separately priced items already include everything required under this heading, but equally we cannot accept at face value the scale of costs proposed by Mr Dunne. Mr Dunne explained that his 21-week programme contained a “two to four week” float period. A safety-net of four weeks is beyond what the respondent should be expected to pay for. And in any event as we have said above Mr Dunne, and Mackleys, have in general made generous and cautious allowances, to which contingency is going to be added later.

We will therefore reduce the cost of plant and machinery by removing the weekly costs of the dumper etc (£2,750, see above), the 13-tonne excavator and labour, at £11,150, for five weeks, being a little under 25% of the time leaving 16 weeks; that is still cautious and, in our judgment, makes full provision for what is required. That amounts to a reduction of £55,750 to give a total cost for plant and machinery of £127,000.

Mr Dunne had made an allowance of £5,190 plus inflation (making £5,444) for reinstating the flap values already installed in the new river wall. By the end of the hearing it was agreed that the respondent would carry out this work so we make no allowance for it.

Mr Dunne allowed £7,868 for three surveys: to test soil for disposal; a ground penetrating radar survey to identify any underground services and structures; a UXO survey to reduce the risk of encountering unexploded bombs. Mr Steel considered that these surveys would be accounted for in the separately costed investigation costs. We do not see them specified anywhere else and allow for them in our costs at £7,868.

This brings our total for the items under ‘Works costs’ to £407,143.

The next four items are preliminaries, overheads and profit, contingency and professional fees. It will be recalled that when considering the work to the house, Mr Steel assessed all four of these items simply as a percentage of the running total, whereas Mr Dunne took a more detailed approach (see paragraph 61 above). We find the same divergence of approach here where Mr Steel has calculated all four items as a percentage, albeit with additional sums for access and monitoring, whereas Mr Dunne has looked at specific costs for preliminaries and professional fees.

Preliminaries including flood protection and access

For both experts these items were significant. Mr Dunne’s £729,784 is by far the largest component of his total estimated cost for the dock wall works, accounting for about 40% of the whole. Mr Steel allowed £131,842 over 30% of his total, less than 20% of Mr Dunne’s figure. We have grouped these costs under three headings of (a) flood protection works, (b) access from the river and (c) preliminaries.

Mr Dunne allowed £37,500 (adjusted for inflation to £39,342) for a Portadam (a portable dam) to provide flood protection to the garden during the works. The figure had been provided by Mackleys and was based on £20,000 for installation and removal, together with hire for 14 weeks at £1,250 per week. Mr Steel acknowledged that flood protection would probably be required but made no specific provision for it and considered that it would be included within his general 37.5% allowance for preliminaries. That is not credible when that allowance amounts in total to £49,845. We agree with Mr Dunne that a Portadam will be required but as we said above we are not persuaded by his time provision. We have allowed 16 rather than 21 weeks for the whole project, and accordingly we allow for the Portadam to be used for 10 weeks rather than 14. We have no way to assess whether the installation and removal costs are realistic, so we accept them, bringing the total to £32,500.

Mr Dunne costed this at £313,342 (£328,739 after inflation), while Mr Steel allowed £81,996. That latter figure was composed of £69,746 for the jack-up barge (see below), £10,000 for a temporary works engineer and £2,250 for monitoring over the winter after the work was finished. Neither of those two latter sums is about access from the river; we assume they were included in this sum and in the preliminaries for convenience and so we include both here.

Mr Dunne’s scheme allowed, as advised by Mackleys, for a jack-up-barge (“JUB”) to be used for carrying out the dock wall works, and a 60-tonne crane to operate from it, for a period of eight weeks. A JUB is, we understand, a mobile, self-elevating platform designed for use in tidal waters, which is delivered in sections to be constructed on site. Mr Steel agreed that a JUB would be required and agreed the cost of delivery and collection (around £27,000 twice), and the weekly hire cost at just over £14,000, but because he felt it was not possible to determine the number of weeks that would be needed he allowed just one week’s hire, on the basis that that figure could then “be applied across the weeks in a developed programme”. So Mr Steel’s figure was expressly not a realistic costing.

Mr Dunne’s £313,342 comprised, first, £181,962 for the JUB itself (£52,554 for delivery and collection, insurance at £1,500 per week for 10 weeks, £3,000 for a barge survey and £111,408 for eight weeks of hire). He had received a quote from Mackleys for this work and obtained an alternative one from Red 7 Marine which was similar.

Second, Mr Dunne added £131,380 in associated costs for that eight week period: £7,500 for a “site specific survey and calculations”, £13,440 for a welfare unit, bowser, container and fuel; £24,000 for a work boat and boatman; £42,000 for delivery, collection and hire of a 60 tonne crane to operate from the JUB; £3,840 for a timber deck; £4,600 for scaffold edge protection. A further £15,000 was allowed for the use of a 100 tonne crane at Highway Marine (on the opposite side of the river from the property) to assemble and dismantle the JUB over two separate weeks. Finally a provision of £21,000 was made under this heading for a safety boat at £1,000 per week for 21 weeks, the time Mr Dunne expects the project to take from start to finish.

Mr Steel made no express allowance for these additional items because he thought they would be included in his all-in preliminaries percentage, which we discuss in the separate heading below.

Mr Dunne was careful and methodical in identifying all elements of cost that would arise in connection with a JUB, having been advised by Mackleys from their own experience of carrying out projects locally. As before, we cannot criticise the thoroughness of Mr Dunne’s approach, but we remain concerned about the resulting elaborate provision and the absence of any means to test or compare Mackleys’ figures. We allow £230,000 under this heading, which is approximately 75% of Mr Dunne’s figure, which we intend to include Mr Steel’s two additional items (paragraph 157 above).

Mr Dunne costed preliminaries at £344,760 (£361,701 after inflation) and Mr Steel at £49,845, being 37.5% of his cost estimate of £132,922 for the dock wall works. In cross-examination he relied on this figure to cover some of the items costed by Mr Dunne under the heading of access to the river, such as the 100 tonne and 60 tonne cranes, but acknowledged that using Mr Dunne’s costs for any one of those items would leave little provision under preliminaries for anything else.

Mr Dunne’s £344,760 covered the cost of a “full management and supervision team” at £272,800, a compound at Highway Marine for them to operate from at £61,560, and £10,400 for mobilising and demobilising fencing, temporary road mats and welfare cabins. These prices had been obtained from Mackleys. The largest figure of £72,900 was provision for a project manager to work full time for 27 weeks (i.e. from six weeks before the 21 week project would commence and all the way through it) at £2,700 per week. A works manager, a foreman and a gateman/traffic marshal at the compound were all costed for 21 weeks totalling £98,700. Other professionals, including a contracts manager, a senior engineer, a planner, a quantity surveyor, a temporary works design (engineer) and an environment advisor were costed for shorter periods.

We understand that the dock wall project will be complex and specialist work, but we find it very difficult to envisage how so many specialists would need to be involved in it for such lengthy periods. Mr Dunne’s preliminaries costings are heavily reliant on advice and figures he has received from Mackleys, and we understand that the management and supervision costs listed under the heading of preliminaries are for people engaged by the contractors. The Mackley costings were not produced in evidence for us to see so we must take a cautious view of the accumulated totals, particularly where specialist costs are multiplied by generous time periods, and we are asked to assume that each specialist would be employed full time for that period on the dock wall project at the property. That seems to us a highly unlikely and indeed exaggerated scenario and, doing the best we can, we reduce the allowance for management and supervision under this heading by 50% to £136,400.

We can see that a works compound across the river at Highway Marine would be a key facility for the project, although we understand that it has been costed by Mackleys without any specific evidence that Highway Marine would be open to such an agreement. Most of the itemised costs are based on a 21-week period which we have found to be excessive; but we acknowledge the uncertainty surrounding arrangements with Highway Marine. We allow a cost of £50,000, to which we add £8,000 for Mr Dunne’s final component for mobilising and demobilising the fencing, roadway mats and welfare cabins.

This brings our total for (c) preliminaries to £194,400.

Our total under the heading of (ii) preliminaries including flood protection and access, is therefore £456,900 (£32,500 + £230,000 + £194,400).

Our sub-total of (i) works costs and (ii) preliminaries etc is £864,043(£407,143 + £456,900) which we will round to £864,000. It is from this figure that (iii) overheads and profits will be calculated.

Overheads and profit (OH&P)

Mr Dunne applied a rate of 15% to his sub-total of £1,270,091 to produce an allowance of £190,513. His choice of 15% was taken from the lower end of the range of 10% to 35% suggested by Mackleys for undertaking tidal works. Mr Steel applied a rate of 10% to his sub-total of £264,763 to produce an allowance of £26,476. His choice of 10% was based on his understanding of the respondent’s framework rates, and he commented that this rate was an additional 7 percentage points above the rate of 3% referenced in Spon’s, which he considered would be unlikely to be attractive to the market. He acknowledged that when the claimants instructed contractors they would not have access to the respondent’s framework agreement rates. In view of that we adopt Mr Dunne’s rate of 15%.

15% of £864,000 is £129,600, which is our figure for OH&P. This brings our running sub-total to £993,600.

Contingency

Mr Dunne applied 10% to his cost sub-total (including OH&P) of [£1,460,604] to give £146,061; Mr Steel also applied an allowance of 10%, but to his much smaller sub-total of £291,239. We adopt the same contingency allowance of 10% on our running sub-total of £993,600 to arrive at a sum for contingency of £99,360.

This brings our final running sub-total to £1,092,960.

Professional fees

The difference between the experts’ figures is – to repeat our refrain - striking. Mr Dunne’s total of £190,467 included a stand-alone figure (before inflation) of £114,400 comprising: £60,000 for a structural engineer to prepare the detailed design, with £10,000 for site support; £10,000 for an independent checking engineer to check the final design of the works; £24,400 for monitoring of movement in the house while the dock works are done (including 14 visits at £1,600) and £10,000 for a clerk of works to make 13 visits at £750. A further £76,067 was added based on 4.75% of the projected cost (including contingency allowance) of £1,606,665. This comprised 1% for a project manager, 1.5% for a contract administrator, 2% for a quantity surveyor and 0.25% for a principal designer required under the Construction (Design and Management) Regulations 2015 (“CDM Regulations”).

Mr Steel’s total of £40,046 was simply a provision of 12.5% on his total cost of £320,364. He called this a provision to cover “expected costs, given the high-level understanding” of the potential works, he thought it was not possible to check that there was no overlap with Mr Dunne’s provision for similar professional fees under the heading of preliminaries.

We appreciate and prefer Mr Dunne’s thorough approach to the costing of the professional inputs required to carry out and complete the dock wall works, but we were provided with no evidence to support the figures amounting to £114,400. The round figures for engineering input appear generous, and provision for 14 visits to monitor tell-tales in the house at £1,600 per visit seems to us to be very excessive indeed. We allow £70,000 for engineers and monitoring and add £10,000 for visits by a clerk of works, to reach a stand-alone figure of £80,000. However, we agree with Mr Steel that where Mr Dunne added 4.75% for other professional fees there is potentially duplication of provision already made under preliminaries for project and contract managers and a quantity surveyor. We adopt the same approach as we did for preliminaries of reducing by 50% the allowance of 4.5% to 2.25% of our sub-total, but increase that to 2.5% to include 0.25% for the principal designer required under the CDM Regulations.

An allowance of 2.5% on our running sub-total of £1,092,960 gives a sum of £27,324, so our total allowance for professional fees is £107,324.

Consents

Mr Dunne’s figure of £13,670 included £670 for listed building consent, £10,000 for a conservation architect to manage the process of listed building consent and £3,000 for an archaeologist’s watching brief during the works. These were mainly required ahead of the works so had not been adjusted for inflation. Mr Steel had made a provision for investigation costs, consents and professional fees across all the works, of which £10,000 was for “consents as referenced by architects/conservation experts”. This was a general provision and no allowance had been made specific to the dock wall work. In the hearing Mr Steel accepted that listed building consent would be necessary and agreed the logic of an archaeological watching brief. We also accept these as necessary costs but consider the allowance of £10,000 for an architect to manage the listed building consent process as generous. We allow £10,000 under this heading.

Our assessment of costs for the restoration of pre-works water levels amounts to £1,210,327, which we round to £1,210,500. We refer to the point made above at paragraph 43; we will be adding VAT to all our figures later, but it should be noted that £1,210,500 plus VAT amounts to £1,452,600, which is well below the current unblighted value of the property (see paragraph 39 above), and so no issue about the measure of damages arises.

Other costs

It has been obvious from when the works were first proposed that the garden would suffer; the respondent needed a working corridor parallel with the river, and it was always going to be necessary to replace the planting in that section of the garden – an area that the claimants’ horticulture expert, Mr Wilson, put at about 167 square metres and the respondent’s expert Mr Harrison put at 240 square metres. As things turned out, because of the repeated need for remedial works after 2014 that part of the garden was rather more thoroughly excavated than originally envisaged, as well as having been inundated by salt water on the occasions of the extra-high tidal inflows. The Deed said this:

“4.1

Landscaping to be lost will be recorded by the Owners’ chosen specialist. A schedule of plants and trees to be replaced on a like for like basis will be agreed. This includes matching the maturity and size with the Owners prior to entry for site clearance works. The trees that are proposed to be retained and those which are proposed to be removed are shown on the plan in Appendix A. This is indicative and will be confirmed as above.

4.2

The landscaping works will be cared for and monitored for a period of 5 years after the construction works are completed and replaced if necessary.”

The Deed also provides that if the yew tree dies as a result of the works the respondent will replace it “with a yew tree as near as possible of similar size and maturity.”

It is not in dispute that the respondent has to pay for the restoration of the garden. It will be a major operation; soil has to be removed and replaced, plants and trees have to be installed, and then the Deed provides for a 5-year period of monitoring and an obligation to replace plants that fail during that period. The work cannot be done until the dock wall is replaced and all the work described in the earlier sections of this decision has been done.

Expert evidence about the garden, both as to what needs to be done and what it will cost, was given by Mr Gareth Wilson for the claimants and by Mr Karl Harrison for the respondent. Each runs their own business; both are gardening and landscape consultants with many years of impressive experience. Mr Wilson relied heavily on quotations from Ruskins, a specialist and highly qualified horticultural firm within reach of the property; Mr Harrison tended to rely on his own experience in providing quotations, on the basis that he knew the prices for which he could get work done. The latest version of the quantum experts’ Scott Schedule puts the figure that the respondent agrees is payable in respect of the garden at £728,828 + VAT. That sum includes, to give just a few examples, about £50,000-worth of soil, over £25,000 for a trellis, over £190,000 for plants and trees plus £62,000 for a replacement yew tree, £17,300 for the hire of an excavator for 24 weeks, the cost of a site foreman and office management, labour and a 10% contingency provision.

The parties also agree that three further items must be paid for, but they are in dispute as to the amount required:

A further source of dispute at the hearing was Mr Wilson’s view that if Mr Brinkworth’s drainage solution was adopted the respondent should pay a further £302,500 to cover the possibility that the drain would not work and the garden would have to be dug up and re-planted all over again. In light of our decision above that does not arise and that additional sum will not be payable.

So we turn to the three remaining garden items in dispute:

The two experts agreed that an irrigation system was needed to care for the new plants, and that a four-channel timer system was appropriate. Their costings are about £2,800 apart. Mr Wilson’s £10,070 is a quotation from Ruskins; he confirmed in cross-examination that it was the cost of initial installation and did not include a figure for maintenance. The quotation did not provide any further breakdown.

Mr Harrison’s £7,200 is not derived from a third-party quotation; it is what he said he would quote to a customer, and he was confident that he would be able to find at least one subcontractor who would be willing to provide a system for that sum, plus or minus 10% or 15%. He had calculated the £7,200 by adding the cost of materials to the cost of labour at £350 per person per day, plus the cost of de-commissioning for winter and re-commissioning for spring for five years, and £600 per year for five years to allow for accidents (such as a gardener accidentally digging into a pipe with a fork).

Mr Wilson was asked if he thought that a sufficiently good irrigation system could be provided for the £7,200 that Mr Harrison thought was appropriate and he agreed that that was “a possibility”. In light of that, of the fact that Mr Harrison’s figure was helpfully broken down so that we could see how it was derived, and of the fact that Mr Wilson did not point to anything specific in the Ruskins system which could not be obtained elsewhere, we take the view that Mr Harrison’s £7,200 would provide proper compensation for the claimants.

Here the experts were a long way apart, Mr Wilson assessing the cost at nearly £70,000 and Mr Harrison at £13,100, both figures excluding VAT.

It is worth beginning by looking at what the Deed says. Clause 4.2 promises that “the landscaping works will be cared for and monitored for a period of 5 years after the construction works are completed and replaced if necessary.” So the respondent is obliged to care for the “new landscaping”, which we take to mean all elements of the garden that are replaced and replanted. It must also monitor it for five years and replace it if necessary. We take the latter to mean that the respondent, not the claimant, is responsible for replacing plants that fail. There is some dispute about what “cared for” means, with Mr Harrison regarding it as monitoring rather than maintenance and Mr Wilson seeing it as actual practical care.

It is easiest to start with Mr Harrison’s costing of the after-care plan. He set out the work required in each of the five years, comprising a number of visits and the preparation of a report. He provided for 12 day-long visits by a qualified horticulturalist in year 1, 6 in year 2, 4 in year 3, and 2 in each of years 4 and 5. The horticulturalist would inspect and check on plant health and structure, water and feed the garden, prune the trees, test the soil and keep records. Mr Harrison saw the respondent’s obligation as being care rather than maintenance, which he took to mean deciding and reporting on what needed to be done rather than doing the actual digging and so on, on the basis that the claimants would have incurred the cost of practical gardening work in any event. He made no provision for the replacement of plants that failed, on the basis that that would be covered by the overall contingency figure for garden costs (see paragraph 198 above), and no provision for mulch on the basis that the agreed restoration of the garden allows for 100mm depth of mulch which is twice the usual amount. We do not have Ruskins’ actual quotation so we cannot see what is allowed for those items within it.

Mr Wilson on the other hand offered Ruskins’ quotation, in which the principal difference from Mr Harrison’s plan was the number of visits. Ruskins quoted for two people – a horticulturalist and an arborist – to visit the property for 16 days in each of the first two years (fortnightly from April to September and then four visits during autumn and winter), then three weekly visits in the next three summers with three over winter, making 60 visits. Ms Jackson who cross-examined Mr Wilson suggested that he was advocating a Rolls Royce service, and he very fairly agreed. His figure of £69,930 is a reduction from Ruskins’ actual quotation of £81,348 on the basis that he thought the cost could be reduced if the work were handed over to a less expensive company in year 3, with Ruskins providing training for the successor company.

Even with that modification we take the view that Mr Wilson has indeed provided a Rolls Royce service. It is telling that Ruskins in correspondence with him said “we will have almost a parental relationship with these specimens”. At the heart of the difference between the two experts is the level of visiting for which they provide. Ruskins provide for fortnightly visits by two qualified professionals throughout the first two summers and every three weeks for the next three summers; Mr Harrison realistically pointed out that the claimants would have to be gardening, or paying a gardener, in any event and it is not the respondent’s obligation to relieve them of that cost. On the other hand, Mr Harrison agreed in cross-examination that his horticulturalist would call in an arborist where necessary and that his plan made no provision for the cost of that. We fail to see the need for a qualified arborist to spend anywhere near the number of days provided for in Ruskins’ quotation, but some provision for specialist attention to the trees is appropriate. Moreover we think that the phrase “cared for and monitored” in clause 4.2 of the Deed requires rather more hands-on work than Mr Harrison provides for, and that his plan does not go far enough beyond the meaning of “monitored”; but on the other hand we take Mr Harrison’s point that these plants will be taking root in fresh soil and absent the unforeseen there is no reason why they should not do well without an unusual level of nursing.

It is also important that Ruskins is a firm within reach of the claimants’ property and is willing to do the work. On that basis, in our judgment the starting point should be Ruskins quotation, but without the extravagant provision for the work of an arborist and without so many days’ visiting overall. Ruskins’ quotation should be scaled down to provide for 50 days’ visiting by one person rather than 60 by two. The actual frequency of the visits and the division between a horticulturalist and an arborist can appropriately be decided as time goes on and by reference to the conditions and to the health of the plants and trees. The calculation of the reduction to be applied is difficult but as Ms Jackson pointed out in cross-examination each day’s care within the Ruskins’ programme (total cost £69,930) costs just short of £1,165 on the basis of two people visiting, which is just over £580 per person. On that basis we are reducing the days of care provided from 120 (= 60 x 2) to 50, we are reducing the cost by £580 x 70, i.e. by £40,600. Taking that from Mr Wilson’s total, rounded to £70,000, we allow £29,400 for the after-care plan.

In line with the approach we explained above (paragraph 48) since the parties have agreed that an allowance for inflation is required, we adopt the lower of the two experts’ figures, Mr Harrison’s £48,974 + VAT. We see that some of the prices incorporated in the agreed £728,828 are 2027 prices, so the claimants are abundantly protected from future price rises.

The total payable is therefore £728,828 + £7,200 + £29,400 + £48,974 = £814,412, which we round to £814,500.

It will be recalled that there is a swimming pool at the north end of the garden. It is a Fox pool, built of steel panels with a liner, installed in the 1980s and re-lined in 2004. It was in use up to 2012 but was not used thereafter because it needed to be serviced and it was obvious that there was no point in having that done at a point when the respondent’s works were about to be done. The pool played an interesting evidential role in the 2023 hearing and decision; it is of a kind that cannot be installed, or re-lined, when groundwater levels are above the level of the deep end of the pool, which is about 1.6m AOD. When it was re-lined in 2004 it was drained and then left empty and unlined for three weeks; and that was an important piece of evidence for the groundwater levels at the property before 2014 (see paragraphs 129 to 136 of the 2023 decision).

It is agreed that the pool is now beyond repair.

The Deed said this:

“6.2

Prior to entry a pool specialist will survey the pool and share the results with the Owners.

6.3

On completion of the works any damage to the pool will be repaired to the same state as shown in the condition surveys under the supervision from the swimming pool specialist.

6.4

If the pool is damaged beyond repair then the pool will be reconstructed.”

Although it is agreed that the pool is beyond repair, the respondent says that it was nearing the end of its life in 2014 and would have deteriorated beyond repair by now in any event. Accordingly we have to decide whether the respondent is liable under the Deed for the provision of a new pool; if it is, we have to decide what that will cost.

Expert evidence about the pool was given by Mr Peter Corcoran of the Morton Partnership for the claimants and by Mr Phil Gordon of Jim Gordon Associates LLP for the respondent. Mr Corcoran has had a long involvement with the property and it will be recalled that he gave evidence at the 2023 hearing; as we said above he also gave evidence about the condition of the house for the 2026 hearing. Mr Gordon has been involved only since 2025.

The pool experts made a joint statement in which they recorded a considerable level of agreement. They agreed that the pool has been damaged by the downward movement and settlement of the deep (north) end of the pool surround, and by the displacement of the steel panels that form the wall of the pool, as well as cracking to the pool surround and movement of the facing wall. They agree that it is not practicable to repair it due to the extent of the movement and of the displacement, or separation, of the panels. In view of its situation next to a river they agree that it should be replaced by a cast in-situ reinforced concrete structure, and not before 6 to 12 months after the flood wall remediation works are carried out so that the groundwater levels have returned to where they used to be before 2014. They agree that the pool plant cannot be re-used.

The experts’ disagreement is only as to the state of the pool in 2014 and the cause or causes of the movement and displacement that mean it cannot now be repaired.

We found in 2023 that the pool had been damaged by the respondent’s work (paragraph 75 of the 2023 decision). The issue is whether, as an aged pool that was agreed to have been in poor condition in 2014, it would need to be replaced now anyway. Mr Corcoran’s evidence was that despite its age the pool could have been repaired. Fox pools are no longer made but Mr Gray of Grayfox Pools maintained them; in correspondence with Mr Corcoran he indicated that he was still maintaining a pool built in the 1970s.

Mr Gordon pointed to storm Xavier in 2013, the age of the pool, its construction, the materials used in constructing it, and the respondent’s works as five possible causes of its deterioration. But he agreed that the reason why it cannot now be repaired is the movement at the deep end of the pool and the separation of the steel panels. Ms Daly in cross-examining him took him to the available evidence for the condition of the pool in 2014.

There is certainly evidence that the pool was in poor condition before the respondent’s works took place. The Ridge report (see the 2023 decision) said that it was “in an aged and poor condition”, but deferred to the report to be provided by Mr Hollingworth of Oasis Swimming Pools. In that report in November 2014 Mr Hollingworth said:

“Pool water was cold and out of condition. No view of the pool bottom was possible and as no circulation was taking place I cannot confirm if all was in working order. …

The steel panels are original and when we have worked on the pool in the past, some degrading of the galvanising has and is taking place.

When last we replaced the liner in the pool the 50mm soft sand screed was in an acceptable condition although approaching end of life.

The preformed step unit has begun to de-laminate and spares for this are no longer available. Should replacement of the liner become necessary again in the future, then we would not expect the step unit to be re-usable again.”

Mr Hollingworth mentioned movement of the pool surround and retaining wall, but did not suggest that the pool itself had moved or settled, nor that the panels had started to separate. He concluded:

“A pool of this kind of construction has a limited life expectancy, whilst all looks to be in reasonable order and the pool, if cleaner, usable, I would consider that it nears the end of its life and will at some time in the near future require some considerable refurbishment or re-build. The pool plant and equipment is well installed, adequate and very usable. If the pool required some renovation or replacement I can see no reason why the circulation equipment and pipe work could not be re-used.

So while Mr Hollingworth regarded the pool as near the end of its life in 2014 he clearly did not think the end of the road had been reached. He still contemplated renovation, including replacement of the liner.

The Morton Partnership report in 2014 (see the 2023 decision) said nothing about the pool. In its updated report in 2016 it referred to deterioration of the pool surround and settling of the north-east corner, with a difference in level of some 70mm from one end to the other.

That is the first mention of settling or downward movement, and it appears after the respondent’s works had been carried out. There is no evidence of downward movement before that point, nor of separation of the steel panels. Finally the Polaris Pools Ltd report in March 2018 said:

“On inspection 6 and 21 February 2018, it is clear that there has been a serious amount of on-going subsidence to the pool surround and the actual pool itself since the Environment Agency (EA) Flood Defence works started in 2014. This is particularly evident in comparison to the photographs contained in the Oasis report of 19 November 2014 as referenced above.

The pool level at the deep end (nearest the river) has dropped significantly This is evidenced by the noticeable cracks and displacement to the concreted pool surround, and much widened gaps between it and the retaining wall, but most significantly subsidence of the pool itself.”

The experts agree that it is the settlement, or downward movement, of the pool and the separation of the panels that means it is beyond repair. There is no evidence of either before the respondent’s work. There is evidence that it was in poor condition by the end of 2014, but not that it was beyond repair. Given the sensitivity of this type of pool to groundwater levels (see paragraph 213 above) we find that the probable cause of the settlement of the pool and of the separation of the panels was the respondent’s works, which means that the respondent must pay the cost of removing and replacing the pool.

Evidence of the cost of a new pool was given for the respondent by Mr Gordon, not by the quantum expert Mr Steel. Mr Gordon thought that in 2025 the cost of removing the present pool and building a new concrete one would be £140,000 to £150,000, excluding VAT. He had not obtained quotations to support that view. The respondents’ case is that if they are found to have been responsible for the damage to the pool – as we have found – then £150,000 + VAT is payable.

For the claimants, evidence about the cost of the pool was given by Mr Dunne. He had obtained one quotation, from Beyond Swimming Pools, for £199,990 plus VAT. Mr Dunne explained that he had found one other company willing to quote, but they had pulled out. He said that in supplying rectangular swimming pools he was currently quoting £200,000 plus VAT, but that a kidney-shaped pool like the one at the property would be more expensive.

Mr Dunne’s final position was that the respondent’s liability for the pool should be £213,988 + VAT. He started from the quotation for £199,990 + VAT; added £1,000 for a new armoured cable from the house to the pool; subtracted £3,500 because the current pump still works, to get £197,490. He then added an allowance of 8.35% (using an index agreed with Mr Steel) to give a total of £213,988 + VAT.

We are unimpressed by Mr Gordon’s unevidenced estimate for the cost of the pool, and have to adopt Mr Dunne’s opinion. The respondent does not concede that a figure for inflation should be added, so we find that the respondent’s liability in respect of the pool is £197, 490 + VAT. We round this to £197,500.

We bring together the totals already calculated as follows:

Accordingly the compensation due to the claimants is £3,315,200, including VAT.