(“the George S” case)
In 1983 it first came to light that some ship’s crews were siphoning off their cargo of crude oil to use as fuel. The practice was the subject of a circular to all governments, issued by the Marine Safety Committee of IMO, warning them of a serious danger to both ships and persons by the use of low flash-point fuel. It was established that in several known cases the transfer of cargo oil to the bunker system had been carried out by means of a cross connection between the cargo and bunker piping systems. It was urged that, where illegal transfer was suspected, surveyors should check the bunker fuel and, if necessary, take samples to be analyzed by an approved laboratory to ascertain if the flash-point was below 60°C. If so, this would be proof of bunker contamination. Under the terms of the Bulk Oil Clauses, however, theft of cargo in this manner is not one of the perils covered by the Risks Clause.
The difficulty of ascertaining the amount of shortage and its cause is well illustrated by the case of (the George S.). This concerned a claim by the owners of a cargo of North Sea crude oil against the carriers, who had undertaken transport from Teesside to Curacao. The sole issue was whether the vessel delivered at the discharge port all the oil which had been loaded on board her.
In the Commercial Court carefully studied all the methods of measuring the oil, and ascertained that at the loading port a meter was inserted in the pipeline from the shore tank, and also a temperature recording apparatus. These enabled a computer to calculate continuously the volume of material flowing through the pipe, and its temperature. In addition the oil in the shore tank was measured by ascertaining the depth of oil and its temperature, and reading off the resulting volume from a table prepared for that tank in the past. This served as a check on the measurement by meter, producing a figure of 862,838 barrels, which the learned judge found “inherently impressive”.
The ship’s loading figure, on the other hand, was 867,470 barrels, which excluded the quantity on board before loading (OBQ). This figure, it will be noted, exceeded the shore measurement, and caused Mr. Justice to reflect on the possibilities of inaccuracies occurring when measuring oil in a ship’s tank. The common method is to ascertain the “ullage”, i.e. the distance between the surface of the liquid and a measuring point at or near the top of the tank, and to take the temperature of the liquid. The volume can then be calculated from tables prepared for the ship when she was built. Naturally some allowance requires to be made for a vessel’s trim or list, and possible changes which may occur in a ship’s tanks from the day when the ship was built. In this respect it is useful to know the vessel’s experience factor (VEF) which is computed from the record of cargoes previously loaded in the same vessel. “In adjusting practice this is calculated from the ratio of the sum of the vessel’s loaded quantities and the sum of the relative shore lank quantities from at least half a dozen consecutive previous voyages.” Such evidence was not available in this case; in fact it transpired that the ullages had not been measured by tape at the loading port, and reliance had been placed on the automatic ullage gauges in the control room. Unfortunately, it was discovered at destination that one of these gauges was faulty, requiring the extraction of air pockets out of the mercury.
The discrepancy was discovered at Curacao when the inspector there ullage the tanks directly and found that the measurement did not correspond with the automatic ullage gauge. After the gauge was attended to the readings corresponded and the ship’s discharging figure was calculated to be 861,062 barrels after allowing for the quantity remaining on board (ROB). According to the shore tank measurement, calculated from the depth of oil in the tank and its temperature, the quantity delivered was 854,335 barrels after deduction of bulk sediment and water (BSW). Mr. Justice did not find it proved, on the balance of probability, that the ship’s discharging figure of 861,062 barrels was wrong.
On the available evidence he concluded that the probable inevitable loss by evaporation on board ship was about 0.2%, or roughly 1720 barrels for the cargo. The difference between the shore loading figure and the ship’s discharging figure was, however, only 1,776 barrels, which was for all practical purposes within the margin of 0.2%.
Since, however, the shore tank measurement at destination was equally free of criticism he felt obliged to see if there was any explanation for the possible loss of 8503 barrels in transit, i.e. the difference between the shore tank readings at the two ports.
There was no evidence to show that some 7000 barrels had leaked out during the voyage, and there was no intermediate port involved. He did not think it probable that such a quantity could be concealed on board without it being detected, and neither was any charge brought to the effect that the crude oil was used as fuel for the vessel’s engines, if, indeed, it was capable of being used for that purpose. If the oil had been diverted, deliberately or inadvertently to other shore tanks it would have been necessary for the plaintiffs to adduce evidence to the standard required. However, a statement from the independent surveyor at the destination port asserted that the pipeline would have been full prior to discharge and thus negative any suggestion that it was empty or half empty; at the same time it also by implication negative the possibility of diversion. The learned judge could only conclude that the plaintiffs’ claim failed.
In the Court of Appeal, nevertheless, it was observed that, by comparing like with like i.e. shore measurements with shore measurements, and ship measurements with ship measurements, at both the loading and discharging ports, a substantial short-delivery resulted. In particular, Lord Judge stated that accurate shore tank measurements provide powerful, but not conclusive, evidence of the quantity discharged from the vessel. If the cargo-owners rely upon such evidence, it is for the ship to displace it. There was no evidence whatever that the pipeline was other than full when discharge began or that any part of the oil was in fact diverted to unmeasured tanks, consequently the situation imposed a very considerable burden of proof upon the ship to show beyond reasonable doubt that none of the goods were lost or stolen after receipt, and that all received had been delivered. Here the shipowners were unable to counter the evidence of short-delivery or to establish beyond reasonable doubt that a short-delivery was impossible. The appeal was allowed on the basis of the shore figures, i.e. a loss of 8,503 barrels less an expected evaporation of 1,720 barrels a net claim of 6,783 barrels.
Under a policy of insurance containing the Bulk Oil Clauses, of course, the burden of proof is on the assured to bring themselves within the insured perils. In circumstances like those above a claim against Underwriters could not be substantiated.
Source: Witherby Marine Insurance Claims – J. Kenneth Goodacre
