Rodriguez & Sons Pty Ltd v Queensland Bulk Water Supply Authority trading as Seqwater

Background

1. Somerset Dam has four regulator valves, eight sluice gates and eight crest gates. The regulator valves and sluice gates allow for a discharge of water from towards the bottom levels of the dam which has an elevation level of 60.96m Australian Height Datum ("AHD"). The spillway fixed crest level which is at the bottom of the crest gates has an elevation level of 100.45m AHD. The point of the dam adjacent to the top of the crest gates is known as the crest level. It has an elevation level of 107.46m AHD. The top of Somerset Dam's deck has an elevation level of 112.34m AHD.
2. Section 9.2 of the Manual specifies that once a flood event is declared all of Somerset Dam's crest gates are to be fully opened and all of the sluice gates and regulator valves are to be closed. Various parts of the Manual specify flood strategies for Somerset Dam that, at least in some circumstances, required the crest gates to remain open during a flood event. The defendants contend that they can, or at least should, never be opened during a flood event. When the crest gates are opened then uncontrolled discharge will occur for water stored above 100.45m AHD subject to whatever control can be exercised by releases from the sluice gates and regulator valves.
3. In describing flood event Strategy S2 for Somerset Dam, the Manual states, inter alia (at p 41): "The levels of 109.70m AHD and 80.00m AHD represent the likely failure level for Somerset Dam and the level at the top of the Wivenhoe Dam Wave Wall respectively. Note that the failure level of 109.70m AHD for Somerset Dam assumes all radial [ie, crest] gates are fully open and this failure level will be reduced if this cannot be achieved."
4. The fact that the crest level is 107.46m AHD but that the likely failure level of Somerset Dam is 109.70m AHD reflects the fact that the Manual contemplates the possibility of Somerset Dam being "overtopped", that is its water level being above 107.45m AHD. In that regard, section 10.2 of the Manual states (at p 44): "Somerset Dam should not be overtopped by flood water, but if Wivenhoe Dam is threatened by overtopping, the release of water from Somerset Dam is to be reduced at the risk of overtopping Somerset Dam in order to prevent the overtopping of Wivenhoe Dam".
5. I have described the scope of the evidence of the plaintiff's expert on dam operations, Dr Ronald Christensen, in previous judgments: see Rodriguez & Sons Pty Ltd v Queensland Bulk Water Supply Authority trading as Seqwater (No 9) [2017] NSWSC 1116; Rodriguez & Sons Pty Ltd v Queensland Bulk Water Supply Authority trading as Seqwater (No 10) [2018] NSWSC 149. Dr Christensen has produced a number of simulations which form the basis for the actions that the plaintiff contends a reasonably competent flood engineer should have undertaken in conducting flood operations at Wivenhoe Dam and Somerset Dam during the period 2 January to 12 January 2011. The simulations adopt different start dates for flood operations and have different operating assumptions which are unnecessary to describe.
6. There are numerous aspects of Dr Christensen's simulations that are contentious but two in particular are of present relevance. First, two of the simulations assume that under some circumstances the crest gates at Somerset Dam can be closed during a flood event, being Simulation I and Simulation J. In Simulation I, the water level in Somerset Dam rises to 106.95m AHD. [8] In Simulation J the water level in Somerset Dam rises to 106.93m AHD. [9] Second, the remainder of the simulations do not close the crest gates at Somerset Dam however they take the water level in Somerset Dam higher than the level it actually reached in January 2011 namely 105.11m AHD. For example, both Simulation F and Simulation H take the water level in Somerset Dam to a height of 106.25m AHD. [10]

First Category of Objection to Mr Dann's evidence: the PRA 2013 and the URS 2014

1. Mr Dann's letter of instruction asked him to agree or disagree with certain statements made in Dr Christensen's reports concerning the operation of Somerset Dam and then to address the following questions: [11] "…From a dam safety engineering perspective, what, if any, are the risks if the level of water in Lake Somerset exceeds 107.46m AHD …From a dam safety engineering perspective, what, if any, are the risks in adopting an approach to the operation of Somerset Dam which requires that some or all of the eight radial gates be operated (i.e. closed and/or opened) at a time when the level of Somerset Lake is exceeding 100.45m AHD? "
2. Mr Dann was instructed that he was entitled to consider information and material obtained as part of his previous retainers with Seqwater, which thus included the PRA 2103 and the URS 2014 provided he clearly identify the information relied on. He was also asked to state whether or not his knowledge of any of the matters covered in his report had been identified before the January 2011 flood event.
3. In his report Mr Dann describes, amongst other matters, the process of revised risk assessment of Somerset Dam that was undertaken as part of PRA 2013 and the URS 2014. As part of this a new potential "failure mode" was identified namely an "assessed increased erosion potential of the rock foundations when flows start to pass over the non-overflow sections of the dam at EL107.46." [12] Mr Dann relevantly concluded: "Since the original risk assessment studies at Somerset and Wivenhoe Dams were undertaken in 2000 the key failure modes identified have substantially remained the same for each dam. The level of engineering assessment and analysis has generally increased during each subsequent phase of risk assessment and this has influenced the assessment of the annual failure probability of the failure modes as well as the assessment of the consequences of dam failure. However, some level of uncertainty still remains for some of the key failure modes and this is the purpose of adopting a risk framework to manage these uncertainties. The outcomes of the 2014 Dam Safety Review (URS 2014) are a good example that shows how one key failure mode - erosion of the Somerset Dam abutments under dam overtopping flows - can change substantially based on the level of engineering assessment undertaken. A key conclusion from URS 2014 was that there is no safe level of overtopping of Somerset Dam and this was a substantial change from the previous assessments, reflected in the Flood Mitigation Manual that assessed 2.2m of overtopping of Somerset Dam as the likely dam failure level." [13] (emphasis added)
4. Thus, the effect of the studies was in effect a revision of what the Manual contemplated namely a potential to operate Somerset Dam by overtopping it at 107.46m AHD but not allowing it to exceed 109.70m AHD. This conclusion appears to have been in part based on a further engineering assessment and in part based upon a revised approach to risk analysis.
5. Seqwater did not contend that this aspect of Mr Dann's evidence was relevant to any causation analysis of Dr Christensen's simulations. [14] As noted, this material only suggests that Somerset Dam should not be overtopped and Dr Christensen does not overtop Somerset Dam in any of his simulations. In other words, this material does not support a suggestion that, had any of the flood engineers adopted Dr Christensen's approach, then Somerset Dam would, or even might, have collapsed.
6. Instead, Senior Counsel for Seqwater, Mr Pomerenke QC, contended that this material addressed the question of breach of duty. He contended that the evidence suggested that in January 2011 the flood engineers acted on the basis that there was a level of risk in operating Somerset Dam above 107.46m AHD, and below that if the crest gates were closed, that was not properly recognised by the Manual or by Dr Christensen in his simulations and this material served to demonstrate that the engineers' belief was "real and not fanciful". [15] If accepted this might tend to negate any suggestion that a reasonably competent flood engineer would take Somerset Dam as high as Dr Christensen did in his simulations.
7. Mr Owens SC contended that none of the engineers who had given evidence prior to 14 June 2018 had expressed a belief in the terms submitted by Mr Pomerenke QC. However, even if they had, Mr Owens SC submitted that the reasonableness of any belief held by any of the flood engineers in January 2011 concerning the maximum height at which Somerset Dam could be operated with or without its crest gates open could "only be determined by reference to the factual material that was available [to them] at the time". [16]
8. In circumstances where the evidence of only two of the four flood engineers was complete and where it appears that much of their evidence is contested, I determined this objection on the assumption that Mr Pomerenke QC correctly characterised the belief that at least one flood engineer held about the risks of operating Somerset Dam at elevated levels with and without the crest gates closed. However, as noted, this material is only relevant to an assessment of breach and such an inquiry is prospective in that it "seek[s] to identify what the response would have been by a person looking forward at the prospect of the risk of injury" (Vairy v Wyong Shire Council [2005] HCA 62; 223 CLR 422 at [126] per Hayne J). An assessment of whether any actions or omissions of the flood engineers in January 2011 informed by any belief they held about the maximum operating level of Somerset Dam breached any duty of care they owed must be undertaken prospectively and not by reference to matters they were unaware of and could not reasonably be expected to have been aware of at that time. The PRA 2013 and URS 2014 were such matters. The same applies to any assessment of Dr Christensen's evidence and simulations.
9. It was for that reason that I rejected so much of Mr Dann's report that address the risks associated with the operation of Somerset Dam by reference to studies and assessments undertaken after the flood event in January 2011.

Third Category of Objection to Mr Dann's Evidence: Precautionary Releases

1. In section 8.3 of his report Mr Dann stated: [17] "8.3 Precautionary Releases Dr. Christensen's report presents a number of alternative scenarios for managing flood operations during the January 2011 events. These scenarios include 'precautionary releases' from both Wivenhoe and Somerset Dams to lower the reservoir levels in the dams below Full Supply Level on the basis of forecast rainfall and expected high catchment runoff due to previous rainfall that had wetted the catchment. The term 'precautionary release' is not part of the approved Flood Mitigation Manual. In fact a key objective of the Flood Mitigation Manual is to retain the storage at Full Supply Level at the conclusion of a Flood Event. In my view a formal revision of the Flood Mitigation Manual would be needed to include a basis for making releases from the dams prior to a flood event, for this option to be available to the Flood Operation Engineers. A revision to the Flood Mitigation Manual would also need to be approved under the Water Supply Act 2008."
2. There is no doubt that an assessment of the obligations imposed on the flood engineers by the Manual is a hotly contested issue in these proceedings. To give evidence on that topic it must be demonstrated that Mr Dann had "specialised knowledge based on his training, study or experience" and that the opinions he expresses are "wholly or substantially based on that knowledge" (see Dasreef Pty Ltd v Hawchar (2011) 243 CLR 588; [2011] HCA 21 at [35]). It may be that a civil engineer in Mr Dann's position is capable of bringing expertise to bear upon a determination of what the Manual required of the flood engineers. However, the absence of any substantive reasoning in Mr Dann's report to support the above conclusions left me unsatisfied that the conclusions he expressed were "wholly or substantially based on that knowledge".
3. Accordingly I rejected the above passage in Mr Dann's report (and its equivalents elsewhere).

Endnotes

1. EXP.SEQ.003.0113.
2. SEQ.004.036.7950; SEQ.004.036.8522.
3. SEQ.004.036.7520.
4. QLD.001.001.0146.
5. T 7242.32.
6. T 7246.15.
7. T 7252.23.
8. EXP.ROD.015.0461 at 0995.
9. EXP.ROD.015.0461 at 1069.
10. EXP.ROD.015.0461 at 0855 and 0940.
11. EXP.SEQ.003.0113 at 0183.
12. EXP.SEQ.003.0113 at 0142.
13. EXP.SEQ.003. 0113 at 0144.
14. see T 7221.
15. T 7221.35.
16. T 7234.18.
17. EXP.SEQ.003.0113 at 0155. 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. Decision last updated: 04 July 2018