HCAL184/2002

HCAL184/2002

由此 A A HCAL184/2002 B B IN THE HIGH COURT OF THE C HONG KONG SPECIAL ADMINISTRATIVE REGION C COURT OF FIRST INSTANCE D D CONSTITUTIONAL AND ADMINISTRATIVE LAW LIST E NO.184 OF 2002 E F --------------------- F BETWEEN G G SHIU WING STEEL LIMITED Applicant H H I and I J J DIRECTOR OF ENVIRONMENTAL PROTECTION Respondent K AIRPORT AUTHORITY OF HONG KONG Interested Party K L ---------------------- L M Before : Hon Burrell J in Court M Dates of Hearing : 5, 8-11, 15-17 September 2003 N N Date of Judgment : 30 September 2003 O O ------------------------- P JUDGMENT P ------------------------- Q Q 1. The Hong Kong Airport Authority (“HKAA”) needs a new R R permanent air fuel farm (“PAFF”). During the late 1990’s many possible S sites were considered. The result was that a site known as Tuen Mun 38 S was chosen as the most suitable. Accordingly, they commenced the T T numerous procedures necessary to embark on such a project. If all the U U V V 由此 - 2 - A A procedures are successfully completed they have estimated that the PAFF B B would be operational by 2006. C C 2. One of the many procedures is compliance with Cap.499, the D D Environmental Impact Assessment Ordinance (“EIAO”). The EIAO sets E out a number of steps to be followed by the proposed developer. As part E of their purported compliance with the EIAO the Director of F F Environmental Protection (“the Director”) has made decisions which have G permitted the continuation of the project. Two of his decisions are now G the subject of judicial review by the applicant (“SWS”) in these H H proceedings. The first decision was made on 2 August 2002. That was I a decision to approve an EIA report submitted by the HKAA in relation to I J the PAFF for the Airport under section 8(3) of the EIAO. The second J was on 28 August 2002. This was the decision under section 10(2) to K K grant an Environmental Permit with conditions (pursuant to section 10(5)) L to the HKAA. L M M STATUTORY FRAMEWORK N N 3. The EIAO is a relatively new, relatively short but very O important ordinance. It is clear from an overview of the whole enactment O that it has a dual objective. The “overview” reveals a number of steps P P that a developer must take, each of them has a specified time limit. The Q result is a fairly tight timetable to which both the developer and the Q director must conform. The dual objectives strike a balance. The R R EIAO’s primary concern is the protection of the environment. Its S framework provides a regime in which, and this is the second objective, S important projects may be completed in a timely and efficient way, T T provided all the requirements of the EIAO are complied with. The U U V V 由此 - 3 - A A Tribunal’s report in the recent “Long Valley” appeal (KCR Corp. v. B B Director of Environmental Protection; EIA Appeal Board 2 of 2000) C helpfully commented as follows : C D “… The Ordinance gives the EIA process a legal structure. It is D not a process which lies comfortably within a detailed legal framework. Much of its success depends therefore upon the E E manner in which it is implemented within the outline structure provided assisted by the Technical Memorandum. The F consequence is that all involved are learning how best the F various steps required can be implemented. … There are two main matters of public interest involved. Both are important. G The first is the public interest in the protection of the G environment upon which the quality of life in Hong Kong will H increasingly depend. The second is the public interest in H ensuring that major designated projects are brought to fruition in a timely and efficient manner. The time constraints put upon I the Director for steps in the process and for his decisions show I that the Ordinance aims to satisfy both interests.” J J K 4. The key provisions in the ordinance which demonstrate this K step by step approach and dual objectives are these : L L (1) By section 4 certain projects (such as the PAFF) become M “designated projects”. M N (2) By section 5 the developer applies to the Director for a study N brief (“SB”) and at the same time submits and advertises a O project profile. The Director also informs the Advisory O Council on the Environment (“ACE”). ACE is an P P independent advisory committee comprising 23 members. It Q is a sort of watch-dog and is peculiar to Hong Kong in Q environment impact regimes. ACE and the public have R R 14 days within which it may comment on the project profile. The Director must issue the SB within 45 days. S S (3) By section 6 HKAA prepares an EIA report in accordance T T with the SB (section 6(1)) and delivers it to the Director. U U V V 由此 - 4 - A A Within 60 days thereafter the Director decides if the report B B meets the requirements of the SB and the technical C memorandum (“TM”) (section 6(3)). If he does, the report is C then made available for a 30-day public inspection. D D (4) By section 7(5) ACE may give its comments within 60 days. E E (5) By section 8(3), within 30 days after the public inspection the F Director either approves, or approves conditionally or rejects F the EIA report. G G (6) By section 10 HKAA may apply for the environmental permit H for the project which the Director may grant (under H section 10(2)). I I (7) By section 16 the Secretary for the Environment Transport J and Works may issue TM setting out principles, procedures, J guidelines, requirements and criteria for whether the EIA K report meets the requirements of the SB. K L L 5. In the present case SWS’s key complaint is that the Director’s M M decision to approve the report (section 8(3)) and grant the permit (section N 10(2)) were either unlawful and/or Wednesbury unreasonable because the N report did not meet the requirements of the SB and TM. SWS in this O O framework has no right of appeal against the Director’s decision. Only a P party to the project itself has such a right. The SB and the TM are key P documents. The TM is the framework of criteria out of which an SB is Q Q created. The SB is the template of objectives and obligations against R which the EIA report is prepared. They will both be considered in more R detail later in this judgment. S S T T U U V V 由此 - 5 - A A BACKGROUND B B 6. The background to the issues in this judicial review is best C C understood by reference to a chronology of key events. D D 7. In April 2001 HKAA applied with a project profile for an SB. E E On 31 May 2001 the Director issued the SB and HKAA engaged Mouchel F to prepare the EIA report (in June 2001). The report took nearly a year to F produce. It is 480 pages long. The “Hazard to Life” section, with which G G this judicial review is concerned, runs to 55 pages. H H 8. The report was submitted to the Director for his consideration I I under section 6 on 3 May 2002. It had been signed as a complete report J on 26 April 2002. The day before, 25 April, an important meeting had J taken place between representatives of the Director and HKAA. At this K K meeting a decision was made that one of the particular hazards to life that L the report had identified, namely an instantaneous loss of 100% of the L contents of a fuel tank, need not be quantitatively assessed. It was M M decided that it was sufficient to assess such a scenario qualitatively. This N N decision lies at the heart of this judicial review. O O 9. In short, a quantitative assessment requires a statistical P P analysis. It results in a numerical representation of the likelihood of Q deaths occurring from a particular type of accident occurring and at what Q frequency and in what numbers. A qualitative analysis, on the other hand, R R is more subjective. It does not require an input of statistical data. It is S an assessment made by experts in the field. It is the report’s failure to S carry out a quantitative assessment (“QRA”) of the 100% instant loss T T scenario upon which SWS relies as the lynchpin to their case. They U U V V 由此 - 6 - A A submit that failure rendered the report deficient so that the Director’s B B approval of it was unlawful. C C 10. During May, June and July the various provisions of the D D EIAO were complied with. The report entered the public domain. SWS E wrote its letter of complaint on 13 July 2002. The key extracts from that E letter, which summarize SWS’s opposition to the project which pre-date F F the Director’s decision are as follows : G G “‘… SWS believes that the EIA report on the proposed PAFF has not addressed adequately the impact of the PAFF on the Mill as H H an existing user of a high temperature operation.’ I ‘Spillage I The effects of a major spillage incident on water at both the J Castle Peak Power Station in the proximity and the Mill would J be catastrophic. Both facilities use seawater intakes for cool-down process, but no reference is made in the EIA report as K K to how the location of the jetty, the discharge of the fuel and anchoring of tankers will affect these. Besides, the impact of an L oil spill on adjacent industries have not been considered in L respect of the seawater intake.’ M ‘Risk of Fire M In the EIA report, several references are made to the potential N N emissions from both the transfer (working losses) and storage (standing storage losses) of the aviation fuel. Yet the total O volume of the likely emissions is not discussed. The more O important shortcoming is that the EIA Report has not made references to the fact that the Mill is sited immediately adjacent P to the proposed PAFF and will engage in processes which P o involve very high temperatures (up to 1300 C). The siting of a Q facility which discharges aviation fuel vapours immediately Q adjacent to a facility of very high temperature processes and furnace operations is unsafe. In this case, any spillage incident R would be catastrophic.’” R S S T T U U V V 由此 - 7 - A A 11. ACE met to consider the report and all public comments on B B 29 July 2002. It declared that it was satisfied that the risk was within C acceptable levels and endorsed the report. C D D 12. On 2 August 2002 the Director approved the report under E section 8(3). The section 10(2) permit was granted on 28 August, with E conditions. F F G 13. At some stage during this process SWS had commissioned its G own report. That report (the “Maclnnis” report) is dated 16 October 2002. H H It was sent to the Director on 21 October. The accompanying letter I summarizes the report’s conclusions as follows : I J “(1) The EIA Report does not meet the requirements of the J Technical Memorandum. K K (2) All reasonable hazardous scenarios have not been considered, and certain assumptions/methodologies used in L the EIA are questionable and/or inaccurate. L (3) Construction of the Proposed PAFF at Area 38 may produce M hazardous conditions that would result in a greater hazard M to life than intended by the EIA Ordinance. N N The points raised have been put to the Town Planning Board and the Airport Authority. Given the nature of the risks Shiu Wing O have decided that it would be in the interest of all concerned to O make the report available in its entirety. The Airport Authority have been asked to confirm that it has thought of these risks and P P that it has adequate measures in place to safeguard safety. Q Safety Q As stated above, this is obviously a very special situation, and R strict compliance with all required procedure and technical R memoranda is vital. Shiu Wing have no wish to be obstructive, but we clearly have a major interest in the outcome of your S decision. The Macinnis Report confirms that the proximity of S the Steel Mill’s operations is not discussed in the EIA report but T may be of importance if there was a fuel spill inside or outside T the bund area, and that there are a number of possible ignition U U V V 由此 - 8 - A A sources which have not been considered. In all the B circumstances, you should not hesitate to suspend, vary or cancel B the permit if you are in any doubt regarding the information previously provided, or the safety of the project. C C Notwithstanding the importance of the project, we would suggest that you cannot be too cautious in these circumstances.” D D E 14. After a brief exchange of correspondence between the various E parties SWS issued its notice of application for judicial review on F F 1 November 2002. Leave was granted by Hartmann J on 13 November. G G 15. Since the granting of leave further reports from all sides have H H been prepared. In response to Maclnnis the HKAA filed a report by AEA I Technology (“AEAT report”) on 5 February 2003, and the Director filed a I report from Shell Global Solutions (“Shell”) on 12 February. AEAT J J concluded that : K K “(a) None of the issues raised by Macinnis identify errors or omissions in the EIA Report which AEAT consider would L significantly affect the predicted risk levels, which lie L within the acceptable region of the criteria established in M Annex 4 of the Technical Memorandum; … M (d) AEAT consider that the EIA Report identifies and discusses N all reasonable hazardous scenarios associated with the N PAFF. In particular, AEAT do not consider it was necessary for the EIA Report to consider specifically the O O scenarios suggested by Macinnis in which : (a) hot surfaces within the Steel Mill could ignite a release from P the PAFF; (b) smoke carrying inflammable vapours could P ignite on reaching the Steel Mill; and (c) hot gases emitted from the Steel Mill could lead to ignition of Jet A-1 at the Q PAFF. AEAT consider that these potential scenarios are Q reasonably covered within the general scenarios adopted R by the EIA Report; and R (e) AEAT consider that the assumptions and methodologies S used in the EIA Report are reasonable within the scope of S the assessment. In particular, AEAT consider that the thermal impact criterion used in the EIA Report is a T T reasonable approximation for the PAFF, which will not U U V V 由此 - 9 - A A significantly affect the predicted risk levels compared B with the criteria established in Annex 4 of the Technical B Memorandum, when compared to other reasonable calculation methods.’ C C Based on this review, AEAT consider that none of the issues D identified by Macinnis would change the key conclusion in D paragraph 10.9.7 of the EIA Report, namely that ‘Based on the analysis presented in this section, including the E recommendations, it can be concluded that the offsite individual E and societal risks posed by the activities at the PAFF tank farm and associated marine environment are acceptable according to F F the criteria set out in Annex 4 of the EIAO-TM (ie the Technical Memorandum).” G G H 16. Shell concluded that : H I “The EIA for the PAFF at Tuen Mun Area 38 has been I challenged for not taking sufficient account of the neighbouring Shiu Wing Steel Mill. Although the presence of the mill and its J workforce is specifically mentioned, Macinnis claim that the J effect modelling described in the EIA is not sufficiently K conservative, and that worse case scenarios deemed incredible by K the authors of the EIA should have been included. L It has been shown in this report that the extra scenarios L mentioned by Macinnis are indeed incredible, and that the scenarios and assumptions in the EIA Report are comprehensive M M and robust. The worst-case scenario addressed in the EIA (tank overfill) is the worst foreseeable, and was modelled N conservatively in that the assumed damage was significantly N worse than is observed in practice. The hazard to life assessment in the EIA remains sufficient, and retains its validity. O O Even if the extra hypothetical scenarios postulated by Macinnis P after the EIA was published were to be included as a sensitivity P study, they do not pose a risk to the people in the steel mill or to its structure, and the total risk for the PAFF is still acceptable.” Q Q R 17. SWS were clearly not satisfied with these conclusions. As a R result, their notice of application was re-amended and a further report S S filed — the “HSL” report (Health & Safety Laboratory, a much respected T T U U V V 由此 - 10 - A A UK organization) written by Mr Marc McBride. A summary of that B B report’s findings are : C C (1) The Hazard Assessment is fundamentally flawed because it D fails to identify and assess the risk associated with a D catastrophic failure of one of the aviation fuel storage tanks, E E i.e. instantaneous loss of 100% of the tank contents. This is contrary to common practice for Hazard Assessment studies F F in Hong Kong and published international good practice. G Furthermore, there have been past occurrences of catastrophic G tank failure relevant to the PAFF. H H (2) The Hazard Assessment fails to comply with the requirements I of the EIA Study Brief because it does not identify all I hazardous scenarios, in particular the catastrophic tank failure J J scenario. K (3) HSL’s assessment is that this scenario could cause almost K 200 fatalities at the SWS site, due to the significant fraction of L L aviation fuel which would overtop the bund (as demonstrated M by theoretical models, large scale experimental studies and M past incidents), together with the strong likelihood of ignition N N associated with activities at the steel plant. O (4) The Hazard Assessment grossly underestimates the risk O associated with the proposed PAFF. Using the figure P P contained in section 10.5.2 of the Hazard Assessment on the frequency of catastrophic tank failure, the overall risk for the Q Q PAFF (2040 case) is calculated to lie with the upper ‘ALARP’ R region of the Risk Guidelines, rather than in the “acceptable” R region as the EIA report would suppose (ALARP stands for S S As Low As Reasonably Practicable). T T U U V V 由此 - 11 - A A (5) There is a significant off-site risk associated with the B B proposed PAFF. Compliance only with minimum separation C distances does not provide an acceptable basis for the siting of C the PAFF. Good practice would require that the decision as D D to its siting be risk-based, having regard to the location and E surroundings of the installation, irrespective of separation E distances. F F (6) The proposed PAFF design does not comply with relevant G international standards, namely the Institute of Petroleum G Model Code of Safe Practice, Part 2 : Design, Construction H H and Operation of Distribution Facilities. I I SWS acknowledge that (1) and (2) are inaccurate. The EIA report did J identify the 100% instantaneous loss scenario and it did assess it. What it J did not do is assess it quantitatively. The report had concluded that the K K worst case scenario was a 10% loss when it said “The catastrophic failure L of the tank is therefore assumed to result in a release of 10% of tank L contents (i.e. about 3,900 m3) on to the bund”. M M N 18. There is no complaint that the assessment of the “10% worst N case scenario” was not fully addressed in considerable detail in the report. O O The conclusion that the QRA be confined to the 10% scenario is preceded P by a wealth of scientific data. P Q Q 19. I have set out the conclusions to these reports for two reasons. R Firstly, to demonstrate one of the few things about which there is R agreement between the parties in this case, namely that there is a S S difference of opinion between the experts. (In passing, it is also agreed T that the scientific or technical debate between the experts cannot be and is T U U V V 由此 - 12 - A A not to be resolved by the court. The parties are unanimous that the court B B should not get involved with the debate between the experts). Secondly, C the filing of these reports (together with two affirmations on behalf of C SWS from Mr T. Maylor and Libor Rostik which provide, inter alia, D D evidence about the workings of the steel mill) prompted a preliminary E hearing which was heard in this court on 24 July 2003 in which SWS E applied for leave to re-amend the notice of application and file the HSL F F report and the two affirmations. The application was opposed. At the G end of a relatively short hearing, which only scratched the surface of the G issues and the reports, I granted leave to re-amend and admitted the fresh H H evidence relied on by SWS de bene esse. One reason for admitting it I de bene esse was to give the court an opportunity at the full hearing to I J consider the report and its relevance in greater depth. This has now J happened and I consider the matter more fully later in this judgment. K K L 20. Throughout the above events, many affirmations and L affidavits have been filed. 12 deponents have filed 18 affirmations M M totalling 177 pages. N N THE ISSUES O O 21. SWS on the one hand and the Director and HKAA on the P P other do not agree what the issues to be resolved by the court are. The Q prime reasons for the disagreement are that : Q R (a) there is no consensus as to the legal boundaries set by judicial R review proceedings. SWS submit they are changing and S widening, S T T U U V V 由此 - 13 - A A (b) the admissibility of the HSL report remains an issue. B B Without it the issues are narrower. C C D 22. Defining the issues is a highly significant exercise. In some D ways it goes a long way towards determining the outcome of the judicial E E review itself. F F 23. Erring on the safe side all counsel have addressed the court on G G all issues, that is, those advanced by SWS. Counsel’s submissions have H been comprehensive. Mr Charles Haddon-Cave, QC and H Mr Nigel Pleming, QC leading Mr Anthony Chan, SC have appeared for I I SWS. The Director has been represented by Mr Benjamin Yu, SC J leading Mr Anthony Ismail and Ms Gladys Li, SC has represented HKAA. J Their instructive and helpful skeleton submissions run to 250 pages. The K K research of all three legal teams has been impressive. Whether the court L proceeds therefore on the wide list of issues contended for by SWS or the L more conservative list proposed by the Director is a very important M M decision. Mr Haddon-Cave’s list of issues and consequential issues pose N 24 questions to be answered. Mr Yu lists five issues. N O O 24. As will be seen from what follows in this judgment, and after P anxious deliberation, I have preferred the narrower approach submitted by P the respondent and HKAA. I will not, therefore, set out all the issues as Q Q advanced by SWS. Mr Pleming, an acknowledged and respected expert R in administrative law commenced his submissions as follows : R S “ This is, I hope your Lordship will accept, a very important S case for environmental law in Hong Kong. Your Lordship’s T judgment will no doubt significantly assist in a number of ways, T apart from assisting the particular parties. U U V V 由此 - 14 - A A Could I respectfully suggest what those ways might be. B First of all, it will be the first opportunity for the court in Hong B Kong to express its view as to the meaning, if it has a view, and can express one, on the language and the purpose -- and C C I emphasise the purpose -- of the ordinance and the technical memorandum. D D The second area of significance is that your Lordship’s judgment will provide an opportunity for the determination of E the true role and expectations of the public. By ‘the public’ E I also include the advisory council as the representative of the public, in what Mr Haddon-Cave has described so far as, and F F I here add as part of my submissions, a tripartite process. G The third area of importance is that your Lordship’s G judgment may be a step in the direction of determining the standards to be applied in the approval process. I cover by that H all stages in the ordinance, from compliance through approval to H permit, the degree of rigidity or laxity that is involved.” I I J 25. The view of this court is that first instance judgments should J be primarily, if not solely, directed towards the decision which the parties K K want to know. Should that decision necessitate matters of statutory L construction or applications of principle then so be it. However, such L determinations should be confined to only those which are necessary for M M the court to carry out its primary task which in this case is to apply the N Hong Kong law of judicial review to the two decisions under challenge N and decide if any grounds have been made out to review them. O O P 26. It seems to me therefore that the respondent’s issues better P reflect the task of this court in this case. They are : Q Q (1) Whether the evidence in the form of expert reports are R R admissible in judicial review proceedings by way of exception S under R. v. Secretary of State for the Environment ex parte S Powis [1981] 1 WLR 584? T T U U V V 由此 - 15 - A A (2) Whether the Director acted lawfully in approving the EIA B B report and allowing the project proponent not to carry out a C quantitative risk assessment on instantaneous release of 100% C of a tank content? D D (3) Whether the Director acted reasonably in the public law sense E in approving the EIA report and allowing the project E proponent not to carry out a quantitative risk assessment on F F instantaneous release of 100% of a tank content? G (4) Whether the challenge that the EIA report does not comply G with the SB and the TM is out of time? H H (5) Whether the court in the exercise of discretion should grant I relief having regard to : I J (a) delay; J (b) prejudice to the Airport Authority; K K (c) prejudice to good administration. L L M 27. It will be noted that the first issue concerns the admissibility M of the HSL report. I therefore deal with this as a preliminary issue next. N N Before doing so I should add that SWS’s list of issues has not been O shredded. On the contrary, they remain a useful guide to structure this O judgment. The 24 questions have been helpfully grouped under the P P six heads of challenge contained in the re-amended notice of application. Q I propose to deal with each of those challenges separately, although in Q many instances, briefly. Before, therefore, I turn to the important R R question of the admissibility of the HSL report, it is convenient to set out, S in summary form, SWS’s six heads of challenge. S T T U U V V 由此 - 16 - A A SWS’S GROUNDS OF CHALLENGE B B 28. SWS’s case is that the EIA report is deficient in that it failed C C to comply with the legal requirement of EIAO, in particular the D requirements of the SB or TM. If that proposition is correct the D Director’s approval of it, under section 8(3) was unlawful because it was E E not a proper or valid report under section 6. For the same reason the F section 10(2) permit was not lawfully granted either. F G G 29. Six separate grounds are identified. In truth, grounds 2 to 6 H flow from ground 1. Ground 1 is the crux of the matter. Ground 1 H depends on how the ordinance is construed. Counsel for SWS I I acknowledge that there is considerable overlap between the various J grounds. In spite of submissions to the contrary I am not persuaded that J any of grounds 2 to 6 can be freestanding. K K (a) Ground 1 is described as the Quantitative Risk Assessment L L argument (“the QRA argument”). Paragraph 3.3.10 of the M SB states : M “3.3.10 Hazard To Life N N 3.3.10.1 The risk to the life, including the workers of nearby plants, due to marine transport, jetty transfer, tank farm O storage and pipeline transfer of aviation fuel shall be O assessed. The Applicant shall follow the criteria for P evaluating hazard to life as stated in Annexes 4 and 22 P of the TM in conducting hazard assessment and include the following in the assessment: Q Q (i) identification of all hazardous scenarios associated with the marine transport, jetty transfer, tank farm R storage and pipeline transfer of aviation fuel, which R may cause fatalities; S (ii) execution of a Quantitative Risk Assessment S expressing population risks in both individual and societal term; T T U U V V 由此 - 17 - A A (iii) comparison of individual and societal risks with the B Criteria for Evaluating Hazard to Life stipulated in B Annex 4 of the TM; and C (iv) identification and assessment of practicable and cost C effective risk mitigation measures as appropriate.” D D In short, having identified a 100% instantaneous loss as a E E scenario under (i), they were required to execute a QRA in F respect of it. Once identified the TM required that a QRA F must be done in respect of all identified hazards. G G (b) Ground 2 is described as the Material Facts argument. It is H submitted that prior to making his decision the Director failed H to acquaint himself with the nature of the works at the steel I mill and the number of people who worked there. I J (c) Ground 3 is the Applicant’s letter argument. It is submitted J that the Director disregarded SWS’s letter of complaint dated K K 13 July 2002. L (d) Ground 4 is the 25 April 2002 meeting argument. The L decision not to carry out a QRA in respect of 100% M M instantaneous loss was taken at a meeting between the Director and HKAA just prior to releasing the report. It is N N submitted that the effect of this decision was to remove a key O issue, namely 100% catastrophic failure, from public debate O and consideration. P P (e) Ground 5 is the Safeguards argument. It is submitted that Q the Director should not have relied on the various safeguards Q that were incorporated into the design and operation of the R R PAFF when deciding not to do a QRA for the 100% loss S scenario. S (f) Finally ground 6 is the Conditions argument. It is submitted T T that a condition which was imposed on the granting of the U U V V 由此 - 18 - A A permit (“Permit Condition 3.9”) was unlawful as it fell B B outside Schedule 4 of the EIAO. C C D 30. Before considering these grounds in more detail, I now turn to D the preliminary issue of the admissibility of the HSL report. E E F THE HSL REPORT — ADMISSIBILITY F G 31. The quality and expertise of the HSL report, together with G independence and excellent reputation of HSL itself is not doubted. H H I 32. The report has been admitted de bene esse. The court has I now had a full opportunity (as opposed to a superficial opportunity at the J J preliminary hearing on 23 July) to consider the purpose and content of the K report in the light of issues to be determined. K L L 33. Counsel for SWS submit that the report and the fresh M evidence is admissible as an exception to the general rule. Authorities M and learned texts have been cited in support of the general proposition that N N the circumstances in which post decision materials can be admitted is O widening. Bearing in mind the importance and complexity of this case, it O is submitted that the court should lean in favour of including it rather than P P excluding it. Q Q 34. The general rule can be found in R. v. Secretary of State for R R the Environment ex parte Powis [1981] 1 WLR at p.595 : S S “Finally there was an application on behalf of the tenant to admit fresh evidence which the Divisional Court had refused to admit. T T Like the Divisional Court we considered the evidence de bene U U V V 由此 - 19 - A A esse. What are the principles on which fresh evidence should B be admitted on judicial review? They are (1) that the court can B receive evidence to show what material was before the minister or inferior tribunal : … (2) where the jurisdiction of the minister C C or inferior tribunal depends on a question of fact or where the question is whether essential procedural requirements were D observed, the court may receive and consider additional evidence D to determine the jurisdictional fact or procedural error : … and (3) where the proceedings are tainted by misconduct on the part E of the minister or member of the inferior tribunal or the parties E before it. Examples of such misconduct are bias by the F decision making body, or fraud or perjury by a party. In each F case fresh evidence is admissible to prove the particular misconduct alleged : …” G G H 35. There are many examples of the application of the general H rule in Hong Kong. For example in Nguyen Ho v. Director of I I Immigration [1991] 1 HKLR 576, Cons VP said : J J “From that exposition alone I am satisfied that unreasonableness as a factor by itself, what might be termed the second limb in K Wednesbury, can only be judged with regard to what was known K to the inferior tribunal at the time.” L L 36. In Yu Chee Yin v. Commissioner of ICAC [2001] 2 HKC 91 at M M 104, Hartmann J said : N N “It is common cause that the evidence contained in the affirmations was not before the Commissioner when he made his O decision. Indeed, that is the foundation of the applicant’s O third challenge. But a claim of unreasonableness in the P ‘Wednesbury’ sense is judged in light of the information that was P before the decision maker when he made the decision. It is not judged in the light of information that has only become known Q after the event.” Q R R 37. SWS submits there are four reasons to treat the HSL report as S an exception to the general rule. I consider it sufficient to simply recite S T T U U V V 由此 - 20 - A A those four reasons without reviewing the many authorities to which I was B B referred in the course of submissions. C C 38. First, where it is alleged that the decision maker failed to take D D account of a relevant fact, evidence can be admitted to show (a) why it is E relevant and (b) the significance of the disregarded material in the E decision-making process. F F G 39. Second, this case falls within the second head of Powis : G Where the jurisdiction of the decision maker depends on a question of fact H H evidence is admissible to challenge the fact and to show that he had no I jurisdiction or authority to do what he did. I J J 40. Third, there is an “emerging” category for admission of K K material not before the decision-maker. Where a decision is made in L ignorance of or contrary to the true facts, evidence is admissible to L demonstrate the true facts upon which the decision should have been based, M M provided that it is shown that the facts placed before or found by the N decision maker were “plainly wrong”. N O O 41. Fourth, where it is alleged that the defendant has failed in its P duty to acquaint itself with relevant information, it will often be essential P for the applicant to demonstrate (in a case such as this by reference to both Q Q factual and expert evidence) that there was material with which the R decision maker should have been acquainted at the time of the relevant R decisions but was not. Here, that category covers the true factual S S situation at the steel mill described by Maylor and Rostik. T T U U V V 由此 - 21 - A A 42. In short it is argued that HSL establishes that a QRA on the B B 100% loss scenario could have been carried out, but was not, and had it C been the Director’s decision could have been different. C D D 43. I have considered the authorities relied on by SWS and in E particular the passage from Fordham Administrative Law under the E heading “Fresh evidence in Judicial Review”. For the reasons which F F follow I have come to the conclusion that the report is not admissible G because the grounds for categorizing it as an admissible exception to the G general rule have not been made out. Any apparent inconsistencies H H between the reasons for excluding it now and the reasons for admitting it I de bene esse on 23 July stem, quite simply, from the full nature of the I J argument in this hearing compared with the brief nature of the argument in J the former hearing. It is worth mentioning also that SWS’s primary K K argument, namely that the section 8(3) decision was unlawful because the L EIA report was deficient is a question of construction and remains in tact L with or without HSL. M M N 44. I agree with the general argument advanced by the Director N and HKAA that, in reality, the HSL report is an expert opinion in support O O of SWS’s latest case (I deal later with the point that SWS’s latest case is P significantly different from its original case upon which leave for judicial P review was granted). As such, it merely reinforces the undoubted truth Q Q that expert opinions differ. The fact that a QRA was not being done on R the 100% loss scenario was known from 3 May 2002. The alleged R deficiency, which is the crux of the HSL report, could have been placed S S before the Director before he made his decision but was not. SWS submit, T as they must do, that the report contains fact not opinion which should T U U V V 由此 - 22 - A A have been before the Director. To separate fact from opinion in this B B particular report is almost impossible. Taken as a whole its conclusions C plainly attract the label of opinion rather than fact. In its own summary C the report states that its purpose is to establish whether the Hazard D D Assessment complies with the requirements of the SB and TM. The E author gives his opinion on this issue. That question was the question the E Director had to decide. It was his judgment call. Whether or not that F F decision was made at the time unlawfully or irrationally is for the court to G determine. The answer to that question does not require a consideration G of the subsequent competing opinions as to whether the 100% loss H H scenario is credible or not, whether a QRA on such a scenario is feasible or I necessary or not, or whether the result of such a QRA, had it been done I J would have been acceptable or not. It does not constitute “true facts upon J which the decision should have been made” nor does it contain matters K K which “the decision maker should have acquainted himself with, but did L not.” L M M 45. Having applied the basic legal principles to the issue and N having decided, as a result, that HSL is not admissible, further analysis of N the report becomes unnecessary. O O P 46. Two final points require mention. Firstly, one of SWS’s P arguments is that because no challenge was made to the Maclnnis report, Q Q which also post dates the Director’s decision, the respondent cannot now R challenge other post decision material. I disagree for two reasons. As a R statement of principle it is wrong. Simply because they did not challenge S S one report does not act as a waiver of a right to challenge subsequent T reports. Also, the Maclnnis report did contain factual matters which were T U U V V 由此 - 23 - A A relevant to SWS’s original complaint as set out in their 13 July letter. B B Secondly, it was submitted at the preliminary hearing that the HSL report C merely confirms and enlarges upon the Maclnnis report. Having now C considered them both I think this understates the true position. I agree D D with Ms Li counsel for the Interested Party that a fairer reflection on the E matter is that Maclnnis on its own, in the light of the AEAT and Shell E reports could not establish that the Director’s decision was either unlawful F F or irrational. SWS therefore side-lined Maclnnis. HSL focused on the G matter differently, namely on the failure to execute a QRA on the 100% G loss scenario. SWS’s focus shifted accordingly. H H I THE CONSTRUCTION ARGUMENT I J 47. I turn now to the competing submissions on how the EIAO is J to be constructed in answer to the question — was it mandatory for the K K EIA report to contain a QRA on all hazards to life that had been identified? L L M 48. Where the meaning and effect of a statutory provision is the M subject of legal argument the court should construe the provision N N purposively. The court’s interpretation should be consistent with the O objects of the legislation as a whole. I have already set out what the dual O objectives of the EIAO are. The key provisions which are under scrutiny P P are sections 6(1), 6(3), 8(3) and 10(2). They should be read together as Q integral parts of an overall regime. Q R R 49. Section 6(1) requires the EIA report to be “in accordance S with” the SB and TM. Then, section 6(3) requires the Director to decide S if the report “meets the requirements” of the SB and TM or not. T T U U V V 由此 - 24 - A A 50. The section 6(3) decision comes before sections 8 and 10. It B B is a decision to be made by the Director at a time when there is no public C involvement. It has been argued by SWS that the whole EIA regime is a C tripartite process involving the government, the developer and the public. D D In fact, the regime is not tripartite throughout. At the time the section 6(3) E decision is made the first public window has closed (advertising the project E profile) and the second public window has not yet opened (public F F comment on the signed EIA report). G G 51. This is perhaps an appropriate moment to deal briefly with the H H complaint about the 25 April meeting, which immediately preceded the I section 6(3) decision, at which it was decided that the 100% loss scenario I J did not warrant a QRA. It was suggested that this somehow J circumvented public involvement. In my judgment this complaint does K K not stand up. The decision taken at the 25 April meeting was a L clarification of, not a dilution or variation of, the criteria. The EIAO L provides that the report is prepared at a non public involvement stage in M M the regime. Any decision about its content will be without public N consultation. The public consultation may later affect its approval but not N its substance. After the section 6(3) decision the absence of a QRA on a O O 100% loss scenario is open to debate and comment. In support of SWS’s P case that an important part of the section 6(3) decision cannot be P “surrendered to a private agreement” Mr Pleming cited two authorities Q Q from the UK (Gillespie v. First Secretary of State and Bellway Urban R Renewal Southern [2003] EWCA Civ 400 and Smith v. Secretary of State R for the Environment [2003] EWCA Civ 262.) Whilst these cases are S S instructive, their particular facts are wholly different and the UK statutory T regime about which they are concerned is also very different. T U U V V 由此 - 25 - A A 52. Like all decisions, section 6(3) requires judgment. Whether B B the answer is “Yes” or “No” does not depend on a robotic or check-list C approach. Whether or not it “meets the requirements” of the SB and TM C naturally depends on what the requirements of the SB and TM are. Those D D documents are not to be read as a strict list of things which must be done E before it can be decided that its requirements have been met. It is now E necessary to consider those two documents in a little detail. F F G The Study Brief G H 53. Section 3.3.10 has been set out at page 16 of this judgment. H The key parts are that when conducting a Hazard Assessment the applicant I I shall “follow the criteria for evaluating hazard to life as stated in J Annexes 4 and 22 of the TM”. Included in the assessment is to be J (a) identification of all hazard scenarios and (b) execution of a quantitative K K risk assessment. L L 54. SWS submit that the TM must be followed rigidly and M M completely and that the word all in (a) should be read into (b) as well. N N O 55. I do not agree. O P P 56. The SB requires an “assessment”. An assessment is defined Q as “an evaluation or an estimate of the nature of”. A QRA is a tool to be Q used in making the assessment. Section 3.3.10.1(ii) does not require a R R QRA for all hazardous scenarios. It is sufficient to execute a QRA for all S those scenarios, which in the Director’s judgment, need to be addressed S and assessed. Mr Yu points out that the next section of the SB, T T section 3.3.11 “Risk Assessment on Fuel Spillage” deals with the point U U V V 由此 - 26 - A A specifically. (i) says “identification of all fuel spillage scenarios … and B B (ii) says “quantification of the impacts … due to fuel spillage scenarios C identified in (i)”. Section 3.3.10 could very easily have said the same C thing but did not. Mr Haddon-Cave submits that sections 3.3.10 and D D 3.3.11 differ because section 3.3.11 does not have the same preamble as E section 3.3.10 (in which the TM is specifically mentioned). This is E correct but, in my judgment, does not diminish the point that if all F F scenarios in section 3.3.10 had to be quantitatively assessed it would have G said so. (The reliance on the TM as being the document which says that G all hazards require a QRA is dealt with in the following section.) H H I 57. To require a QRA for all scenarios would include all I J incredible scenarios and scenarios which it is known will not occur. For J illustrative purposes it is permissible to dip into the HSL report to support K K the Director’s decision that this cannot be the requirement. HSL accepts L that certain hazards can be eliminated qualitatively, such as those L described as “boil over” and “brittle failure” because they have no possible M M relevance to this project in this location. Moreover, HSL does not explain N why 100% instantaneous loss scenario could be a credible event. It does N not answer the conclusions in the Shell and AEAT reports that it is not O O credible for a Jet A-l storage tank to split from top to bottom when one P takes into account the actual design features and standards of construction P and testing being applied in this particular project. In short, HSL says, it Q Q would be possible to do a QRA for the 100% loss scenario but does not go R on to say that such a scenario is a credible event. R S S 58. In the course of argument reference has been made to to T SRAG — the UK Safety Report Assessment Guide. I refer to it to extract T U U V V 由此 - 27 - A A one passage only. The entire document has only a limited value in the B B present case because it relates to Potentially Hazardous Installations (PHIs) C whereas the PAFF is not so designated and also it relates to “Highly C Flammable Liquids” whereas Jet A-1 fuel in the PAFF is not classified as D D highly flammable. The passage which, in my judgment, describes a E correct approach in our case is : E F “Proportionality will influence the type and level of analysis F detail that Assessors might expect to underpin the various demonstrations in the safety report. …. It is important for G G Assessors to realise that QRA does not mean that a detailed and full numerical analysis resulting in iso-risk contours and F/N H societal risk curves is needed. Rather the extent of the H quantification and the form it takes will depend on the site specific circumstances determining the level of proportionality I that applies.” I J In other words, it is expected that the decision of expert assessors will be J K founded on professionalism and common sense. K L L 59. SWS’s answer to the question — why should a QRA be done M for all scenarios is because the TM says so. I move on therefore to a M consideration of the TM. N N O Technical Memorandum O P 60. The first point to establish about the TM (and indeed, the SB P also) is that it is not legislation. Section 16(12) of EIAO makes this Q Q clear : R R “16. Technical Memorandum …. S (12) A technical memorandum is not subsidiary S legislation.” T T U U V V 由此 - 28 - A A 61. Mr Pleming’s submission advances theories as to why that B B particular subsection may have been included. He submitted that at the C very least the TM had legal effect. He said that the TM certainly looked C like subsidiary legislation. In my judgment, the reason section 16(12) D D appears is plain. It is to put the matter beyond doubt lest anyone (here E SWS) might think otherwise. E F F 62. The fact that it is not subsidiary legislation provides the G starting point for how it is to be treated. It is a guide. The TM itself, at G paragraph 1.2.2 says so : H H “The Technical Memorandum on the Environmental Impact I Assessment Process is a guide for the Director in deciding on I matters under sections 5, 6, 8, 10, 12, 13 and 14 of the Ordinance. J The Director is the Director of Environmental Protection. He J will follow this technical memorandum as appropriate according to the circumstances of a case.” (emphasis added) K K Words and phrases such as those emphasized do not appear in legislation. L L There are many other examples which illustrate that, whilst an important M document, it provides for a degree of flexibility and common sense. It is M not a straightjacket to be slavishly and mechanically followed, in which N N discretion and judgment play no part. O O 63. Other examples of the language of the TM providing a pointer P P to how it should be used can be found in part 4 which sets out the criteria Q for the EIA report. Q R R 64. 4.1.1 states : S S “An EIA report shall comprise a document or series of documents providing a detailed assessment in quantitative terms, T T wherever possible, and in qualitative terms of the likely U U V V 由此 - 29 - A A environmental impacts and environmental benefits of the project. B The requirements for the EIA report shall be set out in B accordance with this technical memorandum. The EIA report shall be produced in accordance with the EIA study brief issued C C by the Director to the applicant.” D D 65. SWS relies on the words “wherever possible”. They submit E E that as a QRA on the 100% loss scenario was possible (in support of which F they rely on HSL) it should have been done. F G G 66. However, this must be read in context. The whole sentence H suggests that both quantitative and qualitative assessments will be done. H They will be done in respect of the likely environmental impacts. The I I word “likely” appears in the next section also (objectives and contents of J the report) : J K “to identify and describe the elements of the community and K environment likely to be affected by the proposed project(s), and/or likely to cause adverse impacts to the proposed project(s), L including both the natural and man-made environment and the L associated environmental constraints;…” (emphasis added) M M N 67. The same section reminds the compiler of the report of the N purposes and objectives of the report : O O “The contents of an EIA report shall fully meet the purposes and P objectives set out in the EIA study brief issued by the Director, P and shall adequately address all the issues set out in the EIA study brief.” Q Q R 68. The general approaches and methodologies are set out in R section 4.3. SWS point to 4.3.1(vi) : S S “predicting the likely nature, extent and magnitude of the T anticipated changes and effects such that an evaluation, in T U U V V 由此 - 30 - A A quantitative terms as far as possible, can be made with respect to B the criteria described in Annexes 4 to 10 inclusive.” (emphasis B added) C C 69. This should be read together with the section dealing with the D D review of the report (section 4.4) which again illustrates that the reviewer E E of the report is permitted to use his judgment subject to the guidance set F out : F “Quality of the EIA Report: The quality of the EIA report shall G G be reviewed having regard to the guidelines in Annex 20 and in Section 4.3. The report shall be considered as adequate if there H are no omissions or deficiencies identified which may affect the H results and conclusions of the assessment. …” I I 70. Both counsel have helpfully referred the court to several J J authorities on the issue of the construction of the TM. I refer to just four K and very briefly. The purpose is not to either follow or distinguish a K particular decision but to borrow some apposite phrases which illustrate L L the correct approach to how the TM guide should be applied. M M N 71. R. v. Rochdale Metropolitan Borough Council ex parte Tew N [1999] 3 PLR 74 was a planning case in which Sullivan J held that the O O regulation under discussion should be interpreted purposively rather than P literally : P “ While regulation 3 of the applications regulations is Q Q expressed in mandatory terms — an application for planning permission ‘shall’ include the particulars specified in the form R provided by the local planning authority — I do not accept that R failure to include one or more of those particulars means that the application is invalid, so that the local planning authority is S unable to grant planning permission even if it wishes to do so. S The mandatory terms of regulation 3 are addressed to a practical T problem and are intended to ensure that the local planning T U U V V 由此 - 31 - A A authority is able to require that certain particulars are provided as B part of the application for planning permission. ….” B C C 72. He later went on to say that to interpret in any other way D would be “an unduly literal reading of the regulation”. D E E 73. Similarly, the conclusions reached by Mackay J in R. v. F Dorset County Council [2002] All ER (D) 68, another planning decision, F are to point : G G “The adequacy of environmental information contained in an H H Environmental Statement is a matter for the judgement of the planning authority, with which the Court will only interfere if it I is proved to have been exercised irrationally. I …. J J It is not ‘every scrap of information’ which has to be considered …. (a flexible attitude to compliance with a given K K paragraph’s requirement may be appropriate, …. L I was not in the result persuaded by the decisions that Part II L paragraphs 1, 2 and 3 have to be approached in the rather mechanistic, sequential and discrete way contended for by M Mr Fordham ….” (emphasis added) M N N 74. The same theme is expressed by Jowitt J in R. v. Wakefield O Metropolitan District Council ex parte Pearl Assurance plc (QBD) 1997 : O The document under scrutiny was a planning policy guide (PPG). P P Jowitt J said : Q Q “…. PPGs are not delegated legislation and do not have the force of statute. They are guidance and not tramlines. They do not R R purport to deal definitively with every situation which may arise. …. S S …. Whether a piece of guidance amounts as a matter of law to a material consideration has to be judged by reference to the T content of the guidance seen in the factual context of the T U U V V 由此 - 32 - A A particular case. Moreover, there is an important distinction B between having regard to guidance and being bound to follow B it.” (emphasis added) C C and quoting Lord Hoffman from an earlier decision : D D “This distinction between whether something is a material consideration and the weight which it should be given is only E E one aspect of a fundamental principle of British planning law, namely that the courts are concerned only with the legality of the F decision-making process and not with the merits of the decision. F If there is one principle of planning law more firmly settled than any other it is that matters of planning judgement are within the G exclusive province of the local planning authority or the G Secretary of State.” H H I 75. Adopting such an approach to the present case makes SWS’s I task an uphill one. In my judgment, it is simply not sensible to extract the J J two words “wherever possible” from paragraph 4.1.1 of the TM and K conclude that because HSL said that a QRA on the 100% loss scenario was K “possible”, then the failure to carry it out renders the entire process L L unlawful. That would be “unduly literal”, “mechanistic” and not in M accordance with the objectives and purposes of the ordinance. Rather, it M should be read in its context in 4.1.1 which anticipates both quantitative N N and qualitative assessments as part of the guidance. O O 76. Of course, this does not mean that the TM or any other similar P P guideline is to be treated in a cavalier way. It is to be applied responsibly Q and with care, but not unduly literally and with blinkers. Failure to Q comply with every requirement will not automatically render it invalid (as R R submitted by SWS). The court’s decision in the present case is that the S particular non-compliance alleged is not to be construed as a S non-compliance and thus the complaint that it was an invalid report fails in T T limine. What remains therefore is a possible challenge that the decision U U V V 由此 - 33 - A A not to assess the 100% instantaneous loss scenario quantitatively was B B Wednesbury unreasonable. C C Annex 4 D D 77. Before considering the case under “Wednesbury E E unreasonableness”, Annex 4 should not go unmentioned (although, absent F HSL, its significance is diminished). Annex 4 provides the criteria for F carrying out a QRA should it be deemed appropriate. Annex 4 states G G “The criterion for hazard to human life is to meet the Risk Guideline as H shown in Figure 1. Figure 1 is in two parts : H I (1) Individual risk guideline for acceptable risk levels. I J Maximum level of off site individual risk should not exceed J one in 100,000 per year, i.e. 1 x 10-5/year. K K (2) Societal risk guidelines for acceptable risk levels. L The societal risk is in the form of a graph. The horizontal L axis is the number of fatalities (N) from one to 10,000. The M M vertical axis is the “frequency of accidents with N or more N fatalities per year with one in a billion at the bottom and one N in a hundred at the top. Roughly speaking the graph has O three regions. If, statistically, a risk is in the top right hand O region it is “unacceptable”, if it is the bottom left it is P P “acceptable”. If it is in a band between the two it is in the Q “ALARP” region, namely “as low as reasonably practicable”. Q This means that an ALARP risk should be mitigated to as low R R as reasonably practicable. The QRA done in the EIA report was all in the “acceptable” region. HSL’s assessments S S entered into the ALARP region. Thus no risks, even those T T U U V V 由此 - 34 - A A considered by HSL 10 months after the Director’s decision, B B were shown to be quantitatively unacceptable. C C D THE DIFFERENCES BETWEEN SWS’S CASE IN 2002 AND THE CASE D NOW PRESENTED IN THIS JUDICIAL REVIEW HEARING E E 78. As I have come to the conclusion that there are material F differences between the case as first put and the case now, it is important F to highlight those differences even though their primary relevance is G G towards the admissibility issue of the HSL report with which I have H already dealt. It is useful however to contrast the two cases as a prelude H to the next section which deals with each of the heads of challenge based I I on the information which was before the Director at the time of the J decisions. J K K 79. In the original notice of application it was stated that Jet A-1 L fuel was a “highly volatile substance”. The case was that the proximity of L a storage depot of such a fuel to SWS’s steel mill posed an unacceptable M M risk because of the nature of the fuel and the nature of the works carried on N N in the mill and the number of people employed there. Leave to apply for O judicial review was granted on this basis. O P P 80. In the case as now presented however, it is accepted that Q Jet A-1 fuel is correctly categorized on “a flammable liquid” and not Q highly volatile and that by complying with all safety requirements in the R R design, operating and testing of the PAFF it is safe in normal working S conditions. The original focus of SWS’s case was addressed, in evidence, S by an affirmation form Dr J. Wrigley, a Senior Environmental Protection T T U U V V 由此 - 35 - A A Officer, in February 2003. That affirmation dealt exclusively with the B B case being advanced at that time. A summary of his evidence was : C C “HS/APG approved the Study Brief for the Hazard Assessment which relates to risk to life. Other types of risk, such as D personal injury or property damage are outside the scope of the D Hazard Assessment (paras 6-9). E E The Hazard Assessment has been carefully reviewed by HS/APG (para 10). The PAFF containment and drainage systems are key F safety features which ensure that accidental spills of aviation fuel F are contained on the PAFF site and then drain to sea (para 10v). HS/APG advised EAND that the Hazard Assessment meets the G requirements of the Study Brief and EIA TM from risk point of G view (para 12). H H HS/APG and EAND have reviewed public comments including SW comments on ignition sources at the Mill, ‘JW-2’. HS/APG I conclude that ignition sources at the Mill have no impact or I relevance to the Hazard Assessment due to adequate separation distance between PAFF and the Mill (paras 13-18).” J J (HS/APG is Hazard Section of Air Policy Group). K K L L 81. The simple point is that virtually the entire focus of SWS’s M amended case, the 100% loss scenario, had not been mentioned. Mr Yu M submits that had it been raised at the public consultation period it would N N have been considered and responded to in the same way as the concerns O which were raised. O P P 82. What is new is not the fact that no QRA was done but the Q emergence of the issue that it should have been. Q R R 83. In a recent 2nd affirmation Dr Wrigley reinforces the S Director’s stance that, from the outset, the 100% loss scenario was S regarded as incredible. T T U U V V 由此 - 36 - A A “ ... when the Environmental Protection Department took the B view that the Environmental Impact Assessment Report did not B have to present a QRA on the instantaneous loss of 100% tank content scenario, that was made on the basis that such an event C C was incredible. According to the experience that I and my team have, it would not have made any difference to the risk D assessment whether one attempts to set out some sort of D quantitative assessment. E Based on the meeting discussions, it was agreed that the E ‘100% instantaneous tank failure’ was incredible and could be assessed qualitatively in the EIA. The ‘10% instantaneous tank F F failure’ was regarded as the worst credible event for the QRA process, as explained in EIA para 10.5.2.8.” G G By implication therefore, we see that it was decided that by “all” was H H meant “all credible events” and by “credible” was meant all those from a I realistic worst case scenario basis downwards. It was that decision which I in my judgment cannot be impugned. J J K K 84. 10.5.2.8, which is a key passage in the report, is as follows : L L “The tank vapour space could be in the flammable range due to vent opening to atmosphere and therefore ignition of tank M vent due to lightning could result in a tank fire and M subsequent failure at the roof to shell connection (API 650 tanks are provided with a weak roof to shell N N connection which will fail preferentially to any other joint). Such failures could also occur in the event of fire O impingement to relieve excess vapours. It may be assumed O that in the event of roof failure, the top most plate of the shell connecting to the roof may also fail resulting in spill P onto bund. Each plate is about 3m high which is about P 10% of the tank height. The catastrophic failure of the tank Q is therefore assumed to result in a release of 10% of tank Q contents (ie about 3,900 m3) on to the bund. Aircraft crash is the only conceivable incident that can result in more than R 10% of tank contents but this can be discounted, as the R proposed site does not lie near to the flight path.” S S T T U U V V 由此 - 37 - A A SWS’s original case was based on the proximity of a highly volatile fuel B B farm. What remains is the proximity of a flammable liquid farm should C an incredible event occur. C D D 85. Another difference in SWS’s case is that, it now seeks to E introduce a new ground of complaint based on the concept of “legitimate E expectation”. The basis of the argument, in outline, is that as the EIA F F report was not in full and complete compliance with the requirements, the G Director’s approval of it may have constituted an abuse of power. Put G more mildly, if the Director did not understand the requirements that he H H had to comply with he misused his power in allowing it to be approved. I I J 86. I accept that this argument is not at the forefront of SWS’s J case and that it is included not so much as a separate argument but within K K the principle that arguments based on unfairness, unreasonableness and L abuse of power can be intermingled (a note from Wade’s Administrative L Law, 8th edition, p.385). M M N 87. It can nonetheless be answered shortly. The leading Hong N Kong case of Ng Siu Tung v. Director of Immigration [2000] 5 HKCFAR 1 O O deals comprehensively with the doctrine of legitimate expectation. In our P case it has been advanced very late in the day, it was not in the notice of P application, it is unsubstantiated and it falls outside the criteria in the Ng Q Q case, namely where : R R “... the legitimate expectation arose from a promise or representation, the expectation being that the promise or S representation would be honoured, should be properly taken into S account in the decision-making process so long as to do so fell within the power, statutory or otherwise, of the T T decision-maker ...” U U V V 由此 - 38 - A A It is difficult to understand what clear promise has been ignored in the B B present case in the context of the guidelines which the Director is required C to follow in a non mechanistic and purposive manner. C D D WEDNESBURY UNREASONABLE? E E 88. Having decided the issues of the “admissibility” of the HSL F report and the “construction” issue in the respondent’s favour all that F remains is to examine SWS’s six heads of complaint in the context of G G Wednesbury unreasonableness. H H 89. The basic legal principles to be applied in each case are : I I (1) The court is not concerned with the correctness or otherwise J J of the decision. K (2) The court will examine the unreasonableness (Wednesbury) K on the basis of the material before the Director. L L (3) A challenge to the decision should be made promptly. M M (4) Remedies under judicial review are discretionary. N N (a) No QRA done on 100% loss scenari O O 90. Having decided that neither TM nor the SB requires a QRA P P for every identified risk it cannot, in my judgment, be argued that a Q decision not to do a QRA for an incredible scenario was unreasonable, let Q alone Wednesbury unreasonable. The decision concerned technical R R matters, the report was compiled by experts, the Director was advised by S experts. S T T U U V V 由此 - 39 - A A 91. To reach the high threshold at which a decision will be B B regarded as Wednesbury unreasonable is an uphill task for an applicant. C As Sir Derek Cons VP in Nguyen Ho v. Director of Immigration [1991] 1 C HKLR at p.583 put it : D D “I would most respectfully suggest that this approach must be in E accordance with basic principles, for if the Court may properly E interfere when the inferior tribunal has not taken into account some matter which it should have done, the Court must also be F F able to do so when the inferior tribunal has got that matter wrong. But it must be something that is plainly wrong or, as the judge G below put it, ‘established unassailably to be erroneous’. Courts G must in no circumstances allow themselves to be enticed into the evaluation of a fact which is properly within the exclusive H H jurisdiction of the tribunal.” I I The appropriate test for the court to adopt is as stated by Keith J in Tran J Van Tien v. Director of Immigration (No.1) [1996] 7 HKPLR at p.177 : J “... The test is not whether the evidence upon which the K K applicants seek to rely is capable of demonstrating that the facts on which the decisions were based were plainly wrong. The L test is whether the evidence demonstrated in fact that the facts on L which the decisions were based were plainly wrong. …” M M In 1996, Lightman J in R. v. Director General of Telecommunications N ex parte Cellcom (unreported) said : N O “If (as I have stated) the Court should be very slow to impugn O decisions of fact made by an expert and experienced decision-maker, it must surely be even slower to impugn his P P educated prophesies and predictions for the future. ... Q The Court may interfere with a decision if satisfied that the Q Director has made a relevant mistake of fact or law. But a mistake is not established by showing that on the material before R the Director the Court would reach a different conclusion. The R resolution of disputed questions of fact is for the decision-maker, and the Court can only interfere if his decision is perverse eg if S S his reasoning is logically unsound. ... The Court may interfere if the Director has taken into account an irrelevant T consideration or has failed to take into account a relevant T consideration. But so long as the Director takes a relevant U U V V 由此 - 40 - A A consideration into account, the weight to be given to that B consideration and indeed whether any weight at all should be B given to that consideration is a matter for the Director alone, so long as his decision is not perverse: ...” C C In our case the attempt to impugn the Director’s decision is precisely as D D described by Laws J in R. v. Secretary of State for Transport ex parte E Richmond L.B.C. “a disguised through elegant plea upon the merits”. E F F 92. With these principles in mind I turn to the remaining heads of G G challenge of which, as I have already found, none are truly free-standing H and all are closely linked to the “main” challenge. I have concluded that H none of them can survive the fact that ground 1 has failed on all I I grounds — unlawfulness, unreasonableness and irrationality. J J (b) The material facts argument K K 93. The facts known to the Director at the relevant time were that L L there was a steel mill next to the proposed site where approximately M 300 workers worked in 2001 and that the number would be substantially M the same in 2016. The precise distance of the mill boundaries and the N N PAFF were also known. The expert assessment, based on the hazards O which were quantitatively assessed (i.e. all those that it was necessary to O do) was that the risk to life was contained within the PAFF site. P P Q 94. I reject the argument that such a decision is rendered Q Wednesbury reasonable, or invalid, by virtue of the decision maker’s R R failure to visit the steel mill and consider what goes on inside the mill. S To be told that the steel mill contained molten metal, sparks, very high S temperatures, many workers and frequent site traffic would not take T T U U V V 由此 - 41 - A A anyone by surprise. Moreover, no survey of the mill was required by the B B SB and it is not suggested that the SB should have required one. C C (c) The applicant’s letter D D 95. SWS’s case at trial is quite different from its case as set out in E E its 31 July 2003 letter. This particular ground can only have substance if F it can be shown that the Director disregarded SWS’s concerns as set out in F the letter. To be Wednesbury unreasonable the Director would have to be G G shown to have turned a blind eye to their proper concerns. The reality of H the situation is that SWS’s complaint is not that their complaints were not H addressed but that the Director disagreed with them. There is ample I I evidence that the concerns were addressed. The following are selected J references to this issue from the affirmation evidence : J K “… The comments discussed at that meeting included the K concerns set out in Shiu Wing’s letter to the EIA Ordinance Register Office dated 13 July 2002 referred to by Mr Pong at L L paragraph 6.8 of his Affirmation on behalf of Shiu Wing. The main purpose of the meetings was for the Airport Authority, with M its consultants, to show the EPD where in the EIA Report each M comment was addressed. I and my colleagues, with our consultants Mouchel and their sub-consultant, considered N N carefully all the comments raised, included Shui Wing’s concerns. The EIA Report was considered to have covered all the O comments.” (Mr W.O.W. Roberts, the Civil Engineering O Manager for HKAA) P P “HS/APG have reviewed and commented on EAND response to Q public comments including SW. There is now produced and Q shown to me a copy of the comments marked exhibit ‘JW-2’. I will now give further explanation of the important technical R R issues …” (Dr J. Wrigley) S S “During the public consultation period between 14 June 2002 and 14 July 2002, a total of 13 sets of comments were received T T by the EPD. The comments included comments from Shiu U U V V 由此 - 42 - A A Wing Steel Limited (‘SWS’) contained in a letter dated 13 July B 2002… During this period, the public comments received were B made available to the Secretary for the EIA Subcommittee though there was no obligation on the part of EPD to do so. At C C the end of the consultation period, I conducted a preliminary review of the comments and considered there were potentially D major concerns expressed by the public. I therefore alerted all D senior staff of the EPD including the Director, the Deputy Director of Environmental Protection (‘Deputy Director’) and E the Assistant Director of the concerns expressed, highlighting the E major issues including those raised by SWS… F F After the public consultation period, staff of the EPD carried out a thorough review of the issues raised by the public, collated G them under different categories, discussed them with the HKAA G and checked them against the EIA report. The staff of EPD including myself were satisfied that the concerns appeared to H H have been fully addressed and, a set of detailed responses were subsequently drafted …” (Mr Hui Yat Ming, a Principal I Environmental Protection Officer) I J J (d) 25 April meeting K K 96. This issue has been addressed earlier in this judgment. I do L not consider there to be merit in the complaint that a decision was made L behind closed doors and therefore to the disadvantages of the public. As M M already stated the EIA report is prepared in a “non public” period in the N whole process. The basis of the risk assessment became public (namely N the 10% loss of tank content scenario) and therefore open to scrutiny the O O moment the report was published. It is wrong to describe the Director as P making a “belated admission” to this effect in August 2003. The P consequence of the decision has been in the public domain since May 2002, Q Q and was affirmed to by Dr Wrigley in February 2003. Neither the TM R nor the SB require disclosure of all the circumstances, considerations, R deliberations and reasons which led up to any particular judgment or S S decision being made. T T U U V V 由此 - 43 - A A 97. In short it is suggested that because a decision concerning B B the content of the EIA report was made between the Director and the C HKAA, (at the highest level and by acknowledged experts) at a time when C there was no provision for public involvement, was Wednesbury D D unreasonable because it “surrendered the process to private agreement”. E I consider this submission to be circular and illogical. E F F 98. In any event, it is worthy to note that this complaint is not set G out in SWS’s notice of application. The respondent and interested party G have therefore not had an opportunity of filing evidence about it. H H I (e) Safeguards and conditions I J 99. SWS’s complaint is that the Director has granted J environmental approval and permission without ensuring that the full K K details of the project are fixed with certainty. L L 100. Section 10(5) of the EIAO permits the granting of a permit M M with conditions. Section 9 makes the conditions enforceable and N N section 14 provides for the suspension of the project if the conditions are O not complied with. O P P 101. SWS refers to an extract from Mr Hui Yat Ming’s affirmation Q as the basis of the complaint : Q “[w]here the particular [mitigation] measure is complex, difficult R R to express, or contain details not readily available during the EIA stage, further submissions may be required to ensure that the S project will be designed in line with the assumptions in the EIA S while providing the necessary flexibility for minor project changes.” T T U U V V 由此 - 44 - A A 102. It is argued that this amounts to an admission that the Director B B did not have sufficient information on which to base a decision that the C qualitative assessment was sufficient and that the Director allowed HKAA C to delay providing important technical information about the project. D D E 103. To interpret Mr Hui’s evidence in this way reads too much E into it. No complaint can be made that the Director in acting in F F accordance with his powers acted unreasonably in the Wednesbury sense. G That it could have been unlawful and outside his powers is unsustainable. G H H 104. The role of conditions was considered and commented on in I the “Long Valley” appeal : I J “… Where a report is approved and permit granted, however, J important steps in the process follow. These are:- K K a. The Director will often impose conditions on both the report and the permit (see Schedule 4 of the Ordinance). These L are to ensure (for example) that Habitat Creation and L Maintenance Plans and Environmental Monitoring and Audit Programmes outlined in the Report are later approved; that M M mitigation measures and the like are in place and satisfactory before the project proceeds. N N b. The report will be placed in the register under section 8(5). This is available for public inspection and if relevant may be O relied upon for the granting of a permit for other projects O without a further EIA study and report. P P c. The Director continues to monitor compliance before, during and after construction and will if necessary use his Q considerable powers under the enforcement provisions of the Q Ordinance. R There is a difference in emphasis between conditions to the R approval of a report, which usually are to ensure that matters dealt with in the report are completed. The Director’s approval S S of a Habitat Creation Plan for example. The conditions to a permit are related to the construction work and are to ensure that T the report is complied with before, during and after construction. T U U V V 由此 - 45 - A A For example a condition that a mitigation area is functioning B before construction begins.” B C Conditions enable matters of detail to be dealt with at a later stage. C D D 105. Before I turn to the final section of this decision (namely the E issue of discretion), I think it appropriate at this stage to deal briefly with E an important point, hitherto unmentioned. The exhaustive EIA process, F F starting with the project profile, going through the SB, TM and report stage, G on to the approval, permit and conditions stage is but one statutory control G and safeguard in the construction of this important but sensitive installation. H H The Buildings Department, the Fire Services Department, the Marine I Department and the Lands Department are others who exercise stringent I statutory control over the project. The approval of the report and the J J granting of the permit under the EIAO is just one of a number of approvals K K to be sought. This court’s decision that the Director’s decision should not L be reviewed is one piece in a bigger picture. That is not to say that L because there are other statutory schemes in place anything less than the M M highest standards of expertise and competence will do when the crucial N decisions are made. It merely serves to show that these same tanks, bund N wall, fuel properties and so on, will be scrutinized from a variety of O O different viewpoints. P P DISCRETION Q Q 106. For the sake of completeness only, I make brief reference to R R the question of discretion. Remedies in judicial review are discretionary. S In view of the decisions made I do not have to consider whether I should S exercise a discretion not to grant any remedies even though a ground of T T judicial review has been made out. U U V V 由此 - 46 - A A 107. However, had I found in favour of SWS on any one of the B B grounds of complaint, I address the issue of whether or not I would have C exercised my discretion to decline to grant any relief. It would have been C a difficult decision. D D E 108. At the end of the day I would have exercised the discretion in E the Director’s favour. The outline reasons being : F F (a) that there was delay in commencing the proceedings. It is G G very arguable that the first decision which SWS submit was H wrong was the section 6(3) decision. However no H proceedings were commenced in relation to that decision I I although SWS’s original complaints plainly stem from the content of the EIA report regardless of whether it was J J accepted or not. The section 6(3) decision was in June 2002, K the section 8 and section 10 decisions were August. SWS’s K letter of complaint had been in July but their original notice of L L application was not until 1 November 2002. M (b) the crux of SWS’s case as presented at the hearing was largely M technical in nature. It attacked the technical soundness of N N the EIA report based on expert opinion produced in the O summer of 2003. Had the report been admitted and had the O court decided a strict compliance with the letter of the TB and P P SB was required the outcome would have been that the judicial review would have succeeded on the basis that it had Q Q been unlawful for the Director to carry a QRA on a scenario R which was, based on the information available to the Director R at the time, considered to be incredible. S S (c) In any event the case, as put in 2003, could have been made at T the time of the public consultation but was not. T U U V V 由此 - 47 - A A (d) It is the court’s view that the sum of the evidence supports the B B submission that the same decision would be made now as was C made in August 2002 notwithstanding the additional material C which has been submitted. D D E 109. In short this court’s view would have been that the balancing E exercise would have tilted in the respondent’s favour so that this particular F F major designated project would have been allowed to continue in the G G interest of the public at large and in the interest of good administration. H H 110. All considerations concerning discretion are academic in view I I of the court’s primary conclusions. J J 111. To conclude : K K (1) The Director’s decisions have been considered on the basis of L the materials before him. L M (2) SWS’s challenge that the decisions were unlawful based on M the alleged failure to meet the requirements of the SB, the TM N N and the Ordinance have not been made out. O (3) That challenge has been based, to a large extent, on the O opinions in the HSL report which is inadmissible. P P (4) Were the post decision materials admissible they would have Q resulted in a difference of expert opinion which itself would Q not have altered the fact that the Director was acting within R R his power when he decided that the EIA report should be approved as it complied with the SB and TM and that decision S S was not Wednesbury unreasonable. T T U U V V 由此 - 48 - A A The answers to the five questions posed on pages 14 and 15 of this B B judgment are : C C (1) No. D (2) Yes. D E (3) Yes. E (4) Not addressed as a separate issue. F F (5) Had the answers to (2) or (3) been different the discretion to, G nonetheless, grant no relief would have been exercised. G H H 112. This application is dismissed. The respondent’s and I I interested party’s costs are to be paid by the applicant. J J K K L L ( M.P. Burrell ) M Judge of the Court of First Instance, M High Court N N O O Mr Charles Haddon-Cave, QC, Mr Nigel Pleming, QC and Mr Anthony Chan, SC, instructed by Messrs Simmons & Simmons, P P for the Applicant Q Mr Benjamin Yu, SC and Mr Anthony Ismail, instructed by Q the Department of Justice, for the Respondent R R Ms Gladys Li, SC, instructed by Messrs Lovells, for the Interested Party S S T T U U V V