DOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL MAHONEY AP, MEAGHER and SHELLER JJA 19-21, 24-25 July 1995, 12 September 1995 [1995] NSWCA 128 NEGLIGENCE — construction of a flood levee duty of care — reasonable engineering standards causation — s582A of Local Government Act 1919 — defence of good faith. MINISTERIAL CORPORATION In April 1990 the town of Nyngan was flooded. The flood overtopped a levee which had been erected by the Bogan Shire Council (the Council) after seeking advice from the Water Administration Ministerial Corporation (WRC). The greater part of the levee was built to a height equal to the level reached by a 1976 flood plus one metre, referred to as the freeboard. The Tottenham Road Section and other parts of the levee including the Kokoda Trail and Bexon's Corner, in all about 1600 metres, were built to a lower height with a freeboard of 0.4 metre. During the days before the 1990 flood, sandbags were used to raise the height of these parts of the levee. Flood waters entered the town after the sandbag levee at the Kokoda Trail and Bexon's Corner were breached. The plaintiffs were residents of the town. They sued the Council and the WRC to recover damages in negligence for loss of and damage to their property. They submitted that the Council which had built the levee, and WRC which approved the plans and supervised the construction, were negligent in failing to build the whole levee, other than the Tottenham Road Section, to the higher level. The defendants denied the allegations of negligence and relied upon statutory defences based on good faith. On 10 March 1994 Cole J entered judgment in favour of both defendants. The appellants challenged the judgment in favour of WRC. Cole J described the allegation of negligence as amorphous as it failed to specify any standard of protection, unless it could be said that the standard was that required by "reasonable engineering standards". His Honour said that . if the duty was to protect against the 1976 flood of record, the question in issue was whether provision of 0.4 metre freeboard where the levee comprised a made road as a matter of engineering design constituted a compliance with that duty; . if the duty was to protect against some higher flood it was necessary to define that different flood of design. The appellants accepted that if the chosen design was one to protect the town against a flood up to the level of the 1976 flood, the levee complied with reasonable engineering standards. However, the appellants submitted that the levee was designed to withstand future floods exceeding the 1976 level. It was argued that the levee banks were designed not to withstand a particular level of flood but to provide reasonable protection on a cost benefit basis against future floods. HELD: (1) WRC, in exercising its statutory power pursuant to s11(1) of the Water Resources Commission Act 1976 placed itself in a position where it owed a duty to the residents of Nyngan to exercise due care in approving plans for and supervising the construction of the levee; Sutherland Shire Council v Heyman (1985) 157 CLR 424. (2) There was evidence upon which his Honour could conclude that: . the level of flood against which the levee bank was to provide protection was the 1976 flood; 2 UNREPORTED JUDGMENTS . the appellants had failed to discharge the onus of showing that the decision to have the 0.4 metre freeboard along the gravel road was not in accordance with accepted engineering principle at that time. The design was to deal with the 1976 level of flood with an added safety margin. This meant the levee was able to withstand a higher flood level and met current engineering standards. The reduced freeboard of 0.4 metre was part of this design. (per Mahoney AP) It was the duty of the designer of the levee to decide how high it was to be, taking into account considerations of safety and cost. The decision to take the last highest flood level as a starting point and to add a one metre freeboard was a reasonable accommodation of safety and cost and accorded with the standards of reasonable engineering practice. Although the selection of the last highest flood will not always be a proper basis for the design of a levee against future floods, in these circumstances, the 1976 flood was a proper basis for the design of a levee to protect against future floods. (3) There was no reason to disturb his Honour's conclusion that even if the levee bank had been built as originally designed the flood would have overtopped it, inundated the town and caused the same damage that it ultimately suffered. (4) The statutory defence of good faith provided for in s582A of the Local Government Act 1919, does not require the plaintiff to prove dishonesty; Mid Density Developments Pty Ltd v Rockdale Municipal Council (1993) FCR 290. Given ie conclusions on negligence, it was unnecessary for WRC to rely on the defence of good faith. ORDERS Appeal dismissed with costs. Mahoney AP In April 1990 the town of Nyngan was flooded. The flood overtopped a levee bank erected by the Bogan Shire Council ("the Council") and partly financed by a government body which ultimately became the Water Administration Ministerial Corporation ("the Corporation"). A number of residents of Nyngan ("the plaintiffs") sued the Council and the Corporation in negligence for the losses they suffered. After a long trial Cole J gave judgment for the defendants with costs. The plaintiffs have appealed to this Court against the judgment given in favour of the Corporation. No appeal has been pressed against the judgment in favour of the Council: the Court has been told from the Bar table that the plaintiffs have agreed that, if the Council will not enforce its order for costs, they will not appeal against that judgment. Accordingly, the judgment in favour of the Council remains. Thave had the opportunity of reading the judgment of Sheller JA. I agree with the orders which his Honour proposes and, subject to what I shall say, with his Honour's reasons. What his Honour has said enables me to go directly to what, in my opinion, are the essential difficulties which confront the plaintiffs' claim against the Corporation. 1. THE ISSUES: The plaintiffs' claim, as it was pleaded, fought and evidenced at the trial, potentially gave rise to a number of interesting questions. During the appeal, reference has been made to some of them. Ultimately at the trial and on appeal the plaintiffs' claim has been pursued on a limited basis in negligence; it is proper to deal with this appeal upon that basis. To this I shall refer. However, in addition to the difficulties which, on that basis, the plaintiffs' claim has been pursued on a limited basis of negligence; it is proper to deal with this appeal upon that basis, the plaintiffs' claim faced, there were in my opinion other difficulties. In order to prevent misunderstanding of what is, and is not, decided in the present appeal, it is necessary to refer to two of them. UWRQUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Mahoney APB First, the plaintiffs' claim against the Corporation was laid essentially in negligence. At the trial, there was discussion between the trial judge and counsel as to the basis upon which the plaintiffs desired their case to be considered. In order, no doubt, to make clear the basis upon which, at the trial, the claim was pressed, Cole J in his judgment said: "Tt was accepted by senior counsel for the plaintiffs that there was no positive duty upon the Council to build a levee. The plaintiffs' case was grounded upon an allegation of misfeasance, not nonfeasance. Notwithstanding the manner in which the case was pleaded, in final submissions, Mr Gross QC put the negligence of both the first and second defendants as follows: 'The first defendant in the present case was under an ordinary duty of care to the Nyngan residents in carrying out its design and construction activities in relation to the levee. The second defendant assumed a duty of care by embarking upon a course of conduct of supervision, control and advice. The consequence of its intervention was that it adversely influenced the design choices ultimately made by the first defendant in relation to Stage 2 of the contract, and caused the first defendant in respect of Stage 2 of the contract to deviate away from what was required by reasonable engineering practice. The result was a diminution of the protection provided to Nyngan residents by the first defendant which left the Nyngan residents with a lesser degree of protection that (sic) was required by those reasonable engineering standards which the second defendant knew needed to be applied.' This is an amorphous allegation of negligence because it fails to specify any standard of protection, unless it can be said that the standard is that required by 'reasonable engineering standards'. Implicit in any engineering work of the nature of levee banks is the selection of a design criteria, that criteria being the risk against which the structure is designed to protect. The defendants argued that proper engineering practice in 1979-81 reasonably required, or at least permitted, a body in the nature of the Council or WRC to select as the risk to be guarded against the flood of record. For Nyngan that was the 1976 flood. They argued that, in accordance with proper engineering concepts, to design a levee bank to resist such a flood of record it was necessary to have a freeboard above that flood height but that the freeboard whilst giving incidental protection to flood heights in excess of the flood of record was not designed principally for that purpose but was rather an ingredient within the levee design." At the commencement of the appeal, Mr Gross QC, for the plaintiffs, affirmed that the basis on which the plaintiffs' appeal against the Corporation was pressed was that recorded by Cole J. The case as so framed accepted, I think, that it was the Council ("the first defendant") which built the levee. On this basis, it was necessary to determine how far what the Corporation did was the result of (as I shall describe it) money considerations. The plaintiffs' suggestion was that the Corporation intervened and by its intervention "adversely influenced the design choices ultimately made by" the Council. One of the suggestions made was that the levee should have been erected with a freeboard of one metre throughout its whole length: it was, in some areas, constructed with a freeboard of only.4 of a metre. In considering such a case, it would have been necessary to consider how far what the Corporation did was done because of money considerations. 4 UNREPORTED JUDGMENTS Upon this basis, the claim in negligence against the Corporation depended upon the allegation that it assumed a duty of care to the residents of Nyngan and so to the plaintiffs and that, in breach of that duty, it led the Council, in building the levee "to deviate from what was required by reasonable engineering practice". One of the factors proper to be taken into account by the Corporation in deciding what kind of levee bank it was prepared to finance was the amount of money which, as a matter of policy, the Corporation (or perhaps the government of the day) felt should be made available for this purpose. Nyngan could have been protected against any flood of less than Biblical proportions if the Council had erected and the Corporation had financed a levee ten metres high. The Corporation, in the proper exercise of its power, would not have made money available for such a levee: it would have cost too much. It was open to the Corporation - it was, I think, its duty - to determine what was the degree of safety to be afforded to the town by the levee to be erected having regard to such calls as there were on its available money and, perhaps, the competing claims upon the money available for expenditure upon public works generally. The matter may be illustrated by an example adapted from the circumstances of the present case. The submissions suggested, inter alia, that the levee here in question would be in accordance with "reasonable engineering practice" if it was built, in general, to a height which was one metre above the height reached by the previous highest flood which had affected the town. This involved, I think, that (as far as the available information permitted) the varying heights to which the 1976 flood had reached at particular places round the town should be ascertained, that the levee bank should be erected there to that height plus one metre, and that otherwise the works be carried out in a proper way. But there was a qualification to this which was of significance in the present case. That qualification was to the effect (I state the matter generally and therefore to an extent inaccurately) that where the levee was "incorporated with a constructed road formation providing stable foundations and permanent access... a freeboard of.4 metre above design flood level was appropriate". (I have quoted from the Report which the court-appointed experts, under the chairmanship of Mr D Dole, provided to the trial judge). It would, of course, have been possible for the Council to build and the Corporation to finance a levee with a freeboard above the design flood level of one metre at all places: it would, indeed, have been possible to provide for a freeboard level of two metres and more. To do so would, of course, have reduced the risk of the floods overtopping the levee. Such increases in the design level of the levees were not, I think, suggested by the Council or by the Corporation: at least, there was no suggestion of a two or three metre freeboard. But, in what it did, both the Council and the Corporation took into account the additional cost which would be involved, not merely in the much higher general levels of the levee but also in, for example, erecting a levee with a one metre freeboard at all places. If it had been necessary to pursue to a conclusion matters of this kind, it would, in my opinion, have been necessary for the Court to decide to what extent the Corporation and the Council were influenced in adopting the design that they did by money considerations. As at present advised, I believe that each body was influenced by money considerations in adopting the design and building the work as it did. On that basis it would have been necessary for the Court to determine UWRQUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Mahoney APB the extent to which it was proper to take into account money considerations in this way and whether money considerations would have justified the taking of the risk of the overtopping of the levee which the adoption of the present levee design involved. The significance of money considerations in determining the existence of a duty of care in negligence and in determining whether that duty has been breached is, in my opinion, a matter of considerable importance. In recent years, the courts have, by their decisions, clarified and perhaps extended the nature and extent of the duty in negligence which councils and other statutory corporations have to those affected by what they do. The expenditure which they must incur has, I think, been increased by the liability imposed upon such bodies in respect of the provision by them of information or similar services: compare the Shaddock litigation (1981) 150 CLR 225; by the extent of the duty of care now imposed in relation to swimming areas and other public facilities: cf Nagle v Rottnest Island Authority (1993) 177 CLR 432; and by the increase in the duties of councils in the exercise of their power to grant or refuse development and/or building approvals: cf Sutherland Shire Council vy Heyman (1985) 157 CLR 44. What they do will no doubt be affected by the extent to which they desire to and can provide money to do these things. It is, in my opinion, proper to record that the Court's decision in this matter does not involve the determination that the Corporation owed or breached a duty of care of the kind alleged in this proceeding. The second matter of difficulty arises from the fact that, there being no appeal against the judgment of Cole J in favour of the Council, the present appeal has been conducted upon the basis that what the Council did was not negligent. Upon the case as Cole J recorded it, the claim against the Council was that it owed a duty of care to the plaintiffs "in carrying out its design and construction activities in relation to the levee'. It was held at the trial that it was not liable in negligence in that regard. I shall not pursue the distinction between a finding of no duty and a finding of no breach of the duty. It is sufficient to note that the case against the Corporation was that by its "intervention" it "caused the first defendant... to deviate from what was required by reasonable engineering practice" in the relevant respects. I find difficulty in concluding that, if the judge's findings and verdict for the Council are to be accepted as correct, a finding should (or, perhaps, can) be made against the Corporation. The finding for the Council, in the factual context of the present case, would appear to involve that it did not depart from "what was required by reasonable engineering practice". If that be so, then what was erected did not depart from "what was required by reasonable engineering practice" and accordingly, whatever the "intervention" of the Corporation produced, it did not produce a departure which constituted negligence in relation to the levee. In view of the conclusions which otherwise the Court has reached, it is in my opinion not necessary to pursue this aspect of the appeal in detail. It is sufficient to record, as I do, that as at present advised this aspect of the matter imposes difficulty in principle on the case now pressed for the plaintiffs. And, in my opinion, upon a full consideration of the factual basis of the claim now pressed against the Corporation, I would not conclude that what the Corporation did caused the Council to depart from reasonable engineering practice. On that basis, as at present advised, I would find for the Corporation. 6 UNREPORTED JUDGMENTS It is proper to refer to two other matters. As I understand the judgment of Cole J, he was not satisfied that, if there was a duty of care and a breach of that duty, that breach caused the damage of which the plaintiffs complain. Reference has been made to this aspect of the matter in the judgment of Sheller JA. I agree with what Sheller JA has said in this regard. lagree also with his Honour's observations in relation to the "statutory defence of good faith". I do not desire to add to what was said by his Honour or by Cole J in this regard. 2. DEPARTURE FROM REASONABLE Engineering PRACTICE: A great part of the argument in this appeal was directed to what reasonable engineering practice required as to the design of the levee and whether what was constructed conformed with that practice. As I differ, to an extent from the approach adopted by Sheller JA in this regard, I shall indicate the view which, in principle, I have taken in relation to it. I shall confine what I say to the design of the levee. There were suggestions m the evidence and in the argument that the Council, in what it had its contractor build, departed from the design of the levee in that, at certain points, it allowed the levee to be constructed of a lower height than the design required. Counsel have placed before the Court diagrams showing the height of the levee above ground level at various places, its height above the level which the water attained in the 1976 flood, the position of the relevant roads, and the places where, it was contended, the height of the freeboard was less than the one metre or.4 metre respectively which the design required. It is not necessary, I think, to deal with the actions of the Council or the Corporation in this regard. The unappealed verdict for the Council establishes that the Council was not negligent in what it did, in this or other respects. Accordingly, I shall confine what I say to matters concerning the design of the levee. I shall, in this regard, assume that the Council and the Corporation each had a duty of care in negligence in relation to the design of the levee. On these assumptions, it was the duty of those designing the levee to decide, inter alia, how high it was to be. In deciding how high the levee was to be, the designer had to exercise judgment. If he knew in advance what the height of the highest flood which would thereafter occur would be, he would of course design a levee of such a height as would cope with that flood. He could not know that. Accordingly, the height selected was one determined by his judgment. But that judgment was one of some complication: the factors to be taken into account were of different kinds. Over-simplified, the choice involved the reconciliation of two things: safety and cost. The higher the levee, the greater the safety and the greater the cost; the lower the levee, the more likely it was that a flood would occur which would overtop the levee, but the cost would be less. The design which was selected was, in my opinion, influenced by money considerations. This may be inferred from the terms of the correspondence between the Council and the Corporation. In addition, the Council, in deciding what its own contractor was to build, was influenced by the cost of the levee and the consequent effect of that cost upon the ratepayers to which it was responsible. The design which the Council adopted and which the Corporation determined to finance involved, inter alia, three things: the adoption of the 1976 flood levels as the basic level; the adoption of a freeboard of one metre; and the acceptance that the freeboard might be reduced to.4 of a metre where, in the sense to which I have referred, roadworks were involved. It was these matters to which, qua design, attention was mainly directed. UWRQUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Mahoney APY Before coming to these three aspects of the design, one more general comment may be made. What I have said relates to the general approach or principle adopted in the design of the levee. I confess to some difficulty in ascertaining definitively what precisely the plaintiffs allege to be the departure from reasonable engineering practice charged against the Corporation. One matter of which they complained was, I think, the approach or principle adopted in determining what was to be the height of the levee. One way in which the case for the plaintiffs was put was that the Corporation erred in causing the Council to adopt a design which involved a height determined by the height of the last highest flood (the 1976 flood) with a freeboard of one metre (or up to one metre). The suggestion was, as I understand it, that to adopt that approach was not reasonable engineering practice. There is, to a person not skilled in engineering, some cogency in the argument that, if an engineer be given the task of designing a flood levee of the present kind, it is wrong for him merely to adopt the approach that the levee should be one which, in the sense to which reference was made in argument, merely protects the town against the last highest flood and the incidental incidents of a flood of that height. Ordinary prudence would, perhaps, require that there be a measure of protection against a higher flood. This in turn requires an examination of the function of the freeboard and of that against which it is designed to offer protection. The term freeboard was used in the evidence (I take the description used in the Report made in respect of the Nyngan flood) as "that incremental difference in height between the height of the flood the levee is designed to protect against and the design crest height". The approach adopted involves a decision be made as to the level of protection to be given by the choice of the height of the water reached in a previous flood: in this case, the 1976 flood was chosen, it being the highest recorded flood level. The approach involves, in addition, the construction of a freeboard, ie, it involves that the levee be built higher than the previous flood level to the extent of the freeboard. The justification for the addition of a freeboard, as appearing in the expert evidence, was, in my opinion, mixed. On the one hand, the experts referred to the fact that, where a levee is constructed to the height of the last highest flood, it must take into account factors such as waves, accidents, variations in height at different places, and matters of that kind. Thus, as the evidence suggested, this approach to design assumes that the level which the flood reached at various places can be accurately measured; this, of course, is not necessarily true. It assumes that the level of flood waters at the surface will be the same at different places on the same level and that the waters will be flat. In fact, winds, water flow and other factors cause waves on the surface of flood waters and accordingly the height of the flood water at the levee at particular times will vary according to the waves. In addition, accidents may occur. One matter referred to in the evidence, was the necessity to put additional sandbags on the top of the levee. The suggestion was, I think, that the design of a flood levee envisaged that those safeguarding it during the flood would find that, at various times or places, the flood waters would be temporarily or permanently higher than the levee. The suggestion was that it would be necessary to increase the height of the levee above the designed height by the addition of sandbags and that this was an integral part of the design expectations. 8 UNREPORTED JUDGMENTS And, I think the suggestion was, the designed height of a levee should take account of the fact that accidents or other factors might result in those defending the levees being from time to time unable to place on the top of the levees the required quantities of sandbags. These and other factors were advanced in explanation of the function of a freeboard in the design of the levee. It was said for the plaintiffs that, on such an understanding of the function of a freeboard, a levee designed at 1976 flood level plus a freeboard would in substance protect the town only against a flood no higher than a 1976 flood. But there was, in the expert evidence, an additional factor. It was suggested in the evidence that the function of a freeboard was also to guard against the likelihood or "inevitability" of higher floods. On this basis, the levee design involved protection not merely against the last highest flood but, to the extent of the freeboard, against a somewhat higher flood. However this be, the suggestion for the plaintiffs was, I think, that to design a levee protecting only against the last highest flood or giving only such limited protection against a flood higher than the last highest flood, was a design not in accordance with reasonable engineering practice. Reference was made to the "inevitability" of higher floods - as the 1990 flood was - and it was said that accordingly the levee designed by the Council and accepted by the Corporation did not satisfy the Corporations's duty of care. I accept that - money considerations apart - the duty of the Council (if it had a duty of care) was to design a levee which would provide proper protection against future floods and that a design which determined its height merely by a rule of thumb would not satisfy that duty. But I do not think that that was what the Council did. As I have said, an engineering design of this kind involves an accommodation of safety and of cost. A design is not negligent because, in the event, it does not achieve the purpose to which it was directed, viz, keeping the flood waters from the town. Accordingly, in this context the question becomes whether an approach which, to achieve this effect, selects the last highest flood as the appropriate starting point and then adds a one metre freeboard is a reasonable accommodation of safety and cost in this regard. That, in my opinion, was what the experts saw as reasonable engineering practice and what, inter alia, the Corporation sought to achieve when, as the letters of 30 October 1976 and 30 November 1978 indicated, it had adopted as its approach to design a levee having one metre freeboard above the highest recorded flood level. The rule of thumb was adopted, not as such but because, in the judgment of the engineers, it provided a proper basis for securing an accommodation between safety and cost. The second matter to be referred to in this regard is that, to the extent to which this approach or principle was adopted, it was accepted by most if not all of the expert witnesses as proper for determining reasonable engineering practice. This, I think, emerges from the report of the Dole committee; it was accepted by the judge as the basis of the expert evidence. I see no error in what the trial judge did in this regard. In any event, I do not think that this Court should, on a rehearing, depart from the estimate which the trial judge formed of the expert witnesses who gave evidence before him. Upon this basis, I am satisfied that as far as concerns the height of the levee, the design of it accorded with reasonable engineering practice. I have had regard to the evidence in this regard of, inter alia, Professor Pilgrim. If and insofar as Professor Pilgrim's evidence suggested that the approach or principle here URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA) 9 adopted was not in accordance with reasonable engineering practice, the trial judge did not accept his finding. There was evidence from other experts which supported the approach or principle adopted. His Honour saw the relevant witnesses before him in evidence. As I have indicated, I do not think the Court can or should depart from his Honour's assessment of them. But, considering the expert evidence merely by reference to the material appearing in the appeal papers, I am not satisfied that it has been established that that approach or principle was wrong or not in accordance with proper practice. No doubt that selection of the last or highest flood levels will not always be a proper basis for the design of a levee against future floods: the unthinking adoption without more of a previous level would in principle be wrong. But that was not suggested to be the true basis in this regard. Such a suggestion was not, I think, pressed upon the expert witnesses. The 1976 flood was such a flood as, apparently, experts saw as a proper basis for the assessment of the height of future floods and for the design of levees against them. The evidence does not establish that this was not so. I am not satisfied that any alternative approach or principle for the design of the levee has been shown to be preferable to that here adopted. As I have said, I have dealt with the matters to this point by reference to the design of the levee. Reference was made in the evidence to aspects of the construction work done on the levee and what was done during the flood. For reasons to which I have referred, I do not think that such matters are open upon this appeal: the issues were defined and argued differently. If I be wrong in this, I do not think that the plaintiffs can succeed on any basis arising from such matters. Assuming that, in the design or in the construction of the levee, the assumption was involved that the levee banks would be strengthened by sandbags or the like during the course of a flood, I see nothing in what occurred to warrant the conclusion that the Corporation was in breach of its (assumed) duty of care. The extent to which it was involved in the supervision of the building of the levee or was aware of the details of it was minimal. I do not accept that, by what it did or failed to do in that regard, it was in breach of any such duty of care. I have dealt with these matters at length in deference to the arguments of Mr Gross QC, for the plaintiffs, and because of the significance which the decision in this appeal may have for those damaged by the flood. Buts in the end, I am satisfied that the conclusions reached by Cole J were correct. I agree with the orders proposed by Sheller JA. Meagher JA I agree with Sheller JA. Sheller JA INTRODUCTION The Bogan River flows from south to north a short distance west of the township of Nyngan. In the evening on 23 April 1990 flood water from the river broke through the temporary sandbag levee built on a section of the permanent flood levee protecting the town. In the next six hours water inundated the town causing considerable damage to public and private property. About 2,300 people were evacuated by helicopter. Fortunately no lives were lost. Three weeks passed before permanent re-occupation of Nyngan could begin. Nyngan is in mid-west New South Wales at the lower end of the Bogan River catchment area, which stretches from 200 kilometres to the south and south east over an area of about 18,000 square kilometres. Floods have been recorded on occasions since 1887. The 1990 flood reached 5.23 metres on the town gauge and 10 UNREPORTED JUDGMENTS was the highest on record. Previous record floods in February 1955 measured 4.42 metres and January 1976 4.33 metres on the town gauge. In the wake of the January 1976 flood the Bogan Shire Council (the Council), after seeking advice and financial assistance from the Water Administration Ministerial Corporation (then called the Water Conservation and Irrigation Commission and later the Water Resources Commission), decided to build a flood levee over six kilometres long on the western, southern and eastern sides of the town. For convenience I shall refer to the Corporation as WRC. Nyngan is crossed in a straight line from south-east to north-west by the embankment of the main western railway line between Dubbo and Bourke. The Mitchell Highway runs parallel to the railway. The western levee was divided into three sections, the Airport Section, north of the railway line, and the Golf Course and Tottenham Road Sections, south of the railway line. The southern levee was divided into the West and Central Sections (in part comprising Boundary Street) and Cemetery Sections (Oatley Street). The eastern levee, which extended north until it reached the railway embankment, was divided into the Showground (Bexon Street) and Grain Store Sections. Between the West Section and the Central Section the southern levee diverted in a loop to the south to include the electricity substation. The levee enclosed Nyngan south of the railway line. North of the railway line the Airport Section ran in a northerly direction for approximately 1.3 kilometres. The town north of the railway line was not enclosed, was on higher ground and had not been so much affected by earlier floods. The original 1978 plans specified that the levee, other than the 800 metre Tottenham Road Section, be built to a height equal to the level reached by the 1976 flood plus one metre, referred to as the freeboard. The plans specified a freeboard of 0.4 metre for the Tottenham Road Section. The greater part of the levee was built to the higher level but, in addition to the Tottenham Road Section, other parts of the levee, in all about 1600 metres, on the south western corner of Tottenham Road and Boundary Street in the West S(the dog leg), along Boundary Street south of the showground in the Central Section (the Kokoda Trail) and along the right angle formed by Bexon and Oatley Streets on the Showground and Cemetery Sections (Bexon's Corner) were built to the lower height with a freeboard of 0.4 metre. During the days before 23 April 1990 volunteers from the town and beyond used sandbags to raise the height of these parts of the levee. Flood entered the town after the waters broke through the sandbags at the Kokoda Trail and Bexon's Corner to an extent which made further repair impossible. PROCEEDINGS The plaintiffs were residents of the town. They sued the Council and WRC in the Common Law Construction List to recover damages in negligence for loss of and damage to their property. Apart from denying the allegations of negligence, the defendants relied upon statutory defences based on good faith. Cole J, as he then was, heard the proceedings in February 1994 and on 10 March 1994 entered judgment in favour of both defendants. The plaintiffs appealed against the judgment in favour of the second defendant, WRC. THE PLANNING AND BUILDING OF THE LEVEE Put in a simple form, the residents' claim was that if the whole of the southern and eastern levees had been constructed to the higher level with a one metre freeboard, instead of the lower level with a 0.4 metre freeboard, the residents could and would have placed sufficient sandbags on the permanent levee to have URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA) 11 defended the town against inundation even if the 1990 flood had overtopped the permanent levee. The plaintiffs argued that the Council, which had built the levee, and WRC, which approved the plans and supervised the construction, were negligent in not building all sections of the permanent levee, other than the Tottenham Road section, to the higher level and providing the residents with a "platform" upon which they could build temporary levees to defend the town against the flood. The plaintiffs accepted that the 800 metre Tottenham Road Section was properly and reasonably designed and built to the lower level. The proposals to build the levee, first advanced after the 1955 flood, were revived after the 1976 flood. On 25 March 1976 Mr RC Forgie, then the Shire Engineer, reported to the Finance and Works Committee that an estimate of the cost to construct permanent levee banks, having a minimum crest width of eight feet, three to one batters, and two feet freeboard above the level of the 1955 flood, had been prepared in the sum of $140,000. On 11 June 1976 the Council wrote to WRC stating that it had decided to construct permanent levees around the town of Nyngan to keep flood waters from the Bogan River out of the town and wished to make application to the Commission for a subsidy to carry out this work. WRC wrote back in July 1976 saying that in order that consideration might be given to the matter it would be necessary for Council to submit to the Commission a firm proposal and an estimate of cost of the work and that: "The normal minimum standard of construction for levees required by the Commission, when a grant is involved, is for the crest height to be a minimum of one metre above the highest known flood, the crest width of three metres and batters of one on two on the inside face and one on three an the outside face of the levee." The Commission also required the Council to apply, under PtII of the Water Act 1912, for a licence for the levee. On 16 August 1976 in response to a request from the Council for standard levee plans and flood gates on culverts through levees, the WRC wrote that it had no standard plans and "the design of such or the use of proprietary makes being left to the discretion of the Shire Engineer subject of course to the approval of the Commission when a grant is involved." At a Council meeting on 17 February 1977 the plans and estimate of cost of the levee banks were tabled for the Council's information. On 30 March 1977 the Council provided these to WRC. In its letter the Council referred to the history of floods. The height reached was greatest in 1955. Mention was made of the State Government's interests to keep flood waters out of the town and the Council's belief that it was unsatisfactory and uneconomical to put up temporary levees each time there was a major flood. At about the same time the Council applied for a licence under the Water Act. The application was not in evidence but in WRC's licensing inspection report the purpose of the work was stated as the prevention of inundations of land by flood waters. Under the heading requiring full details of the proposal, the approximate maximum flood level was shown as RL [reduced level] 173.65, the level of the 1976 flood. On 15 June 1978 the Council resolved that the matter be urgently pursued. In June 1978 David Mittelheuser, a WRC engineer, went to Nyngan and met Bob Whatham who had become shire engineer in March 1977. They discussed the alignment of the levee. Mr Whatham questioned the necessity for a levee on the south east of the town. According to his evidence, Mr Mittelheuser said he could give him an outline of the Commission's requirements for the proposed levee and some advice regarding the construction and funding of the works. The 12 UNREPORTED JUDGMENTS Council would have to submit a completely new proposal if it wanted funding. "The new proposal must include extension of the levee to the south and east of the town. A further cost estimate will have to be resubmitted. The extension to the south and east is necessary so that there is a one metre freeboard above the highest known flood level, over all sections of town." Mr Whatham was not able to say which was the flood of record. Mr Mittelheuser said it would be necessary to clarify which flood was the highest and WRC would undertake further investigations. He then said: "The freeboard will need to be the highest known flood level whatever that height was, plus one metre. Later Mr Whatham inquired whether WRC required the levee to have a freeboard of one metre where the levee was constructed as a roadway. Mr Mittelheuser replied: "No. In those places the levee can have a lower height of 0.4 metre freeboard as the width of the road will form an additional barrier against wave action. Questioning why WRC insisted on the levee having a one metre freeboard on the eastern part Mr Whatham said that in that area the waves would be broken up by the timber and any effect the waves might have had would be neutralised. Mr Mittelheuser said: "So far as Iam concerned Council will have to construct the levee having a one metre freeboard except where the road forms part of the levee." On about 11 October 1978 a WRC officer, either Mr Mittelheuser or Peter Wem, informed Mr Whatham that the Commission was satisfied that the 1976 flood was the flood of record. Mr Whatham asked if WRC was still insisting on a one metre freeboard to the east of the town. The officer replied: "You must submit the plans with a one metre freeboard as a standard requirement for all areas except where it forms part of the road. That includes the eastern section as allowance has to be made for wave action there as well. If Council considers the freeboard to be excessive then this fact should be pointed out in an accompanying letter to WRC. Another Council has raised a similar objection to that requirement. We are considering that Council's objection now. We will consider Nyngan's position once you put it in writing." Mr Whatham stated that having discussed the matter on numerous occasion with more than one officer of WRC it was his understanding that in circumstances where a roadway formed part of a levee including but not limited to that section where Tottenham Road formed part of the levee, there was no need to have the levee with the height of one metre above the highest known flood because the wave action would not be as severe where there was a road. In circumstances where there was a road the levee needed only to have a height of 0.4 metres above the 1976 flood level. On 19 October 1978 Mr Whatham reported to the Finance and Works Committee that the design for the proposed levee banks had been extended to provide protection on the eastern side of the town. "The design of the extension provides for a freeboard of one metre to the top of the bank above high flood level, even where the total bank height is only 1.2 or 1.3 metres. This is considered quite unnecessary for that section as the purpose of the freeboard is to allow for wave action and the timber in that area would prevent wave action from developing to any extent. With the metre freeboard the levee bank would join onto the railway embankment with the top of the levee at a higher levee than the rails." URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA) 13 He referred to the Commission's suggestion that, if Council considered the freeboard in this section to be excessive and resulting in unnecessary cost, this fact might be pointed out in an accompanying letter. The Committee recommended to Council that provision be made for a half metre, instead of a one metre, freeboard on the extended section at the eastern side of Nyngan. On 30 October 1978 the Council wrote to WRC protesting against the requirement to have one metre of freeboard on the eastern side of Nyngan, firstly because ie area was well timbered and any floodwater in that location would be very shallow so that wave action would be minimal and secondly because the levee bank with the metre freeboard would be approximately 0.5 metres higher than the railway embankment at which it terminated. The Council requested that the requirement be altered to provide a freeboard of 0.5 metres on the eastern side of Nyngan. On 23 November 1978 WRC replied as follows: "Because of the major expenditure by the Government on works of this nature it is necessary to ensure that the standards of construction are high and that the likelihood of the levee being overtopped and destroyed leading to catastrophic damage and possible loss of life is minimised. Whilst the Commission is aware that the area in question is well timbered this may not always to be the case and when considered with the inevitability of higher floods, Council will appreciate that relaxation of levee standards in the east of town would represent weakness in the levee system as a whole. Accordingly the Commission has recommended to the Minister of an approach being made to the Treasurer seeking his approval of a State grant of $186,000 being made available to Council, representing two- thirds of the total estimated cost based on the full one metre freeboard over the entire length of the levee." Cole J remarked that the letter was not quite accurate because it was never intended that the Tottenham Road section should have a freeboard in excess of 0.4 metre, that being at that time, according to Mr Whatham's evidence, the only portion of the levee comprising a roadway. On 21 December 1978 the Council noted the contents of WRC's letter and set about obtaining approval from land owners who were told that the levee would have a height of one metre above the highest known flood. On 21 March 1979 WRC advised that a grant equivalent to two-thirds of the cost of the levee had been approved subject to the Council's entering into an agreement with the Commission. The schedule to the agreement referred to plans showing a levee with a freeboard of one metre except at Tottenham Road. On 21 June 1979 Council resolved to execute the agreement and shortly thereafter did so. The agreement recited that the Commission had agreed to make available to the Council out of public moneys a contribution of two-thirds of the actual cost or two-thirds of $279,000 whichever was the lesser. CL1 provided that the Council should carry out and complete the work to the satisfaction in all respects of the Commission and the standard of work and the rate at which the work was to be carried out and completed should at all times be subject to the control of the Commission. CL7 provided that nothing in the agreement should in any way be construed or be capable of being construed as imposing upon the Commission any responsibility for the carrying out or construction of any part of the work but notwithstanding the terms of the agreement the whole of the responsibility for the carrying out of the work and for any consequences thereof should be and remain solely that of the Council. Maintenance of the works was the responsibility of the 14 UNREPORTED JUDGMENTS Council (CL8). The works were to be carried out by the Council subject to such supervision by officers of the Commission as the Commission might require (CL9). Although the work was to be carried out generally in accordance with the alignment shewn on the Council plans which extended to the full length of the proposed levee, the Council, for financial reasons, chose a staged construction. Stage | was for the western levee to the point where it joined Tottenham Road. On 1 October 1980 the Council wrote to Ralph Hunt MP. After referring to the plans and the WRC approval, the letter continued: "When these levees are completed, the town will be protected from any flood which reaches the highest known level in the Bogan River in the past." On 16 October 1980 the Council accepted the tender of Ric Laing Excavations. WRC approved. Further plans and specifications were prepared for Stage 2. Because of objections from landowners and others, the Council decided to build what became known as the Showground and Cemetery Sections as roads, Bexon and Oatley Streets. On 5 August 1981 the Council wrote to WRC, for the attention of Mr Mittelheuser, enclosing a revised estimate of cost of the permanent levee. The letter stated that there had been minor variations due to a realignment of the eastern end of the levee resulting "in a reduction in earthwork quantities and a small increase in the number of culverts required." Two copies of the specification for Stage 2 of the construction prepared in accordance with the Commission's requests were enclosed. Mr Mittelheuser received the letter. A clerical officer in WRC noted "tender document and plan in folder behind file". Cole J found that Mr Mittelheuser received the letter and considered the plans and schedule of quantities for he altered the word "reduction" to "increase". His Honour said: "The plans for Stage 2 clearly showed to any competent engineer glancing at them that the realignment of the levee bank to the east of the electricity substation traversed roadways being part of Boundary Street, Bexon Street and Oatley Street, and that where roads were so traversed, the levee freeboard was reduced to 0.4 metres. There can be no doubt that the WRC officers received plans for Stage 2 showing this reduction. Subsequent to such receipt, officers of the WRC from time to time carried out inspections of the works and approved payment for them. The evidence is overwhelming that, explicitly or implicitly, WRC approved Stage 2 plans with the lesser freeboard where the levee bank traversed formed roads, and saw the levee constructed with such lesser freeboard." His Honour said that the reason the Stage 2 plans were drawn with the reduced freeboard where they traversed made roadways was because Mr Mittelheuser had told Mr Whatham that such a reduction was acceptable where the levee bank traversed formed roads. Mr Whatham instructed Radford Davies, a senior engineering assistant with the Council, that a one metre freeboard above the 1976 level was to apply where the levee was a bank but where the levee was a road, which was wide, a 400 mm freeboard would apply. Mr Davies drew the Stage 2 plans accordingly except for the dogleg at the corner of Tottenham Road and Boundary Street where a small section of the levee was drawn to have a freeboard of 0.4 metre because of the presence overhead of high voltage power lines. For similar reasons the freeboard was reduced on the eastern end of Boundary Street between the substation and Hoskins Street, part of the Kokoda Trail. In September 1981 the Council received tenders for Stage 2. On 26 November 1981 URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA) 15 WRC approved Council's recommendation to let Stage 2 to Ric Laing Excavations. The quotation was formally accepted on 26 January 1992. The work proceeded. Pivotal to Cole J's decision was a finding that the Council's decision was to build a levee designed to protect Nyngan from the recurrence of the 1976 flood and that it acted on advice from WRC that an appropriate levee would be of a height equivalent to one metre above the 1976 flood as the flood of record. There was an exception. Where the levee bank comprised a made road, WRC would provide funding even though the freeboard was only 0.4 metre. The appellants did not contend that the levee bank built was not of a suitable design to protect the town against a recurrence of the 1976 flood. The appellants challenged the finding that the levee bank was designed and constructed to do no more than keep out this particular level of flood. THE RISING FLOOD From 29 March to 6 April 1990 rain fell which wetted the Bogan catchment area. At 12.10 pm on Wednesday 11 April the Commonwealth Bureau of Meteorology issued a preliminary flood warning advising that "peak below moderate flood level (3.5 metres) next week". The gauge reading at noon that day was 3.05 metres. The warning was reissued with a slight change the next day. For the period from 13 April to 17 April the Bureau predicted a peak of 4.2 metres for Wednesday 18 April. At 3 pm on 17 April the gauge height was 4.01 metres. On Wednesday 18 April the Bureau's prediction had risen to "4.3 metres by tomorrow". At 1 pm the gauge height was 4.25 metres, the highest reading ever recorded. The Council appealed for volunteers to raise the levee height. On Thursday 19 April a Flood Emergency Committee met at the Nyngan Police Station. A further 0.13 metre increase in water level occurred over a twelve hour period and volunteer work on the levees commenced in earnest. At 9 am on Friday 20 April the river height was 4.54 metres. At 11.40 am the Bureau advised that flood levels would "remain in the 4.5 to 4.6 metre range for at least another week". The Committee met at 9.10 am and again at 4.20 pm. At 7 am on Saturday 21 April the river level was 4.58 metres. The Bureau's prediction was that levels would "remain in the 4.6 to 4.7 metre range for several days". At 1 pm on Sunday 22 April the river level was 4.6 metres and rising. At 1.05 pm the Bureau predicted a water level "peak near 4.8 metres during Tuesday 29 April". At the Emergency Flood Committee meeting at 10.10 am Mr Whatham reported that a 5 metre level was essential right around the levee "by tonight. This should be 0.3 of a metre above the expected height." Amongst the matters to be included in a press release was the statement: "The Bogan River has again started to rise". At a meeting held at 4.15 pm it was reported: "Should achieve 5 metre level of bags by tonight and work will continue after that". At 12.15 pm on Monday 23 April 1990 the Bureau issued a flood warning as follows: "The Bogan River at Nyngan was 4.74 metres at 10.40 am and is expected to peak at 4.85 metres overnight." From 12.40 pm to 3 pm Premier Greiner visited Nyngan and inspected the levee. At 2 pm water entered houses in the northern part of the town. At 3.30 pm the southern Central Section breached near the substation but was repaired. Between 3.30 pm and 4 pm three further breaches were repaired. The Mitchell Highway to the east of the town was impassable. By Monday afternoon, sandbags could not be placed fast enough to cover the rapid rise in flood levels. 16 UNREPORTED JUDGMENTS On the basis of the Saturday prediction of a flood level of 4.7 metres the levees were raised to a height believed to be sufficient to protect the town, that is to a level equivalent to 5 metres on the Nyngan gauge. The sudden accelerated rise in flood levels however overwhelmed all attempts to keep water out. According to 5 the Nyngan Flood Investigation Report over the previous four days the rise had been 0.11 metre, 0.03 metre, 0.01 metre and 0.09 metre. On 23 April the rise exceeded 0.5 metre. At 5 pm uncontrollable breaches developed in the Showground Section near Bexon's Corner, followed by additional uncontrollable breaches in the Central Section levee between the substation and Hoskins Street, 10 the Kokoda Trail. Work on the levees was abandoned. At 6 pm the Tottenham Road levee opposite Oatley Street breached. Within six hours the town was fully flooded. Overnight those residents who had left their homes sheltered in a picture theatre, the RSL Club and a hotel while others slept in their ceilings or climbed onto their roofs. The peak gauge height of 5.23 metres occurred about 2 am on 15 Tuesday 24 April 1990. At daylight the residents were airlifted by helicopter from the town. MEASURING THE FLOOD The measurements set out above were taken from the town gauge located immediately upstream from the Mitchell Highway bridge. Cole J directed that the 20 experts retained by the parties meet under the chairmanship of David Dole to provide to the Court a report specifying any issues of a technical nature upon which there was disagreement together with a brief summary of the opposing views in relation to those issues. On 9 February 1994 Mr Dole provided a final report to which all experts agreed. The report addressed, among other things, the 25 question whether the levee even if constructed in accordance with the original agreement, that is to say with a freeboard one metre above the level of the 1976 flood except in the Tottenham Road section, would have been defensible against the 1990 flood. The parties agreed that the translation of the peak level of the flood to a gauge height and thus to a reference level had inherent uncertainties 30 associated with it. The report identified these. There were differences between the parties as to how various hydraulic effects should be quantified. The parties agreed that river levels reached, at least, the following levels at the times specified: 35 | "Time (23 April 1990) Nyngan Town Gauge m River level near gauge AHF (m) 0600 hr 4.6 172.24 1400 hr 4.8 172.44 Midnight 5.0 172.74 40 [sic] Notes: Gauge zero taken as 167.64 AHD or 169.075 Bogan Shire Council datum. . The Department believes that the level at the Nyngan Town gauge could have been 5.2m as early as 2000 hrs on 23 April." AHD is the acronym for Australian Height Datum. 45 The parties agreed that the information shown on a plan the Council had prepared (Paterson Consultants 90.052 [sic] -Recorded 1990 Floor Levels) was an accurate record of flood levels. This plan (in fact Job No 93-052) became exhibit 9A. The parties also considered a report of a simulation Water Studies Pty Ltd produced in November 1993. This assumed levees built to the original 50 specifications (that is to say one metre above the 1976 flood level except in the Tottenham Road section) and with modelled flows modified by limiting the URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA) 17 maximum computed differences above the modelled April 1990 flood to 0.15 metre. 0.15 metre was an agreed compromise estimate of the increase in flood water level which would be brought about by the levee's excluding flood water from the town (the exclusion factor). Mr Dole reported: "The outcome was that additional heights above the originally specified levee would have been required to prevent over-topping. On a zero freeboard basis these heights would have been: . Western levees - up to 0.7 metre* . Southern levees - up to 0.5 metre . Eastern levees - up to 0.3 metre * Council advises that, in its view, the western levee would have been overtopped by up to 0.9 metre." In February 1994 Water Studies Pty Ltd prepared a revised report which was in evidence. Cole J, on the basis of additional evidence, found that at its peak the river reached a gauge height at Nyngan of 5.2 metres. The appellants did not seriously challenge this conclusion. In fact it was slightly low. The evidence suggested the peak level was 5.23 metres. The simulated studies showed that at a town gauge height of 5.23 metres the flood level would have exceeded the height of the levee built with a one metre freeboard at nine of thirteen locations around the levee by between 0.13 and 0.57 metre. If 0.15 metre, the exclusion factor, was added, the flood would have exceeded the height of a one metre freeboard at ten of the thirteen locations by amounts as low as 0.3 metre and as high as 0.72 metre on the western levee, as high as 0.32 metre on the eastern levee and as high as 0.48 metre on the southern levee, which is consistent with and explains the expression "up to" in the table in Mr Dole's report. Cole J said: "As exhibit 9A demonstrates and the computer simulations predict, and as all of the hydrological experts now accept,...... overtopping would have been at least up to 0.7 metre in some portions of the levee and down to 0.3 metre in others." The appellants took exception to the words "at least" but his Honour accurately stated by that expression what exhibit 9A and the simulations demonstrated. The appellants criticised two other passages in his Honour's judgment on this topic. His Honour said, after quoting part of the Dole report and referring to exhibit 9A: "Tt is to be recalled that the parties agreed (and the plaintiffs have pleaded) that with an assumed 0.5 metre flood, overtopping between 0.3 and 0.7 metre would have occurred. It would have been greater with a 5.2 metres flood, and with an additional 0.15 metre in consequence of the assumed scenario." The respondent accepts that this passage involves a misunderstanding of the agreement reached and recorded in the Dole report. The agreed heights ranged from between "up to" 0.3 metre on the eastern levee, "up to" 0.5 metre on the southern levee and "up to" 0.7 metre on the western levee. These heights assumed the actual level reached by the 1990 flood, and were not relevantly related to the separate agreement that river levels reached, "at least", 5 metres or 172.74 metres AHD at midnight on 23 April 1990. Accordingly, with respect, it does not follow from the Dole report that the overtopping agreed would have been any greater or indeed any different if the experts had agreed or assumed that the flood reached 5.2 metres on the town gauge. Moreover as appears from the part of the report which I have quoted the exclusion factor of 0.15 metre was included in the heights agreed upon. The appellants did plead in para(vii) of their "Additional Particulars of Negligence Against Both Defendants" in the amended statement of claim: 18 UNREPORTED JUDGMENTS "The additional heights of original designed 1979 design levels that would have been required, would have been, on a zero freeboard basis (as stated in the report of Dr Dole para5.4.42) on the western levees - up to 0.7 metres; on the Tottenham levees [sic] - up to 0.5 metres and on the eastern levees up to 0.3 metres." In other words the appellants conceded, though inaccurately described, that degree of calculated overtopping. Later in his judgment his Honour said: "Tt follows that even had the levee been constructed as designed but with a freeboard of one metre except at Tottenham Road, the town would, unless additional measures had been taken, have been inundated. Assuming a 5.0 metres flood, with no freeboard the overtopping would have been AT LEAST: Western levees up to 0.7 metres Southern levees up to 0.5 metres Eastern levees up to 0.3 metres With a flood of 5.2 metres height, or 5.3 metres height, the town would have been subjected to severe inundation from the north. The temporary levees in fact constructed would have been inadequate to repel water from the north as in fact they were." (emphasis added) The appellants argue that his Honour's misunderstanding of the agreement the experts reached and Mr Dole recorded led, when he came to the question of causation, to his substantially overestimating the gravity of the task the residents would have faced in resisting the flood if the levee had been built with a freeboard one metre above the level of the 1976 flood except on the Tottenham Road section. However the passage later in his judgment, which I have already quoted, shows that his Honour proceeded on the basis that the overtopping would have been "at least up to 0.7 metres in some portions of the levee and down to 0.3 metres in others." This is what exhibit 9A demonstrates. THE ALLEGED NEGLIGENCE At the time the levee was built s494 (1) of the Local Government Act 1919 empowered the Council to make provision - "(b) for the prevention or mitigation of menace to the safety of life or property from floods........." S11(1) of the Water Resources Commission Act 1976 provided that the Commission "in particular, shall have and may exercise and perform the following powers, authorities, duties and functions: (g) the investigation, evaluation, implementation and co-ordination of proposals for flood control and flood mitigation.......... (m) the formulation, co-ordination and implementation of plans for flood control works and the ascertaining of the nature an extent of land likely to be affected by those plans if implemented;" In Sutherland Shire Council v Heyman (1985) 157 CLR 424 at 459 Mason J said that it has been generally accepted that, unless ie statute manifests a contrary intention, a public authority which enters upon an exercise of statutory power may place itself in a relationship to members of the public which imports a common law duty to take care. For present purposes I accept that WRC owed a duty to the residents of Nyngan to exercise due care in approving plans for the construction of the levee and in supervising that construction. URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA) 19 In deference to some of the appellants' arguments based on reliance, both specific and general, I refer to some further parts of Mason J's judgment dealing with occasions when a public authority, vested with a power, may come under a duty to exercise it. At 459-60 his Honour said that, generally speaking, a public authority which is under no statutory obligation to exercise a power comes under no common law duty of care to do so. But an authority may by its conduct place itself in such a position that it attracts a duty of care which calls for exercise of the power. A number of illustrations followed. If a public authority in the exercise of its functions creates a danger it may subject itself to a duty of care for the safety of others which must be discharged by an exercise of its statutory powers. At 463 Mason J said: "The American experience........ furnishes support for the view that a public authority is liable for negligent failure to perform a function when it foresees or ought to foresee that: (a) the plaintiff reasonably relies on the defendant performing the function and taking care in doing so, and (b) the plaintiff will suffer damage if the defendant does not take care." His Honour referred to the interrelated question, one of several arising, of whether the concept extended to general reliance or dependence by those in the position of the plaintiff, as distinct from specific reliance by the plaintiff. At 463-4 he said: "Tn the case of a public authority, the foreseeability of the plaintiff's reasonable reliance is a sufficient basis for finding a duty of care, subject to such dispensations as may arise from the special character of a public authority exercising statutory functions..... If this be accepted, as in my opinion it should be, there will be cases in which the plaintiff's reasonable reliance will arise out of a general dependence on an authority's performance of its function with due care, without the need for contributing conduct on the part of a defendant or action to his detriment on the part of a plaintiff. Reliance or dependence in this sense is in general the product of the grant (and exercise) of powers designed to prevent or miximize a risk of personal injury or disability, recognized by the legislature as being of such magnitude or complexity that individuals cannot, or may not, take adequate steps for their own protection. This situation generates on one side (the individual) a general expectation that the power will be exercised and on the other side (the authority) a realization that there is a general reliance or dependence on its exercise of power; see Shapo, The Duty to Act (1977), at 95-96. The control of air traffic, the safety inspection of aircraft and the fighting of a fire in a building by a fire authority (but cf Bennett and Wood Ltd v Orange City Council (1967) 67 SR (NSW) 426 may well be examples of type type of function." L Shaddock and Associates Pty Ltd v The Council of the City of Parramatta (1981) 150 CLR 225 is a case in which a Council provided erroneous information about road widening proposals to a person who had contracted to purchase the land affected. The Council was held liable to the purchaser in negligence on the basis that the purchaser relied upon it and the Council should have been aware that it would do so. In Parramatta City Council v Lutz (1988) 12 NSWLR 293 at 306 Kirby P found the element of reliance to be present in the judgments of Mason, Brennan and Deane JJ in Heyman. At 318 Mahoney JA said that to establish liability, it was necessary that the plaintiff show that she relied on the advice given to her and acted on it, to her detriment. McHugh JA at 330 explained why in his opinion the Court should adopt the general reliance rule. However his Honour acknowledged 20 UNREPORTED JUDGMENTS that the reasons of Mason J in Heyman did not constitute the ratio decidendi of the case, were in conflict with the reasons of Brennan J and received no support from the judgments of Gibbs CJ and Wilson J. Deane J may only have endorsed the concept of specific reliance and not that of general reliance. In Bryan v Maloney (1995) 69 ALJR 375 at 377 Mason CJ, Deane and Gaudron JJ said: "The cases in this Court establish that a duty of care arises under the common law of negligence of this country only where there exists a relationship of proximity between the parties with respect to both the relevant class of act or omission and the relevant kind of damage. In more settled areas of the law of negligence concerned with ordinary physical injury to the person or property of a plaintiff caused by some act of the defendant, reasonable foreseeability of such injury will commonly suffice to establish that the facts fall into a category which has already been recognised as involving a relationship of proximity between the parties with respect to such an act and such damage and as 'attracting a duty of care, the scope of which is settled'." The appellants' case was one of misfeasance, not non feasance. They accepted that a decision whether to provide finance for a flood control or mitigation project was one of policy dictated, amongst other things, by the availability of financial resources. I did not understand the appellants' argument to be that each specifically or generally relied upon WRC to ensure that the Council built a levee which would have been, with the addition of sandbags, sufficient to withstand the 1990 flood, that such reliance imposed upon WRC a duty to ensure the Council built such a levee and that the Council's failure to do so, under WRC supervision, gave rise to a breach by WRC of its common law duty of care. If this were the argument it was not made out. The circumstance of a Council empowered to order the owner to demolish a building in a condition prejudicial to the property in, or the inhabitants of, the neighbourhood may, as this Court has held, require the Council to exercise the power, amongst other things, because of its relationship with its ratepayers. The position of a public authority concerned with flood control and flood mitigation throughout New South Wales in the way described in the Water Resources Commission Act is not comparable. In my opinion the question is a simpler one. Was WRC, having entered upon the exercise of its statutory power and having, therefore, placed itself in a situation where it was bound to exercise due care in doing so, in breach of that duty of care? Reliance, either specific or general, in the circumstances of this case, is of no real significance. Before Cole J, Mr Gross QC, who appeared for the appellants, put the negligence alleged against the Council and WRC in the following way which I have broken up into sections: 1. The Council was under a duty of care to the Nyngan residents in carrying out its design and construction activities in relation to the levee. 2. WRC assumed a duty of care by embarking upon a course of conduct of supervision, control and advice. 3. The consequence of WRC's intervention was that it adversely influenced the design choices ultimately made by the Council in relation to Stage 2 of the construction contract and caused the Council in respect of Stage 2 to deviate away from what was required by reasonable engineering practice. URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA)21 4. The result was a diminution of the protection provided to Nyngan residents by the Council which left the Nyngan residents with a lesser degree of protection than was required by those reasonable engineering standards which WRC knew needed to be applied. Cole J described the allegation of negligence as amorphous because it failed to specify any standard of protection, unless it could be said that the standard was that required by "reasonable engineering standards". His Honour said: "Implicit in any engineering work of the nature of levee banks is the selection of a design criteria [sic], that criteria being the risk against which the structure is designed to protect. The defendants argued that proper engineering practice in 1979-81 reasonably required, or at least permitted, a body in the nature of the Council or WRC to select as the risk to be guarded against the flood of record. For Nyngan that was the 1976 flood." Later his Honour said that the plaintiffs had considerable difficulty in defining or specifying the scope of the duty of care arising in the circumstance that even if the levee had been built with a one metre freeboard, except for Tottenham Road, the town would nonetheless have flooded unless additional temporary levees had been built. "If the duty was to protect against the 1976 flood of record, the question in issue reduced to whether provision of 0.4 metre freeboard where the levee comprised a made road as a matter of engineering design constituted a compliance with that duty." The appellants accept what his Honour said in this quoted sentence. At the trial the appellants argued that where the levee was built with only a 0.4 metre freeboard, it did not comprise a made road, apart from the Tottenham Road section. His Honour found that those sections were formed roads. The appellants do not challenge that finding. The appellants accepted that if the chosen design was one to protect the town against a flood up to the level of the 1976 flood the levee that was built complied with reasonable engineering practice. Fundamental to the appeal was the submission that the levee was not designed or intended to be designed only to withstand a flood of the 1976 level but to withstand even higher future floods. The appellants argued that levee banks are designed and constructed not to keep out a particular level of flood but to provide reasonable protection on a cost benefit basis against future floods. The design level was the top of the crest of the permanent levee because that was the level at which protection against the water ceased. Factors of safety and uncertainty as to the height of the future flood must be allowed for. The purpose of the design was to give a permanent levee at a particular height at a given point. With respect I agree with Cole J when he said that, if the duty was to protect against some higher flood, it was necessary to define that different flood of design. Cole J said that the evidence was overwhelming that, explicitly or implicitly, WRC approved Stage 2 plans with the lesser freeboard where the levee bank traversed formed roads, and saw the levee constructed with such lesser freeboard. Later he said that since WRC had acquiesced in or approved the levees being constructed with the reduced freeboard of 0.4 metre upon portions of Boundary Street, Bexon Street and Oatley Road, it followed that the levee banks as constructed were in accordance with an agreed variation in alignment and height to that initially specified in the schedule to the agreement between the Council and WRC. He noted that the experts agreed that, when designing levees, it was necessary to choose a flood level as the basis for design of a levee, and further that the 1976 flood level was the appropriate basis for design. These remarks were based on the following quotation from the Dole report: 22 UNREPORTED JUDGMENTS "Tt was agreed that: . IN 1978/79. THE RECORDED 1976 FLOOD LEVELS WERE AN APPROPRIATE BASIS FOR DESIGN OF PROTECTIVE LEVEES: . there were no formal freeboard standards adopted in Australia at that time but the freeboard specified was within an accepted band of practice; . WHERE LEVEES WERE INCORPORATED WITH A CONSTRUCTED ROAD FORMATION. PROVIDING STABLE FOUNDATIONS AND PERMANENT ACCESS (IN PARTICULAR. ALONG TOTTENHAM ROAD). A FREEBOARD OF 0.4 M ABOVE DESIGN FLOOD LEVEL WAS APPROPRIATE: . for other locations as proposed in 1978/79, freeboard of 1 metre above design flood level conformed with accepted practices at the time; . in the particular circumstances at Nyngan, the 1978/79 crest level specifications were reasonable and conformed with accepted practices;. the 1978/79 specifications were adequately defined in the various plans approved by the Bogan Council Engineer between October 1978 - March 1979." (emphasis added) According to the appellants this meant only that the 1976 flood level was a starting point for determining what levee height should be chosen to resist even higher future floods and that the agreement on a 0.4 metre freeboard was limited to a section up to a length of 800 metres, namely the Tottenham Road section. The language of the Dole report does not support the appellants' contention that the agreement was so limited. The report recorded that there was agreement that the Tottenham Road levee was constructed in accordance with reasonable engineering practices applicable at the time. The report continued: "Tt was also agreed that at other locations where the constructed levee was lowered over a short distance where the works intersected specific features eg road or rail crossings, the localised lower freeboard was reasonable having regard for accepted engineering practice. There is no agreement between the parties that the reduced freeboard at other locations is in accordance with reasonable engineering practice. This relates specifically to the southern and eastern levees between RD 3600 - 7500." The report then summarised various views. Those of the residents were stated as follows: "The levees were not constructed to be access roads as well as levees. Whilst the alignment followed road reserves, there was no original road formation or construction. Some of the road reserve incorporated graded tracks without formation or longitudinal or cross drainage arrangements. The constructed levee had a general crest width of 3 metres and was clearly and primarily constructed as a levee and not as a road. IN THESE CIRCUMSTANCES the Residents hold the view that it was not in accordance with reasonable engineering practice to reduce freeboard below 1 metre. In particular they are of the view that over the total length of 1,600 metre, where freeboard is 0.4 - 0.5 metre, a significant departure from acceptable engineering practice has occurred having regard for the nature and purpose of the works as flood protection levees." (emphasis added) This was the appellants' contention at the trial. Apart from the Tottenham Road Section, the levee with the lower freeboard did not comprise made roads. They argued that the 0.4 metre freeboard had been used on sections "clearly and primarily constructed as a levee and not as a road." URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA)23 The appellants also challenged the conclusion that WRC had acquiesced in or approved the Stage 2 plan. They submitted that the plans went no further than Mr Mittelheuser who had no authority to approve them and was negligent in not referring them to his superiors so that they could have been given proper consideration in light of WRC policy. THE DESIGN CRITERION Understandably available finance was a major consideration in determining what should be done after the 1976 flood to provide flood protection. On 25 March 1976 Mr Whatham reported to the Finance and Works Committee on the cost of a levee with a "two feet freeboard above the level of the 1955 flood" and after government grant how much money the Council would have to find to meet that cost. On 11 June 1976 the Council wrote to WRC to find out the Commission's "requirements for subsidised levees" and advice "when Council might expect to receive a subsidy." The July 1976 reply stated the normal minimum standard of construction for levees "required by the Commission, when a grant is involved." That standard is stated to be "a minimum of one metre above the highest known flood." The need for Commission approval depended upon whether the Council wanted a grant towards the cost of the levee. The relevant maximum flood level was agreed to be the 1976 level. The Council's protest on 30 October 1978 was based as the shire engineer's report to the Finance and Works Committee showing the unnecessary cost of a one metre freeboard where a lower freeboard was sufficient. The Commission's response is that, acknowledging the area in question was well timbered, this might not always be the case and when considered with "the inevitability of higher floods, relaxation of levee standards in the east of town would represent a weakness in the levee system as a whole." Sheets 1 - 6 of the plans complied with this and Stage 1 was built. On 1 October 1980 the shire clerk wrote to Mr Ralph Hunt that when the levees were completed the town would be protected from any flood which reached the highest known level in the Bogan River in the past. When the contractor, Mr Laing, tendered for Stage 2, Mr Whatham told him: "It is too much money the Council cannot afford that price. We want you to do the job but we would like you to have a look at a new schedule of rates." Mr Whatham showed Mr Laing another drawing of the proposed levee in places at the lower level. Mr Laing quoted a new and reduced price. At the very least the reduced price for building parts of the levee to the lower level played a substantial part in the Council's decision to go ahead with Stage 2. Cole J addressed the question whether it accorded with proper engineering practice to have a freeboard of 0.4 metre along the levee where it formed portion of Boundary Street, Bexon Street and Oatley Street. He took account of the following factors: (a) All parties and all experts accepted that it was proper practice to have a freeboard of 0.4 metre along Tottenham Road, although that was closer to the Bogan River than the other 0.4 metre freeboard levee sections, was subject to greater river pressure flows, at least equal attack from wind driven or surge waves, would be subjected equally to any unforeseen hydrological circumstances due to the different nature of flooding and obviously would be equally threatened by any flood greater than the 1976 design flood. (b) The water height and thus water pressure and likely wave action adjacent to the southern and eastern levees during the 1976 flood was much less. 24 UNREPORTED JUDGMENTS The experts expressed various views about the purpose of building a freeboard. In the American Water Resources Bulletin of August 1982 Mr Charles E Yoe, an economist with the US Army Corps of Engineers, wrote an article "Design Freeboard and Economic Efficiency". For him design freeboard was "an increment of height added to a dam, flood wall, levee, or channel, above and beyond the design flood height, intended to serve as a safety factor to account for unforeseen errors in design." Later he quoted from a publication by the Hydrologic Engineering Center of the Corps of Engineers about freeboard: "After the design water surface profile is determined for a levee or channel improvement the profile of the top of the levee or of channel wall is determined by adding some amount of freeboard. This freeboard constitutes a safety factor to allow for wave action, errors in measuring cross sections, errors in estimating roughness coefficients, and errors in the overall water surface profile determinations..... Selection of the amount of freeboard is usually dependant upon experience during actual flood-fighting operations...... There is a tendency to depend on freeboard for carrying more water than the project is designed for. This is an unwise use of freeboard. Freeboard is intended to compensate for computation uncertainties, inadequacies of data, and other sources of error." Mr Yoe referred to an engineering regulation of the Corps of Engineers which said: "Statements or implications that a flood damage prevention project, because of freeboard, will protect against any flood larger than design flood shall not be included in feasibility, post authorisation, or other related reports or otherwise communicated to the public." WRC called Dr Christopher Joy, an eminent engineer, the principal of Water Studies Pty Ltd and responsible for the reports that body prepared. Dr Joy gave evidence that the Hydrologic Engineering Center of the Corps of Engineers was a world authority pre-eminent in hydraulic studies. He agreed with what that the Center said about freeboard in the passage quoted in Mr Yoe's article. The following interchange took place in his evidence: "HIS HONOUR: Q.It does not, I do not think, refer to the prospect that the flood may be experienced in excess of, although not a great deal in excess of, the flood of record. That seems, I gather from Dr Dole's report, to be something that has been agreed to be taken into account in selecting a measure of freeboard? A. What the US Corps of Engineers is saying here is when they undertake the economic analysis when they work out the benefits attributed to a levee, they only consider the levee provides protection up to the design flood level. They could not attribute any additional protection to the freeboard in that although you will obviously gain some, they say no it is incorrect to do that. The levee has been designed to provide protection for a flood and any added protection we get out of it is a bonus but that shouldn't be taken into the ensuing cost calculations attributed to that calculation. HIS HONOUR: Q. But the reality is it does give greater protection? A. It does give a certain degree of greater protection. A lot of the problem centres around just exactly what that degree of protection is." Cole J said there was a body of evidence before him to the effect that provision of a 0.4 metre freeboard along the gravelled road levee section was an acceptable and appropriate engineering decision. In addition to Dr Joy's evidence he referred to the evidence of other experts. Mr Paterson, in considering the crest level of the levee, stated: "Review of Figure 1 indicates that provision of 0.5 metres URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA)25 freeboard above the 1955 flood would provide protection up to the once in 1000 year event. Whilst accepting the inaccuracies of the above method, the data available in 1979 suggest that 0.4 to 0.5 metres freeboard provided reasonable protection against uncertainty of the designed event." According to Mr Paterson freeboard was intended to cover: ". Hydrological uncertainties regarding the design flood; . local minor variations in actual flood surface levels; . an element for delayed maintenance (vis delay in repair of settlement or stock crossings); and/or . wave action fretting the crest." Cole J said: "Whilst not espousing the 0.4 metres freeboard, Mr Patterson noted the acceptance in principle by WRC of the lower freeboard where roads were used to form the levee as indicated by its acceptance of the Tottenham Road position. He expressed the view that reduction in freeboard along roads generally could be justified upon the basis that: ". roads are relatively well maintained and thus would not suffer from reduction in levels due to stock crossing and traffic; . the crest of the levee (the road formation) is about double that of a rural levee and thus provides protection against fretting and attrition from wave action; . access along roadways is generally better than levee crests so that sandbag or "flood fighting' measures are easier than along long narrow rural levee crests." Mr Howell, of Unisearch, after reviewing literature showing different freeboard criteria selected by different authorities in Australia and the United States concluded: "The design and construction of the permanent levee complied with standards which were reasonable and appropriate at the time." As an introduction to his statement Mr Eagles, who, before his retirement, was employed as Assistant Principal Engineer Dams with the Public Works Department and was responsible for the design of the upgrading of the Nyngan levee after the 1990 flood, said that the formulation of the design and proportions of a protective levee system around rural towns, requires initially an assessment of the design flood level; "that is the flood level which the levee embankments are required to withstand without failure." He said that freeboard was a factor of safety against failure at design flood conditions. The allowance catered for shrinkage, cracking or low density zones in the crest area, settlement of fill materials and wave action at design flood level. He regarded a flood situation which exceeded the adopted design flood level and encroached into the freeboard zone as representing an imminent failure situation. Mr Eagles considered that the design involving a gravel road embankment with a crest/road surface eight metres wide constructed to a height of 0.4 metre above a design flood level adequate in 1979 for a rural town levee embankment for the design flood level adopted. Matters he took into account were that in a road embankment the materials would be well compacted, usage and road maintenance would reduce concerns about low density areas and shrinkage or cracking below the design flood level and the crest width of eight metres. He said further that the eight metre wide road embankment configuration was about twice the width of fill embankment and higher densification of fill materials was as satisfactory as the narrower crested (three metre) levee with the greater freeboard of one metre. The appellants relied upon evidence that protection against a flood exceeding the level of the design flood was included amongst the purposes of the freeboard. Cole J said that at the hearing it was left to Mr Mathlin from Sinclair Knight and 26 UNREPORTED JUDGMENTS Partners Pty Ltd, consulting engineers, and Professor David Pilgrim, professor of civil engineering at the University of New South Wales, "to carry the argument that, it being acceptable to have a freeboard of 0.4 metres in Tottenham Road, it was nonetheless unacceptable to have the same freeboard over an additional 1,570 metres of Boundary Street, Bexon Street and Oatley Street." His Honour said that Mr Mathlin had not dealt with that matter in his tendered reports. The appellants submitted that his Honour had overlooked a report by Mr Mathlin dated 10 August 1993. That report did not address the argument his Honour mentioned. However it did deal with what were described as reasonable engineering standards for levee construction. It stated that the appropriate standards included "freeboard of one metre over the | percent flood or flood of record." Of freeboard Mr Mathlin said that the generally accepted freeboard provision of one metre is to take account of the following: "1. Water levels exceeding the design flood, through;. the design flood frequency being exceeded. the actual flood slope differing from the design flood, eg from inaccuracies in survey or prediction . Wave action 2. Reduction in the actual levee height over time possibly combined with lack of maintenance by: . settlement. wear and attrition of the crest through traffic, stock movement etc . erosion due to wind and or local rain It is not meaningful to separately quantify each of these components which make up the total requirement for freeboard. Indeed their relative importance depends on the immediate circumstances and environment. For example freeboard is often reduced where levees cross made roads because access is generally superior (for emergency works) and the risks of settlement, wear and erosion are considerably less than for a levee which is located in a remote rural environment. Likewise, wave action is more likely for a fast flowing coastal river than an inland river." None of this deals with the acceptability of a 0.4 metre freeboard in Tottenham Road and the suggested unacceptability of such a freeboard on the other made roads. In the course of cross examination Mr Mathlin gave evidence about the design of levees to accommodate sandbags on top and said: "There is no difficulty constructing that additional height where the road is parallel with the levee over a short distance, but when the length is excessive clearly it becomes very difficult and it is an unreasonable reduction of freeboard in those circumstances." He mentioned the resources required to construct the additional freeboard as well as factors of access. He gave the following evidence in answer to questions by Mr McClennan QC, who appeared for WRC at the trial and on the appeal: "McCLELLAN: Q. So would this be generally the position, so if you are talking about lengths of 100 or 200 metres to contemplate providing temporary levees.5 of a metre or higher, it would be your expectation that there is a real probability that they would fail? A. It is very difficult to give an answer in the abstract to that question, I am afraid. I believe that 800 metres on the Tottenham Road section was at the upper limit. The maximum length of acceptable practice in this situation, and that's why we formed the view in the early 80's that Tottenham Road ought to be raised in due course when it could be funded, that is getting at the upper limit of acceptable practice." This became the appellants' case. Professor Pilgrim said that where long continuous records of stream flow and floods were not available as at Nyngan: URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA)27 "The adopted procedure of designing for the highest flood level of record plus one metre is considered to be sound, and the best approach available. In flat areas such as Nyngan, the one metre provides a reasonable allowance. Although it is referred to as 'freeboard', it has three functions. It provides for the fact that the design flood should be larger than the highest flood of record, that some allowance should be made for wave action, and that even with compaction during construction, some settlement of the levee is likely to occur. Any allowance significantly less than one metre would not give sufficient safety..." Incorporated in this statement is a suggestion that the design flood should have been larger than the flood of record. No other expert supports this suggestion. Professor Pilgrim said of the 800 metres reduced freeboard along Tottenham Road: "While it would have been better to have maintained the top of all the levee at one metre above the 1976 flood, the reduction at this site seems reasonable. With the already constructed sealed road, this would have saved considerable costs and been convenient to the residents. The section would have also been very stable physically. More importantly, this relatively short section would have been able to be defended by sandbags in the major flood." Professor Pilgrim turned to consider the other three reduced sections covering 1,570 metres. He described these as unauthorised and said: "Where considered in conjunction with the authorised reduction section, it would have always been improbable that these long sections could be sandbagged sufficiently in the time available as a defence against a flood with levels near the top of a full levee, or above this. In the 1990 flood, this was proven to be the case. I therefore agree with the Sinclair Knight advice that the three additional reduced sections were unsound, and that without them, there would have been a high probability that with the available resources in the 1990 flood, the levee bank would have been sandbagged with sufficient strength to withstand the flood." Cole J said that the passage quoted might be an accurate statement of fact. However it assumed, as did the whole of Professor Pilgrim's evidence, that the obligation of the Council was to construct a levee sufficient to defeat a flood of a dimension of at least one metre above the flood of record. In that sense the statement was contrary to the agreed position before Mr Dole. Furthermore Professor Pilgrim advanced no empirical basis for his judgment that any distance greater than 800 metres was indefensible. His Honour said: "Tt seems to me to be wrong in concept to say that if a levee is designed to defeat a flood of, say, 4 metres, and a levee properly designed to effect that object and to take into account hydrological uncertainties, wear and erosion, wind and wave action as well as the prospect that there may be a greater flood than the flood of record, some obligation attaches to the constructing authority to provide a freeboard which would not only defeat the flood it was designed to defeat but must also provide a freeboard of greater dimension so as to defeat a flood which it was not designed to defeat. It may be a matter of engineering judgment as to what height a freeboard should be, but I am not able to accept that if a constructing authority decides to provide some protection up to a design level and constructs a levee with a freeboard additional to that design protection level which would have defeated the design flood, but which would not defeat an unforeseen flood which the levee was not designed to defeat, there can attach any tortious responsibility to the constructing authority." 28 UNREPORTED JUDGMENTS Professor Pilgrim accepted, indeed advanced, the view that the selection of the height of a particular freeboard over the design flood was a matter of engineering judgment to be approached with conservatism in the light of unknown future factors to which I have referred, but particularly having regard to the accessibility of any lower freeboard and its defensibility. Accepting that, I do not think it can be said to be any error of engineering judgment to select a freeboard above the design flood desired to be defended against, which freeboard in fact would have defeated that design flood. Before me there was no evidence that erosion, wave action, hydrological uncertainties or absence of constructional integrity of the freeboard was such that the 0.4 metres freeboard along the gravel road sections would not have defeated the equivalent of a flood or record, or slightly greater. What it could not defeat, even with temporary measures, was a flood exceeding the flood of record by at least 1 metre." Professor Pilgrim accepted that a gravel road, properly compacted, was a better starting point for a sandbag operation than a sealed road. He accepted, as did the other experts, that where levees were incorporated with a constructed road formation, providing stable foundations and permanent access, a freeboard of 0.4 metre above design flood level was appropriate. The appellants relied upon various papers of Mr Wood, a WRC officer and the Manager, Catchment Management, who prepared the Nyngan Flood Investigation Report. In an earlier paper Mr Wood pointed out that the consequences of a failure of a levee protecting an urban centre are potentially disastrous in terms of the financial costs to private individuals, the State and the Commonwealth and the potential for loss of human life so that the setting and adoption of appropriate flood standards and the provision of adequate freeboard are of utmost importance. In the course of notes for a meeting in July 1990 to discuss policy matters and funding for levees in the western part of New South Wales. Mr Wood said: "Tt has been and will continue to be Department of Water Resources practice to insist that one metre freeboard be provided on all levees protecting urban areas. Freeboard is that additional height which is provided over and above the design flood level to account for factors which cannot rationally be accounted for in the design process. It: . provides a factor of safety - to guard against floods higher than the design event." Mr Wood then went on to refer to other factors and concluded: "Bearing in mind these factors, as well as the need to reduce the risk of overtopping, the freeboard provision of one metre is fixed as a minimum requirement. In paral5 of the Nyngan Flood Investigation Report, "Lessons Learned from the Nyngan Flood" appears the following: "15.2.1 DESIGN At the time of design and construction of the Nyngan levee the Department's policy with respect to the height of levees was to provide 1m freeboard above the highest recorded flood level. This policy was not found wanting during the April 1990 flood. Those sections of the Nyngan levee that were constructed to provide 1m freeboard above the highest recorded level (1955 and 1976) did not fail. Failure occurred through the breaching of temporary sandbag levees constructed along low portions of the permanent levee where the freeboard was only 0.4m. Freeboard along these sections was sacrificed to enable the levee to be used as a URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA)29 permanent roadway. Whilst this appeared to be a justifiable engineering decision at the time, it was found to be wanting when the levee was called upon to meet its prime purpose." In appendix 4 under the heading "Freeboard" it is said that freeboard takes into account a variety of factors that are not known with precision including safety", "waves", "variable flood patterns", "embankment settlement" and "embankment attrition". Of "Safety" it is said: "It is inevitable that at some time there will be a flood greater than the highest yet recorded." Dealing with freeboard standards and as a result of a review by the government re-evaluating its approach to flood plain management after widespread flooding in the 1970's "and taking into consideration both the increased hydrologic data available in overseas practice, it was determined that the minimum levee height standard which should be adopted should be fixed such that the crest of the levee was equivalent to the one in 100 year flood level plus 1.0 metre freeboard." Cole J was not at all satisfied that the appellants had discharged the onus of showing that the decision to have a 0.4 metre freeboard along the gravel road was not in accordance with accepted engineering principle at that time, or indeed that the decision was wrong. "In my view the evidence established that the decision was a reasonable one which accorded with the generally accepted engineering principle. That view is reinforced by the circumstances that the levees so incorporated in that road would have successfully defended the township of Nyngan against the flood of record or a flood up to 0.4 metres in excess of any previously known flood. To select such a freeboard was not a breach of any duty which it might be thought Council owed the residents in consequence of its decision to construct levee banks to protect the town." There was evidence upon which his Honour could reach this conclusion and in my opinion it was correct. I am satisfied that in order to design a levee bank it is first necessary to determine the level of flood against which the levee bank is to provide protection. The Dole report makes plain that the level chosen was that of the 1976 flood. I am satisfied that it was good and accepted practice to build what was called a freeboard which increased the crest of the levee above this level. This provided a safety margin. If one starts with the proposition that the levee bank was designed to protect Nyngan against a flood at the 1976 level, the acceptance by the parties and their experts of the sufficiency of a 0.4 metre freeboard on the 800 metre Tottenham Road Section demonstrates that such a freeboard was sufficient elsewhere where the levee consisted of a made road. I am satisfied this accords with the experts' agreement recorded in the Dole report. The evidence that the length of the lower freeboard should be limited to no more than 800 metres is unconvincing and as Cole J pointed out proceeds on the basis that the levee was designed to protect against some higher flood level. The Dole report recorded agreement that in 1978-79 when the levee was designed the recorded flood levels were an appropriate basis for design of protective levees. The appellants vigorously attacked his Honour's conclusion that WRC acquiesced in or accepted the Stage 2 design with the reduced freeboard. We were told that at the trial the appellants' case was, understandably, that WRC had approved the plans thereby fixing it with responsibility for them. It is not entirely clear why on the appeal the appellant was anxious to say the contrary. However what is important is that the design was to deal with the 1976 level of flood, with an added safety margin which as a 30 UNREPORTED JUDGMENTS consequence meant that the levee was able to withstand a higher flood level, and with such a purpose the design met current engineering standards. The reduced freeboard was part of the design. On this aspect of the case two further matters should be mentioned. First the simulated flood patterns show that at a gauge height of 4.33 metres the town south of the railway line was protected by the permanent levee and the town north of the railway line protected by the permanent levee on the western side and needed no protection to the north or east. This was the level of the 1976 flood. At a gauge level of 4.85 metres and above the town to the north of the railway line was flooded. This supports the conclusion that the levee bank was designed to protect the town against a recurrence of a 1976 level of flood. Second if the design flood had been something greater than the 1976 flood level WRC design-principles would have required a one metre freeboard above that level in areas other than those where the levee consisted of a formed road. This supports the same conclusion. If the design was for a flood greater than the 1976 level, the crest of the Tottenham Road Section must have been higher even with a reduced 0.4 metre freeboard. CAUSATION The respondent persuaded Cole J that even if the Council had constructed levees with a freeboard of one metre above the previously known record flood, except at Tottenham Road, the town would nonetheless would have been flooded by the April 1990 flood and the appellants would have suffered the same damage they sustained. Central to this finding is an understanding of what happened on the afternoon of Monday 23 April 1990. At 12.15 pm the river was, according to the Bureau, expected to peak at 4.85 metres overnight. By that stage it appears from the forecast reported in the minutes of the meeting of the previous Sunday afternoon that at least a 5 metre level of bags had been achieved. During Monday work was done on various supplementary levees. The Premier visited Nyngan and inspected the levee between 12.40 and 3 pm. During that time at about 2 pm the river reached approximately 4.8 metres on the gauge. In the next twelve hours it rose more than 0.4 of a metre. This surge was totally unexpected and occurred largely after dark. It had been acknowledged that supplementing the levee with sandbags in the dark was dangerous. Assuming that the permanent levee had been built with a freeboard of one metre except on the Tottenham Road section, Cole J said: "A rise of 0.4 metres or 0.55 metres during the afternoon and evening of Monday 23 April 1990 could not have been successfully resisted because whilst it was accepted that there were adequate numbers of sandbags and resources to fill them, the emergency committee and the residents would have been taken by surprise by the surge of between 0.4 and 0.55 metres in such a short period of time. In those circumstances, which were unknown, there would not have been in place any strategy to provide a levee height equivalent to at least 5.2 metres or, under the assumed scenario, 5.35 metres. Shortly put, the emergency committee and the residents would have been defeated by the entirely unexpected final surge which was not predicted or planned against." His Honour's reference to 0.55 metre and 5.35 metres takes account of the 0.15 metre exclusion factor. The appellants argued that if the residents had not had to devote time and energy to raising the 0.4 metre freeboard in the areas other than Tottenham Road to a height of one metre they would have been able to devote time to raise the level of an assumed permanent levee with a freeboard of one metre to a greater height. I do not think this realistically answers the conclusion that what would URJOUGLAS v WATER ADMINISTRATION MINISTERIAL CORPORATION (Sheller JA)31 have defeated them was the unexpected, sudden and huge surge in the flood level. Until some time, probably after the visit by the Premier, the situation was seen to be manageable with a levee bank raised to 5 metres. Suddenly and as night approached this changed. As I have said it was agreed that at its peak in the early hours of the morning of 24 April the flood would have overtopped the levee with a one metre freeboard by up to 0.7 metre on the western levees, up to 0.5 metre on the southern levees and up to 0.3 metre on the eastern levees. I see no reason for disturbing his Honour's conclusion that even if the levee bank had been built as originally designed the flood would have overtopped it, inundated the town and caused the same damage that was ultimately suffered. GOOD FAITH WRC relied upon a statutory defence of good faith. S582A(1) of the Local Government Act 1919 provided that a Council should not incur any liability in respect of - "(b) anything done or omitted to be done in good faith by the Council insofar as it relates to the likelihood of land being flooded or the nature or extent of any such flooding." Subs(2) provided that, without limiting the generality of subs(1), that subsection applied to: "(e) the carrying out of flood mitigation works; (f) any other thing done or omitted to be done in the exercise of a Council's powers, authorities, duties or functions under this or any other Act." The section applied to and in respect of the Crown, a statutory body representing the Crown and a public or local authority constituted by or under any Act; subs(5). This defence does not require the plaintiff to prove dishonesty; Mid Density Developments Pty Ltd v Rockdale Municipal Council (1993) FCR 290 at 298 and following. However the conclusions I have reached on negligence make it unnecessary for WRC to rely on a defence of good faith. I am quite satisfied that WRC did act in good faith. CONCLUSION In my opinion the appeal should be dismissed with costs. Representation: COUNSEL: Appellant: BJ Gross QC/TJ Boyd Respondent: PD McClellan QC/BM Green SOLICITORS: Appellant: Mathews Williams Respondent: Legal Section - Department of Water Resources