The Shetland Oil Disaster: Could the ship have been saved?: Tim Kelsey asks why it all went so wrong for 'a good boat' in familiar waters

Jim Dickson knew the Braer well. As oil pollution and safety adviser to Shetland Islands Council, he had monitored her sailing past the island 'more than a hundred times'. He had inspected her himself at Sullom Voe oil terminal on many occasions. 'It was a good boat,' he said.

But shortly after 5 o'clock on a windy Tuesday morning, he received a call from Sullom Voe reporting simply that 'a vessel was drifting off the shore at Sumburgh'. That vessel was the Braer, carrying 85,000 tonnes of light crude oil.

The 17-year-old tanker, owned by a Connecticut-based shipping firm, started its last journey last Sunday. The passage from Mongstad oil terminal, near Bergen in Norway, via Shetland to Quebec was one she had made, without incident, twice in the last three months. She had passed a stringent inspection in May last year by the Norwegian authorities.

But port inspections in February and August identified a variety of small problems: leakages from the main engine and waterpipes, leaks in the pumproom, corroded pipe covering, and leakages in the hydraulic system. All were promptly rectified. On 19 October, when the tanker was taking on oil at Sullom Voe, a blocked vent pipe - allegedly due to poor maintenance - caused some oil to be discharged.

It is a port's responsibility to ensure seaworthiness before a vessel departs, and it was with misgivings that technicians for the Norwegian firm Statoil, which controls the Mongstad terminal, noted on 3 January problems with the steam boiler room piping.

The Greek captain of Braer, Alexandros Gelis, 45, with 10 years' experience as a tanker skipper, reassured the Norwegians that the problem had been fixed. Statoil did not conduct an independent investigation.

There is a slim chance that faulty pipework running to the ship's boiler could have contaminated the fuel tanks with water. Tony Redding, a spokesman for its owners, B&H Management, said yesterday that the piping in question had no connection with the fuel or engine systems. But the question remains as to whether it was effectively repaired before the ship sailed.

The vessel did not take on fuel for its own engines in Norway. According to B&H Management, the Braer was refuelled on 21 December in New York, taking on enough fuel to cover the round trip from Norway to Quebec.

But a Norwegian environmental group, Bellona, has claimed that a senior Mongstad source said that the emergency fuel tank, vital for allowing power to be generated after the main engine had stopped running, was empty when the ship sailed.

Capt Gelis set his course through the 22-mile wide channel between Fair Isle and Sumburgh Head, a common choice for tankers. In high seas, however, the Fair Isle channel has a notoriously high swell. The question also arises of the sort of weather forecast Capt Gelis used. It is common for ocean-going vessels to leave port with only a three-day shipping forecast but it is much more appropriate for a large boat to use a 10-day forecast, which gives more flexibility in planning her route. If Capt Gelis used a 10- day forecast which correctly predicted the gales, why did he set course through the tricky Fair Isle channel?

Early on 5 January, the Braer lost power. The captain claimed, shortly after being winched off, that this had happened at around 5.30am. The following day, the ship's owners said that the ship was disabled at 4.40am.

These timings are contradicted by a trawlerman, in the area several hours beforehand. Sam McCullough, skipper of Stephens, claims that, at 11.45pm on 4 January, he passed a large tanker in the Fair Isle channel in virtually the same position as where Capt Gelis says he lost power - five hours later. The only tanker in the area at the time was the Braer. Police have the logbooks of the Braer, which should identify its position at the time of the sighting.

Central to the disaster is why the engines failed. There are several possible explanations. Ulthie Roldan, the chief engineer, said after being rescued that 'there were problems with the generator and boiler. The bad weather meant seawater had got into the fuel tank'.

But experts have expressed surprise that both the fuel tank for the main engine, as well as that for the diesel-run auxiliary engines, could have been contaminated. There are other theories. The settlement tanks, which separate water from the fuel before it reaches the engine, require regular maintenance, as do the purification centrifuges which clear water out of the fuel.

B&H Management has insisted that the tanker's crew was correctly trained and would have been vigilant in monitoring the fuel tanks. It has admitted that a four-strong 'riding gang' of Polish maintenance engineers was working in the engine room, but Mr Redding said they were repairing the pipeline on the ship's freshwater generator.

It is not clear when news of the engine failure reached the local coastguard. According to officials, Capt Gelis radioed the Maritime Rescue Co-ordination centre in Aberdeen at about 5.05am. Aberdeen then contacted Shetland Coastguard at Lerwick. According to Capt Kenneth Low, district controller of the coastguards on Shetland, the initial contact indicated that the Braer was 'in no immediate danger'. This may have been an underestimate on his part.

Shortly afterwards, Shetland Coastguard offered tug assistance. Capt Gelis hesitated. Ordering tugs would involve huge costs. He contacted his employers in New York via the radio- telephone, and was given the necessary authority within 15 minutes. He then radioed the coastguard and asked for a tug. A coastguard spokesman said afterwards: 'Had he called for tugs earlier, who knows what might have happened?'

It probably would have made no difference. The coastguard turned first to Sullom Voe where tugs are always berthed. There were two immediately available, but the operations room warned it could take five hours for them to reach the tanker. The boats were small and could not make good headway in the huge swell.

By now Capt Gelis was very concerned. 'I told them that in five hours the ship would be in trouble.'

Somebody at Lerwick coastguard then remembered that the Star Sirius, an oil rig support tug, might be at dock. The harbourmaster confirmed at about 6am that it was available, and roused the skipper. It took a further hour for the tug to prepare itself and change towing gear on the stern. Barry Cork, managing director of Star Offshore, which owns the Sirius, said the tug had left Lerwick by 6.30am. But the harbourmaster is adamant that it left around 7.10am. Did the coastguard take too to order the Sirius out?

Capt Gelis claims, in contradiction to other statements, that the coastguard told him the Star Sirius would be arriving at about 7.30am. Some coastguard officials refer to the existence of language difficulties. The Captain speaks only broken English, and could have misunderstand when the tug would be available.

The Braer was drifting towards the shoreline and Capt Gelis decided to winch off most of his crew. At 6.39am a coastguard helicopter had arrived above the tanker, which was about eight and a half miles off the coast with no power on board at all. Twenty minutes later it prepared to winch off the crew. By just after 7am, the first crew member was hauled to safety.

At 7.40am, a second helicopter arrived from RAF Lossiemouth. At about 8.30am, the coastguard helicopter flew to Sumburgh, and Capt Gelis told the RAF helicopter to winch off the rest of his crew. The ship was now within three-quarters of a mile of the shoreline.

Both the coastguard and the ship's owners have applauded his decision to abandon ship at this stage. The coastguard urged Capt Gelis to do so. Bob Driver, the Aberdeen Rescue co-ordinator, said: 'There was every indication that she was going to drive straight ashore. I would hate anyone to get the impression that we are not environmentally conscious, but my first concern is the safety of life.'

By 8.54am, the last of the crew had been winched off. At 9.38am, the Star Sirius arrived - four and a half hours after the first call to the coastguard from the tanker - but there was nobody left aboard.

'There was very little I could do,' said the tug captain, David Theobald. 'We had rocket lines ready, all the gear for making a connection to the tanker, but the only way we could was to have somebody on the tanker.'

Most experts agree that the Star Sirius could have held the tanker clear of the shoreline. Capt Gelis asked to be dropped back on the ship to try and establish a towing line. But a team was not airlifted back on to the vessel immediately the tug arrived.

At 10.58am, four men, including Mr Dickson, were winched on to the deck when the vessel was only about 50 yards off the rocks, and 17 minutes from disaster. None of the power winches on board worked. 'We tried to haul the rocket line on board, but the swell ripped the rope from my hands,' said Mr Dickson. 'You could not even walk on the decks because of the waves. There was nothing we could do.'