FEDERAL COURT OF AUSTRALIA

 

Strong Wise Limited v Esso Australia Resources Pty Ltd
[2010] FCA 240


Citation:

Strong Wise Limited v Esso Australia Resources Pty Ltd [2010] FCA 240



Parties:

STRONG WISE LIMITED v ESSO AUSTRALIA RESOURCES PTY LTD (ACN 091 829 819), BHP BILLITON PETROLEUM (BASS STRAIT) PTY LTD (ACN 004 228 004) and PERSONS WHO MAY HAVE A CLAIM WITHIN THE MEANING OF ARTICLE 2 OF CONVENTION ON LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976 (AS AMENDED BY THE 1996 PROTOCOL TO AMEND CONVENTION OF LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976) AS GIVEN FORCE OF LAW BY THE LIMITATION OF LIABILITY FOR MARITIME CLAIMS ACT (1989) (AS AMENDED) IN RESPECT OF LOSS SUFFERED OR DAMAGE CAUSED BY AN ANCHOR OR ANCHOR CABLE OF THE VESSEL "APL SYDNEY" IN PORT PHILLIP BAY, VICTORIA ON OR ABOUT 13 DECEMBER 2008



File number:

VID 1060 of 2008



Judge:

RARES J



Date of judgment:

18 March 2010



Date of Corrigendum:

18 May 2010



Catchwords:

ADMIRALTYLIMITATION OF LIABILITY – application under s 25 of the Admiralty Act 1988 (Cth) by shipowner to limit liability under Limitation of Liability for Maritime Claims Act 1989 (Cth) and the Convention on Limitation of Liability for Maritime Claims 1976 as affected by the 1996 Protocol to amend that Convention – meaning of “claims arising on any distinct occasion” in Arts 2(1)(a) and 6(1) of the Convention – Arts 6(1)(b) and 11 of the Convention permitting shipowner to apply to limit liability for all claims arising on a distinct occasion – maritime liens –  multiple claims alleged to arise from ship’s anchor fouling submarine gas pipeline, ship then going astern, ship later moving ahead, pipeline then fractured, later ship moving astern and further bending pipeline – claims by pipeline’s owners for repairs, loss of gas and economic loss – consumers of gas claiming economic losses – longer repair period because of further bending – whether the different events causing damage to the pipeline on one or more distinct occasions


ADMIRALTYSTATUTORY INTERPRETATION – construction of international convention given force of law by an Act of the Parliament – construction of Convention on Limitation of Liability for Maritime Claims 1976 as affected by the 1996 Protocol to amend that Convention  – construction of international conventions – application of principles in Vienna Convention on the Law of Treaties of 1969 – purpose of limitation Convention to protect shipowner from financial ruin, encourage investment in shipbuilding, international trade and commerce and provide for limited and certain insurable risks – unbreakable limitation – purpose only to limit the liability of shipowner for each separate act, neglect or default from which claims arise – history of limitation of maritime claims laws and conventions – use of domestic law precedents


PRACTICE AND PROCEDUREEVIDENCE – concurrent evidence – direction that experts in each relevant discipline confer together, without the parties or their lawyers to prepare joint report setting out issues on which they agree and disagree, giving brief reasons for their differences


Held:  Whether one occasion is distinct from another will depend upon whether the causes of the claims that arise from each act, neglect or default are sufficiently discrete that, as a matter of commonsense, they can be said to be distinct from one another. Claims arose on two distinct occasions.



Legislation:

Admiralty Act 1988 (Cth) s 25

Civil Aviation (Carriers Liability) Act 1959 (Cth)

International Convention on Civil Liability for Oil Pollution Damage (CLC), 1969

Convention on Limitation of Liability for Maritime Claims 1976 as affected by the 1996 Protocol to amend Convention on Limitation of Liability for Maritime Claims of 19 November 1976

International Convention relating to the Limitation of the Liability of Owners of Sea-Going Ships, and Protocol of Signature (Brussels, 10 October 1957)

International Convention for the Unification of Certain Rules Relating to the Limitation of Liability of Owners of Sea-Going Vessels, 1924

Limitation of Liability for Maritime Claims Act 1989 (Cth)

Merchant Shipping (Liability of Shipowners & Others) Act 1900 (Imp.) (63 & 64 Vict. c. 32)

Merchant Shipping Act 1894 (Imp.) (57 & 58 Vict. c. 60), ss 503, 504

Navigation Act 1912 (Cth) s 410

Protocol to Amend the Convention for the Unification of Certain Rules Relating to International Carriage by Air, signed at Warsaw on 12 October 1929, done at the Hague on 28 September 1955 (The Hague Protocol to the Warsaw Convention 1955)

Transport Safety Investigation Act 2003 (Cth)

Vienna Convention on the Law of Treaties of 1969 [1974] ATS 2



Cases cited:

Ballast Trailing NV v Decca Survey Australia Ltd (NSWCA, unreported, 29 September 1981) considered

Banque Commerciale SA (In liq) v Akhil Holdings Ltd (1990) 169 CLR 279 applied

Barde AS v ABB Power Sytems (1995) 69 FCR 277 referred to

Braverus Maritime Inc v Port Kembla Coal Terminal Ltd (2005) 148 FCR 68 applied

British Transport Commission v United States 354 US 129 (1956) referred to

Caltex Oil (Australia) Pty Ltd v The Dredge “Willemstad” (1976) 136 CLR 529 considered

Chapman v Hearse (1961) 106 CLR 112 distinguished

China Ocean Shipping Co v South Australia (1979) 145 CLR 172 applied

Chiropedic Bedding Pty Ltd v Radberg Pty Ltd (2008) 170 FCR 560 referred to

Distillers Co Biochemicals (Aust) Pty Ltd v Ajax Insurance Co Ltd (1974) 130 CLR 1 considered

Dovuro Pty Ltd v Wilkins (2003) 215 CLR 317 distinguished

Exxon Shipping Company v Cailletau 960 F. 2d 843 (1989: CA5) considered

Foots v Southern Cross Mine Management Pty Ltd (2007) 234 CLR 52 referred to

Forney v Dominion Insurance Co Ltd [1969] 1 WLR 928 referred to

Fothergill v Monarch Airlines Ltd [1981] AC 251 referred to

Gulf Air Company GSC v Fattouch (2008) 251 ALR 183 referred to

Hughes v Lord Advocate [1963] AC 837 distinguished

Immigration and Multicultural and Indigenous Affairs v QAAH of 2004 (2006) 231 CLR 1 applied

James Patrick & Co Ltd v Union Steamship Co of New Zealand Ltd (1938) 60 CLR 650 cited

Jones v Dunkel (1959) 101 CLR 298 applied

Lewis v Lewis & Clarke Marine Inc 531 US 438 (2000) referred to

LK v Director-General, Department of Community Services (2009) 237 CLR 582 applied

Luke v Lyde (1750) 2 Burr 882 cited

Mahony v J Kruschich (Demolitions) Pty Ltd (1985) 156 CLR 522 distinguished

March v E & MH Stramare Pty Ltd (1991) 171 CLR 506 applied

Norwich Company v Wright 80 US 108 (1871) followed

Pilkington (Australia) Ltd v Minister for Justice and Customs (2002) 127 FCR 92 referred to

Povey v Qantas Airways Ltd (2005) 233 CLR 189 applied

Qenos Pty Ltd v Ship “APL Sydney” (2009) 260 ALR 692 considered

SS Pharmaceutical Co Ltd v Qantas Airways [1991] 1 Lloyd’s Rep 288 considered

Tasman Orient Line Ov v New Zealand China Clays Ltd [2009] 3 NZLR 58 disapproved

The Bramley Moore [1964] P 200 followed

The Creadon (1886) 5 Asp. M.C. 585 considered

The Leerort [2001] 2 Lloyd’s Rep 291 considered

The Lucullite (1929) 33 Lloyd’s List Rep 186 considered

The Rajah (1872) LR 3 A&E 539 considered

The Schwan [1892] P 419 considered

The Scotland 105 US 24 (1882) considered

The ‘Saint Jacques II’ [2003] 1 Lloyd’s Rep 203 referred to

The Stream Fisher [1927] P 73 cited

The Tolten [1946] P 135 followed

Vale v Sutherland (2009) 237 CLR 638 applied

Victrawl Pty Ltd v AOTC Limited (1993) 45 FCR 302 considered

Victrawl Pty Ltd v Telstra Corporation Ltd (1995) 183 CLR 595 applied


REFERENCES


Cleton, R., Limitation of Liability for Maritime Claims in Essays on International & Comparative Law in Honour of Judge Erades:  Maritime Niihoff 1983

Conférence Diplomatique de Droit Maritime, Dixième Session, Bruxelles 1957, Doc. No. 10. Royame de Belgique Ministere Des Affaires Etrangeres Et Du Commerce Exterieur (1958)

DVD:  Concurrent Evidence – New methods with experts (2005):  Produced by the Judicial Commission of New South Wales and the Australian Institute of Judicial Administration

F Berlingieri, Unification and Harmonisation of Maritime Law Revisited, (2006) 59 Revue Hellenique de Droit International 603

Gortius, H., Law of War and Peace (1625) Book 2 c. 11 §13

Griggs, P., Limitation of Liability for maritime claims:  the search for international uniformity [1997] LMCLQ 369

Griggs, Williams and Farr, Limitation of Liability from Maritime Claims, 4th Edition (2005)

Hare, J., Shipping for the Best Admiralty Bargain in Jurisdiction and Forum Selection in International Maritime Law, M Davies ed Kluwer Law Int. 2005

Holmes, Oliver Wendell, The Common Law, 1881 Lecture 1 Dover, NY

Lilar, A. and van den Bosch, C.:  Le Comité Maritime Intérnational 1897 - 1972

Limitation of Liability for Maritime Claims Act 1989 (Hansard, 12 April 1989 at pp 1482 – 1483)

Marden’s Collisions at Sea, 6th Edition Stevens & Sons London 1910

Meeson, N., Admiralty Jurisdiction and Practice, 3rd edition, LLP London 2003

Ordinance de la Marine of 1681

Rares, Steven, Expert Evidence in Copyright Cases – Concurrent Expert Evidence and the “Hot Tub” 15 October 2009, Sydney, 4th Biennial Copyright Law and Practice Symposium

Selvig, Prof E., An Introduction to the 1976 Convention;  published in The Limitation of Shipowners’ Liability:  The New Law, London, Sweet & Maxwell, Institute of Maritime Law (1986)

The Travaux Préparatories of the LLMC Convention 1976 and of the Protocol of 1996, Published by CMI Headquarters: Antwerpen, Belgium, November 2000

 

 

Dates of hearing:

3-6, 10-14, 17-21, 24, 27 & 28 August 2009

 

 

Place:

Melbourne

 

 

Division:

GENERAL DIVISION

 

 

Category:

Catchwords

 

 

Number of paragraphs:

364

 

 

Counsel for the Plaintiff:

Dr G Griffith QC, Mr S Horgan SC and Mr M Scott

 

 

Solicitor for the Plaintiff:

Holman Fenwick Willan

 

 

Counsel for the First and Second Defendants:

Mr P Murdoch QC and Dr AP Trichardt

 

 

Solicitor for the First and Second Defendants:

Blake Dawson

 

 



IN THE FEDERAL COURT OF AUSTRALIA

 

in admiralty

 

NEW SOUTH WALES DISTRICT REGISTRY

 

GENERAL DIVISION

VID 1060 of 2008

 

BETWEEN:

STRONG WISE LIMITED

Plaintiff

 

AND:

ESSO AUSTRALIA RESOURCES PTY LTD (ACN 091 829 819)

First Defendant

 

BHP BILLITON PETROLEUM (BASS STRAIT) PTY LTD (ACN 004 228 004)

Second Defendant

 

PERSONS WHO MAY HAVE A CLAIM WITHIN THE MEANING OF ARTICLE 2 OF CONVENTION ON LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976 (AS AMENDED BY THE 1996 PROTOCOL TO AMEND CONVENTION OF LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976) AS GIVEN FORCE OF LAW BY THE LIMITATION OF LIABILITY FOR MARITIME CLAIMS ACT (1989) (AS AMENDED) IN RESPECT OF LOSS SUFFERED OR DAMAGE CAUSED BY AN ANCHOR OR ANCHOR CABLE OF THE VESSEL "APL SYDNEY" IN PORT PHILLIP BAY, VICTORIA ON OR ABOUT 13 DECEMBER 2008

Third Defendant

 

 

JUDGE:

RARES J

DATE OF  CORRIGENDUM:

18 MAY 2010


CORRIGENDUM


1.         On page 26 paragraph 66, first line, insert the word “Convention” before the word “operated”.


2.         On page 31 paragraph 80, second line, insert the word “a” before the word “question”.


3.         On page 31 paragraph 83, fourth line, the word “proceeding” should read “preceding”.


4.         On page 35 paragraph 97, under Structural Engineers, “Director of Braemar Staege Ltd” should read “Director of Braemar Steege Ltd”.


5.         On page 38 paragraph 105, second last line, the word “were” should read “where”.


6.         On page 40 paragraph 114, “s 410” should read “s 410B”.


7.         On page 49 paragraph 140, sixth line, the word “date” should read “data”.


8.         On page 50 paragraph 143, delete the word “of” after “15:43:09”.


9.         On page 57 paragraph 162, first line, the word “ot” should read “not”.


10.       On page 76 in the heading above paragraph 225, the word “ESSSO” should read “ESSO”.


 

I certify that the preceding ten (10) numbered paragraphs are a true copy of the Corrigendum to the Reasons for Judgment herein of the Honourable Justice Rares.



Associate:


Dated:         18 May 2010




IN THE FEDERAL COURT OF AUSTRALIA

 

in admiralty

 

NEW SOUTH WALES DISTRICT REGISTRY

 

GENERAL DIVISION

VID 1060 of 2008

 

BETWEEN:

STRONG WISE LIMITED

Plaintiff

 

AND:

ESSO AUSTRALIA RESOURCES PTY LTD (ACN 091 829 819)

First Defendant

 

BHP BILLITON PETROLEUM (BASS STRAIT) PTY LTD (ACN 004 228 004)

Second Defendant

 

PERSONS WHO MAY HAVE A CLAIM WITHIN THE MEANING OF ARTICLE 2 OF CONVENTION ON LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976 (AS AMENDED BY THE 1996 PROTOCOL TO AMEND CONVENTION OF LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976) AS GIVEN FORCE OF LAW BY THE LIMITATION OF LIABILITY FOR MARITIME CLAIMS ACT (1989) (AS AMENDED) IN RESPECT OF LOSS SUFFERED OR DAMAGE CAUSED BY AN ANCHOR OR ANCHOR CABLE OF THE VESSEL "APL SYDNEY" IN PORT PHILLIP BAY, VICTORIA ON OR ABOUT 13 DECEMBER 2008

Third Defendant

 

 

JUDGE:

RARES J

DATE OF ORDER:

18 MARCH 2010

WHERE MADE:

MELBOURNE

 

 

THE COURT ORDERS BY CONSENT THAT:

 

1.                  The Plaintiff file and serve an outline of submissions and any statements of evidence by 4.00pm on 16 April 2010.

2.                  The First and Second Defendants file and serve an outline of submissions in reply and any statements of evidence in reply by 4.00pm on 3 May 2010.

3.                  The Plaintiff file and serve an outline of submissions in reply and any statements of evidence in reply by 4.00pm on 10 May 2010.

4.                  The proceedings be listed for further hearing on 13 and 14 May 2010.


Note:Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.
The text of entered orders can be located using Federal Law Search on the Court’s website.



IN THE FEDERAL COURT OF AUSTRALIA

 

in admiralty

 

NEW SOUTH WALES DISTRICT REGISTRY

 

GENERAL DIVISION

VID 1060 of 2008

 

BETWEEN:

STRONG WISE LIMITED

Plaintiff

 

AND:

ESSO AUSTRALIA RESOURCES PTY LTD (ACN 091 829 819)

First Defendant

 

BHP BILLITON PETROLEUM (BASS STRAIT) PTY LTD (ACN 004 228 004)

Second Defendant

 

PERSONS WHO MAY HAVE A CLAIM WITHIN THE MEANING OF ARTICLE 2 OF CONVENTION ON LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976 (AS AMENDED BY THE 1996 PROTOCOL TO AMEND CONVENTION OF LIMITATION OF LIABILITY FOR MARITIME CLAIMS 1976) AS GIVEN FORCE OF LAW BY THE LIMITATION OF LIABILITY FOR MARITIME CLAIMS ACT (1989) (AS AMENDED) IN RESPECT OF LOSS SUFFERED OR DAMAGE CAUSED BY AN ANCHOR OR ANCHOR CABLE OF THE VESSEL "APL SYDNEY" IN PORT PHILLIP BAY, VICTORIA ON OR ABOUT 13 DECEMBER 2008

Third Defendant

 

 

JUDGE:

RARES J

DATE:

18 MARCH 2010

PLACE:

MELBOURNE


TABLE OF CONTENTS

Introduction..........................................................................................................

[1]

The scheme of the Convention....................................................................

[11]

History of limitation provisions..............................................................

[22]

Policy considerations......................................................................................

[43]

The construction of the Convention.....................................................

[46]

The use of Judicial Decisions on other Legislation in construing the Convention.........................................................................

[53]

Previous consideration of “arising on a distinct occasion”.

[58]

Use of common law concepts to characterise what is a distinct occasion................................................................................................

[66]

What is a distinct occasion?.........................................................................

[77]

Some of the primary sources of what happened.............................

[89]

Concurrent expert evidence.......................................................................

[93]

The anchoring system on the ship............................................................

[98]

Anchoring................................................................................................................

[103]

The overall context of the four asserted distinct occasions..................................................................................................................

[112]

Dropping Anchor..................................................................................................

[118]

The Ship Drags Its Anchor..............................................................................

[124]

The Cable Leads Out 90°.....................................................................................

[134]

The manoeuvres immediately preceding the anchor fouling the pipeline...........................................................................................

[143]

The anchor fouls the pipeline.....................................................................

[152]

The impact damage to the pipeline...........................................................

[156]

The consequences of the dead slow astern order at 15:46:01..

[175]

The effect of the dead slow astern order at 15:46:01...................

[182]

The windlass fails...............................................................................................

[201]

The events from 15:48 to 16:18:51...................................................................

[208]

Effect of Essso and BHP not pleading that the cable should have been slipped from the bitter end..................................................

[225]

The nautical experts’ contended courses of action...................

[229]

(a)      Use of the Engine.............................................................................................

[231]

(b)      Use of a gas axe...............................................................................................

[232]

(c)      Release from the bitter end.............................................................................

[239]

The pipeline ruptures........................................................................................

[260]

What caused the pipeline to rupture?....................................................

[263]

Was the decision to order the engine to go ahead seamanlike?............................................................................................................

[279]

Claims arising from the rupture of the pipeline...........................

[285]

Claim for lost gas and other potential claims.............................

[292]

How long was the anchor chain at the time of the rupture..

[295]

The events after the pipeline ruptured – The fourth alleged distinct occasion............................................................................

[303]

Was the astern movement a distinct occasion?.............................

[324]

The subsequent damage to the Mordialloc end Of The pipeline.......................................................................................................................

[329]

The aetiology of the missing fractured pipeline...........................

[332]

Conclusion...............................................................................................................

[361]

 

 

REASONS FOR JUDGMENT

Introduction

1                     On the afternoon of Saturday 13 December 2008 the MV “APL Sydney” dragged at anchor during a gale in Port Phillip Bay, Melbourne.  The “APL Sydney” was a two year old 231 metre l.o.a. (length overall), fully cellular container ship with a carrying capacity of 3,534 TEU (i.e. twenty foot equivalent units or containers).  She was then displacing 38,069 tonnes and carrying about 23,000 tonnes of cargo.  The outer and inner anchorages of the Bay are separated by an area through which a submarine pipeline ran.  The pipeline carried ethane gas at high pressure from Mordialloc on the eastern side to Altona on the western side.  These three areas were marked on the Admiralty chart for Melbourne.  The chart carried a warning of the location of the pipeline.  I have reproduced below the part of the chart showing the locale (Figure 1).

 

 

 

Figure 1

Location of the pipeline between the inner and outer anchorages in Port Phillip Bay

2                     The ship dragged her anchor to the north east from her original position west of the Fawkner Beacon in the outer anchorage in the Bay.  Between 15:44 and 15:45 the vessel’s starboard anchor fouled the pipeline.  This arrested the movement of the ship.  The pipeline was a 10 inch thick steel pipe coated with concrete and tar.  It was buried about 3 metres below the seabed.   Shortly after the fouling, at about 15:46, the vessel’s engine was put astern and about one and a half minutes, later, stopped.  “APL Sydney” then yawed from side to side in the wind and waves for about 35 minutes before her engine was put ahead at about 16:20.  Soon after this, the pipeline ruptured and, then, the engine was stopped.  About six minutes later, the engine was put astern.  One end of the ruptured pipeline was pulled back towards Mordialloc, dragging it further out of its trench and bending it almost to a right angle before a piece of the pipe broke off, freeing the anchor.

3                     On 19 December 2008, the owner of the “APL Sydney”, Strong Wise Limited, began these proceedings under s 25 of the Admiralty Act 1988 (Cth).  The shipowner claimed that it was entitled to limit its liability pursuant to the Limitation of Liability for Maritime Claims Act 1989 (Cth).  The latter gives the force of law in Australia to the Convention on Limitation of Liability for Maritime Claims 1976 as affected by the 1996 Protocol to amend that Convention (s 6).  I will refer to the Convention as affected by the protocol simply as “the Convention” in these reasons.

4                     The pipeline was owned by Esso Australia Resources Pty Ltd and BHP Billiton Petroleum (Bass Strait) Pty Ltd.  Esso and BHP are the first and second defendants.  The third defendants have been nominated as persons who may have a claim within the meaning of Art 2 of the Convention.  No-one has appeared or been appointed to represent those persons.  However, the parties agreed that in four other proceedings in this court other plaintiffs have made claims presently estimated at a total of over $66 million.

5                     Article 6(1)(b) of the Convention entitles a shipowner to limit its liability for claims “arising on any distinct occasion”.  This case is essentially about the meaning of that expression as used in the Convention.  The shipowner argues that the whole episode beginning with the initial fouling and ending on the final separation of the anchor from the pipeline was a single “distinct occasion”.  If the shipowner is correct, then it will be entitled to establish one limitation fund, comprised of the value of “APL Sydney”, of about $32 million calculated in accordance with the Convention.

6                     Esso and BHP argue against that result.  They assert that there was more than one distinct occasion in respect of which claims may be made.  They assert that this is because there was more than one act, neglect or default giving rise to entitlements of persons to bring claims against the ship and shipowner.  They contend that in addition to the claims that arose on the occasion of the initial fouling, there were at least three further “distinct occasions” on which additional claims arose.  The four instances on which Esso and BHP rely are:

(1)               the navigational errors leading to the initial fouling of the pipeline by the anchor around 15:44 to 15:45;

(2)               the order at 15:46:01 that the ship’s engine go astern.  That allegedly caused the pipeline to be pulled further out of its trench for an appreciable distance and bent more;

(3)               the order at about 16:19:51 that the engine go ahead.  That allegedly caused the pipeline to rupture and to further deform together with the loss of a volume of ethane gas;  and

(4)               the order at 16:27:59 that the engine go astern.  That allegedly caused the anchor to re-engage with a portion of the severed pipeline on the eastern side, drag it further out of its trench, bending it to almost a right angle before severing about 6 to 7 metres of pipe.

7                     Esso and BHP argue that each of the last three engine movements caused new and separate, additional damage to the pipeline that was not inevitable or a necessary consequence of the initial fouling by the anchor or, in the case of each later engine movement, the immediately preceding engine movement.  If they are correct, then if the shipowner wishes to limit its liability, it will have to establish two or more separate limitation funds, one in respect of each “distinct occasion”.

8                     Thus, the central issue in these proceedings is whether there was a single distinct occasion, or more than one, that gave rise to all claims, within the meaning of Art 2(1)(a) of the Convention, in respect of damaged property in direct connection with the operation of the ship and consequential loss resulting therefrom.  The shipowner accepted that it bore the onus of proving that it was entitled to obtain substantive relief under the Convention.

9                     In separate proceedings in this Court, Esso and BHP have sued the shipowner for damages in respect of the cost of repair of the pipeline and other alleged losses estimated to exceed $27 million.  The cost of repair of the pipeline alone is said to amount to nearly the whole of the value of the ship.  Two other plaintiffs, Huntsman Chemical Corporation Pty Ltd and Qenos Pty Ltd, both large consumers of ethane gas supplied from the Altona end of the pipeline, have each begun other proceedings in this Court against the shipowner claiming damages for pure economic loss estimated at $12 million and $27.6 million respectively.  A trial of a separate question in the latter proceedings was ordered to determine whether the gas customers’ claims were capable of being made subject to a limitation of liability under Art 2 of the Convention.  Finkelstein J has found that they are:  Qenos Pty Ltd v Ship “APL Sydney” (2009) 260 ALR 692.  And the Port of Melbourne Corporation has sued for damages, or indemnity or contribution but has not yet given an estimate of its claim.

10                  The trial of these proceedings took nearly four weeks.  Regrettably, its outcome will not resolve any of the various claims for damages, although it may decide the potential sum recoverable.  The splitting of the several proceedings and the long trial in these limitation proceedings is not a necessarily efficient way of conducting such litigation.  However, the fragmentation of limitation from liability actions is an available, and perhaps necessary, consequence of the provisions of the Convention:  see N Meeson Admiralty Jurisdiction and Practice (3rd ed;  LLP London 2003) at 268-269 [8.83]-[8.85].  Inevitably, there will be appeals from the decisions on the construction of the Convention before any trial on liability and damages can occur.  The question of construction of the Convention raised in these proceedings has not been decided by any court anywhere in the world so far as the researches of the parties or myself have revealed.  At the trial the parties tendered a considerable body of factual and technical evidence requiring analysis.  I will consider first the construction of the Convention before turning to the facts.

The scheme of the Convention

11                  A person who apprehends that a claim for compensation under the Limitation of Liability for Maritime Claims Act may be made against that person by some other person may apply to this Court under s 25 of the Admiralty Act to determine the question whether the liability of the first person in respect of the claim may be limited under the Convention.  (The Supreme Courts of the States and Territories are given a similar jurisdiction under s 9 of the Limitation of Liability for Maritime Claims Act).

12                  When an application has been made under s 25(1), the Court may determine whether the applicant’s liability may be limited in accordance with the Convention.  If it does, the Court may determine the limit of that liability, order the constitution of a limitation fund for the payment of claims in respect of which the applicant is entitled to limit its liability and make such orders as are just with respect to the administration and distribution of the fund (s 25(3)).  A shipowner is entitled under s 25(4) to claim a right to have a limitation fund constituted as a matter of defence to an action against it.

13                  The Convention provides that shipowners (including charterers, managers and operators) of a seagoing ship may limit their liability in accordance with the rules of the Convention for claims set out in Art 2.  In addition, Art 1(4) provides that if any claim set out in Art 2 is made against “any person for whose act, neglect or default the shipowner … is responsible”, such a person shall be entitled to avail himself of the limitation of liability provided for in the Convention.  And, the liability of a shipowner includes liability in an action brought against the vessel herself;  i.e. an action in rem (Art 1(5)).  In addition, an insurer of liability for claims subject to limitation in accordance with rules of the Convention is entitled to the benefits of the Convention to the same extent as the assured (Art 1(6)).  Importantly, by force of Art 1(7), the act of invoking limitation of liability is not to constitute an admission of liability.

14                  Article 2 relevantly provides:

“(1)      Subject to Articles 3 and 4 the following claims, whatever the basis of liability may be, shall be subject to limitation of liability:

(a)        claims in respect of loss of life or personal injury or loss of or damage to property (including damage to harbour works, basins and waterways and aids to navigation), occurring on board or in direct connexion with the operation of the ship or with salvage operations, and consequential loss resulting therefrom.”  (emphasis added)

15                  Article 2(1) sets outs five other categories of claim, none of which are presently relevant.  Claims set out in Art 2(1) are subject to limitation of liability even if brought by way of recourse or for indemnity under a contract or otherwise, with certain exceptions not relevant here (Art 2(2)).  Article 3 provides that the rules of the Convention do not apply to certain classes of claim, such as salvage claims, under other conventions for oil pollution, nuclear damage and contribution in general average.  Importantly, Art 4 provides:

“A person liable shall not be entitled to limit his liability if it is proved that the loss resulted from his personal act or omission, committed with the intent to cause such loss, or recklessly and with knowledge that such loss would probably result.”  (emphasis added)

16                  Persons entitled to limitation of liability can set off claims made by claimants “arising out of the same occurrence” under Art 5.  Article 6 sets out the quantum of limitation of liability commencing with the following important expression:

“1.        The limits of liability for claims other than those mentioned in Article 7, arising on any distinct occasion, shall be calculated as follows  .…” (emphasis added)

17                  Article 7 commences “In respect of claims arising on any distinct occasion …” and provides for quantification of the amount of limitation of claims in respect of loss of life or personal injury.  Article 8 deals with the use of special drawing rights to determine quantification of the limitation fund.  Importantly, Art 9 provides:

“1.        The limits of liability determined in accordance with Article 6 shall apply to the aggregate of all claims which arise on any distinct occasion:

(a)        against the person or persons mentioned in paragraph 2 of Article 1 and any person for whose act, neglect or default he or they are responsible;”  (emphasis added)

18                  A limitation fund may be constituted in accordance with Art 11.  That provides that a fund shall be constituted in the sum of such amounts set out in Arts 6 and 7 as are applicable to claims for which that person may be liable together with interest thereon:

 “… from the date of the occurrence giving rise to the liability until the date of the constitution of the fund.  Any fund thus constituted shall be available only for the payment of claims in respect of which limitation of liability can be invoked.”  (emphasis added)  (Art 11(1))

19                  Where a fund has been constituted by any person named in Art 1(2) or that person’s insurer or a salvor, the fund then is deemed to have been constituted by all persons identified in Art 9 as entitled to constitute it (Art 11(3)).  The fund is then available for distribution among claimants in proportion to their established claims, subject to any priority established in Arts 6 and 7 (Art 12(1)).

20                  Once a limitation fund has been constituted under the Convention, a shipowner and other persons on whose behalf the fund was constituted, such as an insurer or protection and indemnity (P&I) club gain significant protection by force of Art 13.  I will refer to all these persons for simplicity’s sake, as “a limiting shipowner” when describing the Convention’s operation.  First, any person who has made a claim against the fund is barred from exercising any right in respect of such a claim against any other assets of a limiting shipowner (Art 13(1)).  Secondly, after a limitation fund has been constituted in accordance with Art 11, if a ship or other property belonging to a limiting shipowner is arrested or attached within the jurisdiction of a State party for a claim that could be raised against the fund (or any security that that limiting shipowner had given) a court of that jurisdiction may order its release.  And such a court must order the release of the ship or other property if the limitation fund was constituted at the port where the occurrence took place (Art 13(2)(a)) or at some other place nominated in Art 13(2).

21                  Article 15 provides that the Convention shall apply whenever any person, such as a shipowner, referred to in Art 1 seeks to limit its liability before the Court of a State Party or seeks to procure the release of a ship or other property or discharge of any security given within the jurisdiction of any such State.

History of limitation provisions

22                  The Convention was negotiated against a long history of earlier regimes providing some protection to shipowners by permitting them to limit their liabilities.  The development of modern limitation principles can be traced from a number of sources of law:  see the erudite exposition of Scott LJ in The Tolten [1946] P 135 at 149-154;  Marsden’s Collisions at Sea (6th ed Stevens & Sons London 1910) at 147-148;  Prof Erling Selvig, (a Norwegian delegate to negotiations for the 1976 Convention) An Introduction to the 1976 Convention published in:  The Limitation of Shipowners’ Liability:  The New Law (London, Sweet & Maxwell, Institute of Maritime Law (1986) at 15:  see too Robert Cleton (a Dutch delegate to the 1976 Convention) Limitation of Liability for Maritime Claims in Essays on International & Comparative Law in Honour of Judge Erades:  (Maritime Niihoff 1983) at 17-22. 

23                  Mr Cleton traced its development to Spain whence it spread to Italy and then France before spreading to Northern Europe at the beginning of the fourteenth century:  Cleton op cit at 17; see too Marsden op cit at 147.  Mr Cleton noted that previously German and Scandinavian law provided that a shipowner had no personal liability for limitable claims.  In those systems, claims were enforceable only against the ship and freight.  However, a claimant was entitled to a maritime lien conferring priority rights of recovery from those assets.  He described that system as the “execution” system.  In other countries, such as France, and later the United States, limitation of liability was implemented by what Prof Selvig termed the “abandonment” system.  There, the shipowner was personally liable for limitable claims, but was entitled to avoid, and thus limit, liability by abandoning the ship and freight to the claimants.  The consequence was that the claimants were only entitled to recover by enforcing their maritime liens against those assets.  Both those systems provided that the limitation amount could be distributed among the claimants according to priority rules applicable to maritime liens.  Those systems also provided that the limitable liability applied to the aggregate amount of claims accrued up to the time when limitation was invoked.

24                  Thus, in the eighteenth and nineteenth centuries, the predominant European approach to limitation involved a principle of abandonment in natura.  That required the owner to limit by reference to the actual value of the ship plus freight after the accident:  see too Patrick Griggs (President of the Comité Maritime Intérnational (CMI)):  Limitation of Liability for Maritime Claims:  The Search for International Uniformity [1997] LMCLQ 369 at 372. 

25                  In contrast, England had developed quite a different system of limitation during the eighteenth and nineteenth centuries.  This began with an English Act of 1734 (7 Geo 2 c15).    That Act recognised the value of the ship and freight as the limit of liability, but contrary to the law of other countries, that value was arrived at before the accident.  And, to achieve this, the English system used a monetary value for the ship based on an amount per tonne.  It also gave a separate right to recover for personal claims, in addition to rights to recover for property damage.  In general, only the latter class of claims was subject to limitation.  The limitation fund was to be distributed among the claimants in proportion to their claims, not according to the priorities of maritime liens.

26                  Prof Selvig noted that English legislation provided for a separate limitation fund that would be available for claims arising on “any distinct occasion” concluding that:

“… thus, the extent of aggregation of claims for limitation purposes was restricted accordingly.”

27                  In The Tolten [1946] P at 149-154 Scott LJ explained the historical interconnection between maritime liens (or “privilège” or “créance privilégiée”; in the Continental European legal lexicon a “créance” is a debt or liability) and limitation of liability.  He described “… an integral – almost organic – connexion between the two in the history of our own admiralty law, and that connexion comes from the ancient law of the sea in which it is deep-rooted”:  The Tolten [1946] P at 149.  The expression “creances privilégiées” described the secured right of the sea creditors which are relevantly the same as the Anglo-Australian concept of maritime liens.  He explained that the principle of limitation was given operative effect by the “droit de l’abandon” that permitted the shipowner to acquit himself of all the “créances du voyage” by abandoning his ship to his creditors with a view to it being realised by the Court and the proceeds distributed rateably among the creditors in accordance with the several priorities of their “privileges”:  The Tolten [1946] P at 150.

28                  Scott LJ related this history to the scope and function of the action in rem.  He explained that when the Court sells the res or ship, first, such a sale passes a clear title to the purchaser that extinguishes all pre-existing maritime liens and, secondly, enforces distribution of the proceeds among the lien creditors in order of their priorities, and subject to those, rateably:  The Tolten [1946] P at 145-146.  He discussed the 1924 Convention saying (The Tolten [1946] P at 153):

“On the Limitation Convention, agreement was achieved by combining the economic effect of the Continental (and original) system of allowing the shipowner to clear his liabilities by abandoning his ship and freight to his creditors (privileged and unprivileged) with the English system of a maximum money liability dependent on the size of the ship, substituting in effect a conventional value of ship and freight for abandonment; but it is an essential principle of the international concordat that the existing correlation of limitation of liability with maritime liens, inherent in the general law of the sea, should be preserved, so as to ensure that the proceeds of ship and freight, or the fund coming from the statutory payment, should be distributed by the court in strict accord with the rights and priorities of the lien creditors. Adherence to the principle of correlation appears in art. 6 which provides for distribution of the limited fund in strict accordance with the order of ranking of all liens on the ship. Likewise in the requirement of art. 8, namely, that if proceedings are taken against the same ship in courts of different states, its owner shall be entitled to bring to the notice of any such court all the claims, whether "privilégiés" or not, already lodged against him or his ship in all the other courts, so as to ensure that through proceedings being taken in more courts than one, the total limit of his liability shall not be exceeded. In this convention there is again manifest the tacit assumption that the law of the seas shall prevail whatever the national court, and whatever the secured claim (créance privilégiée), and whatever the country where the claim originated.”


29                  In Norwich Company v Wright 80 US 108 (1871) Bradley J gave the unanimous opinion of the Supreme Court of the United States.  He traced the history of limitation in maritime law in a learned opinion that has been approved by that Court on subsequent occasions:  Lewis v Lewis & Clarke Marine Inc 531 US 438 (2000) at 446-447 per O’Connor J for the Court:  British Transport Commission v United States 354 US 129 (1956) at 133 per Clark J.  Bradley J referred to Hugo Grotius’ observation in his Law of War and Peace (1625) Book 2 c. 11 §13, that Holland had rejected Roman Law and had applied a regulation that shipowners should be bound for acts of the master no further than the value of their ship and freight.  His Honor then noted that the French Ordinance de la Marine of 1681 had also provided that the shipowner was responsible for the acts of the master but would be discharged by abandoning the ship and freight:  Norwich 80 US at 116.

30                  Bradley J identified the policy reason behind these laws as being the need to protect shipowners so as to encourage investment in trading ships:  Norwich80 US at 116-117.  He then traced the development of English legislation from 1734 to Congress’ enactment in 1851 of the United States law for limitation (formerly 9 Stat. at Large 635 and now included in the 2006 revision as 46 USC §30501-30512).  He saw the Congressional purpose as informed by the prism offered by the law maritime, together with European and British limitation legislation (Norwich 80 US at 121), finding that:

“The great object of the law was to encourage ship-building and to induce capitalists to invest money in this branch of industry. Unless they can be induced to do so, the shipping interests of the country must flag and decline.”

31                  Lord Denning MR once said that limitation of liability was not a matter of justice but “… a rule of public policy that had its origin in history and its justification in convenience”:  The Bramley Moore [1964] P 200 at 220, Donovan and Danckwerts LJJ agreed;  see too Cleton op cit at 15.  In China Ocean Shipping Co v South Australia (1979) 145 CLR 172 at 185 Barwick CJ observed of ss 503 and 504 of the Merchant Shipping Act 1894 (Imp) (see too at 200 per Gibbs J):

“The policy, evident in these sections is the protection of the owner engaged in the maritime carrying trade from financial ruin where his vessel causes damage of the described kind.”

I am of opinion that this purpose still informs the policy reasoning that underlies the present Convention.

32                  The first international convention providing for limitation was the International Convention for the Unification of Certain Rules Relating to the Limitation of Liability 1924 (the 1924 Convention).  It gave shipowners an option of limiting their liability to the value of the ship and freight or an amount of £8 per tonne.  Article 6 of the 1924 Convention provided that the “… various claims connected with a single accident …” would rank equally with one another against the limitation amount having regard to the order of any maritime liens.  This was described by Albert Lilar (the Belgian Minister for Justice and his country’s representative at the 1957 Brussels Conference) as being “… the result of a laborious compromise between the traditional limitation system applied on the European Continent … and the system in force in Great Britain”:  Griggs [1997] LMCLQ at 372 quoting from A. Lilar and C. van den Bosch:  Le Comité Maritime Intérnational 1897-1972 at 1.  Mr Griggs concluded that the 1924 Convention was not much of a compromise because it had effectively adopted s 503 of the Merchant Shipping Act 1894 (Imp), so much so, that Great Britain did not amend that Act.  Perhaps more charitably, Prof Selvig said that the 1924 Convention “was a most unhappy compromise between the existing systems”.  The CMI regarded the 1924 Convention as a failure, having attracted only 15 States Party:  Griggs [1997] LMCLQ at 372;  see too Cleton op cit at 19. 

33                  Next, the Convention Relating to the Limitation of Liability of Owners of Seagoing Ships done at Brussels in 1957 (“the Brussels Convention”) was agreed in 1957.  Article 2(1) of the Brussels Convention provided that the limitation of liability applied to the aggregate of personal and property claims “… which arise on any distinct occasion without regard to any claims which have arisen or may arise on any other distinct occasion”.  Ultimately, it attracted 46 States Party.  Prof Selvig concluded that it had the result that “… the English system for limitation of liability received full international recognition” (Selvig op cit at pp 3-5).  He noted that the Brussels Convention had made a few changes to refine that system.  And other commentators also have observed that the Brussels Convention had adopted the basic principles of the British limitation system:  Cleton op cit at 19;  J Hare:  Shipping for the Best Admiralty Bargain in Jurisdiction and Forum Selection in International Maritime Law:  M Davies ed Kluwer Law Int. 2005 at 142 [1.2.2].

34                  Revision of the Brussels Convention began after the negotiation of the Convention on Civil Liability for Oil Pollution Damage in 1969.  Prof Selvig observed that the 1976 Convention introduced Art 4 to close what he described as “an escape route” that was present in the Brussells Convention.  He said (Selvig op. cit p 15) that the overall purpose of Art 4:

 “… is to prevent the right of limitation granted to shipowners, and enjoyed indirectly by P & I Insurers, being frustrated one way or the other.

The net effect for injured parties, however, is that the limitation amounts become the only source of compensation available from the persons responsible for the various aspects of the operation of the ship causing the damage.”  (emphasis in original)

35                  The travaux préparatories for the Convention suggest that the capacity of the insurance market to provide cover was an important consideration when new limits of liability were selected in the negotiation of Arts 6 and 7:  The Travaux Préparatories of the LLMC Convention 1976 and of the Protocol of 1996 (CMI Antwerp November 2000;  F, Berlingieri ed) pp 151-153.  The counterpoint in the delegates’ considerations was that, as a general rule, the new limits would be “unbreakable”, in contrast to the experience under the Brussels Convention.  Thus, the IMCO legal committee (until 1982 the International Maritime Organisation or IMO was called the Inter-Governmental Maritime Consultative Organisation or IMCO) reported after its 28th session in December 1975 (op cit at XIII:  151, 153 [72]):

“The consensus in the Committee was that, in choosing the figures for eventual insertion in the Convention, the Conference should recognize that the capacity of the insurance market was an important consideration to be taken into account, having regard to the fact that other liabilities would also have to be covered within the same capacity.  It was also generally recognized that as a general rule the more “unbreakable” the limits of liability were in the Convention, the higher these limits could be while a provision making it relatively easy to “break” the limitation levels would entail a corresponding decrease in the limits to be provided.”  (emphasis added)

36                  Prof Selvig reasoned that the Convention’s adoption of a global, “unbreakable” limitation regime ensured its overall purpose being achieved.  He identified that purpose as preventing “… the right of limitation granted to shipowners, and enjoyed indirectly by P&I insurers, being frustrated one way or the other” (Selvig:  op cit p 15).

37                  Neither the travaux préparatories nor the learned commentary on the evolution and sources of each of the three international limitation conventions examined the concept behind the criteria of “any distinct occasion” or “an occurrence” (or, for that matter, the French text’s single usage of the words “l’évenment”).

38                  One significant difference between the British and two other systems of limitation was that the former involved the use of a pre-determinable valuation of the ship, unaffected by the circumstances in which the claims arose.  In contrast, the Continental and United States systems valued the ship and freight as they were, following those circumstances and allowed the owners to abandon that property in its then state so as to constitute the sole means of satisfying all their liabilities.

39                  This difference entailed two consequences.  First, under the British system, the owners were fixed with an ascertainable, readily insurable, maximum liability.  That liability responded to claims made on one distinct occasion and it refreshed or revived in full for claims made on a second or subsequent distinct occasion.

40                  Secondly, under the other systems, the owners’ liability varied from the British system according to how damaged the ship was and what the value of her freight was after the circumstances from which the claims arose.  And that liability was not apparently refreshed or revived for claims arising on a second or subsequent distinct occasion.  Claimants’ maritime liens would take priority in respect of the ship and her freight in accordance with the substantive law in those systems.  Thus, once abandoned, the ship and freight either ceased to be available to respond to claims arising on a later occasion beyond what, if anything, was left of their value after it had satisfied the earlier maritime liens of claimants from the first occasion.  This balance constituted the only property or fund to which claimants on all subsequent occasions could resort.  At least from the materials to which I was referred by the parties, there did not appear to be any means of refreshing or reviving the amount available beyond the single total value of the, by now, damaged ship and freight.  Thus, if the ship were totally lost, the liability of the shipowner was at an end, since the ship (and unless it were still payable, her freight) then had no value:  The Scotland 105 US 24 (1882) at 28 per Bradley J giving the opinion of the Court.  Under the then provisions of the United States statute Bradley J explained (The Scotland 105 US at 29, see too at 34) that:

“… our law adopts the maritime rule of graduating the liability by the value of the ship after the injury, as she comes back into port, and the freight actually earned; and enables the owners to avoid all responsibility by giving up ship and freight, if still in existence, in whatever condition the ship may be; and, without such surrender, subjects them only to a responsibility equivalent to the value of the ship and freight as rescued from the disaster.”

41                   Subsequently, as a result of the decision in The Scotland 105 US 24, the Congress enacted a new measure, now reflected in 46 USC §30506(c) that prevented a shipowner from limiting its liability for all losses of life or personal injuries on one voyage to a single fund.  For that purpose it adopted the British discrimen of allowing limitation for any distinct occasion as the Fifth Circuit Court of Appeals explained in Exxon Shipping Company v Cailletau 869 F. 2d 843 (1989:  CA5) at 846-847.

42                  In the negotiations for the Convention, the delegates realised that the level and availability of insurance were related to the degree to which the ability of claimants to “break” limitation, could be constrained beyond the relative freedom they had had under the Brussels Convention.  Hence, the Convention used the severe restriction in Art 4 to create a virtually “unbreakable” limit of liability on any distinct occasion.  Insurers (including P&I clubs – although they are not technically insurers) could then provide higher levels of protection based on the notional value of a ship fixed in accordance with the valuation rules in the Convention.  Insurers, of course, insure against casualties, or events or “distinct occasions”.  The Convention and its predecessors had used the concept of claims arising on a distinct occasion, drawn from the British system.  That usage did not import the British jurisprudence, however instructive it may have been to common lawyers.  But the States Party understood that the use of the concepts from the British system involved employing a theoretical pre-incident valuation of the ship and having that valuation as the basis for establishing limitation for each and every occasion on which claims arose.

Policy considerations

43                  The Minister said in the second reading speech when introducing into the House of Representatives the Bill for what became the Limitation of Liability of Maritime Claims Act 1989 (Hansard, 12 April 1989 at pp 1482-1483):

“As a trade-off for the vastly increased amounts of compensation available, the Convention provides for avirtually unbreakable system of limiting liability.  It declares that a person will be deprived of his ability to limit liability only if it is proved that the loss resulted from his personal act or omission, committed with the intent to cause such a loss or recklessly and with knowledge that such loss would probably result.”  (emphasis added)

44                  This was cited with approval by Gummow J in Victrawl Pty Ltd v AOTC Limited (1993) 45 FCR 302 at 306, Lockhart and Cooper JJ agreeing;  see also Barde AS v ABB Power Sytems (1995) 69 FCR 277 at 290E per Sheppard J;   Griggs, Williams and Farr Limitation of Liability from Maritime Claims 4th ed;  (2005) at pp 1-2).

45                  The concept of the “virtually unbreakable” system is reflected in the provisions of Art 4.  Lord Phillips of Worth Matravers MR appositely, in my opinion, referred to the “very heavy burden” that Art 4 imposed, namely, identifying an act or omission of the shipowner committed with the intent to cause the loss complained of or recklessly with knowledge that such a loss would probably result:  The Leerort [2001] 2 Lloyd’s Rep 291 at 294 [10]-[13] Henry and Brooke LJJ agreed:  see too Hare op cit at 147 and The ‘Saint Jacques II’ [2003] 1 Lloyd’s Rep 203 at 207-209 [16] per Gross J.  That burden is similar to the one imposed by Art 25 of the Warsaw Convention as amended at The Hague 1955 with respect to the right of an air carrier to limit its liability (cp:  Civil Aviation (Carriers Liability) Act 1959 and SS Pharmaceutical Co Ltd v Qantas Airways [1991] 1 Lloyd’s Rep 288 at 290-291 per Gleeson CJ and Handley JA.

The construction of the Convention

46                  The Vienna Convention on the Law of Treaties of 1969 [1974] ATS 2 (“the Vienna Convention”) is an authoritative statement of customary international law for the purposes of construing the ConventionVictrawl Pty Ltd v Telstra Corporation Ltd (1995) 183 CLR 595.  There, Deane, Dawson, Toohey and Gaudron JJ used the Vienna Convention to construe the Convention even though the Vienna Convention had not entered into force until 1980 and, by Art 4, it did not apply directly to treaties concluded (as the Convention was) before it came into force:  see too Qenos 260 ALR at 695-696 [11]-[13].

47                  The Convention should be construed by reference to the principles stated in the Vienna Convention even though the latter has not been enacted as part of the law of Australia:  Immigration and Multicultural and Indigenous Affairs v QAAH of 2004 (2006) 231 CLR 1 at 15-16 [34] per Gummow A-CJ, Callinan, Heydon and Crennan JJ:  Povey v Qantas Airways Ltd (2005) 223 CLR 189 at 202 [24]-[25] per Gleeson CJ, Gummow, Hayne and Heydon JJ;  see too Chiropedic Bedding Pty Ltd v Radberg Pty Ltd (2008) 170 FCR 560 at 568 [34]-[35] per French, Rares and Besanko JJ:  Pilkington (Australia) Ltd v Minister for Justice and Customs (2002) 127 FCR 92 at 100-101 [26]-[27] per Mansfield, Conti and Allsop JJ;  Fothergill v Monarch Airlines Ltd [1981] AC 251 at 276D-278E per Lord Wilberforce; see also at 282B-283D per Lord Diplock 294A-295D per Lord Scarman, 302A-B per Lord Roskill;  Gulf Air Company GSC v Fattouch (2008) 251 ALR 183 at 190-191 [24]-[26]per Allsop P, Hodgson and Campbell JJA agreeing;  F Berlingieri,'Unification and Harmonisation of Maritime Law Revisited' (2006) 59 Revue Hellenique de Droit International 603 at 613-615.  Recently a majority of the New Zealand Court of Appeal, without referring to the Vienna Convention (esp Art 31(2)(a)), has taken a different approach rejecting the relevance of the history and travaux préparatories:  Tasman Orient Line Ov v New Zealand China Clays Ltd [2009] 3 NZLR 58 at 68-69 [31] per Baragwanarth J, 76-77 [69] per Chambers J, contra 83-84 [106]-[108] per Fogarty J.  (The Supreme Court of New Zealand granted leave to appeal on 1 July 2009).  The Vienna Convention provides in Arts 31 and 32:

“Article 31

General rule of interpretation

1.         A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.

 

2.         The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:

(a)        any agreement relating to the treaty which was made between all the       parties in connexion with the conclusion of the treaty;

(b)        any instrument which was made by one or more parties in connexion with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.

3.         There shall be taken into account, together with the context:

(a)        any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions;

(b)        any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;

(c)        any relevant rules of international law applicable in the relations between the parties.

4.         A special meaning shall be given to a term if it is established that the parties so intended.

Article 32

Supplementary means of interpretation

Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:

(a)        leaves the meaning ambiguous or obscure; or

(b)        leads to a result which is manifestly absurd or unreasonable.”

48                  The right of a shipowner to limit liability conferred by the Convention is a substantive right enforceable by independent proceedings:  Victrawl 183 CLR at 622 at 619 per Deane, Dawson, Toohey and Gaudron JJ;  see too at 616-617.  Their Honours held that the overriding limitation imposed by the Convention “attaches to”, limits and confines the rights of affected claimants to recover compensation through the international regime which the Convention establishes.  Their Honours held that the Convention involved a significant alteration of substantive domestic law.  Additionally, it had imposed a new regime that replaced that in the 1924 Convention.  The Brussells Convention had been given the force of law by previous provisions in the Navigation Act 1912 (Cth):  Victrawl 183 CLR at 618.

49                  Deane, Dawson, Toohey and Gaudron JJ said that subject to specific exceptions or exclusions, the Convention effectively transformed claims against a shipowner and the shipowner’s property into claims against a limited fund established under the ConventionVictrawl 183 CLR at 617.  The ability to limit liability in this way is an exceptional right of a shipowner and other persons who are entitled to invoke the provisions of the rules under the Convention for their protection.  Once a limiting shipowner has constituted a limitation fund under the Convention, then subject to any specific exceptions or exemptions, that shipowner is relieved of any further liability for claims.  And, the fund assumes a status, in effect, in rem to answer all claims arising from the one occurrence, whether or not they have already passed into judgment:  cp James Patrick & Co Ltd v Union Steamship Co of New Zealand Ltd (1938) 60 CLR 650 at 673 per Dixon J who explained how a right to establish a limitation fund operated under ss 503 and 504 of the Merchant Shipping Act 1894 (Imp);  see too The Tolten [1946] P at 149.  This is given effect by the provisions of Arts 13 and 15.

50                  A significant purpose of the Convention is to protect international trade and commerce conducted through the operation of seagoing ships.  Because ships may be arrested or attached in most maritime States, unless the international community provided a regime limiting shipowners’ liabilities a ship, or sister ship, could be arrested at every port in each jurisdiction at which she called.  The shipowner would be required to put up security each time his ship was arrested for the full amount claimed.  A shipowner, demise charterer, insurer or P&I club ordinarily will be required to pay up to the value of the ship into court as security for any claim that might be made against her in order to arrange for her release.

51                  The general law of the sea that has for centuries informed the exercise of Admiralty jurisdiction has recognised that absent a means for limitation of claims, a ship would be susceptible to an unlimited number of arrests arising from one casualty, or attachments arising out of one occurrence:  Luke v Lyde (1750) 2 Burr 882 at 887 per Lord Mansfield CJ;  The Tolten [1946] P at 155-156 per Scott LJ;  The Stream Fisher [1927] P 73 at 80-81 per Bateson J.  The consequence would be that claimants could obtain security up to the total value of their claims on each arrest or attachment.  The security so obtained may far exceed the monetary limits fixed in the Convention or the value of the ship.  If this were readily permitted it would be likely that once the ship was first arrested or attached, the shipowner may have little practical choice but to leave her where she lay.  In the absence of the protection to a ship and those interested in her offered by the Convention, substantive insurance and protection and indemnity arrangements would also be discouraged.  These have proved vital in the conduct of international maritime trade and commerce.

52                  The Convention ensures that a globally recognised and enforceable cap will be placed on the liability to which a shipowner is exposed in respect of all claims arising “on any distinct occasion”.  The critical question is, what is a “distinct occasion” or “occurrence”?  (The French text uses only one noun, “l’êvenment”, for the English text’s use of two, namely “occasion” and “occurrence”.  This, too, suggests that the differing expressions in the English text should be read cognately.)  The language of the Convention connotes that an occasion is where some act, neglect or default, of the shipowner, or a person for whom the shipowner is responsible, gives rise to the claims.

The use of Judicial Decisions on other Legislation in construing the Convention

53                  The parties in these proceedings urged that the meaning of “distinct occasion” in the Convention should be treated as if it were governed by judicial decisions, principally in the United Kingdom, based on the use of that expression in that nation’s domestic legislation, such as the Imperial Merchant Shipping Acts of the 19th Century.  While those decisions may assist in cases where there is ambiguity or be illustrative of the circumstances in which there may be a separation between different acts, neglects or default giving rise to different maritime claims, I am of opinion that that the meaning of the Convention must be arrived at from an examination of its text and structure:  cf LK v Director-General, Department of Community Services (2009) 237 CLR 582 at 596 [36] per French CJ, Gummow, Hayne, Heydon and Kiefel JJ:  see too Foots v Southern Cross Mine Management Pty Ltd (2007) 234 CLR 52 at 75 [62]-[64] per Gleeson CJ, Gummow, Hayne and Crennan JJ.  Each of the 1924, 1957 and present Conventions has used a different form of expression to address the aggregation of claims and a particular factual circumstance.

54                  The Convention uses the concepts of an individual “distinct occasion” (Arts 6(1), 7(1), 9(1)), “occurrence” (Art 5), “occurrence giving rise to the liability” (see Art 11(1)) and an “act, neglect or default” of a person for whom a shipowner is responsible (Art 9(1)(a), (c)).  The expressions “claims arising on any distinct occasion” (see e.g. Art 9(1)), “the same occurrence” (Art 5), and “the occurrence giving rise to the liability” (Art 11(1)) focus on a particular act, neglect or default of the shipowner or person for whom the shipowner is responsible.  I am of opinion that the natural and ordinary meaning of those expressions as used in the Convention is substantially the same.

55                  Although these are different expressions, their use in an international convention, should not be construed narrowly as in a domestic statute.  The purpose of the Convention is clear.  It is to allow the shipowner to limit his liability for claims arising out of a particular, identifiable act, neglect or default that gives rise to claims of the kind described in Art 2(1).  Such claims will often have attached to the ship as maritime liens at the moment they arose;  e.g. on a collision (Art 2(1)(a)).  If the shipowner’s conduct involves more than one act, neglect or default, each of which would entitle a person to bring a separate claim under Art 2(1), the shipowner cannot limit his liability in one fund for all the claims arising from the second or subsequent act, neglect or default.  The shipowner must constitute, if he wishes to limit his liability under the Convention, separate funds for each such individual act, neglect or default.

56                  This construction is reinforced by the exception provided to a shipowner’s entitlement to limitation in Art 4.  That disables a person liable from limiting his liability where it is proved that the loss resulted from “his personal act or omission” committed with the intent to cause that loss or recklessly with knowledge that the loss would probably result.  Once again, the focus of that exception is upon an act or omission.  An “omission” is similar to or connotes the concept of “neglect or default” used elsewhere in the Convention.

57                  The Convention allows the shipowner to limit his liability in respect of each particular, distinct occasion.  A shipowner who commits more than one distinct navigational error, even within a short period of time, each of which causes damage to third parties or even to the same third party, ordinarily will not be entitled to assert that all of the acts, neglects or defaults amount to a cognate “distinct occasion”.  As a matter of commonsense, usually, they do not.  There is more than one “distinct occasion”, and the shipowner is liable for each.  The ordinary and natural meaning of the Convention accords with the commonsense meaning.  The shipowner cannot use the Convention as a shield to escape that second source or occasion of his liability.  However, there may be instances where the occasions will not be “distinct”.  This will depend upon all of the circumstances, but, in particular the existence of some connection or lack of distinctiveness.  I will discuss these concepts in more detail under the heading “What is a distinct occasion?

Previous consideration of “arising on a distinct occasion”

58                  In Ballast Trailing NV v Decca Survey Australia Ltd (New South Wales Court of Appeal, unreported, 29 September 1981) Hope JA, with whom Hutley and Mahoney JJA agreed, considered whether the owners of the dredge “Willemstad” could  invoke the right to limit their liability under s 3 of the Merchant Shipping (Liability of Shipowners & Others) Act 1900 (Imp).  That provided that the limitation of liability afforded under that Act related to the whole of any losses and damages “which may arise upon any one distinct occasion”. The same dredge had caused similar damage that had led to the High Court’s decision in Caltex Oil (Australia) Pty Ltd v The Dredge “Willemstad” (1976) 136 CLR 529.

59                  There, a track plotter chart had been prepared for the dredging operation using a device that incorrectly marked the location of four pipelines.  However, the dredge struck the pipelines on each of the nine passages across a stretch of water that had been separated by intervals of approximately an hour and a quarter.  The striking occurred because of the error in the perceived location of the pipelines in relation to the dredge’s intended course.  Hope JA held that the master and pilot of the dredge had had an opportunity before each passage to make use of conventional navigational methods to determine the position of the dredge in relation to the pipelines but had failed to do so.  He concluded that there were nine separate occasions each of which produced damage in respect of which a liability arose.  That was because an action could have been brought in respect of each occasion based on a separate cause of action.  He observed that:

“No doubt incidents may be so close to one another that, as a matter of degree, it may be possible and proper to regard them as involving one distinct occasion.”

60                  Hope JA applied the leading decision of the English Court of Appeal in The Schwan [1892] P 419.  There, the master of The Schwan manoeuvred her in a river so that she collided with two different vessels.  However, each collision was avoidable had the master not acted negligently.  And, the first collision was not the cause of the second, despite the short interval between the two.    Lord Esher MR said that in such a case, the ship could not limit its liability to one fund, but was liable to each of the two plaintiffs to the full extent of her statutory liability ([1892] P at 438).  He found that there were two mistakes of bad seamanship that had caused accidents to two ships.  Lord Esher MR had posited the following test (The Schwan [1892] P at 439):

“So if you run into one ship half an hour before you run into another, what difference does it make? It is not the time which is the substantial thing; but whether both are the result of the same act of want of seamanship, and, if they are not, the [Merchant Shipping] Act does not apply, except as to each of them separately.”

And Bowen LJ added (The Schwan [1892] P at 441):

“It is clear that you must examine the section in each case to see what particular damage is caused by the same act of improper seamanship; that if you find two acts which are distinguished one from another, which lead to loss or damage, then the double loss or damage is not entirely due to the same act. It is due to two acts instead of to one act. Otherwise, as has been pointed out, a ship might after making one blunder go blundering up the whole river. It is quite impossible to take that view. The question is, what unseamanlike act of the person in charge of the ship has caused a particular accident?”  (emphasis added)

61                  Their Lordships applied the decision of Butt J in The Creadon (1886) 5 Asp. M.C. 585 at 586.  He held in that case that a second collision was inevitable after the first had occurred.  Butt J allowed the owners to limit their liability to one fund.  He found that the two collisions were so close together that

“…:  the first was the substantial and efficacious cause of the second, and that there was no separate act of negligence on the part of those in charge of The Creadon in respect of the second collision.”

62                  That approach harked back to the similar reasoning of Sir Robert Phillimore in The Rajah (1872) LR 3 A&E 539 at 542.  He held that although the Rajah had collided with two ships, each collision had taken place “… if not at the same moment in time, yet substantially at the same time and on the same occasion;  and the whole damage done seems to have been caused by the one act of improper navigation on the part of the Rajah”.

63                  The Scottish Court of Session followed The Schwan [1892] P 419 in The Lucullite (1929) 33 Lloyd’s List Rep 186 at 187 per the Lord Justice-Clerk (Lord Alness), 188 per Lord Ormidale agreeing, Lord Hunter and 189 per Lord Anderson).  There, Lord Anderson identified the tests as being first, whether distinct acts of negligence had caused damage to each of the two ships or, secondly, whether the later act of negligence causing the damage to the second ship had not been necessitated or rendered inevitable by the earlier.

64                  In Exxon 869 F 2d at 846-847, the Fifth Circuit Court of Appeals noted that the law for limitation of liability in the United States of America (then contained in 46 USC App §185 and now in 46 USC §30506(c):  the United States has not ratified the 1976 Convention) had devolved from British law based on the “distinct occasion” rule.  They held that where successive collisions had occurred as a result of the same negligent act or constituted one “distinct occasion” but added “… if there is time and opportunity after the first collision to take action which would avoid the second collision, each is a ‘distinct occasion’”:  Exxon 869 F 2d at 847-848.

65                  In Distillers Co Biochemicals (Aust) Pty Ltd v Ajax Insurance Co Ltd (1974) 130 CLR 1 at 19 Stephen J (with whom Gibbs J at 10 agreed on this aspect) discussed the meaning of a clause in an insurance policy limiting the insurer’s liability to the insured, the distributor of the drug “thalidomide”.  The policy wording limited the insurer’s liability for all compensation payable to all claimants “… in respect of or arising out of any one occurrence”.  Stephen J observed that because the policy contemplated that a number of claims could arise out of the one occurrence, this indicated that the word “occurrence” there meant the mishap, not its consequences, following a similar construction on another insurance policy made by Donaldson J in Forney v Dominion Insurance Co Ltd [1969] 1 WLR 928 at 934. Gibbs and Stephen JJ held that the occurrence was the injury to the plaintiff, not the original distribution of the drug:  Distillers 130 CLR at 10, 20.  Menzies J came to the same conclusion: Distillers 130 CLR at 6.

Use of common law concepts to characterise what is a distinct occasion

66                  The shipowner argued that the question for determination was whether, after the fouling, there was any act, neglect or default which gave rise to a claim distinct from any claim that had by then arisen by reason of the fouling.  It argued that this formulation of the test derived from the terminology and concepts employed in the Convention and recognised an assumption that the fouling was attributable to some act, neglect or default.  It contended that because the purpose of the Convention was to limit the liability of a shipowner, the concept of claims arising “on any distinct occasion” as used in Arts 6 and 9 of the operated so that  “… if a particular act, neglect or default is sufficient to give rise to a claim (notwithstanding subsequent acts, neglects or defaults) then it can be said to arise on one distinct occasion.”

67                  I am of opinion that this argument distorts the proper approach to the construction of the Convention.  The degree of connection between events or circumstances is often examined in order to ascertain whether one is a cause of another.  The Convention uses the concept of all claims arising on any distinct occasion to identify the degree of connection between an act, neglect or default and the causation of the class of claims for which a shipowner may limit its liability.  The determination of whether a claim arises on a particular occasion, involves the application of a test for causation and the formation of a judgment about the existence or absence of the fact of a sufficient connection between the two.  That judgment is arrived at after a balancing of fact and degree.  Its formation involves the selection, consciously or unconsciously, of values used to establish the presence or absence of a sufficient degree of connection between the claim and the occasion.

68                  In one sense, the law of the forum (lex fori) will be relevant in ascertaining whether a claim arises on the asserted occasion.  But, because the Convention applies internationally, the determination whether something is a claim at all may be affected by the law of the places where the act, neglect or default occurred or it may be affected by the law of the place where the person making the claim asserts that he, she or it has suffered a loss.

69                  The means of classification of whether a particular assertion of a right to sue a shipowner amounts to a claim for the purposes of the Convention must be flexible enough to allow for variations between the laws of any State Party at which the ship may be arrested or attached.  This flexibility or extensive reach of the word “claims” serves the purposes of the Convention, because the establishment of a limitation fund in accordance with Arts 11 and 13 will create a fund as a substitute for the res as a means of satisfying those claims.  This is complemented by Art 14 which provides that the rules relating to the constitution and distribution of a fund and all rules of procedure in connection therewith are governed by the law of the place (i.e. of the State Party) where the fund is constituted.

70                  The right to make a claim on the fund may be a matter of substantive, not procedural, law.  If a ship were arrested and then established a fund under the Convention in a jurisdiction that did not have a substantive law equivalent of liability for many claims that arose in the place where the casualty occurred, the procedural law of the place of the fund could not be used to frustrate claimants’ substantive legal rights to recover in the place of the occurrence.  Obviously, if the law of the places of each of the occurrence and the fund are those of the same State Party to the Convention, as in these proceedings to establish a fund here, then the substantive and procedural law to determine whether a claim has been caused by an act, neglect or default should be the same as the test in the Convention for a claim arising on a distinct occasion.

71                  The shipowner contended that the factual or temporal limit of an occasion could not be less than what was sufficient to give rise to a claim.  It argued that the words “arise” or “arising” in Arts 6 and 9 embodied the key concept in the Convention.  Thus, the shipowner argued that all the claims that have been made, or are capable of being made, by Esso, BHP and third parties based on what occurred on 13 December 2008 arose from the one occasion of the initial fouling of the pipeline.  It argued that this was the occasion that gave rise to all possible bases of its liability including the subsequent engine movements, the rupture of the pipeline and its further deformation.  To support this argument the shipowner relied on common law decisions on the tort of negligence which held the original tortfeasor liable, first, for independent, but reasonably foreseeable, actions of third parties as in cases such as Chapman v Hearse (1961) 106 CLR 112 at 120-122;  Mahony v J Kruschich (Demolitions) Pty Ltd (1985) 156 CLR 522 and Dovuro Pty Ltd v Wilkins (2003) 215 CLR 317 at 336 [60] per Gummow J or, secondly, an unforeseeable consequence of a reasonably foreseeable breach of a duty of care:  Hughes v Lord Advocate [1963] AC 837 at 855-856 per Lord Guest.  The shipowner argued that these cases supported the characteristics of the three subsequent engine manoeuvres from 15:46:01 on which BHP and Esso relied as distinct occasions separate from the initial fouling, as merely a consequence of, and so assimilated into, the occasion of the initial fouling.

72                  Esso and BHP contended that the real issues were whether the subsequent acts, neglects or defaults on which they relied were the inevitable consequence of the initial fouling or whether the ship, her master and the pilot had the time and opportunity to avoid each of them.  They relied on the common law authorities on “a distinct occasion” set out above.  They argued that questions of causation and the application of the domestic law of negligence were matters that would need to be decided in proceedings seeking to establish the liability of the shipowner.  Esso and BHP argued that the common law cases had considered whether the ship could have taken a different course of action as being determinative of whether the subsequent claims were part of the original occasion or arose on a distinct occasion, that was distinct because the master or pilot had committed a new error or had failed to take a step that he could have, and should have, taken to avoid the second or subsequent casualty.

73                  However, even the common law cases recognise that a second tortfeasor’s involvement in a subsequent foreseeable injury or the consequences of a breach of a duty to take reasonable care following the first tortfeasor’s negligence does not necessarily create a causal connection between the two justifying the first tortfeasor bearing liability for all the subsequent consequences.  In Mahony 156 CLR at 528 Gibbs CJ, Mason, Wilson, Brennan and Dawson JJ said that a negligent tortfeasor did not always avoid liability for the consequences of a plaintiff’s subsequent injury, even when that subsequent injury was tortiously inflicted.  The original tortfeasor’s liability depended on whether or not the subsequent tort and its consequences were themselves “… properly to be regarded as foreseeable consequences of the first tortfeasor’s negligence”.  They said that the boundary of the damage for which the tortfeasor was liable in negligence could be drawn either because the relevant injury was not reasonably foreseeable or because the chain of causation had been broken by a new intervening act.  But the original tortfeasor will be liable unless a line could be drawn clearly separating his or her negligence from that of a subsequent tortfeasor adding:

“Whether such a line can and should be drawn is very much a matter of fact and degree.”

74                  In Chapman 106 CLR at 122 Dixon CJ, Kitto, Taylor, Menzies and Windeyer JJ said that:

“… the term ‘reasonably foreseeable’ is not, in itself, a test of ‘causation’;  it marks the limits beyond which a wrongdoer will not be held responsible for damage resulting from his wrongful act.”

They concluded that the criterion of reasonable foreseeability embraced all foreseeable intervening conduct whether it be wrongful or otherwise:  Chapman 106 CLR at 125.  And, the common law test for causation, in general, and in Australian law, in particular, is in many cases a question of fact:  March v E & MH Stramare Pty Ltd (1991) 171 CLR 506 at 515 per Mason CJ, 523 per Deane J, 524 per Toohey J and 525 per Gaudron J.

75                  I reject the shipowner’s construction.  First, common law cases dealing with the classification of a tortfeasor’s liability for events subsequent to an initial tort are not appropriate to use in construing the operation of the Convention or the meaning of the concept of a distinct occasion used in it.  That is not to say that the law of the place of the wrong (lex loci delicti) is irrelevant to the characterisation of what is or is not a “claim”.  Claims ordinarily can arise only under the law of the place of the wrong and that law, in general, must provide the means by which the legal classification of a claim will be ascertained.

76                  Secondly, the right to limit liability provided by the Convention is associated with an act, neglect, default, occasion or occurrence.  These are concepts of a concentrated or discrete event quite different to an ambulatory or wide ranging spectrum of activities.  In one sense, a voyage between two ports can be an occasion.  In another sense, we speak of the “occasion” of a person’s marriage.  That meaning can extend to everything that happened on the day including the ceremony, wedding reception and all circumstances around those two events.  Yet the Convention could not have been intended to have an unlimited ambit.  Indeed, Art 4 permits the shipowner’s right to limitation to be broken if “… the loss resulted from his personal act or omission”.  This suggests that the search for an appropriate discrimen for a “distinct occasion” should focus more narrowly on the connection between the loss (or claim) and the act or omission (in Art 4) or “act, neglect or default” in Arts 6 and 7.  The fact that further damage may be foreseeable after one casualty eventuates because it is foreseeable that other acts, neglects or defaults may occur subsequently, does not necessarily entail that the later event or the claim to which it gives rise arose on the happening of the former or is sufficiently connected to be assimilated into it.

What is a distinct occasion?

77                  Here, the question of classification of a particular claim as arising on a particular distinct occasion must be considered in light of:

(1)        the evident intention of the Convention to protect shipowners from ruinous claims;

(2)        the Convention’s invocation of an occurrence or distinct occasion as the circumstance from which claims arise.

78                  I am of opinion that a claim arises on a distinct occasion within the meaning of the Convention in the following way.  Where a single act, neglect or default of a shipowner places him in such a relationship that, as a matter of commonsense, it is a cause of loss or damage suffered by a third party, that third party will have a claim under Art 2 of the Convention.  And, such a claim will be caused by an occurrence and, so, will arise on that distinct occasion for the purposes of Arts 6, 7, 9 and 11.

79                  But where a subsequent act, neglect or default of the same shipowner separately operates to cause different or separately identifiable loss or damage to the same third party, or to others, then a new claim or claims will arise on that later distinct occasion.  The latter occasion is distinct because, first there is a new event (the separate act, neglect or default), secondly, there is new loss or damage and thirdly, the new cause is, as a matter of commonsense, not a necessary or inseparable consequence of the earlier act, neglect or default.

80                  Thus, whether one occasion is distinct from another will depend upon whether the causes of the claims that arise from each act, neglect or default are sufficiently discrete that, as a matter of commonsense, they can be said to be distinct from one another.

81                  It is possible in many accidents to see a concatenation of errors that lead to the particular casualty.  When more than one act, neglect or default occurs causing or giving rise to liability of a shipowner to one or more third parties, it will be a question of fact and degree for a court to determine whether these circumstances amount to more than one distinct occasion.  One error can cause many persons to suffer loss or damage;  e.g. when a passenger ship sinks causing many losses of life or personal injuries.  Each casualty or injury may have its own associated causes, e.g. an explosion of a gas cylinder negligently handled by the crew that may cause a chain of events to occur.  There may be a lack of life boats, a fire and a pierced hull letting in seawater, causing differing problems for the passengers in different parts of the ship.  But the distinct occasion is the explosion, even though the losses that arise do so by different related causes.  Whether there is a dividing line between a number of acts, neglects or defaults that occur in a particular situation, so as to constitute more than one distinct occasion, is question of fact and degree.

82                  The sufficiency of the connecting and dissociating factors has to be assessed in an overall factual context.  This is another instance of a need to search for a cause in order to characterise the legal consequences that flow from that cause.  The Convention requires that the cause from which each claim arises be attributed to a particular factual context, namely the distinct occasion resulting from an identified act, neglect or default.  The Convention also has regard to its purpose of limiting the liability of the shipowner to the amount of the fund constituted under the Convention for the purposes I have described above.

83                  The Convention focuses attention on an act, neglect or default that constitutes the occasion from which claims arise.  In such a case, it is often the last avoidable error that, not having been averted, will be seen as the commonsense cause of the casualty.  Then, the proceeding errors are not the relevant act, neglect or default because they do not, as a matter of commonsense, necessarily lead to what followed.  There is a difference between this and the “but for” test of causation because the Convention seeks to broaden the protection afforded to the shipowner by the width both of the meaning of “claims” in Art 2 and the concept that such claims “arise” on a distinct occasion.

84                  The purposes of the Convention would be subverted if shipowners or claimants could identify any antecedent, act, neglect or default as giving rise to a claim merely because of its logical association with the loss or damage and so invoke or break the limitation provisions of the Convention by this artificial mechanism.  The act, neglect or default must, as a matter of commonsense, be sufficiently discrete from an earlier occurrence (or act, neglect or default) and sufficiently related to the particular loss or damage the subject of a claim within Art 2 to make the claim one that arises from that occasion.

85                  On the other hand, the drafters of the Convention would have been well aware that within a short space of time several errors may be made in the navigation or conduct of ships, each one of which could be a cause of different losses or damage and each one of which was avoidable.  In such a case, each occurrence giving rise to the separate liabilities will be (see Art 11(1)) the avoidable act, neglect or default, not the associated conduct which, itself gave rise to the other claims.

86                  The shipowner also contended that here, everything following the fouling was attributable to what it had done to cause that original act, neglect or default.  It argued that this approach gave a purposive construction to the Convention and avoided Esso and BHP’s attempts to subvert the unbreakable limit of its liability.  I reject that argument.

87                  In my opinion, the Convention provides a very wide protection for shipowners.  This is reinforced by the use of the words “arising” and “arise” in connection with the claims that are to be limited to each distinct occasion.  The word “arise” is a word of wide connotation and will embrace a variety of circumstances and causes leading up to the identification of “a distinct occasion”.  However, the Convention also recognises that there will be circumstances in which more than one distinct occasion will occur from which claims against a shipowner can arise.  It will be a question of fact in each case as to whether the particular act, neglect or default said to give rise to “a distinct occasion” has been identified.  This is not to encourage the unmeritorious and over analytical dissection of circumstances in each marine casualty in which a shipowner seeks to limit its liability.  Where a single act, neglect or default gives rise to or causes a particular incident, such as a single collision with another vessel, a wharf, or even a pipeline, commonsense will say that minute dissection of the circumstances will not lead to identification of more than one “distinct occasion”.

88                  Here, the pipeline did not rupture on the initial fouling.  It stayed intact for at least another 35 minutes.  There were three subsequent manoeuvres of the ship using her engine that may have caused new loss or damage.  These engine movements and the other circumstances in which the pipeline came to rupture and the Mordialloc end to be pulled out of the trench and bent in almost a 90° angle on the seabed, raise issues of fact and degree requiring an examination of whether there were any further distinct occasions in which damage occurred.  I will apply these principles in considering the facts below.

Some of the primary sources of what happened

89                  The ship had a number of instruments that recorded important information about its handling and movement on 13 December 2008.  The course recorder, engine telegraph log, ship’s log and a number of other records including harbour control’s voice data records between the ship and harbour control were available.

90                  But some records were not available to the parties, including the ship’s records of helm or rudder orders (what the orders were, when and for how long they lasted) in the voyage data recorder, and of wind direction, speed and duration.  One of the nautical expert witnesses said that this data was “absolutely necessary to understand, from the master’s perspective, what was happening with the starboard anchor”.  Those records presumably had been taken by Australian Transport Safety Bureau while investigating the incident in the exercise of its powers under the Transport Safety Investigation Act 2003 (Cth).  However, no application was made to obtain that information under the powers given to the Court in Div 1 of Pt 6 of that Act, esp in s 53(4).  The absence of this important, objective information is regrettable.

91                  An important record that was used frequently in the trial was the Automatic Identification System computer record which the parties described, as will I, as “the AIS video” of APL Sydney’s movements.  This was in the form of a computer disk recording a video display of information relayed automatically by many seagoing ships.  The AIS video depicted a physical shape of the ship on the Admiralty chart background as she moved in Port Phillip Bay.  This was an approximation by the computer program of where the ship appeared to be, based on the location of the transmitter (in the stern superstructure) and other AIS data transmitted, such as the heading, speed over the ground (in knots), latitude, longitude, course over the ground, rate of turn and time.  The AIS video display was available to, and used by harbour control contemporaneously with the events as they occurred.  However, this display was not available to those on the bridge of APL Sydney, although most of the information from which it was composed was available there but in disparate form on the various instruments that collected the data.

92                  Each of the available records indicated a time for the various entries.  These times, however, were not all co-ordinated to the exact same time.  Thus, a voice data record of harbour control may nominate a time for a conversation that will match to an event in some other record which gives a different time.  In general, the parties proceeded on the basis of the time in whatever record was used, as will I.

Concurrent expert evidence

93                  There were eight expert witnesses who gave oral evidence over five separate areas of specialised knowledge.  Each had prepared at least one principal report, some prepared a responsive report.  In the pre-trial phase, I directed that the experts in each relevant discipline should confer together, without the parties or their lawyers, and prepare a joint report that set out the issues on which they agreed and those on which they disagreed, giving brief reasons for their differences.  I also directed that the experts, in each discipline would give evidence concurrently.  I will briefly describe the process:  see also a DVD:  Concurrent Evidence – New methods with experts (2005):  produced by the Judicial Commission of New South Wales and the Australian Institute of Judicial Administration, and a paper I wrote:  Expert Evidence in Copyright Cases – Concurrent Expert Evidence and the “Hot Tub”  15 October 2009, Sydney, 4th Biennial Copyright Law and Practice Symposium [http://www.fedcourt.gov.au/aboutct/judges_papers/speeches_raresj10.html].

94                  The joint reports were extremely useful in crystallising the real questions on which the experts needed to give oral evidence.  Experience in using this case management technique generally demonstrates considerable benefits in practice.  First, the experts usually will readily accept the other’s opinion on the latter’s assumptions.  This position is often lost in long reports that debate, not that opinion, but the assumptions which, in turn, usually depend on the facts that need to be found.  Secondly, the process then usually identified the critical areas in which the experts disagreed.

95                  When each concurrent evidence session began, I explained that the purpose of the process was to engage in a structural discussion.  Each expert was asked to summarise what he (all were male) thought were the principal issues between him and his colleague(s).  Each was free to comment on or question his colleague on what he had said both during the introductory part and throughout the process.  After each expert had outlined the principal issues (usually one did this and the other agreed that it was a fair summary or added some brief further remarks), counsel identified the issues or topics on which they wished to cross-examine.  I then invited whichever counsel wished to begin questioning to do so.  The experts sat at a table where they had ample room to place their reports and materials.  They had a single microphone for whomever was speaking, so that the transcript would record the relevant evidence and they would exercise self-discipline in responding.  Often when one had given an answer, the other would comment, or agree, thus narrowing the issues and focussing discussion.  From time to time counsel could and would pursue a traditional cross-examination on a particular issue exclusively with one expert.  But, sometimes when one expert gave an answer, counsel, or I, would ask the other about his opinion on that same question.

96                  The great advantage of this process is that all experts are giving evidence on the same assumptions, on the same point and can clarify or diffuse immediately any lack of understanding the judge or counsel may have about a point.  The taking of evidence in this way usually greatly reduces the court time spent on cross-examination because the experts quickly get to the critical points of disagreement.

97                  Here, the experts and their fields were:

Master Mariners

·       Capt John Third

·       Capt Dale Cole

·       Capt Kerry Dwyer

 

 

Naval Architects

·       Professor Martin Renilson Professor of Hydrodynamics at the Australian Maritime College

 

·       Mr Holge Kelle Senior Naval Architect Worley Parsons Group

 

 

Structural Engineers

·       Dr Bent Lyngberg Group Technical Director of Braemar Staege Ltd

 

·       Dr John Price principal of Integrity Partners

 

 

Metallurgical Engineers

·       Mr Barry Gartner managing director of AMAT Materials Engineering Pty Ltd

 

·       Mr Kelle

 

 

Mechanical Engineers

·       Dr Lyngberg

·       Dr Price

 

A number of other experts gave written reports that were accepted without the need for cross-examination.

The anchoring system on the ship

98                  The ship’s two anchors weighed about 8.325 tonnes each.  They were Admiralty cast type 14.  Each anchor had a shank to which was attached, at the top, the anchor chain and at the base, two flukes that could move about 35° away from the shank.  (A depiction of the anchor is in Figure 5 at [342] below.

99                  The anchor chain or cable was attached to the anchor by an anchor crown “D” shackle, to which was attached a separate “D” type shackle, then an open end link and a swivel.  This configuration enabled the anchor to be detached from the anchor chain when the ship was to be moored to a buoy using the chain.  The anchor chain consisted of 81mm stud chain links and kenter links.  These links are heavy in themselves.  The kenter links were placed at intervals of 27.5 metres.  That was the length of one shackle (equivalent to 15 fathoms), a measurement used by a ship’s crew to describe the length of anchor chain that had been deployed or is in the water (i.e. ignoring the chain between the water surface up to the deck).  The first shackle length measured from the anchor crown “D” shackle consisted of the particular type of securing links described above and 10 stud chain links making up a section of chain called a “forerunner” or “ganger” totalling 4.995 metres.  This was connected by a kenter link to a further 22.505 metres of 69 standard chain links to what was called the first shackle’s kenter link.  This was located 27.5 metres, or one shackle away from the top of the anchor shaft.  The starboard anchor chain was 12.5 shackles long.  The total weight of the anchor and chain was about 61 tonnes.

100               The anchor chain or cable is stored in a chain locker.  The extreme end of the chain from the anchor is connected to the ship by the bitter end.  The bitter end is a quick release mechanism.  It is intended to be used in an emergency, in effect, to release the ship from the anchor by letting the anchor chain pull itself free from the ship.

101               The windlass is an hydraulic winch system that mechanically raises and lowers the anchor.  The windlass was capable of heaving (or lifting) a maximum of three shackles of anchor chain per minute.  The windlass is connected to a cable lifter by a drive shaft and clutch mechanism.  The cable lifter is also called the gypsy.  It has sprockets that engage with the chain links of the anchor cable.  The clutch mechanism on APL Sydney was hydraulic.  The windlass has a manually operated drum brake.  The ship’s drum brake is operated by a wheel which the bosun, or another crew member, can turn on or off while standing on a raised platform on the aft (i.e. stern) side of the gypsy.  The cable then passes over a fairlead which is a structure that aligns the cable, over the forecastle deck, with the gypsy and the hawse pipe.  The hawse pipe passes through the hull.  The cable passes down the hawse pipe towards the water.  The shaft of the anchor is drawn up into the hawse pipe.  The fairlead usually has a cable stopper which is a jamming device that can be deployed between cable links to stop the further release of the cable.

102               The controls for the windlass, other than the windlass brake, were located further to starboard at the end of the windlass drum.  The cable passes down from the gypsy along a pipe towards the anchor chain locker.  The crew can see the chain passing up or down the gypsy into that pipe.

Anchoring

103               Two features of anchoring are important in order to understand the evidence.  These are the concepts of bringing a ship up to the anchor and of the catenary effect on the cable when it has been paid out in the process of anchoring the ship.

104               A ship is brought up to the anchor when at the completion of the anchoring process she holds fast.  Capt Third explained that when the master orders the mate in the forecastle party to prepare to let go an anchor, the crew first clears away all securing arrangements for the anchor and the cable that are in place while  the ship is at sea.  The anchor is then ready in the hawse pipe.  When the order is given to let go the anchor from this “cleared away” position, a clutch mechanism engages the cable lifter to the windlass motor.  Then, the windlass brake is released so that the motor drive takes the weight of the anchor and cable.  The mate, usually the chief officer, orders the bosun or crew member to pay out the anchor cable.  The windlass motor operates in reverse.  The bosun usually then uses it to pay, or let, out enough cable to lower the anchor from the hawse pipe to just above the waterline.  Then the handbrake on the gypsy is applied manually.  The windlass motor then moves a small amount to release the pressure on the clutch so that the brake, not the windlass, now holds the anchor in place.  The clutch is then disengaged.  On the next order, the bosun lets go of the anchor by releasing the brake.  The mate and the bosun observe how many shackles are let go and, when the required amount is reached, the brake is reapplied.

105               Thus, the weight of the anchor and cable paid out draws down more cable simply by the force of movement of the ship away from the anchor.  When the required length of cable has been let go, the chain can be secured using a stopper (or guillotine stopper) mounted above the fairlead between the windlass and the hawse pipe.  And, ordinarily, the master will drop the anchor while the engine is astern and so the ship is moving astern, away from were the anchor will have been buried in the seabed.

106               As David J House explained in his work Anchor Practice, quoted by Mr Kelle, once a predetermined amount of cable has been paid out, the master will then order the engine to stop:

“The cable will have been allowed to run and the brake would be applied to check the amount of scope.  This should serve the purpose of digging the anchor into the seabed, over the ground.  The ship is described as being ‘brought-up’ to the anchor and it would be the duty of the anchor party officer to determine when the vessel is ‘brought up’ and not dragging her anchor.”

107               The ship is “brought up to the anchor” when the anchor shows resistance to the movement of the ship, such as by pulling the bow into the wind.  That is, when the cable (after the brake has been applied) rises up from the surface of the water towards the hawse pipe and the ship moves forward towards the anchor (i.e. the direction in which the cable is leading) causing the cable to drop so as to make a catenary – i.e. a curve formed by the chain hanging freely between the hawse pipe and either the top of the anchor shaft or a point on the seabed where the chain rises toward the surface even though there is more chain between there and the anchor lying on the seabed.

108               Earlier, Capt House said:

“The fundamental principle of anchoring is that it is the weight of cable and the lay of the ‘scope’ that anchors the vessel successfully and not just the weight or design of the anchor. …’

109               Heaving the anchor, however, always requires the use of the windlass to raise the heavy cable paid out and anchor mechanically.  The crew prepares by turning the windlass motor to operate in the opposite direction to when it lowered the anchor.  The clutch mechanism is again brought into alignment with the cable lifter.  Then, when the order to heave anchor is given, the bosun engages the clutch and the windlass takes the weight of the cable.  The stopper is removed.  Next, the brake is released so that the cable then commences being heaved in by the windlass motor.

110               The speed of retrieval depends on a number of factors, including the prevailing weather conditions, the direction in which the cable is leading from the bow (i.e. it need not be straight in front of the bow but can be at any angle including leading astern or across the bow) and the movement of the ship (e.g. the master may try to move the ship towards where the cable is leading to assist in its recovery).

111               Ordinarily, when a ship is brought up to the anchor the cable will not stretch out taut in a straight line between the hawse pipe and the anchor.  This is because the weight of the chain will cause it to curve in a catenary even if the ship is pulling astern from the anchor.  Part of the process in anchoring successfully requires sufficient cable to be paid out so that, when the vessel swings in the wind or is lifted by the waves, the chain will not pull the head of the anchor shaft upwards toward the surface.

The overall context of the four asserted distinct occasions

112               The shipowner accepted that, for the purposes of invoking its right to limitation under the Convention, claims against it arose from the initial fouling.  It contended that this circumstance and its sequels were all but one “distinct occasion”.  I will consider each of the four pleaded scenarios that Esso and BHP asserted amounted to a distinct occasion.  They also contended that these scenarios were not exhaustive of other possible distinct occasions, but did not deal with this any further.  They, too, accepted that the initial fouling was a distinct occasion but they contended that immediately after it a second distinct occasion occurred when at 15:46:01 the master put the engine dead slow astern.  The following key engine orders are associated with the four alleged distinct occasions which I consider below:

                        The first alleged distinct occasion:  The Anchor fouls the pipeline 

                                   

15:30:27

Dead Slow Ahead

15:33:31

Stop

15:35:55

Dead Slow Ahead

15:36:31

Stop

15:40:29

Dead Slow Astern

15:41:17

Stop

15:43:09

Dead Slow Astern

15:43:35

Slow Astern

15:44:53

Stop


                        The second alleged distinct occasion: The astern order after the fouling

                       

15:46:01

Dead Slow Astern

15:47:31

Stop


                        The third alleged distinct occasion:  The pipeline ruptures 

16:19:51

Dead Slow Ahead

16:20:09

Stop

16:20:13

Dead Slow Ahead

16:21:27

Stop


                                    The fourth alleged distinct occasion:

                        The Mordialloc end of the pipeline is further bent

                                               

16:27:59

Slow Astern

16:33:03

Stop

113               A considerable time during the hearing was taken up with the events leading up to the initial fouling.  Apart from setting the scene for what followed in the next hour, it is not necessary to explore the detail of that evidence in great depth.

114               Pilotage was compulsory in the Port of Melbourne.  However, by force of s 410 of the Navigation Act 1912 while the pilot may have the conduct of the ship, he or she is subject to the authority of the master.  The section provides that the master is not relieved from responsibility for the conduct and navigation of the ship by reason only of the ship being under pilotage.  And, that responsibility is determined according to the principles applicable to voluntary pilotage, where the pilot is treated as the servant of the shipowner, despite pilotage being compulsory (s 410(2)):  Braverus Maritime Inc v Port Kembla Coal Terminal Ltd (2005) 148 FCR 68.

115               The master of APL Sydney was Capt Xu Pingfeng.  Capt Xu had 22 years seagoing experience and had held a master’s certificate for seven years.  He had been in command of container vessels since 2002.  He signed on as master of the ship for the first time just before the shipowner took delivery of her from the shipyard in October 2006.  He remained as master until November 2007 and signed on again in August 2008.  He had called at Melbourne with the ship on at least seven prior occasions.

116               Capt Xu gave evidence mostly through an interpreter.  He had basic functional English, but not sufficient to fairly or fully give evidence.  At times he became frustrated while being cross-examined.  He was also very upset and embarrassed by the events leading to the rupturing of the pipeline.  Although Esso and BHP criticised Capt Xu’s evidence and asserted that he was unresponsive, I formed the view that he was a practical and generally honest man who nonetheless found the experience of giving evidence trying and difficult.  As will appear, he had some justification for his frustration and his actions, because of the conduct of both harbour control and, the pilot, Capt Craig Murchie.

117               On 13 December 2008 at 11:12 the ship was at the pilot station off Port Phillip Heads in the Bay.  Capt Murchie came on board at around noon and presented a passage plan to Capt Xu.  He considered and then accepted it.  The pilot advised the master that the ship would anchor in the outer anchorage until a berth became available.  He told Capt Xu that the holding ground was better in the outer anchorage than in the inner one.  Capt Murchie did not initially identify the precise anchorage position.  While the ship was proceeding to the outer anchorage, the pilot marked his proposed anchorage on the ship’s Admiralty chart.  It was to the west of the Fawkner beacon that was located in the middle of the main Port Melbourne Channel.  At the same time Capt Murchie marked the position of another, anchored ship, HS Electra which was south of the area he designated for APL Sydney.  Capt Murchie told the master to keep a close watch on the anchor position.

Dropping Anchor

118               During the early afternoon the weather worsened.  The wind was from south south west to south west up to force 8 or 9 on the Beaufort scale.  This was gale force.  The seas were stormy with wave heights of 2 to 2.5 metres.  At around 14:00 the pilot directed that the ship’s starboard anchor be readied for anchoring.  The master gave the order to the chief officer, Qiu Jian Gio.  Capt Xu was planning to use the ship’s bow thruster to bring her bow into the wind.  The chief officer and the bosun, Liu Weihua, went to the forecastle.  The bosun had a long experience of over 30 years at sea.  He was standing by the starboard windlass.  Then at 14:12 the bosun reported to the master that the starboard anchor was ready to be lowered.  There was no direct line of sight from the bridge to the forecastle or bow and the master, and bridge, communicated by radio with the forecastle party. 

119               During their discussions Capt Murchie warned the master of the presence of the pipeline, but Capt Xu was already aware of it.  By 14:30 the ship had moved to a position in the outer anchorage almost north of HS Electra and to the west of the Fawkner Beacon.  Her engines were then ordered to stop.  The water depth was about 15.5 metres and the ship was heading in an easterly direction (98°).  The wind was south westerly.  The bridge was about 0.6 of a nautical mile south west from the chartered position of the pipeline.  The pilot advised the master to drop the starboard anchor with five shackles in the water.  This command meant that when the ship ultimately was brought up to the anchor, five complete shackles of chain would be in the water, and there would be additional anchor chain between the gypsy and the surface of the water.  Initially, in accordance with standard anchoring procedure, three shackles were in the water and the ship’s engine was ordered to stop but remained on standby.  After the chief officer had seen these paid out, then he ordered the bosun to release the next two shackles.

120               Normally, Capt Xu would have put the ship’s head into the wind.  That is, he would have allowed the stern to swing round about 180° so that the bow was pointing into the direction, here south west, whence the wind was blowing.  The wind would thus blow against the superstructure housing the accommodation and bridge pushing the ship north east away from the anchor.

121               However, Capt Murchie told Capt Xu that he wanted to create a lee (an area sheltered from the wind) so that he could disembark onto the pilot boat, George Tobin.  The pilot said that he did this because the wind was very strong and the waves were about 2 metres high.  The master said, and I find, that the pilot left the bridge very quickly after the order to drop the anchor.  Capt Murchie told the master that he wanted to disembark as soon as the anchor had been dropped.  He radioed the pilot boat at 14:28 saying that the anchor was dropping “… and I’m on my way down”.  The pilot disembarked and the pilot boat was on her way by about 14:36.  Capt Xu could not control Capt Murchie’s decision but the master would have liked the pilot to stay for safety reasons, given the bad weather conditions and his lack of familiarity with the anchoring position.

122               I am satisfied that the pilot’s departure from the bridge before the ship had been brought up to anchor, particularly in the prevailing conditions, was a breach of his obligation to undertake the compulsory pilotage of the ship.  Unless she had been brought up to anchor, the ship was still required to be under pilotage.  If anything went wrong with the anchoring, as it did in this case, the master would be in the invidious position of potentially breaching the compulsory pilotage requirements of the Port of Melbourne were he to use the engines to manoeuvre the ship or attempt to re-anchor her.  That position was exacerbated because the pilot directed the anchoring to be south west of the pipeline, a valuable and potentially dangerous infrastructure resource in the port, in a gale where the wind would blow the ship towards the pipeline if she were not brought up to anchor.  That is just the position in which Capt Xu was placed, contributed to by the conduct of the pilot and later by the port authorities.

123               Capt Xu was aware that the pipeline carried gas and that if the ship got too close to it a dangerous situation could arise.  By shortly after 14:30 five shackles of cable were in the water.  At about 14:42 the chief officer reported to the master that the cable looked normal, there was no vibration and the anchor appeared to be holding.  The cable was leading out very tight at about 1 o’clock (treating 12 o’clock as straight ahead).  The bow had swung round from the east to the south east.  The chief officer said that the wind was strong and that there were waves when he reported this.  He was standing at the top of the starboard forecastle looking through the bulwark in the direction that the cable led.  The fifth shackle’s kenter link was a little below the water and the chain was very tight.  The chief officer said that it was hard for him to judge in his position if the ship had been brought up to the anchor.

The Ship Drags Its Anchor

124               After the chief officer reported to the master at about 14:42, Capt Xu was concerned that the radar indicated that the ship had a minus speed and was yawing.  These were indications that the ship had not been brought up to the anchor.  Had she been so, her bow could be expected to be heading into the wind and the radar would not have indicated any speed.  Capt Xu plotted the vessel’s position at 14:45 and concluded that she was dragging her anchor.  At that time her stern was close to the northern boundary of the outer anchorage.  The ship was moving north east, beam onto the wind (i.e. the wind was blowing the ship which was facing the length of its 231 metre long hull into the wind).  He kept the engine on standby while the ship was being blown towards the north east closer to the pipeline, as the anchor continued to drag and the vessel yawed.

125               At about 15:01 Capt Xu radioed harbour control and sought permission to heave the anchor and to re-anchor further away from the pipeline.  The ship had moved further to the north east.  Just before she reached this position, the anchor had appeared to hold briefly and the bow swung round to face south south west into the wind.  Capt Xu explained to harbour control that he wanted to heave the anchor because the ship was near to the pipeline and the position was dangerous for his ship.  Harbour control had the ability to observe the ship moving on the AIS screen.  The AIS video showed that the ship had passed stern first completely over the northern boundary of the outer anchorage.  There is no evidence of what, if anything, harbour control did in order to safeguard the pipeline other than its radio communications with the master and the pilot.

126               Harbour control told Capt Xu that his request was refused and he could not heave the anchor unless a pilot was on board.  He said that he did not need a pilot and could do this himself, he just wished to move about half a mile ahead (i.e. south west).  Harbour control said:

“Ah, yes, there is a problem you shifting yourself from the anchorage.  You stand by until we send you a pilot.”

127               Capt Murchie heard the master’s request on the pilot boat’s radio.  It was then at Williamstown to the north of the ship.  The pilot boat had a less sophisticated AIS screen than the one at harbour control.  Harbour control confirmed the original anchoring position with Capt Murchie and saw on the AIS screen that the ship had dragged anchor to the northern boundary of the outer anchorage.  The pilot said he would return to APL Sydney and at 15:05 harbour control told the master to maintain position.  Capt Murchie thought, correctly, that it would take at least 40-45 minutes for the pilot boat to return to the ship in the prevailing weather conditions.

128               Capt Xu believed that harbour control could monitor, and was monitoring, the vessel’s position as she was moving under the force of the weather ever closer to the pipeline.  That was not only a reasonable belief, it was correct.  By 15:07 Capt Murchie told harbour control that the stern of the ship appeared on the pilot boat’s AIS screen to be over the pipeline.  Harbour control told him that she was north of the boundary line but south of the pipeline.  It regarded the situation of APL Sydney as a “drama”.  Capt Murchie cautioned about the possibility of her dragging over the pipeline.  Around this time, the AIS showed that the anchor appeared to be holding and the bow was facing towards the south.  At 15:16 harbour control told Capt Xu that the pilot would arrive in about 30 minutes.

129               Despite the master radioing his concern at 15:01, the evidence does not reveal that harbour control took any active step for about 25 minutes to deal with the situation, far less to assist or facilitate the master taking steps to prevent the ship getting too close to the pipeline.  The master of a ship always has, and must have, the ultimate responsibility to ensure the safety of his ship, crew and other people and property vulnerable to the ship.  However, in areas of compulsory pilotage, a master is placed in a difficult position.  First, he or she will not be familiar with the way in which every port State into which he or she has sailed will enforce its compulsory pilotage laws.  A master will be cautious before deliberately overriding a port State’s harbour or passage control authority’s direction for the legitimate fear that he or she could be arrested and jailed, however reasonable or necessary the overriding manoeuvre was.  Thus, harbour or passage control authorities have concomitant responsibilities to pay proper regard to requests of masters to depart from the usual course.  Secondly, a master will not usually be familiar with the area of compulsory pilotage, because the very reason for the regime of requiring a pilot there is the difficulty of the locale for navigation and manoeuvring.  A master is entitled to expect reasonable assistance and co-operation from the harbour or passage control authority when, as here, a significant difficulty arises.

130               There is no evidence of why harbour control reacted, or failed to react, to the developing situation.  With the benefit of hindsight, at least after 15:01, it would have been obvious to a person monitoring APL Sydney on the AIS screen at harbour control that she was inexorably being dragged in the gale towards the pipeline.  Yet, only when Capt Xu, again, radioed at 15:25 and was told that the pilot would not return until about 1600, did harbour control give him permission to use the engine and move the ship.  The weather was worsening.

131               Shortly after 15:25 the master ordered the chief officer to heave anchor.  He passed that order to the bosun, who was operating the windlass and standing about 10 or so metres away.  The chief officer and the bosun counted the shackles of cable as they were heaved in by the windlass.  In the meantime, after the master ordered the engine dead slow ahead at 15:26:45, the ship began to move.  First, the bow began to turn into the wind and the vessel moved ahead toward the south west or south south west.  The master ordered the engine to stop at 15:28:01.  He confirmed to harbour control at 15:29 that the anchor was being heaved in and that he was planning to move ahead about half a mile.  Harbour control confirmed that he had permission to move half a mile to the south east.  He said “at this time our vessel’s position danger to the … pipeline” and asked to move a mile forward.  Harbour control agreed.  At 15:30:27 the master again ordered the engine dead slow ahead.  The forward party continued to raise the anchor.

132               The windlass could retrieve cable at a maximum speed of 1 shackle every 3 minutes (or 9 metres per minute) in ideal conditions.  This information was on the ship’s wheelhouse poster.  Mr Kelle, however, calculated that, on 13 December 2008 with 5 shackles in the water the average speed of retrieval would have been about 5.5 metres per minute.  Mr Kelle based his calculation taking into account the loads on the windlass, the engine movements after 15:25 (which would have had an effect of reducing the load on the cable and so assisting in the rate of retrieval to about 5.5 metres per minute), the AIS video depictions of the ship’s headings, position and other data, and the weather and environmental conditions (that would have slowed the rate of retrieval to less than 4 metres per minute when the ahead engine movements ceased).  The more anchor chain that has to be retrieved the heavier it is and the slower the initial rate of retrieval.  As more chain has been recovered, the lighter the remainder and the quicker the retrieval becomes.  Thus, Mr Kelle used an average retrieval rate of 5.5 metres of cable per minute in his calculation and assumed that the total time that the windlass was operational was 18 minutes.  This resulted in his calculation that about 99 metres had been retrieved in that time.  The length of cable actually retrieved during the period up to the failure of the windlass and the stop engine movement at 15:47:31 has some significance for later expert hypotheses.

133               Capt Xu said that he moved the ship towards the anchor to assist in its retrieval.  This was his purpose in ordering each of the dead slow ahead engine movements at 15:30:27 to 15:33:31 and again at 15:35:55 to 15:36:31.  These movements brought the bow into the wind heading south (176°).  It had swung back towards the north east in between those two ahead movements, after the engine stopped.

The Cable Leads Out 90°

134               The master said that the weather conditions were very bad and changed very quickly.  Capt Xu did not issue any further engine orders until 15:40:29 at which time the ship’s bow was heading south east (131°).  He ordered dead slow astern until ordering the engine to stop at 15:41:17.  He explained that his reason for this order was that the chief officer had told him that the cable was then at 90° to the starboard bow.  The master could not remember precisely which way (to port or starboard) the cable was heading (south west or north east) but he believed that the chain was going to the port or north east side.  If it were, and the vessel went ahead, she would run over the chain.  The bow swung slightly east by 10° to 121° after the engine stopped at 15:41:17.  The ship continued to move north east, beam on to the wind.  Capt Third and Capt Dwyer observed that vessels dragging their anchor in high winds often get into a situation in which the bow goes down wind leaving the anchor cable, and anchor, streaming (i.e. leading or pointing) astern.  Thus the ship will pull the anchor ahead, as it were, with the cable on an angle leading towards the stern.

135               Esso and BHP argued that the master had not been asked in his evidence in chief to explain his reasoning process at this point of time.  They said that this had emerged during cross-examination and in a confusing way, since, initially he said that the chief officer told him that the cable was leading 90° to the port side.  But after a short adjournment the master said that he did not remember whether he had been told it was to port or starboard.  In addition, Esso and BHP relied on the fact that neither the chief officer nor the bosun gave evidence in chief about the cable being at any such angle or of any such conversation between the chief officer and the master.

136               However, I believe Capt Xu.  He gave his oral evidence against a background of being shown the AIS video several times for particular time periods.  In ordinary experience people do not recall every single event, action or communication accurately or at all, even immediately after a crisis or accident.  Sometimes a recollection comes back, genuinely, in the heat of cross-examination.  And as Capt Xu said of what must have been one of his worst moments as a master, if not the worst:

“I can’t remember all the details, and I don’t want to remember all of the details, and it has been a long time and it’s quite painful to remember all of this;  and I want to forget it all, although I couldn’t.  But I don’t want to remember”.

137               When the vessel was not under her engine during the four minutes before 15:40:29, all the nautical experts agreed that the cable was likely to have come into a position where it led out towards the stern.  Capts Third, Dwyer and Cole said that it was leading between 110° to 115° back from the starboard hawse pipe.

138               I find that immediately before the dead slow astern engine order at 15:40:29, the bow was moving ahead of the anchor cable so that the cable was leading astern slightly.  The ship was beam onto the wind.  I accept the master’s evidence that the chief officer told him that the cable was at 90° before he made the dead slow astern order at 15:40:29.  I find that by the time of the stop engine order at 15:41:17 the angle of the cable was closer to 110° to 115°on the starboard side and that the preceding dead slow astern order had had little effect in altering the ship’s apparent general movement or position as it appears on the AIS video or other available records.  The nautical experts said that there was just a slight movement of the bow to port (or the east). 

139               However, that navigational result may have come about because the master had arrested what would otherwise have been a more pronounced movement of the ship in the prevailing conditions.  It is not possible to know without other data, such as the helm orders, voice data and wind data, whether the dead slow astern order did achieve, or was directed to, some intended purpose.  Capt Third explained that one possible purpose for the short duration of this astern order may have been to change the vessel’s position in relation to the direction of the cable by creating a transverse thrust.  This is a manoeuvring technique.  The ship’s propeller turned to the right when the engine was put astern.  The propeller blades were pitched so that if the engine were put astern and the rudder hard to port, as the propeller initially forced water against the rudder, the stern would begin swinging to port (i.e. at this time towards the north) while the bow would swing to starboard (i.e. here to the south and into the wind).  In other words, the effect of the transverse thrust intended to be generated by such a manoeuvre would be to turn the ship clockwise.

140               The ship will also be affected by transverse thrust from an ahead movement, but to a lesser extent and with the bow and stern moving initially in the opposite way to their reaction to an astern movement.  Transverse thrust can be accentuated by the master giving a helm order, which changes where the rudder is turned.  However, it takes a little time after the propeller begins to respond to an engine order for it to have any effect on the vessel;  then there is an initial reaction of transverse thrust, again for a short but not a precise time of around about 20 to 30 seconds or so depending on the ship.  That can be assisted by the rudder and the use of the bow thruster and can change the immediate heading or direction of the ship to a limited extent.  And, the effect works only for a short period of the initial 20 or so seconds after the propeller starts rotating and moving water around it.  The helm order records which were part of the voyage date records were not available to confirm that the rudder had been put hard to port.  However, I accept Capt Third’s evidence that the manoeuvres at this time were designed by the master to relieve the load on the anchor cable and to make it easier for the windlass to recover the anchor.

141               Capt Third observed the fact that the manoeuvre did not work on that occasion is not necessarily conclusive.  Each of the three experts had a different view of what the master should have done at that point.  Capt Dwyer’s view was that given the ship’s proximity to the charted pipeline, it was not wise to use the engine at all.  Capt Cole thought the master should have used the bow thruster to put the ship’s head (or bow) into the wind.  And, as Capt Third said, because the voyage data recorder was not available, one cannot know whether the bow thruster was being used, but not successfully.  He considered, and I accept, that the astern manoeuvre was one open to the master.  This was because Capt Xu had to continue to heave anchor and the situation he was in, where the anchor cable was leading astern, had to be addressed.  I am not satisfied that the manoeuvre and use of the engine dead slow astern at this time was inappropriate.

142               Just after the stop engine order at 15:42:03, the master reported to Capt Murchie, on the pilot boat, that there were still two shackles in the water.  I am satisfied that this was correct.  Whatever theoretical calculations might be made about retrieval rates for the cable, the conditions in Port Phillip Bay, were very difficult.  The way in which the cable had gone to a position leading astern and the movement of the ship north east as she went abeam to the wind made the retrieval operation anything other than a standard one.  During this period between engine movements the bow went slightly further to the north east but without any substantial change in the vessel’s behaviour as compared to when the last engine movement was occurring.  As Capt Dwyer said, there was insufficient information available to explain why the ship was behaving as it was, including why the dead slow astern movement apparently had no discernible effect.

The manoeuvres immediately preceding the anchor fouling the pipeline

143               The master’s next manoeuvre was a minute later at 15:43:09 when he ordered dead slow astern again.  At 15:43:35 he ordered slow astern until ordering stop at 15:44:53.  Before this manoeuvre started the ship was south of the charted position of the pipeline but to the north of the northern boundary of the outer anchorage.  She was beam on to the wind moving laterally to the north east.  The AIS video showed that by 15:43:06, just before the second engine order of dead slow astern, the ship had crossed the charted position of the pipeline.  At about 15:43:09 of the ship’s heading was about 120° and by 15:44:53 it had swung about 10° to the south and to starboard.  This was a change in the direction from the long swing to port that had begun about 15:30 when the heading had been 228°.

144               Esso and BHP argued that the master should have appreciated that the only possible reason for this swing, and its subsequent progression in the next minute was that the anchor had already fouled on the pipeline.  Capt Xu did not accept this narrow view, although he agreed that it was a possible explanation for the vessel’s behaviour.  He said, and I accept, that his concern at this time was to heave the anchor.  He knew, of course, that the direction of the vessel and its heading were changing and that the ship was swinging through a big yaw.  However, given the weather conditions, with large waves and winds, and because  he was concentrating on heaving the anchor, he did not focus on the very quick change in the heading or any explanation for it.

145               I find that Capt Xu ordered the engine movements between 15:40:29 to 15:44:53 in order to take the vessel’s weight off the anchor cable which was leading at about right angles to the bow  or astern.  He was trying to correct the relationship between the angle of the cable and the bow so that the cable would lead ahead and be easier to heave.  Because he was preoccupied with this, the master did not concentrate on the precise, moment by moment, movement, heading or position of the vessel in relation to the pipeline.  He was generally aware that the ship was very close to the pipeline and of the ever increasing danger of her anchor fouling it.  But I am satisfied that, during this time, he was responding to the immediate problem, as he saw it, presented by the angle of the anchor cable and counteracting its effect on the heaving operation.  The astern engine orders at this time were intended to assist bringing the ship’s head back into the wind.  As he said:

“My main purpose [is] to stop the vessel so that I can heave up the anchor.  Whether I use ahead or astern was based on the direction of the cable chain.  And I know in this area it was very dangerous and I didn’t want to use either ahead or astern.”

146               Capt Xu saw the danger in the situation for both the ship and the pipeline.  The head came into, or at least towards, the wind as Capt Xu intended these manoeuvres to achieve.  The nautical experts agreed that objective was, in normal circumstances, a seamanlike one.  Capt Dwyer, however, considered that because of the ship’s proximity to the pipeline, the master should not have ordered any engine movement at that time.  I prefer the evidence of Capt Xu and Capt Third on this issue.  By 15:43 the master was in a very difficult position.  I am not satisfied that his use of the engine in an attempt to bring the head around into the wind was unseamanlike.  Capt Cole would have used the bow thruster.  As Capt Third said even though the use of the engine astern at this time did not work, the master had to try something, such as he did, to assist in retrieving the anchor more quickly.  In Capt Cole’s words between 15:43:09 and 15:44:53 when the ship’s speed over the ground was 2.1 knots:

“The ship is really in trouble, and the only way that I can see that would assist the situation is to try and get the ship’s bow to starboard so it became windrode.”

A ship is windrode when she appears to be subject to the force of the wind rather than the tide.

147               Although he was responding to a question about how to reduce the speed over the ground and bring the vessel up to the anchor, Capt Cole’s answer captured the gravamen of Capt Xu’s position.  He was in a crisis.  He chose to use short astern engine movements to swing the bow towards the wind in very difficult conditions.  The vessel was being blown north east beam on to the wind at 2.1 knots and the anchor chain was leading out at 90° or astern.  I am not satisfied that the master was then acting in an unseamanlike way simply because he may not have referred to any available navigational aids.  Those aids would have told him what he already knew;  namely that the ship was very close to the charted position of the pipeline and the anchor might foul it at any moment.  Hence, the master’s understandable concern to assist in, the then impeded operation of, heaving of the anchor.

148               However, that danger had arisen because of his and harbour control’s failures in the preceding 40 minutes to act sufficiently quickly to prevent the ship being blown to a position of close proximity to the pipeline.  Difficult as it was for him, Capt Xu should have ignored the refusal of harbour control at 15:07 to allow him to move the ship and he should have begun heaving the anchor then and there.  He had seen that the anchor had been dragging since at least shortly after the pilot left the ship.  The wind was strong and was pushing the ship closer by the minute to the pipeline.  Heaving in over 5 shackles of chain would take some time, particularly in the difficult conditions.  It was the time to act.  His instinctual reaction to the situation was correct.  His primary responsibility was to ensure the safety of his ship, her crew and cargo and to prevent the anchor damaging the pipeline.  I do not accept that he was absolved of his right and duty to exercise that responsibility by the unhelpful refusal of harbour control to give its permission.  The dragging of the anchor should have been addressed no later than then.  As events transpired, it was too late to begin heaving the anchor when harbour control belatedly gave its permission at 15:25.

149               The shipowner argued that Capt Cole and Capt Dwyer had had months to consider the formation of their expert opinions on what a master ought to have done in Capt Xu’s position and of their criticisms of what Capt Xu in fact did.  In contrast, it relied on Capt Third’s evidence that the master did not act in an unseamanlike way as events unfolded on the day.

150               I have also been conscious that criticism is easy in hindsight and with the luxury of having a fuller grasp of and insight into the facts than Capt Xu could have had.  Nonetheless, I am satisfied that his concern at 15:01 and thereafter that had prompted him to seek harbour control’s permission to heave anchor is a sure and certain guide that he, as a competent, experienced master, knew that the time had come when he had to do something new to address the dragging of the anchor.  His mistake of accepting the direction of harbour control to do nothing was a human and understandable one.  But it was a mistake.  And at 15:05:41, when he reported to harbour control that the ship was maintaining its position “with danger for our vessel”, it still had a speed over the ground of 1.1 knots although its heading was generally stable to south south-west and it was moving south east with her bow passing along the northern boundary of the outer anchorage as shown on the AIS video.  But, by 15:10 the bow had moved appreciably to the north west and the ship was obviously dragging anchor again.

151               Even if the position and movement of the ship between 15:01 and 15:05 had suggested to Capt Xu that the ship may have been brought up to the anchor, very soon after that point, it was plain that the anchor was still dragging and the ship was moving closer to the pipeline.  The weather was not abating;  the danger of fouling was increasing the closer the vessel came to the pipeline.  Moreover, heaving anchor in those conditions was not a crisp, simple exercise.  The ship was yawing in the strong winds, at times increasing the strain on the anchor cable.  The sooner the heaving operation commenced, the less likely the ship would be blown sufficiently close to the pipeline that the anchor could foul it before it had been heaved off the seabed.  This commonsense position was self evidently the motivation for Capt Xu’s request to harbour control at 15:01.   It was the sound and seamanlike thing to do.  While it is possible to sympathise (as I do) with his dilemma in having to defy harbour control in a foreign port and to risk criminal prosecution for having done so, Capt Xu should have begun heaving the anchor at 15:01 because, as he said, the situation “… is dangerous for our vessel”.  The unstated concomitant of that observation was that the situation was also dangerous for the pipeline.  That scenario only worsened as time progressed.

The anchor fouls the pipeline

152               Esso and BHP’s pleaded defence alleged that the anchor struck and fouled the pipeline at 15:44.  Before the trial commenced, all the relevant expert evidence agreed that the anchor fouled the pipeline at 15:44 to 15:45;  that is, during the period between the last two astern orders commencing at 15:43:09 and the stop order at 15:44:53 or in the succeeding minute.  Esso and BHP suggested in final address that the fouling may have occurred at about 15:41:47 when the ship’s rate of movement north began to slow.  This was based on Capt Third being asked to look at a print out of the electronic data transmitted by the ship’s AIS equipment.  Capt Third’s attention was not drawn to, and he did not comment on, the impact on the vessel’s speed and direction by the dead slow astern order that was executed between 15:40:29 and 15:41:17.  I consider that the effect of this engine movement was more probably the reason for the observed changes, rather than the anchor having fouled at this earlier time.

153               The shipowner argued that it was impossible to fix the time of the fouling precisely.  It pointed to Capt Cole having identified the time, in his written reports, prepared after his due consideration of available material, as being at about 15:46.  The shipowner’s argument based on Capt Cole’s postulation also went to the assessment of the master’s conduct in respect of a second alleged distinct occasion when he gave the dead slow astern order at 15:46:01.  I will consider this issue later in these reasons.

154               All the maritime experts agreed in their joint report and in oral evidence that the anchor fouled the pipeline between 15:44 and 15:45.  Capt Third also referred to the distinct interruption, at that time, to the otherwise smooth recording on its course recorder of the vessel’s swings during that afternoon.  I accept that evidence.  Complete precision in fixing the time of fouling is not possible.  I find that it occurred within the two minute range between 15:44 and 15:45. 

155               There is, of course, a difference between what the evidence after full analysis discloses, and what a reasonable master in the position of Capt Xu on the vessel’s bridge ought to have appreciated, at any particular time on 13 December 2008.  I have arrived at the finding that the fouling occurred between 15:44 and 15:45 because of the following reasons.  First, the behaviour of the ship as recorded in the course recorder and depicted on the AIS video and the AIS data satisfies me that her movement north east was arrested in that period and her bow began to swing back to the south.  Secondly, this behaviour is not inconsistent with the major failure of the windlass that occurred shortly afterwards that I deal with later in [201]-[207].  Thirdly, the joint view of the marine experts, who appreciated the subtleties of the ship’s behaviour and the available data, appears to me to be rational and soundly based.

The impact damage to the pipeline

156               When the anchor struck the pipeline it did so with a very powerful force.  The pipeline (and the dragging anchor) were buried in the seabed to a depth of about 2.5 metres.  The pipeline had an internal diameter of 10 inches or about 254 mm.  It was constructed in the early 1970s.  It was over 78 kilometres in length running overland and undersea from Long Island Point at Hastings on the east of Port Phillip Bay to Altona on the west.  The pipeline was made of steel.  The steel’s external diameter was 273 mm.  The outer steel was encased in a number of coats.  Closest to the steel was a coal tar enamel coating and over it a coal tar saturated asbestos felt that was wound helically.  These were covered by a concrete weight coat reinforced with wire netting.

157               The shank hit the pipeline first, with the flukes then passing underneath it.  The anchor was travelling at about one metre per second or 2.1 knots with the laden weight of the ship pulling it north east.  The impact caused the pipeline to ovalise.  I will explain the process in more detail below (see “The Aetiology of the Missing Fractured Pipeline” at [332]-[360].  The point of impact was on a piece of the pipeline that became an exhibit.  The impact crushed the circular shape of the pipe into a narrow slit on its south west face (i.e where it was pinched between the shank and flukes of the anchor).

158               The impact and continuing force exerted by the moving ship pulled the now distorted pipeline upwards and out of its position under the seabed to a position in the water above that seabed in the immediate area of the impact.  Thus, a portion of the pipeline was pulled up and further distorted by the impact and continuing movement of the ship.  The damage to the pipeline occurred in several different respects.  First, it was crushed, secondly, it physically moved upwards and to the north east, and thirdly, in lay terms, it was stretched from each side of the anchor’s impact to allow a section of the pipeline to move up and out of its seabed trench without it rupturing at that time.  As Dr Price explained in his concurrent evidence:

“Once the anchor had engaged and ripped the pipe out of its trench, we are now talking about very high forces and forces down the chain. During the initial ripping out of the trench we have the entire momentum of the ship being stopped, the entire movement of the ship being stopped, so we have ... forces ... on the chain during that deceleration of the ship, and at that point ... basically the shank and the flukes were then engaged, and for the next half hour that force still exists along the chain because the ship is being propelled by the tides and sea currents and so forth, and the wind, to the north-east.

So that chain was continuously pulling during that period.”

159               Esso had engaged DOF Subsea Australia Pty Ltd to survey the pipeline to ascertain its state and the extent of damage to it.  That work was performed from 17 December 2008.  DOF prepared, among other things, daily field reports, bathymetric surveys of the damaged section of the pipeline and anchor scours as well as of the length of the pipeline to confirm its position.  One important daily field report was that of 27 December 2008 on which much of the expert evidence was based.  I have reproduced it below.

 

 

 

 

 

Figure 2

160               As can be seen from Figure 2, DOF not only took sonar images of the seabed looking from above, but also identified, using sonar, the depths of the sections of the pipeline on either side of the fracture.  The thin lines radiating from north towards the east above the large blowhole in the centre of the sonar image on Figure 2 are the anchor chain scours.  I will continue to return to what Figure 2 shows in these reasons.

161               Dr Lyngberg and Mr Kelle agreed in their concurrent evidence that the pipeline needed repair and was no longer suitable for carrying gas under pressure, in the area of the impact.  Thus, they agreed that a portion of the pipeline was so damaged that it needed replacement as soon as the impact and initial restraint of the ship’s movement occurred.  They also agreed that once the fouling occurred, the ship continued moving for a time but that it would eventually be brought up to the anchor.  The ship then drifted to the north east some metres.  Mr Kelle calculated that the initial impact had dragged the pipeline 18 to 20 metres from where it had lain in the seabed and that it had absorbed a load or force of approximately 240 tonnes.  He said that the anchor stretched the pipeline that lateral distance by dragging it upwards and out of the seabed towards the north east.  Dr Lyngberg considered that the length of this stretch was between 8-12 metres, but he did not perform detailed calculations to support that view.  This was because Dr Lyngberg was more concerned with the overall damage to the integrity of the pipeline.

162               In a number of areas of his evidence, Dr Lyngberg did ot undertake a detailed analysis to support his opinions that differed from Mr Kelle or Dr Price.  This limited to a degree the reliance that I have been able to place on some of Dr Lyngberg’s conclusions.  He was knowledgeable and experienced in the areas in which he gave evidence.  However, his approach of reasoning from general principles without undertaking a scientific analysis that could be examined by his counterpart inclined me to prefer the more transparently reasoned opinions of Mr Kelle and Dr Price where they conflicted with Dr Lyngberg’s.

163               Even though the pipeline consisted of steel overlaid with the protective coverings that I have described, it still had a degree of elasticity.  The engineering expert evidence described a number of concepts that are relevant to understanding the potential for the pipeline to have suffered damage on each of the four alleged distinct occasions.  Dr Price provided some definitions in his report  on which I have drawn.  A material behaves elasticallyif it recovers its original dimensions after a load (i.e. force) has been removed.  In lay terms, an elastic or rubber band appears to return to its previous dimensions after it has been stretched.  A related concept is plasticity.  A material behaves plastically if it is permanently deformed after a load has been removed.  In lay terms, once the material has been stretched, it does not return to its original dimensions but remains in its last shape even when the load is no longer applied.  Material often displays features of both elastic and plastic behaviour.  When a metal is strained beyond its elastic limit a mixture of elastic and plastic deformation occurs.  That is called “elastic-plastic”behaviour.  When a load is removed suddenly from an elastic material, it experiences recoil.

164               I am satisfied that the initial force of the impact of the anchor on the pipeline caused significant plastic deformation so that it was, first, no longer sound to carry gas under pressure and, secondly, a portion was then and there in need of replacement. 

165               Ultimately, Esso and BHP replaced 156.575 metres of the pipeline at a cost of about $27 million.  The whole repair operation took 11 weeks finishing in late March 2009.  Michael Vines, an engineer employed by a company related to Esso, was the manager of the pipeline repair team.  He gave unchallenged evidence of the actual repair operation and its costs.  He also used assumptions to calculate the lesser work, time required, and costs had smaller sections of the pipeline needed replacement.  The assumptions that Mr Vines used were not agreed.  Esso and BHP had asked him to assume that had the initial fouling been the only incident, 42 metres of pipeline would need replacement.  On that hypothesis he estimated that the task would have taken 10 weeks to complete at a cost of $16,650,000.  Dr Price considered that the actual repairs, and the postulated ones on the basis of Mr Vines’ various assumptions, were appropriate.

166               The shipowner argued that Esso and BHP did not lead any evidence of any greater force having been applied to the pipeline than the force of the initial impact.  However, Mr Kelle made the fair and obvious riposte to this.  He was asked what force had been applied to the pipeline immediately before it ruptured later in the afternoon when the engine movement relied on as the alleged third distinct occasion took effect.  Dr Kelle said:  “Well in excess of the capacity of the pipeline”.  And Prof Renilson said that although he and Mr Kelle disagreed on the magnitude of the peak loads imposed on the pipeline while the ship was yawing with the anchor fouled on it:  “It’s a randomish process, and that’s when the thing actually fails” (i.e. the pipeline ruptured).

167               Mr Kelle said that subsequent engine movements may have contributed to the pipeline being dragged out further than his calculation of 18-20 metres for the initial impact, but that this was difficult to establish.  The reasons why this was difficult were:  the sensitivity of his model to the angle that it used for the cable leading up to the hawse pipe, the soil stiffness properties of various parts of the seabed in which the pipeline was buried and the significantly smaller thrust forces applied by an astern engine movement compared with the forces applied to the anchor and pipeline by the environmental loads.  The environmental loads were the forces applied by the action of the wind, and waves as the ship yawed from side to side for about 35 minutes while the anchor remained fouled.

168               Mr Kelle’s model gave a result that 350 metres of the pipeline from the point of impact towards Mordialloc was plasticised, while most of the Altona side reacted elastically.  This was because Mr Kelle assumed the seabed soil on the Altona side was less stiff than on the Mordialloc side.  However, in deciding to replace only 156.575 metres, Esso and BHP relied on a different soil stiffness analysis by JP Kenny.  That analysis was based on assumptions and modelling of the seabed soil that Mr Kelle did not accept.  Dr Lyngberg pragmatically looked at the question of how much of the pipeline was damaged sufficiently to require replacement.  He accepted that the analysis by JP Kenny was of a sufficient quality to justify what Esso and BHP actually did, namely to replace only about 156 metres.

169               I prefer the approach of Dr Lyngberg on this issue.  It accords with commonsense that Esso and BHP had a real interest in ensuring that any part of the pipeline that was damaged, so as to affect its integrity as a conduit for the gas, should be replaced.  Had there been any indication during the course of the repair work that a greater length of plastic deformation had occurred requiring replacement, Esso and BHP would have taken the prudent course of replacing that portion in addition to what they intended to (and did) replace based on, among other information, JP Kenny’s analysis of soil stiffness.

170               The soil analyses referred to by Mr Kelle and Dr Lyngberg were not prepared for the purposes of identifying how the pipeline reacted during the period that the anchor was engaged with it.  Nor were they very detailed analyses of the consequences of the fouling and later movements of the ship as she moved under her engine or the forces of the environment, as Dr Lyngberg explained.  Rather, as both Mr Kelle and Dr Lyngberg agreed, the analyses were prepared to give Esso and BHP an initial order of magnitude of damage although Mr Kelle pointed out that he maintained his dissatisfaction with the soil data JP Kenny used.  Mr Kelle, however, acknowledged that he was a structural, not a geotechnical, engineer and was interpreting the work of JP Kenny who were the geotechnical engineers, on his understanding of the effect of differing soils on their resistance properties in reacting to the forces exerted by the impact of the anchor when it fouled and then held the vessel on the pipeline.

171               There is no evidence that Esso and BHP used the analysis of JP Kenny in forming their decision as to how much pipeline was to be replaced.  Moreover, Mr Kelle conceded that he had no disagreement with the repairs Esso did or the length of pipeline repaired.  He said that some analysis was done for Esso on damage to the pipeline by Pearl Street Laboratories and that he (Mr Kelle) understood that Esso had considered this in concluding that the length of pipeline replaced was sufficient.  Esso and BHP could have led evidence of the actual soil and other analysis, such as that of Pearl Street Laboratories, on which they relied in determining to replace the 156 or so metres of pipeline that they did replace.  I infer that such an analysis would not have assisted their case or supported Mr Kelle’s hypothesis that over 350 metres was damaged by the initial impact.

172               The various experts acknowledged that their models depended on a variety of assumptions and that in the conditions experienced on 13 December 2008 there was no certainty that any particular selection or combination of assumptions was accurate.

173               The force of the initial impact of the anchor on the pipeline was very significant.  The anchor chain markings in Figure 2 lead into the Altona end blowhole.  Thus, after the fouling and 15:46:01 engine movement, while the ship yawed the position of the anchor on the pipeline must have been located just to the south west of the point to which each chain marking on Figure 2 leads.  This is a position near the north east of the Altona blowhole.  And, the pipeline must have been elevated to some degree from the seabed having been wrenched from about 2.5 metres below it to the north east.  The Altona end of the pipeline that was found in the position identified in Figure 2, was likely to have been located at about, but perhaps a little south west of, the position of maximum distortion of the pipeline when the vessel was yawing.  An in situ picture of the Altona end before it was brought to the surface shows that it remained above the seabed.  The Altona end was measured to be 11 metres outside and to the north east of the trench left after the pipeline had been wrenched from its original resting place below the seabed.

174               I find that when the anchor struck and fouled it between about 15:44 and 15:45, the pipeline came laterally out of its existing trench by 12 to 14 metres (thus moving it upwards and to the north east from about three metres below the seabed by a total of between 15 to 17 metres).  As a consequence, the pipeline was no longer fit to carry gas under pressure and required immediate repair by replacing the length of pipe that had been plastically deformed to the degree that adversely affected its structural integrity.  Mr Kelle’s evidence that I have generally accepted below (see [360]) was that the additional damage caused when the ship went astern after 16:24 required between 15 to 30 metres more pipe to be replaced.  I find that the initial impact damaged about 125 metres of the pipeline so much that it would have had to be replaced.

The consequences of the dead slow astern order at 15:46:01

175               At 15:46:01 Capt Xu ordered the engine to dead slow astern.  One and a half minutes later he ordered stop (at 15:47:31).  He gave no evidence directly explaining his purpose in undertaking that manoeuvre.  By the commencement of this astern manoeuvre the ship’s heading was 150°.  During the preceding engine movements (dead slow astern, slow astern and slow) commencing at 15:43:09, the ship’s heading had swung to the sought from 119° to 138°  and that swing continued in the next minute to 150°.

176               As I have found, the earlier astern movements had been undertaken to take the vessel’s weight off, and ease tension on, the cable and to realign the angle of the cable leading to the bow to assist in heaving the anchor.  In that context, the 31° change in heading should have been observed by Capt Xu.  Moreover, an astern movement that was not undertaken to realign the cable’s direction in relation to the bow, would impede, not assist in, heaving the anchor.  I do not consider that Capt Xu undertook such an illogical manoeuvre.  As the nautical experts observed, the previous behaviour and headings of the ship shown in the AIS video and data, suggested that the cable was earlier at 110°-115° leading astern, not just tending to 90° as Capt Xu had said in his evidence.  The 31° correction or change in the heading would have brought the cable angle closer to 90°.  Because the earlier manoeuvres had only partially achieved the master’s goal, I infer that he initiated the dead slow astern order at 15:46:01 to continue his attempts to assist in heaving the anchor.  Capt Cole observed that the preceding orders beginning at 15:43:09 were made because at that time, “the ship is really in trouble”.  That emergency still continued at 15:46:01.  As Capt Xu said, before he ordered the anchor to be heaved, he was able to consider the ship’s position by looking at the radar:

“MR MURDOCH:  The radar showed clearly, didn’t it, the rapid change in the heading of the ship?

THE INTERPRETER:  If you staring at the radar it is true, but I didn’t staring at the radar at the time.  I was focusing on the heaving up the anchor and not a lot of other things, and I was focusing my whole attention on that and there is one point I was focusing the most, which is heaving up the anchor, and I can’t focus on everything else.  I’m just a normal ordinary person, and I have one point for my focus, not everything else, and I couldn’t do that, and my focus was to heave up the anchor and the biggest hope was to leave the position as soon as possible.”


177               I accept Capt Third’s evidence that there was no major change in the period from 15:43 to the order at 15:46:01.  As he said, there may have been different conditions operating at that time, such as how the wind was gusting, and the master did not then know that the anchor had fouled the pipeline.

178               Capt Dwyer expressed the opinion that the order astern at 15:46:01 was “inexplicable”.  He said the order should not have been made because he considered that no engine movements were justified in the position the vessel had reached.  He reasoned, in his concurrent evidence, that because of the change  of the ship’s heading from 15:44:06 towards the wind the master “… was able to make an assumption that he was hooked up or very close to hooking up to the pipeline” (emphasis added).  Capt Cole expressed a similar view.

179               In my assessment each of Capt Dwyer and Capt Cole came to that conclusion with the benefit of hindsight and the opportunity for detailed reflexion that the master did not have.  Moreover, neither of them related the astern order at 15:46:01 to the position in which the cable had been leading just beforehand.  And that “assumption” was qualified by Capt Dwyer to cover a situation where the anchor had not yet fouled the pipeline.

180               While the master could have made the ambiguous assumption that Capt Dwyer said he was able to make, that would only have left him in a state of uncertainty.  I am not satisfied that at 15:46:01 a competent master in Capt Xu’s position had to make the positive assumption that the anchor had fouled the pipeline and that this was the only explanation for the change of heading, rather than the ship’s behaviour being explained by his immediately preceding engine movements in the wind and weather she was experiencing.  As Capt Xu said, it was possible that the change came about because the anchor had fouled the pipeline.  He said that he noticed that a big change in the heading of the vessel had happened, and he was also aware of the danger, but “… my purpose was to heave up the anchor, and I didn’t have much spare energy to notice anything else”.  The other limb of Capt Dwyer’s assumption, that the ship was very close to fouling the pipeline, was still open at 15:46:01.

181               The second officer plotted the ship’s position and marked it on the Admiralty chart at 15:47 as being tens of metres north east of the charted pipeline.  Capt Xu knew then that the ship had crossed the charted position of the pipeline.  But he was not certain at that moment that the anchor had fouled it.  However, Capt Third and Capt Dwyer suggested, at about the time of the stop order at 15:47:31, the reality that the anchor had fouled had begun to dawn in Capt Xu’s mind.  I infer that Capt Xu learnt of the second officer’s report of the ship’s position at about the same time as the chief officer reported that there was a problem with the windlass.

The effect of the dead slow astern order at 15:46:01

182               The parties devoted considerable attention to the issue of whether or not the engine order for dead slow astern at 15:46:01 was a distinct occasion.  Almost instinctively one might think that by the ship exerting the force of its engine astern for 90 seconds it would pull the pipeline further out of the seabed, and so cause plastic (or permanent) deformation of more of the pipeline than had been damaged by the force of the impact of the initial fouling.

183               Esso and BHP, however, led no direct evidence that this engine movement did cause any additional damage to the pipeline.  Mr Kelle considered that on the initial impact the pipeline absorbed a load of 240 tonnes.  As I have found above, this impact pulled the pipeline about 15 to 17 metres upwards and north east of where it was in the trench, which was a little less than Mr Kelle’s calculation of 18 to 20 metres.  So, it may be that his calculation of a load of 240 tonnes could have been too high, although he also based his conclusion about the movement of the pipeline on an assumption about the resistance offered by the seabed that I have not accepted.  Mr Kelle calculated that between the anchor fouling and the pipeline rupturing, environmental loads below 95 tonnes were imposed on the pipeline.  These loads were the consequence of the movements of the wind and water that affected the position of the ship as she yawed in three wide swings from starboard to port.

184               Mr Kelle said that the force vector due to the engine movements between around 15:44 to 16:19 would only impart an additional load of 10 to 30 tonnes on the pipeline.  I understood him to mean that the latter force operated at a different time and without adding to the 240 tonne load previously absorbed by the pipeline on the initial impact.  He then said in his first report:

“Subsequent engine actions may have contributed to the pipeline being dragged further.  However, this is difficult to establish due to the sensitivity of the model to the pull out angle which was set to 13° (based on engineering judgment and a number of iterations) and the soil stiffness and the significantly smaller engine astern thrust values estimated at Section 4.1.2.

In general it needs to be noted that the results provide an order of magnitude estimate confirming what is shown by the vessel’s behaviour in the AIS video.  It is highly likely that the pipeline would be able to withstand the initial impact and would also be able to provide sufficient restraint to the vessel to allow the vessel to swing around (weathervane) as shown by the AIS video.”  (emphasis added)

185               Thus, the initial impact had generated a load on the pipeline of about 240 tonnes.  Mr Kelle opined only that the dead slow astern engine movement at 15:46:01 “may have contributed to the pipeline being dragged further”, not that this was likely or had happened.  The significantly lesser additional load generated by that engine movement (10 to 30 tonnes) even when combined with the largest environmental load (95 tonnes) is unlikely to have caused any discernible further sections of pipeline to have plasticised or been damaged.  Prof Renilson agreed with Mr Kelle that the latter’s  estimate of the magnitude and direction of environmental loads on the ship was a reasonable approach and was probably within plus or minus 30% of their maximum value.  Also in their joint report, Prof Renilson and Mr Kelle said that it was not certain to what level the engine orders after the initial impact of the anchor contributed to the events.

186               Esso and BHP argued that Dr Price and Mr Gartner had agreed that in the period before the rupture, the anchor dragged and lifted the pipeline.  They agreed that this resulted in the pipeline becoming locally distorted, ovalised and flattened in the region where it eventually severed.  Esso and BHP also relied on the joint report of Mr Gartner and Mr Kelle that the pulling of the anchor during the period between the initial fouling and the rupture was likely to have led to deformations of the pipeline that cannot be discretely identified in what was recovered.

187               Dr Lyngberg said that if the pipeline were dragged further out of the trench by more force, then, depending on the degree of that extra force, the axial strain could cause it to bend and to increase the length of pipeline that had been deformed.

188               In his initial report, Mr Kelle estimated that after the pipeline had been dragged by the initial impact about 18 to 20 metres laterally, it was dragged laterally an additional 7 metres (a total of 27 metres) at 15:47 and another 7 metres (a total of 34 metres) at 16:00.  He used this estimate to assert that the engine movement commencing at 15:46:01 had the effect of further displacing and bending the pipeline transversally towards Altona (i.e. north west) and that this caused more pipeline to be pulled out of its trench and damaged than occurred on the initial impact.  I reject that evidence.  Mr Gartner drew a diagrammatic representation of the displacement of a total of 34 metres of which Mr Kelle spoke in his initial report in reply.  The two experts in their joint report agreed that this diagram reflected what Mr Kelle was asserting and that the deformations of the pipeline that this would have caused, had Mr Kelle’s postulations occurred, were not evident in the recovered pipeline.  Even, if the total 27 metre displacement postulated by Mr Kelle had occurred, the pipeline on the Altona side would have been displaced to a lesser, but noticeable, extent by being bent more to the north or north west than it was.

189               Mr Kelle’s estimate of greater displacement caused by the 15:46:01 engine movement was not supported by any evidence, as he acknowledged.  He had simply assumed that at this time there were 69 metres (i.e. 2.5 shackles) of cable in the water in calculating that displacement.  I have found below that once the windlass broke and cable paid out until stopped by the bosun applying the hand brake, about 2.5 shackles were in the water.  That seems to have occurred about 15:48 (see [207]).  None of the experts appeared to correlate the rapid paying out of cable when that occurred with the data from the AIS video.  And, all the experts proceeded on the hypothesis that the transcript time was reasonably related to the AIS data time.  It may well be that Mr Kelle’s theory is, like the curate’s egg, good in parts and that the conversations about one shackle being in the water and the windlass breaking, that the transcript puts at 15:48, actually occurred contemporaneously with the fouling incident.  As best the nautical experts could discern, the AIS video showed that the fouling occurred between 15:44 and 15:45.  I infer that, rather than the pipeline yielding extra metres as Mr Kelle thought, the cable paid out altering the position of the ship as she went astern.  However, this possibility was not explored in the evidence or address and I am not able to make any finding on it.

190               While the engine movement commencing at 15:46:01 may have caused some deformation to the already displaced pipeline, I am not satisfied that any further additional pipeline was affected or damaged by that movement.  And, any further damage to the already damaged part of the pipeline would not give rise to a claim on a distinct occasion.  That is because the initial impact was of such a force that all of the pipeline it damaged required replacement then and there.  Subsequent deterioration of that same section of the pipeline was not of any consequence at all – the section had been rendered worthless and useless by the fouling.  It would defeat the purpose of the Convention to treat superfluous and irrelevant deterioration in the section of pipeline, that for all practical purposes had been destroyed on the initial impact, as giving rise to a new claim on an occasion distinct from the fouling.

191               Although this finding disposes of the availability of the engine order at 15:46:01 to constitute a distinct occasion, I should also address other arguments on this issue.  Esso and BHP argued that in ordering the engine movement, the master acted in an unseamanlike way because he failed to act as if there were a risk that the ship had by then fouled the pipeline.  They contended that he should have concluded, by use of navigational aids available to him and the bridge team, as well as his own observations, that the vessel’s anchor may have fouled the pipeline;  and, if so, going astern would be likely to add to the damage already done and increase the danger of the pipeline rupturing.

192               The master acknowledged that he appreciated that there was a possibility that the anchor had already fouled the pipeline at 15:46:01 when he ordered dead slow astern, based on the changes to the heading that he had observed over the preceding minutes (T 386).  However, I have accepted his evidence that while aware of that danger, he was concentrating on heaving up the anchor in a fraught situation.  Awareness of such a possibility informs, but does not dictate, the decision making of a reasonable person in the position of the master.  The law cannot impose duties of perfection, for these are not the reactions of reasonable persons – they are quite unreal.  Here, just two minutes before, as Capt Cole graphically said, the ship was in trouble.  He and Capt Dwyer changed their evidence on a number of significant aspects as they reflected on matters.  Capt Xu did not have that luxury of reflection;  his ship was in trouble.  Its anchor was dragging towards, and possibly may have reached and fouled, the pipeline.  The gale was intensifying.  The anchor cable had been leading in a direction that made heaving the anchor difficult.  The bow was swinging round, as his preceding engine movements had sought to achieve, but also consistently with the independent possibility that the anchor had fouled the pipeline.

193               At this point, the question is whether any competent, reasonable master in Capt Xu’s position properly could have ordered dead slow astern to swing the bow to facilitate the heaving operation.  Absolute deference to the mere possibility that the pipeline had already been fouled would preclude a master from taking action that might prevent a later fouling, had it not already occurred.  Thus, much will depend on the actual signs that the master should have been alert to, and the navigational information he should have considered on the bridge, just before he made the order.

194               Having seen and heard Capt Xu, I am satisfied that he was a sensible, prudent person.  Had the alarm bells been ringing in his mind from the change in heading he was observing (and I am satisfied that he was paying proper attention) he would not have made the order.  I infer that the change in heading he observed was consistent with that which he had sought to bring about by the two immediately preceding astern orders.  I am not satisfied that these signs were sufficiently clear to manifest to him a strong enough possibility that the fouling had already occurred to deny him the option of ordering the engine to dead slow astern at 15:46:01 in order to assist in heaving the anchor.

195               Esso and BHP relied on the evidence of Capts Cole, Dwyer and Murchie to demonstrate that Capt Xu should not have made that order.  Capt Murchie, however, referred to the fact that while the master would have been able to observe the 25° change in heading in the two and half minutes before 15:46, one would need to know what engine movements the master had been making before concluding that the change in heading indicated that the anchor had been caught on something.  He was not asked about the position of the cable at the time and, as he said, the master may have been distracted by doing other things.

196               Capt Cole accepted that in an emergency, where time was short, a master might have to make a best judgment based on some uncertainties and thus run certain risks.  And, Capt Dwyer said that while it was human nature to use a method that one was familiar with, a master may have to make a choice that involves taking a calculated risk against his better judgment.

197               Capt Cole said that, in addition to proximity to the pipeline, one feature of the ship’s behaviour that would have alerted the master to the possibility that the anchor had fouled the pipeline, after 15:46 was the length of time she had been windrode.  But, he accepted that the master would only have arrived at that conclusion at about 15:54.  Capt Dwyer referred, on this issue, to the chief officer’s evidence that the cable was at about 10 o’clock to port as indicating fouling.  But, as he said, this was in the context that the master would have realised this only at about 15:47:31 when the stop engine order was made.

198               I am not satisfied that a competent master acting reasonably in Cap Xu’s position could not have made the order for dead slow astern at 15:46:01.  Rather, I am of opinion that this was finely balanced judgment which he made with due regard to the change in the vessel’s heading and all the factors, including the efficacy of the immediately preceding engine and helm orders in bringing the bow round.  I am satisfied that the order Capt Xu gave was within the range of possible courses of action open to a competent master acting reasonably.

199               Esso and BHP made a number of other arguments based on Capt Cole’s evidence as to what Capt Xu should have done at this point.  However, these were based on a factual premise that I have not accepted and indeed, Capt Cole recanted from;  namely, that the master should have realised by 15:46 that the anchor had fouled.  In that context, I need not consider those arguments.

200               No new or further relevant damage was done as a result of the engine orders between 15:46:01 and 15:47:31.  They did not give rise to new damage or claims different or distinct from those caused by the fouling of the pipeline:  cp  The Schwan [1892] P at 439, 441;  The Creadon 5 Asp MC at 586;  The Rajah LR 3 A&E at 542;  Ballast Trailing (NSWCA unreported, 29 September 1981).  I am not satisfied that those orders gave rise to a distinct occasion.

The windlass fails

201               The chief officer gave unchallenged evidence that as the cable was being heaved in he was counting the shackles while standing on the starboard bulwark about 10 to 12 metres from the bosun.  At about the time that the kenter link marking the second shackle had been hauled up to between the hawse pipe and the gypsy, the bosun told him that the windlass was unable to raise the cable any further.  This meant as a practical matter that there was about one shackle of cable in the water with most of the rest of the second shackle going up from the water’s surface to the ship.  The chief officer immediately told the master.

202               Soon after, at 15:48:07, Capt Xu radioed the pilot boat stating “… one shackle in the water, waiting for you”.  Capt Murchie responded:  “The anchor all the way home, all the way home”.  At that time the ship’s heading had moved to 164°.  Next, at 15:48:43 Capt Murchie called APL Sydney on the radio again instructing that the anchor should be fully raised.  Now Capt Xu’s voice had a real note of concern.  He said:

“Harbour control, is trouble, is trouble for our vessel, may be our anchor cannot upper, cannot upper.”

He said that he did not know the reason that the anchor could not be raised and that about two shackles were in the water.  The conversation continued:

“HARBOUR CONTROL:  Roger.  Can you use ship’s engines to maintain a safe distance from the pipeline?  Over.

APL SYDNEY:  I don’t know the pipeline – where are the pipeline.  Maybe our vessel is above the pipeline.  So I is in this trouble.  I cannot use the engine now.”

203               At that time the chief officer saw that the cable was drawn across the bow of the ship to the port side at about 10 o’clock (i.e. leading forward to port at about 60°).

204               This unchallenged observation of the chief officer about the direction of the cable at this time did not feature much in the considerations of the nautical experts in their concurrent evidence.  It gives some insight into the navigational difficulties with which the master was contending.  As the cable shortened the bow was swinging to the south.  And, with the cable leading in the position described by the chief officer when the windlass stopped, there was tension on the cable.  The cable had been realigning itself in relation to the bow, over the preceding 8 or 9 minutes.  During the dead slow astern movement beginning at 15:46:01, the bow moved first back slightly toward the north and then came forward to the south as the stern swung to port (eastwards).  As the cable was shortening while being retrieved, it could have come across the bow to the port side in the way that the chief officer described in his unchallenged evidence.  That evidence was confirmed by him in cross-examination.  This general movement of the ship to the south, and the stern to port (and east), continued so that the ship would be heading southwards toward the cable but just north of where the cable had led out of the water (i.e. so that the cable would lead to port and be drawn across the bow, as the chief officer described).

205               During the period between 15:47 and 15:49, but probably just before the time that the master informed the pilot boat that one shackle was in the water, the chief officer instructed the bosun to pay out 1 to 2 metres of cable and then to start raising the cable again with the windlass.  It is a common practice to pay out a small amount of cable in such situations to take pressure of the windlass and allow it to recommence lifting the cable once more.  However, on this occasion, after the bosun paid out 1 to 2 metres, the cable suddenly began paying out very fast.  The bosun was standing at the windlass controls.  The chief officer estimated about half a shackle had paid out before he turned, first to the windlass controls, and then he saw that the bosun had moved about 10 metres to the left (port) to apply the hand brake above the gypsy.

206               The bosun said that he applied the hand brake instinctively to stop the chain running out completely.  He did not know how much cable had paid out before he had been able to apply the hand brake.  The bosun said that the only noise he heard then was the sound of the cable.  The chief officer now told Capt Xu that they could not raise the anchor and there were about two shackles in the water. 

207               I infer that as the chief officer was alerting Capt Xu to this important development, he broke off his communication with the pilot boat that had begun at 15:48:07. By 15:48:43 Capt Xu was able to piece what was happening together.  The proximity of the pipeline, the inability to raise the anchor and, no doubt, the continued swing of the bow into the wind, led the master to his decision to await the pilot’s return.  He understood that in the circumstances that had unfolded, he should not use the engine and said this to the pilot boat and harbour control.  Capt Xu knew then (at 15:48) that the windlass was broken, but he did not know the nature of the problem with it or whether it would be easy or difficult to fix.  He was now aware that it was possible that the anchor had fouled the pipeline or that it was holding under the seabed.

The events from 15:48 to 16:18:51

208               During the preceding 45 minutes the weather had been worsening.  But worse was to come.  The bosun noticed an oil spill on the forecastle deck.  He and the chief officer immediately inspected the windlass and discovered that its hydraulic motor had broken. Pieces of the motor were lying on the deck amongst the oil.  The oil presented two immediate problems; first, it made working on the forecastle deck hazardous, particularly in the prevailing heavy weather conditions;  secondly, if it ran off the ship into the water, the ship and master would have been liable for offences of causing pollution.  Accordingly, the chief officer immediately gave instructions to the forecastle crew to contain and clean up the spilled oil.  I infer that the clean up operation was conducted expeditiously and efficiently but went on for some time, beyond the time the pipeline ruptured about half an hour later.

209               Four minutes later at 15:52:23 the pilot radioed Capt Xu asking him to put the engine ahead so as to bring the ship over the anchor and cause the anchor to go up and down before recommencing to heave it.  The master said that they could not heave the anchor “… now because there is trouble”.  Then at 15:55:05 the master reported to harbour control that the windlass was broken and could not be used to heave the anchor.  I find that Capt Xu passed that information on as soon as the chief officer had made him aware of how badly the windlass was broken.  The chief officer said that at this time the cable was very tight and leading out from the bow at 10 o’clock.  I find that he also told this to Capt Xu.

210               In a joint expert report on the failure of the windlass, Brian Doherty and Jack Hunter agreed that the hydraulic motor driving the windlass failed because of over-speed and over-loading.  They concluded that these had been due to large forces being applied to the motor by the cable as a result of the sudden change in the movement of the vessel when the anchor fouled the pipeline  For the reasons I have given above, the state of the windlass became evident to the forecastle party within a few minutes after the fouling.

211               At 16:00 the pilot boat reached APL Sydney and reconnoitred around her to assess the best embarkation point for Capt Murchie to climb up a pilot ladder in the conditions.  As he approached on the ship’s starboard side in the pilot boat, Capt Murchie saw some hydraulic oil on deck and a lot of activity near the windlass.  He saw people racing about and then using mops and buckets.  Capt Murchie also observed then the cable to be on a relatively short stay of at 45°.  This was the angle at which the cable entered the water (not the direction in which it was leading).  He said a short stay was consistent with there being 2 shackles in the water.  By 16:03:53 Capt Murchie was on the pilot ladder.  The pilot arrived on the bridge a few minutes later and discussed the situation with Capt Xu.  In their discussion Capt Xu told Capt Murchie that the starboard anchor was still out with two shackles in the water and the windlass was inoperable.  Initially, Capt Murchie said that the anchor should be jettisoned.  Capt Xu agreed.

212               The master said in evidence that he then instructed the chief officer to make arrangements to decouple the chain and jettison the anchor.  I think Capt Xu was mistaken about the time of this instruction.  I find that he only gave an instruction to the chief officer to decouple the anchor cable later.  This was at about 16:50, after the pipeline had ruptured, when he told the chief officer to go to the chain locker and disconnect the cable at the third shackle kenter link.  This is what the chief officer and bosun had said in their affidavits and was also recorded in the master’s statement of facts of 13 December 2008.

213               I find that the pilot told the master shortly after he arrived back on the bridge at about 16:10 that he should buoy the cable and break it by disconnecting it at a kenter shackle or cutting it.  Capt Xu told the pilot that he needed to get permission to take that course of action.  Capt Murchie was referring to two of the three ways in which an anchor that cannot be heaved in may be disconnected from ship.  Breaking the cable would involve the crew on the forecastle either finding a kenter shackle link on deck and disconnecting it or using an oxyacetylene torch, or “gas axe” to cut or “gas axe” the steel chain.  A buoy can be attached to the cable just on the side that is to be released to mark in the sea where the cable and anchor are so that they can be retrieved later.  The pilot said that disconnecting at a kenter shackle link usually required heat and was a long and drawn out process.  He never considered and did not discuss with Capt Xu the third way, namely releasing the cable from the bitter end.  I will deal with these methods of releasing the anchor cable in detail below (see “The Nautical Experts’ Contended Courses of Action” [229]-[259]).

214               Capt Murchie told harbour control at 16:10:57 about his suggestion to the master of breaking the cable.  He said that he could not see any other way around the problem of the broken windlass.  He said that the master was then speaking to his principals.  The pilot enquired about how deep the pipeline was.  Neither he nor harbour control knew whether it was on or below the seabed.  Harbour control said that his plan to drop the anchor (cable) sounded sensible.  Capt Murchie said that he was open to suggestions.  Harbour control told him that the ship was “right in the pipeline area now”, and he replied that they were virtually right over the top of it.  Significantly, as will soon appear, he said that the ship was north of the pipeline but the anchor was south of it.  Harbour control, presumably looking at the AIS, and knowing that there were only two shackles of cable in the water said:  “Yeah, I can see, yeah” and confirmed that breaking the cable “… sounds to be the most sensible thing to do” and “the best course of action”.  It enquired whether the ship could be held in position by using the engine.  Capt Murchie replied that he thought so, but it was hard to tell.  He said that the language difficulties, however, were “extreme”.

215               At the same time, Capt Xu used a satellite phone to call Sun Fagen in Shanghai, China.  Mr Sun was the executive in the shipowner’s office to whom Capt Xu reported.  Mr Sun had been a chief engineer, not a master.  He told Mr Sun that the vessel was very close to the pipeline, the anchor could not be heaved up, the windlass was broken and it was possible that the anchor had fouled the pipeline.  At that stage the master remained uncertain whether the anchor had fouled the pipeline, but was very concerned about the vessel’s proximity to it, the possibility that the anchor had fouled it and the associated danger.  Capt Xu thought that he also had spoken to Mr Sun a little earlier, between the times the chief officer had told him that the windlass could not heave the anchor up and when he reported to harbour control that the anchor could not be raised and two shackles were in the water.  He wanted Mr Sun’s approval to jettison the anchor.  Mr Sun gave him permission to jettison the anchor by decoupling it.

216               The master said that the anchor could not be jettisoned immediately and the procedure to decouple the chain required preparation, which took time.  He was familiar with the procedure of decoupling the shackle kenter link.  Capt Xu agreed with the pilot’s suggestion of jettisoning the anchor because he feared that it might be connected to the pipeline.  He knew that if the anchor was fouled, the only way for the ship to move was to jettison the anchor.  Capt Xu was aware that the anchor could be let go at the bitter end, but this was not raised by him, the pilot or Mr Sun.  He was going to use decoupling because he was familiar with that method.  He would also have sought permission from Mr Sun before letting the anchor cable go from the bitter end, because these were his orders.  He considered that this step was a large responsibility.  He had never cut an anchor cable in his over 20 years at sea.  He said that he did not think that the situation the ship was in was an emergency justifying letting the cable go from the bitter end because the pilot said that he could use the engine.  I do not accept Capt Xu’s evidence concerning use of the bitter end at this time.  I find that he did not think of using the bitter end before the pipeline ruptured.

217               After he had spoken to Mr Sun, Capt Xu came back to see that Capt Murchie was plotting the position of the ship on the Admiralty chart.  Capt Xu told him that he was now ready to jettison the anchor.  The pilot, however, replied that he had plotted the position and the ship could leave safely without needing to jettison the anchor.  Capt Murchie showed the master where he had plotted the position of the anchor to the south of the charted position of the pipeline (which, unknown to them, was charted about 30 metres south west of its actual position).  The pilot drew a line on the chart between the ship and the anchor.  Capt Murchie agreed in his evidence that if he had not used a triangulation to calculate the ship’s position there was a chance of error in what he marked on the chart.  Although he could identify one bearing on the chart that he took from the Fawkner Beacon, he could not identify another bearing he had used to make a triangulation.  I find that he did not accurately plot the position of the ship and failed to use a triangulation.

218               Capt Xu said that the line drawn by the pilot on the chart was in the same direction as he understood was then the direction of the cable.  He said that he maintained regular contact with the chief officer to be updated on the direction of the cable.  The chief officer said that he was busy dealing with the oil spill at this time and he was not monitoring the cable.  Capt Xu did not tell the pilot, and Capt Murchie did not ask, him what the direction of the cable was.  The master was aware that, earlier, the pilot boat carrying Capt Murchie had approached the bow of the ship from the starboard side and gone round the stern to the pilot ladder on the port side.  Thus, he understood (as was the case) that Capt Murchie had seen the direction of the cable at that time.  However, the ship was swinging in a series of yaws that changed its relationship to the direction of the cable as the swings progressed.

219               Shortly before 16:20, the pilot advised Capt Xu that the engine be put to dead slow ahead.  Capt Xu said, and I accept that he just followed the pilot’s advice and believed the pilot’s advice to be correct because of his knowledge of the pipeline.  But, before using the engine to go dead slow ahead as Capt Murchie advised, the master asked him once again:  “Can we do this?”.  Capt Xu said that he knew nothing about the pipeline and wanted to be sure because he was worried.  After all, once he had realised at about 15:47:31 that the anchor might be fouled on the pipeline, he had dared not use the engine.  The pilot replied that they could.  Capt Xu gave this evidence which I accept:

“MURDOCH:  When the pilot directed the vessel to go dead slow ahead you were still worried about the possibility that the anchor was attached to the pipeline, but you were hoping that it wasn’t.

THE INTERPRETER:  Because at that time when the pilot was quite adamant very clear about what he said, he told me the position about vessel and also about the anchor, and also he told me about the direction of the cable. If I could leave the position safely, why I would not give it a try?

MR MURDOCH:  So for you the fact that the pilot said to you that the anchor was on the other side of the pipeline was enough to quell all of your fears?

THE INTERPRETER:  Well, I still had worries, but there are a lot worries at the vessel and we can’t just don’t do anything because there were worries.”

220               Capt Xu implemented the pilot’s advice and ordered the engine dead slow ahead at 16:19:51.  When the pilot gave that advice he was aware that the ship had been static and windrode for the whole time since he reboarded at 16:03.  And Capt Murchie understood that in the circumstances he needed very accurate information before he could give advice to the master to go dead slow ahead, when there was a real possibility that the reason the ship had been windrode was that the anchor had fouled the pipeline.  But he simply made an assumption that the cable was leading “pretty much ahead”.  He did not even ask Capt Xu to ascertain whether that was correct.  The ship, however, had been through three large swings or yaws between the north west and south east between 15:48 and 16:19.  The heading went from 164° to 232°, back to 187°, then to 233°, then to 184° and at 16:20:06 was 193°.

221               Capt Murchie gave his evidence candidly.  But, his plot of the position at 16:15 and his advice to Capt Xu to order dead slow ahead at 16:19 had no justification in the circumstances.  Capt Murchie did not check that his assumption as to where the cable led was accurate.  He did not do any calculations to see if, with 2 shackles in the water, his assumption was possible, and he did not triangulate properly to plot the position of the ship.  In addition, he knew that the vessel had been windrode for the preceding quarter hour since he had reboarded.  Thus, he knew something was holding the anchor in place and that either was because it had grounded or had fouled the pipeline.  The pilot did not discuss with Capt Xu why he had thrown caution to the wind, but I infer this was because Capt Murchie was confident in his plotting of the anchor’s position as being clear and south west of the charted pipeline.

222               Capt Xu should have checked the position plotted by the pilot himself before implementing the advice to put the engine dead slow ahead.  Had he done so, he would have discovered the mistake made by Capt Murchie in plotting the position of the ship.  Both master and pilot were negligent.  Both had ample time to check the position plotted by Capt Murchie at 16:15 before commencing to go ahead.  Neither did.  The pilot’s advice did not have to be carried into an order by the master before any check of his plotting was made.  This decision was made with both the master and the pilot being aware of the background of the earlier navigational errors and their consequence that the vessel possibly had fouled the pipeline.

223               I  accept Capt Cole’s opinion that Capt Murchie did not provide the master with the level of advice that a seafarer was entitled to expect from a person with an unlimited pilot’s licence for Port Phillip Bay.  As Capt Third pointed out, the decision of the pilot to suggest the engine movement of dead slow ahead indicated a change in the navigational priorities.  This demonstrated that at this time a distinct occasion had occurred.  What made the occasion distinct?  It was because new claims arose from an avoidable and unnecessary act, neglect or default of the persons responsible for the conduct and navigation of the ship.

224               I find that the ahead movements of the ship initiated as a result of Capt Murchie’s decisions to abandon the plan of jettisoning the anchor, to plot the position of the ship and anchor, and to advise Capt Xu to order the engine ahead, together with Capt Xu’s implementation of the pilot’s advice amounted to a wholly separate and distinct occasion from the earlier fouling of the pipeline at 15:44 to 15:45.  The consequence of these acts and omissions was that the pipeline ruptured and this allowed a not insignificant quantity of gas to escape.

Effect of Essso and BHP not pleading that the cable should have been slipped from the bitter end

225               During final address, counsel for the shipowner raised an objection against Esso and BHP being allowed to rely on the master’s failure to order, after APL Sydney’s anchor had fouled the pipeline, that the cable be slipped from the bitter end.  The shipowner contended that because this point had not been pleaded Esso and BHP could not rely on it.

226               This argument has no substance.  It does not reflect the way the case was conducted.  In his initial report, Mr Kelle raised this course of action as an appropriate means of freeing the ship from its continuing engagement with the pipeline.  Capt Third in his report, for the shipowner, discussed the use of the bitter end as one of three options open to the master and pilot.  And, the shipowner’s written opening submissions expressly referred to this evidence and asserted that the master and pilot had considered the three options discussed by Capt Third.  Those written submissions then identified a disagreement between the nautical experts (Capts Cole, Dwyer and Third), in their reports about the length of time that would be taken in using the bitter end.  During the hearing, evidence was led by both sides about the use of the bitter end without objection.  And during the concurrent evidence of the nautical experts, the bitter end was a topic that counsel agreed would be examined, as it was, at length.  Both parties made extensive reference to this issue in final submissions.

227               The trial had, and I always understood it to have, as a real issue, whether the master and pilot should have slipped the cable from the bitter end and what that operation would have involved.  The parties fought the trial with that as a substantive issue.  In Vale v Sutherland (2009) 237 CLR 638 at 651 [41] Gummow, Hayne, Heydon, Crennan and Kiefel JJ approved what Dawson J had said in Banque Commerciale SA (In liq) v Akhil Holdings Ltd (1990) 169 CLR 279 at 296-297):

“But modern pleadings have never imposed so rigid a framework that if evidence which raises fresh issues is admitted without objection at trial, the case is to be decided upon a basis which does not embrace the real controversy between the parties ... cases are determined on the evidence, not the pleadings.”

228               Their Honours applied what Dawson J had said, concluding that whatever view were taken of the range of issues tendered by the pleadings, the trial judge was entitled to decide the case as he had and that this caused no unfairness.  I am satisfied that the issue concurring the bitter end was a substantive part of the controversy that the parties fought at the trial and that I can, and should, consider it.

The nautical experts’ contended courses of action

229               I have found some difficulty in assessing the nautical expert evidence offering criticism of the conduct of Capt Xu in the crisis in which he found himself.  Each expert had the ability to reflect for some time before offering an opinion.  While they did not have access to the voyage data recorder information (that would have recorded helm and bow thruster orders) or the communications between the master and crew, the experts had a wide range of data, including affidavit evidence from the crew, and, after their initial reports, the expert opinions of the others.  Even so, as the shipowner’s submissions identified, Capt Cole and Capt Dwyer changed their opinions on important matters radically.  And, so did Capt Third.  This may well be a mark of their intellectual integrity;  that is, each was honest and sensible enough to revise his view when flaws appeared.  But, these changes underscored the very difficulty of the situation in which Capt Xu had to act.  Since three experienced and knowledgeable experts changed their already differing views of what a reasonable master in Capt Xu’s position ought to have done, there is an immediate incentive to caution in my making criticisms of his conduct.

230               This dilemma was graphically illustrated in the following cross-examination about the advisability of use of the engine between the fouling and 16:15.  Capt Cole had the opinion that the engine should have been used to position the ship statically over the anchor.  Capt Dwyer changed his initial opinion which had been the same as Capt Cole’s, after he considered Capt Third’s opinion that trying to position the vessel statically using the main engine was dangerous, if not impossible.  Capt Dwyer’s new view accorded with Capt Third’s, namely that the engine should not have been used.  That, of course, was also the decision Capt Xu made at the time.  Capt Dwyer said that the risk from using the engine was that if the master made a wrong assessment of what movement to order, he may cause the pipeline to rupture with consequent risks of explosion, fire and loss of life.  Then Capt Cole and Capt Dwyer gave this evidence, each having affirmed that he regarded his own view to be seamanlike:

“MR SCOTT:  If the two of you were on the bridge of this ship at 15:44 through to 16:15, is the court to understand your position to be that you would be giving conflicting advice to this master? Is that correct?

CAPT DWYER:  Well, there would be no position where the two of us would be - - -

MR SCOTT:  I’m asking you to make any assumption that you’re there and that you know what you now know, which is more than he knew, and you are giving him advice about what is seamanlike. Do you agree that one would be saying “Go” and the other would be saying “Stop”?

CAPT DWYER:  I would be saying “Ease the cable”.

MR SCOTT:  No, I’m sorry, I should have been more precise. One of you would be saying “Manoeuvre the engines” and one of you would be saying “Do not manoeuvre the engines”.

CAPT DWYER:  Yes.

CAPT COLE:  And I would be saying that that is a strategy we’ve got to try.

MR SCOTT:  I understand. Now, Captains, I will put this to you, Captain Cole, first.  The difference of opinion that you two have, based on the same assumptions of fact, and after all these years of experience that you’ve got reflects, does it not, the fact that in an emergency situation two people applying seamanlike principles can reach different conclusions about the appropriate course of action?

CAPT COLE:  Mr Scott, yes.

MR SCOTT:  And Captain Dwyer, you also - - -

CAPT DWYER:  I agree.”

(a)        Use of the Engine

231               Esso and BHP argued that the master should have used the engine to statically position the ship, based on Capt Cole’s evidence.  I am not satisfied that the master made any error of seamanship in his decision not to use the engine during the period between 15:48 and 16:19:51.  I accept the reasoning that Capt Xu expressed at the time – namely that this would be dangerous – and the reasons of Capt Third and Capt Dwyer.  And, in any event, this divergence of expert views, satisfies me that the course advocated by Capt Cole was not the only one necessarily required in the circumstances and that Capt Xu acted reasonably in adopting the position he took.

(b)        Use of a gas axe

232               Esso and BHP argued that the master should have used a gas axe to cut the cable on the forecastle based on Capt Cole’s evidence.  Capt Cole said in his second report that the master should have taken the weight off the cable by using engine, helm and bow thruster movements before releasing additional chain to have about six shackles in the water while preparing to jettison the anchor and chain.  He thought this would place APL Sydney at a sufficient distance from the pipeline and prevent its rupture.  That would have moved the ship about 110 metres to the north east with the wind blowing towards her.  I am not satisfied that if the pipeline ruptured this was a safe distance to protect against the consequences of a naked flame on the ship coming into contact with the escaped gas.

233               Capt Cole also agreed with the pilot’s initial view that the master should give orders to break the anchor chain and buoy the cable.  In his oral evidence, Capt Cole said that he would consider using a gas axe to cut the cable as one of his options.  As he said, he would weigh up his options before deciding to use it in preference to other options.  From his other evidence I infer that he would have preferred to use the bitter end to jettison the anchor rather than a gas axe, but held the latter method open as an option to be seriously considered.

234               I am of opinion that, the use of a gas axe was not a reasonable option in the circumstances.  It was a course self evidently fraught with danger.  I reject Capt Cole’s evidence.  I accept Capt Dwyer’s evidence in the following passage for the reasons that he gave, which are similar to those of Capt Third, which I also accept:

“MR SCOTT:  Can I get one thing straight. Captain Dwyer, you first.  One thing you would not do in proximity to a fouled high pressure, highly flammable gas pipeline, is get out the gas axe and cut the cable on the focsle. Is that correct?

CAPT DWYER:  That’s correct.

MR SCOTT:  And that’s because of the fire and explosion risk?

CAPT DWYER:  Well, if there’s gas escaping you’ve got a fire and explosion risk.

MR SCOTT:  Well, if you don’t and you start cutting and then the gas escapes you’ve got a fire and explosion risk, don’t you?

CAPT DWYER:  Yes.

MR SCOTT:  Thank you. Captain Cole, do you agree with what I just put to Captain Dwyer?

CAPT COLE:  The short answer is no.

MR SCOTT:  You would be prepared to get the gas axe out on to the fo'c'sle deck?

CAPT COLE:  Yes, I would.”

235               Capt Cole’s view failed to have reasonable regard to the risk of any flame or fire igniting gas from the pipeline if it ruptured.  Even if 110 metres (four more shackles) of cable had been released, the ship would have been only that distance downwind away from the pipeline.  There was no evidence, even from Capt Cole, that, if the pipeline ruptured, this position was a safe distance at which it would not be possible to ignite the gas.  Capt Dwyer and Capt Third thought that about 90% of the crew would be smokers.  The master had no knowledge of the risk of rupturing, nor did he receive any assistance on that issue from harbour control or Capt Murchie.

236               I reject the use of the gas axe as an unsound and dangerous method.  Indeed, later in the concurrent evidence when dealing with the master’s decision to go astern away from the gas cloud after the pipeline ruptured, all three nautical experts, including Capt Cole, agreed that Capt Xu had a reasonable concern to avoid the gas cloud.  This was because the master could not be certain that all flames were out throughout the ship so that there was a risk of fire or explosion if it passed through the gas cloud.  Similar considerations applied, in my opinion, to the risk of fire or explosion, before the pipeline ruptured if a gas axe were being used, particularly since the forecastle was closer to the anchor and pipeline about 150 meters forward of the bridge and accommodation.

237               Moreover, I cannot conceive how any reasonable person could even have imagined using an oxyacetylene torch to cut the cable if the ship remained where she was, when there was a real possibility that the anchor had fouled, and might fracture, the gas pipeline.  If the pipeline broke, the danger from erupting gas meeting any sort of flame is just obvious, let alone the heightened risk from an oxyacetylene torch and setting off a catastrophic explosion, fed by the continuing emission of gas from the pipeline.  I accept Capt Dwyer’s evidence to this effect.  The lives of those on the forecastle, and possibly everywhere else on board would have been at immediate peril were Capt Cole’s suggestion implemented.  What the possible explosion may have done to the ship was not the subject of evidence, but the risk of a significant fire was also obvious.

238               It is disturbing that the port authority, through harbour control, not only did not prohibit the pilot’s suggestion of breaking the cable instantly, it did not even appear (at least from the radio communications with the pilot and APL Sydney) to have had any emergency plan for the contingency that was unfolding, let alone a sound plan that absolutely prohibited using oxyacetylene or gas axe equipment if there were the slightest risk that the pipeline had been fouled or could be ruptured.  The port authority’s pilots and harbour control radio operators should have been trained to meet such a contingency and have had instilled in them that on no account should any source of fire be used to release an anchor possibly fouled on the gas pipeline.  The gas pipeline was a significant item of infrastructure for Melbourne.  It was also a marine hazard marked on the Admiralty chart and, no doubt, was one reason for there being compulsory pilotage in Port Phillip Bay.  The evidence before me disturbingly does not suggest that the harbour control authorities, who had dismissed Capt Xu’s sensible request to move his ship to safety at 15:05, in fact had any training or emergency plan to deal with the consequence of their direction to him that unfolded. 

(c)        Release from the bitter end

239               Each of the nautical experts and Mr Kelle, at various points in their thinking came to consider releasing the anchor cable from the bitter end.  Again, a range of opinions emerged about how this operation would be conducted.

240               At 15:48 the master was aware that something was wrong with the windlass.  Soon after the chief officer reported the oil leak and the master ordered that it be cleaned up.  Shortly afterwards he radioed harbour control at 15:55:05 informing them that the windlass was broken.  During that period of about six minutes the master had a great deal to occupy his mind;  first, the realisation that his ship’s anchor may have fouled the pipeline;  secondly, his consideration that there were two shackles of cable in the water, the windlass was broken and the ship was being buffeted by the gale;  thirdly, the oil spill on the forecastle and the danger that presented, first, to the crew in attempting to repair the windlass, if that were possible, and secondly, in polluting the water below.

241               In order to let the chain go from the bitter end some preparation and precautions had to be taken.  The bosun had to release the hand brake to allow the cable to pay out.  There were two methods of release;  controlled, where the bosun remains near the hand brake, and uncontrolled where the bosun releases the hand brake and the chain runs free, at increasing speed.  With either form of release, as the last links of the cable come up through the spurling pipe to feed into the gypsy, they flail about so there is a real danger of serious injury to anyone standing near the hand brake.  Oil was over the forecastle deck around the windlass, in a quantity noticeable to Capt Murchie as he approached on the pilot boat.  The waves were large and irregular and the wind was gusting at gale force.  The oil had to be cleaned up thoroughly before the forecastle would be safe for the bosun and other crew to arrange for and conduct a release of the cable from the bitter end.  In addition, the crew would need to be instructed carefully as to their roles. 

242               The nautical experts differed as how quickly these steps could be undertaken and what was involved in them.  Each of the nautical experts agreed that the master should have acted, but they differed on the timing.

243               Capt Cole had the view that the pilot should have prepared a contingency plan in advance from about 14:50 to 15:00.  Initially, Capt Cole did not even consider releasing the cable from the bitter end;  he suggested using the engine and the gas axe.  The latter suggestion resulted, perhaps, because he had been advised, incorrectly, by Mr Doherty that the anchor chain could neither be heaved in nor paid out when the windlass broke.  Significantly, in his first report Capt Cole said that the cutting and buoying process “is time consuming” and “… would have required a high degree of concentration on the part of the master”.  The gas bottles were stored in the engine room towards the other end of the ship.  He also suggested that the pilot could have assisted when he reached the bridge at around 16:11.  In concurrent evidence, he thought it would take about 15 minutes to get the equipment from the engine room and a further 13 minutes to cut the cable based on the crew’s evidence of how long it took later on 13 December 2008, a total of 28 minutes.  He said the chief engineer needed to organise an appropriate crew member to use the gas axe.  Capt Cole said that the oil on the forecastle could be cleaned up by using the firehoses “… to eliminate the prospect of a fire on deck”.  He did not know how much oil was on the deck.  Capt Cole did not explain how that necessarily would have made the deck safe for the crew to walk over, especially in the weather conditions and if they needed to take quick action to deal with a cable running past or flailing.

244               Capt Dwyer also did not assess the amount of oil on the deck.  But he accepted that the forecastle deck had to be made safe for the crew’s future operations.  He suggested using some absorbent substance, like kitty-litter, to soak up the oil.  There was no evidence that, first, such a substance was available on board or, secondly, that it would render the forecastle deck safe in the conditions (the weather and the presence of the oil).

245               I prefer Capt Third’s approach to releasing the bitter end.  I was impressed by his more measured approach to the evidence he gave in contrast to each of Capt Cole and Capt Dwyer.  They were more dogmatic in their responses than Capt Third who, I found, had a more appropriate concern for the reality of the effect of oil on the forecastle deck and the safety of the crew.  Capt Third had observed, just before Capt Dwyer gave his answer about using an absorbent substance, that the hydraulic oil on the deck would make it difficult to keep one’s footing and that it would be necessary to clear the area near which the chain would run of both any impediments and of the crew.  He said that the deck was painted and with oil on it “… it’s like an ice rink”.  He also described the steps that needed to be taken to ensure that the release of the chain from the bitter end would be successful.  These were the selection of a controlled or uncontrolled release method.  That required the crew involved to discuss what they felt they could achieve.  And the crew members would need to be briefed properly on the proposed operation.

246               Capt Third regarded good planning as very important to ensuring that the operation succeeded.  The crew would need to make safe and secure anything that the cable could make contact with so that it would not cause a problem.  Thus, the guillotine stopper on the deck between the gypsy and the hawse pipe would have to be doubly secured so that it did not become dislodged and so the fast running chain did not become jammed in it.  The grating behind the hand brake was not a safe place to stand as the cable came out at speed.  The chain, coming out of the spurling pipe from the chain locker at speed, would be jumping and whipping, creating danger for anyone in proximity to it.  The bosun or other crew member operating the hand brake to release the cable would need to be able to get away down the small ladder onto the forecastle deck and move towards the port side of the deck after the release.  That had to be done without any risk from the oil, or its residue on deck, that he would slip, fall or injure himself. 

247               The bitter end is held by a pin that is designed to be released under tension.  The bitter end pin on APL Sydney was contained behind a hatch cover in the forecastle that Capt Cole had inspected.  Mr Kelle said the cover had between 12 to 16 brass butterfly nuts.  An emergency hammer was stored in the vicinity of the hatch.  The pin could be hammered out if the cable was under tension.  Once the pin was driven out, Capt Third said that the bitter end should be held by a light rope, before the cable was slipped (i.e. released by the hand brake).

248               Each of the nautical experts agreed that use of the bitter end to jettison the anchor and cable is not at all a common occurrence.  Each had one direct or indirect experience of its use or some similar event on a ship on which he had served.  As Capt Third said it is, if not a once in a life time experience, “a pretty rare occurrence”.  That evidence, again, assisted me in accepting Capt Third’s general approach to the use of the bitter end in preference to that of Capt Cole and Capt Dwyer.

249               Mr Kelle said that with a reasonably experienced crew the whole operation should take about 10 minutes.  Initially Capt Third said in his report that organising the whole operation would take about 20 to 30 minutes.

250               Capt Cole said that a controlled release, which was his preferred option, would take five minutes and an uncontrolled release less than three minutes.  He would have used a sacrificial pennant in a controlled release to lessen the possibility of the last 20 to 30 metres of chain fouling the gypsy, compressor or hawse pipe.

251               In his first report, Capt Dwyer did not mention the use of the bitter end, referring only to the need to use the engine to take tension off the cable while awaiting a tug to hold the ship until the cable could be parted.  However, he changed that opinion after reading Capt Third’s report.  Capt Dwyer’s second report expressed the view that after 15:48 the master should have begun the process of jettisoning the anchor.  He said that the master’s only action should have been to increase the scope of the catenary effect by paying out more cable.  This would reduce tension on the chain and weight on the pipeline.  He said that the most appropriate action was paying out the cable and letting it go from the bitter end.  Capt Dwyer said that this would take not more than a few minutes.

252               Capt Third agreed that a bosun would carry out controlled releases all the time in the process of dropping anchor.  However, each time the hand brake is used it gets hot because it is a friction brake.  And, he said the hotter it gets the less effective it becomes.  Thus, in a controlled release of the whole cable the brake may not stop later releases as precisely as when it was cooler, and thus more cable may have paid out, making precision in calculating the amount left more difficult.  He considered that Capt Cole’s idea of using a sacrificial pennant an over-complication.  Capt Third conceded that the actual controlled release would probably be done in less than two minutes by making one stop and then letting the cable go.  Capt Cole agreed.

253               Ultimately, all three nautical experts accepted that a controlled release of the whole cable from the bitter end could be achieved (after the preparatory steps had been done) in about two or three minutes.

254               Capt Third also cautioned against first letting cable out to have six shackles in the water as a means of reducing tension on the chain before letting the cable go from the bitter end.  As he said, this was a difference in professional judgment.  His opinion, was that this could create particular danger if, in the conditions of the day, the ship was taken by the wind and started to yaw.  By the time she reached the extremity of the yaw the tension in the cable would be significantly greater.  As he said:

“The more cable you let out, the bigger the yaw.  The bigger the yaw, the higher the spike [of tension into the mooring system connected to the pipeline].”

255               Capt Cole and Capt Dwyer said that the master should have been able to control the ship with five or six shackles out in the conditions of 13 December 2008 using the engine, bow thruster and helm.  While Capt Third exposed a risk, I prefer the opinions of Capt Cole and Capt Dwyer on this issue.  The master had been able to control the ship by his orders between 15:30 and 15:36 when he first began heaving anchor with five shackles in the water.

256               Significantly, Capt Third said that in all the circumstances from 15:48 to 16:19 the master, and the pilot, had the time and opportunity to initiate and complete the process of letting the cable go from the bitter end.  I accept that evidence.  It conforms to the time frame he had outlined and it allowed enough time to ensure that sufficient crew would be ordered to assist in it while the forecastle deck was also cleaned of the oil and rendered safe.  However, that period did not include time for the master to reflect and decide upon a course of action after he would have realised at 15:48 that the anchor may have fouled on the pipeline.  It is significant that letting the chain go from the bitter end did not occur at all to Capt Dwyer when he wrote his first report.  Nor did this course of action occur to Capt Xu, Capt Murchie or harbour control at the time.  Because of the erroneous information he had received about the inability of the chain to pass either way through the windlass, Capt Cole did not address the issue in his first report.

257               I am satisfied that following a realisation at about 15:48 that the anchor may have fouled the pipeline, after a reasonably short period of consideration, of about 10 to 20 minutes, a reasonable master in Capt Xu’s postion could only have arrived at a conclusion that letting the cable go from the bitter end was the correct and necessary course of action.  I include in the course of action he should have followed paying out about four more shackles and making appropriate engine, helm and bow thruster orders.  The time frame for thought that I have allowed, takes account of the problems confronting the master at that time;  first, the immediate issue of oil on the forecastle deck, secondly, the dilemma of selecting the appropriate course of action, eliminating others and then formulating a plan in his own mind.

258               Thus, Capt Xu should have arrived at, and begun implementing, this conclusion by no later than when Capt Murchie returned to the bridge with his inappropriate suggestion of breaking the cable.  Capt Xu, and indeed harbour control and Capt Murchie, knew nothing about the size of the pipeline, its capacity to withstand the yawing of the ship, or its physical location below the surface.  This made it urgent that action be taken sooner rather than later in case the pipeline proved insufficiently strong to withstand the loads imposed on it if the anchor had fouled it.  With the benefit of hindsight, Capt Murchie gave evidence that if he had known that the ship had been weathervaning for half an hour before he advised the dead slow ahead engine order he would have dropped or broken the cable.

259               The necessity to take a further decision about what to do soon after 15:48 arose because of the fouling, but in a new context.  Unless the pipeline would have ruptured before the cable were slipped from the bitter end, the choice of another course of action, such as that made by Capt Murchie and followed by Capt Xu of moving the ship ahead was likely to cause the pipeline to rupture and present them with more problems, including the escaping gas.

The pipeline ruptures

260               As I have described above, Capt Xu implemented Capt Murchie’s advice to order APL Sydney to go ahead at 16:19:51.  The engine went dead slow ahead but was stopped at 16:20:09 and then put to dead slow ahead again at 16:20:13.  The ship began moving ahead to the south south west (or between about 193°-200°).

261               Shortly before a second stop order was given at 16:21:27, Capt Xu, from his position on the bridge, saw gas bubbling to the surface off the starboard bow at about 30°.  He pointed this out to the pilot.  The engine was then stopped.  The master concluded that the pipeline had been ruptured by the anchor.  The chief officer also said that the gas was visible at about 1 o’clock off the starboard bow (taking the centre of the bow to be 12 o’clock).  The gas was bubbling on the surface about 50 metres away from the bow.  The pilot boat was to south west and took photographs of the ship and the gas bubbling.  These show the anchor chain leading towards the direction of the gas bubbling on the water surface at about 40°, as Mr Kelle calculated.  This angle also suggested that the anchor chain deployed was longer than the length of chain recovered later.

262               The pilot instructed that the bow thruster be used to keep the ship from moving towards the gas plume.  Capt Xu ordered the crew to put out all sources of flames, including cigarettes and stoves.  He was concerned to ensure the safety of the crew.

What caused the pipeline to rupture?

263               The AIS data shows that at 16:18:07 the yaw to port finished and then the bow began gradually swinging to starboard from 184° to 200° at 16:21:27.  During this period the speed over the ground increased slightly from 0.9 knots to 1 knot falling back to 0.8 knots and then increasing to 0.9 knots.  The ship’s course over the ground was initially south east going towards the east between about 142° to 122° at 16:20:36, but then, at 16:21:27, it moved back to 168° towards the south.  Thus, as the engine manoeuvre to dead slow ahead began, the heading of the bow swung slightly towards the west, while the course over the ground suggested the ship was moving generally south east, going more to the east (or port).  But as the transverse thrust of the ahead movement began, it probably moved the stern initially to starboard somewhat (but not to a great extent).

264               As Prof Renilson said, both he and Mr Kelle agreed that the effect of the propeller at this time could be negligible or major and the force of this could be affected by the rudder direction.  Mr Kelle said that the dead slow ahead engine orders beginning at 16:19:51 “… brought the ship on a sideways to port and straight ahead movement further towards Mordialloc”.  I accept that evidence as being consistent with the detailed record of AIS data transmitted from the ship (which formed the basis of the AIS video).  In other words, the effect of the dead slow ahead engine orders at this time after the slight initial effect of the transverse thrust, was to cause the ship to move sideways to the east (or port) away from the anchor and south just after she had reached the extremity of her last swing to port.

265               Capt Cole said that the ship appeared to be responding to the engine orders to go dead slow ahead at this stage.  He noted that the head was coming round to starboard and moving in a southerly direction towards the pipeline and anchor.  He said that regardless of whether the cable was under tension when the manoeuvre began, the tension was being reduced during its course ahead.  Capt Cole said that during the period of the two engine movements the ship moved about 39 metres ahead.  He emphasised that before undertaking such a manoeuvre and during it, the master would ascertain the direction in which the chain was leading and the amount of weight on it (i.e. whether it appeared to be taut or there was any visible catenary effect).  Capt Dwyer agreed.  The three nautical experts all agreed that such a manoeuvre should not be commenced when the vessel was moving towards the extremity of a yaw.  Rather, the ahead engine movement and associated orders would be put into effect after the extremity of the yaw had been reached and the ship was moving back towards the anchor.

266               The shipowner argued that the two dead slow ahead engine movements at 16:19:51 and 16:20:13 were seamanlike and that there was no navigational evidence to tie the rupture to these engine movements.  Capt Dwyer had said in his first report that the vessel appeared to be windrode between 16:19 and 16:25.  He opined that the master and pilot should have noticed this and concluded that the anchor was still entangled with the pipeline, including during the period after it had ruptured.  Capt Cole did not agree that the behaviour of the vessel in that period would have indicated any conclusion about whether the anchor was engaged with the pipeline.  The shipowner argued that the nautical experts did not know what was occurring below the surface.  It noted that Capt Cole and Capt Dwyer had said that they had no basis within their expertise to connect the rupture with the ahead movements commencing at 16:19:51. 

267               Prof Renilson observed that an ahead engine movement would only reduce the tension on the cable if the cable were leading ahead.  He said that if the cable were leading astern at the time of the ahead movement, the force on it would increase.  But he said, and I accept, that there was no direct evidence of the helm order, any effect of transverse thrust, or the actual direction in which the cable was leading at 16:19:51 when the ahead engine movements were ordered.  Both he and Mr Kelle agreed that during a yaw, the peak load on the anchor and pipeline occurred when the chain became taut and the heading of the vessel began to change.  After that point the tension would lessen.  They also agreed that the actual peak load at the extremity of each yaw was large but could not be accurately calculated.

268               However, they disagreed about the magnitude of those loads.  Prof Renilson and Mr Kelle accepted that those loads were affected by the environmental and dynamic loads.  They also agreed that each of them had estimated a load of an order of magnitude that was within a range of loads, they considered to be possible.

269               Dr Price and Dr Lyngberg, with Mr Gartner, agreed that the anchor was engaged with the pipeline at the Altona fracture point, but the latter two disagreed with Dr Price on its exact position.  When the Altona end was found on the seabed it was located at the position indicated on the sonar image pointing into the crater-like blow hole to its north.  This piece of pipe was an exhibit and all three experts discussed its features as I explain below (see “The Aetiology of the Missing Fractured Pipeline” at [332]-[360]).  I will call it “the fractured exhibit”.

270               Dr Price said that there was insufficient force available to cause the pipeline to fracture, until it actually did.  Thus, the pipeline had withstood the initial impact, the dead slow astern engine movement at 15:46:01 and the yawing of the vessel prior to the fracture.  He then explained that the force applied to the pipeline was sufficient to cause the fracture on the third occasion of optimum geometry (i.e. when the cable was about perpendicular to the bow at the extremity of the yaw to the east).  He said that this force could have been caused by:

·               force increases due to changes in the wind, sea and or vessel orientation;

·               the slow emergence of the pipe from the viscous silt that would reduce the required failure force;

·               engine movements.

271               Dr Price concluded that the fracture had been caused by the pipeline being pulled sideways when the cable was leading from the starboard hawse pipe between perpendicular and 30°.  He concluded that the fracture occurred about 100-200 mm away from the ovalised deformations of the fractured exhibit that had been caused by the initial impact of the anchor’s shank and flukes and their ongoing contact with the pipeline.  He said that the anchor slightly reoriented on the pipe after the initial impact so that the shank made two impressions on the pipe just near the Altona end.

272               In their joint report Dr Price and Mr Gartner agreed that the initial fracture of the pipeline occurred as a single rapid failure and that the precise location of the point on the fractured surface where the failure commenced was of academic interest only.  While a flaw in an area of the surface of the Altona end of the fracture may have been the source for the rapid cracking of the pipeline (this being the point of the academic contention between Dr Price and Mr Gartner), both agreed that if it were, that flaw still needed a high energy force to initiate the fracture.

273               Dr Price opined that the fracture occurred between 16:18 and 16:20 by using a chart that Mr Kelle had prepared based on the AIS data and Mr Kelle’s calculation of the location of the starboard hawse pipe.  Figure 6, which is at [353] below, is an adaption of this chart.  However, that chart was shown to have some errors in relation to its estimate of the distance between the AIS transmitter on the ship’s superstructure and the hawse pipe.  He illustrated angles of the chain leading ahead from the starboard hawse pipe at about 15° at 16:18 and 60° at 16:20, based on his interpretation of Mr Kelle’s chart.  Dr Price said that the minimum force to cause a rupture would occur when the cable was perpendicular to the general direction of the pipe.  He opined that the features on the fractured exhibit, being the Altona end of the pipe, were consistent with the break occurring while the pipe was being pulled sideways.  He considered that the fracture occurred no later than when the chain was at most 30° off perpendicular.  This was because he concluded both ends of the pipe had recoiled on the occurrence of fracture.

274               Mr Gartner considered that the fracture occurred when a new, “extra high force” was applied, additional to the forces operating when the vessel was yawing.  And he accepted that this force could have come from the ship’s engine starting and causing the vessel to move towards the south.

275               The chief officer’s observation of the cable being at about 1 o’clock when he noticed the gas bubbling is consistent with Dr Price’s evidence of the position of the cable when being pulled sideways, as causing the pipeline to fracture.  Although the chief officer said in his evidence that he was occupied with the oil spill and windlass at this point, I find that he noticed the gas bubbles very soon after the fracture.  That is consistent with his observation of the direction of the cable and later observations of the escaping gas becoming plumes and forming a white cloud.  After the fracture, the ship continued moving south south west and, as it did so, the cable would have led more and more towards the stern until the effect of the ahead engine movement’s momentum was spent.  By then, the master would have spoken to all the sections of the crew, including the chief officer, about the gas and the consequent fire hazard.

276               The shipowner also argued that the fracture of the pipeline was inevitable and was not caused by the dead slow ahead engine movements.  However, the ship owner led no evidence of calculations or tests to make good this proposition.  Nor did it lead other evidence to explain why the fracture occurred when it did.  It relied on the nautical experts saying that there was a mere coincidence between the dead slow ahead engine movements and the fracture.  Indeed, there was literally a coincidence, but the cause of the fracture was outside the expertise of the nautical experts.

277               The shipowner also relied on Prof Renilson’s view that the analysis of the cause was too complex and there were too many factors to enable a conclusion to be drawn about the consequence of the engine movements.  But, as he said, he did not know enough about the strength of the pipeline to say whether it would have survived more yaws than the three before the fracture.  That was outside his expertise.  Mr Kelle, did some calculations of the forces involved, but he did not calculate the force created by the engine movement.  His view was that the pipeline had survived the three previous yaws and would be very likely to survive further ones.  Mr Gartner postulated either that there was a weak point on the pipe which triggered the fracture or the engine movement supplied an extra high force to cause it.  The shipowner argued that this established that the pipeline had been “flawed fatally on the initial impact in a way that initiated the final rupture”.  However, both Mr Gartner and Dr Price agreed that the fracture occurred, not progressively, but in a split second when an extra peak force was applied.  I prefer and accept Dr Price’s evidence.  I am not satisfied that it was inevitable that the pipeline would have fractured had the ship continued to yaw and the engine not been put to dead slow ahead.  In any event I am not satisfied that the fracture would have happened in that scenario before the cable could have been let go from the bitter end.

278               I find that the movement sideways, soon after the ship had reached the end of the yaw to port, combined with the power imparted by the engine movement and the chain being at about 1 o’clock caused the pipeline to fracture as Dr Price and Mr Kelle (albeit on slightly differing timing bases) postulated.  At this point the cable was relatively taut.  The significant new force imparted by the engine moving the ship ahead pulled the pipeline sideways soon after the second engine order ahead, and caused the fracture.  I accept Dr Price’s evidence, although on my findings the fracture occurred around 16:20 or just after it while the engine movement ahead was exerting force on the cable which was then leading at no more than 30° (or 1 o’clock), as the chief officer observed.  The engine movements beginning at 16:19:51, aided possibly by helm orders, were significant new forces that had occurred immediately before the fracture.  The yaw had finished and the ship was swinging back toward the northwest.  The cable was leading close to 30° within the range Dr Price had identified as optimum to cause the pipeline to fracture.  Mr Kelle, Gartner and Dr Price all agreed that the engine movements could have caused the fracture.  I am satisfied that they did.

Was the decision to order the engine to go ahead seamanlike?

279               The shipowner contended that regardless of the actual reasons that the pilot and the master had for undertaking the manoeuvre of going ahead commencing at 16:19:51, the decision was seamanlike.  It argued that the manoeuvre could be “a legitimate, if risky, way of escaping the pipeline” depending on the information available to a master in Capt Xu’s position.  The shipowner asserted that the manoeuvre began when the vessel’s head was close to being into the wind and would have relieved tension on the cable, as Capt Cole and Capt Dwyer accepted.  It posited the test as being whether a reasonable master would have done what Capt Xu did.

280               I reject this argument.  First, before ordering the engine ahead, a reasonable master would have ensured that he had accurately plotted the position of the ship in relation to the pipeline.  Secondly, the master needed to know the direction of the cable and the weight on it before ordering such an engine movement.  The information about the direction and weight on the cable was important to enable the master to assess what helm, bow thruster and engine orders were appropriate, as Capt Cole and Capt Dwyer said.  The chief officer said that he was not monitoring the cable at this time and he also said that the master did not ask him for information about the pipeline.  There was no evidence of precisely what information Capt Xu was aware.    

281               Earlier, when speaking with harbour control at 16:11:21 just after he had arrived on the bridge, Capt Murchie said that he had “… no idea of where exactly the cable is leading” and he complained of the language difficulties he was having as being “extreme”.  He gave evidence that he did not receive any such information before he gave his advice to Capt Xu to order dead slow ahead 8 minutes later.  I infer that he did not enquire or seek this information from Capt Xu before he gave that advice.  Capt Murchie should have made those enquiries.  I am satisfied from the recordings and transcripts of his communications with harbour control and my observation of him giving evidence, occasionally in English, that Capt Xu was sufficiently proficient in English to have been able to answer any enquiry about the direction of the cable or the weight on it, had Capt Murchie asked.  Thirdly, if there were any risk that the anchor had fouled the pipeline, then for the reasons I have given, the only reasonable course of action in the circumstances was to let go the cable from the bitter end.

282               The use of the engine was dangerous given that there was a real likelihood that the anchor was fouled on the pipeline.  The order put more tension on the cable which would be dragged by the ship as she moved ahead.  And, as I have found, at the time the manoeuvre was executed, its effect was to apply a further substantial force on the pipeline.  That was the opposite of a seamanlike approach.  If the manoeuvre were to be attempted, some cable should have been released to avoid the very thing that happened, an increase in tension and the concomitant danger of rupturing the pipeline.

283               While some aspects of the manoeuvre, if taken in isolation, could be appropriate, the question is whether, in all the circumstances, ordering the engine to dead slow ahead at 16:19:51 and 16:20:13 was a reasonable course of action.  I am of opinion that it was not reasonable for Capt Xu to have ordered, or Capt Murchie to advise, the engine to be used at this time.  Each had time and opportunity to consider an appropriate course of action.  Over 30 minutes had passed since the anchor had fouled the pipeline.  The circumstances and navigational priorities had changed in that time.  There had been sufficient opportunity for the master and pilot to reflect on an appropriate plan.

284               Instead, Capt Murchie negligently plotted the ship’s and the anchor’s positions.  Capt Xu negligently failed to check those plots before both took a decision that was fraught with risk if the pilot had made a mistake.  In addition, neither of them considered or implemented a decision to let the anchor cable go from the bitter end.  The pipeline ruptured as a result of their want of seamanship on an occasion distinct from and independent of the initial fouling that had occurred earlier at about 15:44 to 15:45:  cf The Schwan [1892] P at 439, 441;  Ballast Trailing (NSWCA, unreported, 29 September 1981);  The Lucillite 33 Lloyds List Rep at 189;  Exxon 869 F 2d at 847-848.  The rupture set in train further events including a loss of gas and the subsequent movement of the vessel astern which bent the Mordialloc end of the pipeline further.

Claims arising from the rupture of the pipeline

285               Mr Vines’ evidence of the cost of repairs to the pipeline was based on a number of critical assumptions.  The most critical was the length of pipeline that would have required replacement on the hypothesis that nothing more had happened respectively after each of the first three occasions pleaded.

286               As I have found, the length of pipeline damaged on the initial fouling that required replacement was substantially that which was in fact replaced.  Mr Vines said 4 lengths of pipe of 42 metres each (i.e. 168 metres) were used in the repair.  However, only 156.575 metres were laid.  The difference may be explicable because one or two of the 42 metre sections were cut down to fit the pipeline sections replaced.  Importantly, Dr Lyngberg said that 51 metres towards Altona from the estimated point of fracture were replaced while 105 metres were replaced on the Mordialloc side.  I infer that this includes the missing section of pipe.  And as the measurements on Figure 2 show, the apparent length of pipe on the Altona section outside the trench was about 29 metres, whereas, on the Mordialloc section that length was 47 metres.

287               If the anchor had not pulled the Mordialloc side back later, then it is likely that about 29 metres would have been outside the trench on that side as on the Altona side.  This suggests that there was a need to replace a greater distance on the Mordialloc side because of the further engagement of the anchor with that side and the greater bending and displacement of it than on the Altona side.  Dr Price estimated that an additional 20 metres had been dragged out of the trench by the vessel later going astern.  Mr Kelle estimated that an additional 10% to 20% more pipe (i.e. between 15 and 30 metres) had to be replaced as a result of that movement.

288               I am not satisfied that the additional force that was applied to the pipeline to cause it to rupture caused any substantive further damage to the pipeline at that time.  The fracture was not itself any additional damage, because it occurred on part of the whole length of pipeline that had been so damaged by the initial impact of the anchor that it required replacement in any event.

289               I am of opinion that, taken by itself, the fracture damage is not capable of giving rise to a claim for loss of, or damage to, property on a distinct occasion within the meaning of Arts 2(a), 6 and 9 of the Convention.  Such a claim would be artificial in the extreme. To allow it would subvert the purposes of the Convention that permit a shipowner to limit his liability.  Once sufficient damage had been done on the initial impact to require the section of pipeline in which the fracture later occurred to be replaced, the subsequent fracture to the pipeline was of no consequence.  No more cost or expense was involved in replacing it merely because, in addition to having been rendered useless as a pipeline to carry the gas by that initial event, the now useless pipeline was fractured.  In this respect, I accept Dr Lyngberg’s evidence that the effect of the initial impact rendered the affected section of pipeline useless for all practical purposes.

290               Moreover, the maritime lien that attached to “APL Sydney” at the instant of the initial fouling would have entitled Esso and BHP to recover the full cost of replacement, leaving aside the possible further damage caused by the astern movement of the ship soon after 16:21:  cf  The Tolten [1946] P at 153.  No new lien would have attached merely in respect of replacing what was fractured later.  I am of opinion that to allow recovery on this artifice, would undermine the evident policy of the Convention to establish “unbreakable” limits of liability.

291               Different considerations, however, apply to the damage done as a result of the subsequent astern movement.  For the reasons I give later (see [322]-[328]; [352]-[360]) I have concluded that this damage occurred as part of the distinct occasion arising on the ahead orders commencing at 16:19:51.

Claim for lost gas and other potential claims

292               There was evidence of a new and distinct claim that arose on the fracture.  Mr Vines estimated that approximately 30 tonnes of ethane gas escaped between when the pipeline fractured and when the supply line could be closed.  He estimated this loss at $6,000 based on a notional value of $200 per tonne.  He said that this value was used because of the commercial sensitivity of the actual prices for which the gas was sold to Qenos and Huntsman.  Mr Vines believed that the actual sales price was greater than his notional one.

293               Neither Qenos nor Huntsman has appeared in these proceedings.  However, they have each made separate claims against the shipowner arising out of the interruption to their gas supplies while the pipeline was not functioning.  Once again, the greater part of these claims arose on the initial fouling that caused the need for repair and consequent loss of supply.  But Qenos and Huntsman may have some distinct claims that arose because of:

·               the sudden cessation of gas supply due to the rupture on 13 December 2008 causing a loss that would not have been incurred if Esso and BHP had been able to give notice of the imminent need to stop that supply later that day after the initial impact;

·               any additional time for the repair work caused by the need to replace more pipe due to the additional distortion of the Mordialloc end of the pipe.

294               On the evidence before me the values of these potential claims arising on the distinct occasion resulting in the fracture appear to be relatively small, however they are new claims distinct from those that arose on the initial fouling.

How long was the anchor chain at the time of the rupture

295               One curious issue that arose concerned the length of the anchor chain during the period between the fouling and rupture of the pipeline.  The master reported two shackles were in the water at 15:48:43, after the bosun applied the hand brake.  After the pipeline ruptured, the ship sailed north under her engine, dragging the anchor.  At about 21:50 the chief officer supervised, in a safe manner, the gas axing of the chain on the forecastle deck.  The chain had a buoy attached.  The master’s statement of facts dated 13 December 2008 recorded that 2.5 shackles were dropped into the water when the gas axing had occurred.

296               Sometime after the pipeline ruptured the chief officer said that the master instructed him to decouple the anchor chain at the third shackle’s kenter link so as to disconnect and jettison the anchor.  The chief officer took the bosun and other crew members to the chain locker.  He said that it took them about 30 minutes to open the chain locker.  Next, the chief officer said that he saw a kenter link, which he identified as linking the second and third shackles on top of the pile of chain in the locker.  The crew tried to decouple this for two to three hours but could not do so.  After this the chief officer supervised the cutting of the cable on deck.  I accept that evidence.

297               However, when the operation to retrieve the anchor and cable from Port Phillip Bay occurred on 10 January 2009, the anchor and only one and one sixth shackles of cable (32.5 metres) were brought on shore.  Mr Kelle had been retained by Esso and BHP’s solicitors to observe the retrieval operation, arranged by the shipowner or its P&I Club.  But when he arrived on the fishing vessel towing the anchor retrieval barge, the anchor had been recovered and was held on the barge partially submerged.  On his earlier visit to the scene on 8 January, he was told that the divers had located the chain and that it was buried about one metre below the seabed.  The weather made it impossible to continue the retrieval operation which was rescheduled to 10 January.  On the latter day, Mr Kelle was told by the divers that they estimated that the chain was about 50 metres long.  They told him that they would not hang the chain on deck but underneath the barge and hence there was no point in Mr Kelle remaining.

298               On 12 January, Mr Kelle arrived at Lacelles Wharf, Geelong shortly before the time the divers had informed him the anchor and chain would be brought ashore.  However, the anchor and chain had already been brought on shore and had a total length of about 32.5 metres.  The shipowner did not call any of the divers or others involved in the retrieval.  I infer that their evidence would not assist its case as to the length of cable.

299               Obviously, the 32.5 metre length of chain brought ashore is significantly different to the 2.5 shackles, or about 69 metres, that the crew believed were dropped into the water when the cable was cut.  A theoretical dispute arose because Mr Kelle prepared his first report using the longer cable length.  He considered that the shorter length did not accord with his calculations of the vessel’s behaviour taken from the AIS data, the evidence from the sonar images of the anchor’s path through the seabed, the plotted position of the ship when the chain was cut and the rate at which the windlass would have heaved in cable before fouling.  In addition, the forces that would have been imparted by a very short cable were many times those of a longer length.  This was in the context that, even after the fouling the windlass retrieved some cable, before it broke and then some cable was let out before the bosun could apply the hand brake.  If all the cable that was ultimately out from the gypsy was 32.5 metres, Prof Renilson and Mr Kelle calculated that the anchor would barely have touched the seabed.

300               This theoretical dispute evaporated when Prof Renilson and Mr Kelle were asked to assume that the marks radiating to the north and towards the east from the Altona (northern) blowhole on the sonar image in Figure 2 were anchor chain scour marks.  These had approximate lengths of about 21 to 31 metres based on the measurements of distance appearing on Figure 2 itself.  Prof Renilson did not know what the markings were but said that if the 21 metre ones were chain marks made while the ship yawed between 15:48 and 16:19 then the chain would have had to be quite a lot longer than the amount brought ashore.  Indeed, he said that it would probably be even longer than Mr Kelle thought.  Mr Kelle saw the markings as confirming his postulation.

301               Dr Lyngberg identified these marks in his written report as most likely to have been caused by the anchor chain.  He said that the sonar image and survey in Figure 2 was consistent with what he would expect to see where an anchor had been dragged along in the seabed and engaged a buried pipeline.  He confirmed that these radiating lines “… look like to be an anchor chain being touched on the bottom”.  Dr Lyngberg had very considerable experience in, among other fields, damage assessment and the investigation of the causes on pipelines damaged by anchors and fishing trawling equipment.  I accept his evidence that those marks are anchor chain marks.

302               It follows that the chain brought ashore with the anchor was not the full length of chain that was cut late on 13 December 2008.  Because the shipowner did not call evidence of the condition of the chain and anchor as they were found in the seabed, I infer that this evidence would not have assisted its case that the 32.5 metres recovered was the entirety of the anchor chain that was cut from the ship:  Jones v Dunkel (1959) 101 CLR 298.  The objective evidence of the anchor chain marks in Figure 2, that appear to be about 21 metres or more in length and the fact that they radiate in the direction of the yaws that could only have occurred after the windlass failed, demonstrate that it was not possible for only 32.5 metres of cable to have been cut from the ship later on 13 December 2008.  I find that there was at least 70 metres of cable cut, and possibly more.

The events after the pipeline ruptured – The fourth alleged distinct occasion

303               Once he saw the gas bubbling to the surface, Capt Xu ordered stop engine at 16:21:27.  No subsequent engine order was given until 16:27:59 when Capt Murchie advised the master to put the engine to slow astern.  It is important to understand what happened both above and below the sea surface in the six and a half minutes between the two engine orders.  When APL Sydney went astern the anchor pulled the Mordiallic side of the pipe back nearly to a right angle and broke off some 6 or 7 metres of pipe including the portion that had connected to the fractured Altona end.  How did the anchor do this?  I will discuss the events on the surface first, before turning to the submarine ones.

304               Capt Murchie radioed harbour control at 16:22:11 and informed them that the pipeline had burst.  He said, and I find, that he ordered the bow thruster be used to keep the bow clear of the gas by pushing the bow to port.  Capt Xu could not remember if this occurred.  As he explained at that point, he had a great deal to do, including ensuring that the ship and crew were safe from the risk of fire.  He broadcast a message to the crew to stop any cooking and smoking.  He wanted to ensure the safety of the crew. He said he was also concerned not to pollute the environment.  He was, understandably, very concerned about the risk of an explosion at this time.  He also reported to his principals.  Capt Murchie also said that he felt that the ship was in a great deal of danger.  The gas was being blown by the wind into the accommodation area.

305               In the meantime, the ship continued moving ahead, bow into the wind toward the south west as appears in the AIS video.  Mr Kelle said that by about 16:24 the bow was above the pipeline.  He observed that at 16:24:18 the effect and momentum of the ahead manoeuvre had ceased.  The vessel then began to drift astern to the north north east, bow almost into the wind at a heading of 218°.  Mr Kelle estimated that the drift astern covered about 67 metres until 16:27:59.  He opined that the chain would have been slack during the period between 16:24 and 16:28 because of the proximity of the ship to the area where the anchor was located.  Some idea of the movement of the ship during this period can be seen in Figure 6 at [353] below.

306               Shortly before 16:28 Capt Murchie advised Capt Xu that the engine should be put to slow astern to get the ship away from the gas.  Both of them were still worried about the safety of the crew and ship.  Capt Xu agreed to this manoeuvre because he was not sure that all naked flames on board had been extinguished and was fearful of a very strong possibility of an explosion.  He was conscious that the accommodation and bridge were towards the stern of the ship and thus farther away from the source of the gas.  He considered that the pilot’s advice to go astern was reasonable and did not involve the risk of taking the accommodation area (with the potential of it containing a naked flame) through the gas plumes which were substantial.  He said, and I accept, that initially the gas had been just bubbling but it had become “like water plumes … big … and it wasn’t just a few little bubbles”.  The gas plumes were on the starboard side.  The pilot was also concerned that the gas was being blown directly at the accommodation section of the ship.  This, of course, included the bridge.  Like the master, he rejected the suggestion put in cross-examination that he should have ordered the vessel’s engine ahead as a more expeditious way of escaping the gas.  As he said:

“I didn’t consider that it was a viable thing to do.  If you were going to move the vessel ahead, you would be driving the engine and the accommodation through the gas cloud.”

307               The engine remained at dead slow astern from 16:27:59 until it was stopped at 16:33:03.  During that period the ship moved astern towards the north east.  The bow began to rotate towards the north from a heading of about 225° to 293°  and the ship came round towards being beam onto the wind, with the anchor being dragged across the bow.

308               Before they undertook the slow astern manoeuvre, both the master and the pilot were aware that there was a possibility that the anchor might re-engage with the pipeline.  However, they both said that rather than continuing ahead, through the gas cloud, they thought that it was safer for the crew and ship to go astern, away from the gas, even though there was a risk of damage to the pipeline.  Capt Murchie said that the pipeline was of secondary importance at that time.  When speaking to the harbour control at 16:31:23 Capt Murchie graphically described the position the ship was in:

“Not good.  There’s a f…ing great cloud of gas here. … The pipe is already damaged, so what I’ve done I’m just backing away from it just to get away from the gas cloud because I think it’s pretty dangerous … .  So whatever damage is done, I just have to get out of the way.”

309               I infer that when the engine was stopped at 16:33:03 the ship was clear of the gas and both the master and pilot considered that they were safe, despite the fact the vessel was downwind but to leeward of the gas.

310               The nautical experts agreed that when the astern order was given those on the bridge were right to be apprehensive about the proximity of highly flammable gas.  The nautical experts disagreed about the appropriateness of the astern manoeuvre and about whether it was possible for the anchor to have moved from its position immediately after the fracture of the pipeline to a position further to the south or south west of the Mordialloc end of the pipeline.

311               Capt Third noted that as a result of going astern the vessel successfully achieved a safe distance.  He considered that the astern manoeuvre was seamanlike and that going ahead after the pipeline fractured would have been dangerous.

312               In his first report, Capt Cole opined that it was possible for the anchor to have re-engaged with the Mordialloc end of the pipeline after moving with the ship to its south.  He relied on the AIS data and Mr Kelle’s chart showing the starboard hawse pipe’s location.  As I have noted above, that chart was shown not to be accurate during the hearing because it had not kept an equal distance consistently between the plotted positions of the AIS data transmitter in the accommodation section of APL Sydney and the starboard hawse pipe.  In that report, Capt Cole made the assertion that if the anchor had fouled the Mordialloc section of pipe, then the master and pilot had only one option to prevent further damage.  That was to hold the ship statically in position putting no weight on the anchor while the chain was broken either by disconnecting it at a kenter shackle link or cutting it using a gas axe.  That view was untenable.  It ignored the threat that the escaping gas posed to the crew and ship.  It elevated preservation of a fractured pipeline to a priority over the safety of the crew and ship.

313               Capt Cole changed that opinion in his second report where he now said that the only option at this time was to move the ship away from the down wind area of the escaping gas.  He said this could be achieved using the bow thruster and rudder to move the ship to port and putting the engine to half ahead.

314               Capt Dwyer said in his first report that between 16:19 and 16:25 the vessel was still windrode and its behaviour should have indicated to the master and pilot that the anchor was still entangled with the pipeline.  He also relied on Mr Kelle’s inaccurate chart saying:

“It is apparent from [Mr Kelle’s] reconstruction that the vessel having moved ahead at 1620 and severed the pipeline it was highly likely in the circumstances that it would re-engage with the pipeline if further movements were made.  Once reconnected with the pipeline, the astern movement at 1627 dragged the pipeline further North and East.

The behaviour of the vessel with such a short amount of anchor cable out was in keeping with a vessel wind rode and partially anchored to something.”  (emphasis added)

315               Capt Dwyer did not explain how the anchor could “re-engage with the pipeline” if the ship were still windrode after the pipeline had ruptured and gas appeared at the surface by 16:22.  He said that the decision to move the ship astern had been shown in Dr Price’s first report to have further damaged the pipeline.  Capt Dwyer’s first report offered the equally untenable view that the proper course of action for the pilot, after the gas began escaping, was to order a tug to hold the ship in position until the cable could be let go or parted.  He too ignored the question of safety.

316               Capt Dwyer reaffirmed those views in his second report but also asserted that the priority, after the pipeline had been ruptured, was still to jettison the anchor.  He criticised, as unseamanlike, the decision to go astern because “… it would have moved the aspect of the gas erupting on the starboard bow, to an aspect right ahead and the vessel immediately down wind”.

317               The written criticisms by each of Capt Cole and Capt Dwyer of the master and pilot’s conduct of the ship after the stop order at 16:21:27 were not reasonable.  First, each expert had had the benefit of considerable time in which to consider, and in his second report reconsider, what his opinion should have been.  Secondly, each of them in their first reports ignored the most obvious and pressing consideration affecting a seamanlike response at and after 16:22;  namely, the escape of gas which was increasing.  Thirdly, each initially suggested that the ship be held in position in a situation that was so obviously dangerous that he could not have had any regard to the actual circumstances confronting Capt Xu and Capt Murchie.  It is one thing to opine that they should never have acted so as to bring about the rupture of the pipeline.  But, after they had arrived at that point their available seamanlike options had to be assessed having regard to the self-evidently dangerous position they, the crew and APL Sydney occupied near the escaping gas.  These considered, but unwarranted, criticisms made by Capt Cole and Capt Dwyer have made me approach their other opinions critical of the master or pilot with caution.  I have tended to prefer the more sanguine and reasonable view of Capt Third, especially in relation to the conduct of the master and pilot after the rupture.  I accept his evidence that:

“It is unreasonable to expect Masters and Pilots to react in such emergency circumstances with absolute precision and to make the correct decisions immediately.  Due allowance must be made, variously, for the processes in which danger and particular difficulties are recognised, for a situation to be appraised, for the possible actions and consequences to be evaluated and for the appropriate decisions to be taken.”

318               And, during his concurrent evidence, Capt Dwyer ultimately was forced to accept that by 16:35 the astern manoeuvre, as executed, had achieved exactly one objective that all three nautical experts had acknowledged was appropriate;  that was to move APL Sydney significantly astern to port (south east), and to put as much distance between her and the source of the gas as possible.  In their concurrent evidence, all three nautical experts accepted that the actual astern manoeuvre could not be criticised as unseamanlike.  I accept that evidence.  The nautical experts also accepted that another, but not the only, option for the master and pilot was to put the helm hard to starboard, order the engine to half ahead and the bow thruster to move to port, causing the ship to turn to port and go to the south east away from the gas.

319               The shipowner used this evidentiary position as a basis to argue that there was no distinct occasion or fourth event after the pipeline ruptured.  In opposition, Esso and BHP argued that the master and pilot had both the time and opportunity to avoid the anchor re-engaging with the pipeline.

320               However, Capt Cole and Capt Dwyer next asserted in concurrent evidence that the master and pilot should not have waited for about 6 minutes from first noticing the escaping gas before manoeuvring away.  Each of Capt Cole and Capt Dwyer now asserted that he would have undertaken the ahead manoeuvre that I have just described.  I do not accept that evidence.  Neither of them suggested that immediacy of action or any action like it in their two written reports.  Neither expert gave any written analysis about such a manoeuvre.  It simply did not occur to them within 6 minutes, and only did so close to the hearing.

321               As I have set out above, each of them wrote the contrary initially, namely that they would have held the ship in position so as to avoid damaging the pipeline.  By the time of preparing their joint report with Capt Third they recognised that that contention was untenable.  In the concurrent evidence Capt Dwyer said of the 6 minutes “I don’t believe waiting was an option”.  He now said that he considered that waiting with the gas in close proximity to the vessel was dangerous.  I accept that the gas posed a significant present danger and that the ahead manoeuvre was a reasonable alternative.  But, I do not accept Capt Cole’s or Capt Dwyer’s most recent evidence that a reasonable master had to undertake an ahead movement immediately or within less than 6 minutes of first seeing the gas escaping.

322               I prefer Capt Third’s evidence as a more balanced and reasonable approach.  First, he said that while he accepted the alternative of the manoeuvre ahead, the astern manoeuvre was a logical one.  He said that the master and pilot had to work out a manoeuvre in the context of all that was then happening.  He said, and I accept, that it was “… better to have a considered manoeuvre than to attempt something off the cuff, as it were, which might work but you’re not certain of”.  Capt Third made the point that the master and pilot would have been aware that any manoeuvre had to pull the anchor and that it would be better to pull it in line with the ship.  He considered that using the powerful main engine to bring the anchor into line with the bow by moving astern was likely to be quicker than using the less powerful bow thruster to pull it to port.  However, Capt Third accepted that if the master or pilot had a good sound reason to believe that the pipeline was then astern of the position, going astern was not appropriate.  He also acknowledged that the risk of fouling the pipeline again would be a consideration in deciding on the appropriate manoeuvre.

323               I infer that Capt Third was drawing a distinction between giving consideration, as Capt Murchie had, to a situation in which there was a risk that the anchor had moved to a position where it might re-engage with the pipeline if the ship went astern, and one where that risk had matured into a good sound reason to believe that the anchor had so moved.  I find that the risk confronting the master and the pilot could only be assessed in circumstances that it was uncertain where the anchor was in relation to the pipeline.  Throughout the six and a half minutes the ship was moving and changing its position in relation to the anchor and the two ends of the fractured pipeline.  I find that because of that uncertainty, and the unavailability of any means of ascertaining the exact location of the anchor in relation to the pipeline, Capt Murchie and Capt Xu took a calculated and seamanlike risk of adopting a reasonable manoeuvre to escape the gas.  Indeed, even with all the resources of the parties devoted to this issue for the trial, the exact location of the anchor in the period immediately after the initial fracture was a mystery.  The master and pilot were in the middle of a gale with gas bubbling to the surface, a broken windlass and the need to protect the lives of the crew and the ship.  They could not be expected to calculate or ascertain the anchor’s position with precision at any point of time during this period.

Was the astern movement a distinct occasion?

324               Esso and BHP argued that the evidence showed that the anchor’s re-engagement with the pipeline could have been avoided by an appropriate manoeuvre ahead.  They contended that this possibility was sufficient to establish that the master and pilot had time and opportunity to avoid that occurring and, because they had not avoided it, a fourth distinct occasion had occurred for the purposes of the Convention.

325               I reject that argument.  I am not satisfied that in the circumstances Capt Xu’s and Capt Murchie’s awareness of the mere existence of a risk that the anchor could re-engage with the pipeline allowed them time or opportunity to avoid that risk.  The uncertainty of where the anchor was as the ship moved initially ahead and then astern (including the period before the engine order astern) coupled with the proximity of the escaping gas and the prevailing conditions required a considered but relatively quick response.

326               I am of opinion that this situation, and the need for a response, was inseverably connected to the events that had caused the pipeline to rupture.  Once that fracture had occurred, the previous, and now obviously wrong, understanding of the pilot as to the anchor’s location had to be reassessed.  In addition, the master had to do what he could to ensure that the risk of a naked flame was averted.  Both he and the pilot had to work out how to take the ship away from danger in the least risky manner.  Because the location of the anchor and the pipeline was uncertain and changing as the ship moved – it could be north or south of the pipeline – neither an ahead or astern movement at a particular point of time was necessarily going to avoid the risk of the anchor re-engaging.  Indeed, had an ahead movement been made, soon after the rupture I am far from certain that the anchor would have avoided re-engaging the Mordialloc end of the pipeline as the ship headed south east.  This issue was never considered because Capt Cole and Capt Dwyer only first raised it very late, in a sentence in the joint experts report.

327               The question of what actually happened to the anchor and where it moved on 13 December 2008 is difficult enough with at least some actual evidence of the result and expert analysis.  There was no analysis to show that at a particular time, however shortly after the gas was noticed, an ahead order would certainly have avoided the anchor re-engaging with the pipeline.

328               I am not satisfied that Esso and BHP have established that the actual astern manoeuvre was unseamanlike or that the master and the pilot had the opportunity to avoid the risk of the anchor re-engaging with the pipeline after it had first fractured.  Indeed, I am satisfied that the astern manoeuvre, as executed, was a seamanlike and appropriate response in all the circumstances.

The subsequent damage to the Mordialloc end Of The pipeline

329               What happened underwater immediately after the pipeline ruptured is an intriguing mystery.  Like the old genre of murder mystery novels, in which the usual theme involved the butler as the murderer, some critical facts are known;  but there are gaps.  By the time the ship was safe the anchor had created the damage that left a 6-7 metre piece of pipe missing and it caused a vertical ovalisation of the Mordialloc end of the pipeline that emitted ethane gas under pressure.  But how the anchor behaved in the 10 or so minutes between the initial rupture and the Mordialloc end fracture, is a mystery.  It left no tell tale sign of its movement between the Altona and Mordialloc blow holes in the sonar image of the seabed.  Yet the anchor, itself, weighed 8.325 tonnes and the anchor chain, of course, added to that weight.  How did it get from the first point to the second point on the sonar image in Figure 2?  I have included a smaller copy below with some annotations for ease of reference for the following description.

Figure 3

330               At the south of the Altona blow hole, just near the superimposed marking of 11 metres on Figures 2 and 3, is the drag mark of the anchor showing its path as it headed north east to the point of it fouling the pipeline about 11 metres south west of the Altona blowhole.  Then, at the Mordialloc end blow hole shown on Figures 2 and 3 the drag line resumed its north easterly route.  But, as I have said, there is no mark on the sonar image between the two blow holes that any witness suggested was made by the anchor or its chain.  The missing piece of pipeline may have been able to tell the story, but in the discovery of the chain of causation in this proceeding, it is very much the missing link.  Both sides called experts who hypothesised, or speculated, about how the anchor behaved between the Altona and Mordialloc fractures.

331               A further integer to this perplexing problem was the fact that the crew asserted that about 2.5 shackles (or about 69 metres) of anchor chain was between the hawse pipe and the anchor (as I have found above), yet the anchor, when recovered, had only 32.5 metres (or a little under 1.5 shackles) of chain attached to it.  Despite the very considerable resources at the disposal of the parties, these significant gaps or areas of doubt in the evidence must be filled by my making findings of fact.

The aetiology of the missing fractured pipeline

332               How did the missing piece of pipeline come to fracture twice, once at the Altona end and the second time at the Mordialloc end?   I will start with the Altona fracture because all the experts agreed that it occurred first. I have reproduced in Figure 4 below a photograph of the Altona end viewed from the Mordialloc side on which Dr Price drew a representation of the anchor shank and fluke to assist in the following description.

 

 

 

 

 

 

 

Figure 4

333               Dr Price gave concurrent evidence separately with, first, Dr Lyngberg during which Mr Gartner was present in court, and, secondly, with Mr Gartner.  All the experts agreed that the fractured exhibit had been ovalised in the horizontal plane and drawn upwards from 3 metres below the seabed to some distance into the water above and, horizontally, 11 metres from its original trench towards the north east.  This had happened between about 15:44 or 15:45 and the time of rupture around 16:21.

334               In lay terms, the fractured exhibit appeared physically as follows.  On the underside of the Altona end, on the south western edge (where the flukes and shank met), there are scrape marks running north east between 45º and 60º from that ledge back across the pipe (suggesting that the flukes left the pipe as the anchor pulled with the chain and shank pointing north.  That is consistent with the direction of the indentation on the top of the ovalised pipe closest to the fracture in the direction of Altona.  The scrape marks extend up past a small piece of tar coating that has been left on the underside of the pipe and move towards the top of the oxidised part of the pipe.  There is an area of metal to metal contact at the base, near the commencement of the fracture on the south west side.  This indicates that the concrete weight coat covering the pipeline was missing here when the anchor gripped it pulling it towards the ship.  Further to the north-west on the south western edge of the pipe there is an area of indentation and oxidation where (the experts agreed) an anchor fluke was engaged.  There is a pronounced indentation at the north western edge furthest away from the fracture and a second, slightly less pronounced, indentation adjacent to it closer to the south east.

335               Looking through the pipe towards the fracture, when it is sitting with the top side up, there is a pronounced indentation about 30cm in from the fracture.  This is a prominent marking that appears to match where the shank of the anchor impressed itself.  It is further along towards the fracture than the fluke indentation on the underside.  This suggests that (looking from the vessel to the top of the anchor shaft) the starboard fluke was compressing the pipe from below while the shank created the indentation above.  That means that the port fluke was on the Mordialloc side because there is no other place on the recovered Altona section of pipe where both flukes left indentations.

336               As one approaches the fracture site, the angle of indentation appears to soften and the upper lip of the pipe abutting the fracture appears higher than the most compressed part of the pipe further along to the north west.  The area of compression where the fluke and anchor shaft engaged the fractured exhibit is very pronounced down towards the south west.  Thus, the pipe appeared to have been squeezed at an angle of about 35º between where the shank and flukes met on the far (south western) side of the pipe further away from the ship.

337               At the end of the fractured exhibit closest to Altona, the pipe is clearly ovalised, appearing to be squashed from above.  The ovalisation becomes more compressed and distorted as one progresses along towards the point of fracture.  The shank appears to have moved across the upper part of the pipe towards the point of fracture and again impressed itself quite deeply with a further depression of the pipe about 20 centimetres from the edge of the fracture zone.  Dr Price said during his concurrent evidence with Mr Gartner that this was consistent with the shank and flukes making a small rotation but did not indicate any longitudinal movement along the pipe.  At the top on north east, there is an area of metal on metal contact which appears to run across to the actual fracture itself for about 20 centimetres.  The whole of the 40 centimetres section that is oxidised has been almost totally cleared of the signs of tar coating on the upper side, in contrast to the balance of the fractured exhibit to the north west which retains most of its tar coating.

338               Looking through the fractured exhibit from its Altona facing end, one can also see the point of first impact, where the shank of the anchor first arrested the north easterly drift of the ship, as Dr Price pointed out in evidence.  This impact occurred when the vessel, and so the anchor, was travelling at about 1 metre per second.  The impact was very powerful.  It caused the pipeline to ovalise, not in a vertical plane, as it would otherwise, but in a horizontal plane.  The pipeline was gripped on its south west side between the anchor’s shank on top and flukes below and pulled out of its trench towards the north east.  The normal process of ovalisation when a pipe is struck is simply demonstrated by rolling up a piece of paper and striking it in the middle, on the face opposite the direction in which the striking object is heading.  The paper, like the pipe, kinks vertically in the middle around the impact point.  The fractured exhibit could not react in this way because of the force to which it was subjected.  The shank made an impression rising up at an angle of about 35º from where it met the flukes and pipe on its south west face.  This is the point of upper damage on the fractured exhibit furtherest from the site of the fracture (about 30 centimetres).  Underneath and further away from the fracture is a flatter but also distinct impression caused by the starboard (or right or north western) facing fluke (if one is oriented looking toward the anchor from the vessel).  The surface of the pipeline was not gripped by the anchor further to the north-west on the recovered sections of pipeline.

339               Dr Price’s view was that the small amount of metal to metal contact on the fractured exhibit indicated that the weight coat or pieces of it had been held in place by the shank and flukes while they remained engaged over the pipeline.  This protected the surface of the fractured exhibit from the metal to metal scraping that would have occurred as the anchor partly rotated with the yaws of the ship.  Mr Gartner considered that all the gouge marks from metal to metal contact would have been on the missing section of pipe.  I reject Mr Gartner’s view.  First, it did not explain how the actual state of the fractured exhibit could have been caused, especially the significant ovalisation of the fractured exhibit and obvious points of compression by the shank and flukes.  Secondly, Mr Gartner’s theory did not explain how the anchor having had its first impact on the point of the fractured exhibit identified by Dr Price, moved further to the south east without leaving gouging marks on the exposed metal.  Thirdly, as Dr Price noted, no calculations had been put forward to explain how Mr Gartner’s postulated movement could have occurred.  The very gouging that Mr Gartner asserted was missing because it was on the fractured exhibit, was necessary to support his theory that the anchor, during the yaws, moved south east and generally, but not completely, off the fractured exhibit.

340               When the pipe fractured, each end recoiled away from the other in a partly elastic response to the release of tension.  Dr Price used classical stress wave analysis to calculate  that at the moment of rupture, the initial velocity of the longitudinal recoil of each end of the severed pipe was 54.6 kph (kilometres per hour).  This was a sudden and powerful release of energy.  He said that it was hard to estimate the distance of the recoil (which released the elastic strain on the pipe that the vessel was exerting through the pull of the attached anchor).  This was because the stress wave was both reflected (by breaks in the concrete weight coat) and absorbed (from coatings on the pipe).  He considered that the effect of each of those factors was poorly known.  He said that there was no indication that the sideways movement of the pipe was significantly affected by the escape of the gas it carried under pressure.  Overall, Dr Price estimated that the gap that opened up between the two pieces of pipe was approximately 2 metres.  However, he said this recoil distance was very difficult to estimate and it could have been greater.  The other experts generally accepted this result and Dr Price’s methodology.  Although Mr Kelle also calculated a slightly lesser velocity of separation, nothing turns on that and I accept Dr Price’s analysis.

341               During the concurrent evidence of Drs Lyngberg and Price, the former explained his theory of how the anchor slid along the pipeline towards the south-east of the point of fracture.  They agreed that the key difference between their opinions was whether the anchor remained engaged (as Dr Lyngberg contended) or disengaged (as Dr Price argued) from the pipeline at the time of the initial rupture.  Dr Lyngberg based his analysis on his understanding of the sonar image.  He used his mechanical engineering expertise, acknowledging that Dr Price had both that and expertise in metallurgy.  They agreed that on the initial impact the pipeline’s uniform stiffness in all directions was changed.  The action of the shank and flukes flattened the pipe at the area of contact increasing its stiffness in the direction of the upwards and north-easterly pull caused by the vessel and significantly decreased its stiffness in the vertical direction.

342               Dr Price explained, based on Mr Kelle’s calculations, that the loads applied to the pipeline by the anchor varied between about 10 and 100 tonnes, depending on the position of the vessel as it yawed and reacted to the environmental forces.  Dr Lyngberg, Dr Price and Mr Kelle agreed that a significant force, somewhere around 100 tonnes, was applied to the pipeline at the time of rupture although none of them regarded the exact amount as making a relevant difference.  So, after initially hitting the pipe, in a basically vertical plane, the anchor squashed and flattened it as it pulled it sideways out of its trench from about three metres below the seabed.  As it was dragged upwards towards the vessel the pipe was in very high tension.  It exhibited some elasticity (i.e. its ability to return to its previous shape;  in this case when the fracture occurred the ruptured pipe ends moved longitudinally back whence they had originated but to a smaller degree) and some plasticity (i.e. its inability to return to its previous shape because the forces that had been applied to it had caused it to deform permanently) – the movement back being the elasticity;  the extent to which the pipe did not resume its previous shape before impact, being its plasticity.  The pipe ultimately buckled and failed while still gripped by the anchor.  The experts differed as to the anchor’s position as shown in the following illustration agreed between Dr Price and Mr Gartner.

Figure 5

 

 

The Altona end fracture mechanism:  Altona is on the left hand side.  Mr Gartner’s initials (BG) and Dr Price’s (JP) indicate where each says the anchor gripped the pipe at the time of the initial fracture.

 

343               Dr Lynberg’s theory for the anchor’s position at the time of the initial fracture involved it sliding or working its way from the Altona side of the fracture to crimp in a “W” shape on the missing section of pipe immediately to the south east of the fracture;  i.e. for him, the fracture occurred at the point marked “BG” in Figure 5.  That is once the anchor moved to this position, it wedged the missing pipeline on the Mordiallic side of the initial fracture, by indenting portions of the pipe to correspond with the converging upwards and downwards actions of the shank bearing down on the pipe as the flukes pushed upwards.  This resulted in indentations like a “W”, although perhaps not as pronounced.  Dr Price drew a rough sketch of what he understood Dr Lyngberg contended was the result of the indentation action by the anchor on the Mordiallic side of the initial fracture showing a “W” shape in the pipe as it was crushed between the anchor’s parts.  First Dr Lyngberg, and later Mr Gartner agreed that the sketch correctly conveyed the concept behind their explanations of the position of the anchor at the time of and immediately after the initial fracture.

344               In his first report, Mr Gartner described the Altona and Mordialloc ends of the recovered pipe as though they had separated from one another as part of the same event.  Mr Gartner realised, after his first report, that the two fractured ends of pipe at the Altona and Mordialloc blow holes were unrelated and that over 4.6 metres of pipe was missing, that had originally been between the two existing ends of Altona and Mordialloc.  Nor had he even considered in his first report whether there would have been any longitudinal recoil of either end of the pipe after the initial break.  And Mr Gartner did not attempt such a calculation when he saw Dr Price’s approach, even though he accepted that the effect of any longitudinal recoil was critical to his theory that the anchor retained a grip on the Mordialloc section of the pipe after the initial rupture.

345               I found Mr Gartner’s evidence to be less persuasive and less well reasoned than Dr Price’s.  First, I consider that it is telling that Mr Gartner initially failed to identify that the mechanisms which produced the two fractures at the Altona and Mordialloc ends could not have been the same.  One was an ovalised horizontal fracture and the other an ovalised vertical fracture.  While Mr Gartner was not mechanical engineer, he was a metallurgist.  The physical deformation evident on each piece of existing pipe should have alerted him to the difference in the origin of the two very different fractures he was examining.  Secondly, after he realised that his initial report (which he described in evidence as “very preliminary”, relying on his assertion in it that he needed to get microscopic and chemical analyses of the surfaces of the two fractures) he developed a second theory that involved the anchor remaining engaged with the pipe after the initial fracture.  In order for that theory to work the anchor had to be sufficiently engaged with the pipe to withstand the forces of the initial rupture.  Thus, having worked its way across the site of the rupture towards the south west, the anchor had to grip the pipeline so as to create the “W” crimp.  When giving concurrent evidence, Mr Gartner advanced, for the first time, his explanation as to how, while the anchor was so engaged, the Mordialloc end came to fracture.  He asserted that the anchor pulled the Mordialloc side of the ruptured pipe initially towards the south east in the general direction of the ship.  That was a clockwise movement.  But in order to create the vertical ovalisation and fracture of the Mordialloc end of the pipe that was recovered, an anticlockwise movement of the missing pipe had to occur.  In addition, Mr Gartner maintained that this anticlockwise movement occurred with the anchor remaining wedged in the “W” crimp some four to five metres away from the Mordialloc end fracture.

346               I accept Dr Price’s opinion that Mr Gartner’s suggested mechanism is impossible.  First, the postulated mechanism confused, and did not account for, the two necessary forces of movement, clockwise and anticlockwise T.  Secondly, it was contrary to the joint view of Dr Lyngberg and Dr Price that the anchor had to be very close to the Mordialloc end for that second fracture to occur.  As Dr Price pointed out, if Mr Gartner’s latest theory were correct, as soon as the anticlockwise movement occurred the pipeline would snap, not where it did four or five metres away from the anchor, but next to the anchor, as he and Dr Lyngberg agreed.  I had the impression that Mr Gartner was arguing from a conclusion, rather than trying to explore what the conclusion, based on evaluation, ought to have been.

347               Dr Lyngberg’s theory was equally unsatisfactory.  His involved the proposition that shortly after the rupture, at sometime between about 16:22 and 16:24, the anchor managed to move slightly to the south east (towards Mordialloc) while still remaining engaged with the pipe but leaving itself enough room to be able to slide out of the “W” crimp and proceed the next four to five metres along the pipe to be sufficiently adjacent to the Mordialloc end fracture.  He then said that the second fracture occurred when the anchor was dragged upwards in a shear-like fashion as he had understood Mr Gartner’s written report to describe.

348               Again, I accept Dr Price’s evidence that the strength necessary for the anchor to be engaged with the missing piece of pipe in the “W” crimp was inconsistent with its suggested ability to move south east by sliding along the pipe.  The latter movement does not seem logical.  Dr Price thought it was not possible that the anchor could have moved south east a very small distance from the Altona end fracture while remaining engaged with the pipe, so as to disengage itself from the “W” crimp yet be able to slide along the pipe to be adjacent to the second fracture site.

349               The weights of the anchor and the pipe would be likely to have caused both to fall onto the sea bed and into the Altona blow hole area.  The subtlety of movement of the anchor necessary to give effect to Dr Lyngberg’s theory however is, as he recognised, important.  He said that there was no evidence that the anchor had fallen onto the seabed after the initial fracture until it left the Mordialloc end fracture showing its drag mark, as recorded on the sonar image in Figures 2 and 3 to the east.  So Dr Lyngberg postulated that the anchor moved out of the “W” crimp while it was somehow keeping the pipe in the water above the seabed.  Like Dr Price, I have difficulty envisaging how an 8.325 tonne anchor, its heavy chain, and the fractured pipe all remained above seabed level while the anchor moved back, however many centimetres it needed to, in a south westerly direction (so that it loosened its grip) then disengaged itself from the constraint of the “W” crimp, yet remained in contact with the pipeline and was able to slide about 6 metres along it to the south east.

350               This explanation suggested that the anchor had the dexterity of a ballerina performing a pirouette.  I cannot accept that it is a logical explanation.  Moreover, neither Dr Lyngberg nor Mr Gartner  had performed any calculations to see whether their theories could be justified.  While I accept that there were a large number of assumptions that would be necessary in any such calculations and that a great deal was unknown, given the extraordinary nature of the mechanisms that were being propounded, some test based on assumptions should have been performed.  It was reasonable for Dr Lyngberg to say that the value of such a calculation may not have been great.  However, he had observed that he would have liked to have seen more detail and calculation for some of Mr Kelle’s theories, but was not being critical of them.  Despite this, Dr Lyngberg himself had not performed any scientific analysis to support the dexterous movements he asserted the anchor made along the pipeline.

351               The Altona blow hole is quite wide.  Dr Price estimated that the two ends of the pipe recoiled after the initial rupture leaving a space of about two metres, although in his oral evidence he suggested that it was two metres either side.  I accept his evidence that the precise distance of the recoil was very difficult to estimate.  I find that once the rupture occurred, the two ends of the pipeline separated at high speed leaving sufficient space for the 360 millimetre wide shank  to fall between them.  The base of the anchor with the flukes was nearly 3 metres wide, but was below the pipe it had been gripping from the south west side (away from the ship).  When the anchor disengaged at the Altona blowhole it is likely to have fallen somewhere in the middle of the blow hole immediately to the north east of its prominent drag mark.  The vessel headed south west following the ahead movements commencing at 16:19:51, it is likely that the catenary of the cable created one of the anchor chain marks heading towards the east.

352               I accept as more logical, and better reasoned, Dr Price’s analysis of how the initial fracture occurred.  By reference to the diagram in Figure 5, the Altona end of the pipe was the longer piece in the anchor’s grip immediately before the fracture.  I am satisfied by Dr Price’s evidence that the Mordialloc end moved far enough away at the moment of fracture to disengage entirely from the anchor, as did the Altona end.  Once the pipe fractured and its two ends separated at 54.6 kph, the anchor and chain would have dropped to the seabed at about the point below where the fracture had occurred.  I accept Dr Price’s explanation that the chain could not recoil towards the ship and that the anchor would not move far once the pipe separated from where it had gripped the pipeline before the fracture.

353               Capt Third marked up Mr Kelle’s inaccurate plot of the starboard hawse pipe and AIS transmitter positions of APL Sydney at relevant times in Figure 6 below.

 

 

 

 

 

 

 

Figure 6

Extract of Starboard Hawse Pipe Movement Plot

 

354               The yellow line represents the approximate (and possibly inaccurate) position of the ship’s starboard hawse pipe after the pipeline ruptured.  However, the yellow line is useful to enable a rough visualisation of what happened to the point on the ship from since the cable led out the anchor as the vessel moved south west, even though it is not a precise or accurate representation of the exact path.  The cable led from the hawse pipe to the Altona blowhole as indicated on the red dotted line.

355               As the ship moved south west and away from the blowhole, the chain would have been taut.  The further south west the ship moved the more likely that the anchor was dragged below (i.e. south west of) the Mordialloc end in the disturbed area of seabed on Figures 2 and 3 that appears below the blowhole under the superimposed marking of 11 metres.  The ship began moving north east before the engine movement of slow astern was ordered at 16:27:59.  At some point, once the ship’s movement had changed to head towards the north east, either before or after the slow astern order, Mr Kelle and Dr Price considered, and I accept, that it is likely that the anchor dragged north east across the Altona blow hole and re-engaged there with the Mordialloc end of the pipe.  It may have done so after re-passing the trench line about four or five metres away from the fracture.

356               The water and the seabed were very disturbed during these events.  Both the Altona and Mordialloc ends of the pipe initially were emitting gas into the Altona blow hole until the anchor re-engaged with the pipe and moved the Mordialloc side away.  Thus, evidence of where the anchor had fallen into the Altona blow hole at this time is likely to have been obscured by the force of the gas disturbing the seabed in that area.  Later, after the Mordialloc end, held above the seabed, had retreated with the astern movement of the vessel, the Altona end continued to emit the gas that it still contained, but with decreasing pressure.  That emission of gas on the Altona side is likely to have continued for some time but under less pressure than from the Mordialloc side.  However, the gas from the Altona side is likely to have covered up any disturbance of the seabed that would otherwise have been evident as a result of the re-emergence of the anchor from the Altona blow hole.

357               I am not able to accept the shipowner’s contention that the heavy anchor and chain could have remained attached to the missing piece of pipe (then supposedly connected  by the “W” crimp to the rest of the Mordialloc side of the pipeline) as it remained suspended above the seabed in the water for several minutes between the rupture and the astern movement.  Dr Lyngberg argued that the “W” crimp kept the anchor engaged with the pipe during the rupture.  But, he did not explain satisfactorily how the anchor loosened itself just enough to perform his postulated ballet-like or acrobatic movement along the pipe while it remained suspended above the seabed and thus under tension from the vessel. The Mordialloc end of the pipe could not support the 8.325 tonne anchor and the chain above the seabed if it had slackened to allow the anchor to move out of the “W” crimp following the initial rupture.

358               Thus, I am satisfied that the mechanism proposed by Mr Kelle and Dr Price of a re-engagement of the anchor following the initial rupture is, more probably than not, what occurred.  Each of them attempted to verify that what they were postulating was possible by using calculations.  Neither Mr Gartner nor Dr Lyngberg made any such attempt.  The competing theories are not supported by any evidence from the sonar images that any of the experts was able to identify.  Nonetheless, something happened and, I am satisfied that the anchor disengaged from the pipeline and fell into the area of the Altona blow hole immediately after the initial rupture.

359               Oliver Wendell Holmes once observed:  “The life of the law has not been logic:  it has been experience”:   The Common Law (1881; Lecture 1 Dover: NY) at 1.  Here, there are missing pieces of the jigsaw, the lack of a sonar image identifying the movement of the anchor, the missing piece of pipe and the dispute about the amount of anchor chain in the water.  Nonetheless, the objective facts left behind on the seabed floor after the rupture and two fractures demonstrate that the Mordialloc end must have been pulled back by the vessel as she either drifted astern or after the engine movement commenced at 16:27:59.  Whichever way the re-engagement occurred, the astern engine manoeuvre certainly pulled the Mordialloc end of the pipeline back to almost a right angle from its trench before fracturing it as Dr Price explained.  This pulled and damaged significantly more of the pipeline on that side than the initial fouling.

360               I accept the upper range of Mr Kelle’s evidence that about 30 metres more pipe had to be replaced than would have been if nothing further had occurred to damage the Mordialloc end after the initial rupture.  This would have resulted in the repair cost being increased, based on Mr Vines’ estimates, by about $1,080,000.  He estimated that this would have extended the repair work by about three days, which could also affect the claims by Qenos and Huntsman in respect of their loss of gas supply for that additional period.  I am not satisfied that the anchor remained engaged with the pipeline after the initial rupture.  I find that the anchor re-engaged with the pipeline as a result of the ship moving astern after the south westerly momentum of the dead slow ahead movements ceased or after the slow astern order took effect.  I am of opinion that this re-engagement was inseverably connected to and part of the distinct occasion that arose when the ship’s engine was ordered ahead after 16:19:51 in the ensuing emergency as those in command of the ship sought safety from the escaping gas.

Conclusion

361               I am satisfied that there were two distinct occasions that occurred in direct connection with the operation of APL Sydney on the afternoon of 13 December 2008 that gave rise to claims for loss of or damage to property and consequential loss resulting therefrom within the meaning of Arts 2(1)(a) and 6(1) of the Convention.  The first distinct occasion was the chain of events leading to and immediately following the anchor fouling the pipeline at about 15:44-15:45.  The second distinct occasion was the chain of events leading to and immediately following the rupture of the pipeline at about 16:20-16:21.

362               The parties asked that they be allowed the opportunity to consider these reasons before addressing on the relief that should be ordered as a result.  As it may be of assistance to them, I will indicate tentatively a very preliminary view of the consequences of my findings.  However, that is not a view that I have formed after assistance from argument and is merely a very tentative indication.

363               My preliminary view is that the shipowner is entitled to limit its liability for:

·               the first occasion by establishing a limitation fund in the maximum amount calculated using the formula in Arts 6 and 11 of the Convention.  That fund would be available to pay all claims other than those for which claims arose on the second distinct occasion;

·                the second occasion by establishing another limitation fund.  That fund would be available to pay all claims that arose because of the additional damage done to the pipeline by the ship’s movements ahead causing the rupture and astern after it, the extra time for repair of the pipeline, including additional consequential loss claimed by Qenos, Huntsman and other parties, and the loss of escaped gas.

 

364               I will fix a further hearing to determine the relief that should be ordered.

 


I certify that the preceding three hundred and sixty-four (364) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Rares.



Associate:


Dated:         18 March 2010