Nov 202014
 
FVliberty

Sooner or later the chances were that someone was going to be killed aboard the 13.32 metre Irish registered FV Liberty. Given the long list of safety issues uncovered by Ireland’s Marine Casualty Investigation Board, MCIB, and the fact that an earlier incident involving an injury went unreported so the conditions that resulted in the death of a seafarer on 14 February 2013 went undetected, tragedy was inevitable and preventable.

In port at Dunmore East prior to the voyage, one of the trawl nets on the vessel, supplied by the owner, was swapped for a used net supplied by the skipper. The skipper’s net had been kept in storage and had not been used since October 2012. The net was apparently changed because
it was deemed to be more suitable for the intended fishing grounds  where the vessel was going to fish. Continue reading »

FacebookTwitterGoogle+StumbleUponRedditLinkedInDeliciousEmailShare
Nov 172014
 
ffg

If you don’t look after your lifeboat

It won’t look after you

We want to adapt this for a seven minute video and mobile app to be distributed free of charge to seafarers. PSC surveys to hand out or show during their visits, shipping companies to their fleets, P&I Clubs to their members, seafarers organisations to their members. Video will undoubtedly be more effective at getting the messages across, however, it does cost a lot more to make to a professional standard. We need to raise a modest $5,000 to cover the cost of producing the video. If you’d like to help save seafarers lives, and address a leading cause if seafarer fatalities then check out the project here.

 

Listen to the podcast

ist engineeros.jpg

We’ll call them Paul and Butch. Not their real names but they were real people. They can no longer tell you their story.

Paul was Third Engineer and Butch was an Ordinary Seaman aboard the Lowlands Grace when she anchored in ballast nearly 12 miles off Port Hedland, Australia on the morning of the 6th of October, 2004 to wait for a cargo of iron ore for China. Continue reading »

Nov 102014
 
bridge

Dropped objects don’t come much bigger than the Jefferson Avenue Bridge over the Rouge River about 10 kilometres southwest of Detroit, Michigan. It is not especially unusual for ships to hit bridges but fairly rare for bridges to hit ships,only fairly rare because it has happened before under similar circumstances – an impaired bridge operator.

About 0212 on May 12, 2013, the bulk carrier Herbert C. Jackson was en route to deliver a load of taconite pellets, a type of iron ore, to the Severstal ore processing terminal in Dearborn, Michigan. As the vessel approached the Jefferson Avenue Bridge, the master slowed and sounded one long and one short blast of the ship’s whistle to notify the bridge tender of the approach and request a bridge opening. While waiting, the master brought the vessel to a near-complete stop. About 0205, the master saw the bridge begin to open, and when the drawbridge was fully open and green lights were visible on each bridge section, he increased speed.

Continue reading »

Nov 072014
 
capblanche

Looking out of the window was not really an option for the pilot conducting the 28, 372 GRT containership Cap Blanche on the Fraser River, British Columbia, Canada, on 25 January this year. With fog reducing visibility to 150 metres he could not even see the bow of the 221.62 LOA vessel, but he did have his trusty portable pilotage unit, PPU, which he relied upon exclusively for navigation and connected it to the vessel’s AIS. But the AIS had a secret, one which put Cape Blanche on the silt at the river’s Steveston Bend.

The accident report from Canada’s Transport Safety Board brings to light a little known aspect of navigation by GPS yet one that might not have led to the grounding had the pilot not been essentially left to his own devices even when his actions conflicted with the vessel passage plan.

The PPU had a predictor function that projects the vessel’s future position by performing geometric calculations based on the vessel’s current rate of turn, position, heading, course over ground, COG, and speed over the ground, SOG. The COG and SOG are derived from GPS values that continuously fluctuate, even when the vessel maintains constant speed and course due to inherent errors and inaccuracies in the GPS. To stabilize these values, a GPS smooths these inputs to provides the user with a more stable COG and SOG.

One can often see the GPS fluctuations on a GPS-equipped tablet computer or smartphone.

Continue reading »

Nov 062014
 
scorpio

At 1521 on 3 January 2014 the Liberia registered liquefied gas carrier, Navigator Scorpio, ran aground on Haisborough Sand in the North Sea. The vessel was undamaged by the grounding and there were no injuries or pollution; 2.5 hours later, it refloated on the rising tide. The investigation found that the vessel ran aground in restricted waters after the officer of the watch had become distracted and lost positional awareness. The passage plan was incomplete and the significant effects of wind and strong tidal streams had not been properly taken into account.

Given the proximity to danger, appropriate navigational techniques were not applied and the bridge manning was insufficient. Additionally, weaknesses in the crew’s navigation capability had been identified during an audit of the vessel, however, follow up actions were not sufficient to prevent the grounding.

Continue reading »

Oct 262014
 

This podcast has a special place in MAC’s heart – it was the very first one ever broadcast. At the time we did not have a video production capability or a recording studio so the sound quality may be least than ideal but the lessons remain very current.

An exhausted Captain; single watch-keeping; a warm, cozy bridge at night; the heavy traffic of the Kiel Canal, and pirated navigational software. If you think that sounds like a recipe for disaster, you’d be absolutely right.

Listen To The Podcast

Stripes Continue reading »

Oct 162014
 

cmvavenueMurphy’s Law is more consistent than the Law of Gravity: If something can go wrong it will, and at the most critical moment. An unresolved engine problem, a contined waterway and an overtaking maneouvre bought together the 12,878 dwt Antigua and Barbuda-flagged CMV Conmar Avenue with the 88,669 dwt Netherlands-flagged Maersk Kalmar on the Outer Weser between fairway buoys 29 and 31 in the Fedderwarder Fairway, Germany.

The joint accident report from Germany’s BSU and Antigua and Barbuda’s Inspection nd Investigation Division, emerges a few weeks after video of what appears to be a somewhat similar siuation in the Suez Canal circulated on the internet. That partiular incident remains under investigation. Continue reading »

Oct 142014
 

wildeMost of us like to push the limits often because our experience tells us we can do so safely. Just because we can does not mean we should, a lesson from Ireland’s Marine Casualty Investigation Board in its report into the collision between two ro-ro ferries: Stena Europe and Oscar Wilde in the port of Rosslare.

At 17.45 on 26 October 2012  as Stena Europe approached Rosslare the vessel’s master took over as OOW and the Mate/Master briefed the bridge team on the intended approach to the berth. The OOW called Rosslare Harbour Port Control and confirmed a wind direction of 028° (T) and a wind speed between 29 and 35 knots. The fact that the vessel had the use of only 3 out of four engines was not reported to port control.

Another source for information on wind speed and direction Information of wind speed and direction was also available from an instrument installed by Stena Line on the breakwater; this transmitted the information by
radio to displays on the bridge wings of the Stena Europe. Continue reading »

Oct 072014
 
seascope

Top: Before, note six freeing ports. Bottom: Gone, with the help of a doubler plate. Result=sinking

Delaying making permanent repairs to the 40-year old oceanographic vessel Seaprobe, failing to comply with his vessel’s safety management system nor complying with mandatory load limits proved to be a poor decision by the ship’s owner. Fortunately no-one died but one crewmember was seriously injured, two lightlyinjured and the vessel was lost.

Built in 1974 and owned by Fugro-McClelland of Galveston, Texas, according to the Equasis database, US flagged and in class with the American Bureau of Shipping, the 818 tonne Seaprobe was a converted fishing vessel originally named Northern Empire. It was one of two geotechnical drillships operated by  Fugro-McClelland in the Gulf of Mexico.

The US National Transportation Safety board adds: “Contributing to the accident was the owner’s failure to comply with the vessel’s safety management system and mandatory load line regulations”.

In a Jones Act law suite it is alleged that “despite the years of notice, Fugro continually failed to properly maintain the vessel which subsequently led to the eventual sinking”. Fugro’s recruitment advertisements say “Fugro has a strong commitment to health, safety and the environment. ”

One may bear those statements in mind while reading the last days of the Seaprobe:

In late December 2012, the Seaprobe departed its oceanographic research operations off the coast of South America, en route to Morgan City, Louisiana. While in transit, the captain and crew noted seawater entering the engine room where the exhaust pipe for the starboard main engine exited through the engine room overhead.

The captain told investigators that he maneuvered the vessel to minimize water on the deck in the area of the water entry. Also, the crew made temporary repairs including fastening thin sheets of metal to the starboard-side exhaust trunk where the original metal was wasted. The exhaust trunk housed exhaust pipes from machinery in the engine room and ran forward horizontally above the deck to the stack, just aft of the deckhouse. To address the flooding, Saprobe diverted the Seaprobe to the Gulf Marine Repair (GMR) shipyard in Tampa, Florida.

On 4vanuary, 2013, the Seaprobe arrived at the GMR shipyard, where the vessel underwent temporary repairs. After the accident, the uS Coast Guard learned that shipyard personnel used doubler plates to temporarily repair the starboard main engine and generator exhaust pipes housed within the exhaust trunk. Portions of the exhaust trunk’s bottom plate were not renewed after this plate, some of its metal wasted, was removed to gain access to damaged sections of the exhaust pipes within the exhaust trunk.

The port engineer―a representative of the vessel owner who was at the GMR shipyard during the repair work―told investigators that he did not direct shipyard personnel to replace the bottom plate because he wanted to allow for further examination of the exhaust pipes during the vessel’s next scheduled drydock.

Shipyard personnel had asked the port engineer if he wanted the exhaust trunk repaired or left open, and, if the latter, the personnel would install a doubler plate over the six starboard-side freeing ports near the bottomless exhaust trunk to protect it from seawater. The port engineer believed, as an interim measure, “if the freeing ports were closed it would improve protection against seawater contact in that area.”

Shipyard personnel installed the doubler plate over the six freeing ports and left the exhaust trunk open to the atmosphere on the bottom.

An accident was now waiting to happen.

Seaprobe left Tampa about 1400 on 16 January, 2013, and continued toward Morgan City. Based on vessel draft estimates provided by the captain, the freeboard near the deck over the engine room measured 0.3 metres to 1 metre.

During the west-northwest transit toward Morgan City on the evening of 17 January, the Seaprobe experienced north winds at 26 knots, with gusts up to 34 knots, and seas of five metres. The high seas and the fact that six of the Seaprobe’s freeing ports were closed caused seawater to collect on deck. This water made its way into the open bottom of the exhaust trunk and downflooded into the engine room.

Seaprobe lost power to its main engines and generators about 2000 that evening. The chief engineer discovered water in the engine fuel and in the starboard-side day tank. He also saw water entering the engine space near the exhaust trunk. Initially he kept up with the flooding by using the oily water separator to pump the water to a holding tank.

About 2030, the chief engineer  restored the portside generator and about 2130 brought the portside main engine back online. However, over the next few hours, he noted that the flooding increased and used the bilge pumps to remove the incoming water. One of the deckhands also saw the engine room flooding and told investigators that every time the Seaprobe rolled in the high seas, a large amount of water entered the engine room on the starboard side.

About 0200 the next morning, 18 January, with the Seaprobe listing to starboard and upto  metre of water in the engine room, the chief engineer and the captain determined that the bilge pumps could not keep up with the flooding. They then shut the engine room’s watertight doors to prevent progressive flooding of adjacent areas. Shortlylater, they contacted the US Coast Guard.

Two liferafts were launched, the twelve crew boarded hem and were rescued by two USCG helicopters.

About 1800 that evening, the owner reported to the Coast Guard that the aircraft could not locate the vessel. The Seaprobe is presumed to have sunk sometime around 0315 on 18 January.

Safety Provisions

Although the Seaprobe was not required to undergo US Coast Guard inspection, two other safety strategies were in place: a load line certificate and a safety management system. Says the NTSB: “Because the vessel owner did not adhere to these strategies, they did not prevent the sinking”.

The Load Line Certificate

The American Bureau of Shipping (ABS) issued an international load line certificate for Seaprobe on 29 March, 2012.  Under the load line regulations in 46 Code of Federal Regulations (CFR) Part 42, the operators of the vessel should have told ABS before they installed doubler plates over the freeing ports and departed port without repairing the bottom of the exhaust trunk. Having a load line is contingent on weathertight integrity of the vessel above the freeboard deck. Blocking the drainage of water from the deck and operating at sea with openings to the engine room were changes to the conditions of the vessel requiring ABS to revisit the vessel and reassess the load line assignment. ABS, after examining the condition of the vessel and the operator’s proposal for permanent repairs, would have decided whether to allow the vessel to depart Tampa in the condition that it was.

CFR 42.07-55 (b) states that load line certificates may be canceled due to conditions such as closed freeing ports and wasted exhaust trunks. requiring ABS to revisit the vessel and reassess the load line assignment. ABS, after examining the condition of the vessel and the operator’s proposal for permanent repairs, would have decided whether to allow the vessel to depart Tampa in the condition that it was. CFR 42.07-55 (b) states that load line certificates may be canceled due to conditions such as closed freeing ports and wasted exhaust trunks.

These mandatory measures were not adhered to.

SMS

Seaprobe’s owner owner held a valid International Safety Management Code Document of Compliance and Safety Management Certificate issued by ABS. One of the four objectives of a safety management system is to ensure compliance with the mandatory rules and regulations when developing procedures for the safety management system.

The safety management system requires documents for vessel maintenance procedures, which are used to verify that all company vessels are maintained in conformity with relevant rules and regulations. Seaprobe owner’s failure to discuss the wasted exhaust trunk and closing of the freeing ports with ABS meets the definition of nonconformity under the safety management system.

The Cause

In its report the US NTSB concludes that the probable cause of the flooding and subsequent sinking of the Seaprobe was the decision of the vessel owner to delay making permanent repairs to the starboard-side exhaust trunk and covering six of the vessel’s freeing ports, leaving the Seaprobe susceptible to downflooding from boarding seas.

Contributing to the accident was the owner’s failure to comply with the vessel’s safety management system and mandatory load line regulations.

Seaprobe was subject to regulations found in 46 Code of Federal Regulations Part 3, which require the vessel owner to verify in writing every 2 years that the vessel operates as an oceanographic research vessel and is therefore not subject to inspection as a passenger, freight, or offshore supply vessel. An oceanographic research vessel of less than 300 gross tons does not require inspection.

As is all too often the case teh drive to minimise costs and maximise profits put seafarer’s lives on the line.

Download the NTSB report here.