ÐÓ°ÉÔ­´´

This week Britain has been trying to persuade the international community to modify roll-on roll-off ferries to prevent a repeat of the Zeebrugge tragedy. The British government argues that all ro-ro ferries, not just new ones, should comply with safety standards introduced last year to reduce the likelihood of passenger boats capsizing if they are damaged.

Britain’s eagerness to improve the stability of ferries is in no small part the result of a year’s patient research on models of ro-ros carried out by the British Maritime Technology and financed by the Department of Transport. This research, prompted by the Herald of Free Enterprise disaster of 1987, found that in anything other than calm weather the existing fleet of ro-ro ferries were prone to capsize rapidly. In the case of the Herald, 193 people died in the chill March waters shortly after the ferry left the Belgian port of Zeebrugge with its bow doors still open. Once water had got onto the car deck, the ship turned over within about 90 seconds.

The International Maritime Organisation has, since its foundation in 1948, administered the international conventions on the safety of life at sea – usually referred to by its acronym, SOLAS. Last year, the British government was instrumental in persuading the IMO that it should adopt tougher stability standards for ferries. The crucial concept in the safety of ferries is a measure known as residual stability. According to SOLAS this means that a ferry, or any other passenger ship, should be able to survive damage to two of its watertight compartments. The greatest vulnerability of ro-ro ferries is their open vehicle decks with no watertight bulkheads. Once water gets onto these decks, the vessel can capsize rapidly.

Under the old SOLAS rules ferries had to have at least 76 millimetres freeboard, which on ro-ro ferries is essentially the distance between the vehicle deck and sea level. Under SOLAS 90, which came into effect for all ferries built after 29 April 1990, this distance was increased to 1.25 metres.

At this week’s meeting of a little known subcommittee of the IMO’s Maritime Safety Committee, Britain has been pressing for these SOLAS standards to be applied retrospectively to the existing fleet.

The British government was convinced that higher stability standards are needed following extensive tests on a model of the Herald Free Enterprise at BMT Fluid Mechanics, a subsidiary of British Maritime Technology. Before privatisation BMT was the ship research division of the National Physical Laboratory. Ian Dand, who heads BMT’s ship hydrodynamics department, carried out more than 200 trials with a model of the Herald in BMT’s test tanks at Teddington.

In the aftermath of the Zeebrugge disaster, the government was initially reluctant to admit that the design of ferries should be improved. However, the official inquiry into the disaster led by the judge Sir Barry Sheen reported that it was the sheer speed of the accident which caused the large death toll.

Following the inquiry, the Department of Transport financed two sets of model tests, one at BMT and the other at the Danish Maritime Institute. These were to look in more detail at the vulnerability of ferries, and in particular the circumstances in which the ro-ros would capsize rapidly.

The Danish team used a model of the Saint Nicholas, a ferry used in the North Sea run from Harwich to the Hook of Holland. Dand tested a model of the Herald.

These two ferries are broadly representative of the ro-ros operating out of Britain’s ports. Although the Herald of Free Enterprise was scrapped after Zeebrugge, two other ships in the class still sail across the Channel: the Pride of Free Enterprise, now known as the Pride of Bruges, and the Spirit of Free Enterprise, now called the Pride of Kent.

The Danish tests found that the Saint Nicholas, too, was even more prone to capzise than ships of the Herald class. This, according to Dand, is because their funnel and equipment is located in the middle of the vessel rather than on either side.

Originally the research findings of both studies were to be given at a meeting of the Royal Institution of Naval Architects in April last year. However, Dand’s paper was withdrawn at 24 hours’ notice: he was due to give evidence in the corporate manslaughter case against the owners of the Herald of Free Enterprise and the substance of his paper was sub judice. Now that the case has collapsed the first details of his work can be revealed. His paper is due to be read at the RINA’s next conference in April.

Dand tested a model 3.1 metres long (a scale of 1 in 42) of the Herald with varying ranges of freeboard. The hull was built of glass-reinforced plastic. Dand describes the model as ‘like a Meccano set. You can strip it down and rebuild it.’ This allowed him to change the freeboard or centre of gravity easily, or to simulate damage to the hull.

In most respects, the BMT model faithfully respresents the Herald class of ferries, with one significant difference. Dand built an extra bit of stability into the model with polyurethane foam, so that in cases where a real ship would capsize, the model could be prevented from turning over, to save the expensive monitoring instruments from being drenched. Even so, says Dand, on a couple of occasions the model still capsized so rapidly that no one could intervene before it turned turtle.

His trials showed that a ferry does not inevitably capsize when water gets onto the vehicle deck. One test showed that, if the ferry was undamaged, even a vast amount of water on the vehicle deck – equivalent to more than 1500 tonnes – would not cause the ship to capsize. Dand says: ‘Provided the water did not flood through the vessel, the ferry just wallowed in the waves; it certainly did not capsize.’ He says that the Herald of Free Enterprise capsized because a vast amount of water was coming through the open bow doors as the boat moved forward, and there was nowhere for the water to go.

The critical test was whether the ferry could survive when it was damaged. There is real risk of collision in the busy waters around Britain, especially in the Channel. An earlier BMT study found that in a five-year period there was more than 1350 ‘serious accidents’ around the British coast, any of which could have involved a ro-ro ferry and had the ‘potential’ for disaster.

Lloyd’s Register estimates that every five years a collision could ‘have the appropriate ingredients’ for a repetition of the Zeebrugge turnover. In 1989, the ferry Hamburg came perilously close to being that disaster while crossing the North Sea. A container ship ripped a 20-metre gash in the ferry, above and below the waterline, killing three passengers. Fortunately the master was alert to the danger and adjusted the vessel’s trim tanks, causing it to heel away from the hole to prevent water entering.

In the BMT tests, Dand made a hole in the side of the Herald model, virtually corresponding to the two-compartment damage which is the SOLAS standard. Dand said ‘the purpose of the experiments was to discover if it is inevitably that it capzises, and if not what are the conditions of stability.’

The criterion for a model passing the test was whether it could survive from ‘collision’ to capsize for at least an hour in a moderate to rough sea with waves of up to 5 metres – corresponding to about force 6 on the Beaufort scale. Although the IMO’s standard evacuation time is 30 minutes, in practice there may not be much difference in these times. In real life, it could be some time after a collision before a captain realised that the vessel was doomed, and gave the order to abandon ship.

The most obvious conclusion from Dand’s tests is that if the ferry’s freeboard, after damage, is greater than the height of the waves, then it would almost certainly survive. Where the wave height is greater than the freeboard it is largely a matter of luck whether the damage causes it to capsize.

Dand explains that if a ferry has a hole in its side, then water can flow both in and out. The craft is in danger only if more water flows in than can flow out, and if water begins to build up on the vehicle deck.

If the ferry is heeled away from the damage, any water on the vehicle deck will just lie there, adding to the list of the vessel but not endangering it, because the list ‘would tend to prevent more water getting on.’ On the other hand, if the ferry heels towards the damage, then more water is likely to reach the vehicle deck. As long as more water is getting on board than is leaving, eventually the ferry will capsize.

Dand says that ‘once water is on the (vehicle) deck you have a potentially pretty dangerous situation building up.’ In most of the trials where the model was heeled toward the damage, and the height of the waves exceeded the freeboard, the full-size vessel would capsize in less than 20 minutes.

Ships have a tendency to right themselves after heeling: if this righting moment disappears the ship will capsize. If a ferry has a freeboard, after damage, of only 76 millimetres it would survive only if the righting moment was ‘phenomenally high.’ In practice, says Dand, no one would design a ferry with such a large righting moment because it would be rather like a car with a very stiff suspension. The violent roll would be uncomfortable for passengers, who would be seasick, and it would be rough enough to damage cargo. With a larger measure of freeboard to give protection against rapid capsizing, however, designers could afford to have a smaller righting moment.

Rescue will continue to be a problem with high-sided ferries. These days the cargo is very light ‘all volume and no mass,’ says Dand. In effect ‘you’ve got an inverted iceberg. Modern ferries have huge windages.’ This means that the high sides offer a large surface area to the wind, but with a relatively small amount of the vessel underwater to prevent it from being blown around. ‘If they get holed and start to drift they will get beam-on and drift downwind at quite a high speed,’ he says. This makes it difficult to keep the rescue boats in contact with the ship.

BMT and its precedessors pioneered the use of models to test ship safety. While Dand was carrying out his experiments, he discovered that the ship division of the National Physical Laboratory, BMT’s forerunner, carried out a similar range of tests on the model of a liner, the City of Lucknow, towards the end of the First World War. Nowadays, the physical model tests are complemented by extensive computer modelling. Dand says that the advantage of physical model tests is that ‘they have given us some strong leads – and helped to explain otherwise inpenetrable happenings.’

The Department of Transport is about to commission a further tranche of research. The researchers will study the effect that various devices for improving ferry stability have on the tendency to capsize.

This week, the government has presented its case to IMO for higher stability standards on ro-ro ferries. If it fails to get agreement then Britain will try to reach a regional agreement, with its main trading partners. France, Germany and Scandinavia are generally sympathetic to the need for higher safety standards. If this fails then Britain has the power unilaterally to ban ro-ro feries from its ports if they do not meet the safety standards it requires.

* * *

The options for greater safety at sea

The British government wants the SOLAS 90 standards to be applied retrospectively to all ferries. In practice this means that ferries will have to be fitted with one of the number of devices designed to improve their stability.

Watertight bulkheads would subdivide the vehicle deck and reduce the likelihood of a capsize, but they would also make loading the ferry with vehicles more difficult. A variation on the same theme is to create buoyant wing spaces, a watertight second skin on the vehicle deck inside the hull.

The most popular device is likely to be a fitting sponsons to the side of the hull – a kind of hollow metal compartment stretching about three-quarters of the way round the ship and acting as a permanent lifebelt. Sponsons would increase the righting moment of ferries. The vessels would be less likely to capsize, the passengers more likely to be sick.

The choice of which remedy is adopted is likely to vary from ship to ship. Although bulkheads impede loading, for example, this matters less on long hauls across the North Sea than it does on trips across the Straits of Dover, where a short turnaround time is of prime commercial importance.

According to the shipowners the cost of modifications could reach around 60 million Pounds for the British fleet of ferries operating on international voyages, and 25 million Pounds for the domestic fleet. The figures are effectively the bill for all British ro-ros, both passenger and freight. Although SOLAS rules cover only passenger to ro-ros, they also apply to those freight ferries which carry more than 12 lorry drivers as passengers. There will also be additional operating costs, and some berths may need to be modified to take the extra beam of a ferry with sponsons, for example. The shipowners say that the cost will mean higher fares.