[2016] NSWCCA 37
Director of Public Prosecutions (Vic) v Iliopoulos (No. 2) [2016] VSC 47
Hodgson v Amcor Limited
Amcor Limited v Barnes (No. 3) [2011] VSC 272
Jones v Dunkel (1959) 101 CLR 298
[1959] HCA 8
Papakosmas v The Queen (1999) 196 CLR 297
Source
Original judgment source is linked above.
Catchwords
(1996) 136 ALR 627
Bulga Underground Operations Pty Ltd v Nash (2016) 93 NSWLR 338[2016] NSWCCA 37
Director of Public Prosecutions (Vic) v Iliopoulos (No. 2) [2016] VSC 47
Hodgson v Amcor LimitedAmcor Limited v Barnes (No. 3) [2011] VSC 272
Jones v Dunkel (1959) 101 CLR 298[1959] HCA 8
Papakosmas v The Queen (1999) 196 CLR 297
Judgment (59 paragraphs)
[1]
Judgment
HOEBEN CJ at CL: I agree with Bellew J and the orders which he proposes.
BELLEW J:
[2]
INTRODUCTION
By amended summonses filed in the Land and Environment Court of New South Wales on 1 February 2019, WaterNSW (the respondent) sought orders against Peter James Harris and Jane Maree Harris (the appellants) that they each appear before a judge of that Court to answer a charge contrary to s 91G(2) of the Water Management Act 2000 (NSW) (the WMA). The charge against each of the appellants was in the following terms: [1]
[B]etween 22 June 2016 and 27 June 2016, at the property known as Beemery Farm located at 6104 Kamilaro Highway, Brewarrina, 2839 consisting in total of 4 lots within the Parish of Stonehenge, County of Clyde, being Lots 1/751597, 5/1147705, 2/1147705 and 3/1147705 in the State of New South Wales (Beemery Farm), the [appellant] committed an offence against s 91G(2) of the Water Management Act 2000 in that a term or condition of an approval of which (s)he was a co-holder was contravened by a person.
The particulars of the charge brought against each of the appellants were as follows:
Approval: [The appellants] are the landholders and occupiers of Beemery Farm and the co-holders of:
• Water Access Licence 33692 [reference 85AL753235] (WAL 33692); and
• Water Supply Works and Water Use Approval 85CA753236 (the Approval).
Manner of contravention: Water was taken in contravention of a term or condition of the Approval in that:
(a) It was a term or condition of the Approval that water is prohibited from being taken when the flow in the Darling River at the Bourke gauge is equal to or less than 4,894 ML/day.
(b) Approximately 3147 ML of water was taken in the 2015/2016 water reporting year by means of the water supply works nominated under the Approval and situated on the landholders' land.
(c) In the period 22 June 2016 to 27 June 2016, water was taken at times when the flow in the Darling River at the Bourke gauge was less than 4,894 ML/day.
Each of the appellants pleaded not guilty to that offence.
On 19 March 2020, Robson J found each offence proved and convicted each of the appellants. [2]
The appellants now appeal against their convictions on the grounds more fully set out below.
[3]
THE FACTUAL BACKGROUND
Much of the factual background to the alleged offending was not in dispute. In setting out that background, I have indicated the areas where there were issues between the parties.
[4]
The Barwon-Darling river system
The Barwon-Darling River System is an unregulated river system, or in other words, a river system without major storage facilities or dams. [3] It flows from Mungindi on the New South Wales/Queensland border, to Wentworth in south-western New South Wales, and extends downstream to the Menindee Lakes. The system includes the Barwon River (the river), extending from upstream at Mungindi at the confluence of the Macintyre and Weir Rivers, to where it meets the Culgoa River, at which point the river channel becomes the Darling River. [4]
[5]
Beemery Farm
Beemery Farm (the property) is a parcel of agricultural land located in the Brewarrina Shire in north-western New South Wales. The Kamilaroi Highway runs east/west through the property, which is bordered to the north by the undulations of the river. Cotton growing fields are located within three lots of the property located on the south side of the highway. [5]
From about the mid-1990s until 2014, the property was occupied by Clyde Agriculture Pty Ltd (Clyde). [6] Since July 2014 the appellants, through Clyde Cotton (a partnership comprising the appellants individually, and PJ & JM Harris Pty Limited) have occupied the property [7] and have operated it as a cotton farm. [8]
[6]
The issue of a water licence to Clyde
Until 4 October 2012 (at which time the WMA became operative), [9] the Water Act 1912 (NSW) (the 1912 Act) governed the protection and management of water sources in NSW. That management included the issue of licences and approvals to the occupiers of land regarding the use of water. Sections 10 and 12 of the 1912 Act allowed an occupier of land to make an application for a licence in relation to "any work to which this Part extends". Such work was defined to include (inter alia) work affecting the quantity of water flowing in, to, or from, a river or a lake. [10]
Pursuant to the 1912 Act, Clyde was granted licence number 85SL105059 (the 1912 licence) [11] in respect of the property. [12] The permitted purpose of that licence was irrigation, and the licensed works included the use of a number of centrifugal pumps. [13] The effect of the 1912 licence was to allow the irrigation of the property by those pumps extracting water from the river, through an irrigation channel, to two above ground water storage reservoirs. The water from the reservoirs was then directed to that portion of the property which contains the cotton fields. [14]
Section 17A(2) of the 1912 Act conferred a power on the relevant statutory authority to (inter alia) modify a licence. On 20 February 2012, Clyde was notified of a variation in conditions (5) and (6) of the 1912 licence which, in their varied form, were in the following terms: [15]
(5) SUBJECT TO CONDITION (6) THE LICENSED WORK SHALL NOT BE USED FOR THE PURPOSE OF IRRIGATING THE LICENSED AREA OR ANY PART THEREOF UNLESS THE DISCHARGE OF THE DARLING RIVER AT THE BOURKE GUAGE [sic] EXCEEDS 4894 MEGALITRES PER DAY.
(6) THE LICENSED/AUTHORISED WORK SHALL NOT BE USED FOR THE PURPOSE OF IRRIGATION UNLESS THE FOLLOWING FLOW CONDITIONS EXIST:-
THE BARWON RIVER FLOW EXCEEDS 760 MEGALITRES PER DAY AT THE CULGOA RIVER JUNCTION (UPSTREAM) GAUGE AND 840 MEGALITRES PER DAY AT BREWARRINA GAUGE.
[7]
The water sharing plan
On 4 October 2012, the Water Sharing Plan for the Barwon-Darling Unregulated Alluvial Water Sources 2012 (NSW) (the 2012 plan) commenced. Clause 4 provided that the 2012 plan applied to (inter alia) the Barwon-Darling Unregulated River Water Source. There is no dispute that this water source applied to the river, and thus to water which was drawn from the river at the property during the period of the charges. [16]
On the same day, the Water Management (Application of Act to Certain Water Sources) Proclamation (No 2) 2012 (NSW) (the Proclamation) became operative. Clause 3 of the Proclamation declared that Pts 2 and 3 of Ch 3 of the WMA applied to each "prescribed water sharing plan". This included the 2012 plan.
Since 4 October 2012, the WMA has made provision for the issue of water access licences (pursuant to Part 2 of Chapter 3) and approvals (pursuant to Part 3 of Chapter 3).
The Proclamation enlivened the operation of Sch 10 of the WMA, cl 3 of which specifically provided that any entitlement that was in force under the 1912 Act immediately prior to the appointed day (i.e. 4 October 2012) was taken to have been replaced. The 1912 licence constituted an "entitlement" having regard to the definition of that term in cl 2(a) of Sch 10 of the WMA. Cl 3 was partly in the following terms:
3. Access licences and approvals arising from former entitlements
(1) Subject to this Schedule, an entitlement that, immediately before the appointed day, was in force under the 1912 Act, the 1948 Act or the 1994 Act is taken to have been replaced:
(a) to the extent to which it entitles any person or body to take a specified quantity of water, by an access licence held by that person or body (subject to such of the conditions of the entitlement as are applicable to an access licence):
(i) for the quantity of water so specified, or
(ii) if the relevant management planned, and regulations made for the purposes of this paragraph, indicate that a different quantity of water calculated in accordance with a specified methodology may be taken under an access licence issued in relation to the water management area or water source to which the management plan applies - for a different quantity of water calculated in accordance with that methodology, and
(b) to the extent to which it entitles any person or body to use a specified water management work, by a water management work approval held by that person or body in respect of that work (subject to such of the conditions of the entitlement as are applicable to an approval of that kind), and
(c) to the extent to which it entitles any person or body to use water on any land, buy a water use approval held by that person or body in respect of that land (subject to such of the conditions of the entitlement as are applicable to an approval of that kind), and
(d) to the extent to which it entitles any person or body to carry out a specified activity, by an activity approval held by that person or body in respect of that activity (subject to such of the conditions of the entitlement as are applicable to an approval of that kind).
[8]
The issue of a replacement licence and approval to Clyde
By operation of cl 3(1) of Sch 10 to the WMA, the 1912 licence was replaced and separated into two instruments namely: [17]
1. a water access licence, number 85AL753235 (the 235 licence); and
2. a water management work and water use approval 85CA753236 (the approval).
Clause 20(2) of Sch 10 provided:
(2) a replacement access licence or approval:
(a) is to include any mandatory conditions that are required to be imposed on the licence or approval, and
(b) may include such other conditions ("discretionary conditions") as the Minister thinks fit, including (but not limited to) conditions relating to the protection of the environment.
[9]
The notification to Clyde of the 235 licence and the approval
Clause 20(1) of Sch 10 of the WMA required the Minister to give written notice of the terms of any access licence or approval arising by operation of the schedule.
On 17 October 2012 Clyde was advised that upon the commencement of the 2012 plan, the 1912 licence had been replaced with: [18]
1. the 235 licence; and
2. the approval.
The correspondence said nothing about the conditions attaching to either, but included the following: [19]
Following consultation on its development, I am writing to inform you that the Water Sharing Plan for the Barwon Darling Unregulated & Alluvial Water Source (the Plan) commenced on 4 October 2012.
Water sharing plans and water access licences issued under the Water Management Act 2000 establish clear and secure rights for access to water by water users. Water access licences are held separate from land and have an independent title.
Approvals are different to access licences. Approvals give holders the right to construct and operate water supply works and to use water on land. Approvals continue to be tied to the land.
...
The commencement of the Plan means that your former entitlement/s 85SL105059 [20] issued under the Water Act 1912 has now been replaced with the following access licence/s and approval/s:
Water Access Licence: 85AL753235
Approval/s: 85CA753236
...
A notice containing the details of the water access licence/s and approval/s and conditions will be mailed to all licence and approval holders. Until then, licence and approval holders continue to be the subject to the conditions of their former licence that apply to either an access licence or an approval.
[10]
The transfer of the 235 licence and the approval to the appellants
On 4 July 2014, the appellants' solicitors wrote to the respondent enclosing a copy of a Notice of Sale which confirmed that the property had been transferred to the appellants on 12 June 2014. [21] The correspondence went on to state: [22]
Please note that all further correspondence in relation to their Licences which have been acquired as part of this transaction should be forwarded to the following:-
Mr & Mrs PJ Harris
PO Box 20
BOURKE NSW 2840
That letter attached a schedule setting out the various licences which had been purchased by the appellants from Clyde as part of the sale. That schedule included a reference to both the 235 licence and the approval. [23]
On 11 September 2014, the respondent wrote to the first appellant stating: [24]
You have recently become the holder of the above licence or approval resulting from your purchase of a water access licence (WAL) or the purchase of lands benefitted by approvals.
Licences and approvals issued under the Water Management Act 2000 operate under a set of terms and conditions which are imposed and regulated by NSW Office of Water. Licence conditions define when and how much water you can extract. Approval conditions authorise water supply works - and the circumstances and location of where the water can be used.
We have attached information relevant to the above matter as indicated by the following tick boxes.
☒ We have attached a copy of your Water Act 1912 licence.
☒ The NSW Office of Water is currently finalising the conversion process with reference to approvals and conditions for the BARWON DARLING UNREGULATED AND ALLUVIAL WATER SOURCES. As soon as this process is finalised, a notice containing details of the WAL and/or approval/s conditions that apply will be mailed to you. Until then, licence and approval holders continue to be subject to the conditions of the former licence 85SL105059.
Further information on licences and approvals under the Water Management Act 2000 can be obtained from the Office of Water's website go to: www.water.nsw.gov.au - Water licensing.
Please direct any questions regarding this correspondence to Water Advisory Services on 1800 353 104.
The reference in that correspondence to "former licence 85SL105059" was a reference to the 1912 licence.
[11]
Notification to the appellants of the conditions attached to the approval
Richard Wheatley, a Senior Water Regulation Officer, gave evidence that shortly after the appellants had purchased the property, he had a conversation with the first appellant to discuss a number of matters regarding licences and approvals. [25] Mr Wheatley gave evidence that the following was said: [26]
WHEATLEY: A guy called Mark Adams has made a request in relation to your and [the second appellant's] licences and approvals. Is this something you were aware of and are you happy for Mark Adams to lodge papers, make requests and deal with your licences and approvals?
FIRST APPELLANT: Thanks Richard, yes that's fine. Mark is [sic] employee of ours who helps out in the office. On behalf of [the second appellant], Budvalt and I, I give consent and authority for Mark to deal with all of our licences and approvals. Mark has authority to send and receive documents on our behalf by email and post.
WHEATLEY: Thanks [first appellant]. I will let the Dubbo team know.
Mr Wheatley also said that, consistent with the terms of this conversation, he sent an email to Mr Adams on 23 September 2015 attaching the approval. [27]
The email identified by Mr Wheatley was not in evidence. However, Mr Wheatley produced what he described as a "copy of a screenshot from the WLS entry for Approval 85CA753236". [28] "WLS" was a reference to the respondent's Water Licencing System database which contained (inter alia) details of licence approvals issued by the respondent from time to time. [29]
The screenshot was headed "Email history for document ID 258164" and was dated "23-Wed-2015". [30] The email message was in the following terms: [31]
Please find attached an updated copy of the Statement of Approval for 85CA753236. If you have any questions, please reply to this email.
Mr Wheatley's evidence was that an automated response was received from Mr Adams' email address. [32] It contained a reference to the approval and was in (inter alia) the following terms: [33]
Please be advised that I am on leave from Thursday 10/09/15 returning Wednesday 29/09/15.
The response went on to provide alternative contact details in the event of "anything urgent".
In light of the evidence of Mr Wheatley as to the automated response, the trial judge found that the reference in the screenshot to "23-Wed-2015" was in fact a reference to 23 September 2015. [34] His Honour accepted that Mr Wheatley had sent the email to Mr Adams on that day.
The terms of the approval which Mr Wheatley identified as having been sent to Mr Adams included the following: [35]
Schedule 2 - Water supply works
Part A: Authorised water supply works
Subject to the conditions of this approval, in relation to each numbered work in the table, the holders of this approval are authorised to construct and use a water supply network of the type shown at the location specified:
Work 1
Specified work - 660MM CENTRIFUGAL PUMP x 4
Specified location - 1//751597 Whole Lot
Water Management zone (if applicable) - BREWARRINA to CULGOA RIVER JUNCTION MANAGEMENT ZONE
Water source - BARWON DARLING UNREGULATED RIVER WATER SOURCE
Water sharing plan - BARWON DARLING UNREGULATED AND ALLUVIAL WATER SOURCES
Work 2
Specified work - 610MM CENTRIFUGAL PUMP
Specified location - 1//751597 Whole Lot
Water Management zone (if applicable) - BREWARRINA TO CULGOA RIVER JUNCTION MANAGEMENT ZONE
Water source - BARWON DARLING UNREGULATED RIVER WATER SOURCE
Water sharing plan - BARWON DARLING UNREGULATED AND ALLUVIAL WATER SOURCES
Schedule 3 - Water Use
Subject to the conditions of this approval, the holder(s) of this approval is authorised to use water for the following purpose(s) and location(s):
Purpose 1
Specified purpose - IRRIGATION
Specified location - 1//751597
5//1147705
2//1147705
3//1147705
Schedule 4 - Conditions
The approval is subject to the following conditions:
Plan conditions
Water sharing plan - Baron [sic] Darling Unregulated and Alluvial Water Sources
Take of water
MW0655-00001 - Any water supply work authorised by this approval must take water in compliance with the conditions of the access licence under which water is being taken.
MW1916-00001 - Water must not be taken when the flow of the Darling River at the Bourke gauge (425003) is equal to or less than 4,894 ML/day.
Water management works
MW0491-00001 - When a water supply work authorised by this approval is to be abandoned or replaced, the approval holder must contact DPI Water in writing to verify whether the work must be decommissioned.
The work is to be decommissioned, unless the approval holder receives notice from the Minister not to do so.
Within sixty (60) days of decommissioning, the approval holder must notify DPI Water in writing that the work has been decommissioned.
Monitoring and recording
MW0484-00001 - a logbook must be kept and maintained at the authorised work site or on the property for each water supply work authorised by this approval, unless the work is metered and fitted with a data logger.
MW2338-00001 - The completed logbook must be retained for five (5) years from the last date recorded in the log book.
MW0484-00001 - Before water is taken through the water supply work authorised by this approval, confirmation must be recorded in the logbook that cease to take conditions do not apply and water may be taken.
The method of confirming that water may be taken, such as visual inspection, internet search, must also be recorded in the logbook.
If water may be taken, the:
A. date, and
B. time of the confirmation, and
C. flow rate or water level at the reference point in the water source
must be recorded in the logbook.
MW2337-00001 - The following information must be recorded in the logbook for each period of time that water is taken:
A. date, volume of water, start and end time when water was taken as well as the pump capacity per unit of time, and
B. the access license number under which the water is taken, and
C. the approval number under which the water is taken, and
D. the volume of water taken for domestic consumption and/or stock watering.
MW0482-00001 - Where a water meter is installed on a water supply work authorised by this approval, the meter reading must be recorded in the logbook before taking water. This reading must be recorded every time water is to be taken.
Reporting
MW0051-00001 - Once the approval holder becomes aware of a breach of any condition on this approval, the approval holder must notify the Minister as soon as practicable. The Minister must be notified by:
A. email - ;
or
B. telephone: 1800 353 104. Any notification by telephone must also be confirmed in writing within seven (7) business days of the telephone call.
Other conditions
Use of water
DK1542-00001 - The approved work must not be used for the purpose of irrigation unless the following flow conditions exist: the Barwon River flow exceeds 760 ML/day at the Culgoa River Junction (upstream) gauge and 840 ML/day at Brewarrina gauge.
Water management works
DK1642-00001 - Native vegetation may be cleared only to the minimum extent necessary for the construction and maintenance of the authorised works, that is, the minimum area of clearing to permit access for appropriate mechanical implements to maintain the works. Clearing of land for irrigation must be in accordance with the provisions of the Native Vegetation Conservation Act 1997 or the Native Vegetation Act 2003.
DK1215-00001 - The holder of the approval must not construct or install works used for conveying, distributing or storing water taken by means of the approved work that obstruct the reasonable passage of floodwaters into or from a river.
Additional conditions
DK1362-00001 - The approval holder must not allow any tailwater or drainage water to discharge, by any means including surface or sub-surface drains or pipes, from the approval holders property, into or onto:
- any adjoining public or [C]rown road
- any other persons land
- any [C]rown land
- any river, creek or watercourse
- any groundwater aquifer
- any area of native vegetation as described in the Native Vegetation Conservation Act 1997 or the Native Vegetation Act 2003
- any wetlands of environmental significance
- any identified site of [A]boriginal significance
- any identified site of cultural significance.
Glossary - Cease to take - Cease to take conditions means any condition on this approval, or on the access licence under which water is proposed to be taken, that prohibits the taking of water in a particular circumstance
……
General notes
All conditions on an approval require compliance. An appeal to the Land and Environment Court against a decision to impose certain conditions on an approval can be made within 28 days after the date the decision is made. Conditions identified with the first letter "D" are those that can be appealed during the appeal period.
The words in this approval have the same meaning as in the Water Management Act 2000
Note: The words in this approval have the same meaning as in the WMA.
There was a significant issue before the trial judge arising from Mr Wheatley's evidence concerning his conversation with the first appellant. Whilst the first appellant did not give evidence, his position was that the conversation did not take place.
Mr Wheatley was cross-examined at length, in the course of which he was asked: [36]
Q. Did you call him or did he call you?
A. I believe I would have called him in relation to some applications on another couple of properties on the Macquarie I think it was - -
Q. You say you believe you would have called him. You don't actually remember who called who?
A. No.
Q. Although you swore on affidavit saying you telephoned [the first appellant], the truth is he may [have] called you?
A. Yeah.
Q. Is that right?
A. It's very likely that I called [the first appellant].
Q. That's speculation, isn't it? You don't actually remember.
A. I don't remember [the first appellant] actually ringing me on any occasion.
Q. You say you discussed a number of matters in relation to licences and approvals. What matters were they?
A. Some applications that were afoot for some properties on the Macquarie, Lochinvar and Wombulian(?) and - -
Q. What were the matters?
A. There were some applications that were lodged.
Q. Yes?
A. We were discussing advertising of them, there were a number of objections, I needed some plans and designs. Things of that nature.
Q. You say Mark Adams made a request to you at this time.
A. Mark Adams had been in contact with one of my staff I believe and it was in that context that I asked [the first appellant] who he was or if it was appropriate that he be making enquiries into approval or licences.
Q. Did you talk to Mr Adams at this time?
A. Not at that point, no.
Q. What was your understanding of the matters that he raised? What was the request he raised at the time?
A. I think he was actually lodging some dealings or something saying they needed copies or approval numbers and access licence numbers and things.
Q. You think but you're not sure?
A. I wasn't the one dealing with the applications. I was - made the enquiry as to whether it was appropriate for him to be asking questions of the Harris - -
Q. But you can't rule it out?
A. - - or Budvalt licences.
Q. I'm sorry. You can't remember now the actual request was, can you?
A. I believe it was about dealing. I - to which property or which licence? No.
The cross-examination specifically addressed Mr Wheatley's recollection: [37]
Q. You don't have a clear recollection of the conversation you had with [the first appellant] in 2014, do you?
A. The one that I am saying may have happened plus or minus five - you know - six or eight months before the purchase of the properties?
Q. Yes, that one.
A. Yes, No, No I didn't - don't have any - any diary notes or anything to that conversation. I said it was just part of a discussion.
Q. You don't have a clear recollection of precisely what it was that [the first appellant] said to you during this conversation, do you?
A. He said that Mark Adams worked for him and that it was - -
Q. I'm asking you a very specific question. You don't have a clear recollection as you said in the witness box today, four and half years later as to precisely what [the first appellant] said to you during this conversation, do you?
A. He said words to the - yes, I do - words to the effect of - -
….
Q. You don't have a clear recollection as to the precise words that [the first appellant] used in this conversation, do you?
A. Not a verbatim - -
Q. No.
A. But I have a very good understanding of what he said.
Q. What you're trying to do is to reconstruct the conversation based on your general understanding or recollection of what was said at the time. Correct?
A. Not reconstruct but - -
Q. So Mr Adams had made a request and then you were contacting [the first appellant], weren't you? To ask whether you could deal with Mr Adams in relation to that request. Correct?
A. As I told you, I was contacting [the first appellant] in regards to other matters and I raised it in the - in the conversation.
Q. And the conversation didn't extend beyond that, did it? Namely you asking [the first appellant] whether you could deal with Mr Adams in relation to the request that he had raised. Correct?
A. It didn't extend beyond that.
Q. The sequence of events was Mr Adams had made a request, you wanted to find out from [the first appellant] whether you could deal with Mr Adams in relation to that request and you asked [the first appellant] whether you could do so. Correct?
A. That was part of it. Yes.
Q. That was the extent of it.
A. I was calling [the first appellant] on the same day that we got the enquiry. Conversely, my colleague could have emailed [the first appellant] but I was - you know - going to talk to him anyway, it seemed appropriate to mention it.
Mr Wheatley was then taken to his account of the terms used by the first appellant and was asked: [38]
Q. - - - Now, you don't seriously suggest you remember [the first appellant] saying those words, do you?
A. Words the effect, yeah.
Q. That's not how [the first appellant] speaks, is it, in your experience.
A. I've had plenty of conversations with [the first appellant], yeah.
Q. He doesn't speak in this way, does he? "On behalf of [the second appellant], Budvalt and I, I give consent and authority for Mark to deal with all of our licences and approvals." That doesn't sound like the way he would talk, does it?
A. It's in the context of the way that I pitched the question, yes, probably was.
Q. You can't actually remember the words that [the first appellant] actually used, can you?
A. Words to that effect, yes.
Q. That was your understanding, was it, of what he said?
A. Understanding and interpretation of the words that he said, yeah.
Q. You might have misinterpreted or misunderstood what he said.
A. Certainly not.
Q. You can't remember what he actually said, how can you discount that as a possibility?
A. Like I said, words to that effect.
Q. You said that was your understanding or interpretation of what he said. You may have been wrong about that. Correct?
A. Well, in the ensuing months, Mr Adams lodged several papers on behalf of [the first appellant] that [the first appellant] had signed so I don't think there was any misunderstanding or misinterpretation.
Q. But you're not able to form a view about whether you misunderstood or misinterpreted what [the first appellant] said because you don't actually remember the precise words he used, do you?
A. If I had to quote the words verbatim, no.
Q. You may have misunderstood or misinterpreted what he actually said. Correct?
A. No, I don't - don't agree.
It will be evident from those passages that part of the thrust of the cross-examination was that Mr Wheatley did not have a clear recollection of the precise words that the first appellant was said to have used in the conversation. Whilst that proposition was put, in varying forms, on a number of occasions, Mr Wheatley's consistent position was that whilst he could not remember the precise words, he could certainly remember the effect of what had been said. [39] By way of example, he was asked: [40]
Q. Mr Wheatley, you do not remember [the first appellant] actually getting to the level of detail of talking about whether correspondence could be sent by email and post, do you? You don't actually remember that.
A. Words to that effect were used. This is the words that I have put in my affidavit and I'll stand by them.
The cross-examination then continued on the same theme: [41]
Q. You don't actually remember as you sit here now, four and a half years later, that [the first appellant] in this conversation got right down to the level of detail where he was talking to you about the modes of communication that could be employed including email and post?
….
Q. You don't remember that as you sit there now, do you?
A. And that's why I've used the words, "words to that effect," and haven't put him in quotes to, you know, quote him as saying - -
Q. Because you can't remember? Correct?
A. Can't remember him saying words to this effect? I certainly do. That's why I've written it.
Q. But not about whether it could be email or post. That level of detail doesn't come back to you now, four and a half years later, does it?
A. If I had to dwell on it, certainly may not have used the word "post". But we were using "email".
Q. "May not have used the word 'post' but may have used 'email'". And your basis for saying that is the fact that email was being used at the time. Is that right?
A. Well, it adds weight to my recollection. Yes.
The cross-examination culminated in the following: [42]
Q. You didn't have such a conversation with [the first appellant], did you?
A. I did.
Q. If you had such a conversation, you would have notified persons of the kind that generate these documents on behalf of the department. Correct?
A. Not the extensions and things, no, it's not. And had I of notified them, they still may not have changed it. They would of just used the customer ID that was in the system of which [the first appellant] holds several licences at that one mailing address.
[12]
Gauging stations operated by the respondent
Andrew Cutler has been employed by the respondent as a Hydrometric Coordinator since 2009. In that capacity, his responsibilities have included the management of hydrometric field staff who monitor surface water and groundwater flows. Prior to 2009, Mr Cutler held the position of a supervising Hydrometric Officer which involved performing fieldwork, including recording river gauge readings. [43] Mr Cutler swore three affidavits which formed part of the evidence before the trial judge. It is important to emphasise that his evidence was not the subject of any substantive challenge.
Mr Cutler's evidence established that the respondent owns, operates and maintains a network of gauging stations along the river. Each gauging station is located on the top of the river bank, and houses a number of different instruments which measure the river height level at 15 minute intervals. The instruments include: [44]
1. a series of gauge lines within the river;
2. a pressure line to the river;
3. a pressure sensor;
4. a logger;
5. a modem;
6. a transmission antenna; and
7. a power source (which is generally solar).
In explaining how the instruments operate at a gauging station, and the nature of the data which is collected as a result, Mr Cutler said the following: [45]
[11] Gauging stations measure and record river heights at their individual locations along the river. Their purpose is to contribute to the production of river flow data, which is the quantity of water flowing at a particular point of the river at a given point in time. The data collected by gauging stations over the last few minutes or hours is referred to as 'real time or near real-time' river height data or raw height data. Real time height data collected by gauging stations is transmitted to a central data program known as Hydrotel. Based on the river height information at each point in time, real-time flow data is automatically generated based on 'rating/calibration tables' (Rating Tables)….
Mr Cutler expanded on that process as follows: [46]
[15] …… Real-time river height data is automatically collected by gauging stations and transmitted to a central data distribution system known as Hydrotel. Hydrotel is a telemetry network system which automatically collects and automatically transmits hydrological data from telemetry-enabled gauging stations to WaterNSW's Hydstra database.
[16] The Bourke Gauge is one of the telemetry-enabled gauging stations linked electronically to Hydrotel.
[17] River height data is collected automatically in real-time by gauging stations and then transmitted to the Hydstra database via the Hydrotel system. Once this data has entered into the Hydstra database, it can be viewed and adjusted by my team and I ….
[18] The Hydstra database stores the real-time height data collected by gauging stations (including the Bourke Gauge) for each 15 minute period. Once the real-time data has been automatically collected by gauging stations in the field, it is automatically transferred to Hydstra via Hydrotel. Once the data is transferred to Hydstra, it is automatically recorded within a sub- program called 'Data Managers Workbench'. As soon as the data has been stored in the Data Managers Workbench, it can be exported to various different formats (such as Excel spreadsheets and PDFs) using the Hydstra sub-program called 'HYCSV'.
The data collected by the instruments at a gauging station contributes to the calculation of river flow data, or in other words, the quantity of water flowing at a particular point of the river at a given point in time. [47] That process is discussed in further detail below. Importantly, given the challenge which is made to the accuracy of the data on which the calculation of the flowrate is based, Mr Cutler's evidence was that gauging stations are inspected at two monthly intervals, and that the instruments are re-surveyed for accuracy every two years, and are usually calibrated every three years. [48]
[13]
The Bourke gauge
The particulars of each of the charges centred upon the flow of the river at the Bourke gauge, which was one of the gauging stations operated by the respondent. The Bourke gauge is located approximately 100km downstream from the property, and 6 km upstream from the Bourke Weir, [49] a man-made weir constructed in 1897. [50]
Between 17 and 20 May 2016 the Bourke gauge was relocated approximately 50 to 100m upstream of its previous location. The re-located gauge was formally commissioned on 7 June 2016, [51] and thus a short time before the commencement of the period of the alleged offending.
[14]
The process of conducting gaugings
Mr Cutler explained that at the time of the inspection of a gauging station (which, I have noted, is at two monthly intervals) hydrometric field staff employed by the respondent perform what is referred to as a "gauging". That process entails: [52]
1. carrying out a physical inspection of the river;
2. determining, by reference to gauge posts located within the river, the height of the river at the time;
3. inspecting upstream and downstream for items which may influence the natural characteristics of the river flow, such as the presence of large fallen trees; and
4. by the use of instruments, including an Acoustic Doppler, measuring the volumetric flowrate of the river, or in other words, the rate at which water moves in the river (such measurement being expressed in megalitres per day (MLD)).
Mr Cutler specifically explained the process of conducting gaugings at the Bourke gauge: [53]
[50] When performing gaugings at the Bourke Gauge, my Hydrometrics team and I:
a) insert a flow measuring device into the water (such as an Acoustic Doppler, Flow Tracker and/or Mechanical Current Meter, depending on which device is chosen on the day);
b) read the velocity and depth measurements produced by the measuring device;
c) record observations and some of the measurements onto hardcopy inspection sheets and save data from the measuring devices onto field laptops or USB drives;
d) once the team member has returned to WaterNSW's office, the electronic measurement data collected using the measuring device is then transferred into Hydstra by the following process:
i) downloading the gauging logger file to the computer;
ii) inspecting the logger file via on the computer screen;
iii) if the measuring device:
1. is a Flow Tracker or Mechanical Current Meter: importing the file to the Hydstra Gaugings database via the HYGAUGE sub-program; or
2. is an Acoustic Doppler: creating a summary report via the proprietary software and manually enter the gauging summary report details into the Hydstra Gaugings database;
iv) manually entering any additional comments into the Hydstra Gaugings database which are relevant to the gauging; and
v) saving the raw gauging logger files to Hydstra as a reference to the new entry within the Hydstra Gaugings database; and
e) store the gauging data collected within Hydstra as a record of the gauging for that date, which can be retrieved on command at a later time.
[51] The Hydstra database also stores records of gaugings recorded by my staff at particular gauging stations…… and entered into the Hydstra database following the these site visits. This data is used within the Hydstra 'Ratings Workbench' sub-program for the purpose of preparing Rating Tables.
In light of the matters raised in ground 1 of the present appeal, and notwithstanding Mr Cutler's reference to "my hydrometrics team and I", his evidence was that he did not attend each site visit, and that his staff were responsible for recording the relevant data. [54] None of those staff were called to give evidence before the trial judge as to what was done at a particular gauging.
Mr Cutler also explained the nature of the data obtained from gaugings conducted at the Bourke gauge: [55]
[54] The gauging data for the Bourke Gauge… shows the Hydstra database's records of each gauging performed and recorded since 28 January 1895 at the Bourke Gauge, including the following information:
a. date of the gauging;
b. start time of the gauging;
c. end time of the gauging;
d. start gauge height observed during the gauging;
e. end gauge height observed during the gauging;
f. mean gauge height observed during the gauging;
g. flowrate ML per day measured during the gauging;
h. type of flow measuring device used and the measuring device's number….; and
i. person(s) who performed the gauging.
The historical data to which Mr Cutler referred was contained in a spreadsheet which formed part of the evidence before the trial judge. [56]
[15]
The application of Quality Codes
As noted above, [57] Mr Cutler explained that at the time of conducting a gauging, the officer(s) of the respondent record the data obtained in a field inspection sheet, samples of which formed part of the evidence before the trial judge. [58] The completion of those documents incorporates the use of what are referred to as "Quality Codes" which are applied for the purposes of calculating the flowrate of the river.
The respondent published a document entitled "Hydrometric time series quality codes." [59] That document includes the following: [60]
1. Principle of quality codes
Clients need to understand whether results are measured or estimated, actual or adjusted.
Modellers need to know how to quantify model uncertainty. Resource managers need to know how close the recorded value is to a fixed reference. The fixed reference for stage may be a gauge height, which has been levelled to a recognised benchmark. For a water quality parameter a value may have been measured by a portable instrument, which has been checked against a traceable standard.
Manufacturers generally specify accuracy in terms of a percentage of the full scale range of the equipment, rather than specifying a standard error. Their calibration procedures require that all test readings are within the specified tolerances, and instruments are rejected if they do not meet specifications. For example, Campbell Scientific Australia has advised that this corresponds to a 95% confidence interval.
Thus, provided that instruments are calibrated regularly according to standard procedures, and that calibration check indicate results are within manufacturers' specification, quality codes 1, 3, 5, 7 , 9 can be interpreted to indicate that results are within a 95% confidence interval of the true value as far as instrument calibration is concerned.
The document defines and prescribes the selection of quality codes [61] whilst a separate document also published by the respondent, entitled "Gauging quality codes" addresses the use of quality codes in reliably measuring the flowrate in a way which conforms to "AS3778". [62]
[16]
Compiling of a rating table and a rating curve
With the use of the real time data collected at the gauging station, along with the data obtained in carrying out the process of gauging, the respondent compiles a rating table [63] which correlates the expected flowrate of the river with the height of the river observed by hydrometric staff. [64] The "rating" is the mean fit relationship between the: [65]
1. river height, as recorded at the time that it is physically observed at a gauging station (known as stage); and
2. the volumetric flowrate of water that is physically measured as passing the gauging station expressed in MLD (known as discharge).
The rating table is created by a software program called "Ratings Workbench", and involves that program plotting the data obtained during a gauging as a series of node points on a graph. The graph depicts the stage on one axis, and the river discharge on the other axis.
The process of creating a rating table was explained by Mr Cutler as follows: [66]
[39] Rating tables are created and adjusted within the "Ratings Workbench" sub-program within Hydstra, using the following process:
a. river height versus discharge measurements (also called gaugings, which reflect the relationship shown in the applicable rating curve) are entered into the Hydstra database and appear in the "Ratings Workbench" software on a pre-determined graphing plot;
b. in the Ratings Workbench software, the height versus discharge measurements are assessed by my staff and I for their uniformity of fit with other sequential measurements;
c. the Ratings Workbench software plots the gaugings on a graph, which enables my staff and I to use the program to view the mean fit of the gaugings as they have been sequentially recorded;
d. the Ratings Workbench software allows my team to create a curve of mean fit through these gaugings to adjust an existing relationship of mean fit or to create a new relationship when gaugings begin to deviate from the mean fit previously recorded in Hydstra;
e. when height versus discharge measurements do not uniformly fit with previous sequential measurements, the height versus discharge relationship in the current Rating Table is assessed and adjusted by my staff and I for either an adjustment of the:
i. the current mean fit of the rating curve relationship, or
ii. if a new hydraulic relationship of height versus discharge has been determined, then this is substituted in place of the previous mean fit measurements; and
f. Ratings Tables are created within Hydstra and labelled with a prefix and a suffix convention. The prefix is the rating table number and the suffix is the series release or adjustment number of the mean fit.
[40] The Hydstra database stores the Ratings Tables for individual gauging stations as used in real time and as retrospectively adjusted - this includes a record of the current Rating Table for each gauging station, as well as previous versions used for each gauge. My staff and I are able to access and export both kinds of ratings tables using the Hydstra database. The database stores ratings tables (used both in real time and as adjusted) for the Bourke Gauge.
The rating table therefore represents a pre-calculated flowrate which corresponds to a particular level of stage, based on the observed relationship between stage and discharge. [67]
Bearing in mind the operation of the gauging station, when data is received from a gauging station it is stored in the Hydstra database. Hydstra automatically determines the flowrate for the river at 15 minute intervals by applying the data obtained from the gauge as an input in the rating table and then determines the flowrate that corresponds to the height of the river at that time. [68] That process produces what is known as real-time flow data.
For the purposes of confirming the accuracy of the real-time flow data, the real-time height data for a particular date or period is then compared to the height recorded by way of physical inspection. In instances where the river height measured by the gauging station differs from the actual height observed at an inspection by the respondent's staff, the data will be adjusted to produce what is known as adjusted height data. [69] Mr Cutler explained that: [70]
1. this will usually only occur when the variance between the figures exceeds approximately 10mm;
2. the precise figure (which is usually around 10mm) is also determined by reference to guidelines provided by the gauge manufacturer; and
3. if the actual river height differs by less than 10mm to the height measured by the gauging station, the height data will usually not be adjusted. Mr Cutler explained that adjusted river height data is produced through the adjustment of the real-time height data stored within the Hydstra database by himself and his team through use of the Hydstra software. The adjustment process involves the data collected by the gauging station being assessed against the physical readings taken at the time of an inspection.
Mr Cutler also explained that the respondent assumes that the adjusted flow data which is generated from adjusted height data remains subject to a 10% margin of error which may be due to: [71]
1. the natural deposition of detritus matter compromising pressure sensing;
2. instrumental sensitivity errors; and/or
3. the process requiring the retrospective and uniform application of the flowrate recorded on a particular inspection date to the preceding two months' real-time height data.
Mr Cutler's evidence was that the current rating table used in respect of the Bourke gauge is rating table 300.14. [72] By reference to that rating table, Mr Cutler prepared and produced a number of spreadsheets. They included:
1. a spreadsheet setting out the data and records for each gauging for the Bourke gauge stored in the Hydstra database for the period of 1895 to September 2018; [73] and
2. a spreadsheet setting out the adjusted river flow data generated by the Hydstra database for the period 15 June 2016 to 5 July 2016 (thus encompassing the charge period of 22 June 2016 to 27 June 2016) and which demonstrates the flowrate of the river expressed in MLD. [74]
The second of those spreadsheets demonstrated that on each of the days within the period of the charge alleged against the appellants, the flowrate of the river was substantially less than what was contained in the conditions of the approval, and less than that stipulated in the charge, namely 4,894 MLD. [75] It was the position of the respondent before the trial judge that this evidence established that the recorded flow at the Bourke gauge was at least one billion litres per day less than the flowrate of 4,894 MLD specified in each charge.
Following the preparation of the rating table, a "curve of mean fit", known as a rating curve, is produced. The rating curve is a graphical representation of the rating for a particular gauging station, with the gaugings recorded in the Hydstra database being plotted on a graph so as to show a stage/discharge relationship. [76] The curve is therefore the mean fit of a series of physically measured river height versus discharge measurements. [77]
[17]
Australian Standard 3778
The reference to "AS3778" is a reference to Australian Standard 3778 which sets out the protocols and procedures to be followed when collecting data. Clause 1 of Pt 2.3 of AS3778 is in (inter alia) the following terms: [78]
Scope
This part of ISO 1100 specifies methods of determining the stage-discharge relation for a gauging station. A sufficient number of discharge measurements, complete with corresponding stage measurements, is required to define a stage-discharge relation to the accuracy required by this part of ISO 1100.
Clause 5.2.6 is in the following terms: [79]
[R]egardless of the measuring method, the discharge through the discharge measuring section or reach shall be the same as the discharge normal to the reference stage gauge over the entire range of discharge rates. At a gauging station, different measuring sections or different methods may be used to cover the discharge range.
Clause 6 of the Standard is headed "Stage-discharge calibration of a gauging station". Clause 6.1 is in the following terms: [80]
6.1 General
The primary object of a stage-discharge gauging station is to provide a record of the discharge of the open channel or river at which the water lever gauge is sited. This is achieved by measuring the stage and converting this stage to discharge by means of a stage-discharge relation, which correlates discharge and water level. In some instances, other parameters such as index velocity, water surface fall between two gauges, or rate-of-change in stage may also be used in rating calibrations. Stage-discharge relations are usually calibrated by measuring discharge and the corresponding gauge height. Theoretical computations may also be used to aid in the shaping and positioning of the racing curve. Stage-discharge relations from previous time periods should also be considered as an aid in the shaping of the rating.
Clause 6.2 addresses the general preparation of a stage discharge relation. Clause 6.2.1 is in the following terms: [81]
6.2.1. General
The relation between stage and discharge is defined by plotting measurements of discharge with corresponding observations of stage, taking into account whether the discharge is steady, increasing or decreasing, and also noting the rate of change in stage. This may be done manually by plotting on paper, or by using computerized [sic] plotting techniques. A choice of two types of plotting scale is available, either an arithmetic scale or a logarithmic scale. Each has certain advantages and disadvantages, as explained in subsequent clauses. It is customary to plot the stage as ordinate and the discharge as abscissa, although when using the stage-discharge relation to derive discharge from a measured value of stage, the stage is treated as the independent variable.
Clause 6.2.2 is in the following terms: [82]
6.2.2 List of discharge measurements
The first step before making a plot of stage versus discharge is to prepare a list of discharge measurements that will be used for the plot. At a minimum this list should include at least 12 to 15 measurements, all made during the period of analysis. These measurements should be well distributed over the range in gauge heights experienced. It should also include low and high measurements from other times that might be useful in defining the correct shape of the rating and/or for extrapolating the rating. Extreme low and high measurements should be included wherever possible.
For each discharge measurement in the list the following items shall be included:
a) Unique identification number
b) Date of measurement
c) Gauge height of measurement
d) Total discharge
e) Accuracy of measurement
f) Rate-of-change in stage during measurement, a plus sign indicating rising stage and a minus sign indicating falling stage.
Other information might be included in the list of measurements, but is not mandatory. Table 1 shows a typical list of discharge measurements, including a number of items in addition to the mandatory items. The discharge measurement list may be handwritten for use when hand-plotting is done, or the data may be a computer list where a computerized [sic] plot is developed.
Clause 6.3.3 is in the following terms: [83]
6.3.3 Hydraulic equation curves
The shape of stage-discharge relations can sometimes be defined through the use of hydraulic equations, namely equations (1), (2) and (3). Where section control exists, the weir equation (1) can be used to compute rating curve points. Coefficients of discharge, C, are defined in other International Standards for certain types of weirs and flumes, so that a reasonably accurate rating curve can be computed that will conform to correct hydraulics. For natural section controls, such as rock outcrop or sand bar, the coefficient of discharge can be estimated on the basis of calibration measurements. Widths and depths can be determined from a surveyed cross-section of the control section.
For segments of the rating curve that are influenced by channel control, the shape of the rating can be defined through the use of equation (2) or (3). An average or typical cross-section in the control reach is surveyed to define the channel characteristics of cross-section area and hydraulic radius. The Manning rugosity, n, or the Chezy C is estimated from field observations. The friction slope is estimated from channel surveys, maps, or calibration measurements. Equation (2) or (3) can then be used to compute discharge for a few selected gauge heights to define the shape of the rating curve. This is a simplified procedure which assumes steady, uniform flow. More complex situations involving non-uniform flow can be analysed with various techniques of backwater curve computation. Computer programs are available for such analyses.
For either case, section or channel control, the rating computed by the hydraulic equations is used only for defining the hydraulic shape of the rating. The correct position of the rating is defined by the calibration measurements. This procedure can also be used to aid in determining when measurements define a new rating position, such as may be the result of a shifting control.
Clause A.2.1 of Annexure A is in the following terms: [84]
A.2 Statistical analysis of the stage-discharge relation
A.2.1 The stage-discharge relation, being a line of best fit, should be more accurate than any of the individual gaugings. The equation of the relation may be computed as detailed in 6.1.3.2, which assumes that the relation plots as a straight line on logarithmic paper.
[18]
The expert evidence as to the calculation of flowrate
Glenn McDermott, a Consultant in Hydrology and Hydrometric practice, provided a report which was relied upon by the respondent, in which he explained the difference between a calculated flowrate and a true flowrate: [85]
[35] It is necessary to distinguish between calculated flowrates and true flowrates. These are as follows:
a) A true flowrate is the volume of water which has actually passed a measuring point in a given time period. It is measured with 100% confidence and there can be no statistical doubt as to its value. In practical terms, I am unaware of any river measuring device in Australia which operates at an 100% level of confidence and has a 0% measurement uncertainty at all times. Calculating the true flowrate of a river is, in my view, a practical impossibility.
b) A calculated flowrate is simply the best estimate of the true flowrate, which is calculated using data and industry-accepted methods. By definition, these estimates are subject to some degree of uncertainty at each step in the measurement/estimation process. In practice, the calculated flowrate of water may be slightly greater or less than the true flowrate. Accordingly, widely-accepted hydrometric industry practice is to calculate the upper and lower confidence limits about the calculated flowrate, within which it can be said with 95% confidence that the true flowrate lies.
Mr McDermott explained that any calculation of flowrate carried with it what he described as an associated "uncertainty of measurement", and that the standard, and widely accepted, hydrometric practice is to calculate flowrate by reference to the 95th percentile confidence level. [86] Mr McDermott undertook his own assessment of the flowrate of the river by reference to rating table 300.14 and concluded that any uncertainty was at that confidence level. [87] In carrying out that assessment, having analysed the data underlying the 20 gauge points on the rating curve nearest to a flowrate of 4,894 MLD, [88] Mr McDermott concluded that: [89]
1. the largest deviation was 13.02%;
2. there was a standard deviation of 6.18%; and
3. the true flowrate of the river was within a range of 4,351 MLD to 5,437 MLD. [90]
In cross examination, Mr McDermott expressed the view that because the points on the rating curve were quality coded they were not lacking in measurement uncertainty or integrity. [91] When asked for the basis of that opinion, he said: [92]
Looking at the Water New South Wales quality coding system, they've classified each gauging as basically very accurate - very accurate to fairly accurate but they've put a number of it. So plus minus 5% so about a third of the gaugings in our interest period are very accurate - better than plus minus 5%, and they actually state that in the information on each gauging. So I didn't sort of go into that. I accepted that that's the best they could do with the technology they had in a large river, and none of them are sort of worse than plus minus 10%.
Daniel Martens, a Geotechnical Engineer and Environmental Scientist, provided a report which was tendered in the appellants' case. [93] Dr Martens' evidence was that the calculation of the flowrate of a river was often a technically difficult and time consuming task, and one which, in his opinion, was subject to various sources of error. [94] Dr Martens made reference to the fact that there were available devices which could be installed into a river for the purposes of directly measuring the flowrate at a particular point, and expressed the view that if a series of devices were installed at appropriate locations across a cross-section of the river, and on the assumption that each device was always correctly calibrated and functioning without error, data would be produced that would enable a calculation to be undertaken to produce an estimate of the flowrate at that cross section. [95]
In expressing his opinions, and in proceeding on the understanding that the respondent had not installed such equipment and had instead relied upon a rating curve based upon underlying data, [96] Dr Martens' evidence was that the respondent's approach necessitated adopting a number of core requirements, including: [97]
1. an accurate survey of the river channel;
2. measuring the river flow velocity with a measuring device, such as an impeller or doppler device; and
3. calculating the discharge by reference to an established mathematical equation.
Dr Martens expanded upon this by saying: [98]
Given the nature of the process I have just described, the calculation to be undertaken is one which should involve the individual exercising judgment about which segments across a channel are similar, and how to combine or treat separately different measurements at different points on the cross-section in order to produce one single figure for flow in the river at that location [even though there will inevitably be different flow rates at different parts of the river's cross-section at that location]. This means that the data points on flow curve graphs that are a record of the "measured" relationship between height and flow of the river at a particular point on a particular day are not records of some empirical facts, but instead represent opinions based on judgments made based on data gathered. Mistakes and errors in judgment or calculation will therefore cause the single figure ultimately chosen to be wrong, in the sense of not reflecting the precise actual flowrate of the river at that location on that day.
Dr Martens expressed the opinion that there were "numerous sources of uncertainty and error" associated with rating curves, in terms of both the measurements incorporated in them, and the process of their actual compilation. [99] He cited a number of factors which, in his opinion, affected (or at least had the capacity to affect) the validity of a rating curve, including: [100]
1. an inaccurate survey of the river bed and bank conditions during gaugings;
2. an inaccurate or insufficient measurement of river flow velocities and heights across a channel section;
3. changes to channel cross-section due to scour and fill;
4. growth and decay of aquatic vegetation, log and debris; and
5. variable backwater conditions.
Dr Martens went on to identify what he described as "various sources of error" that in his opinion contributed towards uncertainty of the flowrate based upon river height at the Bourke gauge, including: [101]
1. inherent equipment uncertainty, "in-situ" factors such as the presence of unknown flow obstructions, and the variation in flow at a given point in time;
2. the form of the channel, due to its shape and its condition;
3. sampling errors, and associated failures to follow necessary processes; and
4. the rating curve itself, arising from the uncertainty of the river level.
In specifically addressing the rating table prepared in relation to the Bourke gauge, Dr Martens noted the following: [102]
1. the gaugings used to create the ratings curve and table had been taken from within an 11 km reach of the river;
2. almost no gaugings were reported to have been taken at the Bourke gauge;
3. some gaugings were taken downstream of the Bourke weir, and some upstream; and
4. at each separate gauging location there would be an expected different discharge and river stage/height relationship.
I have previously made reference to the evidence surrounding the movement of the Bourke gauge, along with the evidence of Mr Cutler that approximately every two months, members of his Hydrometrics team attended the Bourke gauge to perform maintenance on the instruments, and to carryout gaugings. The effect of the movement of the Bourke gauge on the reliability of data obtained by the respondent to calculate the flowrate of the river was a matter of dispute between the experts.
Mr McDermott was aware of the fact that the Bourke gauge had been moved but did not address that issue in his report. When asked why he had not done so, Mr McDermott explained that in his opinion, such movement had no bearing on the data which was collected by the gauge: [103]
Because the accuracy of any of these level measuring sites relies on the gauge posts, the series of gauge posts which are hammered in up the bank. And as long as they are the same as wherever they used to be, which is only 30 metres away, they would tend to be the same along the whole 6 kilometre length. Like, that - the - the level pool, particularly in low flows, is a - if you ideally set the gauge posts to that, you know, if it was - it was - if it was 3.98 at the old location, you just set the post and move it up or down till it hits 3.98 at your new location and you are measuring the same thing. Because it's one level pool all the way back. So long as they calibrate their new set of level staff gauges to their old ones, that would be accurate.
Although Dr Martens took issue with Mr McDermott's opinion in this respect, [104] he said the following when cross-examined: [105]
Q Can I ask you this. In your report you expressed concern that the moving of the Bourke gauge might have had some effect upon the accuracy of the readings?
A Yes.
Q But if that were the case in the same way, if the Bourke gauge as moved was giving erroneous results, then the gaugings after that point should form a new line, should they not?
A Possibly.
Q But there's no evidence of a new line on this graph, is there?
A This chart at attachment D is to 25/11/2016. So there are going to be two or three points maybe that would be, if you like, between the charge period when the gauge was moved, between the gauge movement and when this chart was drawn.
Q Yes, but if for example a gauging was taken after that time in that area, then you would expect to see it on this line. Correct?
A Yes.
Q If it was on that line, you could be confident that the gauge was accurate and the movement of it had made no difference.
A I'm not following. If it was on the line, that would mean it would be on the line. It wouldn't mean that it's more or less accurate than other measurement made.
Q What I'm saying is if the Bourke gauge was not giving an accurate reading after it was moved, then gaugings taken from that reference point would start to fall off this curve, wouldn't they, because the relationship between flow and height would be disturbed, because the height would be giving an inaccurate reading.
A Let's say the gauge was wrong by 20 centimetres.
Q Yes.
A On this chart, that would fall within the scatter that this chart already shows. So I don't know whether that answers your question. If it falls exactly on the line, it would fall exactly on the line, but the next measurement might be different again.
Q Yes, but you follow my point, don't you? If the Bourke gauge is inaccurate, the ratings - the readings that it gives and the correlation between the two variables would be disturbed because the height data that the Bourke gauge would be giving would be erroneous.
A It would be disturbed, yes.
Q So the correlation should not fall consistent with the correlation which has existed when the true height was the variable being correlated.
A It would be different - on a different path, yes.
A further issue between the experts was the effect, if any of the Bourke weir on the calculation of the flowrate. In addressing this issue, Mr McDermott said: [106]
My assessment is that the Bourke Weir controls the water level to discharge relationship (rating) for low and medium flows. I assume that the "weir" control (also termed "section control") is gazumped or subsumed in high flows. Like most rivers, I assume that during very high flows, the flood breaks out of the channel confinement and overflows sideways.
Dr Martens took a different view, saying: [107]
Hydraulic conditions upstream and downstream of the Bourke Weir are very different because downstream of the Bourke Weir, flows are not modified by a subsequent weir structure.
When asked about this aspect of Dr Martens' opinion, Mr McDermott said: [108]
The practice of the industry recognises that in a weir pool backed up by a large weir like Bourke, you're measuring the height of some six kilometres back. But when - in any period of steady flow, you can measure the flowrate anywhere. It really doesn't matter where you're measuring it because it's - you after the flow so you can put one point on the rating curve. And the flow is the same flow in the entire weir pool, and even over the weir. Doesn't matter if - cause it's the same flow. It doesn't - the flowrate during that time you're gauging doesn't vary. It's - it's - and that's the normal industry practice is you - you pick the spot to do your gauging that's convenient for you, and it can be anywhere, either side of the actual level measuring point, and that's normal practice.
Mr McDermott was then asked: [109]
Q. You told his Honour earlier that in a weir pool, you could take flow measurements anywhere within the pool?
A. Yes.
Q. Could you tell his Honour why that is so?
A. Cause throughout the pool, the flowrate is the same cause water is incompressible so it just pushes itself at the same rate. Like in the - in the course of going down the pool, there might be a hump in the bed at a deeper part - another shallower part but continuity principle is the product of velocity in the area so as you go down, it's sort of a faster velocity times a small area when there's a hump and a shallow bit. And then when you come to a deeper pool, it's a slower velocity times a bigger area for the deeper pool but the flowrate is the same. And it really doesn't matter where you pick to do the gauging. You'll get the same flowrate.
When directed to the opinion of Dr Martens, Mr McDermott explained why he did not agree with it: [110]
Because the flowrate is the same in that entire section of weir pool. It really doesn't matter if you pick a shallow part of the pool or a deep part to do your gauging. You will measure a flowrate which relates to the level back there. It really doesn't matter that it's a different cross section cause of continuity of mass.
In commenting on the opinions of Mr McDermott, Dr Martens, said: [111]
Well, a weir pool is the pool that's created behind a weir and it obviously can vary in length. But ultimately, the flow in a channel is governed by the catchment area that drains to that channel, irrespective of the weir pool being present or not. The flow in the channel is also governed by the hydraulic properties of the channel, and what I mean by that, factors such as bridges may arrest the flow of water passing through the bridge, indeed, in the same way that a weir would arrest the flow of water.
Other obstructions such as rocks, boulders, the bed and bank conditions, vegetation, all these are hydraulic considerations that have a very significant impact on the flow at a particular station when it's being measured. So, there are catchment area factors, there are hydraulic, or perhaps I could bundle those into frictional related factors. But then there are, of course, other factors such as evaporation, which over a large weir pool could be quite substantial and varies along the length of the weir pool. And possibly some loss through the bed and the banks of the wetted perimeter of the weir pool into the soil below, and that's to ground water below.
So, there are a number of factors and those factors become more and more relevant with the length of - or the distance along the river. But it's probably worthwhile mentioning the word "weir pool", or the words "weir pool" are a generic, sort of general term and they tend to apply to the pool that's created behind the weir when the weir is not flowing, because it's difficult to define where the - where the weir pool actually is when the weir is flowing, and ultimately the weir and the pool, and the areas down side of the weir may be merged into one continuous flow.
When asked to explain what he meant by "catchment area factors", Dr Martens said: [112]
So in hydrology there is a very strong correlation between the area of a catchment that contributes to the flow at a particular point in a river. And at its extreme if there is no catchment to a river at a particular point there'll be no river because there'll be no water arriving, and as the catchment area increases, the flowrate tends to increase.
And there's a particular field in hydrology where - where catchment area is used as a surrogate for flow, in circumstances where the discharge or the flow in the river is difficult to measure or unavailable. In other words, you could, for example, plot the width of the cross-sectional area of a river against its discharge at that point, and then create a chart at different locations along the river.
But you could equally create the similar chart replacing discharge as, if you like, the X-axis or the Y-axis with catchment area. So there's a very strong relationship between catchment area and flow in a river.
[19]
Ground 1 - The trial judge erred by finding that the [respondent] had demonstrated beyond reasonable doubt that the flow rate at the Bourke gauge was less than 4,894 ML/day in the period 22 June to 27 June 2016.
[20]
Ground 2 - In the course of determining that flow issue, the trial judge further erred by:
1. failing to understand the effect of the evidence of Dr Martens in demonstrating why it was that the evidence upon which the [respondent] relied did not establish the alleged flow rate beyond reasonable doubt;
2. treating as relevant whether Dr Martens had identified practicable alternatives to measuring flow;
3. admitting the departmental record of gauge node points, and evidence based upon that record;
4. treating Mr Cutler as having given evidence as to what the field officers actually did when the gaugings in question were carried out;
5. failing to recognise Mr McDermott's treatment of the inherent uncertainty in the calculation of flowrate was one which assumed relevant standards and processes for the gathering of reliable gauge node points had been properly followed.
[21]
THE FINDINGS OF THE TRIAL JUDGE
It is appropriate at this point to set out the principal findings of the trial judge as they relate to each of these grounds.
[22]
Ground 1 - Proof beyond reasonable doubt of the fourth element of each offence
The trial judge commenced by identifying the four elements which the respondent was required to prove in order to establish the charge against each appellant, the fourth being that during the relevant period, water was taken when the flow of the river at the Bourke gauge was equal to or less than 4,894 MLD (the fourth element). [113] In doing so, his Honour noted (inter alia) that: [114]
1. the flowrate of a river can vary across different sections of the river, and at different positions and depths along a channel cross-section;
2. these variations in flowrate can result from a variety of factors, including changes in the frictional characteristics of a river, the presence of other hydraulic controls such as the shape of the channel, or by artificial characteristics such as the Bourke weir;
3. the experts had agreed that some level of uncertainty in calculation of flowrate, by reason of possible instrument or measurement error, is unavoidable;
4. the experts agreed that it is accepted hydrometric practice to assess measurement uncertainty by reference to a 95% confidence level; and
5. AS3778 provides guidance on how to estimate uncertainty in flowrate measurement.
His Honour summarised the evidence of the process of calculating the flowrate, [115] before identifying three particular issues in relation to the fourth element, namely: [116]
1. the reliability of the underlying data;
2. the accuracy and reliability of the rating curve and the calculated flowrate; and
3. the impact of the movement of the Bourke gauge.
In identifying those issues, his Honour emphasised the need to consider the evidence as a whole, and not in a piecemeal fashion. [117]
As to the reliability of the underlying data used to calculate the flowrate, his Honour concluded: [118]
[366] In relation to the reliability of the underlying data, I accept the evidence of Mr Cutler and Mr McDermott and consequently find that the data is not unreliable to the extent that the flow rate cannot be reliably ascertained as being below 4,894 ML per day during the charge period. In making this finding, I note that the combined effect of Mr Cutler and Mr McDermott's evidence demonstrates that a variety of comprehensive procedures were followed in collecting, processing and analysing flow rate data at the Bourke gauge, which, in any event, produced data with minimal uncertainty as shown by Mr McDermott's statistical assessment, which I accept.
His Honour went on to say: [119]
[370] River flow data extracted from the Department's Hydstra system by Mr Cutler shows that, in the period from 15 June 2016 to 5 July 2016, the earliest time at which the flow rate of the Darling River at the Bourke gauge reached 4,894 ML per day was 15:30 on 30 June 2016. The river flow data does not, however, allow for a margin of error. Nevertheless, the impact of that margin of error was quantified by Mr McDermott in his report and he indicated that based on his calculations, when the river flow data is adjusted for measurement uncertainty corresponding to at least a 10% margin of error, the flowrate of 4,894 ML per day may have been reached, at the earliest, by 08:45 on 29 June 2016.
In addressing the accuracy and reliability of the rating curve and the calculated flowrate, his Honour said: [120]
[377] Noting the above, I find that an appropriate margin for error is provided in Mr McDermott's assessment of uncertainty and I accept his calculations which take into account any deviations in gaugings data from Rating Table 300.14 around a flow rate of 4,894 ML per day. Even with this allowance being made, and noting that error is not guaranteed in the first place, or indeed could skew a data point in favour of the [appellants'] position, I accept Mr McDermott's approach toward calculating flow rate. I find that the McDermott Rating Curve provides independent verification of the conclusions derived from Rating Table 300.14 that the flow rate did not exceed 4,894 ML per day before 29 June 2018, as both the McDermott Rating Curve and Rating Table 300.14 calculated the flow to exceed 4,894 ML per day at a similar time.
His Honour's reasons for accepting Mr McDermott's methodology were that: [121]
1. Mr McDermott conduced a comprehensive assessment of the uncertainty associated with flowrate measurement;
2. in conducting his measurement uncertainty analysis for the Bourke gauge, Mr McDermott specifically examined flowrates around the range of 4,894 MLD in order to prevent very high or low flowrate data from affecting his calculation; and
3. any of the possible sources of error identified by Dr Martens had been taken into account by Mr McDermott in his calculations.
His Honour then continued: [122]
[382] As such, I find that Rating Table 300.14, because it is based on gaugings taken by a range of instruments at different times and over a substantial time period, adds to its strength as a reference. To that end, I accept the [respondent's] position that even if an individual gauging may have been inaccurate, a larger averaging of the data would mean that Rating Table 300.14 would give relatively little weight to an inaccurate gauging despite this not being accepted by Dr Martens. I do not find the [appellants'] argument that errors which are repeated on every gauging occasion would be undetectable as being compelling, and insufficient to contribute a reasonable doubt, given that the data upon which Rating Table 300.14 is based has been recorded over approximately 27 years.
[383] I also accept the analysis of Mr McDermott that the adjusted height data and real time height data for the duration of the charge period were practically identical, as the two data sets contained relatively few discrepancies which, in any event, were limited to 1mm. In consequence, I find that the distinction between adjusted and real-time data therefore has no bearing on the overall accuracy and reliability of the data used to calculate flow rate, or was otherwise accounted for in Mr McDermott's statistical analysis.
[384] Although the [appellants] submit that Mr McDermott's evidence offers no assistance as he merely assumes the reliability of the Gauging Log and performs a statistical exercise and, as such, allowing a margin of error of some 10% is irrelevant, and that Mr McDermott did not assess the extent to which the rating curve and Rating Table 300.14 would be incorrect if the Gauging Log was wrong, considering the extent of the analysis conducted by Mr McDermott as described earlier and the existence of estimated uncertainty for each gauging as contained in Gauging Uncertainty Graph, I find that the [appellant's] submission materially understates the evidence of Mr McDermott. I find that he adopted a statistically appropriate methodology to quantify the extent of errors likely to be present in the flow rate data.
…
[386] Given my findings, I consider that each of the [appellants'] arguments in relation to this aspect simply hypothesise possible errors which, even if the case, would yield a relatively minor result (for example, without trivialising the concerns, failing to notice a log in the river or misreading the river height by a matter of centimetres due to human error or time delay would have a negligible impact) and, in any event, these errors have been accounted for in Mr McDermott's statistical analysis. As such, I have no reasonable doubt that a combination of these possible errors is insufficient to create a deviation from the recorded flows during the charge period beyond that which was accounted for by Mr McDermott, such that it would approach or exceed 4,894 ML per day.
[387] I therefore accept that the Department's rating curve and Rating Table 300.14 are capable of determining the relationship between river height and flowrate with sufficient precision for it to reliably prove that the flow rate was below 4,894 ML per day. In light of my findings above, I consequently do not accept the [appellants] submissions that the rating curve is illogical, based upon assumptions which have not been proven, or that the methods used by the [respondent] to demonstrate its reliability lacked in rigour.
[388] For completeness, I also do not accept the [appellants'] submission that the procedure is "primitive", subjective and prone to error. Although the "system" does depend upon various processes which involve some element of subjective input, these are criticisms that do not, either individually or cumulatively, lead to the view that the system itself or its manner of implementation (especially in relation to adherence to various standards and guidelines) leads to any reasonable doubt in relation to the accuracy of the figures. More particularly, I do not find that the numerous matters raised, considered cumulatively, raise a reasonable doubt in relation to the reliability of the figures.
As to the suggested impact of the Bourke weir, and the movement of the Bourke gauge, on the reliability of the data, his Honour concluded: [123]
[402] Although not determinative in my overall findings, I accept the evidence of Mr McDermott in relation to the effect of the Weir, and generally in relation to the "Weir pool". As such, I accept Mr McDermott's evidence that the Weir is a unique hydraulic control producing, in effect, a uniform flowrate across the length of the Weir pool which extends up to and including the Bourke gauge during periods of low to medium flow. I also accept Mr McDermott's evidence in relation to the normal practice in the industry and that consistent flowrate about which he opines is the result of consistency in channel factors throughout the weir pool as well as the incompressibility of water.
[403] In accepting Mr McDermott's evidence, I find that it is unlikely that the possible introduction of "rocks, scour, logs and vegetation and/or evaporation" could eliminate the controlling impact of a Weir which Mr McDermott described. Further, although there was dispute between Dr Martens and Mr McDermott in relation to the amount of the "scatter" in relation to the rating curve, I accept Mr McDermott's evidence that, having conducted a statistical analysis of gauging data nearest to a flowrate of 4,894 ML per day, he found that there was a maximum deviation of 12.4% within the gaugings data from Rating Table 300.14, and Despite Mr McDermott's concession in relation to possible leakage, I accept that potential sources of error such as this have been adequately accounted for in Mr McDermott's uncertainty analysis.
….
[405] In addition to the above, I find that the taking of gaugings at various points along the river (even away from the physical location of the Bourke gauge) is in accordance with the accepted industry practice in circumstances where the flow remains uniform due to weir control and as confirmed by reference to Pt 2.2 cl 5.2.6 and Pt 3.1 cl 5.1(o) of AS3778 and cl 6.1 of the National ADCP Guidelines to which I have been referred. I do not accept Dr Martens' evidence in these circumstances that the taking of gaugings away from the Bourke gauge is not in accordance with accepted practice described in AS3778. In making this finding, I note that the presence of the Bourke Weir and its subsequent effect on the hydrology of the relevant river section was critical and I have considered Dr Martens' concerns in relation to this aspect and accept that, had the factual circumstances of the river been different, these concerns may have manifested a different result.
[406] I also accept the evidence of Mr McDermott that the movement of the gauge would have no impact if the gauge was recalibrated correctly, and that the records of WaterNSW indicate that the relocated site was surveyed prior to the charge period. To the extent that there was further criticism by Dr Martens in relation to Gauge Posts being correctly calibrated, I repeat my finding that the evidence demonstrates that gaugings taken after the Bourke gauge relocation were consistent with the long term trend and were further analysed by Mr McDermott as having a marginal deviation from the rating curve. Further, even if it were the case that the river heights were taken at the gauging location instead of the Bourke gauge, the evidence regarding the weir control and incompressibility of water given by Mr McDermott (which I have accepted) indicates that the calculated flow would remain unchanged.
[407] In relation to the movement of the Bourke gauge in May 2016, the [respondent] relied on gauging records taken after movement of the gauge in order to demonstrate the negligible effect of the relocation. Mr McDermott's statistical analysis of gauging number 590 and gauging number 557 and their respective deviations from the rating curve of 1.5% and 5.62% adds weight to the [respondent's] submission that the movement of the Bourke gauge did not impact upon the accuracy and reliability of the rating curve. Further, when questioned in cross examination on the consistency of these data points with the rating curve and Rating Table 300.14, Dr Martens described each as falling within a "scatter" of dots. I consider that the use of statistical methods by Mr McDermott to quantify the extent of uncertainty and variability, which is founded on a scientific basis and outlined in the corresponding Australian Standard, is the approach which ought to be adopted in the circumstances. As such, I have no reasonable doubt that the river height data obtained at the Bourke gauge after its relocation in May 2016 and therefore during the charge period was accurate.
[408] For the above reasons, I find that movement of the Bourke gauge and the taking of gaugings at other locations within the weir pool did not have an impact upon the accuracy of the rating curve, Rating Table 300.14 or the ultimate flowrate calculated at the Bourke gauge beyond the level of uncertainty that was appropriately accounted for by Mr McDermott. Further, I do not accept Dr Martens' evidence that factors such as friction, evaporation and leakage have a relevant or material effect or were otherwise not accounted for in Mr McDermott's statistical assessment.
Ultimately, his Honour concluded: [124]
[409] Whilst I have made separate findings above in relation to specific areas of disagreement between the experts, I consider that Mr McDermott did, as the [respondent] submits, provide clear and reasoned explanations for agreeing (and on occasion disagreeing) and included detailed reference to AS3778 and the processes and practices employed by WaterNSW and confirmed the veracity of these practices including the underlying gaugings data.
[410] Both parties made submissions concerning the manner in which the experts had approached and gave their evidence. The [respondent] sought to contrast the evidence of Mr McDermott, who on its submission had provided clear evidence supported by reference to applicable standards that he was well familiar with, with Dr Martens' evidence which the prosecution submits does not make any hard findings, was based on a limited understanding of the details of the case and the relevant section of river, contained important concessions, and was prepared within a relatively short four day period. Further, the [respondent] submitted, and I accept, that Mr McDermott was also not significantly challenged in cross examination. In response, the [appellants] submit that it is not open to the [respondent] to make something of the date of Dr Martens' report and the time taken to prepare it given that this was not put to Dr Martens in cross examination. The [respondent] submit that the issues addressed in Dr Martens' report were not complex in light of his expertise and the limited data available.
…
[412] Although I accept that the [respondent] bears the onus beyond reasonable doubt throughout, I have been presented with the evidence of Mr McDermott which quantifies, in accordance with accepted industry practice and a comprehensive Australian Standard, the full and cumulative extent of error and uncertainty, which in contrast Dr Martens raises as concerns, if not possibilities. As such, I prefer the evidence of Mr McDermott in relation to assessment of error and uncertainty in gauging data and in calculation of flowrate at the Bourke gauge….
[413] In accepting the evidence of Mr McDermott and light of my findings noted above, including in relation to the accuracy and reliability of the underlying data, the accuracy and reliability of Rating Table 300.14, and the evidence in relation to the movement of the Bourke gauge and how gaugings were undertaken, I find beyond reasonable doubt that the flow of the Darling River at the Bourke gauge was below 4,894 ML per day for the duration of the charge period.
Finally, his Honour determined: [125]
[414] The WM Act provides a comprehensive statutory regime for the sustainable and integrated management of the water sources of New South Wales for the benefit of both present and future generations and, relevantly, encourages the sharing of responsibility for the sustainable and efficient use of water between the Government and water users. Despite its noble intentions, the legislative regime is not without complexity.
[415] I have considered all of the evidence closely. The [appellants] bear no onus and the matters that need to be established beyond reasonable doubt are the essential elements of the offence.
[416] For the reasons I have given, I find beyond reasonable doubt, first, that the [appellants] were holders of a Water Supply Works and Water Use Approval applying to the Barwon-Darling Unregulated River Water Source during the charge period; second, that the Approval contained a condition that water must not be taken when the flow of the Darling River at the Bourke gauge was equal to or less than 4,894 ML per day; third, during the charge period the [appellants] took water from the water source; and, fourth, that that water was taken when the flow of the Darling River at the Bourke gauge was equal to or less than 4,894 ML per day.
[23]
Ground 2(a) - Failing to understand the evidence of Dr Martens
The trial judge observed that although the respondent had acknowledged that both the real time flow data and the adjusted flow data was subject to a 10% margin of error, [126] Dr Martens' opinion was that the margin of error was much larger, and therefore could not be used to confidently estimate the flowrate. His Honour specifically addressed this aspect of Dr Martens' opinion as follows: [127]
[340] The [appellants] rely on the evidence of Dr Martens, who raises concerns regarding the Department's procedures used to calculate flow rate and the nature of the data used to create Rating Table 300.14. Dr Martens deposes that the assessment of river flow is subject to a range of possible sources of error and uncertainty.
[341] Dr Martens' concerns may be broadly grouped within three primary categories:
(1) Uncertainty derived from instrument and equipment error and sensitivity;
(2) The effect of human intervention and judgment, including: first, human error and the subjective nature of data collection; second, channel form uncertainty in relation to how channel surveys have been undertaken; third, inconsistent gauging locations and movement of the Bourke gauge in May 2016; and, fourth, the reliability of historical data; and
(3) The calculations and processes used to produce Rating Table 300.14 and the rating curve from which that table is derived, which result in uncertainty in flow velocity, rating curve uncertainty, and uncertainty in calculated flowrate.
His Honour went on to observe that Dr Martens had given evidence that: [128]
1. each of the measured inputs required to assess flowrate were necessarily inexact;
2. an inadequate assessment of the nature of the river channel could generate unreliable data;
3. the locations at which the gaugings were taken varied;
4. there was significant "scatter" above and below the rating curve;
5. uncertainty may result when calculating flowrate as a consequence of inappropriate or unrepresentative flow velocity sampling; and
6. the respondent's own procedures considered and recognised levels of uncertainty and error.
His Honour generally considered the evidence of Dr Martens on the basis that such submissions "simply hypothesise possible errors which, even if the case, would yield a relatively minor result…". [129]
Having noted the dispute between Dr Martens and Mr McDermott as to the rating curve, his Honour concluded: [130]
[403]… I accept Mr McDermott's evidence that, having conducted a statistical analysis of gauging data nearest to a flowrate of 4894ML per day, he found that there was a maximum deviation of 12.4% within the gauging data from Rating Table 300.14 and despite Mr McDermott's concession in relation to possible leakage, I accept that potential sources of error such as this have been adequately accounted for in Mr McDermott's uncertainty analysis.
[24]
Ground 2(b) - The failure of Dr Martens to identify practical alternatives
In relation to this ground, his Honour concluded: [131]
[411] Despite the [appellants'] submission that Dr Martens did not merely express "concerns", and acknowledging that Dr Martens, as an expert, opined that the evidence he had been provided with did not enable him to reliably assess the matter, I nonetheless find that Dr Martens' criticisms were largely of a speculative nature. Dr Martens did not quantify or qualify the impact of any sources of error he addressed, nor did he present any reasonable or practicable alternatives, with the exception of installing a series of measuring devices at appropriate locations along the cross-section of the river, to the methodology employed by Department personnel in recording gauging data and producing Rating Table 300.14.
[25]
Ground 2(c) - Admission of the evidence of gauge node points
His Honour identified the issues relevant to this ground [132] and concluded that: [133]
1. the recording of the data at the time of a gauging, and the observations of employees or officers of the respondent, was not opinion evidence;
2. although the expression "opinion" was not defined in the Evidence Act 1995 (NSW) (the EA), it was to be understood, in the context of this case, to be an inference drawn from observed and communicable data; and
3. the recording of the primary data was not an opinion, and thus was not subject to the exclusionary rule governing the admission of opinion evidence.
His Honour went on to say: [134]
[328] In ruling on admissibility of the documents (in particular the spreadsheets) based upon the primary data, I am mindful of the distinction between admissibility and weight. I have therefore taken into account the submissions made in relation to admissibility in considering the weight to be attributed to the evidence which I have admitted. I am of the opinion that in recording the data displayed on the Gauging Devices and Gauge Posts at various times, the officers and/or employees of WaterNSW and its predecessors were merely recording what they directly observed as objective facts.
In respect of s 147 the EA which was relied upon by the respondent, his Honour found that the conditions of admissibility under that section were satisfied, and that the provisions of s 147(3) had no application. [135]
[26]
Ground 2(d) - The evidence of Mr Cutler
His Honour identified the general nature of Mr Cutler's evidence [136] before saying: [137]
[325] While I accept that a primary issue in this aspect of the proceedings is the reliability and use of documentation which includes the primary data in particular, despite the criticisms of Dr Martens (which are nonetheless matters that are relevant to my consideration of the overall reliability of the material), I note that Mr Cutler provided detailed evidence explaining how each of the various documents were created as well as detailing the manner in which the various gauging stations are surveyed for accuracy. Mr Cutler also gave evidence in relation to the procedures followed by WaterNSW concerning calibration of the various instruments to be relied upon.
Having noted the evidence of Mr Cutler and Mr McDermott regarding the quality codes and other procedures that applied to the collection of data: [138]
[326] Furthermore, both Mr Cutler and Mr McDermott gave evidence in relation to the various quality codes and procedures which specifically prescribed the manner in which the data is to be recorded and used. As I have noted, and will consider separately, whether the processes, procedures and methodology are sufficiently reliable to determine the flow rate at the Bourke gauge is a matter of dispute between the experts.
His Honour summarised the evidence of Mr Cutler regarding the calculation of flowrate and, [139] against that background, reached the following conclusions regarding Mr Cutler's evidence: [140]
[365] The evidence of Mr Cutler, as summarised above, details the processes and procedures undertaken by the Department to obtain and process flowrate data. Mr Cutler deposes that the Department has adopted and "consistently implemented" several NSW Office of Water hydrometric procedures relating to the collection and adjustment of river height and flow data. The hydrometric procedures referred to by Mr Cutler include, inter alia: the recording and documenting of gauging station visits; the review and editing of hydrometric data; the management and monitoring of instruments; and procedures for maintenance and calibration of current meters and pressure sensors.
[366] In relation to the reliability of the underlying data, I accept the evidence of Mr Cutler and Mr McDermott and consequently find that the data is not unreliable to the extent that the flowrate cannot be reliably ascertained as being below 4,894 ML per day during the charge period. In making this finding, I note that the combined effect of Mr Cutler and Mr McDermott's evidence demonstrates that a variety of comprehensive procedures were followed in collecting, processing and analysing flowrate data at the Bourke gauge, which, in any event, produced data with minimal uncertainty as shown by Mr McDermott's statistical assessment, which I accept. Further reasons for accepting the evidence of Mr McDermott and Mr Cutler in relation to this aspect overlap with my consideration of the submissions regarding the reliability of the rating curve and Rating Table 300.14 which follow later in the judgment.
Importantly, in reaching these conclusions, his Honour was specifically cognisant of the fact that the respondent had not called any officer who had undertaken a gauging. So much is evident from the following passage of his Honour's reasons: [141]
[385] Leaving aside what may be seen as a challenging task to call each employee who had at any time inspected and recorded each element of the primary data, questioning the ability of Department personnel to observe a gauge post, operate and calibrate equipment correctly, or otherwise prove that the equipment was indeed working correctly, would place an overwhelming evidentiary burden in cases involving technical or scientific data attained through physical observation over a sustained period of time. Despite this, I consider that the reliability of data derived from physical observations will increase when an identifiable correlation or relationship exists between the variables being observed and where that correlation is capable of statistical validation using established scientific processes that account for possible error or uncertainty. As noted above, I consider and find that the validity of any such correlation or relationship will also be dependent on whether the size of the data set is sufficient. Further, I note Mr McDermott's evidence that he compared the manual Gauge Post readings taken by Department staff on four occasions with the corresponding river height data collected by the automatic sensor, noting that the minimal discrepancies in each case (if any) were consistent with the WaterNSW drift tolerance policy and had been accounted for in his calculation of uncertainty.
[27]
Ground 2(e) - The evidence of Mr McDermott
His Honour summarised the general nature of Mr McDermott's evidence [142] and in doing so noted that Mr McDermott had: [143]
1. examined the real time data and adjusted data for the Bourke gauge;
2. assessed the margin for error;
3. explained the role of AS3778 in the calculation process; and
4. created his own rating curve and undertaken his own calculations.
In considering the issues to which Mr McDermott's evidence related, his Honour said: [144]
[363] Before turning to the discrete issues raised by the parties, I note that particular care must be taken when approaching scientific evidence in criminal trials. In assessing the reliability of Mr McDermott's assessment methodology, I am satisfied that his evidence is not of a character which introduces a risk of confirmation bias, as each step within Mr McDermott's uncertainty analysis and the creation of the McDermott Rating Curve has been premised on objective standards. For example, the calculations and process undertaken by Mr McDermott has been derived primarily from an Australian Standard which provides detailed guidance on methods to assess flowrate and measurement uncertainty. Similarly, Mr McDermott's decision to select 20 gaugings as part of his analysis was recommended by a WaterNSW Operational Procedure that indicated that this number should be considered for statistical validity.
In accepting the evidence of Mr McDermott, his Honour found that: [145]
1. an appropriate margin for error was provided in Mr McDermott's assessment of uncertainty and that his calculations took into account any deviations in gaugings data from rating table 300.14 around a flowrate of 4,894 MLD;
2. Mr McDermott conducted a comprehensive assessment of the uncertainty associated with flowrate measurement;
3. Mr McDermott specifically examined flowrates around the range of 4,894 MLD in order to prevent very high or low flowrate data affecting his calculations;
4. in any of the circumstances under which the errors identified by Dr Martens might have been introduced, Mr McDermott had quantified that total error as being not greater than 12.4% for gauging data in a flowrate of 4,894 MLD; and
5. rating table 300.14 was based on gaugings taken by a range of instruments at different times and over a substantial period of time, and that these were factors which added to its strength as a reference.
In doing so, his Honour specifically addressed the proposition that Mr McDermott had simply assumed the reliability of the data, and concluded that such a proposition significantly understated the evidence of Mr McDermott who had adopted a statistically appropriate methodology to quantify the extent of any error present in the data. [146] His Honour accepted the evidence of Mr McDermott regarding the effect of the Bourke Weir, and also accepted Mr McDermott's evidence that the movement of the Bourke gauge had no bearing on the reliability of the underlying data. [147]
His Honour's conclusions in respect of the fourth element included the following: [148]
[409] Whilst I have made separate findings above in relation to specific areas of disagreement between the experts, I consider that Mr McDermott did, as the [respondent] submits, provide clear and reasoned explanations for agreeing (and on occasion disagreeing) and included detailed reference to AS3778 and the processes and practices employed by WaterNSW and confirmed the veracity of these practices including the underlying gaugings data.
[28]
Ground 1
The primary submission advanced on behalf of the appellants in support of this ground, was that the respondent had failed to adduce any evidence to establish the underlying data upon which the calculation of the flowrate depended was reliable, and that the trial judge had erred in not accepting that to be the case, and thus in finding that the fourth element had been established. In summary, the submission of the appellants was that the trial judge had erred in concluding that the fourth element of each charge had been established beyond reasonable doubt, given that the respondent had called no direct evidence of what had been done at any gauging, and in circumstances where:
1. there was substantial scope for human and other error in the gauging process;
2. small errors could have a drastic effect on the reliability of the flow curve and rating table; and
3. the flow curve necessarily depended upon the reliability of the underlying gauging node points.
Fundamental to the appellants' case in support of ground 1 was the proposition that the flowrate of the Bourke gauge could not be reliably assessed by reference to the flow curve, in circumstances where no officer who had undertaken any gauging had been called in order to prove what had actually been done at the time of a gauging, and thus prove that the data on which the flow curve was based, was reliable.
Senior counsel submitted that the trial judge's error in this respect had arisen as a consequence of a series of factors, including the failure of Mr McDermott to address the unreliability of the data on which the flow curve depended. It was further submitted that although the gaugings were meant to be undertaken at the Bourke gauge, the limited evidence adduced by the respondent demonstrated that this had not occurred, and that gaugings had been undertaken at different locations which were significant distances away from the Bourke gauge. This, it was submitted, necessarily affected the data used to establish the flowrate.
It was further submitted that the evidence of Dr Martens had established that gauging was a complex process, and one that involved a series of steps, which included:
1. preparing an accurate survey of the cross-section of the river at the gauge;
2. measuring the flowrate at selected points in the cross-section;
3. determining how a single flowrate figure was to be settled upon, having regard to the variety of individual flowrate measurements which were taken at the cross-section; and
4. determining what river height should be identified as corresponding to the single flowrate figure that was adopted.
It was submitted that as a result of these factors, a properly conducted survey was required before any judgment could be exercised about the point at which a flowrate measurement could be taken, and that such a survey necessarily involved not just a physical inspection and measurement, but the correct use of a properly calibrated echo sounder.
It was submitted that having regard to the evidence of Dr Martens, the accuracy of any survey of the cross-section of the river depended upon a variety of factors, including:
1. the location and number of points at which measurements were taken;
2. the exercise of judgment concerning where the variations were to be charted, and those which could be ignored on the basis that they were inconsequential;
3. the correct recording of measurements; and
4. the skill, time and effort devoted by the officer in question to that particular exercise.
Senior counsel submitted that the trial judge ought to have accepted Dr Martens' opinions, and that the failure to do so was a critical error.
It was submitted that in all of these circumstances, in order to satisfy the trial judge that the rating curve was a sufficiently accurate and reliable tool on which to be satisfied that the flowrate of the river of the Bourke gauge was less than 4,894MLD, it had been necessary for the respondent to adduce evidence to establish, beyond reasonable doubt, that the data obtained from the gaugings, and upon which the flow curve depended, had been properly undertaken and could be relied upon. Senior counsel emphasised that Dr Martens had explained how even a minor error in a gaugings exercise had the capacity to lead to significant errors in the calculation of the flowrate, and had identified that it was critical to know, and to be able to assess, what had actually been done by the officers at the time of the gaugings. It was submitted that the evidence adduced by the respondent going to these issues had been confined to:
1. single page field reports which were limited in number, and which did not contain the data that the trial judge required in order to assess whether the officer in question had acted with sufficient care, and made sound judgments, so as to render the calculations reliable; and
2. a document, namely the rating curve, which recorded the flowrate based upon each of the gaugings on which the flow curve depended.
Finally, senior counsel emphasised that although the trial judge had apparently placed some emphasis on the AS3778, no evidence had been led by the respondent which demonstrated that the requirements set out in the standard were actually followed by the field officers when undertaking the gaugings, let alone complied with in a way which would render the outcome of the gaugings reliable.
[29]
Ground 2(a)
Senior counsel for the appellants submitted that the conclusions reached by the primary judge in relation to the evidence of Dr Martens encompassed four principal findings, namely that Dr Martens:
1. had engaged in speculation;
2. had done no more than raise concerns;
3. did not quantify, or qualify, the impact of the potential sources of error that he had identified; and
4. did not advance any practicable alternatives to the methodology used by the respondent's officers, with the exception of installing measuring devices at the river cross-section.
It was submitted that such findings reflected a failure on the part of the trial judge to understand, and recognise the purpose of, Dr Martens' evidence, which had been directed towards identifying the absence of evidence sufficient to establish the fourth element of each charge beyond reasonable doubt.
It was submitted, in particular, that the reasoning of the trial judge reflected a failure to appreciate the inherent unreliability of the gauging process, which was consequent upon a failure on the part of the trial judge to appreciate the significance of Dr Martens' evidence. It was submitted that Dr Martens had identified what the respondent had been required to, but did not do, in order to discharge the onus that it bore, and that rather than recognising that circumstance, the trial judge had erred by effectively reversing the onus of proof. A related error was said to stem from the trial judge's assessment of the evidence of Mr McDermott which forms the basis of ground 2(e).
Senior counsel for the appellants submitted that the evidence of Dr Martens focused upon the inherent uncertainty in the correctness of the flow curve, and the practical improbability of calculating the true flowrate at any point. It was submitted that whilst Mr McDermott had identified and employed certain techniques to address that issue, he had not purported to provide any mechanism that would serve to address the unreliability of a flow curve in the event that there had been any error in the approach taken by an individual officer when undertaking a gauging, irrespective of how that error may have arisen. It was submitted that any analysis undertaken by Mr McDermott necessarily assumed that any gauging had been undertaken diligently, in accordance with appropriate standards, and with the use of properly calibrated equipment. It was pointed out that Mr McDermott had expressly confirmed that he had assumed that the data reported by the gaugings were reliable, and had accepted the accuracy of a rating curve dependent upon the gaugings actually being performed in accordance with the required procedures.
[30]
Ground 2(b)
Senior counsel submitted that the finding of the trial judge that Dr Martens had not identified any practicable alternatives demonstrated further error because, apart from reversing the onus of proof, the question of whether or not there were practicable alternatives for measuring the flowrate was irrelevant to the determination which the trial judge was required to make.
[31]
Ground 2(c)
It was submitted that the evidence of the gauging node points, and evidence based upon that data, should have been excluded by the trial judge, either on the basis that it constituted inadmissible opinion evidence, or alternatively on discretionary grounds pursuant to s 135 of the EA.
Senior counsel submitted that the evidence amounted to unexplained statements of opinion of the respondent's officers as to the appropriate flowrate figure, and that in admitting this evidence, the trial judge had incorrectly categorised the data as objective facts.
It was further submitted that the determination of the trial judge that the evidence was admissible pursuant s 147 of the EA demonstrated a further error. It was submitted that this provision was not directed towards circumstances of the present kind, and could not operate to overcome the absence of the evidence in the respondent's case as to the reliability of the gauging node points.
[32]
Ground 2(d)
Senior counsel for the appellants submitted that the trial judge had erred in evaluating the evidence of Mr Cutler. It was submitted that, rather than recognising that the respondent had failed to adduce evidence as to what its officers had actually done at the time of performing the gaugings, the trial judge had made findings in terms which indicated that his Honour had proceeded on the basis that such evidence had, in fact, been given.
In advancing that submission, it was emphasised that Mr Cutler had unequivocally accepted that he had viewed the data only after it had been reported and put into the database and that, properly understood, Mr Cutler had done nothing more than give evidence of the respondent's documented procedures, and had said nothing about such procedures having actually been carried out. It was submitted that in these circumstances, the trial judge had overstated the evidence of Mr Cutler, and had treated his evidence as having filled a critical gap in the respondent's case.
[33]
Ground 2(e)
It was submitted that Dr Martens had identified a number of reasons to doubt the reliability of the flow curve. It was further submitted that quite apart from the absence of evidence demonstrating the correctness in the approach and decision-making of the field officers who undertook the gaugings, there were a number of supplementary reasons which gave rise to reasonable doubt as to the reliability of the flow curve as an instrument for assessing the flowrate.
[34]
Ground 1
Senior counsel for the respondent identified the primary complaint in support of ground 1 as one which centred upon an asserted failure of the respondent to prove the reliability of the underlying data upon which the flow curve depended, and an associated failure on the part of the trial judge to recognise that this meant that the charges could not be established.
It was submitted that ground 1 should be rejected because:
1. the data in the rating table had been derived from gaugings taken by a range of instruments, at different times, and at different locations;
2. any possible inaccuracy was, in these circumstances, rendered insignificant, such that the line of best fit produced by the rating curve was more accurate than any individual gauging;
3. the real time height data recorded at the Bourke gauge was regularly reviewed by way of physical observation, and that any variable in excess of 10mm was adjusted;
4. there was a strong concentration of 20 gauging points plotted on the rating curve, disclosing an essentially uniform river height in circumstances where each of the 20 points had a level of measurement uncertainty of 10% or less;
5. gaugings taken after the Bourke gauge was relocated and calibrated were consistent with the established long-term trend, such that, had the movement of the gauge affected its calibration with previous readings, the pattern of node points on the rating curve would have been disturbed, and the absence of any such disturbance confirmed its reliability;
6. the calibration and commissioning of the Bourke gauge on 7 June 2016, which was only 2 weeks prior to the commencement of the offending, rendered it highly unlikely that the gauge had deteriorated in such a short period; and
7. the evidence established that the highest recorded flowrate of the river at the Bourke gauge during the charge period was more than 1,000ML below the 4,894MLD threshold, and that for most of the relevant period it was substantially lower still.
It was submitted that a combination of these factors demonstrated that the trial judge had not erred in his conclusion that the fourth element of each offence had been established beyond reasonable doubt. This was particularly in circumstances where the entirety of the evidence relied upon by the respondent to establish the fourth element had been compiled using the appropriate equipment and applying an industry-accepted methodology, all of which was verified by regular physical examination. It was submitted that in all of these circumstances, the evidence had properly been accepted by the trial judge.
[35]
Ground 2(a)
Senior counsel for the respondent submitted that this ground of appeal overstated the effect of the evidence of Dr Martens, and mischaracterised what the respondent was required to prove in terms of the fourth element. It was emphasised that for the purposes of that element, the respondent was required to establish, beyond reasonable doubt, that water was taken when the flow of the river at the Bourke gauge was equal to, or less than, 4,894MLD, and that it had not been incumbent on the respondent to establish the actual flowrate of the river during the charge period, in circumstances where the evidence established (and the experts agreed) that was impossible to measure the precise flowrate of any section of the river.
It was submitted that the assessment of Dr Martens' evidence by the trial judge was both accurate and fair, and reflected his Honour having fully appreciated its nature and effect. Senior counsel pointed out that the trial judge had specifically referred to concerns expressed by Dr Martens about the processes used to obtain the rating table and that having carefully addressed the competing views of Dr Martens and Mr McDermott, his Honour had properly concluded that the fourth element of each offence was made out. It was submitted that in reaching this conclusion, his Honour had properly concluded that the evidence of Dr Martens had not given rise to any reasonable doubt as to the veracity of the rating table and flow curve relied upon by the respondent.
It was further submitted that nothing in the reasons of the trial judge indicated that he had reversed the onus of proof. It was submitted that, having analysed the evidence, his Honour had determined that the data was not unreliable to the point where the flowrate at the relevant time could not be ascertained. It was submitted that it was evident that the trial judge had considered the possible sources of error identified by Dr Martens, but had properly concluded that they were accounted for by the margin of error applied by Mr McDermott. Senior counsel emphasised that even when that margin of error was applied to the flowrate during the period of the alleged offending, the flowrate remained significantly below the threshold.
Finally, it was further submitted that there was nothing within his Honour's reasons which indicated a lack of understanding of the effect of Dr Martens' evidence. It was submitted that his Honour had considered that evidence, and had properly dismissed the concerns that Dr Martens had expressed. It was further submitted the trial judge had given careful reasons for preferring the evidence adduced by the respondent over that of Dr Martens, and that the rejection of his evidence did not amount to an appellable error.
[36]
Ground 2(b)
Senior counsel for the respondent submitted that the observations of the trial judge that Dr Martens had failed to identify any practicable alternatives to measuring flowrate amounted to nothing more than a recognition of the fact that Dr Martens had done little more than hypothesise as to why the data underlying the flow curve might be inaccurate. It was further submitted that any suggestion that the approach advanced by Dr Martens would have provided a more reliable means of measuring flow was properly rejected because:
1. the permanent installation of devices at various locations within the cross-section of the river at the location of the Bourke gauge was impractical, if not impossible;
2. Dr Martens had not identified the number or nature of any devices, nor the array in which they would need to be deployed, in order to achieve the result that he had proffered; and
3. the evidence of Dr Martens was subject to the important qualification that even if such an impracticable and elaborate approach were adopted, it would produce only an estimate of the total flow, and even then, would do so only if each device was always operating without error.
It was submitted that in circumstances where the evidence of Dr Martens amounted to nothing more than the formulation and expression of a series of hypotheses, there was no proper basis for a complaint that the trial judge had erred in rejecting it.
[37]
Ground 2(c)
Senior counsel for the respondent submitted that in light of the evidence of Mr McDermott, the obtaining of data in the gauging process did not involve the formation of any opinion on the part of any officer of the respondent.
It was further submitted that the unchallenged evidence of Mr Cutler as to what was done when performing a gauging, did not involve drawing any inference as to what data was to be recorded, but involved a process conducted in accordance with AS3778 and the procedures which had been described by Mr Cutler. It was submitted that to the extent that any officer in carrying out a gauging made any kind of determination, such a determination did not amount to an opinion but was simply the recording of data by reference to a method which was designed to ensure that the process was being undertaken accurately.
It was further submitted that in circumstances where there was an absence of evidence which was sufficient to raise any doubt about the data produced, s 147 of the EA facilitated the admission of the evidence, and that there were no grounds for its discretionary exclusion pursuant to s 135 of the EA.
[38]
Ground 2(d)
Senior counsel for the respondent emphasised that Mr Cutler had given evidence as to the procedures applied in gauging, to which no objection was taken, and about which he was not cross-examined. It was submitted that in these circumstances, the suggestion that the trial judge had overstated Mr Cutler's evidence by attributing qualities to it that it did not have, was incorrect, and that to the extent that Mr Cutler referred to the implementation of procedures, the trial judge was entitled to rely upon that evidence in circumstances where it had not been challenged.
[39]
Ground 2(e)
Senior counsel for the respondent submitted that the trial judge had considered the evidence of Mr McDermott on its own terms, and by reference to the evidence of Dr Martens, and had properly accepted the evidence of Mr McDermott. It was submitted that in circumstances where the appellants' true complaint was that the trial judge had preferred the evidence of Mr McDermott to that of Dr Martens, this ground disclosed no legal error and should be rejected.
[40]
Ground 1
Although the submissions advanced on behalf of the appellants encompassed a range of issues, the fundamental proposition advanced was that the trial judge erred in finding the fourth element of each offence proved, in circumstances where the respondent did not call any evidence from any person who had undertaken a gauging to establish what had in fact occurred, so as to establish whether the data underlying the rating curve was accurate.
The submissions advanced on behalf of the appellants had a tendency to simply assume that the data used in the calculations of the flowrate was unreliable in the absence of further evidence. That assumption entirely overlooks the bases on which the trial judge was satisfied of the accuracy and reliability of the data, all of which were open on the evidence.
The omission of direct evidence as to the data obtained at the gauging in each case was not fatal to the respondent's case and did not lead to a conclusion that the fourth element could not be proved. It was open to the respondent to prove the reliability of the data, and thus prove the fourth element, in other ways. In my view, the manner in which the respondent chose to do so provided a sound basis for the finding reached by the trial judge and no error has been demonstrated. It is important to recognise that the trial judge was obviously cognisant of the fact that the accuracy of the data underlying the rating table and the flow curve was an issue between the parties. So much is clear from his Honour's express references to the issue in his reasons. Further, Mr Cutler had made it plain that he did not personally participate in any of the gaugings, and that he had relied upon the data which had been obtained by those officers who did attend. His Honour expressly recognised that fact in his reasons. His Honour was also cognisant of the fact that the respondent had not called any evidence from any officer responsible for conducting a gauging. Once again, that is clear from his Honour's express reference to what would have been involved in such a process. It follows from all of these factors that his Honour was clearly seized of the issue which had arisen on the evidence, and of the respective positions of the parties in relation to that issue.
The findings of the trial judge took into account the unchallenged evidence of Mr Cutler. Based upon that evidence, his Honour had regard to the procedures which were in place for the purposes of ensuring the accuracy of the instrumentation used, and the procedures in place for producing, collating and analysing the data collected. There was no evidence to suggest that these procedures had not been consistently followed.
His Honour also took into account the fact that the data had been obtained by different instruments, at different times, and over a lengthy period, and that all of these factors added to the strength of the rating table as a reference. In addition, his Honour took into account the fact that the data had been validated by the application of a series of scientific processes that accounted for any possible error. Bearing all of these matters in mind, his Honour concluded that the data which was produced had a minimal level of uncertainty. All of those findings were open on the evidence, particularly in circumstances where they were based, in large measure, on the unchallenged evidence of Mr Cutler.
As far as the movement of the Bourke gauge was concerned, his Honour accepted the evidence of Mr McDermott. He did so in circumstances where the effect of Dr Martens' evidence in cross-examination was that had there been such an effect, he would have expected to see a different correlation of data. In those circumstances, it was open to his Honour to conclude that the absence of different data of the kind to which Dr Martens had referred, supported a conclusion that the movement of the gauge had no effect, particularly having regard to the evidence of Mr McDermott.
For all of these reasons, his Honour's finding as to the fourth element was open and this ground is not made out.
[41]
Ground 2(a)
The trial judge identified the concerns raised by Dr Martens as to the procedure which had been adopted by the respondent for the purposes of calculating the flowrate. In doing so, his Honour specifically analysed those concerns as stemming from what Dr Martens had identified as:
1. uncertainty derived from instrument error and sensitivity;
2. potential error arising from human intervention in judgment; and
3. uncertainty in the calculation and processes which had been used to produce rating table 300.14.
That preliminary analysis, without more, runs entirely contrary to the proposition upon which this ground is based, namely that his Honour "failed to understand" the effect of Dr Martens evidence. Such preliminary analysis clearly reflects an understanding of the opinions advanced by Dr Marten's and the basis of those opinions.
The consideration and assessment of Dr Martens' evidence which followed that preliminary analysis was comprehensive, and fortifies the conclusion that there was no lack of understanding on the part of the trial judge as to the effect of anything Dr Martens had said. Such analysis reflected a thorough understanding of the nature of the evidence, the propositions which were advanced by Dr Martens, and the basis on which those propositions were advanced.
His Honour's ultimate conclusions reflect the fact that, notwithstanding the issues raised by Dr Martens, his Honour remained satisfied that the fourth element had been established beyond reasonable doubt. For the reasons set out in addressing ground 1, that finding was open. To the extent that it was based upon a rejection of Dr Martens' evidence, his Honour made clear his reasons for reaching that conclusion.
Like all of the evidence in the case, the acceptance or rejection of the opinions of Dr Martens' evidence was a matter for the trial judge. There is no suggestion that his Honour's explanation for rejecting such evidence was inadequate and for the reasons previously set out, there is nothing to suggest that his Honour failed to understand the effect of anything Dr Martens said.
It follows that this ground is not made out.
[42]
Ground 2(b)
In considering this ground, it is important to bear in mind the context in which the trial judge made the impugned observations.
Dr Martens made specific reference to what he considered to be the shortcomings in the methods adopted by the respondent for the purposes of calculating the flowrate. In doing so, he raised the fact that there were measuring devices which were available to be installed and used for that purpose, but which had not been used by the respondent. He also expressed the opinion that failures in the instrumentation used by the respondent, coupled with errors by a field officer in judgment or calculation, would have the capacity to give rise to error.
It was in this context that the trial judge made reference to Dr Martens not having identified an alternative method of calculation other than the installation of the devices to which he referred. In circumstances where the observation of the trial judge stemmed directly from what had been said by Dr Martens, it could hardly be said that those observations were somehow irrelevant.
Moreover, the observation of the trial judge upon which this ground centres cannot be read and considered in a vacuum. It must be considered having regard to the entirety of his Honour's assessment of the evidence. It will be clear from my finding in respect of ground 2(a) that such assessment was comprehensive, accurate and fair.
It follows that this ground is not made out.
[43]
Ground 2(c)
There are three aspects to this ground.
The first requires a consideration of what is meant by the term "opinion". The term is not defined in s 76 of the EA, which contains the opinion rule and provides that evidence of opinion is not admissible. Exceptions to the operation of that rule are to be found in ss 77 - 79.
In RW Miller & Co Pty Limited v Krupp (Australia) Pty Ltd Giles J (as his Honour then was) considered the distinction between evidence of a fact, and evidence of opinion: [149]
The distinction between fact and opinion, and what is opinion evidence, are not particularly clear, but for present purposes I think opinion evidence can be described as evidence of a conclusion, usually judgmental or debatable, reasoned from facts.
A similar approach was taken by Lindgren J in Allstate Life Insurance Company Ltd v Australian and New Zealand Banking Group Limited (No. 5) where his Honour said: [150]
The expression 'opinion' is not defined in the Act. In the context of the general law of evidence, 'opinion' has been defined as 'an inference from of observed and communicable data'.
Those observations were cited with approval in Seltsam Pty Limited v McNeill. [151]
In Hodgson v Amcor Limited; Amcor Limited v Barnes (No. 3) [152] Vickery J concluded that evidence about the workings of a computer was evidence of a fact, and not evidence of an opinion, [153] such that the evidence of a witness describing what he personally did and observed in relation to information derived from the hard drive of a computer was not opinion evidence. [154] In reaching these conclusions, his Honour said: [155]
Importantly, [the witness] does not arrive at or form any conclusions, nor does he make any judgments by a process of reasoning from the facts that he observed.
A similar approach was taken by Kaye JA in Director of Public Prosecutions (Vic) v Iliopoulos (No. 2). [156]
I have already set out the evidence of Mr Cutler, which I again note was unchallenged, in relation to the steps undertaken for the purposes of carrying out a gauging. Those steps include a field officer (inter alia) reading the velocity and depth measurements produced by the measuring device which is inserted into the water, recording those observations and measurements in an inspection sheet and transferring the data into Hydstra.
Bearing in mind those steps, obtaining and/or recording data during and/or subsequent to a gauging is based upon the direct observations of the officer who carries out the gauging. Properly understood, it does not involve the officer forming an opinion, forming a conclusion, or making a judgment by applying a process of reasoning from facts which have been observed. For those reasons, his Honour's conclusions do not disclose error.
The second aspect of this ground concerns s 147 of the EA which is in the following terms:
147 Documents produced by processes, machines and other devices in the course of business
(1) This section applies to a document--
(a) that is produced wholly or partly by a device or process, and
(b) that is tendered by a party who asserts that, in producing the document, the device or process has produced a particular outcome.
(2) If--
(a) the document is, or was at the time it was produced, part of the records of, or kept for the purposes of, a business (whether or not the business is still in existence), and
(b) the device or process is or was at that time used for the purposes of the business,
it is presumed (unless evidence sufficient to raise doubt about the presumption is adduced) that, in producing the document on the occasion in question, the device or process produced that outcome.
(3) Subsection (2) does not apply to the contents of a document that was produced--
(a) for the purpose of conducting, or for or in contemplation of or in connection with, an Australian or overseas proceeding, or
(b) in connection with an investigation relating or leading to a criminal proceeding.
The Dictionary defines the word "document" as meaning any record of information, including:
1. anything on which there is writing;
2. anything on which there are marks, figures, symbols or perforations having a meaning for persons qualified to interpret them;
3. anything from which sounds, images or writings can be reproduced with or without the aid of anything else; or
4. a map, plan, drawing or photograph.
The Dictionary also contains the following:
References to documents
A reference in this Act to a document includes a reference to:
any part of the document; or
any copy, reproduction or duplicate of the document or of any part of the document; or
any part of such a copy, reproduction or duplicate.
By reference to the provisions of s 147, his Honour found that:
1. the river flow data stored in Hydstra for the charge period, as well as the real-time height data, real-time flow data, adjusted height data, adjusted flow data and rating table 300.14 were all "documents" within the meaning of the EA. That finding was entirely consistent with the definition of the word "document" in the Dictionary to the EA;
2. the documents referred to were based on, and therefore produced by, the process of measurements taken at the Bourke gauge. That finding was clearly open based upon the unchallenged evidence of Mr Cutler, and satisfied the requirements of s 147(1)(a).
3. the outcome set out in the documents had been produced by the gauging devices and gauge posts. Again, that finding was clearly open in light of the evidence of Mr McDermott and satisfied the requirements of s 147(1)(b).
4. the document formed part of the records of the business of the respondent and that the devices were used for the purposes of that business. Again, that finding was open on the basis of the evidence of Mr Cutler and satisfied the requirements of s 147(2)(a) and (b).
In my view:
1. the finding in (i) above was entirely consistent with the definition of "document" in the Dictionary to the EA;
2. the finding in (ii) was open on the (unchallenged) evidence of Mr Cutler and was in accordance with s 147(1)(a);
3. the finding in (iii) was again open in light of the evidence of Mr Cutler and was in accordance with s 147(1)(b); and
4. the finding in (iv) was again open in light of the evidence of Mr Cutler and satisfied the requirements of s 147(2)(a) and (b).
Further, there was no suggestion that at the time of the data being collected, there were any contemplated proceeding and accordingly, the provisions of s 147(3) did not apply.
In circumstances where all of the findings reached by the trial judge were findings upon which the operation of s 147 is predicated, there is no basis on which to conclude that provisions of that section are somehow directed to circumstances other than those in the present case.
Finally, an issue is raised as to his Honour's failure to exclude the evidence, in the exercise of his discretion, pursuant to s 135 of the EA. That section is in the following terms:
135 General discretion to exclude evidence
The court may refuse to admit evidence if its probative value is substantially outweighed by the danger that the evidence might--
(a) be unfairly prejudicial to a party, or
(b) be misleading or confusing, or
(c) cause or result in undue waste of time.
In my view, there was no identifiable basis for the evidence to be excluded in the exercise of his Honour's discretion under s 135.
Firstly, there was no element of unfairness visited upon the appellants as a consequence of the admission of the evidence. The evidence was prejudicial to the appellants only in the sense that it went to prove the respondent's case. That is not the nature of the unfair prejudice to which s 135 is directed. [157]
Secondly, in light of the evidence of Mr Cutler and Mr McDermott, the evidence was neither misleading nor confusing.
Thirdly, it was not suggested at any stage by the appellants that adducing the evidence caused an undue waste of time.
For all of these reasons this ground is not made out.
[44]
Ground 2(d)
His Honour identified the general nature of Mr Cutler's evidence as being evidence going to "the procedures in relation to the collection and recording of hydrometric data" and "the processes regarding data extraction from [the Hydstra] database". Specifically, his Honour observed that Mr Cutler had (inter alia):
1. detailed the manner in which gauging stations are surveyed for accuracy;
2. explained the procedures followed by the respondent concerning the calibration of the various instruments to be relied upon;
3. explained how various documents were created; and
4. given "mostly undisputed" evidence regarding the nature of river gauging stations, the construction of rating tables, the general use of the Hydstra database and the Hydrotel system, the generation of data and the processes and procedures undertaken by the respondent to calculate flowrate.
None of those observations suggest, in any way, that his Honour elevated Mr Cutler's evidence as evidence of what any person had actually done when carrying out a gauging. His Honour's numerous references to Mr Cutler's evidence addressing "procedures" and "processes" are completely at odds with the proposition that he did so. I am fortified in that view by the fact that in the course of his reasons, his Honour made specific reference to the fact that no officer of the respondent had been called to give evidence of what had actually occurred at a particular gauging.
For these reasons, this ground is not made out.
[45]
Ground 2(e)
There was no evidence before his Honour that the relevant procedures for the gathering of the data had not been properly followed. Mr Cutler gave evidence of the processes that are followed by the respondent. In circumstances where that evidence was unchallenged, it was open to his Honour to proceed on the basis that such procedures had been followed.
Bearing that in mind, part of the basis on which the trial judge accepted the evidence of Mr McDermott was that it represented a calculation based upon the adoption of such comprehensive procedures. It is also apparent that his Honour placed considerable emphasis on the fact that in performing his calculations, Mr McDermott took into account the fact that the calculation of a true flowrate was a practical impossibility and that he had built into his calculations an appropriate margin for error.
For these reasons this ground is not made out.
[46]
Ground 3 - The trial judge erred by finding that the [respondent] had demonstrated beyond reasonable doubt that it was a condition of the approval in question that water was prohibited from being taken when the flow in the Darling River at the Bourke gauge was equal to a less than 4,894ML/day (Alleged Term).
[47]
Ground 4 - In the course of determining that condition issue, the trial judge further erred by:
1. finding that the evidence established beyond reasonable doubt, that, in a conversation with Mr Wheatley, the appellant had consented to receiving written notice varying or introducing a term into the approval by way of email addressed to a Mr Mark Adams;
2. finding that the evidence established beyond reasonable doubt that written notice introducing the Alleged Term into the approval was given to the appellant by email dated 23 September 2015 to Mr Adams;
3. erroneously finding that the respondent had made out its case that the Alleged Term was a condition of the approval by virtue of a combination of:
1. the operation of clause 3 of Schedule 10 to the Water Management Act 2000 (NSW); and
2. the water entitlement immediately preceding the approval containing a differently worded condition of the same effect.
3. That finding was erroneous in circumstances where:
4. no such case had been advanced by the respondent;
5. the applicant had not been given the opportunity to consider and be heard on that proposed finding, and had instead conducted his defence on the basis that no such findings was available; and
6. in any event, the differently worded condition in the immediately preceding water entitlement was not to the same effect.
[48]
Ground 5 - By reason of the above errors and each of them, the trial judge erred in finding that the respondent had proved beyond reasonable doubt that the appellants' had committed an offence against s 91G(2) of the Water Management Act 2000 (NSW).
It will be apparent that these grounds overlap to a considerable degree. Ground 4 is essentially a particularisation of ground 3. Ground 5 encompasses the entirety of grounds 3 and 4 without raising any additional issue[s]. In these circumstances, the entirety of the issues raised by these grounds can be addressed and determined by addressing the matters raised in ground 4. It should also be noted that ground 4(c) will only arise for consideration if grounds 4(a) and (b) are both made out. [158]
[49]
Ground 4(a) - The evidence of Mr Wheatley
The trial judge commenced his assessment of the evidence of Mr Wheatley by making reference to his extensive cross-examination [159] before concluding: [160]
[265] Having considered Mr Wheatley's evidence and having observed him giving his evidence, I see nothing in his demeanour during cross-examination that would indicate he did not have an independent recollection of the conversation he deposed to. Although he indicated that some aspects of his memory were unclear, he steadfastly maintained that that the words which he had recorded in the conversation were, at the very least, "to the effect" of the words actually said. While I accept some of the critical commentary made by the [appellants] in relation to aspects of his evidence, I take into account the other evidence that Mr Wheatley gave, more particularly that his employment with WaterNSW since 1999 has involved water licensing and regulation matters, and, as noted in his first affidavit, such roles requiring him to consider and apply the licensing and regulatory provisions of the 1912 Act and the WM Act over many years. I take that experience and familiarity into consideration as a context and background for his conversation with [the first appellant]. He also deposed that he had known [the first appellant] since 2008.
[266] Whilst the precise words deposed to were plain, I do not consider them, in the context of the conversation, as "highly legalistic and idealised" as submitted by the [appellants] and I do not consider the conversation or the words to be artificial. It is not, as the [appellants] submit, simply a conversation between a "farmer and a non-lawyer". Although it is correct that the conversation did not appear in Mr Wheatley's earlier affidavits, I do not consider in the circumstances that that is a matter that is persuasive against the veracity of the evidence he has given.
[267] I also observed that while he was willing to make concessions in cross-examination, he was resolute and clear in that the subject conversation related specifically as to who was able to act on [the first appellant's] behalf or seek information on his behalf and the method of communication. I also accept, as submitted by the [respondent], that Mr Wheatley provided evidence that [the first appellant] nominated Mr Adams as the authorised agent of the [appellants] as well as Budvalt.
[268] I also note that he deposed in cross-examination that "in the ensuing months, Mr Adams lodged several papers on behalf of [the first appellant] and [the first appellant] had signed so I don't think there is any misunderstanding on this interpretation". In this regard, I note that there is evidence, of some relevance but not of determinative weight, of email intercourse between Mr Adams (at his email address) and Mr Campbell, a water regulation officer with the Department, which although occurring on 5 February 2016 involved the provision by email to Mr Adams, consequent upon a conversation between Mr Campbell and Mr Adams, a copy of a Notice of Determination of an application to change a water access licence. I also note that the "application holder" was "Budvalt Pty Ltd" and the "application contact" was stated to be "Budvalt Pty Ltd, Peter James Harris, Kindamindi MOREE NSW 2400". Although after any relevant date his email was used in relation to water access licences held by the [appellants] and Budvalt and it is, at least, indicative of a course of conduct.
[269] I am satisfied beyond reasonable doubt that the conversation occurred in terms to the effect of that which Mr Wheatley deposed and that both limbs of s 8(1) of the ET Act (as applicable at the relevant date) have been satisfied, as I find that at the time the information was given (being the time that the email was sent), it was reasonable in the circumstances to expect that information would be readily accessible and, as a result of my acceptance of the conversation of the evidence of Mr Wheatley, that the provision of the information to Mr Mark Adams was undertaken with the consent of [the first appellant] who, for completeness, provided his consent to that information being given by electronic communication.
[50]
Ground 4(b) - The notice to Mr Adams
His Honour found that the evidence of Mr Wheatley made it plain that he had been told by the first appellant that Mr Adams had the authority to receive relevant notices and that Mr Adams was his (i.e. the first appellant's nominated agent. [161] His Honour went on to conclude: [162]
[259] I am similarly satisfied beyond reasonable doubt that the email attached the Statement of Approval. The Court has been presented with the evidence of Mr Wheatley, who explained the operation of WaterNSW's Water Licensing System database which is capable of being (and, in the circumstances, was) used to send a "Statement" by way of email to a nominated email address.
[260] Further, I note that, first, the screenshot of the WaterNSW Water Licencing System database entry for Approval 85CA753236 shows that Approval number at the top of the page and that it contains a message stating "Please find attached an updated copy of the Statement of Approval" under a dialogue box titled "Email history"; and, second, the screenshot shows an icon beside the email information of 23 September 2015 which was described by Mr Wheatley in his affidavit as a "Statement".
[261] I also find that the incorrect reference to "23 Wed 2015" was instead a reference to 23 September 2015, on the basis of Mr Wheatley's sworn evidence clarifying that the automatic reply from "[xxxx]@pjhsons.com" was received on that same day. It is also apparent that the email address was indeed operated by Mr Adams in light of the later correspondence between that email and the Department.
His Honour subsequently concluded: [163]
[271] While it is clear that the [appellants'] solicitors wrote to the Department on 4 July 2014 in terms which provided that correspondence be forwarded to the [appellants] at a nominated PO Box address and that there was subsequent correspondence to the same nominated PO Box, I consider that the documentary history of the email to "[xxxx]@pjhsons.com" which I accept and find (as deposed to by Mr Wheatley) was properly to be dated 23 September 2015, provided by way of attachment, "an updated copy of the Statement of Approval for 85CA753236" and I accept and I find that the documentary material comprising a business record including what has been described as "screenshot from WLS entry for Approval for 85CA753236 sent to [the first appellant] on 23 September 2015" (being document II of Exhibit J), considered with the "automatic reply" document from email address of Mark Adams, satisfies me beyond reasonable doubt that the email to "[xxxx]@pjhsons.com" attached the approval. I further note that Mr Wheatley deposed that document II of Exhibit J was a "copy of the screenshot from WLS [database] entry for Approval 85CA753236" and that he had given evidence of the manner in which WaterNSW generated and stored documentation on WaterNSW's Water Licensing System database and the manner that such documents are stored and accessed.
[272] For the reasons above, I find beyond reasonable doubt that notification was given by a delegate of the Minister in sending an email with the Statement of Approval attached to Mr Adams. As this notification concerns the Statement of Approval, the effect of this finding is that [the first appellant] was notified that the (unqualified) 4894 Term had been included in the Approval and was therefore operational prior to and during the charge period.
[273] I also accept the [respondent's] submission and find beyond reasonable doubt that Mr Adams was a person duly appointed by [the first appellant] for the purpose of accepting service of documents, and that Mr Adams was notified of the Approval conditions in the circumstances. Accordingly, by operation of s 394(2) of the WM Act, both [appellants'] can therefore be considered to have been notified, noting that they are also co-occupiers of the land and co-holders of the Approval.
[51]
Ground 4(a)
It was submitted that in light of Mr Wheatley's cross-examination, there was a reasonable doubt as to whether or not the conversation given in evidence had in fact taken place. It was submitted that such a doubt emerged from the following specific aspects of his cross-examination:
1. Mr Wheatley was had been unable to ascribe the date of the conversation with any degree of specificity;
2. it was inherently improbable that Mr Wheatley could recall, with any degree of precision, the terms of his conversation, given (inter alia) the time which had elapsed since it occurred;
3. the absence of any reference by Mr Wheatley in his earlier affidavits to the conversation;
4. the inability of Mr Wheatley to recall who had made the call, and any background to the call; and
5. his various concessions and he did not have a clear recollection of the precise terms of this conversation with the first appellant.
[52]
Ground 4(b)
Senior counsel for the appellants highlighted the fact that the actual email identified by Mr Wheatley had not been tendered, that there was no evidence called by the respondent which proved precisely what (if anything) had been attached to the email, and that the data in the screen shot which was tendered, particularly as to the date, raised questions as to the accuracy of the entirety of this evidence.
It was submitted that the trial judge had failed to take into account a Jones v Dunkel [164] submission which had been made regarding the failure of the respondent to call evidence from the person who sent the email to prove precisely what had been sent, and had otherwise failed to take into account the other areas of uncertainty on which reliance was placed. It was submitted that the entirety of this evidence demonstrated that there was a reasonable doubt as to what, if anything, was attached to the email which was purportedly sent to Mr Adams.
[53]
Ground 4(a)
Senior counsel for the respondent submitted that there was no error on the part of the trial judge arising from his acceptance of the evidence of Mr Wheatley, and that it was evident that his Honour had:
1. carefully observed Mr Wheatley;
2. assessed and weighed his evidence; and
3. made a series of factual findings, including findings as to Mr Wheatley's honesty and credibility, which were clearly open, which disclosed no error, and which were not amenable to appellate review.
Senior counsel for the respondent emphasised that such findings were reached in circumstances where the first appellant had chosen to give no evidence of the conversation at all.
Senior counsel also relied on the evidence of Mr Wheatley that in the ensuing months, Mr Adams had lodged several documents on behalf of the first appellant. This, it was submitted, provided independent corroboration of Mr Wheatley's account of the conversation.
[54]
Ground 4(b)
It was submitted on behalf the respondent that there was no question on the evidence that the email of 23 September 2015 had been received by the appellants, as an automated reply had been received by the Department on the same day. It was submitted that this email expressly stated that it had attached the approval that had been created that day and that in those circumstances, the finding of the primary judge that notification of the approval was given at the appellants' verified email address did not disclose error.
[55]
Ground 4(a)
Mr Wheatley was cross-examined at considerable length before the trial judge. It will be evident from the extracts of that cross-examination that I have set out that Mr Wheatley consistently accepted that he did not have a precise recollection of the words used in his conversation with the first appellant. Equally, he consistently said that he had a clear recollection of the effect of what had been discussed. Importantly, he expressly denied the possibility that he may have misunderstood or misinterpreted, anything which had been said by the first appellant. Whilst his evidence as to the conversation was given some five years after the conversation took place, and in circumstances where it was raised for the first time in the third of his affidavits, it was not put to him at any stage of a lengthy cross-examination that his evidence was a recent invention.
In my view, nothing emerged in the course of Mr Wheatley's cross examination which tended, in any way, to erode the evidence that he had given about the conversation. His Honour had the opportunity to observe Mr Wheatley as he was giving his evidence, and used that opportunity in his evaluation of what Mr Wheatley had said. In this respect, one of the specific factors that his Honour took into account in determining that the evidence should be accepted was Mr Wheatley's preparedness to make concessions.
Mr Wheatley's acceptance of the fact that he was unable to remember some of the details surrounding the conversation did not mean that his Honour was compelled to reject his repeated statements that he had a clear recollection of the effect of what had been said. It is also important to bear in mind that in the course of cross-examination, Mr Wheatley made specific reference to the fact that, in the months following the conversation, Mr Adams had "lodged several papers on behalf of [the first appellant] that [the first appellant] had signed" as a consequence of which Mr Wheatley did not think "that there was any misunderstanding or misinterpretation". That evidence provided some independent corroboration of the fact that the conversation had taken place in the terms described by Mr Wheatley in his evidence.
His Honour addressed the evidence of Mr Wheatley in a comprehensive fashion. His acceptance of that evidence was open and none of the matters relied upon by the appellants disclose any error on the part of the trial judge.
[56]
Ground 4(b)
Necessarily, this ground relies, in part, on the error asserted in ground 4(a) which I have rejected.
Leaving aside Mr Wheatley's evidence of the conversation, he also gave evidence that he forwarded the email to Mr Adams on 23 September 2015. Although the actual email itself was not in evidence before the trial judge, the history produced by Mr Wheatley makes reference to an email being sent to the email address of Mr Adams at 1:18:16pm on Wednesday, 23 September 2015. The message contained in that email made reference to approval "85CA753236" being attached.
There was evidence of an automatic response from Mr Adams' email address at 1:17:11pm, which of course is a minute earlier than the time stipulated in Mr Wheatley's email. However in my view, given that the automatic reply was headed with a reference to the approval, nothing turns upon this temporal discrepancy. In circumstances where both emails made express reference to the same approval number and in circumstances where the reply from Mr Adams was apparently an automatic response, the only available inference is that the email was sent by Mr Wheatley, and that it generated the response from the email address of Mr Adams.
A finding that the email was sent, and that it attached the approval, was open to his Honour on the whole of the evidence. Quite apart from the matters to which I have already referred, it is not without significance that the statement of approval forwarded with the email bears, on each page, a reference to the fact that it had been printed on 23 September 2015. That, of course, is the same date as that which appears in the email correspondence. Those facts sustain an inference that the approval was sent under cover of the email.
Bearing all of these matters in mind, his Honour was entitled to conclude that the email attaching the approval was sent by Mr Wheatley to Mr Adams on 23 September 2015.
[57]
Ground 4(c)
Given the conclusions I have reached in relation to those grounds, this ground does not arise for consideration.
[58]
ORDER:
I propose the following order:
1. The appeals against conviction are dismissed.
BEECH-JONES J: I have had the benefit of reading the judgment of Bellew J. For the reasons given by his Honour as well as for the following brief observations I would dismiss the appeal.
This is an appeal under s 5AB of the Criminal Appeal Act 1912. As such it is an appeal in the strict sense, that is the appellants must demonstrate that the trial judge made an error of law or applied the wrong principle in the fact finding exercise. [165]
So far as grounds 3 to 5 are concerned and, as Bellew J demonstrates, those grounds reduce to a challenge to his Honour's acceptance of the effect of Mr Wheatley's evidence about a conversation he had with the first appellant and the subsequent sending of an email. Mr Wheatley was cross examined before his Honour and the first appellant did not give evidence. Mr Wheatley adhered to his evidence as to the effect of what was said in his conversation with the first appellant. His Honour had the benefit of observing Mr Wheatley give his evidence and accepted it. No application of an error of law, application of a wrong principle or even misuse of his Honour's position as the finder of fact was shown by the appellant. That is dispositive of grounds 4(a) and (b). It also follows, that grounds 3 and 5 fail and ground 4(c) does not arise.
In relation to grounds 1 and 2, I agree with Bellew J's analysis. The principal point made by the appellant was that his Honour relevantly erred in failing to address the possibility of error in the readings in the gauging logs from which the rating table was formulated. That was raised as a bare possibility by Dr Martens. Four points should be noted about that possibility, all of which were adverted to by the trial judge. The first point was the evidence of Mr Cutler as to the process by which the readings in the gauging logs were prepared. It may be that he did not see them being undertaken but at the very least it was not suggested that any erroneous direction was given to the staff who prepared them. Second, there was relative consistency in the readings for similar river heights produced by 595 readings over 140 years which was suggestive that, at the very least, any random errors in the gauge reading process were not material. [166] Third, there was the margin of error identified by Dr Martens in his evidence and noted by his Honour (at [99] above). Fourth, there was the very large amount by which the calculated flowrate for the period of the charge was less than the threshold specified in the licence condition, namely 4894ML/day. Throughout the period of the charge the highest recorded flowrate was 3814ML/day being the reading at midnight on 28 June 2021. That figure was 77.9% of the threshold level. In these circumstances His Honour did not err in finding beyond reasonable that the licence condition was breached.
I agree with the orders proposed by Bellew J.
[59]
Endnotes
AB 226-230; AB 231-235.
Water NSW v Harris (No.3) [2020] NSWLEC 18 at [417] (Judgment); AB 225.
Judgment at [13]; AB 96; AB 523 - 524.
Judgment at [12]; AB 96; AB 523.
Judgment at [14]; AB 96; AB 539.
Judgment at [11]; AB 95.
Judgment at [11]; AB 95; AB 320; AB 524,
Judgment at [14]; AB 96.
Judgment at [25]; AB 104.
In s 5, paragraph (b) of the definition.
AB 1034 - AB 1036.
Judgment at [27]; AB 105.
Judgment at [28]; AB 105.
Judgment at [14]; AB 96.
AB 1038.
AB 523.
Judgment at [42]; AB 109.
Judgment at [42]; AB 109.
Judgment at [42]; AB 109 - 110.
The 1912 Licence.
AB 1032.
AB 1031.
AB 1033.
AB 1040; The 1912 licence is the 8th entry in the schedule and is cross-referenced to the Approval.
AB 1393.4 - AB 1393.20.
AB 536.
AB 1041 - 1047.
AB 530.
AB 520 - 521.
AB 1048.
AB 530; 1048.
AB 530.
AB 1048A.
Judgment at [261]; AB 179.
AB 1043 - 1047.
T 1393.36 - T 1394.39..
AB1395.21 - 1396.29.
Commencing at AB 1396.46 - 1397.35.
See for example AB 1396.4; 1396.8; 1397.20.
AB 1401.45 - AB 1401.49.
AB 1402.14 - AB 1402.39.
AB 1407.31 - AB 1407.39.
AB 392.
AB 393.
AB 394.
AB 483.
AB 394.
AB 393 - 394.
Judgment at [292]; AB 187.
AB 492.
Judgment at [311]; AB 191.
AB 394.
AB 492 - 493.
AB 487.
AB 495.
AB 704 - 723.
At [51].
AB 724 - 726.
AB 906 and following.
AB 909.
AB 913.
AB 930.
AB 394.
AB 486 - 487.
AB 486 - 487.
AB 488 - 489.
AB 368 - 369.
AB 394.
AB 394.
AB 394-395.
AB 396.
AB 489. The table is reproduced at AB 699 - 703.
AB 493-494. The spreadsheet is reproduced at AB 704 - 723.
AB 497. The table is reproduced at AB 727 - 786.
Commencing with entry 676 at AB 746 and ending with entry 1251 at AB 763.
AB 487.
AB 487.
AB 1056.
AB 1351.14 - AB 1351.18.
AB 1060.
AB 1060.
AB 1060.
AB 1065 - 1066.
AB 1077.
AB 367.
AB 367.
AB 373.
AB 378.
AB 370 - 373.
AB 380.
AB 1348.42 - AB 1348.44.
AB 1348.47 - AB 1349.4.
Commencing at AB 603.
AB 609.
AB 609.
AB 609.
AB 609 - 611.
AB 611.
AB 612.
AB 613,
AB 613 - 615.
Commencing at AB 620.
AB 1353.31 - 1353.40.
AB 622.
AB 1439.3 - AB 1439.50.
AB 364.
AB 620.
AB 1348.9 - AB 1348.19.
AB 1349.6 - AB 1349.19.
AB 1350.23 - AB 1350.27.
AB 1422.34 - AB 1423.9.
AB 1423.13 - AB 1423.28.
Judgment at [89]; AB 128-129.
Judgment at [306] - [309]; AB 190 - 191.
Judgment at [290] - [311]; AB 186 - 191.
Judgment at [350]; AB 204.
Judgment at [350]; AB 204.
Judgment at [366]; AB 208.
Judgment at [370]; AB 210.
Judgment at [377] - [378]; AB 211 - 212.
Judgment at [378] - [386]; AB 212-215.
Judgment at [382] - [388]; AB 213-216.
Judgment at [402] - [408]; AB 220 - 223.
Judgment at [409] - [413]; AB 223 - 224.
Judgment at [414] - [416]; AB 224-225.
Judgment at [310]; AB 191.
Judgment at [340] - [347]; AB 201 - 203.
Judgment at [340]-[347]; AB 210 - 203.
Judgment at [386]; AB 215.
Judgment at [403]; AB 221.
Judgment at [411]; AB 233 - 224.
Judgment at [313] - [316]; AB 192-193.
Judgment at [327]; AB 196.
Judgment at [328]; AB 196 - 197.
Judgment at [329] - [332]; AB 197 - 198.
Judgment at [62]; AB 114.
Judgment at [325]; AB 196.
Judgment at [326]; AB 196.
Judgment at [338]; AB 199.
Judgment at [365] - [366]; AB 208.
Judgment at [385]; AB 214.
Judgment at [65]; AB 115; Judgment at [326]; AB 196.
Judgment at [339]; AB 199 - 201.
Judgment at [363]; AB 207.
Judgment at [377] - [386]; AB 211-215.
Judgment at [384]; AB 214.
Judgment at [406] - [408]; AB 222 - 223.
Judgment at [409]; AB 223.
(1991) 34 NSWLR 129 at 130.
(1996) 64 FCR 73 at 75; (1996) 136 ALR 627.
[2006] NSWCA 158 at [122].
[2011] VSC 272.
At [48].
At [49].
At [50].
[2016] VSC 47 at [55] - [56].
Papakosmas v The Queen (1999) 196 CLR 297; [1999] HCA 37 at [91].
At [73]; AB 21.
Judgment at [263]; AB 179.
Judgement at [265] - [269]; AB 180 - 181.
Judgment at [257] - [258]; AB 177 - 178.
Judgment at [259] - [261]; AB 178 - 179.
Judgment at [271] - [273]; AB 182 - 183.
(1959) 101 CLR 298; [1959] HCA 8.
Bulga Underground Operations Pty Ltd v Nash (2016) 93 NSWLR 338; [2016] NSWCCA 37 at [96].
See AB 644 and AB 1438 to AB 1439.
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: 09 August 2021