Appeal Outcome: Special leave granted by the High Court 1 May 2009 (S496/2008) Sydney Water Corporation v TuranoSpecial leave dismissed with costs by the High Court 1 May 2009

Case at trial 17 The principal factual issue at trial was what caused the tree to fall. 18 Both Mrs Turano and the Council's case was that the tree's root system was affected by a pathogen due to the soil being waterlogged, at least intermittently, over an extended period of time. They contended that water had reached the roots of the tree via the sand in which the Sydney Water main had been laid. 19 Mrs Turano's case was that as the Council had been undertaking work, or other operations, in the area on at least three occasions prior to the accident, namely, in 1997, 1999 and June 2001, it would, or should, have observed: that the culvert drain was clogged with sediment and debris; that there was water ponding in the culvert pit and that the area around it was waterlogged; the tail-out drain had ceased to function; and that the tree was distressed. 20 She also claimed that the Council should have observed the state of the culvert in the course of the triennial road inspections it was required to undertake as part of its accounting compliance programme. On either basis, the Council should have been aware of the condition of the culvert and that it posed a danger to the health of the surrounding trees, and taken steps to eliminate the ponding. 21 The Council contended that, as the culvert pit was built into impermeable clay, it was not responsible for the water reaching the tree root. Rather, Sydney Water had installed its main in such a way as to breach the impermeability of the pit and the water which introduced the pathogen came via the sand in which the Sydney Water main was laid. The Council also said that as it had no actual knowledge of the weakened condition of the tree, or of the condition of the culvert, it was entitled to the qualified non-feasance protection afforded by s 45 of the Civil Liability Act. 22 Sydney Water resisted the claim against, it on the basis that its water main was laid in accordance with accepted engineering practice; that it was merely speculative that the water had reached the roots via the sand in which the water main was laid; and that as the roots of the tree would have extended beyond the culvert, the water could have reached the roots directly from the culvert. In this regard, it contended that although the clay had a high degree or impermeability, water would still move through it, albeit at a significantly slower rate than sand.

The horticultural evidence 23 The horticultural experts were Messrs Meys, Freeman, Polsen and Castor. All experts except Mr Meys were in agreement that the tree's root system had been weakened due to the introduction of a pathogen. 24 Mr Meys inspected the tree about three weeks after the accident, but could not find any fungal pathogens. He concluded that the tree was in poor health and it was that, together with recent flooding in the area, that caused the tree to fall. None of the parties sought to resurrect Mr Meys' opinion in this regard. 25 Although the experts were substantially in agreement that the tree roots were affected by a pathogen, they were not in agreement as to the cause of the introduction of the pathogen. 26 In the report of CPE Tree Services, dated 20 November 2001 (which, as I explain below, was written by Mr Freeman), the opinion was expressed that: "The tree failed due to a combination of root damage from a root pathogen (Phytophthera), which can be attributed to the permanent rise in the water table from the roadside drain's poor performance to drain away and being located around the tree's root system, and the rotting root system of the remaining root ball from the permanent water logging, (Anaerobic Soil), has resulted in the tree's remaining root system being unable to withstand the large amount of wind movement, which has resulted in this tree's failure." 27 The trial judge, at [27], accepted the evidence of Mr Freeman. This meant that he accepted that the root damage was due to a permanent rise in the water table, which he attributed to the culvert having no tail-out drain. However, the reference to "water table" in the CPE Tree Services report was not relied on at trial. Rather, the phrase "water level" was substituted. Accordingly, his Honour's acceptance, at [26]-[27], of Mr Freeman's evidence that there was a "permanent rise of the water table … located around the tree's root system" was misconceived. 28 Mr Freeman accepted in cross-examination that he had no expertise in soil analysis, which falls within the area of expertise of a geotechnical engineer. Accordingly, his evidence given in cross-examination, that the permanent rise in the water level was due to the poor performance of the roadside drains, ought not to have been accepted by the trial judge. Mr Freeman had no expertise either in soil analysis or drainage. 29 There was a question whether the effect on the tree's health would have been observable, other than by an inspection of the root system. Mr Freeman gave evidence that he had observed "white marks" on the tree, about a metre from the ground. He said that the "white marks" indicated the presence of a secondary infection, known as "phytophthera", within the trunk or lower root system, which would have warranted further inspection and possibly a decision to remove the tree. He also expressed the opinion that the crown of the tree did not show signs of distress. As already indicated, his Honour accepted Mr Freeman's evidence and none of the parties directed any criticism to those aspects of his evidence. 30 His Honour next referred to the report of Mr Polsen of CPE Tree Services, dated 20 November 2001. In that report, the opinion was expressed that a permanent rise in the water table had permitted the introduction of the pathogen and that the root system was dying due to a lack of soil air, which in turn was due to the permanent wetness of the soil. The trial judge accepted Mr Polsen's evidence, where it was not contradicted by "accepted" oral evidence: [39]. 31 Although the trial judge dealt separately with the evidence of Mr Freeman and Mr Polsen, and stated he accepted Mr Polsen's evidence in the terms to which I have referred, Mr Freeman was in fact the author of the report to which Mr Polsen put his name, as the principal of CPE Tree Services. This is apparent from a letter dated 27 September 2006, from Mr Freeman to Mrs Turano's solicitors, where he stated he was the "author" of the CPE report, although he inspected the site with Mr Polsen. Accordingly, Mr Polsen's opinion/evidence is properly to be understood as that of Mr Freeman's. It follows that the reference in the CPE report to the permanent rise in the water table should be disregarded. 32 Mr Castor, arborist, was of the opinion that the tree was subject to intermittent waterlogging, which predisposed the root system to infection. He considered that it was unlikely that the installation of the water main had any impact on the tree failure: [40]-[47]. His Honour made no finding as to whether he accepted these aspects of Mr Castor's evidence, although it is possible that he did, as he accepted Mr Freeman's evidence that the increase in the water level around the tree root was caused by the "poor performance of the roadside drain". 33 Mr Castor was of the opinion that it was unlikely that the problem with the tree would have been detected by a "drive by" type inspection. He also said it was not clear whether the scar on the tree would have been visible from a passing car. His Honour made no finding as to whether this evidence should be accepted, although there was no evidence to the contrary. 34 Mr Castor was also of the opinion that the tree's crown did not show signs of distress: [37]-[39]. He considered that the tree failure would have been very difficult to predict, unless considerable funding was available for a detailed assessment of all roadside trees. He was of the opinion that there was no reasonable evidence that would have required the Council to undertake such work, which would have been well outside the Council's budget and would have been difficult to justify: [43]-[46]. 35 As I have indicated, there was a challenge to the expertise of the horticulturalists to give evidence as to the quality and functioning of the surrounding soil. The purpose of this challenge was to demonstrate that the horticulturalists' evidence as to how the water got to the tree roots and in particular, whether it came from the culvert and why that was so, should not have been accepted by his Honour. 36 I have already indicated that his Honour's acceptance of Mr Freeman's evidence in relation to the water level was flawed, given Mr Freeman's concession that he did not have any expertise in soil analysis. Likewise, he had no engineering expertise and his evidence in relation to the drains needed to be assessed in that context. His Honour failed to make his assessment of Mr Freeman's evidence in that context. It is necessary, therefore, to turn to the geotechnical evidence in the matter, which was given by Mr Lumsdaine, and the engineering evidence.

Geotechnical evidence 37 Mr Lumsdaine is a geotechnical engineer of many years' standing. He inspected the site and undertook fieldwork involving subsurface investigation around the culvert on about 10 occasions in 2006. His Honour noted that Mr Lumsdaine had done investigative work involving test pits; that there was no water when he dug the pits; and that the culvert and discharge pit were dry. He inspected the pit on another occasion after heavy rain. His Honour also recorded Mr Lumsdaine's opinion that the sand under the Sydney Water pipe would act as a drain and that was the likely route of water to the tree root ball. 38 The trial judge, at [85], stated that "Mr Lumsdaine was of the opinion that the water going to the tree root ball almost certainly went through the sand, the backfill of the water main."

His Honour otherwise did not make any express finding as to whether he accepted or rejected Mr Lumsdaine's conclusion or any other part of his evidence. The only other reference to Mr Lumsdaine's evidence in the judgment was at [83], as part of his Honour's consideration of Mr Kiernan's report. Mr Kiernan was a consultant engineer who was not convinced that a tail-out drain had ever been constructed. 39 In fact, extensive investigative work undertaken by Mr Lumsdaine at the site revealed that the outlet pit of the culvert was excavated into very stiff residual clay, which he described as very impermeable material compared, relevantly, to clean sand. He described the difference in the permeability of clean sand to be "at least three orders of magnitude [that is, 1,000 times] greater than the clay". He reiterated that it was "quite possible" that sand's permeability could be five orders of magnitude greater than clay. Importantly, for the purposes of understanding the effect of water on "very impermeable" clay, he explained: "Water ponded intermittently against a face of very stiff residual clay might saturate and soften the outer 'skin' of the exposed clay, but it would not flow (seep) through the body of the clay." 40 Mr Lumsdaine said that there was evidence there had been a tail-out drain at the western end of the culvert. Mr Lumsdaine also observed and commented upon the construction of the water main in a report dated 3 April 2006, in the following terms: "… the water main was probably installed in the following general manner. A trench was dug through the surface filling and the underlying natural clay. The base of the trench would have comprised natural clay. The lower part of the trench was filled with clean sand backfill up to the design level of the pipe base. The pipe was then laid on the sand bedding, and then further sand filling placed around the sides and top of the pipe. The top of the trench was sealed by placing and compacting clay filling. At the location indicated, there was about 300 mm depth of clean sand filling below the base of the pipe. The lower 230 mm of this stratum of sand was observed to be fully saturated with free groundwater, presumably perched on the underlying clay base of the trench." 41 His investigations revealed that in periods after heavy rain, water would be retained at a level that would maintain the sand near the roots of the tree in a saturated state for 23 days. 42 Mr Lumsdaine concluded from his investigations that: "(a) The outlet pit was originally excavated into impermeable clay.

(b) There appears to have been an original tail-out drain, which is now backfilled.

(c) Water that ponded, for whatever reason, in the outlet pit would not have flowed laterally through the surrounding low permeability clay.

(d) The clean sand backfilling around the Water Board main would have acted (and continues to act) as an outlet drain to the outlet pit.

(e) I do not know if part of the explanation for the tree's falling over was saturation of the soil around the tree's root ball by water originating from the outlet pit. But if this is part of the explanation, then it is almost certainly that the route taken by the water from the outlet pit to the tree root ball was through the sand backfill surrounding the water main, and not through the clay strata." (Emphasis added)

Engineering evidence 43 The other expert evidence in the case was that of the expert engineers Mr Bewsher, Mr Burn, Mr Kiernan and Mr Clark. 44 Mr Bewsher is an engineer with 26 years experience in investigation and design of water resources and water-related projects. He was of the opinion that the culvert was constructed between the period 1961-1970. He considered it unlikely that this was a "drowned" or "blind" culvert, as he had never, in his experience, seen such a culvert in a position such as at the accident site. He said that the construction of the culvert with what he described as a "flared wing wall" type headwall, meant that the culvert was to be free-draining and that as originally constructed, there was a tail-out drain. He said that the installation of the utility services (that is, the water main and Telstra line) at a higher level than the invert of the culvert made it impossible for the culvert to be free-draining. 45 In his report dated 6 April 2006, Mr Bewsher said that the sand in which the water main was laid was "highly permeable allowing water to drain easily along the trench, potentially for a very long distance". He said that the sand backfill in the water main trench provided an efficient hydraulic conduit for the passage of water along the trench. By contrast, the clay material was not an efficient medium for the passage of water. 46 He concluded that, given the impermeable nature of the subsurface, it was unlikely that water ponding at the culvert outlet could easily travel to the tree through the clay. However, if water did reach the tree roots, it was more likely that this had occurred via the sand backfill laid in the water main trench. He also commented that water may have originated from sources other than the culvert. He acknowledged that the water in the sand backfill could travel in either a northerly or southerly direction, depending on the grade of the base of the water main trench, and he did not know in which direction the grade fell. 47 Mr Bewsher reviewed a survey that had been carried out by the Council in November 1999 for the purposes of the preparation of drawings for a road widening proposal in relation to Edmondson Avenue. He also reviewed aerial photographs of the area dated 1978 and 1982, which show the presence of a tree with a significant canopy in approximately the same position as that shown on the 1999 survey. Mr Bewsher stated that excavation of a trench for the water main and to a lesser extent, the Telstra cables, would have conflicted with the tree roots, if the tree was the same one present when the Council's survey was carried out in 1999. The 1999 survey and road widening drawings are discussed further below. 48 His Honour, at [52], accepted Mr Bewsher as "a reliable witness and accurate, not being prepared to speculate". This appears to be a finding that his Honour accepted Mr Bewsher's evidence. 49 Mr Burn, consultant engineer, had over 20 years' experience advising council authorities in relation to development projects and planning and had formerly been employed by the Parramatta City and Tumut Shire Council as an engineer. 50 Mr Burn stated, in his report of 1 August 2005, that culverts are installed where a road passes over a clearly recognisable low point, that is, low-lying land, so as to convey surface water beneath the road's surface. He said this served two purposes. First, it prevents water build-up flooding over the road pavement, thereby minimising the risk of motor vehicles aquaplaning during wet weather. Secondly, it ensures surface flows are drained away and do not lie adjacent to the road, resulting in a saturated/weakened road pavement, which would have the effect of shortening the life of the roadway. The trial judge, at [59], stated that he accepted Mr Burn's evidence that the design and construction technique for the culvert was typical of culverts in New South Wales. 51 Mr Burn was of the opinion that the culvert was designed to drain into the ground "downstream", that is, in a westerly direction. His Honour, at [54], accepted this evidence. Mr Burn considered that as originally constructed, the culvert had a tail-out drain. During the course of cross examination by senior counsel for the Council, Mr Burn was shown photographs taken by Mr Bewsher in 2006 which it was suggested showed the tail-out drain. He responded that there appeared to be a "stream path, a dry stream bed or whatever, yes". In later cross-examination by counsel for Sydney Water, he said that from his inspection of the site after the accident, due to the degree of vegetation he observed on site, there was nothing to indicate its existence. 52 Mr Burn agreed that the walls and base of the outlet pit on the western side were of very impermeable clay, which would operate a "little like concrete lining of the culvert outlet pit". He agreed that water will only pass very slowly through very impermeable clay. In this regard, he accepted the expertise of Mr Lumsdaine. 53 His Honour, at [54], accepted Mr Burn's evidence that the utility mains had been laid at a higher level than the invert of the culvert. At [58], his Honour stated that the effect of Mr Burn's evidence was that the ponding of the water was relevant to the ultimate undermining of the tree's root system. This would appear to be a reference to the opinion expressed in his report dated 21 February 2006, where he said: "The outlet from the under road drainage culvert was blocked downstream preventing water from leaving the vicinity of the pipe outlet. The prolonged presence of water saturated the ground weakening soil strength.

The flat grades in the area meant the outlet from the pipe culvert would need to be kept obstruction free to minimise water build-up leading to soil saturation. The growth of vegetation in the pipe outlet area served to block water outflow.

The presence of the utility mains laid at higher levels than the pipe invert further exacerbated the retention of water by raising the 'dam wall' across the pipe outlet discharge path." 54 However, Mr Burn's opinion that growth of vegetation in the pipe outlet blocked the water outflow did not, in my opinion, survive his oral evidence. In cross-examination, he stated that he considered that the outlet pit was not part of the original construction of the drain, but that it had been dug out at a later time or times. He said: "… the fact that … there was no tail out area means that any time a work crew goes to clear the outlet culvert, they're going to be digging down and leaving a - effectively a hole similar in shape to a pit". 55 He continued that any silt flying into the culvert would eventually cause the culvert to fill, or clog, so that during times of flood water would cross the road rather than flow through the culvert, and the Council would then go out and clean the culvert. He conceded, however, that he had no evidence that this culvert had ever clogged. 56 In his examination in chief, Mr Burn stated that both the Telstra main and the water main crossed the outflow area (that is, the outflow from the culvert), and prevented free discharge from the culvert. He said the main obstruction would be by the water main. He agreed with the same proposition put to him in cross-examination, namely, that the installation of the water main would block the tail-out drain and impede the dissipation of the water going through the culvert. 57 His Honour recorded Mr Burn's evidence, given in cross-examination, that water had ponded in the culvert because it could not drain freely. He considered that the obstruction was caused by the water main and the Telstra pit. He said that the water main would aggravate the drainage conditions through the culvert and this would have the effect of keeping the sand surrounding the water main wet: see [59], [60], [62] and [63]. At [69], his Honour noted that Mr Burn said that the same method of installation would be used, no matter at which level the water main was laid. 58 At [69], his Honour accepted Mr Burn's evidence that it would be difficult to know precisely the water levels in the clay beside the road without a detailed scientific analysis. This is another reason why Mr Freeman's evidence in relation to the water levels is problematic, as Mr Freeman did not carry out any tests, or rely on any tests that had been carried out, to support his opinion. 59 At this point of his judgment, his Honour made no reference to Mr Burn's evidence that the water main, being laid at a level higher than the invert to the drain, obstructed the discharge of water downstream from the culvert. However, at [144], his Honour appears to accept that the water discharged from the culvert was partly dammed by the water main and water ran along the sand base allowing water to more readily reach the tree. I will return to this later. 60 Mr Burn expressed the opinion that Sydney Water should have foreseen that the position of the water main would have obstructed the drainage from the culvert and that this could have been avoided by creating a further drain, to direct the culvert discharge westwards. He also considered that the installation of the Telstra pit had contributed to the ponding. He said because of that obstruction, any clearing of the culvert would require digging at the point of the outlet, leaving a pit: [60]. Mr Burn considered that the water main and the Telstra pit might have needed to be adjusted to allow the culvert water to run away and not build up, or pond. 61 Mr Burn concluded that the water main and Telstra line should have been laid below the level of the culvert outlet pipe. He considered that had this been done, the only obstruction to the drainage of water away from the culvert was the "less than ideal downstream conditions, not the 'dam' caused by the utility mains". In an earlier report, dated 29 June 2005, Mr Burn had concluded that Sydney Water had not appeared to make provision to ensure the water main, when laid, did not impact on culvert water discharge flows. 62 Mr Burn said the water main had been laid in accordance with standard practice, that is, by placing the water main on a bed of sand, which itself was laid on the base of the excavated trench into which the water main was to be laid. However, he said that in this case, the effect of the installation of the water main in the bed of sand was to create, in effect, a north/south drain. Mr Burn was of the opinion that anybody who might have considered the effect of installing the drain in this manner at the culvert would have realised that the creation of a north/south drain was a probable consequence of laying the water main in the sand at that place. 63 One solution that would have ameliorated this consequence was for Sydney Water to create a drain so as to provide for further westward movement of the water through the culvert. Mr Burn said this could have been done "quite easily", although it would have required Sydney Water to work on private land. Another 'solution' was that the water main could have been installed in the same manner as in the culvert further down Edmondson Avenue, where the water main was installed under the bed of the culvert. Mr Burn said that would have been a better method of installation. 64 In the report of 29 June 2005, Mr Burn also referred to the survey and other investigations carried out by Council in 1999 as part of preliminary work in respect of a road widening proposal. In June 1999, bore hole investigations had indicated that the ground was moist at all depths below the road pavement and the moisture was in the tree roots. However, the bore hole investigations that showed ground moisture were not at the site of the culvert, but were conducted between approximately 100 m and 500 m away. Drawings prepared from survey data dated November 1999, identified the culvert, the water main and the Telstra service. Mr Burn also said that design details prepared between the date of the survey and the completion of the final design for new work on the culvert, in May 2001, also indicated knowledge of the existence of the drain. 65 Mr Burn provided a further report, dated 28 June 2006, in which he addressed the processes that should have been undertaken by the Council as part of its assets identification programme. The assets identification programme was required to be undertaken for the purposes of the preparation of Council's budget. Mr Burn was of the opinion that an inspection of this culvert should have been undertaken by June 1997, so as to comply with Australian Accounting Standard 27 (AAS 27). 66 Compliance with AAS 27 required local councils to physically identify, inspect and rate all of their assets according to their condition. Mr Burn stated that this required councils, inter alia, to adopt formal asset management procedures to initiate maintenance/rehabilitation programmes. Drainage assets were included in the budget for the financial period ending June 1997. He said that as part of this process, every road and drain, including culverts was inspected. 67 Mr Burn said that in the first scenario, the clearing works "would have identified the public utility infrastructure so approaches to the public utilities to remedy the situation could be instigated". 68 He said that in that circumstance, the person inspecting the culvert in Edmondson Road should have recognised that the discharge area required clearing to facilitate stormwater discharge and arranged for the issue of a works instruction to clear the culvert discharge area. Alternatively, the presence of the water main and Telstra pit blocking the discharge to the culvert should have been observed and steps taken to initiate approaches to the public utilities responsible for that infrastructure, to remedy the situation. 69 Mr Burn also gave evidence that the Council had experts on storm water drains and the effect of drainage on trees and tree roots: [65]. He agreed in cross-examination that it would be his reasonable expectation that one of the things those experts might look at was the question of the culvert's free drainage. 70 His Honour stated, at [68], that the effect of Mr Burn's evidence was that the question whether the tree was suffering distress and whether there was ponding in the storm water outlet were matters that the Council's storm water engineers would take into account. Presumably, his Honour meant that the engineers would have regard to these matters during the course of any inspection carried out in about 1997, when the drainage assets were first included in the Council's budget, or during the course of other inspections, such as whilst carrying out survey work, and take action to deal with any problems that would thereby arise. 71 Except for the evidence to which I have specifically referred as being accepted by the trial judge, his Honour referred to Mr Burn's evidence without comment. However, at [66], his Honour accepted that the Council had the relevant expertise, when the culvert was created and up to the date of the accident. Presumably, the "relevant expertise" was expertise relating to storm water drains and their effect on trees and roots as stated by Mr Burn. 72 Mr Kiernan, consultant engineer, inspected the site on 19 February 2002 and prepared a report dated 5 June 2002. Mr Kiernan reported that at the time of his inspection the culvert was blocked with sediment and the surrounding area was saturated with stormwater from the culvert, which had not drained away, because the culvert was blocked and there was no tail-out drain. 73 Mr Kiernan concluded that the fact that the culvert was blocked with sediment and debris had caused the ground to become saturated. 74 Mr Kiernan concluded that the blockage to the outlet of the culvert was due to the lack of a defined drain and lack of any regular maintenance to remove sediment and debris, which in his opinion had not been serviced for some time. 75 He also considered that the water main and Telstra cables were close to the tree, and there was a possibility of root disturbance at some time by the authorities responsible for laying these utilities, if they were laid when the trees were mature. It is convenient to note at this point that there was no evidence to support this premise. Mr Kiernan concluded that the Council was responsible for not maintaining the culvert properly and allowing water to pond near the tree. 76 Mr Kiernan was not cross-examined and his Honour made no finding as to whether he accepted his evidence. None of the parties made reference to Mr Kiernan's evidence in their oral submissions to the Court on the appeal. 77 Mr Clark is also a consultant engineer. He inspected the site on 19 January 2005 and provided a number of reports to Sydney Water which focussed on the construction of the outlet drain from the culvert. In his opinion, the installation of the water main and the Telstra pit at a higher level than the culvert did not block the discharge from the culvert pipe. Rather, the blockage was due to the fact that there was no outlet channel from the culvert and that the blockage to the water discharged from the culvert was caused by the natural ground in which the services were laid. 78 Mr Clark expressed the opinion that the major cause of the tree falling was a combination of strong winds and the tree's poor root system. He also considered that the soil around the tree had been softened by water and that the water had come from rainfall, rather than from any flooding from the culvert. 79 Mr Clark examined a number of the other expert reports. He was unconvinced by Mr Lumsdaine's view that there had been a tail-out drain at the time the culvert was originally constructed. Mr Clark also expressed the view that the Council would have known the state of the culvert outlet from their inspections of the road from time to time and considered that they had taken no action to establish any free draining outlet. 80 At [84], the trial judge summarised Mr Clark's opinion as being that the area near the tree was wet. His Honour did not make any express finding at that point of his judgment as to whether he accepted Mr Clark's report or not.

Summary of trial judge's findings on the expert evidence 81 It is convenient, at this point, to summarise the express findings his Honour made in respect of the expert evidence. 82 As I have indicated, at [27], his Honour accepted the evidence of the arborist, Mr Freeman. Mr Freeman expressed the opinion that the tree roots were affected by a pathogen. Mr Freeman was also of the opinion that the white mark on the tree indicated the presence of the pathogen which would have warranted further inspection. He was also of the opinion that the crown of the tree did not show signs of distress. None of these findings are challenged by any of the parties on the appeal. 83 His Honour also found that Mr Freeman considered that there had been a permanent rise in the water table due to the poor drainage from the culvert. For the reasons explained above, this finding cannot stand. Nor, in my opinion, can Mr Freeman's opinion as to the rise in the water level, for the reasons I have given. 84 His Honour made no finding as to whether the scar on the tree was visible from the roadway. However, it is apparent from his finding, at [131], that he accepted Mr Castor's evidence as to the type of inspection of the tree that would have been warranted had the Council responded to information that the area surrounding the culvert was wet. 85 His Honour recorded Mr Lumsdaine's opinion, at [85], that the sand underneath the Sydney Water pipe would act as a drain, both north and south, and that the water going to the tree root ball almost certainly went through the sand, although he did not make any express finding as to whether he accepted or rejected Mr Lumsdaine's evidence. However, at [144], his Honour stated that there was evidence that the water from the culvert was probably dammed by the water main and that water ran along the sand base of the water main, allowing water to more readily reach the root system of the tree that fell, than it would have had the water main not been installed. On the assumption that this is a finding of fact, his Honour must be taken as accepting Mr Lumsdaine's evidence on this point. In any event, there was no evidence to challenge Mr Lumsdaine's evidence. Mr Bewsher, whose evidence was accepted, was of the same opinion. 86 If it is not correct to read his Honour's statement at [144] as a finding of fact on this point, such a finding ought to have been made, given Mr Lumsdaine's expertise, the absence of challenge to the evidence and its support in Mr Bewsher's evidence. Mr Lumsdaine's evidence in this regard was also supported by most of the engineering evidence. Likewise, his Honour's statement at [144] in relation to the water being dammed due to the installation of the water main, if a finding of fact, involves an acceptance of Mr Lumsdaine's evidence that the clay acted as an efficient barrier, or wall, to water discharged through the culvert. Otherwise, the water would not have dammed, but rather would have soaked into the surrounding soil. 87 As to the engineering evidence, I understand his Honour's finding, at [52], that he accepted Mr Bewsher as a reliable and accurate witness, to be an acceptance of the totality of his evidence. Mr Bewsher said that: the culvert was originally constructed with a tail-out drain; the water main and the Telstra pit were installed at a higher level than the invert of the culvert (this seemed to be accepted by all engineers), making it impossible for the culvert to be free draining; that the water main was laid in sand which was highly permeable, allowing water to drain along the trench in which it was laid; and that if water did reach the roots of the tree, it was likely that it did so via this mechanism. 88 It was open to his Honour to accept Mr Bewsher's evidence which was the subject of cross-examination and was supported by other expert evidence. 89 His Honour, at [135], specifically made a finding that, as originally constructed, the culvert drainage system included a tail-out drain. It is convenient to note, at this point, that it seemed to be common ground amongst the engineering experts that the culvert was blocked by sediment and that the tail-out drain had ceased to function. 90 His Honour accepted Mr Burn's evidence that the culvert had been laid in accordance with design and construction techniques applying at the time and that it was designed to drain downstream: see at [59] and [54] respectively. 91 Critically for his finding in respect of Sydney Water's liability, his Honour found, at [146], that the water main was laid in accordance with good practice. It would appear that this finding was based on Mr Burn's evidence that the water main would be laid by the same method, that is, on a bed of sand, regardless of the level at which the pipe was laid. It should be noted, however, that in making this finding, his Honour overlooked Mr Burn's evidence that the water main should have been laid at a lower level. 92 His Honour, at [66], also accepted Mr Burn's evidence that the Council had "relevant expertise" in relation to the culvert both at the time of construction and thereafter.

Evidence of Council's budgetary allocations 93 The next factual consideration in the matter relates to the state of the Council's knowledge in respect of both the drain and the tree, as well as Council's budgetary priorities and allocation of funds. These factors relate to the operation of ss 42 and 45 of the Civil Liability Act. 94 The Council's risk management co-ordinator, Mark Brookfield, gave evidence that there were 779 km of roads in the Council area and 1,400 ha of ground under its control. In 2001, at the time of the accident, the Council had 562 employees. There were about 64,000 trees planted along roads in the city and about another 2 million trees on Council property: [72]. 95 As at the date of the accident, the Council did not have a systematised practice in respect of the inspection of trees and culverts, but would react to a complaint or report in respect of the condition of a particular asset. If a complaint was made, the Council would carry out an inspection and take such action as was considered necessary: [73]. Mr Brookfield said that, so far as he was aware, no one within the Council was qualified to inspect a tree and determine if something was wrong with it, although Council employees would have the qualifications or experience to determine if something was wrong with the culvert itself: [74]. 96 Mr Brookfield said that there was no record in Council of any report being made in respect of the tree that fell upon Mr Turano's car. He said that there was no protocol for external authorities notifying the Council of an intention to carry out work or to install utilities and his searches of the Council's records did not reveal any evidence that it knew about the existence of the water main. 97 Mr Brookfield also gave evidence as to allocations in the Council's budget. The annual report for 2001/2002 was in evidence and revealed that demands on the infrastructure budget were substantial. Drainage assets were specifically dealt with in the report and were stated to have an "average overall condition assessed as fair". Expenditure on "preserving or rectifying" drainage assets in the 2001/2002 year was approximately $413,000 and it was predicted that approximately $35 million was required to restore all drainage assets to a satisfactory standard.

Evidence of Council activity in the area 98 It is possible that the Council inspected the culvert sometime prior to June 1997 as part of the identification of its drainage assets. However, there was no evidence whether it had in fact done so, and if so, what the nature and extent of any such inspection was. There was evidence, however, that the Council had engaged in activities in the area on a number of occasions from 1999 up until the accident. 99 As previously mentioned, the Council undertook borehole investigations in June 1999, which showed the ground to be moist at all depths below the road pavement. However, these investigations were some distance from the culvert. 100 In November 1999, the Council surveyed the area in the course of undertaking preparatory work for a road-widening proposal of Edmondson Avenue. Mr Bewsher stated that the 1999 survey identified a large tree approximately 4 m north of the culvert outlet, with a canopy width of 10 m, which appeared to be the problem tree. It also showed the alignment of the water main to be 2 m west of the tree's centre and the Telstra service to be approximately 1 m further west. 101 The 1999 survey formed the basis of various road widening design drawings prepared by Liverpool City Council Design Service in March 2000. Both Mr Burn and Mr Bewsher stated in their expert reports that the drawings identified the 450 mm diameter culvert, the Telstra service and the water main in Edmondson Avenue. 102 The drawings underwent a number of revisions, including the insertion of notes for the proposed modifications to the culvert area. Relevantly, these notes included at 18 June 2001, "match new table drain works into existing low point" and a handwritten note, "replants/replace ? trees". At 22 June 2001, a note stated: "numbers and locations of 45L grey box (Eucalyptus moluccana) trees to be planted shall [be] determined by superintendent on site" and handwritten note, "replace trees to be removed Location to be confirmed on site by the superintendent". 103 On the drawing dated 18 (or possibly 10) July 2001, there were handwritten notes "no tailout presently possible tailout??" and "clean and desilt". 104 The final drawing, dated 31 January 2002, which was specified to be the date of completion, release and acceptance by the client, bore the notes, "clean and clear existing tailout of vegetation and debris", "match new table drain works into existing low point", "clean and desilt existing pipe crossing". The latter note was accompanied by an arrowed point to the "existing pipe crossing" marked on the drawing, which I understand to be the culvert. 105 There was no evidence as to when, or by whom, the handwritten notes were added to any of the drawings, and it should be observed that the final drawing post dated the accident. 106 Mr Brookfield stated that whilst surveyors involved in the plans for the widening of Edmondson Avenue gave consideration to the removal or replacement of trees, this would only have been made in the context of removing trees that were in close proximity to the road and a danger to road users, who might run off the road. Such surveyors would not have the requisite knowledge of a qualified arborist to assess the health of the tree. 107 Road widening works were carried out in August 2002, in which the culvert at the site was extended as part of Variation No 8 for a road reconstruction contract between Liverpool City Council and Power Demolitions. The pipe was extended and the head wall moved, both approximately 1 m west. 108 The Council had a system for regular three-yearly inspection of roads, but that system did not include the inspection of culverts adjoining the roads.

Council's challenge on duty 109 At [100], his Honour stated that the Council conceded that it owed a duty to design and construct a culvert property, but disputed that it had a duty to inspect the trees or the culvert. At [113], his Honour again noted the Council had not disputed it owed a duty to take reasonable care. He said, however, that it was nonetheless necessary to consider the content or scope of that duty, before determining whether there had been breach. On appeal, the Council submitted that his Honour's understanding was wrong and that at all times, it had contested it owed a relevant duty of care. It said this was an essential issue in the proceedings and one to which specific submissions were directed. 110 His Honour expressly recognised the importance of determining the content or scope of that duty. In this regard his Honour held, at [130], that: "… the content of the duty of the Council was to take steps to keep the culvert clear of obstructions affecting the flow of water from east to west. This could have been done by regular inspections of the culvert." 111 Accordingly, it is convenient to consider the Council's challenge to his Honour's finding that it owed a duty of care as raising the question whether it owed a relevant duty. In this regard, the Council contended that his Honour made no clear finding as to the content of the duty and that in any event, the content of the duty of care that his Honour appeared to have found was not open on the evidence. The first of these complaints can be put to one side. His Honour made a specific finding, at [130] (reproduced above), as to the content of the duty owed by the Council. The real question in issue, therefore, is whether the Council owed a relevant duty of care in this case, either in the terms stated by his Honour or at all. 112 The Council submitted it is now settled law that it is not sufficient to merely articulate the duty of care owed by a local council in a broad or generalised way. Rather, what must be identified is a relevant duty, sufficiently specific to be applied to the circumstances of the case: see Vairy v Wyong Shire Council [2005] HCA 62; (2005) 223 CLR 422; Mulligan v Coffs Harbour City Council [2005] HCA 63; (2005) 223 CLR 486. 113 The need to identify the content of a duty of care has been given emphasis in the last decade. Gleeson CJ observed in Graham Barclay Oysters Pty Limited v Ryan [2002] HCA 54; (2002) 211 CLR 540 at 555, that if it is not possible to identify the content of an asserted duty of care with reasonable clarity, that may cast doubt upon the existence of such duty. In Romeo v Conservation Commission of the Northern Territory [1998] HCA 5; (1998) 192 CLR 431, Kirby J said at 478: "It is one thing to hold that a person owes a duty of care of some kind to another. But the critical question is commonly the measure or scope of that duty. The failure to distinguish these concepts can only lead to confusion." 114 In Modbury Triangle Shopping Centre Pty Ltd v Anzil [2000] HCA 61; 205 CLR 254, Hayne J stated at 289-290: "[103] Because the extent of a duty falls for decision in relation to 'concrete facts arising from real life activities': Perre v Apand Pty Ltd [1999] HCA; 198 CLR 180 at 211, it will not always be useful to begin by examining the extent of a defendant's duty of care separately from the facts which give rise to a claim. That may be possible, and useful, in a simple case (like motorist and injured road user) where the duty of care and its content are well established. In other cases, however, it may lead to an insufficiently precise formulation of the duty which obscures the issues that require consideration. That lack of precision may lie in formulating the duty too narrowly: for example, by asking did the defendant owe a duty of care to fence the part of the cliffs in its reserve from which the plaintiff fell?: Romeo v Conservation Commission (NT) [1998] HCA 5; (1998) 192 CLR 431. It may also, as in this case, lie in formulating the duty too broadly: for example, by asking did the defendant owe any duty of care to the plaintiff?

[105] In cases such as the present, where the extent of the relevant duty is not clear, it is useful to begin by considering the damage which the plaintiff suffered, and the particular want of care which is alleged against the defendant. Asking then whether that damage, caused by that want of care, resulted from the breach of a duty which the defendant owed the plaintiff, may reveal more readily the scope of the duty upon which the plaintiff's allegations of breach and damage must depend." 115 See also the discussion of Gummow J in Vairy at [58] ff; and Hayne J at [116] ff, although there is some difference in their Honours' approaches. Gummow J considered, at 442, that the question at issue in that case was whether the Council's duty of care included a duty to warn against the dangers of diving. In answering this question, Gummow J referred to the starting point suggested by Brennan J in his dissenting judgment in Nagle v Rottnest Island Authority [1993] HCA 76; (1993) 177 CLR 423 at 440, namely, looking to the nature of the danger, assessed prior to the accident, with reference to such matters as the functions of the public authority, the obviousness of the danger, and the care ordinarily exercised by members of the public. 116 Hayne J, at [117]-[118], considered that in that case it was not useful to state the duty of care owed by a statutory authority that had the care, control and management of public land to those who entered too precisely. However, he added, at [118], in what appears to be a statement of general application, depending upon the facts of a particular case: "… that it is necessary to recognise that the duty of care, owed by a statutory authority to those who enter land of which the authority has the care, control and management, is not a duty to ensure that no harm befalls the entrant. It is a duty to take reasonable care. Beyond that, however, it is not possible to amplify the content of the duty without reference to particular facts and circumstances. In each case, the content of the duty will turn critically upon the particular facts and circumstances ." (Emphasis added) 117 The question of the content of the duty of care arose again recently in Roads and Traffic Authority of NSW v Dederer [2007] HCA 42; (2007) 238 ALR 761. Gummow J, with whose reasons Callinan and Heydon JJ agreed, stated at [43]: "First, duties of care are not owed in the abstract. Rather, they are obligations of a particular scope, and that scope may be more or less expansive depending on the relationship in question. Secondly, whatever their scope, all duties of care are to be discharged by the exercise of reasonable care. They do not impose a more stringent or onerous burden." 118 It is apparent from the various statements emanating from the High Court that there is no single test or principle as to how the content of a duty of care is to be determined. However, what does emerge, is that the particular circumstances of the case are relevant to that question. 119 In this case, the Council contends that his Honour's formulation of the content of the duty of care was not open on the evidence. Reference was made to the judgment at [65], [113], [126]-[127] and [134]-[136] as being the paragraphs in which his Honour's finding on duty appears to have been based. The challenge to his Honour's judgment in this respect was that the findings and/or conclusions contained therein were either erroneous or not based in the evidence. 120 What follows is an analysis of these paragraphs to determine whether the Council has made out its contention. 121 His Honour accepted, at [66], that the Council had the "relevant expertise" when it created the culvert and up until the date of the accident. I understand this to be a reference to expertise in respect of the installation and maintenance of stormwater drains and the effect of those drains on trees and tree roots, being the evidence of Mr Burn to which his Honour referred at [65]. It also appears from the finding at [66] that his Honour accepted Mr Burn's evidence that one of the things the Council's stormwater drainage experts might look at if inspecting this culvert was "the question of the free drainage of [the] culvert" (also at [65]). 122 The evidence disclosed that there had been no regular or systematised maintenance of the culvert and its drainage system. Indeed, it was unlikely that there had been any maintenance at all. If there was a duty to keep the culvert clear of obstructions as found by his Honour, then, subject to the operation of ss 42 and 45, the finding at [66], combined with the lack of any regular inspection, may have been sufficient to find breach, especially if the evidence of Mr Burn was accepted. 123 However, the present question is whether the Council owed a duty of care, either in the terms found by his Honour, or at all. In this case, that requires a consideration of the purpose for which the culvert was installed and the physical conditions operating at the time of installation. It also requires a consideration of whether an ad hoc duty of care arose, as a result of the inspections that the Council carried out of the culvert on a number of occasions post-1997. I will deal with that question separately, as it raises the question relied upon by Mrs Turano in her Notice of Contention. Both aspects of duty also require regard to be had to the principles stated in s 42. 124 The culvert was intended to drain water under and away from the road surface. It was designed to be free draining across the pasture land to the west by means of the tail-out drain. The road was in a semi-rural area, with no houses or buildings in the immediate vicinity. The immediately surrounding soil conditions were clay. The tree population was sparse. There was no evidence as to when the tree was planted, or whether it was the product of a planting program, although it was likely that it was not, but was naturally occurring. 125 According to Mr Burn, in his report of 21 February 2006, because of the flat grades in the area, the outlet from the culvert needed to be kept clear of obstructions, so as to minimise water build-up leading to soil saturation. If that evidence was accepted, it might be argued it was foreseeable at the time that the culvert and its drainage system was constructed, the drain could become obstructed, so as to prevent the free drainage of water discharged from the culvert. 126 However, there are two more factual circumstances that need to be considered. First, the area into which the culvert discharged was impermeable clay. It would not have been foreseeable in that circumstance that water from the culvert would saturate the surrounding soil, even if the culvert was obstructed for some reason. Further, subject to Mr Burn's evidence as to the culvert silting up, the evidence was that the only relevant obstruction of the culvert was by the water main, possibly contributed to by the Telstra line. Both the water main and the Telstra line were installed long after the culvert was built and there was no evidence that the Council knew that either had been installed. As I have explained earlier, Mr Burn effectively conceded that any obstruction was due to the water main. 127 Accordingly, without needing to consider the other paragraphs to which the Council referred in its submissions, I am not satisfied that there was a duty on the Council in the terms defined by his Honour. Such a duty would only have arisen if the Council knew, or ought to have known, at the time of installation that the culvert and its discharge draining system was likely to become clogged or obstructed, so that water might not drain away but could seep into the surrounding soil, adversely affecting the surrounding vegetation, including trees in the vicinity. 128 None of those things was established: there was no evidence that the culvert was likely to clog, other than the evidence of Mr Burn prior to his concession that there was no evidence that the culvert had clogged; and there was no evidence that the tail-out drain was likely to cease to function. Rather, the evidence was that water would accumulate in the pit after times of heavy rain, but that the ponded water would not seep into the impermeable clay surface. 129 I have reached the conclusion that there was no duty in the terms defined by his Honour without considering the application of ss 42 and 45. As there are separate appeal grounds relating to those sections, their operation is considered below. It is convenient, at this point, to continue with a consideration of the paragraphs to which the Council referred in the submission with which I have been dealing. 130 His Honour stated, at [126], that he was satisfied that the Council's failure to properly maintain the culvert outlet so as to drain water flowing to the west: "… more likely than not resulted in the area around the western end becoming almost permanently damp and undermining the stability of the tree by causing root damage and soil degradation." 131 This finding appears to have drawn on the finding, at [123], that over a period of about forty years, the culvert would become intermittently blocked by debris, causing the water to pond and the surrounding area to become saturated, "leading to a more or less consistent wet area in the vicinity of the tree that failed". His Honour reiterated this at [125], when he referred to the ground being saturated through intermittent or consistent ponding. His Honour also found, at [124], that there was evidence that the rootball system of the tree was badly damaged by water, making it prone to failure. 132 The effect of his Honour's finding at [126], as I understand it, was that water would have soaked through to the impermeable clay wall for a sufficient distance so as to reach the rootball of the tree. This finding would appear to have been made on the basis of the evidence given by Mr Freeman about the increase in the level of the water table, or possibly the increase in the water levels. I have already explained the difficulties with this evidence. 133 The finding is also inconsistent with his Honour's acceptance of Mr Bewsher's evidence and inconsistent with the unchallenged evidence of Mr Lumsdaine, which his Honour also appears to have accepted. On Mr Lumsdaine's evidence, ponded water at the outlet of the culvert would not have flowed laterally through the surrounding clay: see [42] above. At the most, there would have been some minimal water absorption responsible for the shrink/swell function characteristic of clay soils. On Mr Bewsher's and Mr Lumsdaine's evidence, it was the sand from the water main that caused the water to travel. The water in the culvert discharge pit would not have permeated to the tree roots, because of the impermeable nature of the clay soil. Accordingly, as I will indicate later, I am of the opinion that even if duty and breach were correctly found by his Honour, his finding on causation was contrary to the evidence that he accepted. 134 His Honour made a finding, at [134], that as the Council had constructed the culvert and the system of water flow to the west, it had created the potential for the water to pond, if the tail-out drain was not kept clear. His Honour, at [135], accepted that there had originally been a tail-out drain for the western end of the culvert, but this had been allowed to be overgrown, or dug out over time, due to poor maintenance systems. The findings in these two paragraphs were open on the evidence. 135 His Honour concluded, at [136], that if the Council had properly maintained the culvert, the water would not have ponded and thereby affected the surrounding soil, causing damage to the roots of the tree. For the reasons already discussed, that conclusion was inconsistent with the evidence he accepted. His Honour also concluded that had regular culvert inspections been carried out, the condition of the tree rootball would, more likely than not, have been identified and steps taken to eliminate any potential danger by buttressing the tree or removing it. This finding was premised upon the Council having a duty of care in the terms found by his Honour. As his Honour's formulation of duty cannot, in my opinion, stand, this finding, which is a finding of breach, must also fail.

Was there an ad hoc duty of care? 136 Mrs Turano argued that contrary to the Council's submission that the effect of Mr Brookfield's evidence was that the Council had a reactive system which responded to complaints, in fact, it had no system. She submitted that to describe the system as "reactive", wrongly categorised what was, in effect, the complete absence of a system of routine maintenance of culverts and no form of inspection to enable blockages to be identified. 137 She also contended that although the Council's system of three-yearly inspection of roads did not include the inspection of culverts adjoining the roads, the need for a culvert to operate properly was an integral part of road maintenance. It followed that the Council had a duty to inspect the culverts. This argument, in effect, relies upon his Honour's finding, at [127], that the Council, during the course of its periodic road inspections, should have seen that the western end of the culvert needed periodic cleaning or clearing, although that finding is properly a finding on breach. I will return to this below. 138 On Mrs Turano's submission, it was also apparent, from the drawings prepared for the purposes of the road widening, that a professional person, either employed by or engaged by the Council, would have observed that there was no tail-out drain. She also submitted that it was highly likely that such persons would have observed the presence of the water main and that it was embedded in sand and had permeated the 'impermeable' clay walls of the culvert drain. 139 Three things followed on this argument. First, that a professional person, presumably an engineer, must have been on site at some stage for the purposes of the preparation of the drawings and would have observed the water main and that it had been laid in sand. Such a person would know that sand has a high degree of hydraulic conductivity. Secondly, the Council, through its officers, was at least put on inquiry whether there would have been any effect on the surrounding trees. Thirdly, by this chain of knowledge, or constructive knowledge, the Council ought to have called in an arborist. Adopting his Honour's finding on this, at [131], an arborist would have undertaken the kind of inspection which Mr Castor had envisaged. Although not referred to in his Honour's judgment, this inspection would have involved root crown excavation and trunk drilling to quantify the potential for tree failure: see Mr Castor's report of 11 August 2005. 140 It would be reasonable to draw an inference that Council engineers would have inspected the culvert for the purposes of preparing the drawings for the road widening proposal. The inclusion of the culvert in the drawings is sufficient for that purpose. However, there is no evidence that they observed that the water main was laid in sand. Nor is there any evidence as to when they inspected the culvert. Although the handwritten notes referring to cleaning the culvert and the tail-out drain appear on the drawings of 18 July 2001, it is not known when those notes were written. Nor is there any evidence that the manner of installation of the water main was observed at that time. For all that is known, it may have been overgrown by vegetation. 141 These considerations are relevant to the operation of s 45, considered below. Leaving aside the question of actual knowledge, the evidence did not establish that engineers had been at the site of the culvert as at November 2001. Nor was there any material from which it could be inferred that if engineers were on site during this period, such observations as they should have made would be sufficient to put them on enquiry that would have lead to ascertaining that the water main was laid in sand. 142 Further, there was no evidence, one way or the other, as to whether the ground was dry or boggy at the time the survey work was being undertaken, or whether there was any ponding that ought to have alerted them to some problem. It is also relevant that the evidence was that the crown of the tree appeared healthy, and the only evidence of any possible stress or injury was the marking on the trunk, so that there was nothing to draw the attention of Council officers to any risk such as materialised in this case. At most, the drawings established that at some time prior to 31 January 2002, a person within the Council had determined that the tail-out drain was to be cleared, that the water main had been installed and that there was to be cleaning and de-silting of the culvert. 143 Having regard to these facts and circumstances, I am of the opinion that there was no ad hoc duty of care for which Mrs Turano contends. I have reached that conclusion without having regard to s 42, although those principles would not have altered my conclusion.

Section 42 144 Section 42 provides: " 42 Principles concerning resources, responsibilities etc of public or other authorities

The following principles apply in determining whether a public or other authority has a duty of care or has breached a duty of care in proceedings for civil liability to which this Part applies:

(a) the functions required to be exercised by the authority are limited by the financial and other resources that are reasonably available to the authority for the purpose of exercising those functions,

(b) the general allocation of those resources by the authority is not open to challenge,

(c) the functions required to be exercised by the authority are to be determined by reference to the broad range of its activities (and not merely by reference to the matter to which the proceedings relate),

(d) the authority may rely on evidence of its compliance with the general procedures and applicable standards for the exercise of its functions as evidence of the proper exercise of its functions in the matter to which the proceedings relate." 145 His Honour recognised the relevance of the section to the question of duty. He stated, at [132], that having regard to the nature of the inspection that would have been required (that is, Mr Castor's detailed inspection), it was necessary to consider the application of s 42. His Honour concluded, at [133], that the evidence did not permit the court to accept that the section excluded liability. His Honour's reasons for reaching this conclusion appear to be contained in [134] of the judgment. 146 At [134], his Honour concluded from Mr Brookfield's evidence that it would have been unreasonable for the Council to have a general policy to inspect all roadside trees at random, without reason or notice of defect. As I have already indicated, this was part of Mrs Turano's case at trial, but was not pressed on appeal. His Honour considered, however, that was not the case with culverts. His Honour found that the Council, as a road authority, had constructed the culvert and the system of water flow to the western side of the road. Therefore, it had created the potential for ponding of water if any tail-out drain was not kept clear. 147 The Council submitted that his Honour's application of s 42 was erroneous. First, it was said that his Honour treated it as an exclusion clause, whereas it should be properly addressed as an anterior question to the determination of liability. That is not quite correct. Whilst s 42 is certainly not an exclusion clause, it is a clause that specifies the principles that must be applied in determining questions of both duty and breach. 148 The Council further contended that although his Honour made reference to s 42, he did not address the requirements of the section. In particular, it was submitted that his Honour failed to analyse the financial resources available to the Council, or to consider the way in which those resources were currently allocated. He also failed to analyse the many functions required to be carried out by the Council and failed to take into account Mr Brookfield's unchallenged evidence that the Council's system of reacting to complaints worked. 149 Although not formulated in this way, the Council's complaint is that his Honour had failed to give adequate reasons and had failed to deal with the evidence: see Waterways Authority of New South Wales v Coal & Allied (Operations) Pty Limited [2007] NSWCA 276. 150 The Council relied principally upon the evidence of Mr Brookfield to establish both that it did not have a duty of care and that it had not breached any duty of care that it might have owed. It submitted that it was apparent from the Council's annual report that not only were there many demands on its infrastructure budget, but that the general budget itself was in shortfall. There had been no challenge to the manner in which the Council had made its budgetary allocations. Indeed, in accordance with s 42(b), there could be no challenge. 151 The Council also relied upon Mr Brookfield's evidence that there had been no complaint or report relating to the performance of the subject culvert at any time since it was constructed. The Council submitted that Mr Brookfield's evidence was persuasive that in all of the circumstances referred to, an adequate system, that is, to respond to complaints, was in place. 152 Accepting that the Council's challenge on this ground relates to his Honour's reasoning process, I consider that it is well-based. The Council had adduced specific evidence as to its budget, the state of its assets, its budgetary allocations, the expenditure on drainage in the year that the accident occurred and the predicted cost of bringing its drainage assets up to a proper condition. In order to apply the principles stated in s 42 both in respect of the question of duty and breach, it was necessary for his Honour to engage in some analysis of the evidence that related to the principles referred to in paras (a) to (d) of the section. In other words, his Honour had to have regard to the many functions of Council, the actual allocation made in respect of drainage and the fact that there were many demands upon the Council's finances, to determine whether the Council had breached its duty of care (assuming any such duty was owed). 153 It is not readily apparent from his Honour's reasons how he had regard to these matters. Some articulation was necessary. 154 It may be, however, that his Honour reached his conclusion that s 42 did not exclude liability because he had in mind that inspection of culverts should have been part of the three yearly road inspections that were, or should have been, carried out. In this regard, Mr Brookfield said that the culverts were not inspected at that time because the culverts could not be accessed. It appears from his Honour's finding at [134] that he inferred there was no financial constraint on the Council which prevented it from inspecting the drains at the same time. 155 His Honour does not, however, state the primary facts from which this inference was drawn. He does not, for example, specifically refer to Mr Brookfield's evidence to which I have just referred, which might have been the basis upon which such an inference was based. Nor did he refer to Mr Brookfield's evidence as to the extent of the Council's local government area and the vast array of matters to which it had to attend. In this regard, Mr Brookfield had given evidence that he could "go out and find 300 problems" on any given day, which might also be relevant to whether the inference drawn by his Honour was open. However, his Honour made no specific finding in that regard, and the Court is left to speculate whether that was his Honour's reasoning process. 156 The Council also complains his Honour did not deal with the costs of the inspections that he found ought to have been undertaken, as he should have done in order to properly engage with the principles specified in s 42. This may be a variation on the consideration in the previous paragraph. However, when the argument is formulated in this way, a question of onus, at least evidentiary onus, arises. A trial judge can only make such findings of fact (including inferential findings of fact) as are available on the evidence. Mr Brookfield provided a report and gave oral evidence. In addition, the Council's annual report was in evidence. 157 If the cost of carrying out inspections was to be a relevant matter in the Council's defence, the Council should have led evidence that combining the inspection of the culvert with the road inspections could not be accommodated within Council's budget; or that it had allocated the funds required to undertake such work to other projects: see s 42(a), s 42(b). As I have said, Mr Brookfield's evidence on this point related to physical accessibility, not to financial considerations. 158 Nonetheless, the matter was argued in that way and I will first deal with it in the way it was argued. The question is, therefore, subject to those sections, would a duty of care arise? In my opinion, there was nothing in the facts and circumstances as explained above that would give rise to such an ad hoc duty. There was no evidence that the culvert was not draining properly. The evidence of ponding, on the assumption that there was ponding on the approximate half a dozen occasions when it is known that the Council was in the area, was not necessarily evidence that the culvert was not draining properly. 159 The Council made another challenge to the trial judge's treatment of s 42. It points out his Honour did not consider s 42 until after he had in fact made a finding in respect of duty. That finding was made at [130]. His Honour's consideration of s 42 commenced at [132] and his finding that it did not apply was at [133]. It appears that his reasons for so finding are at [134]-[136]. However, those reasons do not address the principles specified in s 42. 160 The analysis of a judgment for appellate purposes does not require a fine parsing exercise and does not require overzealous analysis. The question is whether the trial judge gave adequate reasons for decision. It is permissible to first state a conclusion and to then move to the reasons that justify the conclusion. However, it is possible that in this case, by first stating the conclusion without regard to the particular statutory principles that need to be considered in reaching that conclusion, his Honour may have lead himself into error.

Breach of duty 161 Should I be wrong in respect of duty, the next question which arises is whether there was a breach of duty. 162 As I have indicated, his Honour's finding, at [127], was probably a finding of breach. Then, at [131], his Honour stated that he was not satisfied that regular inspections of the culvert would have alerted the Council to the condition of the tree that fell, but had the Council been in possession of reports that the area surrounding the culvert was wet, it should have realised the need to obtain an arborist's inspection of the nearby trees as to their stability. The inspection required was that indicated by Mr Castor. 163 Mr Castor gave evidence that a detailed tree hazard assessment involving "root crown excavation and trunk drilling" would have had to be undertaken, so as to assess the potential for the tree to fail. He said that if such work was undertaken, the waterlogged condition of the soil would have been identified, on the assumption that it existed at the time of assessment, as would have the scar on the tree trunk. 164 The Council makes three complaints in respect of his Honour's finding as to breach: see written submissions, paras 28-29, which although stated to be errors relating to duty, are properly considered in relation to breach. 165 First, it submitted that his Honour failed to undertake the required analysis of the magnitude of risk as compared to the costs of protecting against such a risk, as required by Wyong Shire Council v Shirt [1980] HCA 12; (1980) 146 CLR 40. This complaint was made in the context that there was no evidence that an event such as this had ever happened before. 166 His Honour, at [137], specifically referred to the well-known passage in Wyong Shire Council v Shirt which specifies the considerations necessary in determining whether there has been a breach of duty. His Honour had found, at [136], that although the likelihood of injury was not great, the magnitude of the risk was. Whilst these statements are conclusory, it is self-evident that a tree with an unstable root system may fall and cause damage to life and property. Accordingly, the magnitude of the risk was high. The degree of probability of it happening was not minimal. It is not unknown for gum trees to fall during high wind, due to some underlying instability. 167 The real complaint, however, is that his Honour did not undertake the next step of the analysis, namely, a consideration of, inter alia, the expense of taking alleviating action, having regard to other conflicting responsibilities. The Council noted in this regard that there was no cross-examination of Mr Brookfield on this issue, he being the obvious person to ask. 168 This complaint, properly characterised, challenges the adequacy of his Honour's reasons. 169 The only 'finding' that his Honour made that might relate to the cost of responding to the magnitude of the risk was at [134], where his Honour said, based on Mr Brookfield's evidence, that whilst it would have been unreasonable to have a general policy to inspect roadside trees, that was not the case with culverts. No reason was given for that conclusion. This, in effect, is the same difficulty with his Honour's reasons as I have referred to above at [154]-[155]. It may be that his Honour drew an inference from Mr Brookfield's evidence that there were no financial constraints which prevented the Council from inspecting the culvert and its drainage system as part of its three-yearly roads inspection program. However, whether his Honour drew that inference is a matter of speculation. 170 If his Honour intended to draw such an inference, it was necessary to identify the facts upon which the inference was based and to state the inferential finding. His failure to attend to these matters was a failure in his Honour's reasoning process: see Waterways Authority of New South Wales v Coal & Allied (Operations) Pty Limited. 171 I should also state that this specific challenge was made in the terms I have identified at [165] above. The Council's submissions did not refer to s 5B of the Civil Liability Act, although his Honour had it in mind on this issue: see [138]. Section 5B relevantly picks up in statutory form the principles stated in Wyong Shire Council v Shirt: see Waverley Council v Ferreira [2005] NSWCA 418; (2005) Aust Torts Reports ¶81-818 at [45]. In terms, the section provides that a person is not negligent in failing to take precautions against a particular risk unless there was a foreseeable risk, which was not insignificant, and in the circumstances a reasonable person would have taken those precautions: s 5B(1). Section 5B(2) specifies the matters to be taken into account in determining whether a reasonable person would have taken precautions against a risk of harm. As I have already stated, those matters include the burden of taking precautions to avoid the risk of harm: s 5B(2)(c). His Honour did not expressly give any consideration to that matter and it is not apparent that he impliedly did so. 172 Secondly, the Council submitted that there was no evidence that the pit was defective. To the extent there was evidence that there was ponding during wet periods, it was submitted that was precisely how the drain was intended to operate. The relevance of this submission is that if there was no evidence that the pit was defective, there was no evidence upon which to base a finding that the Council was negligent, either in the construction or ongoing maintenance of the pit. It will be apparent from the discussion above about the pit at [158] above, that I consider that this submission should be accepted. 173 Thirdly, the Council contends that his Honour failed to give reasons for his conclusion that regular inspections of the culvert should have alerted the Council that there was ponding at the culvert discharge pit and that the culvert needed periodic cleaning or clearing. In my opinion, there is substance in this complaint, although not precisely in the way submitted by the Council. 174 The real difficulty with his Honour's conclusion is that it appears to have been based on the evidence of Mr Burn. I have already explained why that evidence could not have been accepted by his Honour. His Honour's conclusion was inconsistent with the evidence of Mr Bewsher, which he had accepted, and with the evidence of Mr Lumsdaine, which had been the subject of favourable concessions by Mr Burn. This is sufficient to find error in this part of his Honour's reasons. 175 Section 42 also requires consideration on the question of breach. The comments I have already made on the section also apply on this question and it is not necessary to deal with it anymore fully, except to say this. It may be that his Honour's failure to adequately address s 42 may have caused him to overlook the matters of which complaint is presently made.

Causation 176 His Honour dealt with causation at [140]-[142], although, as I have indicated, he made earlier findings as to what caused the tree to fall: see especially at [123]. After having referred to March v Stramare (E & MH) Pty Ltd [1991] HCA 12; (1991) 171 CLR 506 and Betts v Whittingslowe [1945] HCA 31; (1945) 71 CLR 637, his Honour concluded, at [142]: "The tree fell because of the root system coupled with a high wind. The condition of the root system was caused or contributed to by the failure of Council to maintain its culvert to allow water to drain properly and allowing the surrounding soil to be degraded. In my opinion the negligence of the Council created the conditions that caused the injury to Mrs Turano and her children and the death of her husband." 177 As I have already explained, that finding was not open on the evidence accepted by his Honour. Nor was it open on any other evidence, for the reasons that I have also explained, namely, that Mr Burn deferred to Mr Lumsdaine's expertise and also withdrew his evidence relating to a rise in the level of the water table, and no one sought to rely upon the evidence of either Mr Clark or Mr Kiernan. 178 The clear evidence was that water from the culvert travelled along the sand in which the water main was laid. The evidence was also that, to the extent water was blocked or dammed in the discharge area of the culvert, it was due to the water main being laid at a higher level than the discharge area of the culvert. Accordingly, even if Mrs Turano established that the Council owed a duty and that it was breached, she did not establish that the injury caused by the tree falling was caused by any breach of duty of the Council.

Failure to apply s 45 of the Civil Liability Act 179 The Council also complained that his Honour failed to apply s 45 of the Civil Liability Act. 180 Section 45 provides: " 45 Special non-feasance protection for roads authorities

(1) A roads authority is not liable in proceedings for civil liability to which this Part applies for harm arising from a failure of the authority to carry out road work, or to consider carrying out road work, unless at the time of the alleged failure the authority had actual knowledge of the particular risk the materialisation of which resulted in the harm.

(2) This section does not operate: (a) to create a duty of care in respect of a risk merely because a roads authority has actual knowledge of the risk, or (b) to affect any standard of care that would otherwise be applicable in respect of a risk.

(3) In this section: carry out road work means carry out any activity in connection with the construction, erection, installation, maintenance, inspection, repair, removal or replacement of a road work within the meaning of the Roads Act 1993 . roads authority has the same meaning as in the Roads Act 1993 ." 181 The operation of s 45 is fundamental to the determination of the Council's liability in this case. His Honour erred in failing to consider it. That raises the question whether this Court should proceed to determine the question. Although none of the parties submitted that this Court ought not to do so, it is still necessary to consider whether there were sufficiently clear and unchallenged findings of evidence, or alternatively, whether there was other evidence that this Court may act upon. Clearly, if the evidence that related to this issue is not contradictory, or does not depend upon questions of credit, then it is permissible for the Court to consider the matter for itself. 182 In order for the Council to fall outside the protection afforded by s 45, it had to have "actual knowledge of the particular risk the materialisation of which resulted in the harm" that arose from the failure to carry out roadworks, or to consider carrying out roadworks. It was accepted that drainage works were "road works" for the purposes of the section: see the Roads Act 1993. 183 The application of s 45 has been considered in a number of recent authorities in this Court: see Leichhardt Council v Serratore [2005] NSWCA 406; North Sydney Council v Roman [2007] NSWCA 27; (2007) 69 NSWLR 240; Angel v Hawkesbury City Council [2008] NSWCA 130; and Blacktown City Council v Hocking [2008] NSWCA 144. 184 The central issue in each of those cases was what constituted actual knowledge for the purposes of the section. In Serratore, Giles JA (Hodgson and Ipp JJA agreeing) held, at [15], that a finding of actual knowledge could be made by inference and that if the inference was fairly available and the roads authority called no evidence to rebut it, the Court could more comfortably find actual knowledge. 185 In Roman, the question arose as to what level within a Council actual knowledge had to reside. There was no evidence in that case to establish that an officer at the relevant decision-making level had actual knowledge of the defect in the roadway, although knowledge resided in the workers who performed street cleaning work and whose duties included looking out for and reporting upon hazards in the streets that might require repair. 186 The majority in Roman, Basten JA, with whom Bryson JA agreed, held, at [157], that actual knowledge for the purposes of s 45(1) had to be found in the mind of an officer within Council, who had delegated or statutory authority to carry out or consider carrying out the relevant roadwork. McColl JA rejected the majority's approach. Her Honour held, at [60], that knowledge of persons who, acting within the scope of their duties, learned of a particular risk and were under an obligation to report it, as part of the road authority's system of maintaining roads in its jurisdiction, should be attributed to the roads authority. 187 In Angel and Hocking the Court (Spigelman CJ, Beazley, Giles, Tobias and Campbell JJA) sat to determine the correctness of Roman. In Hocking, Tobias JA, at [223], upheld McColl JA's approach to the operation of s 45. 188 In this case, two questions arise for determination. The first is what was the particular risk that materialised. The second is whether the Council had actual knowledge of that risk. The evidence was clear that the Council had no knowledge that any of the trees in the vicinity of the culvert were distressed. Mr Brookfield said that the Council had no notice that the water main had been installed. However, both Mr Bewsher and Mr Burn stated that the November 1999 survey revealed the existence of the water main and the Telstra line. There were also the references on the drawing for the road widening proposal to which I have referred. 189 Mrs Turano identified two risks, which she said resulted in the harm that materialised. First, she contended that the laying of the water main and Telstra line above the level of the culvert outlet pipe, had caused water to be dammed and thus to stay in the pit longer than would have been the case had the culvert remained free-draining, as it was designed to be. The second was that by laying the water pipe in sand, water had been conducted along the sand in the direction of the tree. It was submitted that the combination of both of these factors had caused the tree roots to become compromised, which in turn, had affected the stability of the tree, resulting in it being blown over in high wind conditions. 190 I agree that the particular risk that materialised was as articulated by Mrs Turano. Did the Council have actual knowledge of that risk? It is apparent from the 1999 survey that the Council had knowledge that the water main and the Telstra line had been installed in the vicinity of the culvert. Assuming for the moment that the notification of the water main on the survey constituted relevant actual knowledge of that fact by the Council, there was no evidence that anyone within the Council had ever inspected the water main, or had observed that it had been laid in sand. 191 Accordingly, there was no evidence of actual knowledge in the Council of one of the basal facts constituting the particular risk, the materialisation of which resulted in the harm. That is sufficient for the Council to be protected from liability in this case. I am also of the view that there was no evidence that, as at the date of the accident, the Council had actual knowledge that the water main was laid above the level of the discharge area, so as to cause water from the culvert to dam or pond. 192 I am of the opinion, therefore, that it has not been established that the Council had actual knowledge of the risk, the materialisation of which resulted in the harm in this case. It follows that s 45 operates and the Council has no liability to Mrs Turano.

Case against Sydney Water 193 The trial judge stated, uncontroversially, at [147], that the liability of a public authority is to take reasonable care, having regard to the provisions of ss 5B, 5C, 5D and 5E of the Civil Liability Act and the principles stated by the High Court in Leichhardt Municipal Council v Montgomery [2007] HCA 6; (2007) 233 ALR 200 and in Brodie v Singleton Shire Council [2001] HCA 29; 206 CLR 512. The application of these sections and the principles in Montgomery and Brodie were not disputed and do not call for discussion. It should be noted that s 45 is not relevant to the claim against Sydney Water. Section 42 is relevant, but no argument was addressed to its operation in the case. 194 The trial judge stated, at [144], that there was evidence that the water from the culvert was probably partly dammed by the position of the water main sitting at a higher level than the discharge from the culvert and that water ran along the sand base of the water main, allowing water to more readily reach the tree's root system. There is no challenge to this finding by any of the parties, except that counsel for Sydney Water submitted that this finding was speculative. This was a reiteration of an argument put at trial, to which his Honour referred at [103]. In that paragraph, his Honour accepted Sydney Water's submission that the effect of the installation of the water main on the surrounding area was speculative. However, given his Honour's express finding at [144] I will ignore what his Honour said at [103] and proceed on the basis that his Honour's factual finding was as stated at [144]. 195 His Honour held, at [145], that there was no obligation to continue to monitor the water main after it was installed in the absence of any complaint of a problem with it. His Honour observed that there was no evidence of any complaint since its installation. It followed, in his opinion, that there was no duty to monitor and in any event no breach because there had been no complaint. 196 His Honour, at [146], referred to the manner in which the water main was laid, that is in sand, and held that it was laid consistently with good practice at the time it was installed. He concluded, at [148], that it was not foreseeable that if the water main was laid on sand through impermeable clay, the rootball of the tree was likely to have been compromised by water running down the bed of sand, thus eventually weakening the tree and making it prone to falling. His Honour considered that proposition to be far-fetched and fanciful. 197 His Honour stated, at [149], that there was no evidence that the water main had been laid other than in accordance with a well-recognised and appropriate system at the time it was installed. His Honour added that there was no evidence that anyone from Sydney Water needed to take into consideration the health of any of the trees, or that they needed to have an arborist to give them an opinion before laying the main. His Honour concluded, at [150], that whatever water came from the culvert via the water main to the tree, it was not foreseeable by Sydney Water that such water would undermine the tree to such an extent as to critically affect its stability. His Honour held, therefore, that Sydney Water did not owe a duty of care to Mrs Turano. 198 His Honour then turned the question of the content of the duty of care, should he have been wrong in his determination that there was no duty and held, at [151], that the content of that duty was only to respond to complaints about the condition of the main. As there had been no complaints, there was nothing that Sydney Water was required to do to avoid the risk that the tree would fall and injure a person driving along Edmondson Avenue. 199 Mrs Turano contends that his Honour, in finding that it was not foreseeable that such water would undermine the tree, applied the wrong test of foreseeability. The argument focussed on the principles as stated in Wyong Shire Council v Shirt, rather than on the provisions of s 5B, although this will make no relevant difference to the decision. In Wyong Shire Council v Shirt, Mason J stated, at 47: "In deciding whether there has been a breach of the duty of care the tribunal of fact must first ask itself whether a reasonable man in the defendant's position would have foreseen that his conduct involved a risk of injury to the plaintiff or to a class of persons including the plaintiff."

Section 5B(1)(a) provides in respect of a risk that is foreseeable that it is " a risk of which the person knew or ought to have known ." 200 The particular finding of which complaint is made is at [150]. I have already referred to that but it is preferable to set out his Honour's determination in the terms stated by him. He said: "In addition, I am satisfied that whatever water came from the culvert via the water main to the roots of the tree, it was not foreseeable by Sydney Water that such water would undermine the tree to such an extent that it would eventually become unstable and fall. I am therefore not satisfied that there was any duty on Sydney Water."

His Honour's finding at [148] was to similar effect. 201 Mrs Turano submitted that his Honour determined the question of foreseeability by looking at the actual harm that had occurred, rather than determining whether it was foreseeable that the installation of the water main in a bed of sand involved a risk of injury. 202 The test of foreseeability does not require a determination whether the precise event which caused the harm was foreseeable. Rather, as Barwick CJ stated in Mount Isa Mines Ltd v Pusey [1970] HCA 60; (1970) 125 CLR 383, at 390: "… the rarity of such an injury in the circumstances does not in my opinion deny the foreseeability of an injury of the class of which it forms one. That it is sufficient that the class of injury as distinct from the particular injury ought to be foreseen as a possible consequence of particular conduct in order to establish liability for damages for the particular injury is well established. (See e.g., Chapman v Hearse [1961] HCA 46; (1961) 106 CLR 112, at p 115)" 203 I consider that Mrs Turano's challenge on this point is correct. His Honour had regard to the particular event that caused the harm and determined that the risk that had come to pass was not foreseeable. However, the question to be determined was whether it was foreseeable that, by laying the water main in sand which acted as a conduit for water, in circumstances where the water main was installed in a position that both breached the existing drainage system and obstructed the drainage of water from the culvert, that there could be an effect on the surrounding area such as might cause harm. 204 In my opinion, the harm that resulted from the tree falling was foreseeable in the sense that it was not far fetched or fanciful. It would be well known to a water authority responsible for drainage systems throughout the state that the continuous presence of water is likely to have an effect on the surrounding area. It would also be foreseeable that the introduction of an unnatural source of water would have an effect on the surrounding area and that the soil and/or the surrounding vegetation could be adversely affected. 205 I am also of the opinion that his Honour's formulation of the content of the duty of care was incorrect. His Honour's finding as to the content of the duty of care was based on his finding that the water main was laid in accordance with usual practice. That was tantamount to a finding that there was no negligence in its installation. However, his Honour failed to have regard to all of the facts in reaching that conclusion. Those facts, in summary form, are as I have stated them at [85]-[91]. It was incumbent upon Sydney Water to have regard to the terrain, including the presence of other installations, in which or near which the water main was laid. 206 Mr Burn's evidence was that the persons installing the water main would have realised that the creation of a north/south drain was a probable consequence of laying the water main in sand, through the culvert pit. The horticultural evidence was that the intermittent waterlogging predisposed the root system to infection. Mr Burn had observed that, to the south of the culvert, Sydney Water had laid the water main so as to ensure it did not impact on the water flow of the creeks, but had not appeared to have made any provision to ensure that, when installed, the main did not impact on the water flows from the culvert. 207 As indicated earlier, his Honour did not deal with this evidence, other than to say that the same method of installation would have been used, regardless of the level at which the water main was laid. This finding itself failed to deal with the evidence of both Mr Bewsher and Mr Burn that the level at which the water main had been laid had caused damming, so that water discharging from the culvert could not freely drain away. 208 If this water main had cut through a concrete drain, there would have been no difficulty in articulating a duty of care, the content of which included that the integrity of that drain was not to be compromised by the installation. But that is in effect what happened in this case. The clay which formed the walls of the discharge area provided an impermeable barrier, which, apart from some minimal water absorption, prevented water from soaking through to the north and south of the drain. The culvert drain was designed to function in that way. 209 The evidence that his Honour accepted was that the outlet drain to the culvert was of very impermeable clay and that sand has a high degree of permeability. It was uncontradicted that, given its high level of permeability, the sand would act as a conduit for the water for a considerable distance along the trench in which the water main was laid. The laying of the water main at a higher level than the discharge pipe from the culvert, meant that water could not drain away from the culvert as it was designed to do. 210 Given those facts and circumstances, Sydney Water had a duty to install the water main in such a way that the integrity of the culvert drainage system was not compromised. Accordingly, Sydney Water owed Mrs Turano a duty of care of the content or scope that I have described. 211 In my opinion it is also clear on the evidence that Sydney Water breached that duty in two respects. The first was that by laying the drain at a higher level than the discharge drain from the culvert, it caused periodic damming of the drain. Secondly, by laying the drain in sand, it permitted the water to drain northwards, so as to undermine the roots of the tree. 212 Before concluding on breach, mention should be made of one of the bases upon which Sydney Water sought to resist liability. It contended that given the evidence that the root system was coextensive with the spread of the crown, it was open to conclude that the roots would have reached the pit of the culvert, which flooded from time to time. It followed, on this submission, that it was open to infer that the water which reached the roots and created the environment in which the pathogen could develop, could well have come directly from the culvert pit. 213 The Council submitted that this case was new and had not been advanced at trial. As I understand it, Sydney Water did not contend otherwise. In any event, the evidence does not support the submission. Although the arborists were examined as to whether the root system extended to the width of the canopy, they were not asked whether the roots had reached or penetrated the culvert or whether water ponding in the pit from time to time would have affected the roots in a way that permitted the pathogen to develop. 214 I am also of the opinion that the harm the Turano family suffered was causally related to Sydney Water's breach of duty. In summary, as has been discussed at some length already, the insult to the tree as found by his Honour, at [144], was caused by water reaching the root system as a result of the water running along the sand base of the water main allowing water to more readily reach the root system. The position was aggravated by the water ponding from time to time, because it was dammed by the position of the main, relative to the invert of the culvert. The intermittent waterlogging of the root system facilitated the introduction of a pathogen to the root system, which weakened the root system such that the tree fell in the high wind conditions on the day of the accident. 215 The only evidence not in accord with this, was that of Mr Clark, who was of the opinion the soil around the tree had been softened by rainfall, rather than from any flooding from the culvert. It is apparent from the findings his Honour made that he did not accept Mr Clark's conclusion as to why the area surrounding the tree was wet and none of the parties, including Sydney Water, relied upon Mr Clark's evidence. I should add that counsel for Sydney Water submitted that the water may have or was just as likely to have reached the tree via sand in which the Telstra PVC line was laid. However, there was no evidence to support that hypothesis.

Failure to give reasons 216 Finally, I wish to say something about the trial judge's reasons. Both the Council and Mrs Turano contended in their grounds of appeal that the trial judge failed to comply with the judicial obligation to give reasons. Mrs Turano, in particular, contended that the claim against Sydney Water required his Honour to undertake a detailed evaluation of the opinions of the various experts where those opinions differed. She submitted that his Honour failed to address the technical evidence, so as to sufficiently expose his reasoning process. 217 The Council did not elaborate on this ground of appeal in its written submissions. In oral argument, although senior counsel directed a number of complaints to his Honour's reasoning process, the focus of those complaints was on certain findings made by the trial judge which, it was contended, was contrary to the evidence. There were also complaints relating to a failure to explain his reasoning process, to which I have referred in the course of examining the major issues on the appeal. 218 The legal principles that apply to the judicial obligation to give reasons is well known: Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247; Beale v Government Insurance Office (1997) 48 NSWLR 430; Waterways Authority of New South Wales v Coal & Allied (Operations) Pty Limited. The duty extends to engaging with the expert evidence and to explain why expert evidence is accepted or rejected: Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377; Moylan v Nutrasweet Company [2000] NSWCA 337; Wiki v Atlantis Relocations (NSW) Pty Limited [2004] NSWCA 174; (2004) 60 NSWLR 127. 219 Given that little emphasis was given to this issue in the argument on the appeal, it is not necessary to enter into a lengthy exegesis of the law, or an analysis of his Honour's judgment. It is sufficient to confine my comments as follows. 220 His Honour set out portion of the evidence of each of the expert witnesses. He did not always indicate whether he accepted the evidence recorded in his judgment. At a later stage in his judgment when he was determining questions of breach, duty and causation, his Honour, from time to time made reference to there being "evidence" in relation to a particular matter. It was not clear whether such references were intended to be factual findings. If they were, they were sometimes inconsistent with earlier factual findings. Nor did his Honour identify the differences in the expert opinions or indicate what evidence he rejected, or what evidence he preferred, to the extent that there was a conflict. His Honour also made certain findings of ultimate fact that were inconsistent with his primary fact finding. These matters have been discussed in the course of my reasons. 221 I consider that these problems with his Honour's judgment did amount to appealable error in the sense discussed in the cases to which I have referred. However, I consider that this Court is able to come to its own conclusion on the appeal and the cross-appeal and there was no opposition by the parties to that course. 222 Since preparing my reasons in this matter, I have had the advantage of reading in draft the reasons of Hodgson JA and McColl JA. I agree with the additional comments of Hodgson JA, including those prepared following the receipt of McColl JA's reasons. For the sake of completeness, however, I should add some brief comments in response to her Honour's approach and ultimate conclusion. In doing so, I do not propose to deal with every matter her Honour raises, whether I agree with it or not. To do so would potentially lead to interminable debate about essentially factual matters. In the main, therefore, I deal shortly with certain matters of principle. 223 The first relates to expert evidence generally. This case was a little unusual, in that some experts were not cross-examined. Critically, Mr Clark, engineer, was not cross-examined. Where the evidence of a witness is unchallenged, then there usually needs to be good reasons why that evidence should not be accepted: see Precision Plastics Pty Ltd v Demir [1975] HCA 27; (1975) 132 CLR 362 per Gibbs J at 370-1; Hull v Thompson [2001] NSWCA 359 per Rolfe AJA at [21] (Sheller JA and Davies AJA agreeing); State Rail Authority of New South Wales v Brown [2006] NSWCCA 220; 66 NSWLR 540, per Basten JA at [68]; Uszok v Henley Properties (NSW) Pty Limited [2007] NSWCA 31 per Beazley JA at [121]-[122] (Basten JA agreeing). This has some relevance in this case, because McColl JA prefers certain aspects of Mr Clark's evidence relating to the tail-out drain, over the evidence of Mr Bewsher: see [359] of McColl JA's judgment. The other expert engineers, Mr Bewsher and Mr Burn, were cross-examined. As I have already indicated, the trial judge, at [52], accepted Mr Bewsher as a "reliable witness, accurate, not being prepared to speculate". His Honour also accepted specific portions of Mr Burn's evidence. 224 The fact that two expert witnesses were examined and cross-examined and one was not, raises a question as to how a trial judge should deal with such evidence. In this case, the trial judge accepted the evidence of Mr Bewsher, after his evidence had been explained in his evidence in chief and exposed to scrutiny in cross-examination. His Honour's finding in respect of Mr Bewsher's evidence, that it was reliable and accurate, with an absence of speculation, was, in my opinion, a credit finding of the type in respect of which this Court must be constrained in rejecting. 225 In those circumstances, I am of the opinion that notwithstanding that evidence may not be challenged in cross-examination, it is open to a trial judge (and to this Court) not to accept, or fully accept it, where it is inconsistent with evidence which has been subject to and which has survived challenge by the opposing party. For this reason, I am of the opinion that the Court should act upon Mr Bewsher's evidence. 226 Having said that, the matter on which McColl JA accepts Mr Clark's evidence over that of Mr Bewsher may be of little moment. The contending views in relation to the tail-out drain was whether such a drain had been part of the original construction and whether it would have been in existence (that is, not completely overgrown) at the time that the water main was installed. Mr Bewsher and Mr Lumsdaine considered such a drain had been constructed. Mr Bewsher considered the drain may have extended some 40 m to the west, although he acknowledged that what he observed in aerial photographs may have been a natural depression. Thereafter in his report, he refers to the "drain/watercourse". Mr Lumsdaine considered that the tail-out drain extended some 10 m. If the "extension" that Mr Bewsher identified was a depression, then it is likely that the tail-out drain evolved into that depression. This is implicit in Mr Bewsher's report of 6 April 2006. 227 The real question, however, was whether there was a tail-out drain at all. Mr Clark was quite adamant that there was no tail-out drain. This was inconsistent with all of the other evidence, which eventually accepted that the construction of the culvert initially included a run-off mechanism by way of a tail-out drain. Mr Clark's opinion that there was no tail-out drain was central to the conclusion he reached that the blockage of the water discharged from the culvert was due to unexcavated natural ground to the west of the open pit at the culvert outlet. This was another way of saying the culvert was not free draining, as there was no tail-out drain. To the extent that Mr Clark's opinion depended upon there being no tail-out drain, it should be rejected. 228 Further, as Hodgson JA points out, the evidence did not establish that the tail-out drain had ceased to function by 1981, when the water main was installed. In addition, there was evidence that the tail-out drain had been compacted (both Mr Lumsdaine and Mr Clark refer to this). As Hodgson JA states, it is a reasonable inference that was done by Sydney Water. 229 There are other difficulties with Mr Clark's opinion. Importantly, he does not deal with the question of what caused the problem with the tree's root system. In his report of 24 January 2005, he considered the flooding around the tree observed by Mr Meys probably occurred after the tree had fallen over, that is, it was rainwater that had accumulated in the stump hole and not the water from the culvert. He concluded, therefore: "I do not know to what degree the ground around the tree had been softened by water during this storm or the intensity of the rainfall prior to the tree falling. I understand that strong winds were part of the storm. Meys has described the poor state of the root system. From my experience trees with poor root systems can be blown over by strong winds even in dry conditions. I consider that the major cause of the tree falling was the combination of strong winds and the poor root system ." (Emphasis added) 230 It should be noted that Mr Bewsher and Mr Burn agreed that the "flooding" of the tree hole after the accident was likely to be the result of surface flow, rather than underground seepage. However, that still did not explain the cause of the diseased condition of the roots of the tree. The evidence relating to the tree root system was that it was affected by a pathogen and that was due to the roots being unnaturally waterlogged. The question then arose as to how the roots became waterlogged. This leads to the second difficulty with acceptance of Mr Clark's evidence. Although he was provided with Mr Lumsdaine's report of 3 April 2006, Mr Clark did not deal with the evidence of Mr Lumsdaine relating to the highly permeable nature of the sand in which the water main was laid. His focus was on whether the water main had obstructed the discharge of water from the culvert. He does not deal with the effect of the wall of the discharge area from the culvert having been breached by Sydney Water and the consequence of that, in circumstances where by laying the water in sand, Sydney Water in effect created an underground drain. 231 This brings me to another aspect of the expert evidence, namely, that of Mr Lumsdaine. Mr Lumsdaine's evidence was critical to establishing that water reached the rootball of the tree through the sand. Such a conclusion was a matter of inference, but it was a clear inference drawn from established facts. Mr Lumsdaine's evidence was subject to challenge by way of cross-examination. However, there was no competing expert evidence and for the reasons I have already given, his evidence, if not accepted by the trial judge, should have been accepted. 232 I should also state that I do not agree with McColl JA, at [322] of her Honour's judgment, that Mr Lumsdaine's evidence lacked an essential factual foundation. It was not in contention that the water main was laid in sand. Further, the grade of the water main trench could be inferred from the grade of the road. In any event, Mr Bewsher stated: "Water in the sand backfill could travel in either a northerly or southerly direction, depending on the grade of the base of the water main trench. The details of the water main provided to me … do not indicate the grade. Whilst it is possible that the water could be travelling in either a northerly or southerly direction, if the water main trench at this location was in a lowpoint, water would pond at the bottom of the trench. In this situation it is possible that the water main trench may be acting as a large 'subsoil drain' and directing seepage flows from areas many tens or hundreds of metres away towards the problem tree." (Blue book 413) 233 I also consider that although an expert's evidence should remain within the relevant area of expertise, an expert's experience in matters relevant to their areas of expertise is not to be discounted. In this regard, Mr Freeman said it would be foreseeable to a civil engineer that permanent waterlogging would prejudice trees. Mr Burn considered that there was sufficient expertise in the Council (by which it appeared he meant engineering expertise) for it to be aware of the effect of drainage construction on the surrounding terrain and structures. That, with respect, would seem to me to be a matter of commonsense. For these additional reasons, in addition to those of Hodgson JA, I have found no reason to alter the conclusions I have reached in respect of these appeals. 234 I would add one final matter. McColl JA has been critical of my finding, at [54], in relation to Mr Burn's evidence as to the blocking of the culvert. The evidence with which I was dealing there was his assertion that the culvert itself was so clogged that water would flow across the road in times of flood. That evidence did not survive his cross-examination. The real issue in this case did not relate to the congestion in the culvert itself. If water was dammed at the inlet, that would have prevented water flowing through the culvert. The problem was the cause of any obstruction on the outlet side of the culvert and the consequence of laying the water main, in sand, through the outlet area. 235 I propose the following orders:

1. Grant leave to the Council of the City of Liverpool to appeal from the orders and declarations of Delaney DCJ made 2 May 2007;

2. Grant leave to Maria Turano to appeal from the orders and declarations of Delaney DCJ made 2 May 2007;

3. Appeal by the Council of the City of Liverpool allowed;

Set aside the following orders and declarations stated at [155] of the judgment of Delaney DCJ: (1), (2), (3), and (4);

Order that there be a verdict for Mrs Turano against Sydney Water Corporation on the question of liability;

Order that there be a verdict for Liverpool City Council on the Statement of Claim;

Order that the claim against Sydney Water Corporation be remitted to the District Court for determination of damages;

Direct the parties to file written submissions within 7 days as to the costs orders that should be made on the appeal and in respect of the liability hearing at first instance. 236 HODGSON JA: I agree with the orders proposed by Beazley JA, and I agree substantially with her reasons. 237 As noted by Beazley JA, the culvert and outlet pit were constructed by the Council in the 1960's, and the water main was installed by Sydney Water in about 1981. Later, it would seem, Telstra installed a line running parallel to the water main, together with a nearby pit. 238 The evidence of the geotechnical engineer Robert Lumsdaine, which was not significantly challenged or contradicted, and which either was or should have been accepted by the primary judge, included the following (Blue 389-390): 33. When the outlet pit was originally constructed, I believe that it was excavated through a shallow depth of filling (if it was present at the time), and into very stiff residual clay. The base of the outlet pit would have exposed very stiff residual clay. 34. Very stiff residual clay is a very impermeable material compared to say clean sand or clean gravel or highly fractured rock. Based on presumptive values of soil permeability (i.e. tables of typical values given in soil mechanics text books), I would expect the clean sand backfill to have a permeability at least three orders of magnitude (i.e. 1000 times) greater than the clay. It is quite plausible that the value could be 5 orders of magnitude greater than the clay. Water ponded intermittently against a face of very stiff residual clay might saturate and soften the outer 'skin' of the exposed clay, but it would not flow (seep) through the body of the clay.

35. There is evidence of there once having been a tail-out drain from the outlet pit. The base of the observed (presumed) tail-out drain was [650 BTP], compared to the invert level of the pipe culvert outlet of [750 BTP]. The top of the water main is at about [510 BTP]. The top of the Telstra PVC pipe is at about [160 BTP].

36. If there once was a tail-out, then it appears that at some stage it has been backfilled with clay filling using some compactive effort. I do not know if the backfilling of the presumed drain was carried out as part of the installation works for the water main or the Telstra line.

37. If there was no original tail-out drain, then (prior to Water Board main construction) water passing through the culvert would have merely ponded in the outlet pit, due to the impermeable clay sides and base. There would have eventually been some overflow once the water level reached the lowest part of the lip of the pit perimeter. The remaining ponded water level would eventually fall (albeit slowly) due to evaporation and (to a much lesser extent) some soaking into the clay soil faces of the outlet pit. 38. If there was no tail-out drain, then the installation of the Water Board main would not have made the ponding any deeper than it already was, except possibly for a minor raising of the level of the lowest overflow point, if this was previously located along the western side of the pit.

On the contrary, the construction of the Water Board main probably inadvertently provided an outlet drain to the pit. The clean sand backfill around the water main would act as an efficient outlet drain for the pit, provided the sand backfill detail observed in the pit is typical for the water main beyond the pit location.

Water ponded in the outlet pit would seep into the sand stratum, percolate downwards through the sand, and then flow along the top of the impermeable clay base of the trench. The direction of water flow would depend upon the profile of the base of the trench.

If the base of the trench in clay had been excavated with a uniform fall in level for a considerable distance in a particular direction, then I would expect the water to flow in that direction until such time that the outlet pit was drained.

If the base of the trench in clay had been excavated with localised hollows etc however, then I would expect that the hollow (now sand-filled) would become a "sump", permanently full of water. Subsequent water from ponding in the outlet pit would still escape from the pit and essentially flow over these permanently water filled sumps.

During the investigation, the perched water level in the sand was about 900 mm below the top of the Telstra pit ( [900 BTP]). On Monday 27 February 2006, after a weekend of heavy rain, the ponded level in the outlet pit was about [50 BTP]. Reference to Drawing 6A indicates that this level fell about 500 mm over the next three weeks. I would expect that the ponded water will gradually fall to the originally observed level by means of evaporation and drainage through the sand backfilling. 239 The BTP figures represent the number of millimetres below the Telstra pit cover. 240 Mr Lumsdaine's observations support the inference, which I would draw, that when the culvert and pit were constructed by the Council, there was a tail-out drain from the pit at a level of about 100mm above the lower level of the culvert outlet, so that the pit would be drained above that level. The remaining 100mm of water would be retained by the impermeable clay and gradually evaporate, and perhaps to a small extent drain. To that extent only, the culvert was then a "drowned culvert". 241 As shown by Mr Lumsdaine's drawings (Blue 394-399), when the water main was installed, the main itself effectively dammed the pit to a height of about 240mm above the lower level of the culvert outlet (510 BTP: Blue 395). Filling above the main, which was probably placed there at the same time, dammed it to a further height in excess of 400mm. At the same time, the impermeable clay walls of the pit were breached by the main and by the sand-filled trench in which it was laid; and this trench itself was excavated down to a depth of about 400mm below the lower level of the culvert outlet (that is, down to 1150 BTP: Blue 395). 242 The inevitable result was that water, to a height of more than 640mm above the lower level of the culvert outlet, would not escape freely from the pit, nor be retained there by impermeable clay, but rather would seep away through the sand surrounding the water main. Mr Lumsdaine recorded the water level in the outlet pit falling from about 700mm above the lower level of the culvert outlet (just after heavy rain) to about 200mm above that level over a period of 23 days, during the whole of which period all the sand around the water main would have been saturated (see Blue 399). I note also that, even at the level at which the water was observed before the rain (900 BTP), the sand in the water main trench would have been saturated to a depth of about 250mm; and I would infer that saturation to that depth at least would persist over long periods of time. 243 These were circumstances of a kind readily foreseeable by Sydney Water, whose business involved management of water. The radical changes in disposition of water, caused by the works being carried out as they were, should have caused it either to perform the works differently so as not to interfere with the impermeable pit and drainage from it, or to properly investigate the consequences of periodic saturation of the sand-filled trench in which the main was laid. That in turn should have alerted Sydney Water to the possibility of the roots of the nearby tree being adversely affected. In either case, Sydney Water should have acted so as to avoid the risk that eventuated. 244 But for the intervention of Sydney Water, little water would have been retained in the pit, and that water would have been prevented from saturating the soil where the tree roots were located by the clay walls of the pit. Accordingly, the Council was not shown to be in breach of any duty of care. 245 Since writing the above, I have read the judgment of McColl JA. I respectfully disagree with certain aspects of that judgment, and in the interests of clarity it is convenient to specify the particular matters on which I disagree with her. 246 First, while McColl JA is correct in noting (at [286]-[287]) that Mr Lumsdaine's measurements in 2006 were made after extensions to the culvert in 2002, this does not in my opinion significantly affect the value and importance of his measurements. In particular, in my opinion, there is no good reason to think that the bottom of the culvert drainpipe was previously higher than the level of 750BTP measured by Mr Lumsdaine as the level of the apron extending from the bottom of the culvert pipe (Blue 395). 247 Accordingly, and second, the top of the water main (BTP 510) (Blue 395) was about 240mm higher than the bottom of the culvert pipe, not 100mm higher as stated by McColl JA (at [286]): the oral evidence on which McColl JA based her finding is in my opinion very uncertainly expressed and is generally far less reliable. 248 Third, in my opinion Mr Lumsdaine's evidence strongly supports an inference that an original tail-out drain, down to a level of about 650BTP (about 100mm higher than the lower level of the culvert), was deliberately filled and the fill was deliberately compacted; and in circumstances where there was no suggestion of any such activity, and no occasion for any such activity, prior to the laying of the water main in 1981, it can and should be inferred on the balance of probabilities that this filling and compacting was done by Sydney Water when it laid the water main. Thus, what Sydney Water did had the effect of damming the pit to a further height in excess of 400mm. 249 Fourth, in my opinion, the evidence does not justify McColl JA's conclusion (at [367]-[374]) that the tail-out drain had ceased to function by 1981. Until the drain was filled and the fill compacted, the drain would have functioned to some extent at least, even if affected by sediment, root growth and vegetation. Water would seep rapidly through roots and vegetation, and sediment would only gradually raise the level of the bottom of the drain. 250 Fifth, the probable radius of the tree's feeder roots was about 5 metres (McColl JA at [290] and [305]-[306]) not 10 metres ([385]); and while those feeder roots would have extended to the original pit, they would have had much greater contact with the water main trench. The trench appears to have passed within about 3 metres of the trunk, and thus been in contact with the tree's root system over a length of about 8 metres: assuming a radius of 5 metres, this follows from Pythagoras' theorem, because 52 - 32 = 42.

251 Sixth, and contrary to McColl JA's conclusions at [406]-[410], in my opinion the evidence strongly supports the conclusion that water from the water main trench reached the tree's roots and did so to an extent sufficient to cause the diseased root system. The evidence shows that sand in the trench was saturated over long periods of time to depths in excess (and often much in excess) of 250mm; and since the culvert was at a low point in a relatively flat road, it is clear that whichever way the water main fell, the fall would be slight and insufficient to make much difference over a distance of 8 metres (the length over which it was in contact with the tree's roots). 252 Seventh, and contrary to McColl JA's approach at [385]-[386], in my opinion it was incumbent on Sydney Water, undertaking work substantially affecting drainage of water in the vicinity of a road and trees, to have regard to possible risks and to have or acquire any expert knowledge necessary to assess those risks. The business of Sydney Water is the management of water, and it should make sure it has the knowledge and expertise required to do this safely. 253 Eighth, and contrary to McColl JA's view at [399], laying the water main below the culvert would have greatly reduced the risk, because it would have permitted the tail-out drain to continue to function, and because water which drained into the sand around the water main at the level below the culvert level would not flow into the higher levels of the water main trench on either side of the culvert (see Blue 1). 254 Accordingly, in my opinion, a case in negligence was made out against Sydney Water. 255 McCOLL JA: I have had the benefit of reading Beazley JA's judgment in draft and am indebted to her Honour for her detailed consideration of the evidence. 256 The case Mrs Turano ran at trial was that the tree fell because the activities of either or both the Council and Sydney Water had compromised, by wetting, the soil in the vicinity of its roots in such a manner as exposed it to a risk of destabilising its roots. The instability was said to have been caused by the introduction of a pathogen which caused the roots to rot. In my view this case was not established, as a matter of causation, against either defendant. 257 I agree with her Honour, albeit for slightly different reasons, that the Council's appeal on the finding of liability against it should be allowed. 258 I differ, however, with respect from her Honour's conclusion in relation to Sydney Water. In my view Mrs Turano's cross-appeal should be dismissed as, too, should Ground 2 of the Council's appeal which sought to preserve the Council's position on its cross-claim for contribution or indemnity against Sydney Water. 259 Mrs Turano also sought to argue by Notice of Contention that the primary judge ought to have found that the tree roots were compromised by the carriage of water to them along the sand bed of the water main and that she was entitled to a verdict against the Council on the basis that it was negligent in the circumstances for allowing the occasion for the collection of the water which entered the sand to exist. For the reasons which follow, that contention should also be rejected. 260 Beazley JA has set out the facts. I shall only repeat such as are necessary to explain my reasons.

Statutory framework 261 The case against Sydney Water fell for determination in accordance with Pt 1A, Divs 2 and 3 of the Civil Liability Act 2002 which, relevantly, state: "5B General principles (1) A person is not negligent in failing to take precautions against a risk of harm unless: (a) the risk was foreseeable (that is, it is a risk of which the person knew or ought to have known), and (b) the risk was not insignificant, and (c) in the circumstances, a reasonable person in the person's position would have taken those precautions. (2) In determining whether a reasonable person would have taken precautions against a risk of harm, the court is to consider the following (amongst other relevant things): (a) the probability that the harm would occur if care were not taken, (b) the likely seriousness of the harm, (c) the burden of taking precautions to avoid the risk of harm, (d) the social utility of the activity that creates the risk of harm. 5C Other principles In proceedings relating to liability for negligence: (a) the burden of taking precautions to avoid a risk of harm includes the burden of taking precautions to avoid similar risks of harm for which the person may be responsible, and (b) the fact that a risk of harm could have been avoided by doing something in a different way does not of itself give rise to or affect liability for the way in which the thing was done, and (c) the subsequent taking of action that would (had the action been taken earlier) have avoided a risk of harm does not of itself give rise to or affect liability in respect of the risk and does not of itself constitute an admission of liability in connection with the risk. 5D General principles (1) A determination that negligence caused particular harm comprises the following elements: (a) that the negligence was a necessary condition of the occurrence of the harm ( 'factual causation'), and (b) that it is appropriate for the scope of the negligent person's liability to extend to the harm so caused ('scope of liability'). (2) In determining in an exceptional case, in accordance with established principles, whether negligence that cannot be established as a necessary condition of the occurrence of harm should be accepted as establishing factual causation, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. (3) If it is relevant to the determination of factual causation to determine what the person who suffered harm would have done if the negligent person had not been negligent: (a) the matter is to be determined subjectively in the light of all relevant circumstances, subject to paragraph (b), and (b) any statement made by the person after suffering the harm about what he or she would have done is inadmissible except to the extent (if any) that the statement is against his or her interest. (4) For the purpose of determining the scope of liability, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. 5E Onus of proof In determining liability for negligence, the plaintiff always bears the onus of proving, on the balance of probabilities, any fact relevant to the issue of causation."

Statement of the case 262 Mrs Turano's case against Sydney Water was particularised in a number of ways, all of which, in substance, alleged negligence in the installation of the water main. The Council's cross-claim adopted Mrs Turano's particulars of negligence. It also asserted Sydney Water had been negligent in destroying the tail-out drain and in laying highly permeable sand around the water main thus permitting water from the culvert to pass along the pipe trench to the roots of the tree, causing the roots to rot and the tree to become unstable. 263 Sydney Water's case was simple. It denied it had acted negligently. It put in issue foreseeability and causation. It said the water main was laid correctly and in accordance with correct procedures. It argued that Mrs Turano/the Council's cases depended on the fallacy that the tree's roots were only beneath its trunk and that the clay soil was completely impermeable. It contested the proposition that, if the tree was affected by a pathogen introduced by water, Mrs Turano had established the water main trench as the, or a, source of that water. It argued there was no evidence that the water "travelled" in any relevant sense in a manner which affected the tree's roots. It relied on the evidence of a number of potential water sources including subterranean water naturally occurring and surface water permeating the soil as being equally available inferences as to the source of water, if water damaged the tree's roots. It submitted that the case against it based speculation upon speculation to seek to draw an inference in favour of Mrs Turano/the Council's cases. 264 Sydney Water also argued that it was not reasonably foreseeable to it at the time the water main was constructed that if water did travel along the water main trench, it could damage the tree's roots, let alone cause a pathogen to affect them. It submitted that it was not foreseeable that allowing water to flow past the tree's roots could cause it to fall, 20 years later, in a wind storm. 265 Sydney Water also submitted that Mrs Turano/the Council had to demonstrate that an alternative method of construction would have made a difference. It contended this had not been established. It pointed out that Mr Burn had said that even if the water main had been laid deeper, I infer, than the culvert outlet, water would still have had to pass along it. It also relied upon Mr Lumsdaine's evidence, to which I will return, that the effect of installing the water main was to create a drain for the culvert. 266 Sydney Water also contended that the issue of breach had to be determined in the context of its responsibility to provide water to the Liverpool region. 267 Many of the propositions upon which Mrs Turano's case against Sydney Water depended were uncontroversial and, as found by the primary judge, were:

(a) Sydney Water installed a water main near the western outlet of the culvert in 1981 (judgment at [143]);

(b) the water main was laid on a bed of sand and surrounded by sand which had a permeable quality (judgment at [143]);

(c) the main was adjacent to the roots of the grey box gum which fell and struck the car driven by Mrs Turano's late husband (judgment at [143]);

(d) the root ball of the tree which fell was infected by a fungal infection (judgment at [143]); and

(e) the installation of the water main interrupted the flow of water from the western outlet of the culvert and would have, more likely than not, from time to time, created pools of water that could not escape for various reasons (judgment at [143]). 268 The primary judge made different, and inconsistent findings about Sydney Water's liability. He said (at [88]) that he did not accept a submission by Mrs Turano's counsel that "the water only came to the roots of the tree via the water main". However he accepted (at [103]) Sydney Water's argument that it was a matter of speculation as to whether water from the water main reached the tree's root system. He later found there was evidence that the water from the culvert probably was partly dammed by the main and water ran along the sand base of the water main, allowing water to more readily reach the root system of the tree that fell, than had the water main not been installed (judgment at [144]). In my view his Honour's conclusion that it was a matter of speculation as to whether water from the water main reached the tree's root system was correct. Regrettably his Honour did not expose his reasons for that conclusion. 269 The primary judge held (at [146]), that the water main was laid in a manner which was consistent with good practice at the time. He accordingly rejected Mrs Turano's case that the water main should have been installed in a different manner having regard to the surrounding soil conditions. He also held (at [149]) that there was no evidence indicating the system of laying a water main was other than a well-recognised and appropriate system at the time it was installed. Both the Council and Mrs Turano challenged that conclusion. 270 The primary judge also held (at [149]) that there was no evidence that anyone from the Metropolitan Water Sewerage & Drainage Board (the name of Sydney Water when the water main was installed in 1981) needed to have given any consideration to the health of any of the trees nearby or that they needed to have an arborist give them an opinion before laying the main. No one challenged the latter finding. 271 The primary judge also held (at [148]) that Mrs Turano's contention, that if the main was laid on sand through impermeable clay it was foreseeable that the root ball of the tree which fell was likely to have been compromised by water running down the bed of sand, was far-fetched and fanciful. In his Honour's view (at [150]) whatever water came from the culvert via the water main to the roots of the tree, it was not foreseeable by Sydney Water that such water would undermine the tree to such an extent that it would eventually become unstable and fall. Mrs Turano challenges that finding, contending that his Honour applied too high a test and that the question he ought to have asked was whether it was foreseeable by Sydney Water that water from the water main "could" undermine the tree in the relevant manner. 272 As a result of these findings his Honour concluded that Sydney Water did not owe Mrs Turano a duty of care. 273 The primary judge held (at [151]) in the alternative that if he was incorrect in his conclusion about duty of care, the content of Sydney Water's duty was only to respond to complaints about the condition of the water main which were communicated to it. As no concerns of that nature were communicated to it concerning the water main or the area around it, he was also of the opinion that nothing was required of Sydney Water to avoid the risk that the tree could fall and injure someone driving nearby.

The ambit of the appeal 274 In her Further Amended Notice of Cross-Appeal Mrs Turano complained that the primary judge failed to give reasons or sufficient reasons. The Council did not rely on an express no reasons ground, although Mr Watson complained, from time to time, in a desultory manner in the course of his submissions that his Honour's reasons in some areas were not sufficiently exposed. 275 As Beazley JA has said (at [216]), the gravamen of Mrs Turano's complaint was that the primary judge failed to undertake an adequate evaluation of the expert opinions, where those opinions differed and that he failed to address the technical issues so as to expose his reasoning process. 276 There may be a question as to the utility of a no reasons ground of appeal in an appellate court exercising a rehearing jurisdiction such as that vested in this Court pursuant to s 75A of the Supreme Court Act 1970: see Najdovski v Crnojlovic [2008] NSWCA 175 (at [21] - [22]) per Basten JA (Allsop P agreeing); Ali v Nationwide News Pty Ltd [2008] NSWCA 183 (at [149] ff) per Basten JA. As Basten JA explained in Ali (at [156] - [157), referring to Hayne J's observations in Waterways Authority v Fitzgibbon [2005] HCA 57; (2005) 79 ALJR 1816 (at [129] - [131]), often the real complaint lies in the fact-finding process, rather than in the exposure of reasons for conclusions. 277 The appeal, being by way or rehearing, the Court "is obliged to 'give the judgment which in its opinion ought to have been given in the first instance' [while observing] the 'natural limitations' that exist in the case of any appellate court proceeding wholly or substantially on the record". Within those constraints "the appellate court is obliged to conduct a real review of the trial and, in cases where the trial was conducted before a judge sitting alone, of that judge's reasons". The Court is not excused from the task of "weighing conflicting evidence and drawing [their] own inferences and conclusions, though [it] should always bear in mind that [it has] neither seen nor heard the witnesses, and should make due allowance in this respect": Fox v Percy [2003] HCA 22; (2003) 214 CLR 118 (at [23], [25]); s 75A(10), Supreme Court Act. 278 I also agree with Beazley JA that the primary judge failed to give adequate reasons. All parties' cases depended almost entirely upon expert evidence. 279 The primary judge indicated that he accepted some witnesses (Freeman (at [27]), Polson (at [39]), to the extent his evidence was not contradicted by "accepted oral evidence", Bewsher (at [52]), Burn (at [69])), but was non-committal as to others, including experts whose reports were tendered and who were not cross-examined (Meys, Keirnan, Clark), merely reciting their evidence. To the extent that such witnesses' evidence differed from others, his Honour failed to engage in the process of coherent analysis called for where, as in this case, none of the experts' evidence was challenged on grounds of want of expertise, partiality or any other reliability challenge: Wiki v Atlantic Relocations (NSW) Pty Ltd [2004] NSWCA 174; (2004) 60 NSWLR 127 (at [60]-[64]) per Ipp JA (Bryson JA and Stein AJA agreeing). 280 Accordingly, the primary judge failed to indicate which of the experts he preferred and why. It is significant to comment on his Honour's failure to indicate why he did not accept experts whose evidence was not attacked. A court should not readily decline to act on unchallenged evidence, unless it is glaringly improbable: Markson v Cutler [2007] NSWSC 1515 per Brereton J, citing Precision Plastics Pty Ltd v Demir [1975] HCA 27; (1975) 132 CLR 362 (at 370-1). No-one suggested any of the experts' evidence fell into the latter category. 281 With those observations in mind, I turn to consider the evidence.

The day of the accident 282 The accident occurred at about 4.15pm on 18 November 2001. The circumstances immediately attending the accident were uncontroversial and no lay witnesses were called. However statements they had given were tendered in Mrs Turano's case. According to one of the lay witnesses, Mr Vella, there was a "bad storm" and "big gusts of wind" at the time of the accident. Another gentleman, Mr Montgomery, said that at the time of the accident "there was a storm, large gusts of wind and rain". 283 The police officer who attended the scene, Senior Constable Hughes, attributed the tree's fall to "the severe storm buffeting the metropolitan area at the time". He added: "The area in which this occurred is a rural area and there were numerous other trees fallen over in the vicinity". 284 Mr Burn, a civil engineer to whose evidence I will return, reviewed weather conditions on the day of the accident from which he concluded the rainfall that day was light. He also noted that the maximum wind gust on the day was 56 knots or 103 kilometres per hour, and that those gusts occurred at around 4pm. 285 Mr Burn's evidence in this respect does not appear to have been drawn to the attention of Mr Castor, an arborist to whose evidence I will also return. However he opined that: "[T]he expected wind speed at the time of the accident needs to be acknowledged….at Gale, Force 8 wind speeds (>63 km/h) that 'twigs and branches break and shallow root trees [are] pushed over". (emphasis added)

Dimensions of the culvert, the outlet pit and the water main 286 The culvert pipe was 600mm in diameter. The water main was 300mm in diameter. The height differential between the base of the culvert and the top of the water main was approximately 100mm. The water main trench, that is to say including the sand surrounds, appears on the basis of Mr Lumsdaine's diagram 1A to have been 350mm in diameter at the culvert. 287 There are no clear measurements of the depth of the culvert pit at the time of the accident, let alone at the time the water main was laid. Some measurements appear to have been made by Mr Lumsdaine in 2006, but these have to be approached with caution, in my view, having regard to the substantial work the Council had done in the area in about 2002 when it extended the culvert. 288 The water main was 1.35 metres from the road culvert when viewed by Mr Burn in 2005. The culvert had been extended by approximately 1.3 metres in about 2002. It is a reasonable inference, in my view, that the water main was laid about 2.6 metres from the end of the culvert. 289 According to Mr Burn, public utility services had to be laid a minimum of 900mm below ground. When Mr Clark visited the site in January 2005, he measured the depth of ground above the water main to be 925mm which he opined was consistent with a standard depth for a water main in this location. He also measured the depth of the culvert pipe invert from the surface of the natural ground to be about 1100mm. 290 It is probable that the tree was about 4 metres north of the culvert and had a canopy of 10 metres. A tree of those dimensions was identified on a 1999 survey by Mr Bewsher who also said the water main was about 2 metres to the west of the tree's centre. Mr Bewsher identified a tree "with a significant canopy" in approximately the same position in 1978 and 1982 aerial photographs, however Mr Bewsher could not be certain it was the same tree. None of the arborists reviewed the aerial photographs. The arborists' evidence 291 The arborists' evidence differed. (a) Mr Freeman 292 Mr Freeman, who as Beazley JA has observed, wrote both the report of CPE Tree Services dated 20 November 2001 (ostensibly signed by Mr Polsen) and that penned under his own name of 27 September 2006, was of the opinion in his first report that the tree had failed due to a combination of root damage from a root pathogen which he attributed to the permanent rise in the water table from the "roadside drains poor performance to drain away and being located around the tree's root system, and the rotting root system of the remaining root ball from the permanent waterlogging". In his second report in 2006 he attributed the permanent rise in the water table to the absence of a tail-out drain. As Beazley JA has noted (at [27], [31]) counsel for Mrs Turano, who called Mr Freeman, elicited from him that his reference to the "water table" was, in fact, a reference to the "water level" in the drain underneath the road. He also properly deferred to the opinions of geotechnical experts. I agree with Beazley JA's conclusion (at [28]) that, accordingly, Mr Freeman's opinion as to whether there was a permanent rise even in water levels ought not to have been accepted by the primary judge. 293 This has significant effects in terms of his Honour's findings, in particular that (at [144]) which depended on the proposition that the water levels around the tree's root systems had altered by reason of lack of drainage in the outlet pit. 294 Mr Freeman explained in his second report that his conclusion the tree's roots were affected by a pathogen was based on observations.

(b) Mr Meys 295 Mr Meys inspected the site approximately two weeks after the accident, on 7 December 2001. 296 He took a section of the tree's root system for analysis. He concluded that analysis did not reveal any evidence of harmful insects or fungal pathogens. As Beazley JA has said (at [24]), no party seeks to advance his opinion in this respect. 297 Perhaps the greatest significance of his report lies in the photographs he took of the culvert which show the culvert to be choked by grass and what appear to be roots. Mr Meys commented that the drain was "partially blocked". He observed the area around the tree to be flooded, a condition he attributed to the blocked drain area.

(c) Mr Castor 298 Mr Castor first inspected the accident scene in July 2005. 299 He both inspected the site and reviewed a number of other reports on behalf of the Council. In his opinion, there was no evidence of permanent water ponding adjacent to the tree. He relied in this respect on a photograph taken by Mr Clark on 19 January 2005 which showed a dry culvert outlet which he concluded indicated that the water from the culvert dissipated with time. 300 In his view, the potential for water logging adjacent to the tree had been present for some time, a conclusion he drew from the relative levels on a 1999 survey which revealed "a general low point in the landscape in the vicinity of the tree". 301 Mr Castor noted Mr Meys' analysis had not revealed fungal pathogens, as to which he merely observed, "the method of fungal analysis was not described". 302 He opined that it was possible the tree was accustomed to intermittent water logging and had had to cope with it since the original road formation and culvert installation. He concluded that the primary cause of the tree falling was intermittent water logging leading to a diseased root system in combination with storm and wind. 303 Mr Castor expressed the view that the installation of the water main was not likely to have had an impact on tree failure "given the time period for root regeneration". This appeared to refer to the possibility that the tree's roots were damaged in the process of construction of the water main, a case which is not pursued on appeal. 304 Mr Castor pointed out: " Not all tree failure is predictable. Not all tree failures can be explained even after the event. No tree is completely safe. Trees are living organisms which are anchored to the ground and so are subject, in situ, to activities and stresses from man and nature . … For a tree hazard to exist there must be a potential for failure and a potential for injury or damage to result. Dead trees in remote locations are often less hazardous than healthy trees in built-up areas." (emphasis added) 305 Mr Castor gave evidence. He explained that the pathogen which would have degraded the tree's root system would have entered via the feeder roots initially, then into the woodier roots with the structural strength of the roots degrading over an unknown period. He also pointed out that some trees can recover from invasion by pathogens, while others can die quickly depending on the species of tree and the species of fungus. He said a tree's feeder roots spread up to three times the height of the tree, with most being within the drip line, which I understand to refer to the canopy. On the assumption the tree was 16 - 18 metres in height, he estimated its canopy to be 10 metres in diameter. This was consistent with Mr Bewsher's evidence based on aerial photographs to which I will refer. 306 On that assumption, the pathogen most probably entered the tree's feeder roots at a point 5 metres away from the tree trunk on most points of the circumference of the crown of the tree. This was 3 metres beyond the water main trench at most points. 307 In short the arborists' evidence demonstrated that the tree was affected by a pathogen to which it was rendered susceptible by intermittent water logging. It also demonstrates that the extent to which a tree may be affected by a pathogen is unpredictable. Some trees may recover, others may die quickly. The geotechnical evidence 308 Only one geotechnical engineer was asked to prepare a report, Mr Lumsdaine, who also gave evidence. He undertook investigations around the culvert outlet pit in 2006. He was asked to assess two issues: "(a) The current depth and distribution of filling in the immediate vicinity of the outlet pit to the pipe culvert at the accident location on the western side of Edmondson Avenue. (b) The likelihood that ponded water in the outlet pit would saturate the ground several metres to the north of the pit, where the tree was originally located." (emphasis added) 309 Mr Lumsdaine's investigations revealed that within the culvert, the lower 230mm of the sand on which the water main sat was "fully saturated with free ground water, presumably perched on the underlying clay base of the trench." In his opinion the perched ground water was "trapped" by the "low permeability clay walls and base of the trench". 310 Mr Lumsdaine concluded that the natural clay strata in the area investigated was moist as opposed to saturated. The exception to this was an area exposed in the northern face of the outlet pit which he observed was saturated, a situation he concluded was "presumably due to direct contact with recently ponded water". 311 Mr Lumsdaine undertook an investigation to assess whether there had been a tail-out drain leading from the culvert outlet pit westward. He excavated an investigation trench adjacent to the western boundary fence. He concluded that there was evidence that there had once been such a drain but that at some stage it had been back-filled "with clay filling using some compactive effort". He did not know who had carried out that back-filling. The evidence that satisfied him there had been a tail-out drain was the location of refuse protruding from the western wall of the investigation trench which, according to a plan he prepared (Plan 5A), appears to have been approximately at a projected level of the pipe culvert. He identified this area on his plan as "Possible back-filled tail-out drain (?)". He also identified a greater depth of filling in line with the culvert which he opined "may represent an original tail-out drain (which was subsequently back-filled) to the outlet pit". 312 Mr Lumsdaine identified the depth of the deeper filling at approximately the same level as the existing ground surface level measured about 10 metres west of the boundary fence on the alignment of the pipe culvert. Hence he opined "an original tail-out drain at the level described… would only need to have continued for about 10m to the west before being able to discharge water onto the ground surface". 313 Mr Lumsdaine concluded that if there had been a tail-out drain from the outlet pit, its base was 650 BTP (below top of Telstra pit) compared to the invert level of the pipe culvert outlet of 750 BTP. The top of the water main was 510 BTP. As I interpret those figures they indicate first, that if there was a tail-out drain when the culvert was constructed, it did not drain the base of the culvert, only water which accumulated above 650 BTP and secondly, the water main only overlapped with it to the extent of 140 mm. These measurements would also support the primary judge's conclusion that, assuming the tail-out drain was in existence when the water main was laid, the latter only partly dammed it. 314 During Mr Lumsdaine's investigations there was heavy rainfall in the area leading to the outlet pit filling with run-off water. Mr Lumsdaine investigated the rate of fall in the ponded level in the outlet pipe over the period 27 February - 21 March 2006. He concluded that the level fell about 500mm during that period and opined that he would expect "the ponded water [to] gradually fall to the originally observed level by means of evaporation and drainage through the sand back-filling." No party, as I understand the submissions sought to place any greater weight on these observations than as reflecting what might have been independently inferred, that water ponded in the outlet pit and gradually dispersed. Any greater reliance on these observations in my view would be misplaced having regard to the fact these tests were done 25 years after the water main was laid, and, also significantly, 4 years after major road works in the area including the extension of the culvert by a substantial concrete addition. 315 Mr Lumsdaine concluded that if there had not been a tail-out drain then, before the water main was constructed, water passing through the culvert would have ponded in the outlet pit due to the impermeable clay sides and base with some overflow once the water level reached the lowest part of the lip of the pit perimeter. He concluded that the remaining ponded water level would eventually fall, albeit slowly, due to evaporation and, to a much lesser extent, some soaking into the clay soil faces of the outlet pit. 316 Mr Lumsdaine further concluded that if there had been no tail-out drain then the laying of the water main would not have made the ponding any deeper than it was already, "except possibly for a minor raising of the level of the lowest overflow point, if this was previously located along the western side of the pit". Mr Lumsdaine added: "39 On the contrary, the construction of the Water Board main probably inadvertently provided an outlet drain to the pit. The clean sand back-fill around the water main would act as an efficient outlet drain for the pit, provided the sand back-filled detail observed in the pit is typical for the water main beyond the pit location. " (emphasis added) 317 Mr Lumsdaine hypothesised two possible courses water could take if it left the outlet pit via the water main trench. First, water which ponded in the outlet pit could seep into the sand stratum, percolate downwards through the sand, then flow along the top of the impermeable clay base of the water main trench, its direction depending upon the profile of the base of the trench. If there was a uniform fall in the level of the trench "for a considerable distance in a particular direction", he posited that the water would flow in that direction until the outlet pit, and, presumably the trench, drained. Secondly, if there were hollows in the base of the water main trench, he posited that they would become a sump, permanently full of water. 318 Mr Lumsdaine was not asked to, and did not, undertake any investigations to determine the profile of the base of the water main trench beyond the culvert. He did not determine whether the saturation of the sand back-fill he observed in the culvert outlet pit continued beyond it. Nor, as is apparent from the assumption which underlay his conclusion in his paragraph 39, did he investigate the composition of the sand back-fill beyond the outlet pit to determine the extent to which either drainage effect he hypothesised actually existed. Rather, his investigations focussed on the outlet pit structure itself and whether there had been a tail-out drain at some stage. 319 Beazley JA has set out Mr Lumsdaine's conclusions (at [42]). Of these it is significant, in my view, to note that in Mr Lumsdaine's opinion the clean sand back-fill around the water main "would have acted (and continues to act) as an outlet drain to the outlet pit". This, of course, was subject to the composition of the sand backfill beyond the outlet pit and the grade of the water main trench: see [316] - [317] above. 320 Mr Lumsdaine could not express an opinion whether part of the explanation for the tree falling over was saturation of the soil around its root ball by water originating from the outlet pit. In other words, he did not answer the second question posed for his consideration as to the likelihood of ponded water in the outlet pit saturating the ground where the tree was originally located. He expressed the qualified view that: "44(e) I do not know if part of the explanation for the tree's falling over was saturation of the soil around the tree's root ball by water originating from the outlet pit. But if this is part of the explanation, then it is almost certain that the route taken by the water from the outlet pit to the tree root ball was through the sand backfill surrounding the water main, and not through the clay strata. " (emphasis added) 321 It might be noted that the primary judge (at [85]) stated that: "Mr Lumsdaine was of the opinion that the water going to the tree root ball almost certainly went through the sand, the back-fill of the water main." 322 In my view when one has regard to precisely what Mr Lumsdaine said, his Honour's statement does not accurately reflect the conditional nature of Mr Lumsdaine's opinion in paragraph 44(e) of his report. Mr Lumsdaine's views about the route water might take if it left the outlet pit via the water main trench were, as I have said, qualified by his lack of information about critical facts about the water main trench beyond the outlet pit. There being no evidence of those matters, Mr Lumsdaine's opinion lacked an essential factual foundation: s 76, Evidence Act 1995, Makita (Australia) Pty Ltd v Sprowles [2001] NSWCA 305; (2001) 52 NSWLR 705 (generally, but particularly at [85]) per Heydon JA.

(a) Mr Bewsher 323 Mr Lumsdaine's report was commissioned at the request of Mr Bewsher, an engineer retained to give expert evidence on behalf of the Council. Mr Bewsher reviewed a July 1970 aerial photograph of the area from which he concluded that there was then either a water course, or more likely a dug drain, on the western side of the road which he appears to have concluded was a tail-out drain, although he did not use that precise expression. He also observed the same feature in a May 1978 aerial photograph but commented that "it had become much less distinct". Based on his engineering experience, but not, I gather, the aerial photographs, Mr Bewsher concluded that the culvert was initially free draining (i.e. at the time it was constructed), in part because "it would be a minor excavation task to create a tail-out drain for [it]". 324 Mr Bewsher said the sand backfill in which the water main was laid was highly permeable and allowed water to drain easily along the trench, potentially for a very long distance. In cross-examination by counsel for Sydney Water, he accepted that the extent to which any water which did flow along the sand in the water main trench was absorbed into the clay in which the water main was laid was a matter of some imprecision and, further, a matter for a geotechnical expert. In this respect, therefore, Mr Lumsdaine's views are to be preferred to Mr Bewsher's. 325 Mr Bewsher accepted it was possible that the water would be travelling in either a northerly or southerly direction along the water main trench, but said that the direction in which it would travel depended on the grade of the water main trench of which he was unaware. 326 As I understand his evidence, Mr Bewsher accepted that what he identified as the possible tail-out drain was unlikely to have appeared on topographic mapping available when Sydney Water prepared its designs for the water main. He also accepted that if the culvert outlet was overgrown and if the tail-out drain had become overgrown, silted up or filled in by earthworks by the time the water main was constructed, it would not have been visible to the Water Board's surveyors and designers. However he "expected" that they would have identified the culvert once construction commenced. He observed that "the failure of the contractors to lower the utilities adjacent to the culvert outlet removed the opportunity for the culvert to ever be free-draining again". 327 Mr Bewsher opined that: "31 … If water did reach the tree roots in any quantity it is probable that this occurred by the water main trench." (emphasis added) 328 Mr Bewsher was asked about this passage in his report in cross-examination by Mr Torrington for Sydney Water: "Q. Have you chosen that word deliberately, 'if'? A. Yes … I can't be sure. Q. You can't be sure. Is essentially what you're saying … that a possible way for water to get to the north is along the sand bed? A. It's a possible way, the sand is a much more efficient hydraulic conduit than the surrounding clay. Q. But the short answer is you don't know whether it did or didn't from your investigations. A. No. Q. When you say 'no', you agree with me you don't know if it reached the tree roots? A. I don't know if it reached the tree roots ." (emphasis added) 329 He also agreed that it would have been a reasonable expectation, looking at the water main laid beside the road as at 1981, that water from it would flow along the sand base, north or south from the culvert if it existed at that time. 330 In cross-examination by Mr Toomey, Mr Bewsher acknowledged that the question whether water reached the tree roots was outside his knowledge. He accepted that it would be expected that the water would flow in the direction of the grade of the water main trench, but that he did not know what the grade was.

(b) Mr Burn 331 Mr Burn, a consultant engineer, provided a number of reports and gave evidence on behalf of Mrs Turano. Beazley JA has largely summarised that evidence (at [49] ff). 332 I would add the following to her Honour's summary. 333 Mr Burn reviewed a variety of reports, including Mr Lumsdaine's. In commenting on Mr Clark's observation that there were no table drains in the vicinity of the outlet pit, Mr Burn observed that that was probably due to "a low level of maintenance having occurred that permitted the table drains to silt up and become overgrown". 334 Mr Burn disagreed with Mr Clark's opinion that construction of the water main and the Telstra service did not block the discharge from the outlet pit, a conclusion based on Mr Clark's opinion that there was no outlet channel from it. In Mr Burn's opinion, any object higher than the invert of a pipe outlet had the effect of disrupting water flow from it. 335 Mr Burn examined drainage flow calculation sheets produced by the Council indicating the water flow at the culvert during rainfall events and concluded they showed "comparatively minor flow". He referred to bore hole investigations undertaken by the Council in June 1999 which showed the ground to be moist at all depths below the road pavement and observed that the moisture was in the tree roots zone. He concluded that the groundwater level in the area was "not excessively deep", being just below road pavement level. It might be noted that Beazley JA refers (at [58]) to Mr Burn's evidence that without a detailed scientific analysis, it would be difficult to know precisely the water levels in the clay beside the road. No such scientific analysis was carried out. This is significant because, as Mr Watson acknowledged, and Mr Toomey did not dispute, there was no evidence the water main was leaking. Nor, no doubt having regard to the fact the water main trench was laid in the impermeable clay which was the naturally occurring geological material, was there any suggestion the water main contributed to the groundwater level in the area. 336 Mr Burn's conclusion in his first report was that clearing the vegetation and sediment in the outlet pit would minimise water retention and maximise soil strength. 337 In his second report dated 1 August 2005, Mr Burn observed that photographs taken after the culvert was extended in 2002 showed no sediment or vegetation in it. However when he visited the area in 2005 he observed it to be overgrown and concluded that overgrowth had occurred during "only about three years". He opined, at that stage, that it was the build-up of sediment and vegetation growth which was the "primary reason" the culvert did not adequately discharge water passing under Edmondson Avenue. He concluded that the effect of the three years' growth was to clog and obscure the table drain and tail-out which had been constructed in association with the culvert extension work. He opined that as the pond at the culvert could only dissipate by evaporation or seepage into the ground, the high moisture content in the surrounding ground was maintained. 338 Mr Burn's third report commented on Mr Castor's and Mr Clark's reports. He accepted that the question whether the tree which fell was able to cope with "water logged soil" required advice from a horticulture/aboriculture expert. Referring to Mr Clark's proposition in his report of 24 January 2005 that there was no defined tail-out at the time the water main was installed, Mr Burn commented, again, that the lack of a tail-out when Mr Clark viewed the site was likely to have been due to lack of maintenance. 339 In his fourth report dated February 2006, Mr Burn referred to two surveys, one undertaken in 1999, the other in 2005. The first he identified as showing the outlet from the culvert "though not ideal, [was] clearly evident". It might be noted that the 1999 plan on which he appears to have based that comment is annotated "clean and clear existing tail-out of vegetation and debris", no doubt being the source of Mr Burn's "not ideal" comment. Another 1999 plan was annotated at the same point "no tail-out presently, possible tail-out??". 340 Mr Burn observed that the 2005 survey showed the outlet without any clearly defined outflow path. In his opinion that was due to vegetation growth and sediment build-up. 341 Mr Burn concluded that the flat grades in the area meant the outlet from the pipe culvert needed to be kept free of obstruction to minimise water build-up leading to soil saturation. In his opinion the growth of vegetation in the pipe outlet area blocked water outflow. He also concluded that the presence of the utility mains exacerbated the retention of water by raising the dam wall across the pipe outlet discharge path. In his opinion, observation at the time the utilities were laid should have led to them being laid below the pipe culvert outlet level "despite the less than favourable outlet discharge conditions". As I understand his report, had that been done the obstructions caused by the blockage of the tail-out drain and, possibly, the overgrowth in the pipe culvert would still have obstructed water discharge flows. 342 In his oral evidence, Mr Burn expressed the opinion that the water main would tend to block more flow from the culvert than the Telstra main. Mr Burn did not analyse the extent to which water would drain along the water main trench. 343 Mr Burn had not carried out any investigations to determine whether a tail-out drain had been constructed on the western side of the culvert. However he had concluded from reading Mr Bewsher's report that a tail-out drain existed "some time in connection with this culvert". 344 The Council relied on Mr Burn's evidence in cross-examination by Mr Watson. To the extent it is relevant, that evidence demonstrated:

(a) that drainage water will find the line of least resistance to drain through;

(b) that a probable and foreseeable effect of the installation of the water main in its sand bed was to create a drain running north-south within the highly permeable clay;

(c) that water would enter the sand until the sand itself had become fully saturated;

(d) that the second potential effect of the installation of the water main was to block the tail-out drain;

(e) blocking the tail-out drain would have the effect of impeding water dissipating through the culvert;

(f) the creation of the north-south drain could be "ameliorated" by creating a drain to permit further westward movement of the water drained through the culvert which would have meant working in private land outside the road reserve; Mr Burn also agreed that Sydney Water "can" work in private land but there was no evidence as to what this would have entailed;

(g) Sydney Water could also have put the main under the bed of the culvert, which would have been "a better way"; and (h) laying the water main in the position where it was laid, it was probable, and foreseeable, that the tail-out drain would become blocked. 345 Mr Watson also asked a number of questions related to the possibility that Sydney Water had damaged the tree roots when they installed the water main. I do not understand that that assertion forms any part of either the Council or Mrs Turano's cases against Sydney Water. 346 Mr Watson asked Mr Burn some questions about the effect of water on an Australian gum. Mr Burn's answer appeared to be based on environmental studies he had undertaken which showed that flooded trees died because they were not used to moisture. Based on that observation he opined "I'd expect trees which are in areas wetter than their normal environment they would have problems surviving". 347 Mr Watson then asked Mr Burn to assume that a pathogen was transmitted to the tree via its roots because of the unnatural moisture to which the roots were exposed, then put: "Q. If that was established in accordance with just ordinary civil engineering principles that is which fit into the ideas of drainage, that was a danger created by the person responsible for installing the water main, don't you agree? A. Yes." (emphasis added) 348 In cross-examination by Mr Torrington, Mr Burn agreed that it could reasonably have been anticipated at the time the water main was laid, that there was a height difference of about 100mm (apparently between the top of the water main and the bottom of the culvert), but that the sand around the water main "would act as a dispersal medium for water coming out of the culvert". He also said that the distance water travelled along the sand bed depended upon the grade of the ground, as I understand, in which the water main was laid. As I have said, there was no evidence of the grade of the water main trench beyond the culvert outlet pit. He said that if the water main had been laid deeper, the same method would have been used with the same potential for water to pass along the sand of the water main trench.

(c) Mr Keirnan 349 Mr Keirnan, a civil engineer, prepared a report on the instruction of Mrs Turano's solicitors. He inspected the site in February 2002. He observed the drainage culvert pipe outlet to be blocked with sediment and the surrounding area to be saturated with storm water which had entered the culvert inlet but which, in his view, could not drain away because of the blocked pipe and the lack of a tail-out drain. He opined that the likely reasons for the pipe culvert blockage were the lack of a defined outlet for the culvert stormwater discharge and a lack of routine maintenance to the drainage culvert and the table drain "at the low point". He observed that the general appearance indicated the area had not "been given maintenance attention for some considerable time". He concluded that the likely reason for the tree falling was the lack of an adequate drainage outlet and the lack of routine maintenance. He was not required for cross-examination.

(d) Mr Clark 350 Mr Clark, a civil engineer, prepared three reports on behalf of Sydney Water. He was not required for cross-examination. He inspected the site on 19 January 2005. At the time he inspected the culvert both its inlet and outlet pipes were dry and there was no free water in the drainage system. He considered that although both the water main and the Telstra service were at a higher level than the culvert inlet, they did not block the discharge from the culvert pipe as there was no outlet channel from it in any event. He concluded that the blockage to the water discharge was caused by the natural ground in which the services were laid. 351 Mr Clark concluded from a data sheet for the water main Sydney Water provided to Mrs Turano's solicitors showing the water main was constructed in a "standard depth environment", that that indicated it was laid in a trench. He observed the depth of the ground over the main at his inspection to be consistent with the standard depth he would expect for a water main in this location. He also observed that the Work as Executed Plan dealt specifically with the construction of water mains where watercourses were encountered, those being between Sixth and Seventh Avenues, but that no watercourse had been recorded at the accident site or at all between Eighth and Ninth Avenues. 352 Mr Clark inferred from a note which appeared on the plans referring to the "proposed invert of creek" that inquiries had been made about the future development of the watercourses and steps had been taken to provide for that development. He also inferred that the absence of any special measures for laying the water main indicated either the culvert did not exist at the time it was laid, or that there were no proposals to provide an outlet drain for it. 353 Mr Clark disagreed with Mr Meys' conclusion that the blocked drain caused the apparent flooding around the tree, preferring the interpretation that the water seen was the result of surface flows of rainwater running down the shoulder and being trapped by the hole. 354 Mr Clark concluded that the culvert had been constructed as a drowned culvert, that is to say, one whose invert was below the proposed ground level and with no outlet channel. The use of such a culvert was, in his opinion, not unusual for culverts under shallow embankments in rural areas (the description he would have given to the area in 1961) which did not have to provide for a permanent drainage flow. Such a culvert operated first by accumulating water from the level of the pipe invert to the natural ground surface, allowing the water to seep into the surrounding ground with some additional dispersal by evaporation when wet periods ceased. He concluded that the ponding of water at the outlet was due to the absence of an outlet drain. 355 It appears Mr Clark and Mr Burn were at issue as to how the culvert was originally constructed, that is to say, with or without a tail-out drain. For reasons to which I will come, in my view, the disagreement is not of ultimate significance. 356 Mr Clark concluded that the major cause of the tree falling was the combination of strong winds and its poor root system. While he accepted that the ground around the tree may have been "softened" by water, on his view the water which led to that effect may have been from rainwater rather than culvert flooding. 357 Mr Clark was asked to comment on Mr Lumsdaine and Mr Bewsher's reports. He emphasised that Mr Lumsdaine's investigations had only led him to conclude that there was a "possible" tail-out drain. He also interpreted Mr Lumsdaine's report as stating that any tail-out drain which may have been constructed did not extend to the outlet pit. He appears to have formed this opinion because Mr Lumsdaine did not identify the filling at the culvert outlet pit he had relied upon in identifying the possible tail-out drain further west. Mr Clark also noted that on the levels Mr Lumsdaine deduced for the presumed tail-out drain it only partially drained the culvert, causing it to be a partly drowned culvert. Mr Clark did not consider Mr Lumsdaine's investigations showed that a tail-out drain to provide free draining of the culvert had ever been constructed. Rather, in his view, they demonstrated that the natural ground in which the water main was installed existed between the culvert outlet and the presumed tail-out drain and extended for the full length of the culvert pipe. This reinforced his view that the culvert was a "drowned culvert". 358 Mr Clark disagreed with Mr Bewsher's interpretation of the aerial photographs, relying, in part on Mr Lumsdaine's observations that any tail-out drain only needed to extend about 10 metres into the adjacent land to provide free drainage. On that basis, he concluded that Mr Bewsher's observation that the aerial photographs showed a watercourse, more likely a dug drain extending 40 metres west, to be unlikely. He thought it more likely the feature Mr Bewsher described was the bottom of a depression formed by the opposing slopes of the natural ground similar to a waterway on the eastern side of the road rather than a dug drain. 359 In my view Mr Clark's opinion in this respect is to be preferred to Mr Bewsher's. It accords with Mr Lumsdaine's field investigations which appears to be a better interpretative technique in this respect than observations gleaned from an aerial photograph.

Consideration: breach of duty 360 In order for Mrs Turano to succeed against Sydney Water, she had to adduce evidence supporting a positive inference implying negligence on its part, an inference which arose as an affirmative conclusion from the evidence and one established to the reasonable satisfaction of a judicial mind. The evidence had to rise above the level of conjecture, could not be based on possibilities but had to be established as a matter of probability, and had to do more than give rise to conflicting inferences of equal degrees of probability: Bradshaw v McEwans Pty Ltd (1951) 217 ALR 1 (at 5); Luxton v Vines [1952] HCA 19; (1952) 85 CLR 352 (at 359 - 360) per Dixon, Fullagar and Kitto JJ; Jones v Dunkel [1959] HCA 8; (1959) 101 CLR 298 (at 304 - 305) per Dixon CJ. 361 Assuming Sydney Water owed Mrs Turano a duty of care, a proposition it did not dispute, in my view she did not establish that Sydney Water breached that duty of care or that, if it did, that breach was a cause of the tree falling. 362 The question of breach is to be determined in accordance with s 5B of the Civil Liability Act. Section 5B(1) reflects the common law requirement that a court identify what a reasonable person in the position of the defendant would do by way of response to the reasonably foreseeable risk: Waverley Council v Ferreira [2005] NSWCA 418; (2005) Aust Torts Reports ¶81-818 (at [27]) per Ipp JA, Spigelman CJ and Tobias JA agreeing. Section 5B(2) in substance reiterates Mason J's remarks in Wyong Shire Council v Shirt [1980] HCA 12; (1980) 146 CLR 40; Waverley Council (at [45]). 363 The question of breach determined in accordance with Shirt involves determining whether a reasonable person in the defendant's position would have foreseen that his or her conduct involved a risk of injury to the plaintiff or to a class of persons including the plaintiff. This involves a judgment about reasonableness: New South Wales v Fahy [2007] HCA 20; (2007) 81 ALJR 1021 (at [6]) per Gleeson CJ; (at [62]) per Gummow and Hayne JJ. 364 Further, the Shirt inquiry requires the Court to look forward to identify what a reasonable person would have done, not backward to identify what would have avoided the injury: Fahy (at [57]) per Gummow and Hayne JJ. As their Honours said, referring to Vairy v Wyong Shire Council [2005] HCA 62; (2005) 223 CLR 422 (at [124]), and explaining why it is wrong to focus exclusively upon the way in which the particular injury of which a plaintiff complains came about: "[T]he apparent precision of investigations into what happened to the particular plaintiff must not be permitted to obscure the nature of the questions that are presented in connection with the inquiry into breach of duty. In particular, the examination of the causes of an accident that has happened cannot be equated with the examination that is to be undertaken when asking whether there was a breach of a duty of care which was a cause of the plaintiff's injuries. The inquiry into the causes of an accident is wholly retrospective. It seeks to identify what happened and why. The inquiry into breach, although made after the accident, must attempt to answer what response a reasonable person, confronted with a foreseeable risk of injury, would have made to that risk. And one of the possible answers to that inquiry must be 'nothing'." 365 The inquiry as to breach, accordingly depends first, on determining what Sydney Water's perception would have been of the effect of installing the water main through the outlet pit in 1981. This involves determining the probable state of the outlet pit in 1981. 366 I agree with Beazley JA (at [134]) that the primary judge's conclusion that the tail-out drain had in fact become dysfunctional due to poor maintenance systems was open to his Honour. 367 However neither the primary judge nor Beazley JA has considered whether or not the tail-out drain had become dysfunctional by the time the water main was laid. In my view the evidence establishes that that had, more probably than not, occurred by 1981. This conclusion emerges from the evidence to which I have referred, the critical parts of which I repeat. 368 Mr Bewsher's evidence established that, assuming his analysis of the aerial photographs was correct, the feature which appeared to be a tail-out drain in 1970 was already less distinct by 1978. I have already said that Mr Clark's conclusion that Mr Bewsher erred in this respect having regard to Mr Lumsdaine's evidence is to be preferred. However, assuming for present purposes that Mr Bewsher's analysis was correct, his observation of a less distinct feature in 1978 is consistent with the proposition that whatever had flown through the tail-out drain originally was doing so at a lesser rate. This is consistent with the tail-out drain either ceasing to function or its function being progressively diminished. 369 Even if a tail-out drain had been dug when the outlet pit was constructed it is probable, in my view, that it had ceased to function and was unobservable by 1981. 370 Mr Burn, Mr Bewsher, Mr Keirnan and Mr Clark's evidence established that the culvert outlet pit was probably blocked by sediment and root growth within a few years of its construction. Having regard to the Council's reactive policy in relation to culvert maintenance, which it can be inferred had been in operation throughout the relevant period, it is probable that neither the outlet pit or the tail-out drain had been cleared. 371 It will be recalled that the culvert was choked with sediment and vegetation at about the time of the accident and that no tail-out drain was apparent. Mr Burn opined that this was probably due to "a low level of maintenance having occurred that permitted the table drains to silt up and become overgrown". 372 The relative speed with which the outlet pit and the tail-out drain could become clogged and obscured, is apparent from Mr Burn's evidence that by 2005 after the outlet pit was cleared in about 2002, it had become clogged by overgrowth to the point of obscuring both the table drain and tail-out which had been constructed. There is no reason to suppose the rate of vegetation growth in 2002 - 2005 was any more rapid than it was in the period before the water main was laid. 373 Mr Lumsdaine's evidence establishes that it is probable a tail-out drain was dug at the time the outlet pit was constructed, but disappeared because it had been back-filled by persons unknown at an unspecified time. In my view, it is an available inference that that had been done prior to the 1978 photographs being taken. That, too, would support the proposition that by 1981 when the water main was laid, any evidence of a tail-out drain may long since have disappeared. 374 Accordingly I would conclude that the tail-out drain had ceased to exist by 1981, either because of the Council's reactive maintenance policy leading to vegetation clogging and obscuring it or because it had been deliberately back-filled. 375 Mr Lumsdaine said that if there had been no tail-out drain then laying the water main would not have made the ponding in the outlet pit any deeper than it already was, "except possibly for a minor raising of the level of the lowest overflow point, if this was previously located along the western side of the pit" and, the construction of the water main probably provided an outlet drain to the pit. Although, as I shall explain, whether or not the water main trench actually operated in that manner was not proved, nevertheless it is reasonable to conclude that that possible consequence would have been apparent to a reasonable person in Sydney Water's position in 1981. In other words, a reasonable person in Sydney Water's position in 1981 would not have concluded that laying the water main in the outlet pit dammed any existing drainage outlet, or dammed the outlet pit to any relevant extent. 376 Mr Lumsdaine also said that the construction of the water main probably "inadvertently" provided an outlet drain to the pit. Thus even if there had been a tail-out drain present which laying the water main may have partly dammed, a reasonable person in Sydney Water's position would have foreseen that the water main trench provided another exit route for water. 377 It is convenient at this stage to deal with Beazley JA's conclusion (at [54]) that Mr Burn's evidence that the growth of vegetation in the pipe outlet blocked the water outflow did not survive his cross-examination and with her conclusion (at [126]) that "any obstruction was due to the water main". I disagree, with respect, with this conclusion. In my view, it does not take into account the evidence to which I have referred concerning vegetation clogging the outlet pit and tail-out drain. Further, with respect, it appears to be inconsistent with her Honour's conclusion (at [134]) that the tail-out drain had become dysfunctional due to poor maintenance systems. 378 Mr Burn did give some confusing evidence about the culvert outlet pit. He expressed the view in cross-examination that the "excavated space down stream from the culvert head wall has the shape of a pit, but … that was only there because of the way the ground's been blocked and the fact that water cannot really discharge". He said it appeared someone had dug out the outlet area of the pit by digging a hole. Although he first opined this had been done after the installation of the Telstra pit and water main, he later said he did not know when that might have happened. However, his evidence was clarified, in my view, by his acceptance that the original culvert design would have had a tail-out area (which he called the "discharge area"), and compared with that, the fact that such an area no longer existed meant whenever the outlet culvert was cleared, whoever did so would leave a hole "similar in shape to a pit". 379 Mr Burn assumed the culvert pit would be cleaned out because it would become clogged by silt flowing into the culvert under the road. He accepted he had not seen any document from the Council to the effect that the culvert clogged. The absence of such a document in my view does not support the proposition that the culvert did not clog. There was, as I have said, ample evidence of this. 380 The better explanation for the absence of any Council recording that the culvert was clogged, lies in its policy of reacting only to complaints or concerns raised about culverts. Mr Brookfield gave evidence of this policy and, too, of the fact that an examination of the Council's records indicated that no complaint or concern had been recorded in respect of the culvert in question. That does not mean, in my view, that it was not clogged. Indeed the fact it was clogged by vegetation may well have meant that no-one in the rural environment in which it was constructed, was even aware a culvert existed there. Such an observer may well have concluded that any ponding observed in its vicinity was due to it being at a "general low point in the landscape" (Castor). 381 In my view, therefore, the primary judge's finding (at [123]) that the culvert became blocked with debris over the forty years since it was constructed was correct. 382 The next point to consider on the question of breach is whether it would have been foreseeable to Sydney Water that laying the water main could pose a risk of injury to the class of persons who included Mrs Turano, against which it should take precautions. Both Mrs Turano and the Council's contentions in this respect depended on the proposition that laying the water main exposed the tree's roots to water-logging from the water main trench and, consequently, to the risk of such damage as would destabilise them. 383 In my view the evidence did not establish this was a possible consequence of laying the water main. 384 The first point is that, as I explain in more detail when I deal with causation, Mrs Turano did not establish that the water passing along the water main trench could accumulate in the vicinity of the tree's roots in a manner which might cause them to become water-logged. It is sufficient to note at this point that, in my view, she did not establish that a foreseeable consequence of Sydney Water laying the water main as it did was to cause a risk of injury to the tree's root system. 385 Next it is necessary to bear in mind some matters of measurement. The water main trench, including its sand surrounds, appears to have been 350mm in diameter. The tree's roots, assuming it to have been approximately the height it was at the time it fell (there was no evidence about its likely dimensions in 1981), most probably spread at least 48 metres around it (three times the minimum height estimated by Mr Castor), with most of its feeder roots being within a ten metre range of the trunk. There was no suggestion the water main leaked, or that water from it travelled beyond the bed of sand in which it was presumed to have been laid. There was no evidence, in my view, which supported the proposition that Sydney Water ought reasonably to have foreseen that constructing a water main of those comparatively narrow dimensions in the vicinity of the tree (assuming for present purposes knowledge of the tree's probable root structure could be attributed to Sydney Water) could expose the tree's roots to the risk of water-logging at all, let alone in a manner which might compromise the whole tree. 386 In this respect it is relevant to recall Mr Castor's evidence that not all trees are safe, not all tree failure is predictable, and not all tree failure can be explained after the event. That is the view of an arborist. It is not reasonable in my view to impute an arborist's knowledge to an engineer, nor did I understand that to be a submission. This of itself is a significant concession and, in my view, is sufficient to conclude the issue of foreseeability adversely to Mrs Turano. 387 However even if an arborist's knowledge was imputed to Sydney Water as at 1981, it demonstrates the difficulty of predicting even at the comparatively low threshold Shirt contemplates what might cause a tree to fall, let alone whether putting a narrow water main buried at standard depth along a nature strip might expose the tree to the risk to which I have referred. Laying the water main 388 Both Mrs Turano and Sydney Water relied, on the issue of breach on the proposition that, assuming Sydney Water identified a risk in laying the water main through the outlet pit to the class of persons who included Mrs Turano's family, it could have avoided that risk by laying the water main differently. Consideration of alternative conduct once relevant risk is identified turns on "what a reasonable man would do by way of response to the risk": Fahy (at [58], emphasis in original) per Gummow and Hayne JJ. Further, assuming, contrary to the conclusion I have expressed, that Sydney Water had identified a possible risk to the tree's roots which it could have avoided by installing the water main differently, that does not of itself give rise to or affect its liability for the way in which the water main was laid: s 5C(b), Civil Liability Act. 389 The first point to consider is the Council and Mrs Turano's complaint about the primary judge's conclusion that the water main was laid in a manner which was consistent with good practice at the time. 390 Both pointed to documents apparently produced on subpoena by Sydney Water, the first of which (exhibit G) entitled "Culvert Crossing Concrete Encased Pipe Water Main up to 600mm diameter" apparently dated to mid-1981. It depicts four diagrams. The first, "Sectional elevation A-A", appears to indicate a water main being installed underneath a three pipe culvert. The second is a plan view entitled "Typical Culvert Crossing". The third, "Section C-C" refers to another plan and is not self-explanatory. The fourth depicts an "Alternative End Treatment". 391 Exhibit G was tendered on the fourth day of the trial after all parties' cases had closed and Mrs Turano's counsel had indicated, the previous day, that she had no case in reply. It was not shown to any witness during the trial. To the extent that it demonstrates that it was possible to construct a water main underneath a culvert, it might be noted that there was already evidence, given by Mr Burn, that Sydney Water had put the water main under the bed of a culvert next down from that implicated in the accident. There was no evidence as to why Sydney Water treated that culvert differently from the implicated culvert. 392 Mrs Turano also relied on exhibit H, a document entitled "Water Main Laying Practice Standard Depth" the original of which was said to have been dated 11 August 1994, 13 years after the water main was laid. The relevance of this document was not explained. 393 Next, Mrs Turano relied upon passages in Mr Burn's reports. To the extent that they appeared to be relevant, I note that he referred to:

(a) the Sydney Water Board Works as Executed Plan for laying a 300mm diameter water main showing the use of vertical bends to "lay main at 'Standard Depth' below creek bed between Sixth and Seventh Avenues, but [which] note nothing in relation to [the] culvert between Eighth and Ninth Avenues"; as to this, Sydney Water's expert, Mr Clark, apparently referring to the same documents, observed that the water main was laid at a standard depth according to the data sheets for the water main which Sydney Water had provided to Mrs Turano's solicitors and, too, that the Works as Executed Plan dealt specifically with the construction of water mains where water courses were encountered, those being between Sixth and Seventh Avenues, but that no water course had been recorded at the accident site or at all between Eighth and Ninth Avenues;

(b) Mr Burn also observed that Sydney Water had taken measures to ensure the water main did not impact on the water flow of creeks to the south of the subject culvert, but did not appear to have made provision to ensure the water main when laid did not impact on culvert water discharge flows. This appears to be consistent with the Works as Executed Plan and, no doubt, the fact that there was no water flow at the outlet pit;

(c) Mr Burn also opined that the presence of the utility mains (the water main and the Telstra pit) laid at higher levels than the pipe invert on the culvert "further exacerbated the retention of water by raising the 'dam wall' across the pipe outlet discharge path". As I earlier noted, in his view reasonable observations at the time of laying the utilities should have resulted in them being laid at a level below the pipe culvert outlet level "despite the less than favourable outlet discharge conditions" and that had this been done "the only obstruction to water discharge flows would have been the less than ideal downstream conditions not the 'dam' caused by the utility mains". This depended, in my view, on the tail-out drain being evident at the time the water main was laid. I have concluded it was not.

The "Works as Executed Plan" was not tendered. 394 Mrs Turano submitted that had Sydney Water adopted the method of laying the water main at a level which would have avoided it blocking water dissipating from the culvert, water from the culvert would not have flowed via the sand around the water main to the roots of the grey box tree. She argued that laying the water main immediately adjacent to the culvert so that it blocked its draining was not good practice, the whole point of having the culvert, she contended, being to enable water to be dispersed over a wide area. 395 Mrs Turano also relied on Mr Burn's oral evidence as demonstrating it was foreseeable that if the culvert was dammed by the water main "something such as occurred would happen". The passage of the transcript referred to supported the proposition that blocking the discharge path of the culvert would mean water would pond in the culvert after the next rain storm. 396 Mrs Turano submitted that the primary judge had ignored Mr Burn's evidence, drawing attention to his Honour's reference (at [69]) to Mr Burn's evidence that whether or not the water pipe could have been laid deeper, the same method would have been used no matter at what level it was laid. I do not understand this to have ignored Mr Burn's evidence. As I understand it, it relates to the proposition that the method of laying the water main in a trench with a sand backfill was standard main-laying practice and his evidence in cross-examination by counsel for Sydney Water that the water main would have been laid in the same manner even if laid deeper. 397 The Council also relied upon Mr Burn's evidence that it was the probable consequence of laying the water main through the culvert pit that it would create a flow of water to the north or south of that pit, that this effect could have been ameliorated by taking other precautions, that Sydney Water could have put the main under the bed of the culvert as it had done at the next culvert, that it defied good practice to excavate close to the roots of an existing tree and that the introduction of an unnatural water source in the area of an Australian gum created a potential danger. The Council also submitted that Mr Burn had given evidence that the installation of the water main created the "specific danger of the introduction of pathogens to the tree". It needs hardly be said that this was not an area within Mr Burn's expertise. 398 In considering this submission it is necessary to return briefly to Mr Lumsdaine's evidence. On my interpretation of his report, the water main trench, to the extent it contained saturated water, was below the base of the outlet pit. Thus, as I understand Sydney Water's submission and the primary judge's finding, even if the water main had been laid below the level of the outlet pit it could still have drained water from the surrounding soil which, depending upon the composition of the sand and the grade of the trench may have drained towards the tree, or, indeed any of the large number of trees growing alongside Edmonson Avenue in 1981. 399 Even if a risk to the tree had been identified, in my view, the evidence did not support the proposition that laying the water main below the culvert could, or would, have avoided the possibility of water travelling along the sand bed. Mr Burn agreed that would happen if the water main was laid deeper. In other words it was an inevitable consequence of the practice of laying water mains in sand beds, a practice no party criticised, that the sand trench would become the conduit for water absorbed from the surrounding strata. 400 Further for the reasons I have already developed, it was not in my view foreseeable to a reasonable person in Sydney Water's position that its conduct in laying water mains in what was clearly its standard manner throughout, at least, rural or semi-rural areas such as Edmonson Avenue was in 1981, was such as to require it to take any different approach to that exercise. Consideration: causation 401 If my conclusions on breach be incorrect, I would also conclude that Mrs Turano did not establish that a cause of the tree falling was compromise to its roots caused by water flowing along the water main trench. 402 Section 5E of the Civil Liability Act affirms that it is the plaintiff who always bears the onus of proving, on the balance of probabilities, any fact relevant to the issue of causation. 403 In Jones v Dunkel (at 305), after referring to Bradshaw, Dixon CJ added: "But the law which this passage attempts to explain does not authorise a court to choose between guesses, where the possibilities are not unlimited, on the ground that one guess seems more likely than another or the others. The facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied ." (emphasis added) 404 Accordingly, the issue of causation could not be established on the balance of probabilities by evidence which failed to do more than establish a possibility: see Seltsam Pty Limited v McGuiness [2000] NSWCA 29; (2000) 49 NSWLR 262 (at [80] ff) per Spigelman CJ, (at [201]) per Stein JA. 405 In Seltsam (at [87]), Spigelman CJ referred with approval to Lord Wright's "frequently cited" statement in Caswell v Powell Duffryn Associated Collieries Ltd [1940] AC 152 (at 169-170): "Inference must be carefully distinguished from conjecture or speculation . There can be no inference unless there are objective facts from which to infer the other facts which it is sought to establish . In some cases the other facts can be inferred with as much practical certainty, as if they had been actually observed. In other cases the inference does not go beyond reasonable probability. But if there are no positive proved facts from which the inference can be made, the method of inference fails and what is left is mere speculation or conjecture ." (emphasis added) 406 In my view the evidence that water from the water main trench reached the tree's roots, let alone did so in a manner which would cause the intermittent water logging Mr Castor identified as the probable cause of the diseased root system did not, as the primary judge found (at [103]), albeit without explanation, rise above the level of the conjecture or speculation to which Lord Wright referred. 407 There were, in my view, no evidence of the objective facts from which that conclusion could be inferred. Conditional opinions were expressed as to whether water at the tree's roots could have come from the water main trench but the factual assumptions on which those opinions were based were not established. In particular, there was no evidence to establish:

(a) that water from the culvert reached the tree's roots via the water main trench; or (b) that if water from the culvert did reach the tree's roots via the water main trench, it accumulated there in a volume sufficient to cause damage to the tree or its roots. 408 The highest point the evidence reached on this point was:

(a) Mr Burn, Mr Bewsher and Mr Lumsdaine, the three witnesses who opined on the matter, said that water could flow in the base of the water main trench, but that the direction of flow depended on the profile and grade of the base of the trench; and

(b) the extent to which the water main trench would act as a drain, i.e. whether water did flow in it, depended on whether the sand backfill observed in the pit was typical of the trench beyond the outlet pit (Mr Lumsdaine).

No investigations were undertaken of the water main trench between the outlet pit and the probable location of the tree, four metres to its north, so there was no evidence of either (a) or (b). It was not suggested there was any practical reason why this could not be done. 409 I disagree, with respect, with Beazley JA's conclusion (at [85] - [86]) that Mr Lumsdaine and Mr Bewsher's evidence supported the proposition that water ran along the base of the water main trench and, because the water main had dammed the culvert, allowed water more readily to reach the tree's root system more readily than if the water main had not been installed. Mr Bewsher did not know from his investigations, which I take it included those of Mr Lumsdaine, the geotechnical expert he had commissioned to explore the geology of the area, whether or not water from the outlet pit reached the tree's roots. 410 Further, as I read Mr Lumsdaine's evidence while it was a theoretical possibility, that water could reach the roots of the tree, whether it did in fact depended on the relative permeability of the sand and the clay and the grade of the water main trench. Accordingly to convert this theory into a probability, it was necessary to establish the ability of the water main trench beyond the outlet pit both to act as a drain, and so act in a direction which could cause water logging around the tree's roots. This was never done. In the absence of such investigations, there was no factual basis to support Mr Lumsdaine's opinion that it was "almost certain" the route water took to the tree's roots would be via the water main trench. 411 This deals both with the Council's appeal against the findings in favour of Sydney Water, Mrs Turano's appeal on that point and her Notice of Contention.

Case against the Council 412 While, as I have said, I agree with Beazley JA's conclusion that the Council's appeal should be allowed, I do so for different reasons. 413 First, I disagree, with respect, with her Honour's conclusion (at [126]) that Mr Burn effectively conceded that any obstruction was caused by the water main, and not the general clogging of the outlet pit and the tail-out drain caused by the Council's policy of not maintaining such areas. For the reasons I have given I do not accept he made such a concession. Further, there was abundant evidence that the Council's non-maintenance policy more probably than not led to the tail-out drain having ceased to function by 1981. 414 I would also disagree, with respect, with her Honour's conclusion (at [127]) that the Council did not know, nor ought to have known, that at the time of construction the culvert and its tail-out drain was likely to become clogged or obstructed. In my view the Council would, or ought to, have been aware when it constructed the outlet pit and the probable tail-out drain, that that was the inevitable consequence of that policy. In other words, failure to maintain the outlet pit and the tail-out drain would ensure the outlet pit ceased to be free draining most probably long before Sydney Water's arrival. 415 It follows that I also disagree, with respect, with her Honour's conclusion (at [128]) that there was no evidence that the culvert was likely to become clogged or that the tail-out drain was likely to cease to operate. 416 I do accept, however, her Honour's conclusion (at [128]) that even when dammed because of clogging and/or the absence of a tail-out drain, the evidence established that water would probably pond in, and overflow, the outlet pit. The extent of overflow in the sense of causing relevant water-logging affecting the tree's roots was not established, just as the passage of water along the water main trench was not established. 417 Accordingly, in my view the Council did owe Mrs Turano a duty of care, but the evidence did not establish it breached that duty. I also agree with Beazley JA's conclusions in relation to ss 42 and 45 of the Civil Liability Act. 418 I have read the responses to my judgment written by Beazley and Hodgson JJA respectively. It is, regrettably, necessary to deal with some of the propositions upon which they rely. In so doing, it is important to bear in mind that the Council, and Mrs Turano's, cases against Sydney Water were that the tree's roots were adversely affected by water in the water main trench. Having said this I would, with respect, observe:

(a) As to Beazley JA's conclusion (at [228]) agreeing with Hodgson JA that it was a reasonably available inference that Sydney Water compacted the tail-out drain, Mr Lumsdaine, the geotechnical engineer who identified the compacted drain, did not reach that conclusion nor did Mr Bewsher, the engineer who commissioned Mr Lumsdaine's work. If that was an inference open from the nature of the work Sydney Water undertook, it might be expected either Mrs Turano or the Council would have proved it. Further, no party advanced the proposition that it was Sydney Water which compacted the tail-out drain.

(b) Beazley JA accepts (at [231]) that Mr Lumsdaine's evidence was critical to establishing that water reached the root ball of the tree through the sand. Her Honour concludes that that conclusion was a matter of "clear inference drawn from established facts". Mr Lumsdaine's own report demonstrated that the question of the performance of water in the water main trench beyond the culvert depended upon the composition of the sand in that area, the fall of the trench and the profile of the base of the water main trench beyond the culvert. Mr Lumsdaine did not undertake any investigations of these matters. Beazley JA concludes that the grade of the water main trench could be inferred from the grade of the road. However, this was not an inference drawn by either Mr Lumsdaine or Mr Bewsher, again the experts one would expect to do so, if available. Indeed, the latter said, as I have earlier observed, that he did not know whether or not water moved along the sand bed to the tree roots.

(c) I would also observe, in this respect, that the failure of counsel for the Council or Mrs Turano to ask either Mr Lumsdaine or Mr Bewsher (or any other expert witness) whether an inference could be drawn about the grade of the water main trench means their contentions that water did reach the tree roots via that route may more readily be rejected: Payne v Parker [1976] 1 NSWLR 191 (at 201).

(d) Finally, persuasive as Hodgson JA's diagrammatic representation of the radius of the tree's roots and their probable interaction with the water main trench may be, that diagram does not address the fundamental propositions I have identified as to the failure of the evidence to deal with the composition of the soil in the water main trench beyond the culvert or the direction in which water might have passed along that trench. 419 I agree with orders (1) - (3), (5) (in part), (7) and (9) of the orders Beazley JA proposes. Insofar as Sydney Water is concerned I would not set aside orders (2) and (3) of the primary judge's orders. I would order:

(2) Ground 2 of the Council's appeal be dismissed. **********

08/12/2009 - Amendment to citation - Paragraph(s) 364

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.