Budvalt Pty Ltd v The Minister for Lands and Water; Hospitality and Racing, The Minister administering the Water Management Act 2000

1. [1]

   The plaintiff filed an amended summons on 12 February 2024. It asks for the judicial review of a decision of the defendant, a NSW Government Minister, made on 17 February 2023. The decision concerns the issuing of a replacement floodplain harvesting access licence. The licence seems to have taken effect from 21 March 2023. The plaintiff wishes the decision to be set aside and returned to the Minister to be decided according to law.

2. [2]

   I note that although the licence was referred to as a replacement it was common ground that this was the first instance of the issuing of such a licence.

3. [3]

   The defendant says the attack on the decision is unfounded. The defendant however, did not oppose the granting of leave pursuant to r 59.10(2) of the Uniform Civil Procedure Rules 2005 (NSW) (UCPR). Without leave, the plaintiff would have had three months from the decision to commence the judicial review proceedings. However, in this case the parties were awaiting the decision of the NSW Court of Appeal in Ramsay v Minister for Lands and Water; Hospitality and Racing, The Minister Administering the Water Management Act 2000 [2023] NSWCA 299. This decision was important because it dictated the litigation path open to the plaintiff. It prevented the plaintiff from commencing proceedings in the Land and Environment Court and pursuing a merits-based appeal.

4. [4]

   The Court of Appeal decision was handed down on 12 December 2023. These proceedings were promptly commenced on 20 December 2023. The plaintiff’s actions are described by its solicitor, Mr Neil Archer, in his affidavit of 22 April 2025. Mr Archer points out that the “proceedings in relation to the issuing of a replacement floodplain harvesting licence by the Minister are novel” so that there is a public interest in the granting of leave.

5. [5]

   Having regard to the defendant’s position and to the public interest, I extend the time for the commencement of the proceedings to 21 December 2023.

Background

1. [6]

   The plaintiff is the registered proprietor of two properties adjacent to each other and situated within a floodplain. Generally speaking, a floodplain is “an area of low-lying ground adjacent to a river, formed mainly of river sediments and subject to flooding” (Oxford English Dictionary, online ed, accessed 28 May 2025). The river here is the Barwon-Darling River. The general area is in north-west New South Wales near the town of Moree.

2. [7]

   The two properties are called Geera and Miralwyn, respectively. They are sometimes collectively referred to as Geera/Miralwyn. For the last 30 years the properties have used water from floodplain harvesting. Floodplain harvesting “refers to the collection, extraction or impoundment of any water flowing across a designated floodplain” (WaterNSW, “What is floodplain harvesting” https://www.waternsw.com.au/customer-services/water-licensing/floodplain-harvesting accessed 28 May 2025).

3. [8]

   I will continue the background from the statement of agreed facts:

4. [9]

   After attaching aerial photographs of the two properties, the agreed facts resume:

5. [10]

   I note here that the letter of 17 February 2023 refers to a floodplain harvesting (regulated river) access licence, whereas the licence in this case is probably better described as a floodplain harvesting (unregulated river) access licence. I am not sure if this is an error in the letter, but the parties made no mention of the apparent discrepancy, so I will take the point no further.

6. [11]

   The plaintiff summarised its case in this way:

7. [12]

   Despite the simplicity of the summary, the plaintiff spread its wings far and wide to encompass as many bases for judicial review as possible. These included legal unreasonableness, jurisdictional error, jurisdictional facts and no evidence. In my view, for the reasons I give below, this is a case of jurisdictional error. The error is the defendant not complying with the statutory instructions given to the Minister on the manner in which the licence share entitlement was to be calculated.

Legislative background

1. [13]

   The core legislation is the Water Management Act 2000 (NSW) (the WMA). Equally important are the Water Management (General) Regulation 2018 (NSW) (the Regulations) and the Water Sharing Plan for the Barwon-Darling Unregulated River Water Source 2012 (the Plan).

2. [14]

   Section 4A of the WMA defines “overland flow water” as follows:

3. [15]

   Section 56(1) of the WMA states:

4. [16]

   The share component given to the plaintiff was 9,702 units. A unit is the equivalent of a megalitre (cl 39(d1) of the Plan).

5. [17]

   The category of the licence given to the plaintiff was a floodplain harvesting (unregulated river) access licence or a floodplain harvesting (regulated river) access licence (s 57(1)(k1) and (k2) of the WMA).

6. [18]

   Section 57A of the WMA specifically relates to floodplain harvesting access licenses. The section makes provision for Regulations to deal with floodplain water usage. Subsection (3) may be described as a Henry VIII clause. It states:

7. [19]

   Section 57A(5) of the WMA states:

8. [20]

   Turning to the Regulations: reg 23A defines “eligible water supply work” as “a water supply work specified in cl 23B(2)”.

9. [21]

   Regulation 23B clause (2) states:

10. [22]

    A “water supply work” is defined in the Dictionary of the WMA:

11. [23]

    The definition of water supply work is important. As will be seen, the plaintiff submitted that the share entitlement 9,702 ML was far below the capacity of its water supply work.

12. [24]

    Regulation 23C, together with reg 23I were said to be the key to the plaintiff’s case. The plaintiff had a Barwon-Darling unregulated river access licence as at 3 July 2008 (Reg 23C(1)(c)).

13. [25]

    Regulation 23C(2) states:

14. [26]

    Regulation 23F makes it mandatory for the Minister to invite submissions from the landholder and to consider those submissions. This regulation was said to confirm the submission that neither s 57A of the WMA, nor any other section or regulation, contained a deeming provision to the effect that the product of the application of a model was necessarily the share entitlement to be inserted in the licence. In short, how could there be a deeming provision if the Minister was bound to take account of submissions which could, presumably, influence his decision?

15. [27]

    Regulation 23G(1) describes the three models available to the Minister. They are:

16. [28]

    The model used in this matter was the eligible water supply works scenario model which is governed by reg 23I. Regulation 23I states:

The case put by the plaintiff

1. [29]

   The plaintiff emphasised reg 23I(2)(a) because it meant, said the plaintiff, the share entitlement was obliged to be the capacity of the plaintiff’s water supply work. Again, it was stressed that reg 23I(2) did not contain a deeming provision.

2. [30]

   In regard to the absence of a deeming provision, I was taken to the decision in Ramsay. Bell CJ stated at [65]-[67]:

3. [31]

   The plaintiff submitted that the above passages did not, as applicable to this case, mean that the product of the application of the model was necessarily to be applied. This was because submissions would be superfluous if the model was the only source of the entitlement.

4. [32]

   The plaintiff submitted that “water supply works” includes the means of taking water directly to a crop without the water going through a storage facility. This transfer could be achieved through a channel relying on gravity or assisted by a pump.

5. [33]

   There was no issue but that the correct model was used. Rather the point was that the, albeit correct, model did not produce a result reflecting what was required by reg 23I.

6. [34]

   The plaintiff’s case then moved on to the joint experts’ reports. There are two reports, dated 25 March 2025 and 13 May 2025, respectively. Mr Barma was retained by the defendant. Dr Martens was retained by the plaintiff. Both experts are well-qualified hydrologists. Their reports are not only joint but reflect their complete agreement on all matters covered in the reports.

7. [35]

   The experts were asked to give an “overview of the steps which would have been required to arrive at the Determination” and then to answer some questions. The first question was:

8. [36]

   The experts answered:

9. [37]

   FPH refers to floodplain harvesting. ED Model refers to an eligible development model. Three paragraphs between the above question and answer dominated the submissions of the parties.

10. [38]

    Paragraph 24 of the report states:

11. [39]

    CI model refers to a current infrastructure model. Paragraphs 26 and 27 list a number of assumptions made by the experts. They are:

12. [40]

    The plaintiff relied heavily on paragraph 27 because the omission of water being pumped “directly from the flood runner to crops to enable direct irrigation” resulted in a significant, although unquantifiable, reduction in the amount of water that the plaintiff was allowed to use.

13. [41]

    Marrying the assumption in paragraph 27 with reg 23I(2) led to a conclusion that the model was defective or, at least, misused. This was because leaving out water going from the flood runner to the field diminished the:

14. [42]

    This omission of substantial water volumes, submitted the plaintiff, should have been acknowledged and identified by the defendant. The consequence of not doing so, together with the non-compliant result (in terms of reg 23I) led to the judicially reviewable decision. Two further matters arose from paragraph 27.

15. [43]

    Firstly, the defendant submitted that the absence of consideration of water going from the flood runner directly to crops was no more than an assumption. Generally, if an assumption is not proved it carries no weight. However, the assumptions in paragraphs 26 and 27 were not assumptions provided to the experts, for example by the instructing solicitors. To the contrary, the report states, at paragraph 6:

16. [44]

    Then at paragraph 10:

17. [45]

    The IQQM software is the software used to construct the relevant model.

18. [46]

    I am satisfied that the assumption in paragraph 27 is not a fact that the experts were asked to assume but is rather an observation they made of the model that was used. It follows that I accept the assumption in paragraph 27 as a fact.

19. [47]

    Secondly, what is a flood runner? A flood runner is not defined in the WMA or the Regulations. However, I think its meaning is straightforward. It refers to a depression along which water flows. I agree with the plaintiff’s counsel’s description:

20. [48]

    The above description is consistent with its use in the defendant’s questionnaire and also, more specifically, in another of the defendant’s documents there is this passage:

21. [49]

    Finally, on flood runners, the dictionary to the Water Sharing Plan for the Hunter Unregulated and Alluvial Water Sources 2022, actually defines a flood runner as “a stream or part of the stream that only flows during a flood”.

22. [50]

    Further in respect of the model, the plaintiff submitted that, because the model was defective, and perhaps in any event, the Minister should have considered matters going beyond the model. This was consistent with the obligations imposed by reg 23F.

23. [51]

    As will be seen below, one of the defendant’s responses was that the plaintiff had not taken up the right to make submissions provided by reg 23F.

24. [52]

    In order to explain the extent of the ‘distance from reality’ reflected in the licence allowance of 9,702 units, the plaintiff relied upon the evidence of Mr Peter Harris. Mr Harris is a director of the plaintiff company. He has been a farmer for over 30 years and this experience has included “the farming of irrigated properties to grow crops including cotton, wheat, sorghum, chickpeas; dryland grain growing and livestock rearing of cattle and sheep”. He says that since 2003 and 2016 the properties of Miralwyn and Geera have respectively “included the cultivation of cotton and wheat”.

25. [53]

    Going through Mr Harris’s affidavit, it is evident that in his experience significantly more than 9,702 ML have been used when there has been access to floodplain water.

26. [54]

    In paragraph 6 of his affidavit Mr Harris states:

27. [55]

    I note here that objection was taken to paragraph 6 as well as a number of similar paragraphs in which Mr Harris gives estimates based on his experience. The objection was that the evidence was opinion in nature and not backed up by any reasoning process. I allowed the evidence on the basis that it was to be viewed as Mr Harris’s opinion and therefore submissions on its weight would be open to the defendant.

28. [56]

    The various estimates given by Mr Harris were not contradicted by any other evidence and bearing in mind his extensive experience in the farming industry, and in particular with irrigation, I accept his evidence.

29. [57]

    Returning to paragraph 6, Mr Harris’s opinion is that during flooding up to 500 ML per day could be transferred from a floodplain directly to the fields.

30. [58]

    In paragraph 20, Mr Harris says that he has used floodplain harvested water to irrigate fields via the “flood runners”.

31. [59]

    Mr Harris stated that floodplain water had been taken directly to fields without first going through any type of storage (paragraph 22). Mr Harris said that periods of inundation might last for up to 3 months enabling an extended period of direct irrigation (paragraphs 23 and 24).

32. [60]

    In paragraph 31 Mr Harris sets out a table showing the number of megalitres required to irrigate a summer cotton crop on an annual basis. The table was criticised because, earlier in paragraph 29, Mr Harris had said that 11-12 ML was required for each hectare. The table however applies only a multiplier of 12. I accept the point that the table should have provided the same range (11-12).

33. [61]

    Notwithstanding the use of the 12 multiplier by Mr Harris, it is obvious that the number of megalitres used well exceeds the 9,702 ML provided by the licence. This is not to say that the share component should be based on historical usage, rather that historical usage is an indicator of present capacity.

34. [62]

    From paragraph 40, Mr Harris refers to “eligible works”. Mr Harris concludes at paragraph 42:

35. [63]

    The plaintiff referred me to a letter from the defendant sent to floodplain harvesting applicants in the Border Rivers and Gwydir in April 2020. The purpose of the letter is stated to be “an opportunity to be transparent about the key information which will be used in determining your floodplain harvesting entitlement”. The style of the letter is to answer “frequently asked questions”.

36. [64]

    The first question asked is: “Is the entitlement going to be based on the long-term average diversion result shown in the letters?” The answer is:

37. [65]

    The next question is: “What do the model results mean?” The answer is:

38. [66]

    The plaintiff drew support from the just quoted answer in two ways:

    1. (1)

       The three-year average for Geera/Miralwyn was over 11,000 ML, demonstrating, without more, the deficiency of the model result of 9,702 ML. This obviously incorrect result should have been evident to the defendant.

    2. (2)

       The letter, and its explanation, are inconsistent with the Regulation, in particular reg 23I, which requires a calculation based on the landholder’s water supply works.

    3. (3)

       The reference in the first sentence of the answer to “on-farm storage/s” confirms the deficiency in the model in not taking into account direct irrigation of crops by water which does not go via a storage facility. This confirmation is consistent with the assumption made by the experts at paragraph 27 of their joint report, which I will repeat:

39. [67]

    I was then taken to another of the defendant’s documents, dated May 2022, entitled: “Building the River system model for the Barwon-Darling Valley unregulated river system”. In the chapter entitled “Irrigators” and with the subheading of “Irrigated crops, crop areas and crop water use” it is stated that the range of water used for cotton and other summer crops is 5-13 ML/ha. This was said to reinforce Mr Harris’s figure of 11-12 ML/ha. The point was made that the defendant did not seem to take into account its own estimates when deciding upon the plaintiff’s entitlement.

40. [68]

    On 24 March 2022 the defendant, through the Healthy Floodplains Review Committee, wrote to the plaintiff in response to submissions that had been made by the plaintiff. Table 4 indicates the maximum three-year average of floodplain harvesting for the relevant properties to be 13,006 ML. For five years the total is 10,274 ML. Once again, the plaintiff relied on these figures exceeding the entitlement of 9,702 ML as demonstrating the obvious incorrectness of the entitlement. In addition, and again, it was submitted that the Regulation stipulates the calculation of the capacity of the works and makes no reference to any three or five-year average.

41. [69]

    The defendant wrote to the plaintiff on 20 October 2022 stating that its “share in the resource has been determined to be” for its floodplain harvesting entitlement, a share component of 9,702. The letter includes these paragraphs:

42. [70]

    The plaintiff made the point that no correction had ever been made to the 9,702 figure “due to the plan limit compliance model”.

43. [71]

    In the final result, said the plaintiff:

44. [72]

    In respect of legal unreasonableness, I was referred to the decision of the High Court in Minister for Immigration and Border Protection v SZVFW (2018) 264 CLR 541; [2018] HCA 30 where Gageler J (as his Honour then was) stated at [53]:

45. [73]

    The plaintiff submitted that the dictates of reg 23I resulted in the defendant’s decision being “one that was not exercised within the bounds of reasonableness which is an implied limit on the statutory conferral of power”.

46. [74]

    I note here that legal unreasonableness is sometimes referred to as Wednesbury unreasonableness and that its establishment is rare (see for example Minister for Immigration and Citizenship v Li (2013) 249 CLR 332 at 337-378; [2013] HCA 18 at [113]).

The defendant’s response

1. [75]

   The defendant’s response was on a number of sometimes overlapping levels:

   1. (1)

      The plaintiff’s argument ignored the “accounting rules” that were applicable to the licence.

   2. (2)

      The Minister was required to estimate the capacity to be allowed under the licence. The Minister was not required to be precise. There was a permitted element of flexibility which included the use of averaging in order to arrive at a specific figure. There are two elements to the averaging point: firstly, is averaging permitted by the Regulations and secondly, would averaging produce a greater or lesser result than the actual share entitlement?

   3. (3)

      The plaintiff had not made submissions before the decision, in particular after the defendant’s letter of 20 October 2022, to bring the Minister’s attention to the alleged deficiency in the model. This was despite the plaintiff having notice of the forthcoming decision.

   4. (4)

      The expert evidence, even though joint and agreed, did not support the plaintiff’s assertions. In particular, the important ‘defect’ in the model was no more than an assumption. I have addressed the assumption issue above.

   5. (5)

      Another aspect of the experts’ report was its reference to storage encompassing both temporary and permanent storage.

   6. (6)

      The plaintiff’s complaint was not material to an extent that justified judicial review.

2. [76]

   I will deal with each response in turn and, also make findings about each response.

The accounting rules

1. [77]

   The starting point for the use of accounting rules is s 85 of the WMA. This section requires the Minister to create an account for each licence holder in which the licence allocation is effectively deposited.

2. [78]

   The accounting rules are in the Plan, which is a statutory instrument, essentially arrived at through s 21(c) of the WMA. The relevant rules in the Plan are in cl 42(1) and (10).

3. [79]

   The defendant’s explanation of the accounting rules evolved through the hearing from a somewhat benevolent position to one effectively involving a simple credit with an overall limit.

4. [80]

   The defendant’s initial position was that upon the granting of the licence the plaintiff’s account would be immediately credited with 5 x the plaintiff’s share entitlement. Therefore, the plaintiff would begin with an allocation of 9,702 ML x 5 (48,510 ML).

5. [81]

   Clause 42 of the Plan states:

6. [82]

   As conceded by the defendant, the manner in which the rule would actually operate is as follows. The plaintiff’s share entitlement is 9,702, which is the plaintiff’s allocation in ML in the year in which the licence is granted. Some examples will illustrate the working of the accounting rules:

   1. (1)

      assume in the first year the plaintiff does not use any of the 9,702 ML. Then in the second year the plaintiff will keep that allocation and also receive the new allocation, giving it a total in its account for the second year of 19,404 ML; or

   2. (2)

      assume in the first year the plaintiff uses 5,000 ML. In the second year the plaintiff will receive its annual entitlement of 9,702 ML but also have added to that figure the 4,702 ML not used in the previous year, making a total of 14,404 ML;

   3. (3)

      the same position will continue with each successive year so that the plaintiff will receive another 9,702 ML to go into its account with whatever it has not used in the previous year or years; and

   4. (4)

      the same accounting will continue until the plaintiff, if it be the case, has 5 x its entitlement in its account. Accordingly, the plaintiff’s account can never exceed 48,510 ML.

7. [83]

   The manner in which the rules operate can be seen as the accrual of the licence holder’s asset (the permitted entitlement) into its account which had been created pursuant to s 85 of the WMA, the limit of the asset being 5 x the annual entitlement.

8. [84]

   The accounting rule is therefore not any kind of defence to the share entitlement of 9,702 because the rule is really no more than permitting a carried-over credit from a previous year up to the 48,510 ML limit. Had the defendant’s initial position been correct, so that the plaintiff started off with 48,510 ML, then the plaintiff’s complaint would have been significantly diminished. The plaintiff would have had a starting point well in excess of the licence share entitlement of 9,702.

Estimates, flexibility, and averaging

1. [85]

   The argument is put in this way. Regulation 23G dictates the type of models to be used by the Minister. The defendant pointed out that the Regulations do not specify a “description of the model, the objectives for the model, the matters the model represents, the data the model relies on.” However, the defendant is required to publish these matters on the Departments website. There is no evidence of such publication.

2. [86]

   Then going to reg 23C(2), this regulation requires the use of the three models to determine the share component for the licence. Then, reg 23I requires the determination of the “estimated volume of water”.

3. [87]

   The defendant submitted:

4. [88]

   In final submissions the defendant put the point in this way:

5. [89]

   The defendant’s submission is that reg 23I inherently allows a degree of flexibility. But reg 23I, and in particular the word “estimate”, is in my view, being misconstrued. The estimate being referred to is simply a calculation. In other words, the Minister is being asked to calculate a volume of water. The calculation is to be of a capacity as set out in the Regulation. This is a precise calculation. In my view it does not permit the flexibility, or ‘around about,’ as submitted by the defendant.

6. [90]

   The defendant also submitted, in support of its flexibility argument, that reg 23I(2) did not prescribe a time period. The Minister was not required to calculate a volume over a specified length of time, such as a year or more than a year. Consequently, the Minister was permitted to make the calculation using averages over extended periods of time.

7. [91]

   The defendant submitted that the 9,702 ML should be seen against a background of average use which would create a result much closer to the allocated amount. Although the entitlement is on an annual basis, senior counsel submitted:

8. [92]

   The period over which the averaging occurred was 5 years. The first difficulty for the defendant is that the Regulations make no allowance for averaging. Regulation 23I(2) states the relevant model is to be used “to determine the estimated volume of water capable of being used by …” the relevant eligible water supply works. There is no suggestion of averaging being used to determine capacity. As stated by plaintiff’s counsel when addressing the accounting rules:

9. [93]

   The defendant made the point that pursuant to s 56(2) of the WMA, the share component of a licence could be expressed as “a specified number of units”. This was the case here, but I do not see the significance of the point. As already noted, cl 39(d1) of the Plan equates a unit to a megalitre so that ultimately, as is also permitted by s 56(2) of the WMA, the allowance becomes a “specified maximum volume over a specified period”, in this case the period being a year. I do not see that expressing the share entitlement in units allows averaging to calculate the number of units.

10. [94]

    The second difficulty is that 9,702 ML is not a calculated average. The five-year average “endorsed by the Healthy Floodplains Review Committee” is actually 10,274 ML. The committee was a part of the Department of Planning, Industry and Environment. Another example of averaging, this time on a three-year basis, is to be found in a letter from the Healthy Floodplains Project Manager dated 14 May 2020, where it is stated that the maximum three-year average of total floodplain harvesting is 11,733 ML. Notably the graph included in Figure 1 of the same document shows a capacity in one particular year of floodplain harvesting to be above 30,000 ML.

11. [95]

    In summary, I do not think the Regulations allows for averaging and, in any event, averaging produces a result above the 9,702 share entitlement. It does not save the defendant’s argument to say the 9,702 is an estimate. It may have been estimated, but the instruction is to estimate a capacity not a broadly estimated or average capacity. In addition, I think the term of the share entitlement (over one year), is contrary to the use of an average. The share entitlement is not for an average of 9,702 over, say, three or five years.

Failure to make submissions

1. [96]

   The defendant submitted that reg 23F gave the plaintiff ample opportunity to contest the forthcoming entitlement. The defendant submitted that the plaintiff did not take advantage of correspondence from the defendant, including invitations to comment on its proposed entitlement, so that it could not now be heard to complain about the entitlement.

2. [97]

   The defendant made much of the letter to the plaintiff dated 20 October 2022. I referred to this letter when dealing with the plaintiff’s submissions, above. This letter specifically says that the plaintiff’s floodplain harvesting entitlement would have 9,702 share components. There was no issue that the plaintiff did not respond to this letter prior to being informed, on 17 February 2023, that it was to receive the 9,702 units.

3. [98]

   The letter of 20 October 2022 specifically invites the plaintiff to make a submission if the plaintiff is of the view that “the department has made an error or omission in the application of the licence determination process or with respect to the Regulations”.

4. [99]

   It is unfortunate that the plaintiff did not make a submission, but I do not think its failure is fatal to my conclusion that the model used by the defendant omitted a vital component. I also go back to paragraph 65 in Ramsay. The issue here is not about a discretion which might have been influenced by submissions. It is simply about the correct application of the Regulation.

5. [100]

   There are some important parts to the 20 October 2022 letter that I think support the plaintiff’s case, some of which I have already quoted, for example, under the heading of Draft Floodplain Harvesting Entitlement the defendant wrote:

6. [101]

   The terms of the just quoted sentence highlight the entitlement being worked out through “the farm scale validation process” and the account management rules. As already seen, the account management rules do not affect the initial entitlement but rather the accrual of credits from one year to the next. There is no suggestion of averaging or broad estimation. The target is “the capacity of your eligible works.”

7. [102]

   The letter continues:

8. [103]

   As I read the above sentence, the capacity of the eligible works remains the measure by which the entitlement is to be calculated. Nothing about the Barwon-Darling Valley requires any manipulation or reduction of the entitlement based on the capacity of the eligible works.

9. [104]

   The letter sets out the accounting rules. For the Barwon-Darling Valley this is the accrual provision that has been described above up to the x 5 limit. The letter states:

10. [105]

    I note that 47,561 ML is not 9,702 ML x 5.

11. [106]

    In relation to the submissions on authorities made by the plaintiff, the defendant emphasised the difficulty of achieving success in an assertion of legal unreasonableness. I was referred to paragraph 52 in SZVFW:

12. [107]

    The failure to make submissions was also picked up by the defendant in its submissions concerning Ramsay. Bell CJ, said at [65]:

13. [108]

    And the at [67]:

14. [109]

    The result, submitted the defendant, was that the plaintiff could not go beyond the allotted share component. The defendant was not able to adjust a conclusion reached by the appropriate model. The defendant submitted that reg 23F allowed for submissions to be made “to tell the Department where it has gone wrong”, but once the determination had been made “there is no capacity for the decisionmaker to do anything else”.

15. [110]

    The plaintiff sought to distinguish Ramsay in this way:

16. [111]

    I think there is a distinction with Ramsay, although I would express it in a different way. The current case, at least to the extent that I agree with the plaintiff, is one concerning the application of a model to determine the plaintiff’s entitlement. If the model is defective, that in turn implicates the defendant’s decision on entitlements, not as a matter of discretion in the manner considered in Ramsay, but rather as an expression of the defendant’s failure to exercise the jurisdiction imposed by the Regulations, and in particular reg 23I.

17. [112]

    Put another way, reg 23I tells the defendant what must be done. In this case, because the model left out an important element, namely consideration of direct irrigation to crops, the defendant did not do what the Regulations prescribed. This is the basis for judicial review, not the exercise of any discretion by the defendant.

Various definitions

1. [113]

   The defendant submitted that storage, as used in the model, included flood runners. Before looking at this submission I think it necessary to deal with an opposing concept, albeit one which could favour the defendant’s case. The definition of an “eligible water supply work” in the Regulations is a “water supply work specified in clause 23B(2)”.

2. [114]

   Regulation 23I refers to “eligible water supply works”. This could mean that while flood runners are unquestionably water supply works (as defined in the dictionary of the WMA), they are not necessarily ‘eligible’ unless they fall within one of the descriptors in clause 23B(2). On this basis it could be argued that flood runners are correctly excluded from the model, in particular those relying on gravity to move the water from the floodplain to the field.

3. [115]

   Mr Harris however states in his affidavit:

4. [116]

   I am satisfied, based on Mr Harris’s description, and the forwarding of this information to the defendant in 2014, that the flood runners fall within eligible supply works under reg 23I.

5. [117]

   Turning now to the storage covered by the model, the defendant submitted that the model covered both temporary and permanent storage and that the temporary storage encompassed the flood runners. Essentially the defendant was seeking to contradict the expert report which, to again repeat para 27, states:

6. [118]

   The defendant did make a last-minute attempt to qualify the experts’ opinion but failed (see separate judgment). I agree with counsel’s frank admission:

7. [119]

   The defendant submitted that, in any event, the experts were including direct irrigation because “Mr Harris’s direct irrigation scenario involves diversion into a temporary storage that channels.” I reject this submission because it so plainly contradicts para 27 in the experts’ report.

8. [120]

   In relation to para 24 of the experts’ report, the defendant submitted again, that the model treated storage as including water in the flood runners. But this is not what para 24 says. Firstly, para 24 is clear in referring to water going to storage (temporary or permanent) but not moving along a flood runner, and secondly such an interpretation would directly contradict cl 27.

Materiality

1. [121]

   In his affidavit Mr Harris states:

2. [122]

   Taking these uncontradicted figures as facts, the difference down to 9,702 ML indicates a very substantial decrease in the required use of flood water. The difference makes the plaintiff’s complaint very material. Even if one compares the 9,702 ML to the averaging figures described above, in the order of 11,000 or 12,000 ML, the difference is still material, representing close to a 20 or 25% reduction.

3. [123]

   Accordingly, I disagree that the plaintiff’s complaint is not material.

Further issues

1. [124]

   I think I have dealt with the defendant’s responses, generally not accepting them. There are still some submissions made by the plaintiff to be considered.

2. [125]

   The plaintiff endeavoured to base its entitlement to judicial review on a number of bases, including jurisdictional error, legal unreasonableness, a no evidence ground and a jurisdictional facts ground.

3. [126]

   There may well be a degree of overlap, but I think the present case is one of jurisdictional error as described in Plaintiff M64-2015 v Minister for Immigration and Border Protection (2015) 258 CLR 173; [2015] HCA 50 from [23]-[25]:

4. [127]

   To be clear, the jurisdictional error is the failure to produce the result required by reg 23I. The defendant was required to estimate a capacity as defined in the Regulation. By omitting an integral part of the capacity, the jurisdiction was not exercised. This is a type of error described by Mason J in Minister for Aboriginal Affairs v Peko-Wallsend Ltd (1986) 162 CLR 24 at 39-41; [1986] HCA 40:

5. [128]

   I have included the whole of the propositions set out by Mason J, but emphasise that in the current case the defendant was, by virtue of the Regulations, bound to take into account the dictates of reg 23I, which did not occur. I have already pointed out the power of the Regulations, as stipulated in s 57A(3).

6. [129]

   I agree with the defendant that the no evidence ground should be rejected. As pointed out by the defendant, it is only necessary that there be a skerrick of evidence. That is simply not the case here. I accept the defendant’s submission that the Department “carefully, at each point in the process, sought to consult and obtain submissions from people in the position of the plaintiff”. This is confirmed by the correspondence with the plaintiff as well as all of the questionnaires and fact sheets distributed to potential licence holders.

7. [130]

   The steps taken by the defendant, described by Mr Geoffrey Cameron in his affidavit of 15 May 2025 at paragraphs 54-58 were:

8. [131]

   Later, from paragraph 78, Mr Cameron describes the collection of more information from landholders and then from paragraph 105 describes consultations about the Plan.

9. [132]

   In relation to jurisdictional facts, the plaintiff relied on the judgment of Spigelman CJ in Timbarra Protection Coalition Inc v Ross Mining NL (1999) 46 NSWLR 55; [1999] NSWCA 8, in particular at [40]-[41]:

10. [133]

    Again, I do not think this is a case of inquiring into the reasonableness of the defendant’s decision. This is a simple identification of the defendant not acting in accordance with its jurisdiction, namely the path dictated by the Regulations.

11. [134]

    The result of the above is that the defendant’s decision must be set aside and the matter remitted to the defendant requiring the defendant to assess the plaintiff’s entitlement as dictated by the Regulations.

Orders

1. [135]

   I make the following orders:

   1. (1)

      The time for commencement of the proceedings is extended to 21 December 2023.

   2. (2)

      The decision of the defendant made on 17 February 2023 being a determination of the share component for the Access Licence 44722 in the amount of 9,702 units is set aside.

   3. (3)

      The Geera/Miralwyn Access Licence is remitted to the defendant for determination in accordance with law.

   4. (4)

      The defendant is to pay the plaintiff’s costs of the proceedings.

   5. (5)

      The parties have leave to seek an alternative costs order within 14 days of these orders.