Back to Casualty List | Search The Casualty Database
Precis File
source MEPC 26/18, 29 June 1988
type A
volume 150 tons
material heavy fuel oil

Swedish coastal Tanker Thuntank 5, grounding on the east coast of Sweden on Dec 21, 1986

2.1 The tanker was approaching the port of Gavle on the Swedish east coast with a full load of fuel oil. The weather was bad with snow storm and a wind gusting up to 25 m/s. In darkness and poor visibility due to snow and freezing spray, an navigational misinterpretation led to the vessel grounding on a shoal besides the marked channel.

2.2 The vessel was built with five pairs of wing tanks and four center tanks. cargo tanks 1P, 1S, 2C and 5C were penetrated in the bottom at the impact or later, before the cargo could be salvaged. The vessel was loaded with 5024 metric tons of cargo and fully loaded to the mark. The cargo was heavy fuel oil with a density of 0.93 and a pour point of +12C. Due to the high density of the cargo, not all the cargo tanks were volumetrically full. After a few days of heavy weather when no actions could be taken, the vessel was lightened, pulled off the ground and towed to a repair yard. From the initial amount of cargo, the amount transferred to the lightening vessel and the amount recovered at the repair yard, it was concluded that about 150 tons had escaped to the sea.

2.3 Due to the wave height reported to be 4 to 5 metres, a considerable surging action took place in the damaged tanks, extracting more cargo from these tanks than otherwise would be the case. The high viscosity of the oil when being cooled down by the sea water created some dampening of the surging effect. Figure 1 [not included] illustrates the loading condition of the tanks which were damaged. The quantity lost equivalents as effective low water level of 1.4 metres below the main level which seems likely in consideration of the wave height and the dampening effect of the solidyfing oil.

2.3 The amount leaking was considerably reduced by the fact that only one of the damaged tanks was full. The outflow was also reduced by the fact that the master immediately after the accident decided not to attempt to take any soundings in the tank in order not to admit air into the tanks and thereby increase the outflow. The heavy weather would also have made sounding the tanks as extremely dangerous task.

2.4 The maximum underpressure that would have been needed in the tanks in order to retain the oil despite the heavy surging from the wave action would have been 2.05 metres of water column. Operation with fully closed deck openings and P/V valves with an opening pressure of 2 metres wg would have been fully feasible in this case. In case of no heavy sea, the cargo would have been contained by a maximum underpressure of 0.7 metres wg. The decision in the case to leave the tanks closed with an estimated underpressure in the ullage space of about 0.5 metres wg on the on the other hand have saved some additional 60 tons from escaping into the sea during the prevailing weather conditions.

2.5 The oil that escaped could not be collected due to the heavy weather and no cleanup work was possible before the approaching winter due to the freezing of the sea. The oil was therefore spreading along a large area of the eastern coastline. The cleanup work went on for the major part of the spring and summer season of 1987 and became expensive due to the circumstances. The total cost for shoreline and on-site cleanup was about 30 million Swedish crowns (equivalent to about 5 million USD). A considerable portion of the oil has however sunk and is still laying on the seabed, likely to require further cleanup activities.

2.6 It seems for a study of this accident that the use of a vaccum system to retain oil in damaged tanks would have worked very efficiently. The tanker is typical of a large fleet of coastal tankers operating in the coastal and sensitive waters worldwide.

Tank number 1P 1S 2C 5C
Tank Innage 5.38 5.77 7.47 6.29
Draught at tank inc 0.5 m rise of bow 6.15 6.15 6.30 6.60
Press of cargo, sg = 0.93, P/V vlvs closed at 0.5 m wg 4.50 4.87 6.45 5.35
Press of sea incl 1.4 m wave effect 4.75 4.75 4.90 5.20
Diffential [at tank bottom] -0.25 0.12 1.55 0.15
Outflow of oil, m3 0 8 127 7

source CTX
type A
volume 150 tons
material F

The above description of this spill was taken from the Swedish submittal to IMO basically arguing for regulation to use vacuum to reduce spillage.

  1. It is one of the very few spills for which we have hydrostatic balance data, let alone calculations.
  2. It is the only spill of which CTX is aware in which there was an attempt to quantitatively estimate the oil lost to wave pumping.
  3. The interesting use of and comments on vacuum.

The Swedes have assumed that the damage went no higher than the flat bottom. In reality, grounding damage is never confined to just the flat bottom, it always goes at least a little was up the sides, especially for wing tanks. This is one reason why their hydrostatic calculations are slightly lower (127 m3 versus 161 m3) than the unrecovered cargo measurements.

The Gulf of Bothnia is one of the few places in the world that has virtually no tide, which is the reason why tide was not a factor and not mentioned in the Swedish report.

This ship must not have had an inert gas system. If the ship had had an IG system, there was no way that a vacuum could have been establised in the top of the damaged tanks. It is refreshing to see a master who recognized the importance of maintaining that vacuum. But all P/V valves leak a little bit. CTX suspects that some of the loss abscribed to wave pumping was actually the result of P/V valve leakage.

IOPCF says 150 to 200 tons.

ERC says 58,800 gallons.