Report issued on the Sea Empress   

The British Marine Accident Investigation Branch issued its report on the grounding of the Sea Empress on 16 July. Late 15 Feb., 1996, the Sea Empress (Liberian-registry, 147,273-dwt, 274-meter/900-foot long single-hull tanker built in 1993 by Astilleros Espanoles S.A. in Spain; owned by SeaTankers's Alegrete Shipping Co. and operated by Acromarit (U.K.) Ltd. with 28 Russian crew) ran aground, 180 meters/600 feet off St. Ann's Head near Milford Haven, Wales. Ruptured cargo tanks spilled about 72 million liters/19 million gallons of oil, or 72,000 tons. On 21 Feb., 1996, the ship was refloated with 12 tugs. It was on charter to Dreyfus Energy and was carrying 140.0 million liters/36.75 million gallons of North Sea light crude in 17 tanks from Hound Point, Scotland, to a Texaco Inc. refinery in Wales. The report faults pilot error for the grounding as the immediate cause, but found several underlying problems. There were no tugs at the port able to assist the ship, bad weather hampered clean-up and relations between the pilots and Milford Haven Port Authority was strained. The master and the pilot also failed to agree on a passage plan. The report recommends the port authority conduct better training, examination and management of its pilots. Acomarit (U.K.) should have its masters understand and follow its standing orders on pilots. And the British Marine Safety Agency, British Coastguard Agency and British Department of Environment, Transport and the Regions should ensure that all ships have spill plans in place and can access computerized information on structural integrity and damage calculations when an accident occurs. The report also suggests double-hulls extend to pump rooms and possibly engine rooms. Finally, MARPOL regulations on bottom raking should be tightened. John Prescot, secretary of the Environment, Transport and the Regions, has ordered an independent review of salvage agreements and the port authority. As a result of the report, the British Environment Agency has started criminal proceedings against the Milford Haven Port Authority and its harbormaster, Mark Clive Andrews. Both have received summonses. The authority is charged with being a nuisance to the public, as it "failed to regulate navigation" and prevent the spill under the British Milford Haven Conservancy Act of 1983. It also failed to provide proper pilotage under the British Pilotage Act of 1987. The authority and Andrews were also charged under Section 85 of the British Water Resources Act of 1991 for pollution. Andrews was also charged with being a nuisance to the public.   

 Port staff to be charged over Sea Empress   

Criminal charges are to be brought against senior harbour managers over the Sea Empress oil disaster last year.   

A string of errors and communications failures led to serious delays in trying to stem the flow of 72,000 tonnes of crude oil from the stricken ship. Huge stretches of the South Wales coast were contaminated.   

A highly critical report published yesterday prompted the Environment Agency to take the rare step of prosecuting both Mark Andrews, the harbourmaster who was on duty during the incident, and Milford Haven Port Authority. The decision to bring criminal charges underlined the degree of severity that the Environment Agency attached to the spillage, which killed or injured more than 7,000 seabirds. The penalty for conviction under the Water Resources Act is a maximum of three months in prison or a fine of £20,000 if the case is heard by magistrates, or a two-year sentence or unlimited fine if the case goes to a Crown Court. The charges allege that public nuisance was caused by failures to control the spillage.   

Friends of the Earth, which is still considering bringing a private prosecution against the Government, called for former ministers to be brought to court to explain the reason for shortages of equipment. It took six days to rescue the ship after she had been removed from rocks but ran aground a second time.   

Yesterday, the final report by the Transport Department's Marine Accident Investigation Branch made clear that there were widespread failures, starting with mistakes made by the pilot guiding the tanker into Milford Haven. There were insufficient tugs of the right power and manoeuvrability, and a lack of full understanding of the tidal currents in the area, the report concluded.   

After the initial stages of the accident in February 1996, "the incident was outside the scope of Milford Haven Port Authority's emergency plans", the report said. It added that the onshore management team "became too large and unwieldy to cope with a rapidly moving salvage incident" and suggested that the Government should have been advised by a commercial salvage expert.   

The 147,000-tonne Liberian-registered tanker had sailed from Scotland and was being guided into the port by John Pearn. The report said the pilot failed to take action to keep the vessel in the deepest part of the middle channel leading into the port. It added: "The pilot's error was due in part to inadequate training and

experience in the pilotage of large tankers."   

The report said that the standards of training and examination of pilots at Milford Haven were unsatisfactory. The investigation branch also said that there was a "confrontational relationship between the authority and the pilot" and this could not be "conducive to the safe operation of the port".   

Mr Pearn was found guilty of incompetence at a Milford Haven Port Authority inquiry last year and was demoted. But he successfully appealed against the decision and was able to resume working with large tankers.   

The long-term damage caused by the Sea Empress oil spillage is still being evaluated, but conservationists generally agree that it has not been as bad as feared. The main concern is the lingering effect on birds and fish life. 

MAIB Report of the Chief Inspector of Marine Accidents into the grounding and subsequent salvage of the tanker SEA EMPRESS

PART 1 - FACTUAL ACCOUNT

2. PARTICULARS OF VESSEL AND CREW

2.1 General Description

Name : SEA EMPRESS

Port of Registry : Monrovia, Liberia

Registered Owners : Alegrete Shipping, Co. Inc. Monrovia

Commercial Managers : Seatankers Management, Cyprus

Technical Managers : Acomarit (UK) Ltd, Glasgow

Built : Spain, delivered 1993

Length Overall : 274.30 metres

Breadth Extreme : 43.20 metres

Depth Moulded : 23.80 metres

Loaded Draught : 17.02 metres

Deadweight : 147,273 tonnes

Classification : Det Norske Veritas+1A1 Tanker for Oil

Main Engine : M.A.N. B&W Type 6S70 MC 2-stroke 6 Cylinder oil engine

Maximum power : 13,475 kW

Propeller : Single fixed pitch Right-handed

Steering Gear : EIMAR-WEIP 4 ram type

Service Speed : 14 knots

Manoeuvring Speed : 11 knots (loaded)

Minimum Speed : 4.5 knots (loaded)

Astern Power : 60% of Ahead Power

Rudder : Single plate semi-balanced

She is a single-hull, segregated ballast, suezmax crude oil carrier with the propelling machinery and accommodation superstructure in the conventional aft position. The main engine, which runs on heavy fuel oil at all times, can be controlled from the bridge, which is provided with a comprehensive outfit of navigation equipment. When the vessel arrived off Milford Haven all her Convention, Flag State and Class certificates were valid.

2.2   Cargo and Ballast Tanks and Heavy Fuel Storage Tanks

The hull is subdivided to provide fourteen cargo tanks, which include two slop tanks, with a total capacity of 164,156 cubic metres; the cubic capacity of each cargo tank is as follows:

No 1 Port and Starboard 5,310 cubic metres eachNo 3 Port and Starboard 11,796 cubic metres eachNo 5 Port and Starboard 4,017 cubic metres eachNo 6 Port and Starboard Slops 2,647 cubic metres eachNo 1 Centre 16,423 cubic metresNo 2 Centre 29,471 cubic metresNo 3 Centre 14,735 cubic metres

No 4 Centre 14,735 cubic metresNo 5 Centre 29,471 cubic metresNo 6 Centre 11,781 cubic metres

The cargo pump room is immediately forward of the machinery space, with the motors and turbines for the pumps on the after side of the bulkhead. All cargo pumping operations can be controlled and monitored from the cargo control room, which is on the navigating bridge deck adjacent to the combined wheelhouse/chart room.

The Fore and Aft Peak tanks and Nos 2 and 4 Port and Starboard tanks are dedicated ballast tanks with a total capacity of 52,495 cubic metres; the cubic capacity of each ballast tank is as follows:

Fore Peak 3,968 cubic metres

No 2 Port and Starboard 11,670 cubic metres each

No 4 Port and Starboard 11,700 cubic metres each

Aft Peak 1,787 cubic metres

Heavy fuel oil is carried in four wing storage tanks on the port and starboard sides, located outboard of the slop tanks and pump room, with a total capacity of 4,998 cubic metres.A general arrangement plan of the vessel, together with cargo, ballast and fuel oil tanks is shown at Figure 1(Image 73k).

2.3    The Crew

SEA EMPRESS was manned with a total crew of 27, all Russian nationals. The officers were the Master, Radio Officer, three deck officers, four engineer officers and the Electrical Officer. The ratings were the Bosun, Pumpman, six seamen, a fitter, three motormen and three catering ratings. All the officers and ratings were properly certificated or qualified. Two cadets were also carried.

2.4   The Master was aged 58 and had served as master of tankers since 1966. He joined Acomarit in 1993 and had been in command of SEA EMPRESS, apart from periods of leave, since the vessel was new. He had completed a course on the handling of very large tankers in Grenoble, France.

3. NARRATIVE

Times are UTC (Universal Co-ordinated Time)

Chart extracts showing the main positions of the vessel from the time of approaching the West Channel entrance to Milford Haven on the evening of 15 February to the final float off on 21 February are at

Figure 2 - Positions of Sea Empress on Thursday 15 February (Image 92k)Figure 3 - Positions of Sea Empress on Saturaday 17 February to Sunday 18 february (Image 100k)Figure 4 - Positions of Sea Empress from Sunday 18 February to final refloating on Wednesday 21 February (Image 97k)

The chart extracts are taken from the edition of British Admiralty Chart 3274, which was current at the time of the incident, and not the new edition published in November 1996.

3.1   SEA EMPRESS loaded her cargo of Forties light crude oil at Hound Point, in the Firth of Forth and sailed from there for Milford Haven on 13 February 1996. All the cargo tanks were filled to normal capacity, with recorded ullages on completion of between 1.08 and 1.40 metres. The vessel's Figure for cargo on board was 130,018 tonnes. The dedicated ballast tanks had been emptied except for about 890 tonnes in the Aft Peak tank, for trim purposes. A route via the English Channel to Milford Haven was decided upon and a sea passage plan was prepared to a position 4.5 miles south-southwest of the entrance to the West Channel to Milford Haven. This position was designated 'Way Point 20'.

Thursday 15 February

3.2   The three deck officers and the helmsman/lookouts were keeping the conventional four-on eight-off watch rota. The Master was not standing a watch. During the sea passage, on 15 February, a pilotage passage plan into Milford Haven was prepared. This plan started from 'Way Point 20' with a course almost coincident with the line of the Outer Leading Lights to the West Channel entrance and continued with further courses up the West Channel and thence into the port. The Tanker Check List and Pilot Card were also prepared for arrival at Milford Haven.

3.3   'End of Sea Passage' was at 1905 hrs on Thursday 15 February when SEA EMPRESS was 7.5 miles south-southwest of the West Channel entrance and steering a course of 022°. Both steering gears were running and the main engine was on bridge control. At 1910 hrs course was altered to 012° towards the boarding area requested by the pilot, which was four miles southwest of Saint Ann's Head. The wind was west-northwesterly force 4/5 and it was fine and clear. Engine speed was gradually reduced and the vessel was then manoeuvred to maintain a boarding speed of about 5 knots and a starboard lee for the pilot, who boarded at 1940 hrs. At this time SEA EMPRESS was 3.8 miles southwest of Saint Ann's Head on a heading of 010° and the main engine was on 'dead slow ahead'. The Bosun had cleared the anchors for letting go and the bow mooring party went forward, followed by the Third Officer who had escorted the pilot to the Bridge and would take charge forward.

3.4   On arrival on the bridge, which was manned by the Master, Chief Officer and helmsman, the pilot inspected the information on the Tanker Check List and Pilot Card, which he found in order. The arrival draught had been reported by the Master as 15.9 metres, with the vessel on an even keel. The pilot expected the tidal stream to seaward of the West Channel to be slack at this time, but there was still an outgoing tide running in the Channel, the predicted time of low water in the port being 2130 hrs. At 1944 hrs the pilot ordered 'full ahead' and a turn to starboard to a course of approximately 060°. At 1948 hrs the vessel was 2.75 miles southwest of the Channel entrance and working up to full manoeuvring speed.

3.5   At 1951 hrs the vessel was 2.3 miles southwest of the Channel entrance and the pilot could see both Outer Leading Lights and both Inner Leading Lights for the West Channel (see Section 4 for detailed description of the entrance to Milford Haven). At about 1955 hrs he ordered progressive changes of course to port, by about 5° at a time, until the vessel was on a heading of 035°. A position plotted by the Chief Officer at 2000 hrs showed the vessel to be one mile from the entrance to the Channel and about midway between the lines of the Outer and Inner Leading Lights (see Figure 2 Image 92k). The pilot was steering the course of 035° in order to make a judgement as to which way the tide was likely to affect the vessel. He was satisfied with the performance of the helmsman and had the impression that the course of 035° was being made good. The speed made good had increased progressively to about 10 knots.

3.6   According to published tidal information, the east-southeast running tidal stream to seaward of the Channel entrance was predicted to begin at about 2000 hrs. When SEA EMPRESS was two or three cables from the Channel entrance, the pilot saw from the changing aspect of the Outer Leading Lights that there was a set to the east. He ordered a course change of 5° to port. The vessel, which was now closing with the Mid Channel Rocks Light Buoy, started to turn to port. The pilot saw the Outer Leading Lights close and then open to the east (that is, the rear Leading Light appeared to move to the right of the front Leading Light) as the bows entered the channel. The Buoy was very close to starboard and the pilot now saw that the fixed red and fixed white lights of the Outer Leading Lights, the ones indicating the east side of the deepest water in the Channel entrance, were also open to the east.

3.7   After the bows had passed the Middle Channel Rocks Light to starboard there was a shuddering vibration, then a sound from the deck below of liquid being forced under pressure, accompanied by a strong smell of oil. The time was 2007 hrs (see Figure 2 Image 92k). The helmsman reported that the vessel was not steering and the pilot ordered 'full astern', both anchors to be let go and called out the four harbour tugs which were waiting to assist SEA EMPRESS to her berth. At 2008 hrs the Chief Officer recorded this order and the vessel's position, which was 1.4 cables northwest of Middle Channel Rocks Light. SEA EMPRESS, which was trimming by the head and taking a starboard list, continued to run ahead as the anchors were let go. There were further main engine movements and the vessel finally lost all way at approximately 2015 hrs after she grounded again in a position about 3.3 cables south-by-west of the Mill Bay Buoy (see Figure 2 Image 92k). The main engine was put to 'stop'. SEA EMPRESS was by this time

significantly trimmed by the head and had taken an 18° list to starboard with the starboard fore part of the deck awash and was heading in a northwesterly direction.

3.8   The engineers in the engine control room felt the vessel vibrate very heavily. The starboard list then caused a number of alarms to sound in the engine control room and both the port and starboard boilers shut down automatically due to the extent of the list. The main engine and alternators, however, remained fully operational with no alarm condition being recorded. The starboard boiler was re-started and the engine room and steering gear compartment were checked for damage and leaks. None were found.

3.9   In response to the pilot's request the four waiting tugs proceeded to SEA EMPRESS with all speed. By about 2023 hrs the tugs DALEGARTH and THORNGARTH had made fast their lines to the bow of the casualty and TITO NERI and STACKGARTH were instructed to standby off each quarter in case the crew had to be evacuated. However the Master decided that, because SEA EMPRESS appeared to be in no immediate danger, the crew would remain on board.

3.10   Initial reports of a damage assessment, undertaken by the casualty's crew, indicated that Nos 1 & 3 Starboard and No 6 Centre Cargo tanks together with Nos 2 & 4 Starboard Ballast tanks had been damaged. Some flooding of the pump room was also indicated. The casualty's initial angle of list, recorded as 18° to starboard at 2024 hrs was reduced to about 10° by running sea water into the two port side ballast tanks, Nos 2 and 4, under gravity from the damaged starboard side ballast tanks via the ballast main. An offer of salvage assistance by a consortium comprising SmitÊTak BV, Cory Towage Limited and Klyne Tugs (Lowestoft) Limited on the terms of Lloyd's Standard Form of Salvage Agreement "No Cure-No Pay" 1995 Edition (LOF95) was accepted by Acomarit, the managers of SEA EMPRESS, by 2305 hrs. (See Annex A for general details of marine salvage and LOF, and Annex B for general details of the Government's role in a major incident.)

Friday 16 February

3.11   By 0040 hrs on Friday 16 February STACKGARTH and TITO NERI were made fast to the stern of the casualty. The Master and the pilot agreed that SEA EMPRESS should be manoeuvred into the deeper waters of the 'pool' where she could be held by the tugs (see Section 4 for details of the 'pool'). It was also decided to reposition the anchors. These proposals were passed to the Signal Station at 0055 hrs. High water at Milford Haven was predicted to occur at 0342 hrs.

3.12   Commencing at 0127 hrs, the anchors were weighed and with the assistance of the four tugs and the casualty's engine and steering she was moved to the southwest into the 'pool' and onto a northeasterly heading. The port anchor was let go at 0211 hrs but as sea conditions were deteriorating slightly it was decided to let go the starboard anchor also. By 0427 hrs the two anchor cables had been laid out and the engine stopped (see Figure 3 Image 100k). The tugs were experiencing difficulties with their lines parting but they were reconnected and the casualty's position was maintained.

3.13   At this stage SEA EMPRESS was considered to be in a stable condition and under control. However, there was some debate within the Harbour Authority on the advisability of ordering the casualty out to sea as soon as possible but this idea was dismissed.

3.14   Representatives of the Marine Pollution Control Unit (MPCU) of The Coastguard Agency had began to arrive in the Milford Haven area from about 0030 hrs. MPCU staff boarded SEA EMPRESS with a local representative of Cory Towage and made an assessment of the casualty's condition. As a result of this assessment MPCU recommended to the Harbour Master that SEA EMPRESS should be held in position and lightened.

3.15   One of the first objectives of the salvors was to regain control of the pump room which was flooded to a level of about 5 metres and had a flammable atmosphere. Power packs, pumps and other equipment from the Marine Pollution Salvage Centre (MPSC) in Milford Haven were mobilised and at 0600 hrs began to arrive on board SEA EMPRESS. At 0800 hrs the tug ANGLIAN DUKE arrived and increased the number of tugs made fast to the casualty to five. The presence of a pilot was maintained on board. The tanker STAR BERGEN was on charter to the salvors for an anticipated ship-to-ship transfer of oil.

3.16   In the course of the morning various options for the movement of SEA EMPRESS were considered at meetings held on board and ashore. Shipboard discussions resulted in two options being proposed; either taking the casualty to sea or reducing the angle of list and the draught to bring her further into the Haven. A number of indicators pointed to the need for a lightening operation and preparations to lighten the casualty continued. The time chosen for the commencement of this operation was 1100 hrs on Saturday 17 February in order to take advantage of slack water and daylight. However, a forecast of a period of poor weather suggested the lightening operation might have to be delayed. The Salvage Master and a team from Smit Tak arrived at Milford Haven at 1130 hrs.

3.17   By 1530 hrs portable MPCU pumps and fans were in place pumping floodwater from the pump room and venting it in order to ensure the atmosphere was safe for entry. By this time the Smit Tak Salvage Master had made his assessment of the casualty's condition. The essential elements of the salvage plan chosen by the Salvage Master consisted of pumping out the undamaged centre tanks to a lightening vessel, transferring cargo from the damaged starboard tanks into the centre tanks using portable pumps and pressurising the damaged starboard tanks with inert gas. The objectives of this plan were to right the casualty, reduce its draught and remove cargo from the damaged tanks to prevent pollution. SEA EMPRESS would then be in a suitable condition to be taken to a berth alongside to complete total discharge of cargo.

3.18   Efforts by the salvors continued on board SEA EMPRESS, throughout the remainder of Friday, to achieve this end. Weather conditions deteriorated during the afternoon and evening, however the casualty's position continued to be maintained at anchor in the 'pool', with the use of tugs and the casualty's own main engine.

Saturday 17 February

3.19   Concerns over the deteriorating weather, forecasts of a gale and the anticipated difficulty of holding the casualty's stern to the weather were coupled with a desire to give a lee on the port side to assist the lightening operation. These resulted in a decision being made to turn the vessel during the slack water period of Saturday afternoon, 17 February, so that it headed seawards and with the anticipated gale on the starboard bow.

3.20   Two further tugs, ESKGARTH and ANGLIAN EARL, had arrived in Milford Haven by 0900 hrs and stood by. Pumping and ventilation operations in the pump room had continued and it was declared safe for entry at 1330 hrs.

3.21   Much of the efforts of Saturday morning were directed in the planning and preparation for the turning of the casualty. The operation to turn SEA EMPRESS commenced at 1400 hrs with the manoeuvre under the overall control of a pilot. A total of seven tugs, ANGLIAN DUKE, THORNGARTH, TITO NERI, DALEGARTH, STACKGARTH, ANGLIAN EARL and ESKGARTH, all had lines secured to the casualty. Difficulty was experienced in recovering the anchors of SEA EMPRESS and it was suspected that they were fouled. They could not be brought right 'home' and about one shackle (30 metres) of cable was left from the ends of the hawse pipes. As SEA EMPRESS swung she pointed seawards. The possibility of taking the casualty out to sea was raised by the pilot but the considered opinion of those ashore was that it would be too risky. The turn of SEA EMPRESS was completed at 1555 hrs with the casualty in the southwest corner of the 'pool'. Both of the casualty's anchor cables were then walked out until 11 shackles were streamed on each cable but they were still crossed (see Figure 3 Image 100k).

3.22   At 1700 hrs, with the casualty heading approximately west-southwest, it was decided to reposition the tugs in preparation for the forecast poor weather. During these manoeuvres SEA EMPRESS started moving to the west under the influence of the ebb tide. Efforts to arrest this movement by the tugs resulted in at least two tow lines parting. SEA EMPRESS continued to move beam on to the tide, in a westerly direction. In spite of the use of the casualty's main engine and the efforts of the tugs she grounded off Saint Ann's Head at about 1805 hrs. Both anchor cables of the casualty were then slipped in an attempt to free the bows and allow the stern to be towed up into the tide. The wind was reported as being from the west-southwest, force 6/7, (see Figure 3 Image 100k).

3.23   Due to the casualty's situation, including the loss of her anchors, a request for more tugs was made by the salvors. The tug DE YUE was reported as having departed Falmouth with an ETA at Milford Haven of 0800 hrs on Sunday 18 February. With 54 personnel on board the casualty it was also decided to evacuate non-essential personnel and this operation commenced at 1910 hrs by helicopter with a second lift at 1950 hrs. Further assessment of the situation led to a decision being made to secure the casualty by the bows in her present position by ballasting, so increasing her ground reaction. The casualty's ballast system was prepared and filling of the Fore Peak and No 2 Port Ballast tanks commenced; inerting of the damaged cargo tanks also began at 2245 hrs.

Sunday 18 February

3.24   At midnight the wind was reported as being from the west-southwest, force 9 with a heavy westerly swell. Filling of the ballast tanks continued into the early hours of Sunday morning. Weather conditions were deteriorating and were reported at 0200 hrs as 40 to 50 knot westerly winds; these increased later to 60 knots. Two further tugs, YEWGARTH and ELDERGARTH, arrived in Milford Haven at 0230 hrs and at 0240 hrs a further group of people was evacuated from the casualty by helicopter.

3.25   On the rising tide the casualty started to work in the seaway and generate noises which were associated with structural failure. A decision was made to evacuate the casualty and the remaining 14 persons on board were airlifted from her by 0440 hrs. All tugs, except ANGLIAN DUKE and ANGLIAN EARL, were released due to concern for their safety and the possibility of a fire or explosion on SEA EMPRESS. ANGLIAN DUKE and ANGLIAN EARL remained secured to SEA EMPRESS by 750 metres and 650 metres of wire respectively. The main lighting of SEA EMPRESS was seen to fail, and emergency power take over, at 0607 hrs. At about 0800 hrs the casualty floated off and drifted to the east across the channel.

3.26   SEA EMPRESS was again reported aground at 0840 hrs, in a new position, further southeast by Middle Channel Rocks, after having been driven across the West Channel and the 'pool' (see Figure 3 Image 100k). DE YUE, which had arrived shortly after 0900 hrs and had a pilot and a Klyne tug master on board, spent a period on trial manoeuvres at the stern of SEA EMPRESS. She then anchored in Dale Roads at 1140 hrs and prepared her towing gear. A small group of salvors were airlifted to SEA EMPRESS in order to handle tug lines. They also attempted to restore the casualty's main power but these attempts proved unsuccessful, even with radioed advice from the casualty's Chief Engineer. The conditions at the casualty were very poor with a westerly wind gusting gale force. DEÊYUE again approached the stern of SEA EMPRESS at 1520 hrs and a line was secured with the assistance of a helicopter. SEA EMPRESS was observed drifting in a northeasterly direction between 1500 hrs and 1600 hrs. ANGLIAN DUKE, TITO NERI and ESKGARTH were instructed to assist DE YUE who in turn requested guidance from the shore authorities on action to be taken but no positive instructions were received. The salvors left SEA EMPRESS by 1715 hrs aground in a position approximately 6 cables northeast of Middle Channel Rocks Light, but clear of the West Channel, at the northeast end of the 'pool' (see Figure 4 Image 97k).

3.27   Additional Smit Tak salvage crew, including a Senior Salvage Master, arrived in the Milford Haven area during the Sunday afternoon. It was accepted that there was little possibility of lightening SEA EMPRESS during the following day (see Photograph 1) SEA EMPRESS on Sunday 18 February (Image 35k).

Monday 19 February

3.28   At 0044 hrs, after having experienced great difficulties in maintaining position and threatened with grounding herself, DE YUE slipped her tow line and proceeded to an anchorage.

3.29   SEA EMPRESS was reported as moving towards Middle Channel Rocks at 0810 hrs where she later grounded (see Figure 4 Image 97k). MPCU staff, salvors and some of the casualty's crew boarded SEA EMPRESS between 0830 hrs and 1015 hrs. Main electrical power was restored, an assessment of the casualty's condition was made and approximate calculations showed that 120,000 tonnes of cargo remained on board. However, the pump room was found to be flooded to sea level and more tanks found to be damaged, including the Fore Peak. Pumping out of the pump room was re-commenced and extra equipment was ordered from ashore to handle the additional damaged tanks.

3.30   Later on the Monday afternoon SEA EMPRESS refloated. The five tugs in attendance were unable to control her movements and she moved west across the main channel. At about 1715 hrs, the casualty was in line with the main exit from the 'pool', facing inwards and a request was made to those ashore (MPCU and Milford Haven Port Authority) for permission to take the casualty out to sea. Although this was agreed to, it was not fully communicated to those on board. However, to take the casualty to sea would require the use of her main engine but as it had not been prepared for use, the question was academic.

3.31   SEA EMPRESS was being swept towards Saint Ann's Head where she grounded at about 1815 hrs (see Figure 4 Image 97k). During the early evening efforts were made to introduce inert gas into damaged tanks but with the casualty's hull generating sounds, suggesting that hull failure and a possible explosion might occur, all persons on board evacuated onto attending tugs by 2245 hrs.

3.32   Various options were again discussed by the salvors for the following day. It was decided that the damaged tanks should be sealed on deck at low water to create an air lock in each tank then at the following high water it was anticipated that the casualty would refloat. Once afloat she would be deliberately grounded south of the Angle Buoy (see Figure 5 Image 95k), between the East and West Channels, in order to gain control of her.

Tuesday 20 February

3.33 Two further tugs, PORTGARTH and VIKINGBANK, arrived at Milford Haven at about 0600 hrs on Tuesday 20 February. Members of the salvage team re-boarded SEA EMPRESS, which was still aground off Saint Ann's Head, at 0840 hrs to be joined later by some of the casualty's crew. On several occasions during the following two hours grounding noises were heard throughout SEA EMPRESS and significant losses of oil were observed but the casualty's own engineers prepared the main engine for use. Although the salvors continued to prepare tanks for refloating the casualty their preparations were not sufficiently advanced to allow them to pressurise the damaged cargo or ballast tanks at low water as they had originally intended.

3.34   SEA EMPRESS began to show signs of movement at 1645 hrs and the main engine was tested and found satisfactory. The salvors were given permission to bring SEA EMPRESS to Herbrandston Jetty if the refloating attempt proved successful.

3.35    Eight tugs were in place by this time and over the period to 1936 hrs various main engine and tug movements were made in attempts to refloat SEA EMPRESS. These efforts were unsuccessful. Some of the tugs which were pushing on the port side experienced difficulties due to the oil film on the hull of the casualty causing them to skid and to be unable to apply full power (see Figure 4 Image 97k).

3.36   The salvors judged it would be imprudent to remain on board SEA EMPRESS overnight and all persons evacuated by 2245 hrs but three tugs remained in close attendance. Meetings were held late into the night and plans, produced by the salvors for the following day's operations, were discussed by the concerned parties. It was again decided to increase the casualty's buoyancy, and reduce her list, by pressurising the ballast tanks. The pressures to be used would be higher than for the previous attempt.

3.37   The tanker, TILLERMAN, was made available for lightening purposes after assurances to her managers from MPCU that any costs would be underwritten by HM Government.

Wednesday 21 February

3.38   Salvors, crew, and MPCU staff had all boarded SEA EMPRESS by 0815 hrs on Wednesday 21 February. The casualty's position was estimated as being about 200 metres west of her position on Tuesday. Attempts to pump out the pump room were proving unsuccessful with flood water remaining at sea level.

3.39   As No 2 Port ballast tank was still intact, portable pumps were used to pump out this tank. Preparations were made to pressurise the damaged ballast tanks with compressed air in anticipation of low water. The Aft Peak was also pumped out. Cargo tanks were also pressurised with inert gas at normal system pressure. Shortly before low water at 1350 hrs it was observed that the rate of cargo leakage from the casualty had increased.

3.40   The tug ARILD VIKING had arrived during the previous evening and thirteen tugs were in position by 1630 hrs. Two pilots were on board and the line of on board command was set out for the anticipated refloating operation (see Photograph 2) SEA EMPRESS and tugs on Wednesday 21 February. (Image43k)

3.41   The bows of SEA EMPRESS started swinging to port, under the influence of the tugs, at 1735 hrs. Although the casualty's main engine failed to start when requested it was eventually started, and subsequently operated, from the emergency controls. The casualty then floated free at 1800 hrs, was moved astern into the main channel (see Figure 4 Image 97k) and proceeded to the selected berth alongside Herbrandston Jetty on the north side of the Haven. First mooring lines were ashore at 2155 hrs.

3.42   SEA EMPRESS had lost a total of 71,800 tonnes of cargo during the accident and her salvage.

PART II - ANALYSIS OF INCIDENT (INITIAL GROUNDING)

5. INITIAL GROUNDING

Seaworthiness of the Vessel

5.1   SEA EMPRESS operates under a continuous hull and machinery survey regime with Det Norske Veritas (DNV) the classification society. The latest DNV update of vessel information prior to the incident was dated 1 February 1996 and showed no survey items overdue and no conditions of class. The last annual survey inspection was carried out in early February in Tarragona, Spain. No defects were found but three conditions of class were imposed; these related to the calibrating of testing equipment, the production of certain type approval certificates and the provision of sufficient sealing blanks for the inert gas system and crude oil washing tank openings.

5.2   The vessel has a Shipboard Safety and Environmental Protection (SEP) Management System Certificate and a Shipboard International Safety Management (ISM) Certificate issued by DNV.

5.3   For the arrival at Milford Haven the statutory Tanker Check List was completed. The question asking whether the auxiliary steering gear was in good working order had been answered in the negative by marking the NO box. Investigation of this confirmed that the Master marked the NO box to indicate that the vessel was not provided with such equipment; it was not a requirement since two independent primary units were fitted. It is therefore concluded that prior to the accident SEA EMPRESS was in a fully seaworthy condition with no defects which might have contributed in any way to the causes of the initial grounding.

On Board Communications

5.4   After this accident happened, and indeed after the BRAER accident in 1993, concerns were raised about vessels manned by crews, often of more than one nationality, with no common mother tongue. This was not the case with the crew of SEA EMPRESS.

5.5   All the crew on board SEA EMPRESS were Russian. However, the Master's written Standing Orders, the log books, the movement books and all other written entries on the vessel's papers were in English as is the case on the majority of Liberian registered vessels. The Master's spoken English was good, that of the rest of the vessel's bridge team was not but their understanding of nautical terminology in English was quite adequate

and gave the Inquiry no reason for concern. The pilot had no difficulties in two way verbal communication with the Master. The pilot's helm and course orders were all repeated by the helmsman in English, and then in Russian for the benefit of the Master and officers. After each order had been carried out this was reported by the helmsman to the pilot, again in English and then in Russian. The pilot was quite satisfied with the way the helmsman was steering.

5.6   Based on the interviews carried out with the pilot, the vessel's Bridge Team and the Third Officer, who was standing by the anchors when the grounding happened, there were no communication difficulties which might have contributed in any way to the causes of the grounding.

Timing of Entry

5.7   SEA EMPRESS was required to be alongside the Texaco Refinery Jetty not later than the predicted low water time of 2130 hrs. The preferred latest time to embark the pilot was 1930 hrs. This was to allow one and a half hours from the time of the pilot boarding the vessel to being alongside and making fast. The pilot did not board SEA EMPRESS until 1940 hrs and according to Port records the vessel entered the West Channel about 15 minutes later than any previous vessel of that size, although this fact was unknown to the pilot at the time. However, this delay was not a critical factor because the extra time provided by the period of slack water off the Jetty before the tide turned was additional time in hand, and in any case there would have still been enough depth to get alongside, even at low water. Therefore, even though the entry was slightly later than preferred, there was no justification in aborting and waiting until the next tide. Commercial pressure played no part in this either because there were more than adequate crude oil stocks at the Refinery so SEA EMPRESS's cargo was not required urgently.

Courses Steered Prior to the Grounding

5.8   Prior to SEA EMPRESS grounding, gyro courses were being steered as is usual and 'nil' gyro error was recalled by both the Master and the Chief Officer. This is borne out by the previous entries in the log book and the pilot's own impression. Before the pilot boarded and took the con the course on the chart was 012°, this was the course to the requested boarding position. Five charted positions from 1910 hrs to 1940 hrs show that this course was made good. At the time the pilot took the con the course being steered, as recalled by the Master, was 010°. The turn to starboard which the pilot then ordered was by helm orders, initially "hard-a-starboard" as recalled by the Master and helmsman.

5.9   The pilot told the Inquiry he then ordered a course to steer which was "in a northeasterly to easterly direction". The bridge was not fitted with a course recorder and as there are inevitably some differences between the recollections of the pilot, Master, Chief Officer and helmsman as to what this and the subsequent course orders actually were these have been examined by the Inquiry. The initial course was 065° according to the Master and Chief Officer and "about 065° to 070°" according to the helmsman.

However, a course of 060° was laid off on the chart from the 1948 hrs position plotted by the Chief Officer so this has been accepted as the first course ordered by the pilot.

5.10   The Master said that the pilot's next course order was 070°, which is similar to the helmsman's recollection referred to above. The Master, Chief Officer and helmsman all recalled the next orders as progressive course changes to port of about 5° at a time, to 025° according to the Master and Chief Officer. However the pilot said that he turned to 035° and, taking into account the positions at 1956 hrs and 2000 hrs and his later evidence, this has been accepted as the most likely course steered towards the Channel entrance.

5.11   This course of 035° was steered until about two to three cables (less than two minutes) before entering the Channel, when the pilot became aware that there was a set to the east. There is no doubt that his order then was to steer a course of 030°, not an order for port helm which would have been the appropriate action in this case. The vessel was still turning to that heading when the bows entered the Channel, very close to the Mid Channel Rocks Light Buoy. It was only at this very late stage that the pilot realised prompt action was needed. He said that he was reluctant to order 'hard-a-port' because in a previous accident when this action had been taken control was lost and the vessel grounded off Saint Ann's Head. It was also possible that the starboard quarter would hit the Buoy if port helm was applied. The pilot walked quickly to the starboard side of the wheelhouse to see how close the Buoy was and the first contact of the fore end with the rocks would have happened only seconds later.

5.12   The Inquiry has checked the details of the accident referred to by the pilot to see if there are parallels between that and the SEA EMPRESS incident. The case the pilot cited was the grounding in 1984 of the loaded crude oil tanker MATCO AVON, although he was not personally involved in the incident. On that occasion visibility had deteriorated as the vessel approached the West Channel entrance. She was found to be passing on the wrong side of the Mid Channel Rocks Light Buoy. Full port helm and full astern pitch was applied. Subsequent full counter helm was ineffective, the swing to port could not be arrested and this led to the grounding. MATCO AVON had a right-handed controllable pitch propeller and the effect of full astern pitch with full port helm caused the loss of control.

Planning and Monitoring of the Approach

5.13   The pilot had intended to and initially did approach the Channel entrance within what he termed "the cone of safety" (see Figure 2 Image 92k). The theory of this is that if an inbound vessel is kept within the cone formed by the lines of the Outer Leads and the Inner Leads extended seawards, then she will leave the point of the cone, the intersection of those lines, very near the middle of the Channel entrance and in the deepest water. Other pilots made similar approaches, so the pilot ordered the course of 035° and at 2000 hrs the vessel was indeed within the 'cone' with a mile to go to the entrance. If the course of 035° had been made good the vessel would probably have safely entered the Channel. The pilot (perhaps correctly) said that the tidal stream off the entrance was slack when he

boarded, but according to the sailing directions it was predicted to start running east-southeasterly at 2000 hrs, just 20 minutes later.

5.14   The pilot also told the Inquiry that he was steering 035° to make a judgement as to which way the tide was likely to affect the vessel. This suggests prudence, because it is not unusual for tidal streams to turn earlier or later than predicted. But to make such a judgement when between two sets of leads, with neither in line ahead and with just five or six minutes to go before reaching the 'point' of the cone in the Channel entrance required a high standard of vigilance, not only by the pilot but also by the Master. The effect of the start of the flood (east-southeast running) tidal stream was not detected early enough and, when it was detected, the wrong action was taken to counteract it by giving a small course change order of 5° instead of a helm order of at least 10° or 15°. Any large vessel which is just two minutes or so from a restricted channel entrance and needs a prompt course correction must have immediate and effective helm applied in order to achieve this. The giving of a course change order instead meant that the amount of helm applied and the rate of turn to the new course was left entirely to the choice of the helmsman, whose priority on this occasion was to avoid an excessive swing.

5.15   Clearly the pilot was alert to the danger that tidal streams do not always run as and when predicted and the tidal stream was predicted to start as the vessel entered the Channel. The surest indication of the start of a cross set would have been gained from the use of either the Outer or Inner Leading Lights. Had these leads been used, preferably the Outer Leading Lights, any deviation from the intended track would very soon have been apparent.

5.16   The prior preparation by the vessel's team of their own pilotage passage plan, as recommended in the IMO Bridge Procedures Guide and as required in Acomarit's Navigation Manual issued to their masters, was a regular routine for each port. The plan for Milford Haven provided for an approach to the West Channel entrance along the line of the Outer Leads, as described in the published sailing directions. The vessel was already following this approach, on a course of 022°, when the Master received the request to go to the position southwest of Saint Ann's Head to embark the pilot. This meant a deviation to a northerly course.

5.17   The pilot boarding position for inbound vessels of the size and draught of SEA EMPRESS is stated in the sailing directions as being "at various positions SW of Middle Channel Rocks Light appropriate to vessel's size and weather conditions". Southwest of Middle Channel Rocks Light is generally west of the line of the outer leads, so it follows from this that the request from the pilot launch to the Master to go to the north to embark the pilot should not have been unexpected. Once the pilot had boarded and since the vessel had now left the line of the Outer Leads, the Master should have clarified with the pilot whether or not the vessel was to resume her original line of approach. He should also have shown the pilot the vessel's pilotage passage plan and discussed it. It is a well recognised duty of a bridge team to monitor the pilot's conduct of the navigation, a task made difficult if not impossible without prior knowledge of the pilot's intention and an agreed plan by which the pilot's action can be tested. Indeed, the Navigation Manual

required that the vessel's plan was to be finalised "after consulting the pilot as to his intended route and actions"; it also required the Master to "discuss the ship's proposed movements fully" with the pilot and "satisfy himself that arrangements and conditions were suitable, before allowing the ship to proceed".

5.18   The Master had been to Milford Haven before, in about 1990 on a larger tanker, and in his opinion the correct way to approach the West Channel was to use the 022° Leading Lights; hence the prepared pilotage passage plan. He assumed the pilot intended to return to the 022° Leading Lights sometime before reaching the Channel entrance, although he had not asked him when this would be. It is speculated that had the vessel's prepared plan been implemented the accident would probably not have occurred. It can be seen from the reconstruction (see Figure 2 Image 92k) that the vessel did meet the line of the 022° leads before the Channel entrance, but at a point so close to it that only a substantial application of helm to port would have kept the vessel within the limits of the deepest water. The Master should have realised that the 5° course alteration ordered by the pilot was far short of the action needed, but he appears to have placed total confidence in the pilot's judgement. The Navigation Manual required that "a strict watch must be kept to verify that the ship is handled and navigated correctly". The Master evidently only became concerned when the vessel had crossed the centre 022° leads and was obviously going to pass very close to the Mid Channel Rocks Light Buoy.

5.19   The Chief Officer, who had not been to Milford Haven before, said that his usual duty in pilotage waters was to monitor the vessel's position and report the progress of the passage to the Master. However, he did not ask the Master what the pilot's passage plan was. He did plot the position of the vessel six times after the pilot took the con and prior to the initial grounding, the last position being at 2006 hrs as the bridge was passing the Mid Channel Rocks Light Buoy. He took a further position just after the initial grounding, at 2008 hrs.

5.20   Prior to arrival off Milford Haven all the deck officers had attended a pre-arrival meeting with the Master on the bridge. This meeting, a routine procedure, was about all aspects of the vessel's arrival in the port, mooring at the jetty and discharge of the cargo. Discussion of navigational matters was included. The Chief Officer, like the Master, was satisfied that the vessel's pilotage passage plan included the correct line of approach to the Channel entrance. The Chief Officer was therefore under the impression, as was the Master, that the pilot was going to join the line of the Outer Leads at some stage before the Channel entrance was reached.

5.21   When the pilot ordered the succession of 5° course changes to port the Chief Officer was "not quite happy" since he had anticipated that the pilot would take the vessel nearer to the line of the outer leads before starting the turn to port. However he said nothing because he had been told that the preferred latest time to embark the pilot for berthing on that tide was 1930 hrs, whereas he had not boarded until 1940 hrs. The Chief Officer therefore assumed that the pilot was taking a compromise shorter course to the entrance to save time. Notwithstanding this, there was a failure of the Master to discuss the prepared vessel's approach plan with the pilot and finalise it with him, as instructed in

the Navigation Manual. Whatever was decided should then have been made clear to the Chief Officer. This should have been done before the pilot took the con and need only have taken a few minutes.

5.22   Even if the vessel's prepared plan had been shown to the pilot, it is probable that he still would have preferred his own approach and that the Master would have accepted it. The pilot initially approached the Channel entrance steering what he considered to be a prudent course within the 'cone of safety' explained above. He did not explain his plan of approach to the Master after he boarded, probably because it was not the normal practice. Three of the pilot's senior colleagues, all of whom had an involvement in training him, said that they themselves did not normally tell Masters of inbound deep draught vessels which sets of leads they proposed to use and what specific courses they intended to steer towards the West Channel entrance. Many Masters, not just at Milford Haven, share a similar attitude, saying in effect "she's all yours pilot".

5.23   Merchant Shipping Notice M.854 titled 'Navigation Safety' provides advice on the planning and conduct of passages. It includes the following recommendation - "ensure that the intentions of a pilot are fully understood and acceptable to the ship's navigational staff". This Notice was published in 1978 and continues to be current. Although it is not addressed to pilots, the pilot of SEA EMPRESS should have been aware of it, because written guidelines for the pilotage authorisation oral examination includes a knowledge of M Notices relating to pilotage. A knowledge of relevant M Notices is also in the oral examination syllabus for a Class 1 Certificate of Competency, which the pilot had obtained shortly before joining Milford Haven Pilotage Limited.

Effect of Squat

5.24   The deepest part of the entrance to the West Channel, which is near the middle, is only about 160 metres wide. Since the tidal streams run across the line of approach, even the weakest current strengths are enough to set an approaching vessel across and out of the deepest water unless adequate compensating action is taken. Any approach to this narrow entrance therefore calls for vigilant navigation and this is particularly so in the case of a large loaded vessel which has to be kept within the limits of the deepest water. Such was the case with SEA EMPRESS. The beginning of the east-southeast running tidal stream set the vessel some 115 metres towards the eastern limit of the Channel entrance.

5.25   SEA EMPRESS was drawing 15.9 metres when the pilot ordered 'full ahead'. By about 1955 hrs her engine speed was at 'full manoeuvring' and as she approached the Channel she was making approximately 10.5 knots. During the ten minutes prior to passing the entrance buoys she averaged 10.3 knots. It is estimated that at the time of the grounding one minute later the speed over the ground was approximately 10 knots. There would have been some squat as she left the deep water. As well as the slight reduction in speed, the squat would have caused an effective increase in the draught.

5.26   Squat effect increases with speed, initial draught and the closeness of the vessel's bottom to the ground. The phenomenon has been known about for many years and masters, navigating officers and harbour pilots are generally aware of it. An appreciation of squat is included in the written guidelines for the oral examination for pilotage authorisation. It was left to the individual trainee pilots to ask to see these guidelines; they were not issued to them as a routine practice. Research has been carried out and papers have been published, mainly by Dr I W Dand and Dr C B Barrass who have formulated means for estimating what the effect is likely to be for various conditions. Using the methods of estimation devised by Dr Dand, Dr Barrass and in "A Note on Ship Interaction Effects" prepared by the City of London Polytechnic, results of 0.65 and 0.83 metres were obtained. It is concluded that the increase in SEA EMPRESS's draught due to squat effect would have been about 0.75 metres immediately prior to the initial grounding.

5.27   Acomarit's Navigation Manual stated - "When navigating channels of restricted depth, the effect of increased draught due to squat must be taken into account. It should be borne in mind that the effect increases with speed and is greater when the channel is also restricted in width". The Navigation Manual also required the echo sounder to be run and this was being done, although the recorder was not being run continuously. With squat effect, the draught as the vessel crossed the 30 metre depth contour close to and west of the Mid Channel Rocks Light Buoy would have increased to about 16.6 metres. The height of tide at this time was 2.2 metres. The minimum charted depth over the rocky ground at the southwestern extremity of Middle Channel Rocks (which is just outside the charted limit of the Channel) is 13.7 metres, so there would have been less than 16 metres of water over those rocks at the time of the initial grounding. It is beyond doubt that the vessel, having narrowly missed the Mid Channel Rocks Light Buoy, 'clipped' these rocks, causing the bottom damage and breaching along the starboard side tanks. The position plotted at 2008 hrs confirms this. A depth survey carried out since the accident has found no uncharted obstruction or loss of charted depth in the West Channel entrance and there were more than adequate depths for SEA EMPRESS within the limits of the deepest water in the Channel entrance.

Action taken in Relation to the Initial Grounding

5.28   Consideration was given at an early stage in the inquiry to making an interim recommendation that pilots should use the Outer Leading Lights when bringing in vessels to the Haven via the West Channel. However, it was pointed out by MHPA that the Outer Leading Lights had been established only for use by VLCCs, that is vessels in excess of 200,000 deadweight tonnage (dwt). These vessels inevitably time their arrival off the Channel when the tidal stream is running fairly strongly to the northwest. MHPA suggested that it would not be appropriate to utilise this approach when the tidal stream is running in an easterly direction, which is often the case when tankers of less than VLCC proportions are entering the port.

5.29 It was therefore decided that this proposed recommendation should be held in abeyance until a computer simulation could be developed and exercises run to prove, or

otherwise, the validity of such a recommendation. In conjunction with MHPA and a firm of consultants a simulator programme was developed. A number of simulations employing the characteristics of two tankers, one of 88,425 dwt and one of 133,855 dwt (slightly smaller than SEA EMPRESS), were run which demonstrated the feasibility of approaching the West Channel entrance on the line of the Outer Leads in both ebb and flood tidal stream conditions. Using this approach line has the added benefit that with the configuration of the Outer Leading Lights those on board are given a clear indication should the vessel be straying from the deepest part of the channel. Following on from this successful exercise an interim recommendation was made (see Section 22).

PART IV CONCLUSION

21. FINDINGS

The Inquiry carried out by the Inspectors has covered in great detail the many events which occurred, and throughout the Inquiry the Inspectors received complete co-operation from all parties concerned. Needless to say, with so many people having played a part in the incident and with approximately 100 interviews having been carried out by the Inspectors, there has been some conflict in the evidence. This is not surprising when those people interviewed were having to remember details of many different events which were spread over a period of six days. The Inspectors did not feel that anybody was trying to mislead them, it was simply a case that where no detailed records of events were kept as they took place the recollections of various persons differed. In such cases the Inspectors have had to use their own knowledge and experience to come to conclusions as to what they consider was the most likely chain of events.

To reach their findings the Inspectors have also had to rely to some extent on supposition, but this was consistent with good, unbiased investigatory work. I consider that the findings given in this section of the Report are a true reflection of the actual events which occurred and I support their conclusions.

To give readers of the Report a fuller understanding of the various findings, inserted after each finding are cross references to the appropriate paragraphs in either the main body of the Report or the Annexes where that subject is discussed.

The findings of this Inquiry are as follows.

21.1   SEA EMPRESS first grounded on the western edge of Middle Channel Rocks, 230 metres west of Middle Channel Rocks Light, in a depth of approximately 15.9 metres.

21.2   The immediate cause of the grounding was pilot error, namely his failure to take appropriate and effective action to keep the vessel in the deepest part of the Channel.

21.3   The pilot's error was due in part to inadequate training and experience in the pilotage of large tankers.

21.4    The pilot would have appreciated at an earlier stage in the approach that the vessel was being set to the east if it had been the normal practice to use the Outer Leading Lights for all approaches to the West Channel.

21.5   When the effect of the set was noticed, the Master failed to appreciate that the action then ordered by the pilot would not be adequate.

21.6   The pilot and the Master had not discussed and agreed a pilotage passage plan, as a consequence neither the Master nor the Chief Officer knew what the pilot's intentions were.

21.7   The Master failed to follow the standing orders of his Managers with respect to pilotage matters.

21.8   SEA EMPRESS had no known deficiencies and was in a seaworthy condition prior to the grounding.

21.9   All certification for the vessel and her crew was valid in accordance with the Flag State and Convention requirements.

21.10   The pilot and the ship's Bridge Team were adequately rested before starting their respective duties.

21.11   There were no communication difficulties between the pilot and the Bridge Team which might have contributed in any way to the causes of the grounding.

21.12   The anchors were ready for use, with the anchor party forward, before the vessel entered the West Channel.

21.13   The use of an escort tug in the approach to the West Channel would not have avoided the initial grounding, but it might have avoided the second grounding.

21.14   The standards of training and examination of pilots at Milford Haven are unsatisfactory and in need of improvement.

21.15   There is a confrontational relationship between Milford Haven Port Authority and the pilots. This cannot be conducive to the safe operation of the port.

21.16   Although considered to be an important part of best "practice" in safe operations the fact that the port radar installation was not operational did not contribute to the initial grounding.

21.17   All of the organisations involved with the salvage of SEA EMPRESS had an interest, commercial or otherwise, in seeing that the salvage operation was concluded promptly, at minimum risk to human life and with minimum damage to the environment.

21.18   The damage control efforts of the crew, to correct the list of SEA EMPRESS following the initial grounding, were well considered and effective.

21.19   Initial contractual salvage arrangements between the managers of SEA EMPRESS and the salvors were completed quickly and effectively.

21.20   The decision to move the SEA EMPRESS with a draught of 23.5 metres from the position of her grounding to deeper waters, in the early hours of Friday 16 February, was prudent and allowed the casualty to be secured in a position where a proper assessment of her condition was possible.

21.21   The emergency command/control organisation of Milford Haven Port Authority and of the Marine Pollution Control Unit was set up very soon after the initial grounding of SEA EMPRESS and the Marine Pollution Control Unit's owned salvage equipment was promptly available.

21.22   The salvage consortium mobilised their initial response personnel and resources promptly and efficiently.

21.23   The possible requirement for lightening tonnage was recognised early in the incident and a suitable commercial charter was promptly arranged.

21.24   The initial inspection of the casualty by the crew, and that subsequently carried out by the Marine Pollution Control Unit and their advisers, in the early hours of Friday 16 February, did not fully identify which cargo tanks were ruptured.

21.25   The decision to hold SEA EMPRESS in the 'pool' was understandable and reasonable until the difficulties of holding the casualty were demonstrated on Saturday evening.

21.26   The decisions throughout the salvage operation not to send the casualty to sea were correct.

21.27   The option of bringing the casualty up to Herbrandston Jetty on Saturday 17 February was not fully explored. The draught of the casualty could have been reduced sufficiently to have done so, although this would have had the possibility of leading to the release of some additional cargo.

21.28   The casualty's technical managers had a contract with Lloyd's Register of Shipping, Ship Emergency Response Service (SERS). Due to a lack of accurate data on the damage to the casualty this facility could not be used to its full potential in the early stages of the accident. However, in the latter stages of the salvage operation this facility proved invaluable.

21.29   The flooding and gassing of the pump room delayed any early lightening operation and was a factor in the escalation of the incident.

21.30   The large increase in the draught and list of the casualty due to the rupture of the empty starboard segregated ballast tanks in the initial grounding was a major factor in the escalation of the incident.

21.31   A number of non-essential personnel were allowed to board the casualty in the early stages of the salvage without consideration being given to the associated risks and they were ill-equipped for their evacuation which was later necessary.

21.32   The salvors were diverted from their main task in order to attend meetings ashore early in the incident, partly due to a lack of effective representation of the Marine Pollution Control Unit and Milford Haven Port Authority on the casualty.

21.33   The decision on Saturday 17 February to turn the casualty to face the wind and sea was correct based on good seamanship practices and the information available. The

principal reason for the loss of control of the casualty after the turn was that not enough consideration was given to the effect of the tidal stream and the casualty was positioned further south and across the tidal stream.

21.34   Once the casualty had been swept aground in the evening of Saturday 17 February and had slipped her anchors and cables, it was recognised that insufficient tug capacity was available on scene to hold her in position in the 'pool'.

21.35   The two large tugs under the control of The Coastguard Agency should have been mobilised and moved towards the scene of the incident soon after the initial grounding.

21.36   The salvors placed too much reliance on the total nominal bollard pull available and did not take account of the weather conditions, towage equipment and manoeuvring capabilities of each tug. Powerful AHTS vessels would have been more suitable to contain the situation and the salvors should have taken early steps to contract some of these types of vessels from the spot market.

21.37   The tug DE YUE, although very powerful and closest to the scene after the crisis on the night of Saturday 17 February, was not best suited, by the very nature of her design and her limited manoeuvring characteristics, to the task she was given.

21.38   There were no serious communication difficulties between the salvors' representative on DE YUE and her crew. However there was a difference of opinion, between the tug's Master and the representative of the salvors on how the tug should best be positioned. This disagreement was a factor which led to the failure of DE YUE to contribute to the salvage operations.

21.39   The decisions to totally evacuate the casualty at various times during the salvage operation were all justified due to a real risk to safety of those on board.

21.40   After evacuating the casualty early Sunday morning personnel were in need of rest. Sufficient extra key personnel although mobilised had not arrived by that time to allow for two team working.

21.41   A significant opportunity to gain control of the casualty in the evening of Monday 19 February was lost due to insufficient preparations being made against the foreseeable likelihood of the casualty floating off.

21.42   There was a failure of communication, between the salvors and the representatives of the technical managers of SEA EMPRESS, concerning the possible preparation of the casualty's main engine for the operations on the evening of Monday 19 February.

21.43   The fact that permission was granted for the casualty to be taken to sea on the evening of Monday 19 February without all the facts being made known to those concerned in the decision making process reflects on the poor quality of communications between the Marine Pollution Control Unit advisors on board the casualty and those in authority ashore.

21.44   The refloating attempt on Tuesday 20 February failed, because the tidal current held the casualty firmly against Saint Ann's Head Shoal. Sufficient tug power could not be assembled and attached to the casualty in order to pull her away from the shoal in opposition to the tidal current.

21.45   The tow-off on Wednesday 21 February succeeded because on refloating, the casualty's draught was such that she was clear of the shoal and the tugs were able to turn her end on into the tidal stream which lessened the load on the hull and allowed them to tow her clear of Saint Ann's Shoal.

21.46   The most important factor in the escalation of the incident was the lack of full understanding of the tidal streams within the 'pool' and its immediate vicinity.

21.47   The principal responsibilities of some persons within the command/control organisation ashore for dealing with the salvage operation were not clearly defined and this led to some confusion.

21.48   After the initial stages the incident was outside the scope of Milford Haven Port Authority's emergency plans.

21.49    The full involvement of the Marine Pollution Control Unit in a salvage incident should have been allowed for in the Milford Haven Port Authority's emergency plans.

21.50   Key technical personnel of the Marine Pollution Control Unit should not have been diverted from their primary tasks to brief the media at important stages of the salvage.

21.51   The Marine Pollution Control Unit did not have enough staff dedicated to the salvage operation to cope with all their responsibilities.

21.52   The mechanisms for approving salvage plans by the Marine Pollution Control Unit and Milford Haven Port Authority could not keep pace with the salvage process.

21.53    Notwithstanding the secondment of the Ministry of Defence Salvage and Mooring Officers as advisors to the Marine Pollution Control Unit, it is considered that the Government should have been advised by a commercial salvage expert.

21.54   The use of helicopters, whether military or commercial, for transporting personnel and equipment, proved to be of great value during many stages of the salvage.

21.55    When the Marine Pollution Control Unit were unable to obtain readily the services of any vessel considered suitable for lightening the casualty, in the position where she grounded on the night of Tuesday 20 February, they were justified in underwriting potential commercial losses of an owner in order to ensure the availability of its vessel.

21.56   The onshore management team became too large and unwieldy to cope with a rapidly moving salvage incident, did not have a clear authoritative leader and communications between the team and the salvors was poor. The salvage incident would have been better managed by a small command team, acting as a single unit, with a clear leader and fully operational on a 24 hour basis.

21.57    The Marine Pollution Control Unit's Overall Commander would have been better placed in Milford Haven in direct contact with the casualty and other key members of the onshore management team rather than remaining at the Marine Emergency Operations Room in Southampton.

21.58   The Marine Pollution Control Unit's National Contingency Plan is deficient in that it does not deal clearly with their involvement in the salvage of a vessel within harbour waters, does not define some key terms used which makes it difficult to positively interpret, and does not define how an escalation to a higher stage is accomplished.

21.59  During the night of Tuesday 20 February there was genuine confusion as to whether the Government were about to intervene. The powers of intervention under the Merchant Shipping Act were not invoked at any time during the incident, although

evidence suggests that the possibility of intervention was used to influence the salvage.

21.60   Reputable salvors were engaged and they were clearly expending effort under difficult circumstances. The formal use of powers of intervention or direction would therefore have been inappropriate, either by the Government or Milford Haven Port Authority.

21.61   The salvors were reacting to the developments which were occurring, rather than anticipating them, which indicates a weakness in their strategic planning of the salvage.

21.62   There was no major failure of any machinery, equipment or system on the casualty which contributed to the outcome of the incident.

21.63   Due to the complexity and sophistication of the machinery systems on board SEA EMPRESS, the salvors experienced difficulty in operating on board generators without the assistance of the casualty's crew.

21.64   The casualty is estimated to have lost about 2,500 tonnes of oil because of the initial grounding and from then until she grounded again on Saturday 17 February the oil lost was negligible.

21.65   A total of 71,800 tonnes of oil was lost to the sea and 58,200 tonnes were recovered from the casualty. The total oil lost was an accumulation of individual large oil losses which occurred when the casualty was stranded over low water periods. No evidence could be found to show that oil was deliberately discharged into the sea.

21.66   It is probable that oil pollution would have been avoided in the initial grounding if SEA EMPRESS had been constructed to the double hull design.

21.67   The initial grounding caused bottom raking damage which was 36% greater than that allowed for in the current International Convention for the Prevention of Pollution from Ships (MARPOL).

22. RECOMMENDATIONS

Based on the Inquiry into the whole incident and the findings of the Inspectors, a number of recommendations are being made. If implemented, these recommendations should

prevent recurrence of a similar grounding and help towards a successful outcome to any future salvage operation which might be necessary.

Some matters considered in the course of the Inquiry call for attention and have therefore led to recommendations even though, upon examination, they did not prove to bear upon the incident. It follows that the recommendations below must not be read in isolation as an indication of causative factors.

To give readers of the Report an understanding of the reasoning behind the recommendations, inserted after each one are cross references to the appropriate findings which led to that recommendation or where there is no finding the appropriate paragraphs in the Report.

In the course of the Inquiry four interim recommendations were made. The first of these, which was addressed to the Milford Haven Port Authority, was issued on 20 September 1996. It was:

"Trials should be conducted with a number of large tankers to prove the practicability of making an approach to the Western Channel entrance from sea along the line of the 022 Outer Leads, when the tidal stream to seaward of the entrance is flooding, that is setting in an easterly direction.

Following completion of successful trials consideration should be given to making it the normal practice for inbound vessels under pilotage to approach the West Channel entrance along the line of the Outer Leads, unless there are valid reasons for not so doing."

Milford Haven Port Authority have accepted in principle this interim recommendation by proposing to conduct further simulation trials in advance of any practical trials involving large tankers.

The other three interim recommendations were addressed to the Shipping Policy Directorate of the Department of Transport and were issued on 22 November 1996. They were made at that time as it was appreciated that for action to be taken on them changes to primary legislation might be required and a Merchant Shipping Bill was currently before Parliament. The recommendations were:

"To clarify the status of the Marine Pollution Control Unit's National Contingency Plan by giving it formal recognition in the Merchant Shipping Act."

"To consider clarifying the scope of the Secretary of State's intervention powers to include directions to a Harbour Authority, Harbour Master or pilot under section 137 of the Merchant Shipping Act."

"To review the current powers available to the Marine Pollution Control Unit to charter ships, aircraft and other equipment quickly in the case of a pollution incident, including the possibility of standing commercial arrangements to allow the immediate release of such vessels, aircraft and equipment in the event of such an incident."

The further following recommendations are now made and addressed to the person or body who in my opinion is most fitted to implement them:

Milford Haven Port Authority

1.   The Pilotage Authorisation Committee should amend the qualifying requirements for authorisations to perform pilotage on vessels in excess of 30,000 dwt. The requirements for each of these authorisations should be based upon a minimum number of trips under instruction from another pilot, of which at least half are undertaken at night and at least half are inward trips from sea.

2.    The Pilotage Authorisation Committee should improve the standards of examination of pilots. There should be an examination prior to the granting of any additional authorisation, not just initial authorisation to perform pilotage on vessels up to 30,000 dwt. Each examination should be in two parts, an oral part conducted ashore followed, if successful, by a practical part conducted on board one or more vessels when the candidate should be required to demonstrate his competency in pilotage to the satisfaction of the examiner.

3.   Consideration should be given to the use of simulators as an additional means for both training and examining pilots.

4.   The port radar surveillance system should be returned to a fully operational state and be provided with a continuous recording facility. It should be continuously monitored by a trained operator, fully instructed as to the type of vessel and circumstances when its navigation is to be monitored. In such a case, the intended track of the vessel must be known by the radar operator.

5.   Pilots should be instructed to ask after boarding to see any pilotage plan prepared by the vessel. A plan to be followed, taking the vessel's own plan into account, should be discussed and agreed with the Master and then notified to the port radar operator. The level of detail of the agreed plan, which should either be in writing or drawn on the chart, should be appropriate for the particular pilotage to be carried out.

6.   The boarding position off Milford Haven for pilots should be such that it allows sufficient time to agree the passage plan with the Master of the vessel and sufficient searoom to allow the vessel to be lined up for the agreed approach.

7.   Reforms should be introduced in the management of the pilots. In particular they should consider abolishing their wholly owned subsidiary, Milford Haven Pilotage Limited, so that the pilots become direct employees of the Port Authority and managed by them on a day-to-day basis.

8.   A comprehensive tidal stream survey should be conducted along the West Channel from the entrance buoys to a position on the line joining West Blockhouse Point and East Blockhouse Point, including the waters in the immediate vicinity of the Channel. The information obtained should be provided to all who require it.

9.   A comprehensive revision of the Authority's Emergency and Pollution Plans should be undertaken utilising modern techniques for hazard identification and risk assessment. The revised plans should recognise the role of the Marine Pollution Control Unit (MPCU) and be compatible with their National Contingency Plan. They should also specify clearly the role of pilots on a casualty during a salvage operation and establish a clear procedure for the briefing and debriefing of pilots.

10.   The consideration being given to the employment of an escort tug at Milford Haven should be advanced further.

11.   Consider and discuss with the Hydrographer of the Navy an amendment to the Admiralty Sailing Directions "The West Coasts of England and Wales Pilot", (Approaches to West Channel, paragraph 5.89 lines 29-32), to make it clear which part of the Channel the subsidiary red leading lights indicate.

Acomarit (UK) Ltd

12.   Steps should be taken to ensure that the Company standing orders to Masters on pilotage matters are understood and complied with. In particular, bringing the vessel's prepared pilotage passage plan to the attention of the pilot after he boards and discussing and agreeing with him a plan to be followed, taking the vessel's plan into account. The level of detail of the agreed plan, which should either be in writing or drawn on the chart, should be appropriate for the particular pilotage to be carried out.

Marine Safety Agency/Department of Transport

13.    National minimum standards of pilot training and examination in the UK should be prepared. It is noted that the Sub-Committee on Standards of Training, Certification and Watchkeeping (STCW) at the International Maritime Organization (IMO) has in its work programme an agenda item to consider developing provisions for the training and certification of maritime pilots and VTS personnel.

14.   Procedures should be developed and implemented for the effective monitoring of Competent Harbour Authorities' standards of training and examination of pilots. The involvement of District Marine Safety Committees should be considered for this purpose.15.    To encourage the revision by the International Maritime Organization (IMO) of Regulation 26 of MARPOL 73/78 (as amended) to include the requirement that all oil tankers of 5,000 tonnes deadweight or more have as part of their Shipboard Oil Pollution Emergency Plan pre-arranged, prompt access to computerised, shore-based damage stability and residual structural strength calculation programmes.

16.   To encourage the revision by the International Maritime Organization (IMO) of Regulation 13 of MARPOL 73/78 (as amended) to include the requirement that every oil tanker of 5,000 tonnes deadweight and above shall be fitted with a double bottom over the area of the pump room. Consideration should be given to extending the requirement to include double bottoms over the area of the engine room. The height of the double bottom to comply with the existing regulation 13F.3(b).

17.   To encourage the International Maritime Organization (IMO) to increase the existing MARPOL regulations covering the extent of bottom raking damage in the light of the SEA EMPRESS grounding.

The Coastguard Agency/Department of Transport

18.   To review commercially available systems which exist for receiving significant quantities of oil from a casualty without the need of the services of a lightening tanker. Should any system be considered to be of potential value in the operations likely to be undertaken by the Marine Pollution Control Unit, it should be included in their salvage/transfer equipment stockpiles.

19.   To revise the guidelines covering the use of Intervention Powers to ensure that, in an incident where competent salvors have been appointed, the powers of intervention are used for the primary purpose of assisting the salvors to identify and expedite the salvage plan which best meets the wider public interest.

20.   To review the effectiveness of the Marine Pollution Control Unit's role with regard to the salvage of a casualty and their relationship with the salvors. In particular, to consider the appointment of a specialist (a Marine Casualty officer) who would represent the Marine Pollution Control Unit on board a casualty, report directly to the Overall Commander on salvage matters, liaise with the Salvage Master, and approve and monitor the plans of the salvor.

21.   To review those aspects of the Marine Pollution Control Unit's National Contingency Plan which deal with casualty, salvage, and the relationship between the National Contingency Plan and the emergency plans of port authorities.

22.   To review the procedures for, and the priorities given to, the use of equipment, services and personnel of other government departments and agencies, including those of the military, by the Marine Pollution Control Unit during a marine emergency in order that there successful use, as was the case in the SEA EMPRESS incident can be ensured in the future.

23.   To allocate technical staff dedicated to the task of briefing the media during major incidents, in order that no operational technical staff are distracted from their primary tasks.

24.   To review the procedures for the mobilisation and deployment of tugs under the control of the Coastguard in emergencies which threaten (whether perceived or actual) loss of life or major pollution to the UK coast.