Summary

The Australian bulk carrier River Embley sailed from Gladstone on the afternoon of 8 July 1996 on its regular voyage to Weipa to load bauxite. By 0100 on 10 July, River Embley was approaching Little Fitzroy Light at a speed of 16 knots. At about 0115, three vessels were seen northwest of Little Fitzroy Light, on River Embley's port bow, the closest and the only one liable to pass close to River Embley was at a distance of 8 miles. At about 0119, River Embley altered course from 344 to 320 on the track to Low Isles.

At about 2330 on the night of 9 July 1996, the fishing vessel 'Bronze Wing' sailed from Cairns for Gibson Reef, about 46 miles south-south- east of Cairns, to collect aquarium fish. The vessel cleared Cairns fairway channel and set course for Cape Grafton. From a position about a mile off Cape Grafton and making good a speed of between 6 knots and 7 knots, the vessel was to follow a predetermined track to a position between 1 mile and 1.6 miles off Little Fitzroy Lighthouse. At about 0119, the Deckhand on watch on board Bronze Wing saw a large vessel on the starboard bow at 6 miles on the radar and confirmed the ship's location by a visual sighting.

The two vessels collided at about 0134, with the starboard bow of the fishing vessel coming into contact with the starboard side of River Embley's hull, forward of mid length. Bronze Wing suffered significant damage, but River Embley's paint work was only superficially scratched. Nobody suffered any significant injury and no pollution resulted.

After the collision the two vessels contacted each other by VHF Radio. When it was determined that Bronze Wing did not require assistance and would return to Cairns under her own power, River Embley resumed course for Weipa.

Conclusions

These conclusions identify the different factors contributing to the incident and should not be read as apportioning liability or blame to any particular individual or organisation. The following factors are considered to have contributed to the collision between Bronze Wing and River Embley.

1. The probability is that Bronze Wing initially crossed the bow of River Embley from port to starboard and subsequently, shortly before the collision, altered course towards the approaching vessel to become a crossing vessel on River Embley's starboard side.
2. The Deckhand on watch on Bronze Wing did not have the experience to assess the developing situation and to take action to avoid collision, nor was he qualified to do so.
3. The Deckhand on Bronze Wing had an insufficient knowledge of the International Regulations for the Prevention of Collisions at Sea.
4. The Deckhand on Bronze Wing was not proficient in the use of radar or techniques to determine whether risk of collision existed.
5. The lookout kept on board Bronze Wing was not effective.

6. Bronze Wing was manned in accordance with the provisions of the Uniform Shipping Laws Code, however this standard does not allow a qualified person to be in charge of a navigation watch at all times on planned extended voyages.

   Although the Second Mate on River Embley was placed in an extremely difficult situation by the fishing vessel altering to a collision course so close to the bulk carrier, the following factors also contributed to the collision:

7. The Second Mate on River Embley made assumptions regarding the nature of Bronze Wing's operation based on scanty information and an inaccurate perception of the situation, during a time when other duties may have led to a degree of distraction.
8. The Second Mate's experience of the typical actions of fishing vessels operating in the relatively confined waters of the Great Barrier Reef reduced his sensitivity to the risk of collision.
9. The absence of the Rating from River Embley's bridge between about 0130 and 0134 resulted in the bridge manning being insufficient to manage normal contingencies and the Second Mate alone could not respond to the developing situation, as he was unable to both sound and flash the appropriate warning signals and alter course at the same time in an attempt to avoid collision.
10. In the circumstances immediately before the collision, the Second Mate's only option was to turn to port, away from Bronze Wing.

Summary

On the morning of 5 March 1996, the Indonesian 33,747 tonnes deadweight geared bulk carrier Bogasari Dua and the Panamanian 38,313 tonnes deadweight geared bulk carrier Midas were both lying at anchor in Geraldton Roads, Western Australia. Midas was anchored about 1 mile to the north of Bogasari Dua.

At about 0400, Bogasari Dua, lying to a single starboard anchor with eight shackles in 30 m of water, began to drag anchor in gale force southerly winds and was driven towards Midas. Those on the bridge of Bogasari Dua did not realise that the vessel was dragging anchor until after 0430, when they were alerted by the whistle signals of Midas. Bogasari Dua came into contact with Midas at about 0445, before either the anchor could be weighed, or the engine made ready. As soon as the two vessels moved apart, at about 0450, the Master of Bogasari Dua manoeuvred his vessel clear.

Bogasari Dua received damage to port side shell plating in way of no.5 hold to the main deck bulwark on the port side, to the port side accommodation ladder hoist, to the port side lifeboat davit base and to the port lifeboat, which was split into three sections. Midas received indentation damage to the bulbous bow, also to shipside plating in way of nos. one and two starboard topside tanks and damage to various lengths of shipside railing.

No one was injured in the incident and no pollution occurred.

Conclusions

These conclusions identify the different factors contributing to the incident and should not be read as apportioning blame or liability to any particular organisation or individual.

It is considered that:

• Bogasari Dua's anchor broke out shortly after 0400.
• The Mate of Bogasari Dua was not aware that his vessel was dragging anchor until after 0436.
• When the Mate's attention was caught by the intermittent whistle signals of Midas, Bogasari Dua was already too close to the other vessel for effective avoiding action to be taken.

The following factors are considered to have contributed the collision:

1. The Master of Bogasari Dua had switched off the weather facsimile machine, denying the vessel access to local weather forecasts and the strong wind warnings.
2. Despite his earlier experiences of dragging anchor off Geraldton, the Master of Bogasari Dua did not give any specific directions or warnings to the deck officers to be on their guard or give instructions for the engineers to maintain watches.
3. The watch keeping officers aboard Bogasari Dua did not properly monitor the weather conditions and so did not inform the Master of the increasing wind strength.
4. The Mate of Bogasari Dua did not monitor the vessel's position or keep a lookout.
5. Knowing there was to be a lengthy time at anchor before berthing and knowing the exposed nature of the anchorage, the Master of Bogasari Dua did not consider using more anchor cable to improve the holding capability of the anchor or consider ballasting the vessel down to reduce its windage area.
6. The Master of Bogasari Dua did not consider the option of weighing anchor and standing off the port when the wind increased during the evening of 4 March.
7. The Mate of Midas did not keep a proper lookout in that he did not make sufficiently frequent checks on the relative position of Bogasari Dua.
8. The Master of Midas did not consider the slipping of his anchor cable as an option open to him.

Summary

On Sunday 21 January 1996, the Singaporean flag, 1941 tonnes deadweight livestock carrier Carabao 1, while sailing under pilotage from the privately owned Labroy wharf, Hudson Creek, Darwin, struck Old Man Rock, in the East Arm of the harbour.

No compartments were breached, and no pollution occurred. An inspection carried out by divers ascertained that only minor indentation was visible on sections of bottom plating on the vessel's starboard side and the vessel was permitted to continue on its voyage.

Conclusions

These conclusions identify the different factors contributing to the incident and should not be read as apportioning blame or liability to any particular individual or organisation.

The following factors are considered to have contributed to Carabao 1 making contact with Old Man Rock:

1. Faced with a deviation from the usual operational procedures, the Harbour Master did not fully evaluate the changed circumstances and assess what appropriate action was required.

2. The Harbour Master's decision to make an earlier, quicker turn than normal, which resulted in Carabao 1 being shaped to pass Old Man Rock on the westerly heading at a much closer distance than normal.

3. The phenomenon of tunnel or narrowing field vision while the Harbour Master was executing the turn around Old Man Rock.

4. The absence of any checks to ensure that Carabao 1 was at the correct wheel-over position before, and on the intended track at completion of, the turn off Old Man Rock.

5. A totally unexpected strong northerly tidal flow to the south-east of Old Man Rock.

6. The lack of structured bridge management procedures on board, to assist and monitor a pilot's actions.

7. The lack of a drawn-up pilotage plan, one that included contingency alternatives.

8. The lack of a tidal monitoring regime for the area eastwards of the new port construction, either under the new port construction contract or under Port Authority Management procedures, to ascertain any changes in tidal flows, particularly at Spring.

Summary

In the early morning of 31 January 1996, Casey Chouest, an offshore research vessel engaged in seismic survey operations, was in the prospect area known as Snark 3D, close to Thevenard Island in Western Australia. At about 0700, the fire alarm panel in the wheelhouse indicated a fire in the seismic compressor room space and smoke could be seen coming from the vents.

The crew and seismic staff mustered and closed down the space, starving the fire of oxygen and applying boundary cooling; eventually extinguishing the fire in the space. However, heat from the fire was conducted through the uninsulated steel bulkhead igniting combustible material stored against the common bulkhead in an adjacent space. The firefighting tug, Total Endeavour, was sent to the scene in response to a request from the Master and, using its fire monitors, applied boundary cooling to the hull while the crew continued boundary cooling on board.

The investigation established a sequence of events initiated by a massive short circuit in the cable from a dedicated generator supplying power to the hydraulics for the reels which deploy the seismic gun cables. An incorrectly sized bracket, securing the armoured cable had caused damage to the armouring and the rubber sheath surrounding the three phase conductors. The evidence indicated that the insulation breakdown had proceeded over a long period until the cable failed to perform under normal load conditions. The evidence also indicated that the fault current started to flow from one phase to the earthed bracket. This earth fault increased over time causing heating of the cable conductors in the section of cable adjacent to the bracket. This, in turn, contributed to the complete failure of the cable due to phase to phase and phase to earth fault current.

As the trip settings on the circuit breaker were set far too high, it did not trip, and the cable fault imposed a fault current level on the generator high enough to momentarily cause a locked rotor condition in the generator. The high inertia from the generator caused mechanical failure of the bottom end bearing in no.4 unit. The connecting rod came through the side of the crankcase and pressurisation of the crankcase blew much of the sump oil, out of the hole in the side, over the D399 Caterpillar engine below and to starboard of it. The oil self-ignited on the turbo-charger casing of the D399 engine and burning oil in the bilge melted the polycarbonate bowl of the forward fuel filter on the D399, allowing fuel under pressure to escape and spray over the engine and into the bilge thus adding to the fire.

The burning fuel in the bilge flowed to the starboard aft corner of the seismic machinery room, against the space in the reel-room where large numbers of oil filter elements were stored in cardboard cartons. These stores were ignited by heat conducted through the uninsulated steel bulkhead.

The fire was extinguished at about 1045, approximately 3 hours after the first alarm.

Conclusions

These conclusions identify the different factors contributing to the fire on board Casey Chouest and should not be read as apportioning liability or blame to any particular organisation or individual.

The fire in the seismic compressor room and the adjacent space resulted from a number of factors that combined to initiate and spread the fire. These include:

1. The breakdown in insulation initiated by securing the electrical cable from the seismic generator with a bracket of the wrong radius.
2. An undetected earth fault at the bracket led to a progressive breakdown in cable insulation and eventually to a short circuit between phases.
3. The lack of electrical planned maintenance on board, in particular with regard to the routine testing of electrical cable insulation on board the ship.
4. A lack of appropriate supervision in fitting the circuit breaker and the incorrect trip settings (thermal and magnetic) rendering it unable to protect the cable or the generator under either full load current condition or under fault conditions.
5. The sudden load thrown on to the D343 Caterpillar diesel engine by the short circuit, on an engine in very poor condition caused the failure of no.4 bottom end bearing and the subsequent damage to the side of the crankcase.
6. The pressurisation of the crank case due to worn piston rings and cylinders, causing the sump oil to spray over the turbo charger of the adjacent Caterpillar D399 diesel engine.
7. The lack of a proper maintenance regime and standard operating instructions covering the operation of the Caterpillar D343 engine.
8. The use of polycarbonate, instead of steel, fuel filter bowls and the absence of a remote fuel shut off allowed fuel under pressure to feed the fire in the bilge.
9. Lack of bulkhead insulation allowed the heat of the fire to be conducted through the adjacent bulkhead, igniting combustible stores stowed against it.
10. The containment of the fire and associated smoke and fumes was made more difficult by the number of openings in the seismic compressor room, including the penetrations in the bulkhead to the reel room and the open conduits which ran through to the accommodation.
11. A poorly defined relationship over time between the owners and the various charterers, resulting in the ad hoc selection and fitting of machinery and cables without an overall safety oversight.
12. The lack of the adoption of a suitable standard or practical safety assessment approach, by either the vessel's owners or charterers, when designing, constructing and equipping the seismic compressor room.
13. The lack of any required standard, safety survey or safety overview by the flag State, to vessels engaged in international commercial activities, particularly in view of the nature of the operation and the number of people at risk.
14. The reliance upon a 'grandfather' clause when applying the national tonnage to avoid having to apply appropriate standards to the detriment of the safety of personnel on board.
15. The absence of an appropriate inspection regime to cover vessels operating under the provisions of a "grandfather" clause.
16. The absence of a safety system to take into account appropriate standards for the fitting and maintenance of machinery or other equipment not considered to be ships' equipment.

The Inspector considers, under the circumstances, the fighting of the fire was performed efficiently, by both the ship's staff and PGS personnel, and the decisions taken in the circumstances were sensible and in the best interests of the vessel. However, it should be noted that the risk to personnel and the ship was increased by the limited number of compressed air bottles available to the firefighting team.

Summary

At about 1600 on 2 January 1996, the Owner/skipper of the Queensland fishing vessel Jay Dee was washed ashore, together with the vessel's 'carley' float, about 3 km South of Brunswick Heads, New South Wales.

In a subsequent statement to the New South Wales Police, he stated that on the evening of 31 December (New Years Eve) 1995, he had been trawling for prawns about 16 miles east of Southport, Queensland, the only crew member of his trawler Jay Dee. At about 2200, while trawling in an easterly direction his vessel was hit on the port side by a large trading ship. He was unable to identify the ship in any way.

Jay Dee immediately started to take water into the forecastle accommodation space and the engine room. The Skipper had just sufficient time to release a parachute flare over the stern and towards the bow of the retreating ship, grab a short length of line and free the carley float from the wheelhouse top before jumping clear of the sinking vessel.

The trading vessel did not stop.

The Skipper was able to gain the carley float and secure the ice box, which had floated free and inverted.

He spent that night, the day and night of 1 January, in all over 40 hours adrift, before coming ashore south of Brunswick Heads on 2 January.

Conclusions

These conclusions identify different factors contributing to the accident and should not be read as apportioning liability or blame to any particular organisation or individual.

With the inability to identify any particular trading ship that may have been involved with an acceptable degree of probability, these conclusions are based on the premise that an unidentified trading ship was involved.

The following factors contributed to the causes of the collision:

1. The trading vessel apparently not maintaining a proper lookout by sight, sound and radar.
2. The trading vessel, as the vessel required to give way, not taking action to avoid a vessel engaged in trawling.
3. The Skipper of the Jay Dee not keeping a proper lookout and not detecting the presence of the trading vessel.
4. The unserviceability of Jay Dee's radar which, if operational, switched on and observed, could have shown the approaching target.
5. The decision to operate the vessel alone, which was unsafe and meant that a proper lookout was not possible.

The following factors contributed to the length of time that Jay Dee's Skipper was adrift on the Carley float and/or increased the risk to his life:

1. The lookout/officer of the watch not seeing/responding to the flare fired by Jay Dee's Skipper.
2. The inaccessibility of the lifejackets carried on board Jay Dee.
3. The inaccessibility of the EPIRB carried at the after end of Jay Dee's wheelhouse.
4. The inappropriate nature of the Carley float as an offshore survival aid, resulting in prolonged time in the water, with no means of attracting attention, and the absence of food, water and shelter.

Summary

On the morning of Friday 13 December 1996, the Australian fishing vessel Moonshot was trawling for prawns off Pakhoi Bank, north of Cape Upstart, Northern Queensland. At about 0330, with the vessel trawling in a north-north-westerly direction, the Skipper handed over to a deckhand and went below to get some sleep.

Shortly before 0400, the Skipper was aroused by a shout from the deckhand. Scrambling up into the wheelhouse, he saw the bow of a ship very close on the starboard bow and only had time to brace himself before the stem of Moonshot came in contact with the port bow of the ship. Moonshot was pushed around to port, heeling dangerously to port as the starboard fishing boom came in contact with the side of the ship. As the ship passed clear, the deckhand saw the word ISTANBUL on its stern.

The ship did not stop or respond to the Skipper's calls on VHF 16.

Moonshot received damage to the stem and starboard bow, also to the starboard fishing boom. After ascertaining that the vessel was not taking water, the crew retrieved the fishing gear, then returned to Townsville for Moonshot to undergo repairs.

The Turkish bulk carrier Gumbet was on a ballast passage from Hong Kong to Geelong and had disembarked the Barrier Reef pilot off Cairns at 1230 on Thursday 12 December 1996.

When the Master went to the bridge at 0630 on 13 December, for his customary morning check, he was informed that the Second Mate had experienced a close quarters situation with a fishing vessel at 0340, off Tink Shoal. The Master telexed a brief account of the reported incident to the vessel's owner.

When Gumbet arrived at Geelong on 18 December, it bore signs of a recent contact between the light and load water lines on the port bow.

Conclusion

These conclusions identify the factors contributing to the incident and should not be taken as apportioning either blame or liability.

The main contributing factors are considered to be:

The inexperience of the Watch Officer aboard Gumbet, as a result of which:

• He did not allow a sufficiently wide berth when passing the fishing vessels;
• He did not use visual bearings or the radar to full effect to correctly ascertain the courses of the fishing vessels and to determine whether risk of collision existed;
• After the incident with Moonshot, he did not make contact with the fishing vessel to ascertain whether the crew were injured and whether they needed assistance.

The lack of formal marine training of the person left in charge of Moonshot, as a result of which:

• The progress of Gumbet was not properly monitored, the risk of collision was not assessed and the developing situation was not appreciated;
• An inappropriate course alteration was made, which nullified action taken by the other vessel.
• It is considered that the Master of Gumbet was unaware of the incident and so was not aware that his vessel may have been involved in a collision.

Summary

On 5 December 1996, the sail training ship Concordia, registered in the Commonwealth of the Bahamas, was on passage from Brisbane to Darwin in the Northern Territory. On deck, routine maintenance was being conducted by some of the students.

Shortly before 1145 on 5 December, one of the students, using a rotary grinder, was removing rust from the door of the battery locker, which was situated at deck level, in the after housing, below the wheelhouse. He paused and called over the Bosun's Mate who looked at the work before turning away and walking towards the forward housing. At that instance, the Bosun's Mate recalled being 'moved' over a metre, putting her hands over her ears and hearing a noise. Another student, working on the port side about 5 metres from the bridge front, saw a dark shape, which he took to be a body, being thrown over the port rail.

The Master, who was on watch and fixing the ship's position at the time, and most of the crew heard an explosion and went to the deck. Within a very few seconds life buoys were thrown overboard, one with a smoke marker. The Master could see the student in the water close to the smoke buoy, but as he watched the student disappeared from view.

Within 3 minutes, a rescue boat was launched and was making for the smoke buoy and Concordia was turned about. No trace of the student could be found, except for a pair of shoes on the deck and some evidence of blood on the deck, on the ship's rail and near the ship's side.

The maritime rescue authorities were alerted and Concordia, assisted by a helicopter and a fixed wing aircraft, searched the area that afternoon, into the evening until 1915. The following morning the search resumed at sunrise and Concordia was later joined by a single fixed wing aircraft. The search for the student was abandoned at about 1230, and after a short service Concordia resumed passage for Darwin.

Conclusions

These conclusions identify the different factors contributing to the incident and should not be read as apportioning liability or blame to any particular individual or organisation. The following factors are considered to have contributed to the loss overboard and death of the student from Concordia:

1. A spark from the grinder being used by the student ignited the hydrogen air mixture causing an explosion within the battery locker.
2. The student took the full force of the blast, which threw him through the vessel's port side rail and thence overboard, resulting in his death.
3. The ventilation of the battery locker was totally inadequate and no objective assessment of the ventilation requirements had been made, after the original batteries were replaced. However, the changed battery arrangement was in place in the survey of July 1995 and the inadequate ventilation should have been apparent to the survey authority.
4. There was no safety sign warning of the hazard of explosion due to the build-up of hydrogen gas.
5. There was a lack of perception of the potential dangers presented by the enclosed battery locker and by the accumulation of hydrogen gas.
6. The application of the Classification Society Rules for the Construction of Yachts, to a battery installation of this size and charging capacity, was inappropriate.
7. The Owners, Master and Officers seem to have placed too great a reliance on the Society to act as a de facto company superintendent in matters of ship safety standards, indicating a seemingly common misunderstanding of the role of classification societies.

Summary

Ester J, a steel hulled fishing vessel, based in San Remo, Victoria, had been fishing for shark about 50 miles south of Port Phillip during the 25 November 1997, with a crew of three aboard.

At about 0100 Eastern Australian Standard Time on 26 November, the Skipper anchored the vessel in about 75 m of water in position 39 7.88' South 145 19.25' East, about 33 miles south of Phillip Island. The vessel switched on its anchor light and the Skipper and one of the deck hands went to bed, leaving one deck hand on watch.

At or a little after 0200, the deck hand noticed the lights of a ship approaching from an easterly direction. He realised that the vessel was bearing down towards Ester J and he immediately called the Skipper. The Skipper started the fishing vessel's engine and called the unidentified vessel on channel 16 VHF, with no response. There was no time to weigh anchor or cut the cable, so he put the engine full astern, but the two vessels collided at about 0215.

The fishing vessel sustained damage to the port side. The large vessel did not stop or make any attempt to call Ester J. A quick inspection of the damage showed the vessel was not in immediate danger of sinking.

Ester J's Skipper called Melbourne Maritime Communications Centre on the radio frequency 4125 kHz, reporting the incident. He then called his home by mobile telephone.

Nobody was injured and no pollution resulted from the collision.

The hull remained watertight and the engine was operational, so the Skipper set course for San Remo, where the vessel arrived safely at 0800 on 26 November.

Foreign paint samples found on Ester J, apparently as a result of the collision, were collected into glass jars and sealed, to assist in identifying the other vessel.

Conclusions

These conclusions identify the different factors contributing to the incident and should not be read as apportioning liability or blame to any particular individual or organisation.

The evidence of the paint samples taken from Ester J, together with the circumstantial evidence of the ship's position at 0215 and absence of any evidence indicating any other ship in the area, satisfies the Inspector that the overwhelming probability is that Pacific Premier collided with Ester J.

Prima facie, the lookout maintained by the officer of the watch and any lookout aboard Pacific Premier was not effective in detecting the presence of Ester J by either direct visual means or by radar.

There was nothing to prevent Ester J anchoring in Bass Strait, however, the position of the anchorage was on the direct track from the eastern seaboard to South and Western Australia and hence in a position of greatest hazard.

Summary

In the afternoon of 26 November 1996, the Australian fishing vessel Nimbus was on passage from Cairns to Thursday Island in company with the fishing vessel Anniki, after both vessels had completed a refit. Each vessel was towing a string of aluminium dories or dinghies in line astern - Nimbus was towing five. The Singapore flag bulk carrier Maersk Tapah was on passage from Gladstone to India with a full cargo of coal. The navigation was under the control of a licensed pilot.

Both vessels were making for a point to the east of Low Isles, about 30 miles north of Cairns. At about 1522, while Maersk Tapah was overtaking Nimbus the two vessels collided. Nimbus sustained damage to its bow and wooden hull. Nobody was hurt and no pollution resulted from the collision.

The Pilot on Maersk Tapah ensured that Nimbus required no assistance and the two vessels exchanged details. Maersk Tapah continued on its voyage to India and Nimbus resumed passage for Thursday Island.

Conclusion

These conclusions identify the different factors contributing to the collision between Maersk Tapah and Nimbus and should not be read as apportioning liability or blame to any particular individual or organisation.

The factors leading to the collision centre on the watchkeeping aboard both vessels and the disregard of the International Regulations for preventing Collisions at Sea.

1. Neither the Pilot or Second Mate of Maersk Tapah made a full appraisal of the overtaking situation and risk of collision; their use of objective means to assess whether or not the bearing of the fishing vessel was appreciably changing was not effective.
2. Maersk Tapah's Pilot accepted an unnecessarily close quarter situation in the overtaking manoeuvre, resulting in contact between the two vessels.
3. With Maersk Tapah's automatic radar plotting aid giving inconsistent data for the vessels being overtaken, the Second Mate did not use compass bearings to establish whether the bearings of the vessels being overtaken were altering appreciably.
4. Maersk Tapah's Second Mate having voiced a concern about the overtaking situation did not inform the Master of his concern.
5. The lack of a lookout or any proper watchkeeping aboard Nimbus meant that the Skipper was not in a position to take action within the requirements of the Collision Regulations to take any necessary avoiding manoeuvre.

Summary

Early on 5 November 1996, the 140,086 tonnes deadweight Malaysian flag bulk carrier Giga 2 was nearing completion of discharge of a cargo of iron ore at No. 2 discharge berth, Port Kembla. Due to the vessel's light condition, the unloader could not be positioned over No. 1 hold, to remove the 1080 tonnes of cargo remaining in that particular hold. At the suggestion of the shift supervisor, ballast was pumped into No. 4 hold, which was permissible under the vessel's operations manual. Pumping of ballast into No. 4 hold commenced at 0530.

After the lunch break, two terminal workers descended into No. 5 hold, where the unloader was working, to clear iron ore from around the bottom of the forward spiral access ladder. The spiral ladder terminated at the top of the lower stool, six metres above the tank top. They had just completed this task and were about to climb the ladder when the bulkhead to the starboard side of the centre line, between No. 5 hold and the ballasted No. 4 hold, collapsed. Both workers were immediately engulfed by the deluge and, although one was able to haul himself clear, the other remained submerged.

The shift supervisor was working in his office ashore and, alerted by calls over the radio from the Hatchman, dashed on board and immediately descended the ladder into No. 5 hold. Up to his neck in the swirling water and guided by the Hatchman above, he was able to grab hold of the submerged worker. Assisted by another terminal worker, who had followed him into the hold, he was able to haul the unconscious and apparently lifeless worker clear of the water. The shift supervisor then administered cardio-pulmonary resuscitation (CPR), which was successful in restoring breathing in the worker. Very shortly afterwards an ambulance officer arrived on the scene and administered oxygen, before the worker was lifted from the hold and taken to hospital.

To identify the circumstances which led to the collapse of the bulkhead, the ship's procedures and documentation were examined. Also, a detailed examination of the bulkhead between holds 4 and 5 was undertaken, which included a metallurgical examination and a finite element analysis of the failed bulkhead.

Conclusions

These conclusions identify the different factors contributing to the collapse of the starboard side of the bulkhead at frame 193 aboard Giga 2 on 5 November 1996 and should not be read as apportioning liability or blame to any particular individual or organisation.

1. No. 4 hold was overfilled beyond its maximum allowable depth of water of 14 m.
2. The Mate relied totally on the remote gauging system for filling No. 4 hold, without physically checking on its accuracy.
3. An inaccurate reading was displayed in the ballast control room by the remote gauging system.
4. As the "high level" alarm was not independent of the gauging system, there were no effective defences, other than physical/visual checks, to ensure that the depth of water in No. 4 hold did not exceed the safe level.
5. There was no clear explanation as to the critical nature of the limit placed on the depth of ballast water in No. 4 hold. This was compounded by a lack of clear operating instructions, either in the native language of the ship's personnel or the working language of the ship.
6. The increase in depth of water from 14 m to 18 m resulted in more than doubling the maximum stresses within the bulkhead structure.
7. The specified size of the welds joining the lower stool shelf plate to the structure beneath it was insufficient to withstand the membrane forces developed at the bottom of the bulkhead with the excess water level in the hold.
8. The design and spacing of the webs within the lower stool, in relation to the corrugations of the bulkhead (and depending on the contribution made by the shedder plates), can result in high stress concentrations being formed within the area of failure at the stool shelf plate.
9. Buckling of the bulkhead, due to extensive wastage by corrosion, if not already started at the moment of failure of the welds, was imminent.
10. Extensive corrosion of the webs in the upper stool resulted in the bulkhead and the stool bottom plate being virtually detached from the upper stool. This would have facilitated detachment of the bulkhead along its upper edge during the failure but did not contribute to initiation of the collapse.
11. The quality of structural surveys of this vessel, over a period of time, was not effective in addressing the problem of substantial corrosion as defined and detailed in the International Association of Classification Societies requirements for enhanced surveys, or as recommended in the International Maritime Organisation's Assembly Resolution A.744(18).

It is further considered that:

1. detection of any deficiencies in the structure of the bulkhead was beyond the scope of Port State control inspections; and
2. based on the system for assessing applications for single voyage permits, there was no reason to refuse the application.