NTSB Safer Seas Digest 2021 2022 10

N A T I O N A L T R A N S P O R T A T I O N S A F E T Y B O A R D
SAFER SEAS Digest

2021
Lessons Learned from

Marine Investigations
NTSB/SPC-22-01
Mission
Making transportation safer by
conducting independent accident investigations,
advocating for safety improvements, and
deciding pilots’ and mariners’ certification appeals.
The NTSB is an independent federal agency charged by Congress with investigating

every civil aviation accident in the United States and significant events in other modes of
transportation—railroad, highway and transit, marine, pipeline, and commercial space.
We determine the probable cause of the accidents we investigate and issue safety
recommendations aimed at preventing future accidents.
In addition, we conduct transportation safety studies and coordinate the resources of the
federal government and other organizations to assist victims and their family members who
have been impacted by major transportation disasters.
A Message from the Chair
I
am proud to present the 2021 Safer Seas Digest on Pausing to reflect on a year’s worth of investigations Though the circumstances vary, our mission is the
behalf of the dedicated experts in the NTSB Office presents a unique opportunity to consider the same for every investigation we lead: to determine what
of Marine Safety. Their work over the past year meta-issues threatening safety on our waterways. happened and issue evidence-based recommendations
is represented in the pages that follow — work that In 2021, these issues included the following: to prevent similar events from occurring in the future.
contributes greatly to public confidence in the nation’s It is in pursuit of this mission that the NTSB issued
• Vessel stability
marine transportation systems. dozens of safety recommendations to all parts of the
• Containing engine room fires
marine industry in 2021.
The knowledge we gain from NTSB investigations is • Icing and severe weather
hard-won. The 31 marine accident investigations we • Risk management and project planning But stakeholders at all levels must implement our
closed last year are no exception; each investigation • Cargo preparation and securement recommendations to ensure safety. I hope the pages
was precipitated by a “major marine casualty,” which • Teamwork that follow inspire you to do just that.
can include the loss of human life. • Effective communication
• Standard operating procedures
This was the case with the commercial fishing Sincerely,
• Transiting in narrow channels
vessel Scandies Rose, which claimed the lives
• Distress communications and preparations
of 5 crewmembers when it sank off the coast of
for abandonment
Sutwik Island, Alaska, in 2019. The Scandies Rose
• Identifying navigational hazards
tragedy contributed to our decision to include
• AIS data input for towing operations
“Improve Fishing Vessel Safety” on the NTSB’s
• Continuous monitoring of unmanned vessels
2021–2022 Most Wanted List of Transportation Safety Jennifer Homendy
• Sufficient handover periods
Improvements. NTSB Chair
The US Coast Guard is integral to the NTSB’s marine
We also investigate major marine casualties that result
investigations. Our relationship is an outstanding
in significant property damage. This category will
example of government collaboration focused on
forever be associated with one of the most expensive
saving lives and improving safety. Every accident
marine accidents in history: the 2019 capsizing of the
presented in this report was supported in a variety
656-foot vehicle carrier Golden Ray.
of ways by the men and women of the Coast Guard,
Fortunately, no one died in the Golden Ray accident — and my sincerest thanks go out to every one of them
but that doesn’t mean it was without human risk. Two who assisted us this year. The Coast Guard units that
crewmembers suffered serious injuries and 4 others worked with the NTSB in these accidents are listed on
were trapped for nearly 40 hours in the engine room. page 86.
NTSB SAFER SEAS DIGEST 2021
2 Lessons Learned from Marine Investigations
SAFER SEAS Digest

Abbreviations
AB able seaman
ABS American Bureau of Shipping
AIS automatic identification system
ARPA automatic radar plotting aid
ATB articulated tug and barge
CPP controllable pitch propeller
DOTD Department of Transportation and Development
ECDIS electronic chart display and information system
ECR engine control room
ECS electronic charting system
ENC electronic navigation chart
EOT engine order telegraph
EOW engineer on watch
EPIRB emergency position indicating radio beacon
GPS global positioning system
HMR Hazardous Materials Regulations
ICW Intracoastal Waterway
IMDG Code International Maritime Dangerous Goods Code
LNTM Local Notice to Mariners
mph miles per hour
MT metric tons
NOAA National Oceanic and Atmospheric Administration
NOBRA New Orleans-Baton Rouge Steamship Pilots Association
NTSB National Transportation Safety Board
NWS National Weather Service
OD oil distribution
On the cover: The Golden Ray during Back cover: Fire aboard the OS ordinary seaman
rescue operations (see page 4). Höegh Xiamen (see page 48). PLB personal locator beacon
PPU portable pilot unit
Ro/Ro roll-on/roll-off
rpm revolutions per minute
SMS safety management system
VHF very high frequency
VTS Vessel Traffic Service
Lessons Learned from Marine Investigations 3
Contents
A Message from the Chair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Fire/Explosion
Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Private Yacht Andiamo 44
Towing Vessel City of Cleveland 46
Capsizing/Listing Roll-on/Roll-off Vehicle Carrier Höegh Xiamen 48
Roll-on/Roll-off Vehicle Carrier Golden Ray 4 Dive Support Vessel Iron Maiden 52
Fishing Vessel Scandies Rose 8 Fishing Vessel Lucky Angel 54
Fishing Vessel Master Dylan 56
Collision Offshore Supply Vessel Ocean Intervention 58
Towing Vessel Susan Lynn 60
Tanker Bow Fortune and Fishing Vessel Pappy’s Pride 12
Dredging Vessel Waymon Boyd 62
Offshore Supply Vessel Cheramie Bo Truc No 22 and
Articulated Tug and Barge Mariya Moran–Texas 14 Flooding/Hull Failure
Towing Vessels Cooperative Spirit and RC Creppel and Tows 16
Liquefied Gas Carrier Genesis River and Towing Vessel Voyager and Tow 18 Towing Vessel Alton St. Amant 68
Cargo Vessel Nomadic Milde and Bulk Carrier Atlantic Venus 22 Fishing Vessel Rebecca Mary 70
Contact Grounding/Stranding
Tanker Atina with Oil and Gas Production Platform SP-57B 24 Fishing Vessel Miss Annie 72
Bulk Carrier Atlantic Huron with the Soo Locks West Center Pier 26
Barge Breakaway and Contact with Interstate 10 Bridge 28
Ship/Equipment/Cargo Damage
Towing Vessel Cooperative Spirit and Tow with Containership CMA CGM Bianca 74
Hale Boggs Memorial Bridge Pier 30 Deck Cargo Barge Ho’omaka Hou,Towed by Hoku Loa 76
Bulk Carrier GH Storm Cat’s Crane with Zen-Noh Grain Facility 32
Lessons Learned. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Towing Vessel Island Lookout and Tow with Centerville Turnpike Bridge 34
Liquid Petroleum Gas Carrier Levant with Mooring Dolphin 36 Vessel Particulars by Vessel Group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Towing Vessel Old Glory and Tow with Peter P. Cobb Memorial Bridge 38 Table and Map of Accident Locations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Towing Vessel Savage Voyager and Tow with Jamie Whitten Lock & Dam 40
Towing Vessel Trent Joseph and Tow with Barataria Bridge 42 Acknowledgment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Who Has the Lead: USCG or NTSB?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Table of Figures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2021–2022 Most Wanted List of Transportation Safety Improvements:
Improve Passenger and Fishing Vessel Safety. . . . . . . . . . . . . . . . . . . . . . . . . . . 90
NTSB Office of Marine Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Vessel Group Key CARGO, GENERAL COMBATANT/MILITARY PASSENGER TOWING/BARGE

CARGO, DRY BULK FISHING PATROL/SMALL CRAFT YACHT/BOAT
CARGO, LIQUID BULK OFFSHORE SPECIALTY/OTHER
VESSEL GROUP
Capsizing/Listing
CARGO, GENERAL
Capsizing of Roll-on/
Roll-off Vehicle Carrier
Golden Ray
St. Simons Sound, Brunswick River, near
Brunswick, Georgia
ACCIDENT DATE REPORT NUMBER
September 8, 2019 MAR-21-03
ACCIDENT ID ISSUED
DCA19FM048 August 26, 2021
Figure 1. Ro/Ro vehicle carrier Golden Ray before the Figure 2. The Golden Ray heeled to its port side. SOURCE: COAST GUARD
accident. SOURCE: COAST GUARD
O
n September 8, 2019, about 0137 eastern daylight to enhance the stability of the vessel in anticipation of
time, the Ro/Ro vehicle carrier Golden Ray encountering Hurricane Dorian, the chief officer oversaw
capsized during a starboard turn while navigating the loading of about 1,500 MT of sea water ballast into
the Port of Brunswick. Of the 23 crew and 1 pilot on the vessel’s three double bottom water ballast tanks
board, 2 sustained serious injuries; the remaining 22 (no. 5 port, centerline, and starboard) and the no. 6
were not injured. The Golden Ray and its cargo sustained centerline water ballast tank. The Golden Ray then waited
significant damage due to fire, flooding, and salt water off the coast of Key West, Florida, from September 1–3
corrosion. Total costs for the loss of the vessel were to allow the hurricane to pass.
estimated at $62.5 million, and total costs for the loss of On September 3, the charterer (Hyundai Glovis Ltd. Co.)
the cargo were estimated at $142 million. directed the master to proceed to Jacksonville instead
On August 27, 2019, the Republic of the Marshall of Brunswick. To reduce the Golden Ray’s draft to less
Islands-flagged, 656-foot-long Ro/Ro Golden Ray, than 9.4 meters (about 31 feet) as required by the port,
arrived in Freeport, Texas, to offload a portion of its the chief officer discharged about 1,500 MT of sea water
cargo (vehicles) and load new cargo. The vessel had ballast from the same tanks that were loaded due to the
23 crewmembers, including a master and a chief officer. hurricane.
On August 30, the Golden Ray departed Freeport, Texas, On September 7, 2019, at 0510, the vessel departed
en route to Brunswick, Georgia, after which the vessel Jacksonville, en route to Brunswick, carrying
was scheduled to proceed to Jacksonville, Florida, before 4,067 vehicles with a total cargo weight of 8,407.2 MT
heading to Baltimore, Maryland. After departing Freeport, and displacing 35,044 MT with a midship draft of
30.9 feet (9.4 meters). The Golden Ray arrived outside

the Port of Brunswick that afternoon, and, at 1453, a
state pilot from the Brunswick Bar Pilots Association
boarded the vessel to navigate the vessel into the port.
The pilot and master conducted a master/pilot exchange
to discuss the transit. After the exchange was completed,
the pilot navigated the vessel to the Colonel’s Island
Terminal in Brunswick, docking at 1736.
Shoreside personnel and the vessel’s crew began cargo
operations, offloading and loading vehicles through the
stern ramp. There were no issues reported by shoreside
personnel or the vessel’s crewmembers with cargo
unloading or loading.
Cargo operations were completed by 2330, and the
chief officer supervised preparations for the vessel’s
departure. He transferred 8 MT of water from the no.
5 port double bottom water ballast tank to the no. 5
starboard double bottom water ballast tank, resulting in
the vessel’s list changing from 0.42° to port to 0.03° to
starboard while at the dock.
About 0030 on September 8, the same pilot boarded to
pilot the vessel outbound from the port. The draft of the
vessel was the same as when the vessel entered the Figure 3. Trackline of the Golden Ray after it departed the Colonel’s Island Terminal. BACKGROUND SOURCE: GOOGLE EARTH
previous day (30.8 feet [9.4 meters] forward and 31.2 feet
aft [9.5 meters], which met the required minimum About the same time, the master ordered the crew to
underkeel clearance of 3 feet in the 36-foot-deep open the 7-feet-high-by-7-feet-wide portside pilot door
channel). The vessel was displacing 34,609 MT, with a (side port), located on deck 5, in preparation for the
midship draft of 30.8 feet (9.4 meters). pilot’s planned departure just outside of the Port of
About 0053, the pilot began issuing orders to take in Brunswick at the sea buoy. After supervising the opening
the vessel’s lines and maneuver the Golden Ray off of the portside pilot door, the chief officer went to his
the pier with undocking assistance from the tugboat stateroom. No one remained at the open door as the
Dorothy Moran. By 0100, the Golden Ray was proceeding vessel proceeded outbound.
Figure 4. The Golden Ray departing Jacksonville, Florida,
outbound in the Turtle River Lower Range at 6 knots on a At 0122:43, the vessel approached the Cedar Hammock about 0130 on September 7. The light amidship is the
course of 113°. At 0102:43, the pilot ordered full ahead. Range at a speed of 11.6 knots. The pilot ordered 20° open portside pilot door. (The lights on the blue hull at the
The Dorothy Moran cast off its line but remained with the port rudder to turn left into the Cedar Hammock Range at bow and stern are mooring line stations.)
Golden Ray to provide support as needed. a course of 075° (a change in course of 38°). From this SOURCE: COAST GUARD
About 0108, the vessel passed under the Sydney Lanier turn, it was 1.3 nautical miles (1.15 statute miles) to the
Bridge, where the Dorothy Moran stopped its transit with next left turn into the Jekyll Island Range. At 0128:50, at At 0134:53, at a speed of 12.4 knots and a heading of
the Golden Ray. The Golden Ray proceeded outbound a speed of 12.1 knots, the pilot again ordered 20° port 039°, the Golden Ray approached the 68° right turn
in the Brunswick Point Cut Range on a course of 113°, rudder to enter the range at a course of 037° (a change in at widener 11. The pilot ordered a heading of 044°.
following pilot orders. course of 38°). The vessel made both left turns without About one minute later, at 0136:08, the pilot ordered
incident. “starboard 10” to initiate the turn, and the helmsman
moved the rudder to comply. At 0136:15, with the

Capsizing/Listing
vessel’s speed at 12.9 knots, the helmsman informed the

pilot that the rudder was at starboard 10. Shortly after,
at 0136:39, the pilot ordered “starboard 20” to enter the
Plantation Creek Range, which had a course of 105°
and led to the Atlantic Ocean. The helmsman moved the
rudder to comply with the pilot’s command; the vessel’s
speed at the time was 13.3 knots. Figure 5. Golden Ray 6 hours after the heeling event.
Seconds later, at 0136:47, the pilot ordered the rudder Flame and smoke emanate from the starboard side in the
returned to midships (zero rudder angle). The helmsman area of the cargo decks. SOURCE: COAST GUARD
complied with the pilot’s order, and, according to the pilot,
the “ship just took off.” At 0136:58, the vessel started to Responders initially rescued the pilot and 19 of
heel to port. The pilot stated that as the vessel began the 23 crewmembers on board. Four engineering
to turn, it “felt directionally unstable…meaning when I crewmembers remained trapped in the engine room until
started the turn, she wanted to keep turning.” the following evening, September 9, when responders cut Figure 7. Responders drilled holes into the hull to access
into the vessel’s hull to rescue them. the engineers. SOURCE: COAST GUARD
The pilot and the vessel’s
master began rapidly
issuing rudder commands
in an attempt to counter
the heeling. However, the
Golden Ray continued to
heel over, the rate of turn
to starboard increased,
and the vessel heeled to
port to about 60° in less
than a minute. Water
entered deck 5 through
the vessel’s open portside
pilot door and flooded
through open watertight
doors to the engine and
steering gear rooms. The
vessel eventually settled
on its port side at an
angle of 90°.
The Coast Guard
responded to the
accident, along with
tugboats and pilot
boats from the Port of
Brunswick, as well as
other first responders. Figure 6. Simplified profile of the Golden Ray and 3-D diagram showing a portion of the engine room and engine control room. Exit doors shaded gray.
BACKGROUND SOURCE: GENERAL ARRANGEMENT DRAWING, HYUNDAI MIPO DOCKYARD
Figure 8. Graphs from the Golden Ray performance study by the NTSB showing the vessel's heeling angle, including the Figure 9. Emergency responders work to rescue the
previous two left turns (left), and heading and rate of turn (right) through the transit. trapped engineers and cadet. SOURCE: COAST GUARD
SAFETY ISSUES SAFETY RECOMMENDATIONS

Improperly calculating vessel stability. The operating company, G-Marine Service Co. Ltd., did not provide stability As a result of its investigation into this accident,
software training for its officers who were responsible for using the Golden Ray’s LOADCOM stability calculation the NTSB issued two new safety recommendations
program. The company’s safety management system outlined the chief officer’s duties, including vessel stability to the Golden Ray’s operator, G-Marine Service
calculations, but did not provide any instructions on how to use, or require competency for using, the LOADCOM Co. Ltd. The NTSB found that G-Marine’s lack
computer. Since the company did not provide training on how to use the computer, it had no means to ensure that the of oversight and procedures for auditing and
chief officer was capable of performing his duty to accurately determine the ship’s stability. After the accident, G-Marine verifying the accuracy of their officers’ vessel
implemented several policies to improve safety and reduce the likelihood of another similar accident, including requiring stability calculations before departure contributed
stability calculation training for chief officers. to the Golden Ray not meeting international
Lack of company oversight for calculating vessel stability. The chief officer was the only crewmember responsible stability standards. The NTSB recommended that
for calculating the stability of the vessel. Once the chief officer had calculated the vessel’s stability, he reported the G-Marine revise its SMS to establish procedures for
vessel’s final metacentric height to the master and the company (via the departure report), but neither the master nor verifying stability calculations and implement audit
the company verified that the chief officer’s calculations met stability requirements. The company had no procedures to procedures.
verify stability calculations, so the master and company were unaware that the vessel had been sailing without meeting The NTSB also found that the open watertight
stability requirements during the accident voyage and two previous voyages, and there was no established means for doors on deck 5 of the Golden Ray allowed flooding
the crew or the company to identify and attempt to correct the problem. into the vessel and blocked the primary egress
from the engine room, thus trapping engineering
The probable cause of the capsizing of the Golden Ray was the chief officer’s error entering ballast quantities personnel. Watertight integrity is critical to the
into the stability calculation program, which led to his incorrect determination of the vessel’s stability and safety of a vessel and its crew, so the NTSB
resulted in the Golden Ray having an insufficient righting arm to counteract the forces developed during a turn recommended that G-Marine revise its SMS audit
while transiting outbound from the Port of Brunswick through St. Simons Sound. Contributing to the accident process to verify crew adherence to the company’s
was G-Marine Service Co. Ltd.’s lack of effective procedures in their safety management system for verifying Arrival/Departure Checklist regarding the closure of
stability calculations. watertight doors.
VESSEL GROUP
Capsizing/Listing
FISHING
Capsizing and Sinking

of Commercial Fishing
Vessel Scandies Rose
North Pacific Ocean, 2.5 Miles South of
Sutwik Island, Alaska
December 31, 2019 MAR-21-02
ACCIDENT ID ISSUED
DCA20FM009 June 29, 2021
Figure 11. Scandies Rose arriving in Kodiak, Alaska, before the accident (date unknown). SOURCE: GERRY COBBAN KNAGIN
O
Figure 10. The Scandies Rose wreck, located on n December 31, 2019, Coast Guard took on bait. The captain conducted drills that included
February 11. SOURCE: GLOBAL Communications Detachment Kodiak discussions of the liferaft locations, the vessel’s EPIRB,
received a distress call from the fishing vessel and how to make a mayday call, and a demonstration of
Scandies Rose. The vessel was en route from Kodiak how to don an immersion suit.
to fishing grounds in the Bering Sea when it capsized The captain and crew discussed the weather forecast
about 2.5 miles south of Sutwik Island, Alaska, and sank along the planned route (through the Kupreanof Strait,
several minutes later. At the time of the accident, the then southwest through the Shelikof Strait toward
Scandies Rose had seven crewmembers aboard, two of False Pass en route to the Bering Sea). About 1630,
whom were rescued by the Coast Guard several hours the NWS issued a marine forecast that included a gale
later. The other missing crewmembers were not found warning and a heavy freezing spray warning for the
and are presumed dead. The Scandies Rose, valued at vessel’s proposed route. The crew knew the weather
$15 million, was declared a total loss. “was going to be bad” and that there would be icing
On December 29, in preparation for the Bering Sea pot conditions in which sea spray could potentially freeze to
cod fishery, which opened on January 1, 2020, the crew the vessel and crab pots.
loaded 195 combination crab pots in accordance with At 2035, the vessel departed Kodiak, maintaining an
the vessel’s stability instructions, which laid out sample average speed of about 9–10 knots. An underway
loading conditions that the captain could follow to navigation watch rotation had the captain operating
ensure the vessel met stability criteria established by the vessel for 6 hours and the other six crewmembers
regulators. (The investigation found that the stability for 1 hour each. At 0200 on December 31, the vessel
instructions were inaccurate; therefore, the vessel did entered the Shelikof Strait, and the captain departed
not meet regulatory stability criteria.) The next day, they the wheelhouse; each of the crewmembers on watch
chained the pot stack, secured hatches, tested bilge after him maintained a southwesterly course. Between
level sensors, fueled, topped off potable water, and 0600 and 0800, on-watch crewmembers observed that
wind and waves acting on the starboard bow of the with an estimated 2 inches of ice. Ice coated the inside trying to seek shelter southeast of Sutwik Island but
vessel began to cause ice accumulation on the forward webbing of the starboard pots as well. The captain told was nervous about the “uncharted rock[s].” Although
starboard side of the vessel, railings, and pots. At 0800, the departing crewmember the weather was too rough weather forecasts had projected heavy icing (0.8–1.6
when the captain took the watch, the vessel had no list or to have the crew out on deck chopping ice and that they inches per hour), given the weather conditions observed
heel. At 1118, during a call to another fishing vessel, the would wait until the vessel was in protected waters. by the captain for the two hours between 2000–2200,
captain said his vessel was experiencing light icing and At 1930, the vessel was about equidistant from locations the Scandies Rose was likely experiencing extreme icing
the sea conditions were poor. to shelter to the north and southwest, and the vessel (greater than 1.6 inches per hour) and had accumulated
For about 6 hours, beginning at 1400, the crew rotated was likely experiencing heavy icing conditions. The 6–15 inches on surfaces exposed to wind and icing
watches, and the average speed decreased to 6.5 knots captain was familiar with area southwest near Sutwik during the voyage.
(although the engine’s rpm was constant). The wind Island, which was also along the vessel’s intended route. About 2145, the vessel was about 2.5 miles south of
and weather “progressively got worse all day,” and the Ultimately, the captain decided to maintain course and Sutwik Island when it turned about 50° to starboard and
vessel was “bucking” into the seas, “making a lot of spray, speed and call the onboard engineer (likely to transfer held a northwesterly course in the direction of Sutwik
and the spray was making ice.” About 1915, after being fuel to correct the starboard list). Island’s southern bay. Afterward, the captain reported
woken for his 2000-0200 watch, the captain discussed At 2000, the captain called a friend and said he needed that his vessel’s “list had gotten a lot worse.” The sudden
the worsening weather, the accumulating ice, and the to “tuck in someplace safe.” At 2037, he called the increased list at the time of the course change indicates
development of an approximately 2° starboard list with fishing vessel Pacific Sounder and reported that the that the course alteration to starboard exposed the
the departing crewmember. Scandies Rose was icing “really bad” and he was vessel’s port side to the prevailing wind and waves, which
They considered reducing the vessel’s speed and altering concerned about a 20° starboard list. He also noted exacerbated the starboard list. Although the captain’s
course to limit the freezing spray causing icing on the that the winds were blowing 60–70 knots from the west decision to proceed to Sutwik Island was reasonable,
vessel. At this time, all the pots were glazed over with and the temperature was 12°F. He further stated it was by the time he was close enough to turn into the lee, the
ice. The starboard-side pots were more heavily coated too rough to send the crew out to break ice, and he was icing conditions had accelerated and reduced the vessel’s
stability.
Figure 12. Timeline of the Scandies Rose voyage. For detail of red inset box, see accident Figure 13. Scandies Rose’s positions over last 3 hours before mayday call.
site map at right. BACKGROUND SOURCE: GOOGLE EARTH BACKGROUND SOURCE: NOAA
and deployed a C-130 aircraft. The first helicopter arrived

Capsizing/Listing
at the captain’s mayday coordinates around 0200 and

located the vessel’s empty second liferaft. The two
survivors saw the helicopter’s lights and used a flashlight
to signal. About 4 hours after entering the liferaft, they
were hoisted to the helicopter with the assistance of a
rescue swimmer. No other crewmembers were found.
SAFETY ISSUES
The effect of extreme icing conditions. Sea spray
icing is a serious hazard to marine vessels because
Figure 14. Photo of the bridge of the Scandies Rose taken the ice accumulates over exposed decks and exterior
from the starboard helm station. surfaces of a vessel, adding weight that may ultimately Figure 16. Coast Guard Cutter Mellon crew breaking ice
SOURCE: 2019 CONDITION AND VALUATION SURVEY capsize a vessel. The Scandies Rose was carrying a full during search efforts late afternoon on January 1.
stack of pots that reached about 20 feet above the main SOURCE: COAST GUARD
About this time, the vessel was jolted by a sudden deck, and ice from freezing spray formed asymmetrically
sustained list to starboard, and all the crew rushed to the on the starboard side and built as the voyage progressed. The vessel’s inaccurate stability instructions. The
wheelhouse. At 2155, the captain broadcasted a mayday The added weight of accumulated intent of regulatory requirements
call on high frequency, announcing that they were “rolling ice high on a vessel—in this case, for stability instructions is to
over.” He included the vessel’s position in the call. The up the stack of pots, the fo’c’sle, provide information to vessel
Coast Guard received the transmission and attempted bulwarks, and portions of the owners and operators that enables
to establish communications with the vessel but was house—rapidly raises a vessel’s them to readily ascertain the
unable to do so. center of gravity and diminishes its stability of their vessels under
The first two crewmembers to arrive in the wheelhouse stability. The asymmetrical icing on varying loading conditions and to
got out the vessel’s immersion suits and got them the starboard side of the Scandies operate them in compliance with
up halfway but had difficulty because the vessel was Rose caused the vessel to develop applicable stability criteria, which
“leaning over so hard.” They climbed out the portside an increasing starboard list, and have been developed to provide
door and finished donning their suits outside. They the course change at 2145 brought an adequate level of safety for
attempted to use a line to assist the remaining crew the 60–70 knot winds onto the vessels that are operated prudently.
inside, but the vessel sank lower, and a wave swept them port side, adding to the existing list A margin of safety is built into
over the side. They were in the water about 20 minutes from icing. The sudden increase these criteria to accommodate
before they saw the light from an inflatable liferaft (that in list shortly later indicated that forces that can act on a vessel,
had automatically deployed) and were able to swim to it the vessel’s stability had been such as rolling in waves, heeling
and climb aboard. Their liferaft light went out, and they overcome and that the vessel was due to wind, or limited degree of
fired all their flares. Neither had a PLB. capsizing. The added weight from listing. The investigation found
ice accumulating asymmetrically that the Scandies Rose’s stability
Without the captain’s distress call, the Coast Guard instructions were inaccurate, and
likely would have been initially unaware of the accident on the vessel and the stacked crab
pots on deck raised the Scandies the vessel had “dangerously low
because, when the vessel sank, the GPS-equipped righting energy”—the amount of
EPIRB did not broadcast a receivable signal. Unable to Rose’s center of gravity, reducing
its stability and contributing to the energy that a vessel can absorb
communicate with the Scandies Rose, the Coast Guard from external heeling forces
requested all vessels in the area assist and launched capsizing.
(winds, waves, weight shifts,
a rescue helicopter about 2330 (multiple helicopters Figure 15. Right. Diagram of ice and etc.) before it capsizes—when
launched throughout the day), diverted the cutter Mellon, wind acting on the Scandies Rose. loaded in conditions similar to
those prevailing at the time of the accident. Thus, Lack of accurate weather data for the accident
although the crew loaded the Scandies Rose per the area. The area around Sutwik Island and west of The probable cause of the capsizing and sinking
stability instructions, the vessel had a smaller margin of Kodiak Island is subject to bad weather with northeast of the commercial fishing vessel Scandies Rose
safety than intended by the regulations and was more through northwest winds and cold air moving across was the inaccurate stability instructions for the
susceptible to capsizing. the Alaska Peninsula. The investigation found that the vessel, which resulted in a low margin of stability
Scandies Rose experienced wind conditions exceeding to resist capsizing, combined with the heavy
Need to update regulatory guidelines on calculating asymmetric ice accumulation on the vessel due
and communicating icing for vessel stability those that were forecast or measured by the closest
weather stations. When observation sites are more to localized wind and sea conditions that were
instructions. Stability regulations factor in a minimum more extreme than forecasted during the accident
set amount of added weight for accumulated ice from spread out in remote areas like Alaska, the data do
not accurately represent the entire area, which can voyage.
freezing sea spray on continuous horizontal and vertical
surfaces. However, the regulations do not provide lead to inaccurate and less precise forecasts and
guidance on how to apply ice accumulation on crab pots, weather modeling, and vessels can encounter localized
which consist of tubular frames and mesh and have conditions that are worse than expected.
additional internal ice accumulation. Nor do they account
for reported asymmetric ice accumulation on exposed
vessel surfaces and pot stacks. Additionally, stability SAFETY RECOMMENDATIONS
instructions are currently not required to present the
As a result of its investigation into this accident, the NTSB issued four new safety recommendations to the
accumulated ice thicknesses used to calculate vessel
Coast Guard to improve fishing vessel stability criteria (by evaluating the effects of icing), to improve stability
stability, which, if communicated to masters, would
instructions, and to strengthen oversight of stability calculations. The NTSB reiterated two recommendations
better prepare them in decision making.
to the Coast Guard: first, because the Scandies Rose accident showed that formal stability training would
Figure 17. Crewmembers aboard Coast Guard Cutter be helpful for fishing vessel crews, the NTSB reiterated safety recommendation M-11-24 to require owners,
Polar Star weigh a crab pot following a 3-day freezing captains, and chief engineers to receive training and demonstrate competency in vessel stability, watertight
spray experiment. SOURCE: COAST GUARD integrity, subdivision and the use of stability information. Second, because the investigation found that PLBs
can reduce or eliminate search-and-rescue errors by providing multiple GPS coordinates of survivors to
searchers, the NTSB reiterated safety recommendation M-17-45 to require all personnel employed on vessels
in coastal, Great Lakes, and ocean service be provided with a PLB to enhance their chances of survival.
The weather conditions on the accident voyage and multiple reports indicated that waters west of
Kodiak Island, near Sutwik Island and Chignik Bay, are subject to freezing spray and icing and therefore pose
an increased hazard to the marine community. Thus, the NTSB made two weather related recommendations:
one for NOAA to increase the surface observation resources necessary for improved local forecasts for
the region, and a second to the NWS to make their currently experimental Ocean Prediction Center freezing
spray website—which detailed graphical icing information not currently available elsewhere—operational and
promote its use in industry.
Believing that awareness of the safety issues found in the investigation would benefit fishing vessel crews,
the NTSB recommended that the North Pacific Fishing Vessel Owners’ Association notify their members of
the specifics of this accident, the amount of ice assumed when developing stability instructions, and the
dangers of icing.
O
VESSEL GROUPS n January 14, 2020, about 1537 local time, the
Collision
CARGO, LIQUID BULK • FISHING tanker Bow Fortune was transiting inbound to
Galveston, Texas, in the Outer Bar Channel while
the uninspected commercial fishing vessel Pappy’s Pride
Collision of Tanker was transiting outbound. The two vessels collided in
dense fog, and the fishing vessel capsized and sank. Of
Bow Fortune and the four crewmembers aboard the fishing vessel, there
were three fatalities and one serious injury. There were
Commercial Fishing no injuries to the pilot or the crew of 28 on board the
Bow Fortune. A surface sheen of diesel was reported.
Figure 19. The Pappy’s Pride before the accident.
SOURCE: BALTICSHIPPING.COM
Vessel Pappy’s Pride The Pappy’s Pride, valued at $575,000, was declared a
total loss. The Bow Fortune was anchored about 15 miles offshore
in the East Galveston Fairway Anchorage Area with
The Pappy’s Pride got under way at 1325 from Pier 75 in an estimated visibility at 1.5 miles (the day before, the
Outer Bar Channel, Galveston, Texas the Port of Galveston to shrimp along the gulf coast to NWS had warned of dense fog in the area). At 1415, the
the north. The vessel headed outbound to the east and master navigated the vessel from the anchorage, with
north of the inbound barge lane of the ship channel. The the fog signal energized. At 1500, a pilot boarded the
January 14, 2020 MAB-21-21
Pappy’s Pride captain steered his vessel on the outbound Bow Fortune. He estimated visibility between 0.25 miles
ACCIDENT ID ISSUED transit, through the intracoastal waterway, increasing to port and 0.75 miles to starboard. During the master/
DCA20FM011 October 20, 2021 speed to cross the two channels then transiting through pilot exchange, the master and pilot noted the fog but
anchored vessels in the anchorage, most of which was agreed to continue the inbound transit. The pilot set
Figure 18. Bow Fortune docked after the casualty. done in near-zero visibility, without sounding the required up two PPUs, sent an AB to the bow as lookout, and
SOURCE: COAST GUARD fog signals, initiating or responding to any radio calls, kept the fog whistle sounding from the forward mast.
or effectively using his available bridge equipment to At 1507, with the engine at half ahead, the pilot on the
determine risk of collision. Communication, especially Bow Fortune took the conn. At 1524, the Bow Fortune
in limited visibility, is a vital part of standing an effective was inbound in the Galveston Bay Entrance Channel.
watch.
At 1537:32, the vessels collided just outside the Outer

Bar Channel. The speed over ground of the Bow Fortune
was 11 knots and Pappy’s Pride was 8.4 knots, indicating
the tanker was beginning to slow but the fishing vessel’s
captain did not appreciably change the propulsion engine
speed or direction before the collision. The damage
assessed on both vessels indicates the port side of the
tanker’s bulbous bow struck the starboard side of the
fishing vessel during the collision, which in turn led to the
vessel capsizing but remaining afloat as it passed down
the port side of the tank. One deckhand was rescued, but
the remaining Pappy’s Pride crew perished.
The probable cause of the collision of the inbound

Figure 20. Pappy’s Pride outbound track from Galveston (red) and Bow Fortune inbound track (blue) just before the tanker Bow Fortune and the outbound commercial
collision. BACKGROUND SOURCE: NOAA; TRACKLINE DATA: COAST GUARD fishing vessel Pappy’s Pride was the captain of
the Pappy’s Pride’s outbound course toward the
About 1532, the Bow Fortune was less than 2 miles two vessels may have scraped port sides or avoided ship channel, which created a close quarters
from the Pappy’s Pride. Due to the estimated 0.2-mile contact. The Pappy’s Pride’s outbound course created situation in restricted visibility. Contributing was
visibility, the tanker would not have been visible from the a close quarters situation that was not prudent, and the the lack of communication from the captain of the
fishing vessel, but its position would have been available lack of communication from the fishing vessel created Pappy’s Pride.
to the Pappy’s Pride captain on radar/ARPA and the doubt as to the Pappy’s Pride captain’s intentions.
AIS information on the electronic chart. At this point, the
Pappy’s Pride appeared to be on a course to cross the
channel in front of the inbound Bow Fortune and behind
the outbound tanker Chemical Atlantik.
At 1535:18, the Bow Fortune pilot first hailed the
Pappy’s Pride without response. Twelve seconds after
the initial radio hails, the Pappy’s Pride made a course
change to port (about 19°), indicating the captain was
still actively steering. At 1536:18, the Bow Fortune
sounded five short blasts, then hailed the Pappy’s Pride,
and again sounded five short blasts at 1537:10. VTS, AIS,
and the pilot’s PPU electronic data captured a heading
change of about 15° to port for the Pappy’s Pride in
the seconds before the collision. The captain may have Figure 21. Pappy’s Pride postaccident damage. SOURCE: COAST GUARD.
thought his heading changes would keep him out of the
channel and avoid the collision, respectively. However,
if the last course change had been an attempt to avoid Early Communication
the collision, then it should have been to starboard per Early communication can be an effective measure in averting close quarters situations.
the COLREGS rules. Further, in this collision, if the fishing The use of VHF radio can help to dispel assumptions and provide operators with the
vessel had maintained its previous heading of 113°, the information needed to better assess each vessel’s intentions.
O
VESSEL GROUPS n November 14, 2019, about 0415 local time, the
Collision
OFFSHORE • TOWING/BARGE offshore supply vessel Cheramie Bo Truc No 22

was outbound for sea transiting Sabine Pass
with a crew of five, when it collided with the inbound
Collision between ATB Mariya Moran–Texas, with a pilot and nine crew
aboard, in the vicinity of Texas Point. About 6,641
Offshore Supply Vessel gallons of diesel oil were released, and the waterway
was closed for 12 hours. No injuries were reported. The
Cheramie Bo Truc Cheramie Bo Truc No 22, valued at $1.2 million, was
declared a total loss. The Mariya Moran–Texas sustained
No 22 and Articulated $654,572 in damages.
Tug and Barge

After the Cheramie Bo Truc No 22 left the dock at 0352,
the mate unsuccessfully attempted to use the autopilot
Mariya Moran–Texas
feature for several minutes. He struggled to maintain a
safe course during the first 12 minutes of the passage,
nearly striking a stationary jack-up, and the AB stated
that twice he and the engineer had to remind the mate
Sabine Pass Jetty Channel, Port Arthur, Texas to steer back into the channel. Autopilot use is often
discouraged or prohibited in restricted waters.
November 14, 2019 MAB-21-08 The manual for the Cheramie Bo Truc No 22’s autopilot
specifically warned users not to use autopilot in a
ACCIDENT ID ISSUED Figure 24. Damaged stem of the barge Texas.
“harbor entrance or narrow channel.” After returning to
DCA20FM003 March 25, 2021 SOURCE: MORAN TOWING CORP.
manual steering, the Cheramie Bo Truc No 22 crossed
the channel three times and wound up on the east About 0400, roughly 0.5 miles inside the jetties, a
side of the channel. The AB reported sighting the Sabine pilot boarded the inbound Mariya Moran–Texas
Mariya Moran–Texas to the mate. ATB and, about 10 minutes later, after the chief mate
and pilot completed a master/pilot exchange, the pilot
checked in with VTS, which advised him of the outbound
Cheramie Bo Truc No 22. The pilot first noticed the
Cheramie Bo Truc No 22’s masthead lights about 1 mile
Figure 22. Mariya Moran–Texas. Figure 23. Cheramie Bo Truc No 22. away.
SOURCE: JEFF CAMERON, MARINETRAFFIC.COM. SOURCE: SHIPSPOTTING.COM.
Although both vessels were aware of each other, no
VHF radio passing arrangement or maneuvering signals
were made. The Mariya Moran–Texas pilot assumed a
starboard-to-starboard passing based on the position of
the Cheramie Bo Truc No 22.
About 0414, Cheramie Bo Truc No 22 started a turn to
starboard. The AB and engineer noticed the mate start
the turn and recommended he come left to avoid the
ATB. The mate did not acknowledge them. They did not
take further action, such as summoning the captain,
despite the hazardous situation.
Figure 26. Sabine Pass chart showing path of Cheramie Bo Truc No 22 (red) and Mariya Moran–Texas (green).
BACKGROUND SOURCE: NOAA CHART 11342
Figure 25. The port side of the Cheramie Bo Truc No 22,

post-collision, showing the compromised no. 1 port fuel The bow of the barge Texas collided with the Cheramie stationary jack-ups, weaving across the channel,
tank and oil spillage. SOURCE: COAST GUARD. Bo Truc No 22 at nearly a right angle, aft of the ignoring the warnings from the AB and engineer in the
superstructure. The Cheramie Bo Truc No 22’s no. 1 port wheelhouse, and suddenly turning in front of the ATB all
The Mariya Moran–Texas pilot hailed the Cheramie fuel tank was severely damaged. The no. 2 port fuel indicate a degree of misjudgment, impairment, and/or
Bo Truc No 22 on channel 13. The VTS watchstander tank sustained less damage but was also compromised. incompetence.
noticed the Cheramie Bo Truc No 22’s “course had Neither ATB vessel’s hull was compromised.
changed abruptly,” placing the vessels on a collision
The mate’s postaccident alcohol swab test results The probable cause of the collision between the
course. He reached out to the Cheramie Bo Truc No 22
indicated a blood alcohol concentration of at least offshore supply vessel Cheramie Bo Truc No 22
once, on channel 1A, with no answer.
0.02 grams per deciliter; although indicative that the and ATB Mariya Moran–Texas was the offshore
Evidence showed that there was room to pass safely mate consumed alcohol sometime prior to the voyage, supply vessel mate’s turn across the path of the
starboard to starboard had Cheramie Bo Truc No 22 the test does not demonstrate conclusively that the ATB during a meeting situation. Contributing to
continued parallel to and along the east edge of the mate was impaired by alcohol. However, attempting the accident was a lack of early communication
channel. Inland Navigation Rules require either a port-to- to use the autopilot in a channel, nearly colliding with from both vessels.
port passage or communication either by radio or whistle
signal for an agreed-upon alternate passage between
two vessels. A radio call from the ATB prior to the pilot Teamwork
relieving the mate to confirm the offshore supply vessel’s Safe and effective navigation is not one person’s job. Bridge resource management includes the
intentions may have kept the Cheramie Bo Truc No 22 concept of teamwork, which is an essential defense against human error. A good team should
mate from steering across the ATB’s bow. After the turn anticipate dangerous situations and recognize the development of an error chain. If in doubt, team
to starboard across the bow of the ATB, the collision was members should speak up or notify a higher authority. Vessel operators should train their crews on
unavoidable. and enforce their safety policies.
O
VESSEL GROUPS n January 26, 2020, at 0533 central standard time, At 0522:47, the Cooperative Spirit pilot and the
Collision
TOWING/BARGE • CARGO, DRY BULK the towing vessel Cooperative Spirit was pushing RC Creppel pilot agreed to a port-to-port meeting in the
40 barges upbound on the Lower Mississippi River, bend at 26 Mile Point. Because they planned to meet
and the towing vessel RC Creppel was pushing two in a bend, the high water and strong current increased
Collision between barges downbound when the two tows collided at mile
123, near Destrehan, Louisiana. The RC Creppel capsized
the risk of an accident occurring as both vessels
maneuvered for the turn at the same time. Additionally,
Cooperative Spirit Tow as a result of the collision. Minutes later, the upbound dry
bulk carrier Glory First made contact with the starboard
both operators would have only been able see each
other’s tows and visually assess the situation for a short
and RC Creppel Tow side of the Cooperative Spirit’s tow. All 42 barges from
both tows broke free and were later recovered. One of
time as they approached each other and would have little
time to react, if necessary.
the four RC Creppel crewmembers was rescued; the In such a situation—where two vessels are approaching
Lower Mississippi River, mile 123, remaining three were never recovered and are presumed a bend from opposite directions—navigational tools
Destrehan, Louisiana dead. The accident resulted in the release of about can help to mitigate the risk of collision. Although both
8,000 gallons of diesel fuel into the river and sulfuric the RC Creppel and Cooperative Spirit were equipped
ACCIDENT DATE REPORT NUMBER acid vapors into the atmosphere, and an estimated with AIS, ECS, and radar, each pilot had entered only the
January 26, 2020 MAB-21-16 $3,781,126 in property damage to the 3 vessels and size of his vessel into AIS, rather than length of both the
ACCIDENT ID ISSUED 11 barges. vessel and tow. Since the Cooperative Spirit pilot did not
DCA20FM012 August 12, 2021 At 0433, the 200-foot-long Cooperative Spirit departed inform the RC Creppel pilot of the size or length of his
mile 115.4 of the Mississippi River with 40 barges tow when they planned their meeting, the RC Creppel pilot
arranged six across and seven long (the first two rows was likely unaware of the length of the Cooperative Spirit
each consisted of five barges), headed up river. The tow, and the two pilots had arranged the meeting without
total length of the Cooperative Spirit and its tow was a complete understanding of the developing situation.
1,600 feet long. The vessel’s pilot was in the wheelhouse. About 0531, as the Cooperative Spirit began to transit
About 0514, the 69-foot-long RC Creppel departed the bend, the vessel was tight on the left descending
Hahnville, Louisiana, at mile 126.9, headed down river bank. The pilot used 15–20° starboard rudder for more
with two barges: the SCC-95 ahead of the RHA-2204. than 90 seconds to execute the turn around the bend.
The RC Creppel and its tow measured 514 feet long. The However, the force of the current set the vessel and its
vessel’s pilot was at the helm. tow to port (toward the right descending bank) and into
the path of the downbound RC Creppel, which was in the
Figure 27. RC Creppel under way before the accident. center of the river as it approached the bend.
SOURCE: JEFF L. YATES.
Figure 28. Cooperative Spirit moored after the accident. Figure 29. Glory First anchored after the accident.
Once in sight of one another, there was minimal time

for either pilot to react or respond to the other vessel’s
movements to avoid collision. The Cooperative Spirit
pilot had to assess the risk of collision from more than
1,400 feet behind the head of the tow. He assumed
that the pilot of the smaller RC Creppel tow would
maneuver his tow closer to the right descending bank.
A radio call to the other vessel would have helped both
pilots identify each other’s expectations, but neither
pilot made a radio call after their initial call to arrange a
meeting. The absence of a radio call or “danger” signal
indicates that neither pilot was aware of the impending
collision.
On board the RC Creppel, one of the two deckhands
heard the general alarm sound, immediately grabbed a
lifejacket, and felt an impact as the Cooperative Spirit
tow collided with the RC Creppel’s tow at 0533:04.
The impact separated towing lines connecting the
Cooperative Spirit’s first and second string of barges.
One of the RHA-2204’s pressure relief valves on the
deck was ruptured, releasing sulfuric acid vapors into
the atmosphere.
By 0533:26, the RC Creppel had capsized and begun to
sink. Only one of the four crewmembers was able to
escape.
The Cooperative Spirit continued to drift down and Figure 30. Rose Point ECS display at 0533 screenshot from the Cooperative Spirit, with previous positions of the
across the river into the upbound Glory First, which Cooperative Spirit starting at 0528 (with Cooperative Spirit tow and RC Creppel tow icons added to show overall tow
had slowed and evasively maneuvered toward the right dimensions to approximate scale). BACKGROUND SOURCE: ARTCO
descending bank, and the Glory First struck the aft
corner of the starboard string of barges pushed by the Updating the Overall Dimensions of a Tow in AIS
Cooperative Spirit. The NTSB has previously noted the importance of ensuring that vessels engaged in towing operations
broadcast accurate AIS information regarding tow size. The overall dimensions of a vessel and its tow
The probable cause of the collision of the may change significantly with each transit. For vessels towing ahead or alongside, the dimensions
Cooperative Spirit and RC Creppel tows was the in AIS should reflect the overall rectangular area of the vessel and its tow. Consistently entering
two pilots’ insufficient radio communication the complete dimensions of a tow configuration into AIS for each transit helps to alleviate possible
before meeting in a bend and not broadcasting misinterpretation and thus enhances the situational awareness of all waterway users.
accurate AIS information regarding tow size.
Communication When Meeting in a Bend
When meeting or overtaking a vessel in a bend, especially where high-water conditions can increase the
risk of a collision, early and effective communication is critical to ensuring a successful meeting. The
use of VHF radio can help to dispel assumptions and provide bridge teams and towing vessel operators
with the information needed to better assess each vessel’s intentions.
VESSEL GROUPS
Collision
CARGO, LIQUID BULK • TOWING/BARGE
Collision between
Liquefied Gas Carrier
Genesis River and
Voyager Tow
Houston Ship Channel, Upper Galveston Bay, Texas
May 10, 2019 MAR-21-01
ACCIDENT ID ISSUED
DCA19FM033 March 10, 2021
Figure 31. Voyager moored in Channelview, Texas, Figure 32. Screen capture from wheelhouse video on board the Voyager at the moment that the Genesis River struck the
following the accident. Voyager tow. SOURCE: KIRBY INLAND MARINE, LP
O
n May 20, 2019, the outbound 754-foot-long with Pilot 1 at the conn and an AB from the ship’s crew
liquefied gas carrier Genesis River was transiting at the helm. When the Genesis River departed its berth, it
the Houston Ship Channel in Upper Galveston Bay. had an even keel trim. Had the vessel gotten under way
Immediately after the vessel passed an inbound liquefied trimmed by the stern (there was sufficient depth to do
gas carrier of similar size at the southern end of the so), the shift of the trim toward the bow resulting from
Bayport Flare, it approached the channel’s west bank, the ship making way through the channel would have had
sheered to port, and crossed over to the opposite side of less impact on its path stability and maneuverability.
the channel where, in the barge lane ahead, the 69-foot- During the initial transit through the upper Houston
long towing vessel Voyager was pushing two tank Ship Channel, Pilot 1 determined that the ship’s rudder
barges breasted together side by side. In the ensuing responded sluggishly to the rudder commands. To stop
collision, two cargo tanks in the 297-foot-long starboard the swing of the Genesis River following meetings with
tank barge were breached, spilling over 11,000 barrels inbound vessels earlier in the transit, he used temporary
of petrochemical cargo into the waterway, and the port increases in engine rpm, to increase wash over the
barge capsized. No injuries were reported. rudder and improve its effectiveness.
About 1148 on May 10, two pilots boarded the At 1444, Pilot 2 took the conn from Pilot 1. Pilot 1
Genesis River at the Targa Resources Galena Park Marine remained on the bridge for the next 15 minutes talking
Terminal on the upper Houston Ship Channel and were with Pilot 2. Pilot 1 informed Pilot 2 of the sluggish
escorted to the ship’s bridge. The fully loaded Genesis rudder, and the pilots shared concerns about large ships
River got under way shortly after noon, outbound for sea that were difficult to handle.
Figure 33. Genesis River under way in Bolivar Roads near

Galveston, Texas, two weeks after the accident.
SOURCE: WILLIAM J. LEACH, JR., VESSELFINDER.COM
At 1446, Pilot 2 ordered the Genesis River’s engine to

full ahead. A little over a minute later, he requested an
increase in engine speed to sea speed, which took the
engine control system out of maneuvering mode and
into navigation full (Nav. Full) mode. In Nav. Full mode,
the ability to change speed was limited; under normal
circumstances, the pilot would give 10 minutes’ prior
notice before requesting another speed change. The
crew complied with the pilot’s request, and the engine
speed, which had been at 60 rpm, began to slowly
increase.
At 1500, an OS took the helm under the watch of the AB.
The AB stated that he had requested permission from
Pilot 2 to turn over the helm to the OS, but Pilot 2 stated
that he was not informed that the OS was at the wheel.
The VDR did not capture audio of the exchange.
Five minutes later, Pilot 2 radioed the inbound 740-foot-
long, 120-foot-wide liquefied gas carrier BW Oak to
arrange a port-to-port passing. At the time of the radio
call, the Genesis River was about a mile north of the
Bayport Flare—the intersection of the Houston and Figure 34. The accident location, as shown by the red X. Inset shows Bayport Flare and turn at Five Mile Cut.
Bayport Ship Channels that is funnel-shaped to allow BACKGROUND SOURCE: GOOGLE MAPS; INSET FROM NOAA CHART 1132714.
ships to negotiate the turn from one channel to the
other. Based on information in his PPU, Pilot 2 knew that Figure 35. Typical Lower Houston Ship Channel profile with navigation beacons as viewed by an outbound vessel.
the passing would occur near the southern part of the
Bayport Flare, where the channel makes a 15.7° turn to
the east, but was not concerned, as he had met other
ships there before.
As the Genesis River transited south, its engine speed
continued to slowly increase until it reached between 72
and 73 rpm, which was the programmed rpm setpoint for
Nav. Full. The vessel’s speed over ground was 12 knots.
Beginning at 1509:22, Pilot 2 issued a series of orders to Figure 37. Genesis River bridge
Collision
the helmsman to maneuver the Genesis River to pass the control panel for remote engine
BW Oak. As the Genesis River passed the BW Oak and control system (left) and
entered the turn at the southern terminus of the Bayport EOT lever (right).
Flare at 1512:08, the vessel was on the western side of
the main deep-draft channel at a speed of 12.6 knots. After the pilot’s second order
for more rpm, Genesis River
The channel narrowed and turned, which brought the bridge crewmembers contacted
western channel bank abruptly in toward the vessel, and the ECR and requested
the bank effects on the starboard side quickly increased. maximum rpm, and engineering
The closer a vessel is to the bank, the stronger the bank watchstanders adjusted the
effect forces. These forces were likely exacerbated maximum rpm setpoint to 85.
by unreported shoaling that had occurred on the However, neither the bridge nor
western side of the channel at the turn. Additionally, the ECR watchstanders pressed
the hydrodynamic effects of the 740-foot-long and As the Voyager turned, its speed, which had been 5.3 knots
the program bypass button. Thus, the actual engine rpm at maximum power, decreased to as low as 3.6 knots
120-foot-wide BW Oak acting on the Genesis River were did not change when the pilot ordered the increase.
greater than the effects created by smaller vessels that (a decrease in speed is not unusual during a turn,
the Genesis River had passed earlier. Over VHF radio, Pilot 2 hailed the inbound towing vessel particularly in vessels with inefficient hull forms, such as
Voyager, which was ahead of the Genesis River and barges). As the Genesis River approached the eastern
Figure 36. Hydrodynamic bank effects acting on the pushing ahead two fully loaded tank barges breasted bank of the channel, the hydrodynamic forces of bank
Genesis River: cushion force (left) and suction force (right). together side by side. Both of the Voyager’s engines were cushion pushed the bow of the ship back to starboard.
at full throttle, and the tow was making about 5.3 knots Because of its slow speed, the Voyager was still on the
speed over ground. The Voyager’s relief captain, who eastern side of the main deep-draft channel. The NTSB
was at the helm, answered the call. Pilot 2 informed estimates that the stern of the Voyager was no more
the Voyager relief captain that he was having trouble than 90 feet from the eastern deep-draft channel bank
controlling the Genesis River. In response, the Voyager as the Genesis River approached. Even if the Voyager
relief captain put his engines in neutral. Then, at 1513:25, relief captain had taken a course more perpendicular
Pilot 2 radioed the Voyager, requesting the tow “Go to to the channel, it is likely that the tow would have been
the greens,” that is, cross the channel to the western in a similar position relative to the bank, given the slow
side marked by green navigation beacons. Because the turning speed. Therefore, with its 122-foot beam, the
As a result of the combined hydrodynamic effects of
Genesis River was crossing from the western side to the Genesis River could not safely pass behind the Voyager.
the bank and the BW Oak, the Genesis River sheered to
eastern side of the channel, Pilot 2 intended for the two At 1516:09, the Genesis River’s bow struck the starboard
port. In an attempt to stop the sheer, Pilot 2 ordered and
vessels to pass starboard to starboard once the Voyager barge midship on the starboard side, penetrating through
the helmsman executed a hard starboard rudder, but
reached the opposite side of the channel. the double hull and breaching the no. 2 starboard cargo
the vessel did not respond, continuing to turn toward
the eastern bank of the channel. Pilot 2 twice ordered The pilot’s direction over the radio confirmed what the tank (causing reformate to leak) and capsizing the
increased engine rpm in an effort to improve steering Voyager’s relief captain had already determined was the port barge. The Houston Ship Channel was closed to
effectiveness. Because the Genesis River was at sea best action, so he immediately increased the Voyager’s navigation for two days during response operations
speed in Nav. Full mode, this increase could not be engine throttles back to full power and put the vessel’s and did not fully open for navigation until May 15.
accomplished by moving only the EOT as it could if the rudders over hard to port. AIS data showed that the two The total cost of damages to the Genesis River and
vessel was in maneuvering mode. Rather, an increase vessels were 0.55 miles apart when the head of the the barges was estimated at $3.2 million. The cost of
in rpm while at Nav. Full required bridge watchstanders tow began turning to port at 1513:35. About the same reformate containment and cleanup operations totaled
to contact the engineering watchstanders to change the time, the relief captain sounded the general alarm and $12.3 million. There were no injuries reported.
maximum rpm setpoint in the ECR, depress the engine radioed the deckhand on watch; the captain arrived in the
control program bypass button, and advance the EOT. wheelhouse shortly thereafter to assist the relief captain.
The probable cause of the collision between

the liquefied gas carrier Genesis River and
the Voyager tow was the Genesis River pilot’s
decision to transit at sea speed, out of
maneuvering mode, which increased the
hydrodynamic effects of the Bayport Flare’s
Figure 38. Barges from the Voyager postaccident. channel banks, reduced his ability to maintain
control of the vessel after meeting another deep-
SAFETY ISSUES
draft vessel, and resulted in the Genesis River
Challenges of navigating large vessels in the Bayport Flare area of the Houston Ship Channel. Due to the sheering across the channel toward the tow.
narrowness of the channel, the large amount of vessel traffic, and the size of the vessels transiting the channel, the
Houston Ship Channel is challenging to navigate and requires significant training and experience. The asymmetric
shape of the channel in the vicinity of its intersection with the Bayport Ship Channel, known as the Bayport Flare, makes
navigation particularly difficult due to varying hydrodynamic forces acting on a vessel’s hull. When larger vessels meet
in the intersection while transiting at a relatively high speed, the risk of loss of control is much greater. SAFETY RECOMMENDATIONS
As a result of its investigation into this
Vessel speed while transiting a narrow channel. Transiting a narrow channel at or near a vessel’s maximum speed
accident, the NTSB issued three new safety
provides little room for error and little ability to increase propeller wash over the rudder to recover if control is lost.
recommendations to the Houston Pilots that
The margin for error is even more limited on ships with slow-speed, direct-drive diesel propulsion engines transiting
focused on requiring vessels to be sufficiently
at Nav. Full, an engine mode designed for higher speeds in open ocean waters where the ability to change engine rpm
trimmed by the stern prior to transiting the Houston
on short notice is significantly restricted. The pilot’s decision to transit the wide-beam, deep-draft Genesis River in
Ship Channel; avoiding conducting passing
Nav. Full mode at sea speed subjected the vessel to greater hydrodynamic forces than had it been traveling at slower
arrangements between wide-beam, deep-draft
maneuvering speeds through the shallow and narrow lower Houston Ship Channel. Additionally, the higher speed
vessels in the northern and southern terminuses
resulted in the vessel trimming further down by the bow, and thus reduced path stability with increased speed due to
of the Bayport Flare; and avoiding transiting at sea
the trim change. Finally, the maneuvering limitations imposed by being at Nav. Full prevented a rapid increase in engine
speed in the lower Houston Ship Channel.
speed when needed to improve rudder effectiveness.
Additionally, the NTSB found that placing as OS
Figure 39. Typical steering sequence during head-on meeting in a narrow, symmetrical channel. (Not to scale.) in training at the helm without informing the pilot
was contrary to good bridge resource management
practice and recommended that K-Line Energy Ship
Management (the operator of the Genesis River)
review its SMS and develop formalized procedures
for watch team reliefs to ensure embarked pilots
are informed of changes in personnel.
Believing that the Bayport Flare would benefit from
regular risk assessments and the consideration
of additional vessel routing measures, the NTSB
reiterated Safety Recommendations M-16-16
and M-16-19 to the Coast Guard to develop a
continuous risk assessment program for each VTS
area, and to establish a program to periodically
review each of the 12 VTS areas.
VESSEL GROUPS
Collision
CARGO, GENERAL • CARGO, DRY BULK
Collision of Cargo
Vessel Nomadic Milde
and Bulk Carrier
Atlantic Venus
Lower Mississippi River, miles 114.5,
South Kenner, Louisiana
May 8, 2020 MAB-21-15
ACCIDENT ID ISSUED Figure 41. Nomadic Milde on the right descending bank at the Cornerstone Dock showing a gash in the hull aft of the port
DCA20FM017 August 11, 2021 anchor pocket. SOURCE: COAST GUARD
O
Figure 40. Image from video taken at 1701 from the n May 8, 2020, about 1655 local time, the anchored of chain in the water with a 9 o’clock lead. At 1533,
bridge of the Atlantic Venus with the Nomadic Milde lying general cargo vessel Nomadic Milde collided with the pilot informed the master that the anchoring was
against its bow. SOURCE: ATLANTIC VENUS CREWMEMBER the anchored bulk carrier Atlantic Venus on the finished and cautioned that there was “considerable
Lower Mississippi River near New Orleans, Louisiana, current,” which he estimated to be between 4 and
after the Nomadic Milde began to swing and drag its 5 knots. The pilot departed the ship at 1542.
anchors in the current. After colliding with the Atlantic The on-watch second officer set an ECDIS anchor watch
Venus, which had been anchored directly behind the alarm, which would have sounded had the vessel moved
cargo ship, the Nomadic Milde then struck a nearby beyond the set radius. However, the anchor watch
chemical dock and grounded on the bank. No injuries alarm was set to a radius of 590 feet, while the initial
were reported. The Nomadic Milde released an estimated estimated distance from the stern of the Nomadic Milde
13 gallons of lube oil into the river. Damage to both to the bow of the anchored Atlantic Venus aft of them
vessels and the dock was estimated at $16.9 million. was 490 feet. Thus, the setting was too large to provide
At 1350, the Nomadic Milde got under way en route for a timely alarm of the ship dragging. From 1551 to
to an anchorage area at Kenner Bend, with a NOBRA 1557, the ship began to swing to starboard toward
pilot conning. At the anchorage, the pilot positioned the right descending bank, moving about 730 feet, a
the vessel between and in line with two bulk carriers in distance double the amount of chain that was set on
ballast, and, at 1515, the starboard anchor was let go. the anchor, indicating that the starboard anchor likely
The Nomadic Milde was then maneuvered toward the dragged. The chief officer noticed that the vessel was
right descending bank, and at 1520, the port anchor not in the center of the anchor watch circle when he
was let go. The anchors were configured with the relieved the watch at 1602, but he did not question
starboard anchor at 360 feet of chain on deck with whether the ship had dragged or check whether the ship
a 12 o’clock lead and the port anchor with 270 feet was remaining securely at anchor.
At 1655, the port side of the Nomadic Milde collided

with the bulbous bow and anchor chains of the The probable cause of the collision between
Atlantic Venus. With the Nomadic Milde broadside the Nomadic Milde and Atlantic Venus was the
to the current and pushing on the bow of the bridge team on the Nomadic Milde not effectively
Atlantic Venus, both vessels moved towards the right monitoring the vessel’s position and therefore not
descending bank and closed on a nearby chemical detecting that the vessel was dragging anchor
dock about 650 feet away. During efforts to assist the and had moved from its original position during
vessels, the Nomadic Milde struck the dock before highwater conditions in proximity to other vessels.
grounding on the right descending bank.
Figure 43. GPS positions
of the Nomadic Milde from
Figure 42. The Nomadic Milde loading cargo in March the vessel’s ECDIS, from the
2018. SOURCE: INTERSHIP NAVIGATION CO. LTD. time the Nomadic Milde’s
starboard anchor was let
At 1613, the Atlantic Venus radioed to request that go at 1515 to the collision
the Nomadic Milde monitor their holding position. with the Atlantic Venus at
However, the chief officer did not take any follow-up 1655. Note: the GPS receiver
action. There was no evidence of either watch officer is located on the main
checking the ship’s position at frequent intervals or mast above the bridge of
by means other than the ECDIS watch alarm. Had the the Nomadic Milde. Scale
ship’s radars been used to determine or crosscheck the approximate.
ship’s position, they would have provided information for BACKGROUND SOURCE:
the crew to determine if the range to a vessel or object NOAA ENC 6LA54M
had decreased or if the ship had moved. Had this been

detected, the master could have been alerted earlier and
undertaken necessary measures to address the problem.
About 1637, the Nomadic Milde, then about 350 feet off
the starboard bow of the Atlantic Venus, began to swing
to port, with its stern moving towards the Atlantic Venus.
By the time the master was called to the bridge at 1642,
he had lost about half the original distance between his
vessel and the Atlantic Venus in which to react to and
mitigate the situation. Knowing that a pilot was required
to get under way in the area, he contacted VTS and
the ship’s agent to request a pilot but was informed he
would have to wait hours. At 1648, the main engine was
ready for use on the bridge, but VTS told the master not
to heave anchor and to maneuver only with their engine
until a pilot arrived. This severely limited the bridge
team’s ability to control the vessel in the strong current,
even while using the main engine up to full ahead, the
bow thruster, and rudder.
VESSEL GROUP
Contact
CARGO, LIQUID BULK
Contact of Tanker
Atina with Oil and Gas
Production Platform
SP-57B
Southwest Pass Fairway Anchorage, Gulf of Mexico,
21.5 miles south-southwest of Pilottown, Louisiana
October 17, 2020 MAR-21-24
ACCIDENT ID ISSUED
DCA21FM004 November 10, 2021
Figure 45. The Atina's damaged accommodation ladder. SOURCE: COAST GUARD
O
n October 17, 2020, at 0446 local time, the tanker At 1448 on October 16, the Atina departed the NuStar
Atina with a crew of 21 was attempting to anchor terminal in St. James, Louisiana. About 0110 the
in the Southwest Pass Fairway Anchorage in following day, the relieving master joined the ship via
the Gulf of Mexico, about 21.5 miles from Pilottown, launch. About 0121, the launch returned to transport the
Figure 44. SP-57B preaccident. SOURCE: COX OPERATING Louisiana, when it struck the manned oil and gas departing master to shore. At 0342 the pilot departed
production platform SP-57B. The platform’s four the Atina, leaving the master and second mate alone for
crewmembers and one technician evacuated to a nearby navigation to Southwest Pass Fairway Anchorage area.
platform by helicopter after activating the emergency The accident master wanted to anchor soon after the
shutdown device to shut in wells to the SP-57B platform. pilot’s departure because he was tired. According to the
No pollution or injuries were reported. Estimated master’s 96-hour work/rest history form, he had no sleep
damages to the platform ($72.3 million) and vessel in the 24-hour period before the accident and 19 hours
($598,400) totaled $72.9 million. of sleep during the 96 hours before the accident. The
company’s SMS required a minimum 1-day 0.9 miles beyond SP-57B. It is likely that
turnover between senior personnel aboard the second mate was looking at the S-band
a company vessel if the oncoming senior radar, set at a 3-mile scale, making platform
person worked for the company (and a 7-day SP-57B difficult to see because it was lost in
turnover if the senior person was new to the radar clutter close to the Atina. The master
company), yet the company requested the was likely looking at the X-band radar, set to a
master change out at Pilottown or elsewhere 1.5-mile scale, making SP-57B easily visible at
on the Mississippi River, leaving no room 0.5 miles.
for the SMS-required turnover. The accident Winds were about 25 knots from the northeast.
master, without any handover period, took The pilot had informed the master of the
command of a vessel, under pilotage on the strong westerly set, and the radar indicated
Mississippi River, at night, after having traveled a set and drift of about 247° at 1.5 knots.
for about 54 hours from his home in Turkey, on However, the master did not adequately
a ship he had never served on. He was likely account for the current and wind that pushed
affected by acute fatigue, defined as getting Figure 47. The Atina’s trackline taken from VDR data his vessel toward SP-57B. Likely preoccupied
fewer than 4 hours of sleep over a 24-hour period. An showing the tanker’s position relative to sea buoy SW with bringing the anchor in clear from the hull, the
overlap would have allowed the incoming master to rest and platform SP-57B within the Fairway Anchorage master ordered hard starboard rudder (with SP-57B
and receive his counterpart’s handover information. boundaries. The image also shows the Atina’s originally on the Atina’s starboard quarter and the wind coming
planned anchorage location, actual anchor drop position, from dead ahead), pivoting Atina toward SP-57B and
and the 4-mile radius from sea buoy SW. putting the wind and current on the Atina’s port side,
BACKGROUND SOURCE: GOOGLE EARTH thus causing the vessel to set toward the platform. As
the platform’s relative position to Atina shifted from the
The location the master chose was about 2.5 miles tanker’s starboard quarter to the starboard bow and the
from the sea buoy “SW” and about 0.7 miles from ship pivoted about the anchor chain, the combination
platform SP-57B. He thought that was a sufficient of set and the Atina moving ahead brought the Atina in
distance from the platform to anchor. contact with SP-57B as the amount of chain in the water
After the Atina began anchoring, the Southwest Pass lessened and the ship gathered speed.
pilot station asked the vessel to move more than
4 miles from sea buoy SW. As the crew heaved anchor
to comply with this request, the bridge team lost track The probable cause of the contact of tanker
of SP-57B. Based on VDR audio, it appears that the Atina with the oil and gas production platform
master believed the platform was another vessel. SP-57B was the Atina’s operating company not
When the master asked what the vessel at 0.6 miles ensuring sufficient time for the master’s turnover,
(SP-57B was at that approximate distance) was doing, which resulted in the master’s acute fatigue and
the second mate provided information for the offshore poor situation awareness during an attempted
supply vessel Leader, located 1.5 miles from the Atina, nighttime anchoring evolution.
Handover Period
Vessel operating companies should ensure that joining crewmembers/personnel are given the
opportunity to obtain a sufficient handover period and adequate rest before taking over critical
shipboard duties, such as navigation, that could impact the safety of the crew, property, and the
Figure 46. SP-57B's fractured horizontal and
environment.
damaged leg.
VESSEL GROUP
Contact
CARGO, DRY BULK
Contact of Bulk Carrier

Atlantic Huron with the
Soo Locks West Center
Pier
Soo Locks, Saint Marys River, Sault Sainte Marie,
Michigan
July 5, 2020 MAB-21-10 Figure 49. Atlantic Huron under way before the accident. SOURCE: SAULTSTEMARIE.COM
O
ACCIDENT ID ISSUED
DCA20FM023 April 13, 2021 n July 5, 2020, about 0250 local time, the At 0245, the captain slowed, allowing the vessel to
self-unloading bulk carrier Atlantic Huron was “coast” toward the west center pier. As speed reduced
transiting the Upper St. Marys River, west of the to 3.8 knots, he ordered full astern. He noticed the
Soo Locks, in Sault Sainte Marie, Michigan, with a crew CPP pitch indicator was “erratic” and received a pitch
Figure 48. Postaccident photo of the Atlantic Huron
of 25. While on approach to the locks and attempting differential alarm, indicating the requested propeller pitch
alongside the west center pier. Point of contact and
to slow, there was a propulsion problem involving the from the helm station did not match actual pitch. He
damage indicated by arrows. SOURCE: COAST GUARD
vessel’s CPP system. The vessel subsequently contacted reported the problem to the ECR.
the west center pier at 6.8 knots. Before reaching the The EOW had not received any alarms. The ECR pitch
lock gate, the vessel’s motion was halted, and the crew indicator matched the bridge’s full astern order. After
moored the vessel to the pier. No pollution or injuries calling the chief engineer, the EOW observed the OD box
were reported. Damages to the vessel ($1,633,000) and shifting forward and backward. Moments later, the
pier ($573,000) were estimated at $2.2 million. captain noticed the vessel’s speed ahead increase. The
The CPP allowed vessel movement ahead or astern chief engineer discovered that the OD box assembly,
without changing the propeller shaft direction. which normally sat on the shaft with bearings, secured
High-pressure hydraulic oil acting on a piston within with a “torque stay,” or locking pin, had rotated on the
the propeller hub would alter propeller pitch. A valve shaft.
assembly above the OD box at the forward end of the The captain announced for the chief engineer to call the
shaft controlled oil flow. Hydraulic oil flowed into the OD bridge “immediately.” He sounded the general alarm and
box and oil transfer tube to the piston within the propeller ordered the second mate to drop the stern anchor. The
hub to rotate the blades. As hydraulic pressure moved captain let go the two bow anchors remotely from the
the piston to achieve desired pitch, this movement bridge. The chief engineer reported the OD box failure
transferred to a mechanical follow-up mechanism, and, with the captain’s approval, stopped the main engine
providing feedback to the control valve assembly and from the ECR. At 0250, the Atlantic Huron’s port bow
pitch indication at the CPP, ECR, and bridge. contacted the pier.
Technicians discovered that a bearing within the CPP’s

OD box feedback mechanism had come out of position
and jammed against the feedback arm 9 days before
the accident, when the (non-accident) captain gave an
astern pitch command and the propeller pitch went to full
ahead. They also discovered that the OD box was able to
move axially on the shaft an “inch or more” due to a worn
torque stay. The damage during this previous voyage
likely was caused by the same underlying mechanical
issue that resulted in the unit’s total failure on July 5.
The set screw securing the feedback ring locking
pin was required to have thread-locking fluid applied
when installed. This set screw was last removed and
reinstalled over 4 years before the accident, during
a shipyard period. When examined postaccident,
technicians found no evidence that thread-locking fluid
had been applied.
Figure 50. Postaccident photo of the west center pier.
The location of initial impact is labeled and evident by a
crushed timber fender. Displacement of a timber crib wall The probable cause of the contact between
concrete cap is also labeled. the Atlantic Huron and the west center pier at
SOURCE: US ARMY CORPS OF ENGINEERS Soo Locks was not following the manufacturer’s
requirement to use thread-locking fluid during
The captain could have stopped the main engine installation of the feedback ring locking pin set
earlier; a main engine emergency stop push button on screw on the vessel’s controllable pitch propeller
the bridge was not activated. Given the short period in system, which led to the failure of the controllable
which the accident unfolded, the captain focused on pitch propeller’s oil distribution box.
reaching the chief engineer to diagnose the problem. The Figure 51. Postaccident photo of vessel’s OD box and
company’s SMS procedures did not contain a policy for control valve assembly as viewed looking aft. Lower
loss of propeller pitch control while in restricted waters. right: Postaccident photo showing damage to backed-out
Crewmembers familiar with such a policy would be locking pin. BACKGROUND SOURCES: COAST GUARD, CSL
better prepared to act quickly and, in this accident, may
have been more apt to stop the main engine.
Following the accident, “severe damage” to the CPP’s
valve block assembly was discovered. The pin holding
the feedback ring in place had backed out and contacted Loss of Propulsion Control Procedures
the feedback arm, damaging both. The set screw holding Loss of propulsion control in a critical phase of operation demands crewmembers act quickly to
the pin in place had also backed out. This failure would mitigate potential accidents. Part of a safety management system should address potential emergency
have inadvertently directed hydraulic fluid, producing a shipboard situations, including loss of propulsion, collision, and contact, and establish ways to respond
full ahead pitch on the propeller blades when an astern to them. Due to their unique blade control, vessels with controllable pitch propellers should have
pitch was ordered. Because the feedback mechanism specific procedures for loss of engine and loss of pitch control. These emergency procedures should be
failed, the pitch indicators would have shown the ordered well understood and practiced by crewmembers both on the bridge and in the engine room.
astern pitch.
O
VESSEL GROUP n September 19, 2019, at 2338 local time, during
Contact
TOWING/BARGE historic flood waters and high river current,

11 barges broke free from a San Jacinto River
barge fleeting area just north of the Interstate 10 (I-10)
Barge Breakaway bridge in Channelview, Texas, and 6 barges struck pier
columns supporting the I-10 bridge. No pollution or
and Contact with injuries were reported. Total damages, including repairs
to the I-10 bridge ($5.11 million) and removal of and
Interstate 10 Bridge repairs to the barges ($350,000), exceeded $5.46 million.
The morning of the accident, 11 barges at the Figure 53. Multibeam sonar images overlayed with
San Jacinto River Fleet were secured side by side with above-surface photograph illustrate the damaged
San Jacinto River Fleet, San Jacinto River, fleeting lines and wire ropes and winches that continued columns postaccident (sonar survey taken September 27)
Channelview, Texas outward from tier 3, with the raked bows facing down on the westbound span of the I-10 bridge. SOURCE: TXDOT
river. An empty tank barge closest to shore was secured
to tier 3’s three mooring pilings and two shore-based Usually, when severe weather approached, tier 3 barges
September 19, 2019 MAB-21-14
“dead men” with synthetic fleet mooring lines. Five towing moved to other tiers to reduce exposure to swifter
ACCIDENT ID ISSUED vessels supported fleet operations; their crews regularly currents. The port captain had planned to move barges
DCA19FM052 June 8, 2021 performed tier checks at the fleeting area’s seven tiers. from tier 3 to tier 2; however, the weather changed
As a result of rainfall from Tropical Storm Imelda, the rapidly, and they did not remove the barges.
Lake Houston water level rose drastically, releasing At 2050, a towboat captain and crew observed the
increasing amounts of water into the San Jacinto River 11 barges at tier 3 breaking away. The current’s force on
via the Lake Houston Dam spillway. The dam’s high the barges had significantly strained the tier 3 moorings
discharge rate during this historic rainfall strengthened and caused the lines to part. Investigators later found
the current at the San Jacinto River Fleet 14 miles parted synthetic mooring lines attached to the pilings
Figure 52. Barges resting against I-10 bridge pilings after down river, increasing the force acting on all barges tied and “dead men.” It is likely that the strain was not
striking and damaging a protective cell. SOURCE: KJRH-TV to tier 3. distributed evenly among the lines due to the current,
resulting in the mooring lines exceeding maximum load
and parting. Once one line failed, the strain would have
been placed on the next line, which would also have
failed, until the breakaway occurred.
Towboats and crews worked to control the breakaway
barges and return them to one of the tiers. Crews were
able to transport the two empty barges that separated
from the string of 11 barges safety to tier 2 east. The
towboat crews struggled to control the remaining
breakaway barges as they moved closer to the I-10
bridge. The increasing river current and rising water
caused mooring lines and wires to part on a barge in
the breakaway group. At 2330, the nine barges topped
around and separated into a block of six barges and a
block of two barges, with one grounded barge remaining
in the mud.
Towboat crews corralled the block of six barges as they

drifted down river. At 2337, they contacted and damaged
the bridge’s protective cell closest to the channel on the
eastern bank and then struck and damaged the western
bank protective cell. The barges subsequently struck the
I-10 bridge fendering system and pilings at 2338. The
barges’ mooring lines and wires parted upon contact,
and two barges became lodged between the I-10 bridge’s
concrete columns on the west side of the river while
four barges continued under the bridge, contacting the
bridge’s fendering system and concrete pilings as they
individually headed down river. Meanwhile, two towing
vessels attempted to move two barges to tier 2, but,
according to the captains, they were unable to maintain
control. The two barges passed under the bridge at 2347
without contacting the bridge structure. Several Good
Samaritan towing vessels pushed the six loose barges
into the east bank below the I-10 bridge.
The San Jacinto River Fleet should have followed its
SMS policies relating to severe weather and swift/flood
water plans, implemented its severe weather plan, and
taken earlier action to break down longer tiers and secure
the vessels in the fleeting area. Had the longer string of
barges at the tier been broken down, the resulting shorter
strings would have been less vulnerable to swift currents.
The probable cause of the barge breakaway and

contact with the I-10 bridge was the force of
the river current acting on the moored barges at
the San Jacinto River Fleet, which exceeded the
capacity of the mooring lines, due to the extreme
rise and flow of water in the San Jacinto River
from Lake Houston dam’s uncontrolled spillway Figure 54. Layout of San Jacinto River Fleet’s tiers. Table in lower left corner shows number of barges at each tier at the
release of water during a historic rainfall event. time of the accident. Photo is not from day of accident. BACKGROUND SOURCE: GOOGLE EARTH
Contributing was the operating company not
rearranging fleeting area tiers to mitigate the
effect of current on barge tiers. Severe Weather Planning
Marine operating companies should develop and continuously evaluate severe weather plans to
prepare for challenges accompanied by tropical storms and/or severe weather with the potential
to cause flooding or swift water within their areas of operations. Severe weather can trigger
prolonged periods of weather restrictions in navigable river watersheds and create challenging
conditions due to high or swift water downstream.
VESSEL GROUP
Contact
TOWING/BARGE
Contact of
Cooperative Spirit
Tow with Hale Boggs
Memorial Bridge Pier
Lower Mississippi River, mile 121.6, near Luling,
Louisiana
Figure 56. Cooperative Spirit moored before the accident.
O
March 15, 2020 MAB-21-05 n March 15, 2020, about 0113 local time, the During the voyage, the tow stopped twice to conduct a
ACCIDENT ID ISSUED towing vessel Cooperative Spirit, pushing a crew change, during which the pilot rotated out, and to
DCA20FM015 February 10, 2021 29-barge tow, was transiting downstream on drop off the empty barge. On March 12, the tow stopped
the Lower Mississippi River at mile 121.6 near Luling, again in Vacherie, Louisiana (mile 151) because there
Louisiana, when the port side of the tow struck the was not adequate space in the fleeting area at the tow’s
eastern tower pier of the Hale Boggs Memorial Bridge. final destination of Kenner Bend (mile 115.8). After the
The tow broke apart and began floating down river. One captain was informed that space had been cleared, on
of the barges sank, while the remaining barges were March 14, about 2317, the tow got under way.
recovered by the Cooperative Spirit and other towing At midnight, the pilot relieved the captain for the
vessels in the area. No pollution or injuries were reported. 0000–0600 helm watch. After passing a bulker at 0052,
Multiple barges in the tow, along with other barges the pilot began setting up to maneuver the tow through
moored along the river banks that were struck by drifting a left-hand bend in the river at 26 Mile Point. Electronic
barges, were damaged and required repairs. Two barges charting system data from the towing vessel showed the
were determined to be total constructive losses. The tow moving toward the left descending bank of the river
estimated cost of damages to the barges and cargo was as it approached the bend.
Figure 55. Cooperative Spirit tow configuration at the time $1.65 million.
of the accident. While flanking around 26 Mile Point between 0055 and
On March 6, 2020, the 200-foot-long towing vessel 0108, the Cooperative Spirit pilot worked the throttles
Cooperative Spirit departed St. Louis, Missouri, for the three engines, using varying astern speeds to
downbound on the Mississippi River with a 30-barge tow control the vessel’s movement. As the tow completed the
and a crew of nine. The tow was arranged six barges flanking maneuver, about 0.8 miles upriver from the Hale
wide by five barges long. The total tow size, including the Boggs Memorial Bridge, the pilot brought the throttles to
Cooperative Spirit, was 1,195 feet long by 210 feet wide. the ahead position and began working to line up the tow
One of the barges was empty; the rest were loaded with to pass through the bridge’s channel span. At 0109:00 the
various grain products. vessel’s port, center, and starboard shafts were turning at
155, 148, and 182 rpm, respectively. The tow’s speed over
ground was 5.2 mph and increasing.
The pilot stated that, as the tow came out of the turn at
26 Mile Point, the stern of his vessel was too close to the
left descending bank, and the current was setting the tow
into the bridge pier. About 3 minutes before the accident,
the tow’s heading was 124°, while its course over ground
was 114°, which is consistent with the pilot’s statement.
Due to high-water conditions, the current was stronger
than normal (6 mph), and an eddy may have formed
upriver of the bridge along the left descending bank,
making maneuvering more difficult.
Although the pilot stated that he used starboard rudder
and increased engine speed to counteract the current,
video evidence showed that he used limited rudder Figure 57. The Hale Boggs Memorial Bridge, with location of bridge strike on eastern tower pier indicated by a red X.
as the tow approached the bridge. The pilot chose to BACKGROUND SOURCE: US ARMY CORPS OF ENGINEERS
primarily use increased engine speed in an effort to move
Figure 58. Track of
the tow to starboard away from the bridge pier, stating
Cooperative Spirit
that he “tried to outrun [the current].” However, the tow’s
tow as it flanked the
course over ground did not appreciably change as engine
bend at 26 Mile Point
speed increased, while the increasing speed over ground
and maneuvered
reduced the time the pilot had to maneuver. Ultimately,
prior to the accident.
the pilot’s actions in compensating for the strong current
BACKGROUND SOURCE:
were ineffective, resulting in the tow hitting the bridge
NOAA ENC US5LA52M
pier at 11.9 mph.
The tow immediately broke apart, and one barge
eventually sank, stern first, about 1 mile down river from
the bridge, with its bow remaining above the water.
The remaining barges floated freely down river, some
contacting barges moored along the river banks, before
they were rounded up by the Cooperative Spirit and other
towing vessels that had responded to the accident.
The probable cause of the contact of the

Cooperative Spirit tow with a pier of the Hale
Boggs Memorial Bridge was the pilot not
effectively compensating for the strong current
while navigating a turn and approaching the
bridge in highwater conditions.
VESSEL GROUP
Contact
CARGO, DRY BULK
Contact of Bulk Carrier

GH Storm Cat’s Crane
with Zen-Noh Grain
Facility
Zen-Noh Grain Facility, Lower Mississippi River,
mile 163.8, Convent, Louisiana Figure 60. The ZGC shoreside facility before the accident. An unknown vessel, similar to the GH Storm Cat, is pictured.
ACCIDENT DATE REPORT NUMBER Runway no. 3 is noted between towers 3 and 4. BACKGROUND SOURCE: ZGC
O
November 11, 2020 MAB-21-20
ACCIDENT ID ISSUED
n November 11, 2020, about 0910 local time, the between cargo holds and to and from shore via the one
DCA21FM006 September 30, 2021 bulk carrier GH Storm Cat’s no. 1 crane boom of the vessel’s four cranes, which were operated by a
contacted the Zen-Noh Grain Corporation (ZGC) vessel crewmember.
facility in Convent, Louisiana, while the crew was Figure 61. ZGC-owned
completing corn-loading operations. The vessel was and -operated payloader
moored starboard side to with a crew of 19. No pollution leveling off a cargo hold.
or injuries were reported. The vessel’s crane was SOURCE: ZGC
undamaged; damage to the shoreside conveyor gallery
was estimated to be $481,006. The morning of the
The afternoon before the accident, on November 10, accident, the vessel’s
the GH Storm Cat arrived at the ZGC facility, which crew was removing a payloader from the no. 1 cargo
specialized in the offloading of soybeans, corn, and other hold, which had been filled. One of the ZGC employees
grains from barges, rail cars, and trucks, and the transfer working in the area attached the crane hook to the
of these commodities through a network of storage payloader, exited the hold, appeared to make a hand
bins and elevators to ocean-going ships. Four towers, gesture in the direction of the crane cab, and then
interconnected by elevated, enclosed structures known crossed over to the port side of the vessel and walked
as runways, acted as distribution points to transfer aft, out of the view of the crane operator in the crane.
product to the ships. Following the accident, the crane operator indicated that
he thought the ZGC employee who had attached the
ZGC employees and vessel crew engaged in corn-loading payloader to the crane hook was acting as his signalman
operations from the shoreside facility into the vessel’s for the lift, but the ZGC employee stated he was not.
Figure 59. GH Storm Cat moored in Cork, Ireland, before five cargo holds. As each hold was filled with corn, a
the accident. SOURCE: JOE MOORE small tractor known as a payloader, owned by ZGC After the crane operator hoisted the payloader from the
and operated by a ZGC employee, was used to level off top of the hold, he lost view of the ZGC employee who
the top of the cargo. ZGC employees and vessel crew had made the connection, who he presumed would
coordinated to transfer the 8,180-pound payloader be the signalman for the lifting operation. Instead of
stopping the lift and establishing communications with

the signalman, as industry recommends, he continued
to slew the load over the dock and lower the boom.
After positioning the crane boom perpendicular to the
vessel and suspending the payloader above the pier,
the crane operator began lowering the boom of the
crane to position the payloader to be landed on the pier.
While lowering the crane boom to land the payloader
on the pier, the boom tip penetrated shoreside facility
runway no. 3. Had the crane operator stopped the lift and
attempted to establish communications when he lost
visual contact with the ZGC employee, he would have
discovered that he was operating without the aid of a
signalman, who likely would have noticed the proximity of
the crane boom to the runway and could have intervened
to prevent the crane striking the structure.
Figure 63. Point of contact of GH Storm Cat’s crane boom and ZGC runway no.3. BACKGROUND SOURCE: ZGC
The GH Storm Cat’s company quality, health, safety, and

environment manual policy required crew to ensure that The probable cause of the GH Storm Cat’s crane
“the operator and/or the signaler have a clear view for contact with the Zen-Noh grain facility runway
the whole path of travel for the load” for lifts. Following was the absence of a dedicated signalman, which
the contact, the crane’s cable was cleared of the runway, led to the ship’s crane operator’s misjudgment of
and the payloader was safely lowered to the pier. Before the location of the crane boom while lowering the
removing the boom tip from the runway, staff on scene payloader to the pier.
completed an initial damage assessment and took
measures to temporarily support the runway.
Vessel Crane Operations

All ships’ crane lifts—no matter how routine—should be adequately planned and risk-assessed.
Figure 62. Video footage still image of the GH Storm Cat’s All personnel involved in the lifting operation should be clearly identified and their duties
crane during the initial sequence of the accident lift understood before the start of the lift.
(lifting the payloader out of ship’s no. 1 cargo hold).
BACKGROUND SOURCE: ZGC
VESSEL GROUP
Contact
TOWING/BARGE
Contact of
Island Lookout Tow
with Centerville
Turnpike Bridge
Albemarle and Chesapeake Canal, mile 15.2 of
South Branch of Elizabeth River to Albemarle and
Chesapeake Canal section of Atlantic Intracoastal
Waterway, Chesapeake, Virginia
November 14, 2020 MAB-21-23
Figure 65. Island Lookout, pushing a barge similar to the BH 2903, under way before the accident.
ACCIDENT ID ISSUED SOURCE: BUNGE GRAIN FACILITY
DCA21FM005 November 9, 2021
O
n November 14, 2020, about 0435 local time, the Island Lookout and its tow into the Albemarle and
the towing vessel Island Lookout was transiting Chesapeake Canal from the Elizabeth River.
eastbound on the Albemarle and Chesapeake At 0427:37 the Island Lookout was headed eastbound on
Canal near Chesapeake, Virginia, pushing ahead barge the canal 0.5 miles from the Centerville Turnpike Bridge
BH 2903, which was loaded with scrap steel. As the (a swing bridge). The mate radioed the bridge’s operator
tow was attempting to pass through the Centerville about this time to request an opening. The mate stated
Figure 64. Simple representation of swing span lighting. Turnpike Bridge, the barge struck the swing span of the that he had to call the bridge operator four times before
bridge while it was opening. No pollution or injuries were he received a response. After each unanswered radio
reported. Estimated damages amounted to $2.86 million call, he slowed the towing vessel’s engines; between
for the bridge and $34,000 for the barge. 0428:07 and 0429:27, the vessel slowed from 4.5 to
At 2335 on November 12, the 65-foot-long towing vessel 3.9 knots. Video footage of the bridge at the time of
Island Lookout departed Baltimore, Maryland, pushing the accident showed no vehicle traffic on the bridge or
ahead the loaded 295-foot-long barge BH 2903, en route approaching roadway, and the bridge operator reported
to Hertford County, North Carolina. The vessel had a no delay in opening due to vehicles on the bridge. The
crew of four: a captain, a mate, and two deckhands. warning gates began to close at 0431:46, and at 0432:29,
On the night of November 13, the mate relieved the the swing span began to open. Although VHF channel
captain for his normal 2200–0500 helm watch. After 13 communications were not recorded, the evidence is
a turnover discussion, the captain went to his cabin to consistent with the Island Lookout mate’s account, and
sleep, and the mate was alone in the wheelhouse. Just the bridge operator likely did not respond to his request
before 0400 the next morning, the mate maneuvered for opening until about 3 minutes 30 seconds after the
first radio call.
When the swing bridge was about halfway open, the

Island Lookout mate determined that the bridge would
not be open in time for the tow to safely pass through.
He attempted to avoid contact with the bridge by putting
both engines full astern. However, the barge turned to
port as the tow slowed, risking a collision with boats
moored at a marina located adjacent to the bridge. The
mate responded to the port turn by moving the port
engine throttle to the ahead position and the rudders to
starboard, which arrested the turn but reduced the tow’s
rate of deceleration. By the time the bow of the barge
was about 175 feet from the bridge, the tow’s speed had
reduced to 2.8 knots, but its momentum carried it toward
the bridge. At 0434:39, the forward starboard corner of Figure 66. Centerville Turnpike Bridge in the closed position. BACKGROUND SOURCE: CITY OF CHESAPEAKE
the barge struck the end of the swing span.
Regulations governing bridges over navigable waters
state that bridges must open promptly and fully for the
passage of vessels. However, bridges can be delayed
in opening for a variety of reasons, so vessel operators
must be prepared to slow or stop in time to prevent
an accident. Using the range lights on the bridge, the
mate on the Island Lookout would have been able to
determine the position of the swing span throughout its
opening sequence and therefore should have had a clear
understanding of the status of the bridge as he made
his approach. He had 10 years’ experience operating
the Island Lookout with a barge in the Albemarle and
Chesapeake Canal and thus understood the maneuvering
limitations of the tow and the restrictions of the
waterway. The evidence suggests, however, that the
mate misjudged his speed of approach relative to the
position of the bridge as it opened and did not sufficiently
slow the vessel in time to safely pass.
Figure 67. Illustration of the Centerville Turnpike Bridge at the time of the accident.
The probable cause of the contact of the Island
Lookout tow with the Centerville Turnpike Bridge
was the mate’s misjudgment of the tow’s speed of
approach relative to the status of the swing bridge
opening, which resulted in insufficient time to
slow the tow and avoid striking the bridge before
it was fully open and safe to navigate.
VESSEL GROUP
Contact
CARGO, LIQUID BULK
Contact of
Liquid Petroleum Gas
Carrier Levant with
Mooring Dolphin
Figure 69. The Petrogas Wharf with a tanker of similar size to the Levant docking starboard side to the wharf.
BACKGROUND SOURCE: PETROGAS, LLC
Petrogas Ferndale Wharf, about 5 miles west of
O
Ferndale, Washington
n December 15, 2019, about 0406 local time, saying they would approach the Petrogas Wharf at a
ACCIDENT DATE REPORT NUMBER the liquefied petroleum gas carrier Levant was 65° heading and then turn the vessel to line up with the
December 15, 2019 MAB-21-02 shifting 0.7 miles from its anchorage to the wharf. The pilot explained there would be a tug fast on
ACCIDENT ID ISSUED Petrogas Ferndale Wharf in Ferndale, Washington, the stern and a tug on the port bow, “same as before.”
DCA20FM006 January 19, 2021 when it struck the wharf’s south mooring dolphin. The There was not an effective information exchange
mooring dolphin and catwalk connecting it to the wharf between the pilot and bridge team prior to getting
were destroyed, and the Levant’s forward ballast tank under way. While certain conditions of the December 14
was penetrated and flooded. There were no injuries master/pilot exchange had not changed since the
to the vessel’s crew or persons on the wharf. There pilot last departed the bridge, variables such as wind,
was no release of pollutants or the ship’s liquified current, and tidal conditions were different for the
cargo of propane and butane. Damage to the vessel docking. Unlike a passage where a vessel proceeds to
was estimated at $1.5 million. Damage to the south sea or picks up a pilot for passage to a dock, the Levant
mooring dolphin and adjoining catwalk was estimated was shifting only a short distance back to a wharf
at $6.75 million. where it had been just hours earlier using the same
The Levant first moored at the Petrogas Wharf on pilot and tugs; a condition of complacency likely existed
December 10, 2019, to load a full cargo. After about among the pilot and members of the bridge team.
3 days, the master decided to take the ship to deeper The pilot began issuing helm and propulsion orders at
water and return to the wharf when the tide was high 0354, and two minutes later he ordered slow ahead.
enough to complete loading operations. When the Following that, the pilot informed the master that the
pilot arrived for the maneuver on December 14, he and current was setting to the north-northwest along the
the master conducted a master/pilot exchange and shoreline at a velocity he calculated at 1.2 knots. At
agreed that the pilot would remain on board until the 0401, at a speed of 4.5 knots, the pilot ordered dead
early morning shift back to the wharf. The Levant then slow ahead. For about the next 2 minutes the master
shifted off the wharf and anchored about 0.7 miles and pilot conversed about non-navigational matters.
away.
Figure 68. Petrogas Wharf on the morning of December At 0403 the pilot ordered the rudder hard to port. The
15 with the catwalk and south mooring dolphin missing. At 0341 on 15 December, the master commenced pilot recalled that, “something didn’t feel right,” so he
BACKGROUND SOURCE: COAST GUARD heaving the port anchor. The master recalled the pilot briskly walked out to the starboard bridge wing. The
Levant’s bow was less than two ship lengths from the Figure 70. The Levant’s positions as it moved from the
wharf’s south mooring dolphin. The Levant approached anchorage to the Petrogas Wharf, based on AIS data.
the Petrogas Wharf at too steep of an angle, at an Vessel scale approximate; vessel at pier was not present
excessive speed for the proximity to the wharf. With at the time of the accident.
two high-powered tractor tugs available and no time BACKGROUND SOURCE: GOOGLE EARTH
pressures, the option to approach the wharf at a slower
speed was available. At 0405, at a speed of 4.5 knots, Figure 71. Below: AIS tracks of all liquid petroleum gas
the pilot ordered stop engine and the tug aft to pull carrier approaches to the Petrogas Wharf from September
straight back. About 1 minute later, the Levant struck 19 to December 15, 2019. Included is the initial approach
the south mooring dolphin. of the Levant to the wharf on December 10 (yellow line),
and the accident pilot’s last approach to the wharf on
The lack of shared mental model between the pilot October 18 (white line). Levant scale approximate; vessel
and bridge team diminished the bridge team’s capacity at pier was not present at the time of either approach.
to monitor the passage and alert the pilot should BACKGROUND SOURCE: GOOGLE EARTH
any deviations from the plan occur. Additionally,
both the pilot and master were likely distracted by a
non-pertinent conversation about 3 minutes before the
contact. The vigilance of a pilot and bridge team should
be increased, not decreased, with a ship’s proximity to
dangers.
By the pilot’s own admission, he was not paying
attention to the vessel’s position in relation to the
dangers a short distance ahead and had no expectation
of the master or the bridge team to provide him
any information. The bridge resource management
fundamentals of planning, communication, use of all
available resources, monitoring, and management of
distractions are essential to operations with a pilot
on board a vessel. Had these fundamentals been
effectively employed, opportunities to detect problems
may not have been missed.
The probable cause of the contact of the liquid

petroleum gas carrier Levant with a mooring
dolphin at the Petrogas Ferndale Wharf was the
pilot’s approach with excessive speed and at
too steep an angle, resulting from the pilot’s and
bridge team’s poor bridge resource management.
Passage Planning for Short Transits

Regardless of the length of the transit or maneuver, the master/pilot exchange is a critical opportunity
for a pilot and bridge team to establish and share necessary information (shared mental model) for the
task ahead.
VESSEL GROUP
Contact
TOWING/BARGE
Contact of Old Glory

Tow with Peter P. Cobb
Memorial Bridge
Intracoastal Waterway, Indian River, mile 965,
Fort Pierce, Florida
August 19, 2020 MAB-21-13
ACCIDENT ID ISSUED
DCA20FM025 June 2, 2021 Figure 73. The ICW approach to the Peter P. Cobb Memorial Bridge from the south.
O
n August 19, 2020, about 0251 local time, the At 0248, about a quarter mile from the bridge, both
towing vessel Old Glory, pushing the loaded the Old Glory and Cole were now outside the channel’s
hopper barge Cole northbound on the ICW, western limit by about 65 feet. The relief captain
struck the protective fendering for the Peter P. Cobb explained that the current took the head of the tow to
Memorial Bridge at mile 965 near Fort Pierce, Florida. the west (port), which he was not expecting; he applied
There were no injuries to the four crewmembers on the rudder correction to starboard. At 0249, the tow had
Old Glory, and there was no pollution reported. Damage returned to the channel’s western limit with a speed of
to the barge was estimated at $5,000. Damage to the 6.1 mph. The relief captain observed the current again
protective fendering was $641,000. set the tow to the west just before the vessel reached
About midnight on August 19, 2020, the relief captain the bridge.
took over the watch. At 0245, the Old Glory was about As he approached the bridge, the relief captain said
a half mile south of the Peter P. Cobb Memorial Bridge. he began to slow the tow, which in turn reduced the
According to the relief captain, the current started to set maneuverability of the tow while the current pushed the
the tow from the center to the west side of the channel. tow to the western limit and then outside of the channel
About 0246, the barge’s bow was outside the channel’s into the northern approach to the Fort Pierce City
Figure 72. Old Glory under way. western limit. The relief captain said he was caught Marina channel. The relief captain explained that “the
SOURCE: RIVER VENTURES, LLC
off guard, not expecting the current to be running, boat started bogging down,” and he could “not bring the
considering that the time he was passing through the head back up.” He tried to “twist” the Old Glory’s stern to
bridge was near low tide (which he understood to be get the “head [of the tow] in there, as it is, to go through
0230, rather than 0340, the predicted low tide at the the bridge.” At 0250, the head of the tow neared the
location nearest to the accident). Both the United States southern end of the west protective fendering. The last
Coast Pilot and navigational charts had information on time the relief captain checked, the vessel’s speed was
“strong cross” and “extremely fast” currents near the 4 mph. He noted he was “backing down,” but not getting
Peter P. Cobb Memorial Bridge. the power he expected.
The relief captain steered the tow into the bridge’s in the Coast Pilot publication, he did not use all the
east fendering, which he considered the safest place resources available to him. Towing vessel regulations
to strike since it did not affect any bridge structures. require the officer of a navigational watch to conduct a
At 0251, the barge’s port side touched up against the navigational assessment, using all resources available
west-side fendering, and its starboard bow struck the to gather information on conditions that could impact
middle eastern fender wall. The starboard-side face the safety of navigation. Had the relief captain been
wire also parted. aware of the cautionary note and information contained
The tow became wedged under the bridge, where the in the Coast Pilot, he would have been better prepared
vessel owner, and the crew replaced the starboard face to address the risk of strong currents often seen near
wire. According to the relief captain, “when the tide the bridge.
finally changed,” the tow “straightened itself out” under
the bridge and floated out to the south with the rising The probable cause of the contact of the
tidal flow, and the tow was back under way and passed towing vessel Old Glory and barge Cole with
through the bridge at 0635. the Peter P. Cobb Memorial Bridge protective
Although the relief captain acknowledged the available fendering was an inadequate navigational
navigational information on the vessel’s ECS and assessment that did not identify the risk of strong
cross-currents in the area of the bridge transit.
Figure 75. AIS trackline of the Old Glory and Cole outside

of and alongside the western boundary of the ICW.
Familiarization with Local Information BACKGROUND SOURCE: PORTVISION
The Coast Pilot and navigational charts are valuable sources to mariners that contain amplifying
information on local conditions such as tides and currents, channel characteristics, and bridge
descriptions. It is important to check the Coast Pilot and charts when developing voyage plans to
improve knowledge of an area and prepare for a safe passage.
Figure 76. The approximate position of the tow after

striking the protective fendering system under the
Peter P. Cobb Memorial Bridge.
Figure 74. Postaccident damage to the eastern fendering of the bridge looking to the south (left) and to the north (right). SOURCE: NOAA CHART USFL88M
SOURCE: COAST GUARD
VESSEL GROUP
Contact
TOWING/BARGE
Contact of
Savage Voyager Tow
with Jamie Whitten
Lock & Dam
Jamie Whitten Lock & Dam, Tennessee-Tombigbee
Waterway, mile 411.9, near Dennis, Mississippi
September 8, 2019 MAB-21-06
ACCIDENT ID ISSUED
DCA19FM049 February 19, 2021
Figure 78. A tow with a similar arrangement to the Savage Voyager locking down in the lock chamber after the accident,
with newly repainted warning line. BACKGROUND SOURCE: US ARMY CORPS OF ENGINEERS
O
n September 8, 2019, at 0355 local time, the River from Hartford, Illinois, en route to Tuscaloosa,
towing vessel Savage Voyager and its tow of two Alabama. The tow’s barges, SMS 30056 and the
loaded tank barges were engaged in southbound PBL 3422, were secured in a line, stern to stern, with the
locking operations at the Jamie Whitten Lock & Dam SMS 30056 forward and the PBL 3422 aft, pushed by the
on the Tennessee-Tombigbee Waterway, 6 miles from Savage Voyager. The vessel was 83.5 feet long, and each
Dennis, Mississippi. After lock operations began, the bow barge was 297.5 feet long, giving the tow a total length of
of barge PBL 3422 contacted the lock’s upper gate sill 678.5 feet.
and was hung up as the water level dropped, resulting By 0330 on September 8, the tow had arrived at the
in hull failure and a cargo tank breach. About 117,030 Jamie Whitten Lock & Dam on the southern end of Bay
Figure 77. Preaccident photo of Savage Voyager. gallons (2,786 barrels) of crude oil were released into the
SOURCE: JEFF CUMPTAN Springs Lake, Mississippi. The pilot radioed the lock
lock. No injuries were reported. The damaged barge cost operator to request permission to conduct downbound
$402,294 to repair, and costs to return the lock to service locking operations. About 0345, the lock operator granted
18 days later were about $4 million. permission, and the pilot maneuvered the tow through
On September 4, 2019, at 1620, the Savage Voyager’s the open upper gate and into the 600-foot-long lock
tow got under way southbound on the Upper Mississippi chamber.
Figure 79. Depiction of Savage Voyager tow prior to locking down (towboat position approximate). Figure 80. Depiction of the PBL 3422 in lock chamber
during lock operations (not to scale).
The on-watch deckhand and tankerman followed the 3 minutes and 58 seconds for the water level to drop to Once the PBL 3422’s rake became hung on the concrete
Corp of Engineers’ standard operating procedures and the sill. The PBL 3422’s tow knees were 1.6 feet below miter sill, the deckhand notified the pilot, who sounded
used two lines (one forward and one aft) to secure the the surface of the water; therefore, a 21.4-foot drop the towboat’s general alarm and radioed the lock
two barges inside the lock chamber. The 678.5-foot-long would likely have resulted in contact with the sill. With operator to halt the locking process. However, since
tow would not fit in the lock chamber, so the crew the water lowering at a calculated average rate of about it took over 2 minutes for the emptying valves to fully
“knocked out” the Savage Voyager and secured the 1 foot every 10.3 seconds, after about 3.7 minutes, the close, the water continued to rapidly descend in the lock
towboat alongside the barges, ensuring the tow did not barge’s tow knees would have contacted the sill. The chamber, and the barge became hung on the sill, bending
cross the yellow warning line that marked the upper deckhand stated that he noticed the barge was stuck the rake and breaching the forward cargo tank before the
gate’s submerged miter sill. “maybe a couple seconds after” locking began. However, barge dropped into the water.
The pilot radioed the lock operator to inform him that it would have taken over 3 minutes after lock operations
the crew was ready to commence lock operations. The began for the barge to contact the miter sill, so it is The probable cause of the contact of the
lock operator closed the upper gate, then closed the unlikely that the deckhand was attentively minding the Savage Voyager’s tow with the Jamie Whitten
fill valves and opened the emptying valves so that the stern line. Additionally, the tankerman was not aware that Lock & Dam was the tow moving out of position
water level began to drop. During the locking—a total the vessel was out of position until the deckhand radioed in the lock chamber while locking down when the
descent of 83 feet at the time—the deckhand (on the him. Had the deck crew been vigilantly monitoring the crew did not effectively monitor and maintain the
rake of the PBL 3422) and tankerman (on the rake of the vessel’s position, they would have noticed the barge was vessel’s position during its descent, resulting in
SMS 30056) were responsible for tending their respective out of position before it became stuck on the sill and the aft barge becoming hung on the upper gate
lines to keep the tow within the lock chamber. With the could have alerted the pilot. miter sill.
Savage Voyager at the barge’s side, the tow’s length
was reduced to 595 feet, and the crew had only 5 feet of
clearance in which the tow might safely move, leaving
a very small margin of error and requiring the deckhand Vigilance During Lock Operations
and tankerman to closely watch their respective lines. Although locking operations can seem routine, the margins for safety are frequently low.
Maintaining vessel position and communication with the lock operator are critical practices to
At some point during locking, the PBL 3422 crossed ensure safe lockage. Crews should avoid complacency and vigilantly monitor lines at all times to
the upper gate’s yellow warning line, placing the tow in prevent “running” in a lock.
danger of contacting the miter sill. Postaccident testing
by the Corps of Engineers showed it took a combined
VESSEL GROUP
Contact
TOWING/BARGE
Contact of
Trent Joseph Tow
with Barataria Bridge
Barataria Waterway, Barataria, Louisiana
Figure 82. Postaccident damage to Barataria Bridge looking north. SOURCE: ANONYMOUS
O
November 22, 2020 MAB-21-22 n November 22, 2020, about 2122 local time, About 2100, the tow turned into the Barataria Waterway;
ACCIDENT ID ISSUED the towing vessels Trent Joseph and George C about 0.9 miles ahead was the Barataria Bridge, a swing
DCA21FM008 November 1, 2021 together were towing two barges southbound bridge owned and operated by the Louisiana DOTD.
in the Barataria Waterway near Barataria, Louisiana. Because of the following current, which was estimated
While passing through the open Barataria Bridge, the to be about 1.4 mph, the captain of the Trent Joseph
second barge contacted the bridge’s swing span. The ordered the tail boat George C to “clutch reverse.” About
bridge, which was the only means of road access for the 2104, the captain of the Trent Joseph called the Barataria
community of Barataria, was damaged and remained Bridge tender by radio to request it be opened. The bridge
unusable until November 28. There were no injuries tender logged the bridge as being open at 2110, when the
and no pollution reported. Damage to the barge was Trent Joseph was about 0.5 miles north of the bridge.
negligible, while damage to the bridge was reported to About 2121, when the Trent Joseph was about 100 feet
be more than $500,000. from the bridge, the captain saw the northeast-facing
About 2003, the tow was on the Gulf Intracoastal corner of the open swing span had “over-rotated” and
Waterway west, en route to Grand Isle, Louisiana. The extended past its protective fender wall. He called the
67-foot-long Trent Joseph was operating as the lead boat captain of the George C by radio and told him to keep an
and towing two barges behind in a single string by means eye on the bridge and to “do what he had to do” to keep
Figure 81. Trent Joseph (above) and George C (below)
of a tow bridle and shock line. The combined length the barge from hitting it.
under way before the accident.
of the barges was about 404 feet, with the larger aft As the Trent Joseph passed through the bridge’s
SOURCES: COASTAL TOWING, LLC; CVITANOVIC TOWING, LLC
51-foot-wide barge extending about 3 feet wider on each navigable channel, the barges began to fall to starboard
side than the forward barge. towards the swing span and protective fender. The
Aft of the barges, connected by a tow line, was the captain of the Trent Joseph corrected (steered) to
68-foot-long tail boat George C. Acting based on the port. The George C’s captain was then able to see the
instructions given by the captain of the Trent Joseph, the swing span (his view had previously been obstructed
tail boat would use its propulsion and steering to keep by equipment on the aft barge), and he, too, noticed the
the barges from running over the lead boat if it were to bridge span protruding past the fender. He called the
slow down or stop, or to help move the tow to port and captain of the Trent Joseph but did not receive a reply,
starboard as needed. Overall, the estimated length of the so he kept his vessel in clutch reverse. He noted that
tow was 624 feet. after the correction, with the fair tide, everything went
“sideways” to the bridge.
At 2122, at a speed over ground about 2 mph, the Photos taken at the scene and a postaccident survey Louisiana DOTD maintenance records indicated that, 2
starboard forward side (port aft quarter) of the aft barge report confirmed that the swing span’s beam was struck days before the accident, work had been conducted on
struck the northeast-facing corner of the swing span and by the aft port corner of the aft barge. By design, the the bridge’s limit switches (which prevent the movement
dislodged it from its mounting and supports. After the protective fender was in place to shield the swing span of the swing span beyond a predetermined point).
accident, the swing span was lifted back into the closed from being struck when rotated to the open position. As Additionally, work had previously been performed on
position and opened to vehicular traffic on November 28, such, if the swing span had been behind the fender as the limit switches on four occasions dating back to
2020; the bridge remained closed to marine traffic and it should have been, the tow would have contacted the October 2019. Although no detail was provided in the
was later destroyed during Hurricane Ida on August 30, fender instead of striking the swing span. Although the maintenance records as to what sort of work and return-
2021. fendering system contained previous damage, if the tow to-service testing was conducted, these records indicate
had contacted the fender, damage (if any) to the fendering a recent issue with the span’s opening rotation limit.
would have been minimal due to the tow’s low speed.
The probable cause of the contact of the tow of the Trent Joseph with the Barataria Bridge was a corner of
the bridge’s swing span protruding outside of its protective fendering into the navigable channel after recently
attempted repairs to the limit switch system that controlled the swing span’s rotation limit.
Figure 83. Estimated position of the Trent Joseph tow as Figure 84. Left: Simple representation of fendering and swing span lighting; profile (top) and plan (bottom) views.
it passed through the Barataria Bridge and approximation BACKGROUND SOURCE: COAST GUARD. Middle and Right: Postaccident damage to the corner of barge JMSS Mobile, with
of the contact of barge JMSS Mobile with the northeast damage markings and scrapes about 6 feet above the waterline, and to the northeast corner of the Barataria Bridge, with
corner of the swing span. (Navigation lights not to scale.) points of contact with the barge JMSS Mobile indicated. SOURCE: ANONYMOUS.
VESSEL GROUP
Fire/Explosion
YACHT/BOAT
Contact Fire aboard

Private Yacht Andiamo
Island Gardens Deep Harbour Marina, Miami, Florida
December 18, 2019 MAB-21-17
ACCIDENT ID ISSUED
DCA20FM007 August 27, 2021 Figure 86. Andiamo docked in Miami before the accident. SOURCE: MARINEVIDEOPRODUCTION.COM
Figure 87. Plan view of the lower deck showing the dresser in the port VIP suite where the candles were lit.
SOURCE: BENETTI FIRE & SAFETY PLAN
O
n December 18, 2019, about 1921 local time, a On the afternoon of December 18, the crew of the
fire broke out aboard the privately owned yacht Andiamo was preparing for the arrival of a guest of
Andiamo while moored at the Island Gardens Deep the owner. While preparing the port VIP suite, the chief
Harbour Marina on Watson Island in Miami, Florida. The stewardess and second stewardess noticed that the
Figure 85. Andiamo listing to starboard as fireboats crew of four and a guest on board safely evacuated the lights throughout the lower deck and in the main salon
attempt to extinguish the fire. SOURCE: MIAMI-DADE FIRE vessel as the fire quickly spread. While local firefighters on the main deck above were not working. They reported
RESCUE
and crews from neighboring yachts attempted to the issue to the captain, who believed the problem was
extinguish the fire, the yacht capsized onto its starboard connected to the automated lighting computer on the
side. No injuries were reported, but an oil sheen was bridge. Since the chief engineer was not on hand, the
observed. Total damage was estimated at $6.78 million: captain solicited guidance from an “engineer friend” over
the Andiamo, valued at $6.3 million, was declared a the phone.
constructive total loss; repair costs for the marina and About 1910, the owner’s guest arrived on board the
adjacent vessels were $480,000. vessel and was escorted to the VIP suite. To illuminate
the suite, the chief stewardess lit three candles and Neighboring vessels and responders from the City of
placed them on top of the wood veneer dresser Miami Fire Rescue and Miami-Dade Fire Rescue fought
beneath a porthole decorated with curtains above. After the fire. However, as firewater flooded the upper decks
extinguishing one but leaving the other two lit candles of the vessel, the Andiamo started to list to starboard,
unattended, the chief stewardess followed the guest up rolled, and capsized on its starboard side, coming to
to the sky lounge. rest on the marina’s sea floor at 2130. The fire was
When the chief stewardess went down to the galley to extinguished at 2220.
pick up refreshments for the guest—she estimated about The postaccident fire report revealed the fire originated
3 minutes later—the second stewardess and chef told her in the port VIP suite. As the Andiamo’s flag state warned,
that they noticed “a funny smell” in the main salon. The “leaving open flames such as…candles unattended”
chief stewardess then opened the door to the main salon poses a fire risk. The candles also had not been secured
and saw a plume of black smoke about 4 feet high from on candle holders or any other type of secondary
the deck. containment to ensure they would remain stationary, a
The captain, who was on the bridge, heard the two precaution particularly important on a vessel likely to
stewardesses yelling, “Fire!” He investigated but could sway even within its berth. As the open flames burned,
not determine the origin of the fire. He instructed the the curtains hanging above the dresser nearby likely
chief stewardess to have everyone evacuate the vessel provided the combustible material that started the fire.
and to call for help. At 1923, the second stewardess The vessel’s interior spaces were framed in wood with
called 911. veneer, as well as outfitted and furnished with wood and
other flammable materials, allowing the fire to spread
Most of the crew evacuated the burning vessel. The upward.
captain and chef attempted to connect a fire hose, but
both realized “there was no fighting it,” so they also
evacuated the vessel. The probable cause of the fire aboard the
private yacht Andiamo was burning candles left
Although several of the rooms and adjoining spaces on unattended that resulted in an undetected fire in a
the lower deck were equipped with smoke detectors, Figure 88. Promotional and postaccident photos
guest cabin. Contributing to the severity of the fire
the fire-detection and alarm system for the vessel had indicate location of the candles (yellow rectangle) in the
was the crew’s failure to complete timely repairs
been inoperable during the two months before the fire, unattended port VIP suite. SOURCE [TOP PHOTO]: MCA YACHTS
to a fire-detection and alarm system known to be
as reported by ABS on October 2. While attempts were inoperable for two months.
being made by the crew to repair the system, multiple
visits from ABS indicated the system and alarms were
not functioning. If fully functional, the fire-detection and
alarm system would have alerted the crew of the fire’s
location at its onset and thus provided an opportunity for
a direct response. Earlier detection of the fire likely would Avoiding Candle Use on Vessels
have allowed the crew to suppress the fire with onboard According to the National Fire Protection Association, burning candles results in hundreds
equipment such as handheld fire extinguishers. of millions of dollars in damages ashore in the United States, including injury and loss of life.
Candle usage on a vessel, whether attended or not, also poses a fire risk. Given the dynamic
After all the crewmembers and guest were safely on
environment of a vessel, candles can move, and their open flames can ignite combustible
the dock, the captain instructed them to alert adjacent
materials. The abundance of flammable materials on board can allow a fire to quickly spread out
vessels. He also shut down the shore power to the vessel
of control. Flashlights and battery-powered lighting are safer alternatives to use during a loss of
by opening the breakers at the electrical power pedestal
electrical power.
on the pier.
VESSEL GROUP
Fire/Explosion
TOWING/BARGE
Engine Room Fire

aboard Towing Vessel
City of Cleveland
Lower Mississippi River, mile 348, near Natchez,
Mississippi
ACCIDENT DATE REPORT NUMBER Figure 90. City of Cleveland under way before the accident. SOURCE: JEFF L. YATES
February 26, 2020 MAB-21-04
O
ACCIDENT ID ISSUED n February 26, 2020, about 1600 local time, the The doors from the weather deck to the engine room
DCA20FM014 January 27, 2021 towing vessel City of Cleveland was pushing were closed since the vessel was under way, but
18 dry cargo barges (15 loaded and 3 empty) the exterior windows were open for engine room
upbound on the Lower Mississippi River, approximately ventilation. The crew was unable to close them due to
15 miles south of Natchez, Mississippi, when the vessel heat and smoke from the fire. The steel door between
experienced a main engine failure followed by an engine the rudder room and the upper engine room was open,
room fire. All nine crewmembers safely evacuated to and fire spread aft to that space. The two semi-portable
the barges and were rescued by nearby Good Samaritan extinguishers, stowed on each side of the upper engine
vessels, which worked to extinguish the fire. The room, were the only means for the crew to fight the
Figure 89. Flames coming from the starboard rudder City of Cleveland was later towed to the operator’s conflagration. However, the portside extinguisher
room door. SOURCE: COAST GUARD facility in Rosedale, Mississippi. No pollution or injuries was inaccessible due to the flames, and the hose on
were reported. Damage to the vessel was estimated at the starboard extinguisher failed proximate to the
$2 million. extinguisher.
The vessel left New Orleans, Louisiana, on February 24.
On February 26, about 1600, crewmembers reported Figure 91. Semi-portable extinguisher in starboard engine
hearing a loud or strange noise that “sounded—it felt room following the fire.
like it was a log in the wheel.” The port main engine
stopped, and the pilot and first mate immediately saw
flames from the open engine room ventilation housings
on the upper deck. From his office, the engineer saw
that the upper engine room was engulfed in fire. The
running generator in the upper engine room then
stopped, and the vessel lost power, rendering the
vessel’s fire hoses unusable. The starboard main engine
continued to run.
The engine room was not equipped with a fixed

fire-extinguishing system, nor was it required to
be by existing regulations. If the vessel had a fixed
fire-extinguishing system in the engine room, as
well as a means to close the ventilation and open
windows in the engine room from the outside, the fire
may have been able to be extinguished. Without an
effective means to fight the fire, the crew was forced
to evacuate to the tow’s barges. Between 1700 and
1745, two nearby Good Samaritan vessels arrived
and assisted with securing the City of Cleveland and
its tow, evacuating the crew, and fighting the fire.
Crewmembers told investigators that the fire in the
rudder room proved challenging to fight, as steering
gear hydraulic oil, having spilled from burnt hoses, was
burning while floating on the firefighting water and was Figure 92. City of Cleveland main deck arrangement.
spreading further by the hose streams. The fire was
extinguished by 1900, and the crew was taken ashore.
Figure 93. Holed crankcase at no. 4 left cylinder inspection cover (left), no. 4 left and no. 4 right connecting rods
During a postaccident examination of the wreckage, (middle), and connecting rod clamp (right).
the no. 4 left and right cylinder master connecting rods
and articulating rods were found still attached to each
other outside the crankcase, and the piston pin and
connecting rod clamp were missing. Therefore, the
initial failure was likely of the connecting rod clamp
or the bolt that held the piston pin. Regardless, the
force of the connecting rods driven loose inside the
engine was enough to puncture a hole in the side of the
crankcase on the port side and eject the piston head
through the no. 4 right cover on the starboard side.
The failure of the connecting rod and subsequent The probable cause of the fire aboard the towing vessel City of Cleveland was the catastrophic failure
catastrophic damage to the crankcase likely allowed and crankcase breach of the port main engine resulting from the failure of a connecting rod assembly.
hot pressurized fuel and oil to spread to the lower Contributing to the severity of damage to the vessel was the lack of a fixed fire-extinguishing system for the
engine room and thereby ignite. The significant heat engine room, as well as the loss of electrical power to the single fire pump.
and smoke damage to the upper engine room indicated
that the fire spread up the port side of the engine room,
then aft in the upper engine room, and eventually to the
rudder room. The air drawn through the open windows Engine Room Fires
in the upper engine room likely further exacerbated the Engine rooms contain multiple fuel and ignition sources, making the spaces especially vulnerable
fire’s spread. to rapidly spreading fires. Designers and operators of towing vessels should evaluate fire hazards
and provide effective means to mitigate them. Operators should have equipment and procedures
in place to quickly contain and suppress engine room fires before they can spread to other spaces
and/or cause a loss of propulsion and electrical power.
VESSEL GROUP
Fire/Explosion
CARGO, GENERAL
Fire aboard
Roll-on/Roll-off
Vehicle Carrier
Höegh Xiamen
Pier 20, Blount Island, Jacksonville, Florida
June 4, 2020 MAR-21-04
ACCIDENT ID ISSUED
DCA20FM020 December 1, 2021 Figure 96. Firefighters conducting exterior boundary-cooling the day after the fire was discovered. SOURCE: JACKSONVILLE
FIRE AND RESCUE DEPARTMENT
O
n June 4, 2020, about 1500 eastern daylight time, scheduled 1700 departure for Baltimore, Maryland, to
the crew of the pure car and truck carrier (a type of load the last of its transatlantic cargo.
Ro/Ro vessel) Höegh Xiamen, completed loading While waiting for rain to subside in order to secure the
used vehicles on board the vessel while docked at the stern ramp, the chief mate saw smoke coming from a
Port of Jacksonville, Florida. While preparing the vessel ventilation housing for one of the exhaust trunks that ran
for departure, the chief mate noticed smoke coming from deck 12 to one of the cargo decks.The chief mate
from a ventilation housing for one of the exhaust trunks immediately informed the crew over his radio that smoke
that ran between deck 12 (the weather deck) and one was on cargo decks 7 and 8. The second mate in turn
of the cargo decks. The crew found a fire on deck 8 informed the master and chief engineer. The chief mate
Figure 94. Höegh Xiamen under way before the accident. and attempted to fight the fire before being relieved by reactivated the fire detection system at 1545, which had
SOURCE: HÖEGH TECHNICAL MANAGEMENT INC. shoreside firefighters. The fire was extinguished over a been secured (not activated) in accordance with cargo
week later on June 12. None of the 21 crewmembers on loading procedures. The system immediately alarmed,
board were injured; nine firefighters sustained injuries indicating the presence of smoke.
while responding to the fire. The Höegh Xiamen and
its cargo of 2,420 used vehicles sustained significant Crewmembers discovered a fire on cargo deck 8. The
damage due to the fire and were declared a total loss chief mate sent the vessel’s electrician to close the
valued at $40 million. remotely controlled ventilation dampers, which slowed
but did not stop the smoke (the ventilation system’s
Between June 3 and June 4, 2020, the crew of the manually operated dampers remained open). The crew
600-foot-long, Norwegian-flagged Höegh Xiamen worked attempted to fight the fire but were repelled by heavy
Figure 95. Thermally damaged vehicles after removal with shoreside stevedores to load vehicles on board the smoke. The master instructed the chief mate to close all
from the Höegh Xiamen, Blount Island, Jacksonville, vessel. About 1500 on June 4, loading was completed, the manual cargo deck ventilation housing dampers on
Florida, July 23, 2020. and the vessel’s crew began preparing for the vessel’s deck 12 (the weather deck).
Figure 98. Höegh Xiamen fire zones. The vessel’s gas-tight decks, which separate fire zones, are highlighted orange.
BACKGROUND SOURCE: HÖEGH TECHNICAL MANAGEMENT
Figure 97. Arrangement of portside aft cargo ventilation

trunks on deck 12. Each exhaust is marked with an “E,” and
each supply is marked with an “S.” This only represents a
portion of the exhaust and supply ventilation trunks.
Beginning at 1549, the master made several calls for

help over VHF radio to “Jacksonville Port Control,” an
entity that did not exist (the NTSB and Coast Guard
investigators were unable to determine which channel
was used). At 1554, an unknown vessel answered the
VHF call and advised the master to switch to channel 14
to reach the pilot station, and he did so. The pilot station
relayed the distress call to the Coast Guard.
The Coast Guard hailed the Höegh Xiamen on channel
16 several times beginning at 1555:49. The master
returned to the radio at 1558:20 and informed them that
there was a fire on deck 8 and requested assistance. Figure 99. Damaged aft ventilation housings after the reported explosion of portside vent housing for deck 9 (9E in the
He did not specify where the vessel was moored when photo). BACKGROUND SOURCE: JACKSONVILLE FIRE AND RESCUE DEPARTMENT
asked, nor did he use the radio’s distress button. Neither
the master nor any other crewmembers answered Shoreside firefighters from the Jacksonville Fire and stairwell. Initially, there was no heat or smoke in the
subsequent radio calls. Rescue Department arrived at 1603 and relieved the stairwell, and they found the doors to each deck (except
About 1559, an onshore witness who had observed crew. The captain, after consulting with and receiving deck 12) closed. There was no pressure behind the doors
smoke coming from the vessel called 911 emergency concurrence from the fire department, had CO2 from when they accessed decks 7 and 8.
services to report the fire. Shortly after, the nearby the vessel’s fixed fireextinguishing system released into They encountered two smoldering vehicles and a small
passenger vessel Norwegian Pearl reported to the decks 7 and 8, and the crew then evacuated from the amount of fire on the port bulkhead on deck 8. After
Coast Guard that the Höegh Xiamen was at Berth 20, Höegh Xiamen. The fire continued to spread to the higher cooling the cars and putting out the small fire, the heat
they could see shoreside responders were en route, and cargo decks and the accommodations. Firefighters seemed to increase substantially, so they retreated from
the ship was accessible from shore. decided to enter decks 7 and 8, again from the port aft the deck.
that battery securement crews were unable to gain

Fire/Explosion
access to the engine compartments. If they had followed

Grimaldi’s procedures, these vehicles would have been
rejected and would not have been loaded on board the
vessel. Instead, the stevedores flagged these vehicles
(once loaded) by raising the windshield wipers and
wrapping them in caution tape.
Ineffective oversight of vehicle battery securement.
Grimaldi’s port captain had the ultimate discretion as
to whether to accept any vehicle for loading, as well as
oversight authority to ensure that cargo was properly
secured and in a safe condition. However, during loading
operations, the port captain missed opportunities to
require longshoremen to properly isolate the vehicle
Figure 100. Left: Engine 48 arriving on scene about 1603. Starboard vents 10/11E and 7/8E and port vents 10/11E and electrical systems. The Coast Guard’s postaccident
9E exhibit smoke flow. Right: About 1608, 5 minutes after Engine 48’s arrival, the white paint on the starboard side of the examination of a sample of 59 vehicles did not find
vessel exhibits thermal discoloration (circled in orange). Starboard vents 10/11E and 7/8E and port vents 10/11E and 9E a single battery that was secured in accordance with
exhibit smoke flow. SOURCE: JACKSONVILLE PORT AUTHORITY Grimaldi’s battery disconnect procedure. Even from
random and cursory inspections, it should have been
Firefighters were assigned to deck 12 to search for SAFETY ISSUES immediately obvious to the port captain that the battery
hatches to open for ventilation to evacuate smoke and Lack of training for vehicle battery securement. disconnection crews were not correctly performing
improve visibility on the decks below with fires. Once Grimaldi Deep Sea, the vessel’s charterer, provided their tasks. Additionally, an SSA stevedore gave the
they arrived on deck 12, about 1846, they were ordered SSA Atlantic stevedores with their battery disconnect Höegh Xiamen’s chief mate a “Vehicle Lashing Inspection
to open any doors to the housings around the ventilation procedure. However, after the accident, Coast Guard Procedure” document that indicated that 58 vehicles
trunks. About 60 seconds after firefighters on deck 12 investigators examined several of the used vehicles loaded onto various decks had “incomplete” battery
opened the exhaust vent for deck 9, they heard “a loud loaded on board the vessel and found improperly secured disconnections. Although the chief mate signed the
roar that sounded like a jet engine,” and the ventilation batteries. Stevedores stated that some of the vehicles procedure, he did not take any action to address the
housings for the decks 9 and 10/11 trunks “exploded” stored on decks 7 and 8 had sustained so much damage hazards noted on the procedure.
and were destroyed. Nine firefighters who were working
in the stairwell or who had been staged on deck 5 were
burned, five of them seriously, by the superheated air that
rushed down the stairwell. It is likely not coincidental that
the “explosion” occurred about the same time that the
firefighters opened the exhaust. On their way to deck 12,
firefighters had opened the deck 9 door from the stairwell
and found thick, black smoke just inside. The deck
likely contained a rich atmosphere of heated flammable
vapors, which rapidly combusted when fresh air was
introduced via the opening of the ventilation trunks for
decks 9 and 10/11. This reaction is analogous to an Figure 101. Left: Battery in towed vehicle removed from deck 7. The disconnected bare cable lug was in physical contact
overpressurization event. Following the explosion, the with unprotected battery terminal post. Right: Battery in forklift vehicle removed from deck 5. The disconnected battery
firefighters, assisted by the Coast Guard, transitioned to a cable lugs were located near terminal posts, and the battery terminal posts were unprotected.
defensive strategy, cooling external exposed surfaces. BACKGROUND SOURCE: COAST GUARD
Figure 102. Completed “Vehicle operator) “Cargo Safety Awareness” procedure, to

Lashing Inspection Procedure.” The prevent alarms from continuously activating due to SAFETY RECOMMENDATIONS
document states “all second hand exhaust from the vehicles being loaded onto the vessel.
[sic] vehicles must have the battery Following the completion of loading, the crew left the As a result of its investigation into this
terminals disconnected once in cargo decks and began preparing the vessel for sea. accident, the NTSB issued eight new safety
final stow” (red oval) and lists the Höegh’s procedure did not specify at what point after the recommendations to federal regulators and the
number of batteries on each cargo completion of loading the crew should reactivate the fire companies involved in the accident to reduce the
deck for which the disconnection detection system, so the system remained deactivated. risk of transporting used vehicles, such as those
was “incomplete” (red rectangle) for Without the fire detection system being activated and that were loaded on vessels like the Höegh Xiamen.
the accident loading. without crew on the cargo decks to detect any smoke or Used vehicles are often damaged and present an
fire, there was a gap between the completion of loading elevated risk of fire. Better inspection, oversight,
and the discovery of the fire, during which the vessel was and enforcement are needed to reduce this risk.
Figure 103. Excerpt from Grimaldi
unprotected. The NTSB therefore recommended that the Pipeline
battery disconnect procedure
and Hazardous Materials Safety Administration
showing a vehicle battery secured Ineffective emergency distress calls. A successful eliminate the exceptions in the HMR for used and
with a blue plastic negative terminal emergency response is contingent on early distress damaged vehicles transported by Ro/Ro vessels.
cap and disconnected cable notification and clear, effective communication. To report The NTSB also recommended the Coast Guard
(circled) tucked inside. SOURCE: the fire, the master first attempted to call the ship’s agent propose that the IMO eliminate similar exceptions
GRIMALDI by mobile phone but was unable to reach the agent. He from the IMDG Code.
next went to the bridge and called for help by VHF radio
after sounding the general alarm; however, the entity Additionally, it is imperative that operators of
Regulatory exceptions for used and damaged he attempted to reach, “Jacksonville Port Control,” did similar Ro/Ro vessels engaged in the transportation
flammable-liquid-powered vehicles. The used vehicles not exist. Although the vessel had a non-tank vessel of used vehicles act to ensure that any personnel
loaded on to the Höegh Xiamen were considered response plan, which listed the local Coast Guard’s involved in loading operations—including vessel
excepted from the requirements of the HMR because contact information, and the master should have been crews, stevedores, and longshoremen—be aware of
the vessel’s cargo space had been approved by the familiar with using VHF channel 16 for emergencies in the importance of disconnecting batteries on used
flag state as specially designated and approved for port, he did not use either option. vehicles. To that end, the NTSB recommended that
vehicles, and vehicles with leaking fluids were not the companies involved revise their procedures to
to be accepted. The IMDG Code contained similar improve oversight of vehicle loading and battery
provisions that would have excepted this shipment from securement.
international dangerous goods regulations. However, The probable cause of the fire aboard the vehicle
carrier Höegh Xiamen was Grimaldi’s and SSA The investigation showed that the detection
the circumstances of this accident and others suggest of the fire was delayed because the vessels’
that used vehicles, particularly those that are older with Atlantic’s ineffective oversight of longshoremen,
which did not identify that Grimaldi’s vehicle fire detection systems remained deactivated
unknown maintenance history and/or crash damage, after loading was completed. Additionally, the
require extra protections to mitigate the risk of vehicle battery securement procedures were not being
followed, resulting in an electrical fault from an Jacksonville Fire and Rescue Department’s
fires on board Ro/Ro vehicle carrier vessels. Better response to the accident was delayed because the
inspection, oversight, and enforcement would ensure improperly disconnected battery in a used vehicle
on cargo deck 8. Contributing to the delay in the master seemingly did not know how to report a
effective implementation of battery securement and fire to local authorities. The NTSB recommended
vehicle inspection policies in used vehicles across all detection of the fire was the crew not immediately
reactivating the vessel’s fire detection system that the vessel’s operator further revise their
vehicle carrier operations. procedures to minimize the amount of time
after the completion of loading.
Fire detection system deactivation during cargo vessels’ fire detection systems are deactivated
loading. During loading operations, the crew deactivated and ensure that contact information for emergency
the Höegh Xiamen’s fire detection system, as was response authorities is immediately available.
required by Höegh Technical Management’s (the
VESSEL GROUP
Fire/Explosion
SPECIALTY/OTHER
Fire aboard
Dive Support Vessel
Iron Maiden
Gulf Intracoastal Waterway, mile 36, Larose, Louisiana
April 16, 2020 MAB-21-11
ACCIDENT ID ISSUED
DCA20FM016 April 21, 2021
Figure 106. Fire damage to living quarters (left) and generator room (right). SOURCE: COAST GUARD
O
Figure 104. The Iron Maiden before the accident, with n April 16, 2020, about 0110 local time, a fire number one generator was directly under the starboard
previous name and different owner. SOURCE: COAST GUARD on board the dive support vessel Iron Maiden exhaust fan (which was secured) and had a fire cloth
occurred while the vessel was docked at the over it for protection from falling sparks that could
Allied Shipyard in Larose, Louisiana. Local firefighters come down the exhaust trunk. The shipyard foreman
extinguished the fire at 0225. No one was aboard the examined the generator room at approximately 0930 and
vessel at the time of the fire. No pollution or injuries were determined that the space was safe, since there was no
reported. Damage to the vessel was estimated at greater indication of fire or smoke.
than $900,000. By 1630, all work on the Iron Maiden was concluded,
On April 15, shipyard workers entered the vessel at and all shipyard workers departed the vessel. The vessel
0800 to conduct hot work with acetylene torches on the company representative left about 1735, leaving two
starboard exhaust trunk under the bridge deck and on vessel crewmembers, who finished eating dinner and
the fo'c'sle deck by the starboard-side mooring/securing left about 1800 to return to their hotel for the evening.
bitt. Both work areas were located directly above the The shore power to the Iron Maiden remained energized.
Figure 105. Starboard exhaust trunk fan (post-fire)
generator room on the main deck. The room had three There was no crewmember or shipyard worker staying on
located above the No. 1 generator. SOURCE: COAST GUARD
diesel-engine-driven generators, which were secured. The board the Iron Maiden during the night.
significant damage to the passageway outside the

generator room and living quarters (above the generator
room). They also noted smoke damage to the mess
area and the galley (forward of the generator room) and
the interior stern section of the pilothouse (two decks
above the generator room), as well as sections of burned
exterior paint around the starboard smoke trunk and
pilothouse. Within the generator room, the most severe
damage was observed on the forward bulkhead near the
access hatch from the passageway.
According to the fire investigator’s report, “The fire
started in the generator room on the wall area common
to the mess area.” They could not rule out the possibility
of an electrical short as the potential source of the fire.
Because the battery charger, alarm panel, and generator
push button start-stop panel were in the area of fire
ignition identified by fire investigators, one of these
components may have been the source of the fire as the
result of an electrical short. However, the exact location
of the source of the fire could not be identified by fire
investigators.
Figure 107. General layout of the Iron Maiden generator room and surrounding main deck spaces with area fire started as
identified by the LaFourche Parish Fire District.
The probable cause of the fire aboard the dive
support vessel Iron Maiden was an electrical
On April 16, at 0110, the Lafourche Parish Fire District smoke behind the fuel tank on the starboard side of short from an unidentified source located on the
dispatcher received a phone call from the Larose Bridge the generator room, but it was “dug up” by shipyard forward bulkhead within the generator room.
tender (located roughly 2,000 feet from the shipyard) personnel and quickly extinguished with water from a Contributing to the undetected propagation of the
reporting smoke and flames coming from a vessel at the garden hose. fire was the lack of continuous monitoring of the
shipyard. Firefighters discovered smoke coming from Fire investigators from the LaFourche Parish Fire District vessel while it was docked at the shipyard.
the starboard side of the Iron Maiden’s pilothouse. The noted extensive damage to the generator room and
fire extended from the main deck up to the pilothouse,
encompassing the generator room and the living
spaces on the fo'c'sle deck. As the fire grew, the wood
paneling and furniture in the space above the generator Continuous Monitoring of Inactive Vessels
room ignited and provided a path for the fire to expand Fire and flooding are risks for both crewed and unattended vessels. To protect personnel, property,
from the generator room up into the living quarters. and the environment, it is good marine practice for owners, operators, and shipyard managers
With the vessel’s fire detection system secured for the to coordinate and implement some form of continuous monitoring for vessels undergoing
shipyard period and no continuous or periodic scheduled maintenance in a shipyard, in lay-up, or in some other inactive period without regular crews aboard.
monitoring of the vessel by shipyard or owner personnel, Continuous monitoring can consist of scheduled security rounds and/or active monitoring with
the fire was able to spread undetected. sensing and alarm systems.
The responding firefighters boarded the vessel and
extinguished the fire with water hoses. About 0900,
shipyard personnel found an area still emitting
VESSEL GROUP
Fire/Explosion
FISHING
Engine Room Fire

aboard Fishing Vessel
Lucky Angel
Gulf of Mexico, 20 miles south-southwest of
Pascagoula, Mississippi
Figure 108. Lucky Angel on fire, December 11. SOURCE: COAST GUARD
ACCIDENT ID ISSUED
O
DCA21FM010 December 1, 2021 n December 10, 2020, about 2205 local time, he saw white smoke that “smelled pretty much like
the fishing vessel Lucky Angel was trawling [something] electrical was shorting.” He saw sparks
for shrimp in the Gulf of Mexico, 20 miles from coming from a bundle of wires located overhead and
Pascagoula, Mississippi, when a fire broke out in slightly to the right of the inside platform. Investigators
the vessel’s engine room. The three crewmembers later determined that the group of wires contained 120-
attempted to fight the fire but were forced to abandon volt AC wires to the deck flood lights and the aft bilge
the vessel. They were rescued by the Coast Guard, and pump. It is likely the fire’s source of ignition was the
the vessel sank 2 days later. No pollution was reported, electrical sparking.
but there was one minor injury. The vessel was a total Without maintenance records or a pre- or post-purchase
constructive loss with an estimated value of $120,000. survey, investigators were unable to determine the
About 0600, after 5 days of shrimping, the Lucky Angel condition of the wiring bundle. If the wiring was original,
docked in Bayou La Batre so the captain could attend a dating back to 1968, it may have deteriorated due to
doctor’s appointment. The captain returned to the boat decades of being subjected to the atmosphere and
and sailed about 1455 with two deckhands on board. chemicals found in a hot engine room environment.
The Lucky Angel entered the open waters of the Gulf of Chafing from the material used to support the wiring
Mexico at Horn Island Pass at 1820. could have also caused the wires’ insulation to fail.
Approximately 2000, the captain and the deck hands In either case, a failure in the floodlights and aft bilge
began shrimping operations, and about 2100, the vessel pump wiring insulation likely caused arcing, which was
began to trawl for shrimp. The captain looked into the the ignition source to the ensuing fire. The arcing would
engine room twice (2130 and 2200), and, believing that have ignited some form of nearby combustible material
everything was fine, continued to trawl at 2.7 knots. to initiate and sustain combustion. The fire then likely
spread from the engine room to other combustibles in
About 2205, a smoke alarm for the engine room the house of the Lucky Angel.
indicated on the alarm panel in the wheelhouse. The
captain immediately went to the open engine room door The captain emptied three dry chemical fire extinguishers
in the after part of the house. From the inside platform, into the engine room from the platform, but he did not
close the two access doors or two exhaust fan vents

to the engine room. He tried to turn on the switch for
the deck wash pump but received an electric shock
that caused him to fall backwards—investigators later
confirmed that the deck wash pump wiring was run
separately from the sparking bundle of wires.
The captain returned to the wheelhouse and, in an
attempt to limit the spread of the fire, opened all the
electrical breakers to the wheelhouse, which rendered
all his bridge equipment inoperable. He and the two
deckhands attempted to extinguish the fire by tossing
sea water from 5-gallon buckets into the engine room.
They did this for about 20 minutes until they became
exhausted.
Figure 109. Lucky Angel pierside at Bayou La Batre, Figure 110. Area of accident where the Lucky Angel fire
Next, the captain and crew closed all doors and hatches Alabama, before the accident. started, as indicated by the red X.
to the house and engine room, except the two exhaust SOURCE: JULIAN PRICE BACKGROUND SOURCE: GOOGLE MAPS
fan vents, and went to the bow of the boat because
smoke had now filled the house. The captain used his
cell phone to call two other nearby shrimp boats that he
knew, but he got no answer. He then called his wife on
his cell phone, and she contacted the nearby boats.
He then called 911, who routed his call to the Coast
Guard District 8 command center about 2231, before
his phone went dead. About the same time, the boat’s
main engine and generator shut down.
The crew inflated and launched the liferaft and
abandoned the vessel; the vessel’s EPIRB and
lifejackets were inaccessible due to the smoke and
were left on board. The Coast Guard rescued the crew
at 2327. The Lucky Angel continued to burn through
the next day, and eventually, damage from the fire
likely caused a failure in hoses or piping connected
to a through-hull fitting for a sea water system that
allowed water to enter the hull and sink the vessel on
December 12. The vessel was not salvaged. Figure 111. Engine room layout and area where captain observed sparking.
The probable cause of the engine room fire

aboard the fishing vessel Lucky Angel was the
deterioration or chafing of wiring insulation, which
caused arcing that ignited nearby combustible
materials.
VESSEL GROUP
Fire/Explosion
FISHING
Engine Room Fire

aboard Fishing Vessel
Master Dylan
Gulf of Mexico, about 32 miles west-southwest of
Port Fourchon, Louisiana
ACCIDENT ID ISSUED
DCA21FM009 September 23, 2021 Figure 112. Photos taken from the Coast Guard small boat showing the Master Dylan aground after being towed by the
Master Dustin II (left) and the burning Master Dylan cloaked in smoke (right). SOURCE: COAST GUARD
A
bout 0745 on December 1, 2020, the On the morning of the accident, the crew lowered
fishing vessel Master Dylan was trawling for the shrimping nets into the water around 0730, and
shrimp in the Gulf of Mexico when an explosion the captain conducted a routine inspection of the
occurred in the engine room. Attempts to fight the engine room where the main engine and the starboard
subsequent fire from on board the vessel were generator were operating. He found nothing unusual
unsuccessful, so the crew abandoned ship to a Good and returned to the wheelhouse at approximately 0740.
Samaritan vessel. The fire was eventually extinguished About 5 minutes later, the crew heard a “loud explosion”
by other responding vessels, and the Master Dylan was in the engine room, after which they saw fire and black
taken under tow. However, during the tow, the stricken smoke. The captain attempted to extinguish the fire
vessel ran aground, the fire re-flashed, and the vessel but was unsuccessful. The captain determined that the
later sank. The vessel was a total constructive loss with fire could not be extinguished, so he directed the crew
an estimated value of at $300,000. to prepare to abandon the vessel and to raise the nets
The vessel had a main diesel engine and the two out of the water. The crew engaged the winch, which
diesel generators that were “rebuilt” during a scheduled operated off the main diesel engine, and were able to
maintenance period 5 months before the accident maneuver the nets as directed.
voyage. (The extent of the overhaul and the condition of A nearby fishing vessel rescued the crew from
any replacement parts could not be confirmed through the burning vessel, and an offshore supply vessel
records.) The crew told investigators that there were eventually extinguished the fire. The Master Dustin II,
no operational problems with the main diesel engine or a vessel owned by the same company, proceeded to
the two diesel generators either on the previous voyage tow the Master Dylan to the nearest point of land. The
following the maintenance or during the accident Master Dylan ran aground during the tow and the fire
voyage. re-flashed.
It is likely that the vessel’s engine room hoses,

connected to hull fittings to main engine cooling water
systems, generator cooling water systems, and salt
water service systems, failed due to the long-term
exposure to the heat of the fire, which most likely
resulted in the sinking of the vessel. As the hoses failed,
water would have entered the hull, causing the vessel to
lose stability, roll, and sink.
Because the vessel was not salvaged, the exact cause
of the fire in the engine could not be determined.
Investigators could not determine if electrical power
was lost, so they could not confirm if the fire source
was a generator malfunction. However, the crew was Figure 113. The Master Dylan before the accident.
able to operate the winch, indicating that the main SOURCE: MARINETRAFFIC.COM
engine was still operating and therefore could not have
been the source of the explosion.
The probable cause of the engine room fire on
A mechanical failure could have catastrophically
board the Master Dylan was the catastrophic
damaged the operating starboard generator’s engine,
failure of a diesel generator. Contributing to the
producing the reported explosion. Since the vessel’s
spread of the fire was the location of the fuel
fuel oil supply valves for the main diesel engine and the
shutoff valves within the engine room, which
generators were in the engine room, the fire and smoke
prevented the crew from securing them.
prevented the crew from securing the fuel supply from
tanks to the diesel engines to stop fuel from feeding the
fire. The wooden frames and furniture within the house,
as well as the dry supplies located inside the forepeak,
likewise would have provided additional fuel to sustain
the fire as it spread beyond the engine room.
Accessing Remote Engine Room Shutdowns

Following the initiation of an engine room fire, it is imperative to remove the source of available
fuel to the fire found in the fuel oil and lube oil systems. In this accident, the vessel had no remote
emergency cut-off valves outside the engine room, and thus fuel to the fire could not be stopped
and the vessel was eventually consumed by the flames. Vessel designers, builders, owners, and
operators are encouraged to install, regularly test, and have emergency drills that incorporate
remote cut-off valves for fuel and lube oil lines.
VESSEL GROUP
Fire/Explosion
OFFSHORE
Diesel Generator
Engine Failure aboard
Offshore Supply Vessel
Ocean Intervention
Anchorage B, Mamala Bay, Honolulu, Hawaii
ACCIDENT ID ISSUED
DCA21FM012 December 2, 2021
Figure 115. The damaged section of the engine (left), as indicated by the yellow square, from where the connecting rod
from the no. 3 DG (right) was ejected. SOURCE: OCEANEERING INTERNATIONAL, INC.; COAST GUARD
A
t 1303 local time on December 19, 2020, the Two hours later, the no. 1 DG was taken offline, and the
no. 3 diesel generator engine aboard the Ocean vessel’s electrical load was shifted onto the no. 3 DG—
Intervention sustained a mechanical failure while leaving the no. 3 DG carrying the vessel’s electrical load.
the offshore supply vessel was anchored off Honolulu, At 1303, the chief engineer and engineer on watch
Hawaii. The failure led to the ejection of components heard “an abnormal sound, similar to something heavy
from the engine and resulted in a fire in the engine dropping on the deck,” as the no. 3 DG underwent a
room. The crew isolated the fire before it could spread catastrophic mechanical failure, resulting in cylinder
throughout the vessel. No pollution or injury to the no. 1’s connecting rod being ejected through the engine
16 crewmembers on board was reported. Damage to the crankcase while running at rated speed. The ejection of
Ocean Intervention totaled $3,046,624. the connecting rod allowed atomized oil to be released
Figure 114. Ocean Intervention under way before the About noon on December 18, the Ocean Intervention and ignite, starting a fire in the engine room. Thick, black
accident. SOURCE: OCEANEERING INTERNATIONAL, INC. was docked in Honolulu Harbor awaiting orders with a smoke filled the engine room.
partial crew standing deck and engineering watches. Engineering watchstanders did not receive any alarms
While at anchorage, the crew troubleshot speed variation indicating issues with the operational parameters of the
issues related to the nos. 1 and 3 diesel generators (DGs) no. 3 DG in the minutes preceding the failure. The crew
throughout the afternoon and the following morning, quickly stopped the running engines, isolated all fuel
which involved replacement and calibration of several supplies, shut down engine room ventilation systems,
electrical components and multiple engine restarts. and closed the space’s air dampers to effectively starve
About 1050 on December 19, the no. 3 DG was put the fire of fuel and oxygen, which prevented the spread of
online, sharing the electrical load with the no. 1 DG.
the fire. Additionally, they used the emergency fire pump

to provide cooling water to the exterior surrounding
bulkheads and decks above to reduce the heat in the
engine room. The crew’s quick and effective actions
resulted in the extinguishment of the fire without putting
crewmembers at risk by having to enter the space.
During the postaccident forensic teardown of the no. 3
DG, factory-trained technicians were able to identify the
most likely sequence of events that led to the failure of
the engine but were unable to determine the root cause
due to several unknown preconditions of the engine.
The possibility of fluid, such as cooling water or fuel oil
entering the cylinder, causing a loss of clearance on the
connecting rod bearing and starting the failure sequence
was considered as a viable scenario; however, this theory
could not be verified due to damaged components and
operational alarms not activating before the failure.
The condition of the connecting rod bearings, showing
signs of cavitation erosion (some considered excessive)
was another possible root cause of the failure. If
Figure 117. The yellow bracket identifies the damaged area of the main crankshaft of the no. 3 DG (left), where the no. 1
the cavitation erosion became excessive enough, as
connecting rod bearing failed (right). SOURCE: OCEANEERING INTERNATIONAL, INC.
found on cylinder no. 13’s connecting rod bearing by
technicians, it could have caused the bearings to fail due
to increased tolerances between the components and
excessive movement outside of these tolerances. The probable cause of the diesel generator
engine failure aboard the offshore supply vessel
Ocean Intervention was a cylinder’s connecting
rod bearing adhering to the crankshaft, which
led to the ejection of the connecting rod and
catastrophic damage to the engine.
Containing Engine Room Fires

Engine rooms contain multiple fuel sources as well as mechanical ventilation, making the spaces
especially vulnerable to rapidly spreading fires. The crew of the Ocean Intervention effectively
contained the spread of a fire by removing fuel and oxygen sources. Vessel crews should
familiarize themselves and train frequently on machinery, fuel oil, lube oil, and ventilation shutoff
systems to quickly act to contain and suppress engine room fires before they can spread to other
spaces and/or cause a loss of propulsion and electrical power
Figure 116. Cavitation erosion in upper connecting rod
bearings from the no. 3 DG in cylinder nos. 11, 12, 13, and
14, encircled. SOURCE: OCEANEERING INTERNATIONAL, INC.
VESSEL GROUP
Fire/Explosion
TOWING/BARGE
Engine Room Fire on

board Towing Vessel
Susan Lynn
Barataria Waterway, Lafitte, Louisiana
October 8, 2019 MAB-21-03
ACCIDENT ID ISSUED
DCA20FM001 January 26, 2021
Figure 119. The Susan Lynn at its berth following the fire.
O
n October 8, 2019, about 0600 local time, the The engineer told investigators that at 0600 on October 8,
Susan Lynn was docked and in layup status at he woke to a “beeping fire alarm.” He observed there was
Tom’s Marine & Salvage yard on the Barataria no power and decided to check the engine room. He
Waterway in Lafitte, Louisiana, when a fire started peered through the open interior forward centerline door
in the engine room. The vessel’s watchman could to the upper engine room, observed smoke, and left to
not contain the fire and evacuated the vessel. Local grab carbon dioxide and dry chemical fire extinguishers.
firefighters extinguished the fire. No pollution or injuries He returned to the engine room and discharged the
were reported. Damage to the vessel was estimated at two extinguishers in the direction of the two generators
Figure 118. Susan Lynn (original name Rock Bluff), under
$1,350,000. located forward of the main engines. He said there was
previous ownership. SOURCE: LESLIE JENKINS, JANTRAN INC.
On October 4, 2019, an engineer arrived at the shipyard no way to secure the exhaust trunk ventilation to the
to reside on board the Susan Lynn, relieving the previous engine room, he did not secure any fuel shutoff valves,
watchman. The vessel was not hooked up to shore and he did not attempt to use the semi-portable fire
power. He told investigators that the port generator was extinguisher located near the aft starboard-side engine
operating to power the hotel loads during his stay on room door on the main deck. He left the vessel and
board and had been the only generator used since mid- called 911 and the vessel operator. The Lafitte Barataria
August. Crown Point Volunteer Fire Department arrived on scene
at 0628 and fought the fire with water hoses and foam.
The fire was declared out at 1315.
The heaviest fire damage was in the engine room, near Because of the limited evidence, the exact cause of
the generators, with additional damage on the main and the fire and the generator’s engine failure cannot be
second decks forward of the galley.The fire's spread determined. However, the rupture of the oil reservoir may
beyond the engine room was likely the result of the have been caused when part(s) of the failing connecting
open position of the interior forward and aft centerline rod and/or a piston struck the inside wall of the oil
upper engine room doors, which had been left open reservoir. A rod and/or piston striking and then rupturing
while the vessel was in layup and remained open the oil pan would have released oil into the engine room
throughout the fire. While the vessel was in layup, the under heat and pressure. This ejected lube oil mist may
engine room doors should have been closed as a fire have ignited off a hot surface, potentially the generator’s
Figure 122. Crew quarters on board the Susan Lynn.
safety measure. unlagged (not insulated) exhaust components. The
SOURCE: LOUISIANA OFFICE OF STATE FIRE MARSHAL
intensity and duration of the fire was likely exacerbated
Examinations of the vessel by Coast Guard by the melting of the bowl on the bottom of the
investigators and inspectors, a Louisiana State fire generator’s Racor fuel filter combined with the vessel’s
marshal investigator, and the Susan Lynn’s operator fuel shutoff valves not being closed, allowing additional
provided limited evidence to identify the cause of fuel for the fire.
the engine failure and subsequent fire. They noted
significant damage to the port generator oil reservoir, The probable cause of the fire on board the
including a hole in it, and that the generator engine Susan Lynn was a catastrophic engine failure
exhaust piping had no lagging installed. The fire resulting in an oil reservoir breach and an ensuing
marshal investigator could not identify the cause of fire initiated by ejected lube oil igniting off a hot
the fire, saying he could not “rule out mechanical/ surface. Contributing to the extent of the fire were
electrical failure” nor identify the heat source that the open engine room doors and the unsecured
ignited “combustible materials and ignitable liquid fuel.” fuel shutoff valves.
The vessel operator told investigators that he noticed
the connecting rod had separated from piston number 3
and was hanging down through the hole of the ruptured
Figure 120. The Susan Lynn port generator and lower oil reservoir and still connected to the crankshaft.
engine room following the fire. BACKGROUND SOURCE: Detailed evidence was not available because a forensic
LOUISIANA OFFICE OF STATE FIRE MARSHAL
examination of the port generator engine was not
Figure 121. Ruptured Susan Lynn port generator oil
conducted, nor was the engine rebuilt.
reservoir. BACKGROUND SOURCE: COAST GUARD
Figure 123. Simplified inboard profile of the Susan Lynn.

VESSEL GROUP
Fire/Explosion
SPECIALTY/OTHER
Hazardous Liquid
Pipeline Strike and
Subsequent Explosion
and Fire aboard
Dredging Vessel Figure 125. Post-fire photo of the Waymon Boyd, before sinking. SOURCE: COAST GUARD
Waymon Boyd
O
n August 21, 2020, about 0802 central daylight In 2019, EPIC Crude Terminal Company LP (EPIC) began
time, the US-flagged dredge Waymon Boyd struck planning for a second crude oil loading pier, designated
a submerged 16-inch liquid propane pipeline the East Dock, at the former Interstate Grain Terminal,
EPIC Marine Terminal, Corpus Christi Ship Channel, during dredging operations in Corpus Christi, Texas. A which required the construction of a bulkhead along the
Corpus Christi, Texas geyser of propane gas and water erupted adjacent to shoreline and the dredging of a ship berth between the
the vessel. Shortly thereafter, propane gas engulfed the dock and the main shipping channel. EPIC selected the
vessel, and an explosion occurred. Fire damaged the Orion Marine Group (Orion) for the project. Dredging for
August 21, 2020 MAR-21-05
vessel and surrounding shoreline. A total of 18 personnel the East Dock was conducted in two phases. Phase 1
ACCIDENT ID ISSUED employed by Orion Marine Group were working or occurred from May to June 2019, and phase 2 was
DCA20FM026 December 7, 2021 resting on the dredge and assist boats (tender boats, planned for July to October 2020.
anchor barges, booster barges, and a supply barge) on EPIC commissioned a survey to identify all utilities
the day of the accident. Three crewmembers aboard running through the terminal property. The surveyors
the Waymon Boyd and one on an adjacent anchor barge located one abandoned and two active pipelines that
died in the explosion and fire. Six crewmembers aboard ran parallel to the shoreline along the entire length
the dredge were injured, one of whom later died from of the terminal area. The lines were buried onshore
his injuries. The Waymon Boyd, valued at $9.48 million, but partially exposed in the water, lying in the bottom
was a total loss. The cost of pipeline damage was $2.09 sediment of the waterway. The active pipelines were
million. The cost of physical damage to the EPIC facility owned and operated by subsidiaries of Enterprise
was $120,000. Products Partners LP (Enterprise). The most northerly
Figure 126. Dredge Waymon Boyd before the accident. of the pipelines, a 16-inch-diameter pipe designated
SOURCE: ORION MARINE GROUP TX219, carried non-odorized liquefied propane.
In October 2019, EPIC was granted a permit to extend
the East Dock berth another 167 feet to the west.
Consequently, Orion assigned Schneider Engineering
and Consulting (Schneider), a wholly owned subsidiary
of Orion Marine Group’s parent company, to update
Figure 124. Waymon Boyd preaccident. the phase 1 dredging plans to reflect the revised berth
SOURCE: ORION MARINE GROUP
dimensions.
The revised engineering plans for the East Dock berth

were passed to Orion’s survey superintendent, who used
the plans to build a dredge template in DREDGEPACK,
a module in the hydrographic data collection and
processing software HYPACK. DREDGEPACK was used
by the dredge operator (leverman) to display where the
digging tool that he was controlling (the cutterhead)
was in relation to the dredge template. Using the
software display, the leverman could determine in real
time what areas required dredging and whether the
cutterhead was operating within the dredge template.
The survey superintendent loaded the dredge template
onto the computer on board the Waymon Boyd, the
dredge scheduled to conduct the phase 2 work.
On June 23, 2020, an Orion project engineer made a
one-call notification—a notice of intent to excavate—
for the phase 2 dredging operations. An Enterprise
technician contacted the project engineer to discuss
the project and schedule a site visit. On June 29, the
Orion project engineer sent the Schneider dredging
plans to the Enterprise technician noting in the
accompanying email that the areas where the pipelines
were furthest in the water had already been dredged,
“so there shouldn’t be a need for concern.”
A site visit with the Orion and Enterprise representatives
was set for June 30; however, due to issues stemming
from the COVID-19 pandemic, they did not meet as
planned. The pipeline technician and the project
engineer had a phone call during which they agreed that Figure 127. NTSB depiction showing comprehensive view of dredge template and pipeline locations.
it was not necessary to physically mark the pipelines Figure 128. EPIC dock project area, 1968 and 2016, showing the extent of land loss. The blue line indicates the TX219
because they did not conflict with the dredging area. location, and the red circle is the approximate accident location. SOURCES: US DEPARTMENT OF AGRICULTURE [LEFT];
As a result of their assessment of plans and US GEOLOGICAL SURVEY [RIGHT]; PHOTOS GEOREFERENCED AND OVERLAID ON A MAXAR TECHNOLOGIES IMAGE .
information provided by the project engineer, Enterprise

technicians closed the one-call tickets, believing
pipeline TX219 would be clear of the project and
there would be no dredging near the shoreline. Thus,
technicians concluded that no marking or other
protective measures would be required because
the dredging boundary exceeded the Enterprise
damage prevention program 50-foot distance limit for
mandatory marking.
Two weeks after the one-call tickets were closed,

Fire/Explosion
Orion followed up with an informal request to have the

pipelines marked with cane poles for dredge anchor
avoidance. Enterprise technicians, who had still been
led to believe dredging activities would be outside of
the mandatory marking zone, used an optional marking
technique known as “courtesy marking” in which the
pipeline was delineated with widely-spaced cane poles.
While a portion of on-land pipeline was marked with
color-coded flags and paint, the cane poles were not
color-coded or flagged. On July 16, the Orion project
engineer and the two Enterprise technicians met at the
dredging site, boarded a skiff, and courtesy-marked the Figure 130. Waymon Boyd general arrangements.
location using cane poles provided by Orion.
The Orion project engineer’s supervisor, the project
manager, also reviewed the Schneider dredging plans.
Although the Orion project manager did not expect
that the excavation would be near the pipelines, during
a subsequent discussion with the project engineer,
he suggested that the anchors could be placed near
them. He also discussed this concern with the dredge
superintendent and directed the project engineer to
inform Enterprise about the anchors.
Figure 129. Typical dredging operations.
Figure 131. The Waymon Boyd at the EPIC Marine Terminal East Dock site, August 7, 2020. SOURCE: ORION MARINE GROUP
On July 29, 2020, the Waymon Boyd was towed into Sixty-six seconds after the water eruption began,
position at the East Dock site and began phase 2 an explosion occurred as the propane gas—which
dredging. Through the end of July and the first weeks had been drawn into the dredge’s engine room by
of August, the Waymon Boyd operated at the dredge ventilation fans—ignited.
site, working generally east to west beginning on the Although the leverman attempted to swing the dredge
channel side and moving progressively inshore. away from the geyser of water that was carried with the
In the early morning hours of August 21, the dredge escaping propane, the vessel was less than 200 feet
was working in an area between two existing mooring away from the pipeline breach, and the expanding gas
dolphins located on the western side of the project area. cloud enveloped it. Enterprise later estimated that 6,024
About 0800, the leverman finished a series of side-to-side barrels of propane were released from the pipeline.
swings of the dredge and cutterhead and then operated Within seconds after the breach of pipeline TX219,
the controls to advance the dredge forward about 3 feet. the pipeline controller at the Enterprise control center
When the cutterhead was about 5 feet from the southern identified the pressure drop at the Viola Meter Station.
edge of the dredge template during his swing to port, The pipeline was shut down within 3 minutes, and
the cutterhead struck pipeline TX219, causing a breach technicians responded to valve control facilities within
in the line that allowed propane to escape, and water 13–29 minutes. After the Coast Guard was notified
began shooting up off the surface of the waterway, about of the explosion at 0810, Coast Guard surface and air
2–3 feet landward of the cutterhead. units joined the response effort. Tugboats remained at
the accident site until the fire on the Waymon Boyd was
extinguished about 1300. Residual propane rising from
the breached pipe continued to burn until 1610, when
the pressure in pipeline TX219 equalized with the water
pressure, the release of propane diminished, and the
fire self-extinguished. The dredge, which continued to
smolder, began to founder at 1400. Efforts to stabilize
the vessel were unsuccessful, and it sank at 2151.
Figure 135. Tugboats Ted C Litton and Evelena fight the
fire at the accident scene. SOURCE: COAST GUARD
Figure 132. Above: Waymon

Boyd lever room postaccident.
BACKGROUND SOURCE: ORION MARINE Figure 134. Screen captures from security camera
GROUP footage showing (top to bottom) the water and gas
eruption before the explosion (as noted by red arrow),
Figure 133. Left: Front view the initial explosion, and the ensuing fireball. The
of cutterhead assembly after Waymon Boyd is outside of the frame of the camera.
underwater recovery (looking aft). SOURCE: EPIC CRUDE TERMINAL COMPANY
SAFETY ISSUES
Fire/Explosion
Inadequate project planning and risk assessment.

Orion and its design engineers did not take measures to
address the risk of dredging near the pipelines before
they started dredging. These measures could have
included consulting with Enterprise representative,
conducting a formal risk assessment, or implementing
effective engineering controls. Had Enterprise been
invited to participate in preconstruction and kickoff
meetings, they may have been more aware that the
dredging area was unacceptably close to their pipelines
and could have suggested safer alternatives. Because
Orion did not complete a formal risk assessment for the
EPIC dock project, the hazard presented by conducting Figure 137. Postaccident temporary and permanent pipeline marker survey locations, accident location, and pipeline
dredging operations near pipeline TX219 was never TX219 (left); preaccident photo of marking cane pole (right).
formally identified or documented, and the risk was not BACKGROUND SOURCES: ENTERPRISE PRODUCTS [LEFT]; EPIC CRUDE TERMINAL COMPANY [RIGHT]
completely understood. Had a formal risk assessment
been completed, controls could have been put in place to Ineffective pipeline damage prevention. In information, overestimating the distance between the
mitigate the risk posed by dredging near pipeline TX219. this accident, the tools used for pipeline damage dredging area and pipeline TX219, and communicating
Figure 136. Waymon Boyd AIS data from 0757:25 to prevention—the one-call process, pipeline marking, incomplete and inaccurate information during the
0802:45 on the accident date. BACKGROUND SOURCE: dredging area marking, and tolerance zones—were one-call process, which dissuaded Enterprise from
GOOGLE EARTH either inadequate or ineffective. Marine dredging protecting pipeline TX219 in accordance with the
projects require a greater company’s damage prevention program.
level of collaboration and Proper line locating and marking by the pipeline
review between pipeline operator following a one-call notification are necessary
operators and dredging to ensure that an excavation will be sufficiently clear
companies than the of buried pipelines. Although Enterprise courtesy-
one-call process provides marked the pipelines with cane poles, the markers
because of the challenges did not meet pipeline excavation damage protection
associated with marking standards, nor were they required to, based on the
marine pipelines and the incorrect information provided by Orion, and therefore
lack of precision associated were insufficient to visually warn the leverman of the
with dredging operations. danger of the pipeline. Further, a technique known as
Additionally, the Schneider white-lining (the placement of white paint or flags to
engineering plans provided delineate the boundary of proposed excavation areas)
to Orion for the project did could have been used on this project as an added
not clearly depict the extent measure of communicating the precise location of
of the dredging area or proposed excavation/dredging activity to pipeline
the pipeline location in all technicians in advance of their one-call review. Project
drawings. This resulted in boundary marking requirements for dredging projects
the Orion project engineer (equivalent to land-based white-lining requirements)
misinterpreting the would provide utility operators with additional visual
information about the location of dredging projects to

confirm any encroachment of the proposed project on SAFETY RECOMMENDATIONS
pipelines.
As a result of its investigation into this accident, the NTSB issued ten new safety recommendations to a
Finally, the clearance required by existing state- federal regulator, industry organizations, and the companies involved in the accident.
regulated tolerance zones is not adequate for large-
scale dredging activities because of the inherent The NTSB believes that pipeline operators and dredging companies would benefit from the federal regulator
inaccuracies associated with operating a cutterhead and industry organizations establishing guidance for obtaining and using accurate pipeline location data,
dredge. Dredging safety would be improved if guidelines and for clearly identifying and marking dredging boundaries, during project planning. The NTSB therefore
identified consistent dredging tolerance zones, within recommended that PHMSA collaborate with Coastal and Marine Operators and the Council for Dredging and
which special provisions and procedures are enacted Marine Construction Safety to develop guidance for the industry to follow.
for pipeline avoidance. Believing that the existence of standard minimum tolerance or safety zones for dredging would reduce confusion
Lack of pipeline hazard training. Although 8 months during dredging planning and operations, the NTSB recommended that PHMSA include criteria for minimum
before the accident the Council for Dredging and Marine tolerance or safety zones for dredging in state pipeline safety program evaluation guidelines, and that Enterprise
Construction Safety had published recommended revise its damage prevention program guidelines to include a larger tolerance zone for dredging operations.
actions for a dredge crew in the event of an emergency The investigation revealed that Orion Marine Group did not adequately assess the risk for dredging near
involving a pipeline breach, Orion did not have an underwater pipelines during the planning process, and Schneider Engineering and Consulting, which produced
emergency procedure or crew training for a pipeline the engineering plans and drawings, did not consistently include the pipelines or any tolerance zones in
breach. The dredge crew lacked function-specific project plans. Therefore, the NTSB recommended that Orion Group Holdings incorporate risk assessments,
pipeline safety training and emergency procedures that written policies and procedures for planning dredging operations, and specifications and quality control
could have prepared them to react quicker and more measures related to pipelines and other hazards for engineering plans and drawings into its subsidiary
effectively to the gas pipeline strike. companies’ practices.
Finally, because the dredging industry would benefit from learning about the circumstances of this accident,
The probable cause of the hazardous liquid the NTSB recommended that the Coastal and Marine Operators modify existing pipeline safety training to
pipeline breach, propane release, and subsequent incorporate lessons learned from this accident.
explosion and fire aboard the dredging vessel
Waymon Boyd was Orion Marine Group’s
inadequate planning and risk management
processes, which failed to identify the proximity
of their dredging operation to Enterprise
Products’ pipeline TX219 and resulted in the
absence of effective controls to prevent the
dredge’s cutterhead from striking the pipeline.
Contributing to the accident were deficient
dredging plans provided by Schneider Engineering
and Consulting, which resulted in incomplete
and inaccurate information communicated to
Enterprise Products by Orion Marine Group during
the one-call process, which resulted in insufficient
measures to protect the pipeline from excavation
damage. Figure 138. Post-salvage images of the Waymon Boyd
lever room and captain’s office (above) and engine room
portside bulkhead, bowed outward (right).
VESSEL GROUP
Flooding/Hull Failure
TOWING/BARGE
Flooding of
Towing Vessel
Alton St. Amant
Harvey Canal, New Orleans, Louisiana
May 17, 2020 MAB-21-07
ACCIDENT ID ISSUED
DCA20FM019 March 11, 2021
Figure 139. Alton St. Amant under way before the Figure 140. Alton St. Amant, partially submerged, on the morning of May 18. SOURCE: COAST GUARD
accident. SOURCE: VESSEL FINDER
O
n May 17, 2020, about 0530 local time, a On May 9, after spending about 6 weeks at the Bollinger
shipyard worker reported that the towing vessel Quick Repair shipyard, the Alton St. Amant was shifted
Alton St. Amant was partially submerged while from drydock to a wet berth to complete outstanding
moored at a shipyard in the Harvey Canal in New maintenance items. Among the remaining work, two
Orleans, Louisiana. There were no crewmembers or bilge pumps, which had been removed from the vessel
shipyard workers aboard the vessel. Approximately for overhaul, were to be reinstalled; the sealing rings on
five gallons of diesel fuel were released into the water. several of the vessel’s tank access hatches were to be
Damage to the vessel was estimated at $1.5 million. No replaced; and the sealing surfaces of the hatches were
injuries were reported. to be cleaned.
On Friday, May 15, about 24,000 gallons of fuel were Throughout the remainder of the day and throughout
loaded onto the vessel. The flush hatches to the vessel’s the night, the two potable water tanks continued to
two potable water tanks located on the main deck in the fill with fresh water on the unmanned vessel. After
rudder room had been opened for maintenance, but the the pipefitter departed the vessel, no other persons
covers were not reinstalled at the end of the day. That came aboard to monitor the status of the tank levels,
same day, the port engineer requested that the shipyard and there was no shipyard policy for monitoring the
workers fill the two potable water tanks. filling process. Having been filled for several hours, the
A pipefitter returned to the shipyard the following potable water tanks reached capacity, resulting in an
morning, on Saturday, May 16, and began reinstalling overflow through the open hatches in the rudder room
the bilge pumps with three other shipyard workers (rather than the tank vents as planned). After the rudder
about 0500. About 1000, after completing the pump room flooded, the water spilled over the open doorsill
installation, he started filling the potable water tanks onto the main deck of the engine room and began
from a shoreside water manifold that was connected flooding down into that space.
to the vessel’s potable water fill pipe via a 2-inch hose About 0630 on Sunday, May 17, a shipyard worker
through the open exterior engine room doors. He walking past the Alton St. Amant noticed that the vessel
opened the supply (fill) valve at the shoreside manifold, was sitting low in the water and called the shipyard
and began filling the two tanks, which had a combined general manager.
capacity of 13,233 gallons. Unaware that the potable The general manager found the Alton St. Amant partially Figure 142. Open access hatches and their covers
water tank access hatches were open in the rudder submerged and resting on the bottom of the canal for the two potable water tanks in the rudder room.
room, he left the shipyard about 1030 with plans to alongside the pier. The engine room was flooded, and SOURCE: COAST GUARD
return the next day (although a pre-work safety meeting the main deck was partially submerged. The general
was conducted each day, the status of these hatches manager noticed the potable water hose connected The probable cause of the flooding of the
was not communicated to the pipefitter). He intended to the vessel was charged; he then closed the potable towing vessel Alton St. Amant was the absence
to fill the tanks and then allow them to overflow onto water supply valve on the pier manifold. Fresh water had of shipyard pre-inspection and monitoring
the exterior main deck through their vents to flush out been filling the potable water tanks for over 20 hours. procedures for water transfer, which resulted in
any residual debris inside before turning off the water
Pollution mitigation and recovery efforts began that potable water tanks overflowing through their
supply.
morning. By 1630, the Alton St. Amant was lifted by open access hatches during an unmonitored
crane from the bottom of the canal and refloated. transfer.
Figure 141. Simple profile of the Alton St. Amant (not
to scale), with the potable water tank highlighted. The following morning, shipyard workers who were
BACKGROUND SOURCE: BLESSEY MARINE SERVICES disconnecting electrical power cables from the
recovered vessel found the potable water hose still
connected to the fill pipe on the Alton St. Amant.
Precautions for Tank Filling

Crew and shipyard personnel designated to conduct liquid transfers must be aware of the status
of a vessel’s tanks, including their access hatches and associated piping systems, whether
ashore or at sea. When filling a tank, open access hatches create a risk of unintended flooding.
Pre-inspection and monitoring of transfers provide the opportunity to identify and remedy any
issues in order to ensure they are safely completed
VESSEL GROUP
Flooding/Hull Failure
FISHING
Engine Flooding
and Sinking of
Fishing Vessel
Rebecca Mary
Atlantic Ocean, about 40 miles south of Martha’s
Vineyard, Massachusetts
June 17, 2020 MAB-21-12
ACCIDENT ID ISSUED
DCA20FM021 May 13, 2021
Figure 144. Rebecca Mary before the accident. SOURCE: MARINE SAFETY CONSULTANTS
O
n June 17, 2020, in the early morning, the and contained some seawater that did not drain
Figure 143. Rebecca Mary before the accident.
commercial fishing vessel Rebecca Mary began overboard via the port and starboard freeing ports.
SOURCE: MARINE SAFETY CONSULTANTS
flooding in the aft portion of the vessel while About 0230, the engine room high-level bilge alarm
under way in the Atlantic Ocean about 40 miles south sounded on the bridge. The deckhand on watch exited
of Martha’s Vineyard, Massachusetts. The vessel the wheelhouse, lined up the bilge system, and pumped
capsized and subsequently sank. All four crewmembers out the water in the engine room. He stated that there
abandoned the vessel in their survival suits and was very little water, and everything appeared normal.
were rescued by a Coast Guard helicopter crew with
no injuries reported. The vessel had approximately About 0400, the fish hold high-level bilge sump alarm
3,000 gallons of diesel fuel aboard; after the vessel sank, sounded, which, according to the deckhand, was typical
an oil sheen was visible in the water. The Rebecca Mary’s during a watch, due to melting ice. Crewmembers
estimated value was $375,000. observed that the stern was sitting low enough that
seawater began washing over and covering the non-
On June 17, the Rebecca Mary was returning to New watertight, raised lazarette hatch, which was equipped
Bedford, Massachusetts, after fishing the previous with a cover that could not be latched closed. The
two days. During the return transit, with a typical load captain believed the cover for the lazarette was no longer
of illex and ice on board, the vessel’s freeboard was sitting on top of the hatch after the water level rose
lower than it typically would be with an empty fish hold. above it. It is likely that seawater displaced the lazarette
Crewmembers witnessed waves washing onto the aft cover, causing the lazarette to flood through the open
deck via the ramp into the hog pen area, which had its hatch. As seawater entered the lazarette, the vessel’s
stern boards removed. While some seawater flowed freeboard would have been reduced further.
back overboard via the ramp, the hog pen accumulated
Once the stern lowered to a certain point, water would A survey vessel attempted to locate the sunken vessel
have downflooded into the fish hold after displacing the using a multi-beam echosounder and side scan sonar
raised, rectangular, non-watertight hatch cover. systems, but the Rebecca Mary was not detected.
The crew deployed the liferaft by throwing its canister Because the Rebecca Mary was not salvaged, there was
overboard. After the liferaft inflated, the Rebecca Mary no postaccident vessel examination to determine the
rolled over to port; the vessel’s rigging punctured the initial flooding source.
liferaft, instantly deflating it. As the vessel capsized, all
four crewmembers jumped overboard in their survival The probable cause of the flooding and sinking of
suits. They locked arms and waited about 15 minutes the fishing vessel Rebecca Mary was undetected
until a Coast Guard helicopter arrived. Crewmembers flooding of the lazarette, likely through a non-
reported that the main engine continued to run watertight raised hatch.
throughout the flooding sequence and shut down when
the Rebecca Mary capsized.
Preparing for Abandonment

Early communication with the Coast Guard and preparing to abandon ship by donning survival suits
or personal flotation devices when experiencing significant flooding, fire, or other emergencies
increases the likelihood of survival. When deploying liferafts and other life-saving appliances,
crews should attempt to launch in areas clear of obstructions.
Figure 145. Inverted and mostly submerged
Rebecca Mary during rescue operations on the morning
Figure 146. Working deck of Rebecca Mary. BACKGROUND SOURCE: KEVIN RALPH
of the sinking. SOURCE: COAST GUARD
The deckhand woke the captain, who started a second

bilge pump. He stopped pumping the fish hold bilge
sump and started pumping from the two aftermost
compartments, the net locker and the lazarette. The
deckhand stated that he did not hear high-level bilge
alarms for the net locker or the lazarette when in the
wheelhouse; the captain stated that when he started
the bilge pumps for those spaces, he observed water
exiting through the two overboard discharge pipes,
indicating water was present.
The captain woke the other two crewmembers, and
they all donned survival suits. At 0409, he made a
distress call to the Coast Guard on VHF channel 16 and
activated the EPIRB.
About 0500, the seawater level approached the forward
part of the working deck, and the vessel began listing to
port. It is unclear whether there was progressive flooding
through the bulkhead from the lazarette to the net gear
locker, or the water was from another ingress source.
VESSEL GROUP
Grounding/Stranding
FISHING
Engine Stranding and

Subsequent Loss of
the Fishing Vessel
Miss Annie
Calibogue Sound, Hilton Head Island, South Carolina
ACCIDENT ID ISSUED
DCA20FM008 January 14, 2021
Figure 148. Miss Annie under way. SOURCE: COAST GUARD

Figure 147. Miss Annie shortly after all three crew were
rescued. SOURCE: COAST GUARD
O
n December 19, 2019, about 0700, the fishing during a storm in 2017), which was just over 800 yards
vessel Miss Annie was transiting out of Calibogue southwest of the reported yacht strike location. The
Sound, 2.3 miles north of Tybee Island, Georgia, warning was repeated in weekly LNTMs through the
when the vessel stranded on a submerged wreck. The end of 2019.
three crewmembers aboard remained with the vessel On November 12, 2019, NOAA updated its charts,
until they were rescued by the US Coast Guard, and the using the location that was published in the LNTM, to
vessel later broke apart. No pollution or injuries were indicate that the last known location for the wreck of
reported. The vessel was a total loss. The vessel value the Miss Debbie was “approximate.” On November 21,
was estimated at $60,000. 2019, NOAA conducted a bottom survey of the area
On November 1, 2019—over a month before the near the yacht strike, which showed a submerged
accident date—a yacht owner reported to the Coast wreck submerged less than 200 yards northwest from
Guard the location of a “significant object” that his the reported yacht strike location. About 3 weeks after
vessel struck as he was leaving Hilton Head Island. the survey—on the accident date—NOAA released
Five days later, the Coast Guard published a hazard corrections for their charts covering the area, marking
to navigation warning in the LNTM, reporting the the wreck found during the survey. Following NOAA’s
submerged wreck’s approximate position and advising update for the accident area, the Coast Guard published
caution. However, the LNTM reported the location the information in an LNTM.
of the charted and last known location for the wreck
of the Miss Debbie (a 40-foot shrimp boat that sank
The day of the accident, the 78-foot-long Miss Annie, Although the captain could not confirm if his GPS unit
a single-propeller wooden fishing vessel, departed had the most recent charts, the last software update
from Hilton Head Island at 0530. The captain was had been in April 2019 and would not have included the
following his usual route through the Calibouge Sound surveyed wreck. Similarly, although the initial hazard
entrance channel; two deckhands were also on board. warning and NOAA chart update noted the wreck in the
Although the crew had been aware of the Miss Debbie area, the location was based on the last known position
wreck, they had not sighted it in over a year and were of the Miss Debbie wreck—more than 800 yards
not looking for it. About 0700, as the captain steered southwest from the location of the Miss Annie strike.
southeast at a speed of about 10 knots, the vessel Even if the Miss Annie captain had read the LNTM,
came to a complete stop and listed to its starboard noted the warning, and had the latest GPS updates, he
side. The captain told investigators that it was “like would likely not have altered his route, thinking he was
I hit a rock.” At 0723, the captain sent a distress call clear of the hazard.
to Coast Guard Sector Charleston, and a response For chart correction tools to be useful, mariners must
boat arrived on scene at 0735. After the first crewman read them. Mariners must be alert to new hazards along
boarded the response boat at 0745, the Miss Annie their intended route and adopt a process to identify the
rolled, causing the remaining two crewmembers to slide hazards before getting underway. That process should Figure 149. Image of wreck from NOAA Danger to
into the water. By 0747, the crew of the response boat include viewing the NOAA weekly chart update and the Navigation Report. SOURCE: NOAA
had recovered the two persons from the water. A day LNTM before getting underway. A particular emphasis
later, the Miss Annie had broken apart. should be placed on identifying obstructions, such as a
Based on the locations of the Miss Annie strike and wreck, along the intended track of the vessel.
the wreck charted in the NOAA survey on November
21, 2019 (likely the Miss Debbie), the Miss Annie
The probable cause of the Miss Annie stranding
most likely struck the wreck identified and surveyed
and subsequent loss was the vessel striking an
about a month before the accident. It appears that the
unmarked wreck, whose position had shifted from
previously charted wreck of the Miss Debbie had moved
its previous known position and was yet to be
over the 2 years since its sinking and last charted
charted or announced in the notice to mariners.
position.
Identifying Navigation Hazards

Situation awareness demands a mariner should be alert for new hazards that can appear along their
intended route. NOAA and the Coast Guard track these hazards and publish chart corrections each
week. Mariners should adopt a process for identifying new hazards that are not marked on their Figure 150. Miss Annie top of wheelhouse and debris
paper or electronic chart system, before getting underway. washed ashore. SOURCE: COAST GUARD
NOAA provides weekly chart updates: https://distribution.charts.noaa.gov/weekly_updates/
he U.S. Coast Guard provides NOAA chart corrections each week in Section IV – Chart Corrections
T
in the Local Notice to Mariners: https://www.navcen.uscg.gov/?pageName=lnmMain
F or a list of all chart corrections for paper charts tracked by chart number and edition:
https://ocsdata.ncd.noaa.gov/ntm/Default.aspx
VESSEL GROUP
CARGO, GENERAL
Breakaway of
Containership
CMA CGM Bianca
Napoleon Avenue Container Terminal, Lower
Mississippi River, mile 100, New Orleans, Louisiana
ACCIDENT DATE REPORT NUMBER Figure 152. CMA CGM Bianca before the accident. SOURCE: HENRY KADOCH, HARBORSHOTS.COM
August 2, 2020 MAB-21-18
O
ACCIDENT ID ISSUED n August 2, 2020, about 1402 local time, the inspected by the crew during regular rounds to ensure
DCA20FM024 September 2, 2021 containership CMA CGM Bianca was loading adequate tension. At the completion of mooring, the
cargo while moored at the Napoleon Avenue containership’s main propulsion engine was shut down.
Container Terminal in New Orleans, Louisiana, when At 0712, after a safety meeting, container unloading
a sudden, localized thunderstorm passed through commenced, followed shortly thereafter by concurrent
the area. The vessel’s mooring lines parted in the loading operations, using gantry cranes nos. 5 and 6.
high winds, and the ship moved away from the pier. At 1300, a second shift of longshoremen arrived to
Containers being lifted by shoreside gantry cranes take over cargo operations. The crane no. 5 operator
struck the ship, and one damaged container dropped said that the weather at the time of the shift change
in the water, spilling a cargo of plastic pellets. A crane was “bright and sunny.” However, about an hour into
operator suffered a minor injury; no other injuries were the shift, “it got pretty cloudy.” The two crane operators
reported among ship and shore personnel. The total stated that work normally continued during rain, only
Figure 151. Simplified diagram of gantry crane no. 6. cost of damages was estimated at $15 million for the stopping during reduced visibility or high winds. So,
shoreside gantry cranes and $60,196 for the ship. although the crane no. 6 operator saw the approaching
At 0418 that morning, the 1,099-foot-long weather, he proceeded with operations.
CMA CGM Bianca had moored starboard side to at The NWS issued a special weather statement
the container terminal. Eight lines were rigged from about 40 minutes before the accident, reporting a
the bow, and eight lines were rigged from the stern. thunderstorm in the area moving east toward the
The mooring lines were certificated and in good or accident site. However, the statement made no mention
acceptable condition. The ship was equipped with of the threat of high winds, and a severe thunderstorm
self-tensioning-capable mooring winches; however, the warning was not issued.
auto-tensioning devices on the mooring line winches
were not engaged because, according to the master, At 1350, security cameras at the terminal recorded
the river current in the Mississippi River and wash rain beginning to fall. The rain increased steadily while
from passing vessels could trigger unwanted payout, visibility decreased, completely obscuring the camera
resulting in slack lines. Instead, the mooring winches view. Ten minutes later, “gale force winds and strong
were secured by their brakes, and the lines were rain” hit the containership. The master described the
conditions as “in the form of a tornado,” and both crane container and the forward spreader bars into the ship’s
operators stated that the winds developed “in seconds.” hold. The remainder of the spreader, still attached to the
A vessel located very close to the accident reported crane, then impaled and lodged in another container.
a wind gust at 73 mph, and both the no. 5 and no. 6 Within minutes of the lines parting, the crew had
cranes shut down automatically, indicating sustained dropped both anchors in the water, energized the
winds of at least 45 mph or gusts of at least 55 mph. bow thruster, started the main engine and transferred
The evidence suggests that the CMA CGM Bianca was control to the bridge. The crew then used the thruster
struck by outflow winds from a downburst. and engines, along with the anchors, to hold the
Under the force of the wind, at 1402, seven forward vessel’s position in the river. The quick actions of the
mooring lines and three aft mooring lines on the crew prevented the vessel from drifting down river,
CMA CGM Bianca parted, and the ship moved away where it could have caused damage to other vessels or
from the pier. Due to the wind’s sudden onset and shore infrastructure.
extreme velocity, the crane nos. 5 and 6 operators
had little time to act to move the cranes and attached
The probable cause of the breakaway of the
containers into a safe position. The cranes began
containership CMA CGM Bianca from the
moving forward to aft along their rails. The container
Napoleon Avenue Container Terminal wharf
suspended from crane no. 5 hit other containers stowed
and the ensuing equipment damage was
on the ship and then fell from the ship. The container
the sudden onset of unforecasted severe
hit the pier and broke open before falling in the water.
winds likely originating from the outflow of a
Part of the container’s cargo of plastic pellets was
thunderstorm-generated downburst.
discharged into the river and were “irretrievable.”
The container suspended from crane no. 6 struck a
hatch cover guide in the CMA CGM Bianca’s cargo Figure 153. Right: Security camera footage of Napoleon
bay, puncturing the container. The forward spreader Avenue Container Terminal pier during accident.
BACKGROUND SOURCE: PORTS AMERICA
bars then detached from the spreader, dropping the
Figure 154. Simplified diagram of CMA CGM Bianca's mooring line arrangement.
VESSEL GROUP
TOWING/BARGE
Container Damage
and Loss aboard
Deck Cargo Barge
Ho’omaka Hou,
Towed by Hoku Loa
Pacific Ocean, 6.9 miles north-northwest of Hilo,
Hawaii
June 22, 2020 MAB-21-09
Figure 156. The collapsed row of containers, from the starboard quarter of the Ho’omaka Hou. SOURCE: COAST GUARD
ACCIDENT ID ISSUED
DCA20FM022 April 6, 2021
O
n June 22, 2020, about 0230 local time, the deck dispatcher that the barge was ready for the tug. The
cargo barge Ho’omaka Hou was under tow by cargo consisted mostly of 20- and 40-foot-long dry cargo
Figure 155. Locking cone (left) and stacking cone (right) the towing vessel Hoku Loa off the northeast coast and refrigerated containers but also included ISO tank
similar to ones used on the Ho’omaka Hou. BACKGROUND of the big island of Hawaii en route to Hilo, when fifty containers, wheeled vehicles, flatracks, and palletized
SOURCE: COAST GUARD
40-foot containers stacked on the after deck of the barge cargo. There was no initial barge load plan with weights
toppled, causing 21 to fall into the ocean. There were of the containers because load planning was done “as
no injuries or pollution reported. Eight containers were the day goes on” during loading. Therefore, barge team
eventually recovered by salvors, and 13 remain missing. members were never given a copy of a stow plan to
Cargo loss was estimated at $1.5 million, and damage to assist them in stacking the containers.
the barge and containers was estimated at $131,000. An initial barge load plan showing stratified container
After its last voyage, the 340-foot-long-by-90-foot-wide weights would have been a useful tool to assist the
Ho’omaka Hou had been empty for a few days barge team machine operators in stacking containers on
before loading commenced for the accident voyage. the barge to reduce or eliminate reverse stratification—
The company port engineer performed a thorough meaning that heavier containers were loaded above
inspection of the barge prior to loading and found no lighter containers. Reverse stratification results in stacks
deficiencies that would compromise cargo. having a higher center of gravity than stacks created
On June 20, cargo was driven aboard by the machine by placing the heaviest containers on the deck, with
operators and secured by lashers. The lashings were progressively lighter containers above—referred to as
checked to confirm that all were secure and tight, and, normal stratification. Normal stratification is preferred,
about 1830, the barge superintendent informed the because it creates a stack having the lowest possible
center of gravity.
The reverse-stratified stack’s securing arrangements

would have been subject to increased forces while
moving in a seaway. The containers were secured
primarily with stacking cones, which provided little
protection against the containers leaning or tipping.
About 0200 on June 22, the barge turned about 30° to
a new south-southeasterly course, and it is likely that
the dynamic rolling from the seas on the vessel’s beam
resulted in forces on the container stacks with the
greatest reverse stratification so that, unchecked by the
lashings used solely on outboard stacks of containers
and the stacking cones used as the primary securing
point between containers, the containers tipped over and
caused the row to collapse.
At 0400 on June 22, during the approach to the dock at
Figure 157. Portion of stow plan showing container weights of the toppled container row. All weights are estimated
Hilo, the captain of the Hoku Loa was informed that the
gross weights in pounds. Red containers went overboard. Flatracks are indicated by yellow squares: tare weight is 11,023;
containers on the stern of the barge had toppled over.
gross weight cannot exceed 54,000 pounds.
This was the first time any of the Hoku Loa crew realized
the collapse had occurred. Once moored, shoreside
personnel found that the aftermost row of containers
stratification. In addition, the company did not provide stowed in a fore and aft direction had collapsed and that
the barge team procedures or calculations to determine 21 40-foot containers had fallen overboard. Later that
if the lashing arrangements were sufficient for the afternoon, a salvage company was hired to search for
reverse-stratified container stacks. and recover the lost containers.
About 2004, the 108-foot-long Hoku Loa arrived at
pier 39, where the Ho’omaka Hou was docked. The The probable cause of the collapse of container
master used the barge superintendent’s report to the stacks on board the barge Ho’omaka Hou towed
company dispatcher to determine that the tow was by the Hoku Loa was the company not providing
in compliance with the vessel’s stability letter and the barge team with an initial barge load plan,
applicable loadline regulations. The company did not as well as inadequate procedures for monitoring
Figure 158. The Ho’omaka Hou, loaded for a previous provide the master with the weights of the cargo to stack weights, which led to undetected reverse
voyage, as viewed from the stern. Note the loading of the afford him a means to determine if the lashings were stratification of container stacks that subjected
barge was not the configuration of the accident voyage. sufficient for the way the containers were stacked. The the stacks’ securing arrangements to increased
SOURCE: YOUNG BROTHERS, LLC
tow got under way at 2028 en route to Hilo, entering the forces while in transit at sea.
open ocean about 2115.
Even though machine operators stated they tried to
stack containers with heavy containers on the bottom
and light ones on top, neither the barge team member
Sufficiency of Container-Securing Arrangements on Barges
job descriptions nor the company-provided Container
It is important for cargo planners to have tools, such as stow plans and calculations, to assist with
Lashing Tips included instructions pertaining to the order
determining proper stowage and the sufficiency of securing arrangements for containers stacked
in which to stack containers. Instead, heavy containers
on barges. These tools should address the potential that container stacks may be stacked in a
were often loaded over lighter containers, and stacks
reverse stratified manner.
1, 7, 8, and 10 were loaded almost exactly in reverse
Lessons Learned Vessel Stability

Containing Engine Room
Fires
Accident investigations completed in 2021 demonstrate a variety Through proper design, loading, and Engine rooms contain multiple fuel
of hazards foundational to marine transportation. For example, operation, a vessel should possess sources and are especially vulnerable
stability is critical for any vessel from a canoe to a containership, enough stability to return to its upright to rapidly spreading fires. Following the
position after being heeled over by initiation of an engine room fire, it is
and stability calculations and instructions must be accurate and
any combination of wind, waves or imperative to remove the source(s) of
must be followed. Icing and severe weather are longstanding forces from operations. Stability available fuel to a fire. Designers and
threats to marine transportation, and dovetail with stability criteria, established by regulators, are operators should evaluate fire hazards
concerns when ice is not accounted for in stability calculations. generally recognized as providing an and provide effective means to mitigate
adequate level of safety for vessels them. Vessel owners should encourage
Containment of engine room fires has been a concern since that are operated prudently. The intent crews to familiarize themselves and
vessels have been powered, and proper cargo preparation and of stability requirements is to provide train frequently on machinery, fuel oil,
securement is critical when cargo itself poses a hazard or when information to vessel’s crew that will lube oil, and engine room ventilation
considering the loss of cargo overboard. Other marine casualties enable them to readily ascertain the shutoff systems.
stability of their vessel under varying
closed in 2021 vividly exemplified the need for continuing
loading conditions. A vessel’s stability
improvement to communications, teamwork, and technology. A lack of remote emergency
instructions must be accurate, and the
cut-off valves for the engine room
crew must use the instructions correctly
The NTSB responds to accident lessons by issuing and reiterating was a factor in the Master Dylan
when determining stability to ensure a
safety recommendations, until safety improvements become casualty.
vessel is loaded such that it meets the
realities onboard vessels, throughout the organizations that stability criteria intended by the vessel The lack of a fixed fire-
operate them, and in the Coast Guard’s regulations. designers and approved by regulators. extinguishing system for the
engine room and the loss of
Vessels routinely transit our seas and waterways without incident. electrical power to a single
Inaccurate stability instructions fire pump were factors in the
But when there are tragedies at sea, mariners, masters, and and the effect of icing on stability City of Cleveland casualty.
managers ashore who disregard accident lessons all but invite the were factors in the Scandies Rose
circumstances to repeat. NTSB recommendations can reduce this casualty. Quick and effective actions by the
crew resulted in the successful
risk, once acted upon. But so can you, the individual. Knowing the An incorrect determination of containment and extinguishment
circumstances of the last accident can well be the edge you need vessel stability was a factor in the of the engine room fire aboard the
in preventing the next one. Golden Ray casualty. Ocean Intervention.
View your ship or your company’s operations through the eyes

of our investigators. What lessons might investigators find if your
vessel were in an accident? Have previous investigations yielded
mitigations? We hope that these lessons learned help you, the
reader, to view your own operation with a cold, critical eye and
take appropriate action.
Risk Management and Cargo Preparation and

Icing and Severe Weather Project Planning Securement Teamwork
Severe weather can create challenging A formal risk assessment, which It is important for cargo planners to Safe and effective operations are not
conditions, including strong currents involves identifying hazards and have tools and procedures, such as stow one person’s job. Teamwork is an
and high winds and seas. In cold estimating the risk they pose, is a critical plans, calculations, and preparation essential defense against human error,
weather climates, wave-generated component of casualty prevention. By instruction, to assist with determining and a good team should anticipate
sea spray can cause icing, which can considering the likelihood and severity proper stowage and the sufficiency dangerous situations and recognize
severely affect the stability of a vessel. of each risk, risk matrices increase the of securing arrangements for cargo the development of an error chain.
Additionally, in remote locations where visibility of risks and help managers loaded aboard vessels. These tools and If in doubt, team members should
weather observation sites are more select controls commensurate with procedures must consider the type of speak up or notify a higher authority.
spread out, there can be inaccurate the risk level. With such information, a cargo and the design of the vessel, as Sharing information among crew, pilots,
and less precise forecasts. Marine hazard control plan can be developed well as the potential hazards presented and facility operators and providing
operating companies should develop and implemented. by the cargo. Operators must ensure a thorough turnover are also critical
and continuously evaluate severe that these procedures are followed components of effective teamwork.
weather plans to prepare for challenges during the loading of the vessels.
Inadequate risk management
accompanied by severe weather, and
and project planning were A lack of effective teamwork was
mariners should take caution when
factors in the Waymon Boyd and The improper securing of cargo a factor in the Savage Voyager and
operating in conditions where sea spray
GH Storm Cat casualties. was a factor in the Höegh Xiamen Cheramie Bo Truc No 22 casualties.
icing can occur.
and Ho'omaka Hou/Hoku Loa An ineffective master/pilot
casualties. exchange was a factor in the
Extreme icing and lack of accurate Levant casualty.
weather data were factors in the
Scandies Rose casualty.
Severe weather was a factor in the
I-10 Bridge Barge Breakaway and
CMA CGM Bianca casualties.
80
Lessons Learned Lessons Learned from Marine Investigations
Standard Operating Distress Communications and

Effective Communication Procedures Transiting in Narrow Channels Preparations for Abandonment
Early and effective communication Safety of vessel operations and Narrow channels can be particularly A successful emergency response is
is critical to avoiding close-quarters compliance with mandatory rules challenging to navigate due to the contingent on early distress notification
situations. The use of VHF radio can and regulations can be achieved, in hydrodynamic effects on a vessel and clear, effective communication.
help dispel assumptions and provide part, by establishing clear standard and the substantial amount of traffic Additionally, preparing to abandon ship
bridge teams and vessel operators with operating and emergency procedures. in the waterway. Larger, deeper draft by donning survival suits or personal
the necessary information to adequately In conjunction, regularly training crews vessels are even more prone to the flotation devices increases the likelihood
assess other vessels' intentions. In and personnel involved in operations hydrodynamic forces created by the of survival when experiencing significant
situations where a casualty cannot in standard operating procedures can channel banks and passing vessels. flooding, fire, or other emergencies.
be avoided, early and effective prepare for and mitigate the risk of Transiting a narrow channel, like the
communication can mitigate the effects, emergency situations. Houston Ship Channel, at sea speed,
Early distress communication and
reducing damage, injuries, or loss of life. in which the vessel is at or near its
the crew's preparation to abandon
maximum speed while the engine is less
A lack of specific standard ship in the Rebecca Mary casualty
responsive, provides little room for error
A lack of early and effective operating procedures was a factor contributed to their survival.
and should be avoided.
communication was a factor in in the Atlantic Huron, Golden Ray, An ineffective emergency distress
the Bow Fortune/Pappy's Pride Ho'omaka Hou (below), Hoegh call was a factor in the severity of
and Cooperative Spirit/RC Creppel Xiamen, and Waymon Boyd Transiting a narrow channel at the Höegh Xiamen casualty.
casualties. casualties. sea speed was a factor in the
Early and effective A failure to follow standard Genesis River casualty.
communications during the operating procedures was factor in
Genesis River/Voyager tow casualty the Höegh Xiamen casualty.
likely prevented the loss of the
towing vessel and injuries to its
crew.
Identifying Navigational AIS Data Input for Towing Continuous Monitoring of

Hazards Operations Unmanned Vessels Sufficient Handover Period
Situational awareness demands a To enhance others' situational awareness Fire and flooding are risks not only for Fatigue is a longstanding issue that
mariner should be alert for new hazards and alleviate possible misinterpretation, crewed vessels, but those unattended continues to adversely affect the safety
that can appear along their intended the combined dimensions of a vessel and as well. To protect personnel, property, of marine operations. Failing to get
route. It is important to check the Coast its tow, reflecting the overall area covered and the environment, it is good marine adequate sleep is a high-risk practice
Pilot and navigational charts when by the tow, should be entered into AIS and practice for owners, operators, and that leads to casualties. When joining
developing voyage plans to improve broadcasted while under way. shipyard managers to coordinate and vessels, crewmembers must often travel
knowledge of an area and prepare for a implement some form of continuous long distances, including internationally,
safe passage. monitoring for vessels undergoing and may have little time for rest. It is
Broadcasting inaccurate AIS
maintenance in a shipyard, in lay-up, or critical that vessel operating companies
information was a factor in the
in some other inactive period without ensure that joining crewmembers have
A failure to identify navigational Cooperative Spirit/RC Creppel
regular crews aboard. Continuous the opportunity to obtain adequate rest
hazards was a factor in the casualty.
monitoring can consist of scheduled and allow for a sufficient handover period
Old Glory and Miss Annie
security rounds and/or active monitoring before they take over critical shipboard
casualties.
with sensing and alarm systems. duties.
A lack of continuous monitoring An insufficient crew handover

was a factor in the Alton St. Amant period for a was a factor in the
and Iron Maiden casualties. Atina casualty.
"Though the circumstances vary, our mission is the same for every investigation we lead: to determine what happened
and issue evidence-based recommendations to prevent similar events from occurring in the future…
But stakeholders at all levels must implement our recommendations to ensure safety."
Jennifer Homendy, NTSB Chair
Vessel Particulars by Vessel Group

PERSONS PAGE
VESSEL NAME VESSEL TYPE FLAG LENGTH DRAFT BEAM/WIDTH ON BOARD NO.
CARGO, DRY BULK
Atlantic Huron Self-unloading bulk carrier Canada 736 ft (224 m) 26.5 ft (8.1 m) 26.5 ft (8.1 m) 25 28
Atlantic Venus Dry bulk carrier Panama 590.2 ft (179.9 m) 21.1 ft (6.4 m) 92.5 ft (28.2 m) 20 12
GH Storm Cat Dry bulk carrier Marshall Islands 656 ft (200 m) 43.6 ft (13.3 m) 4106 ft (32 m) 19 40
Glory First Dry bulk carrier Marshall Islands 200 ft (61 m) 13 ft (4 m) 35 ft (10.7 m) 22 76
CARGO, GENERAL
CMA CGM Bianca Containership Malta 1,099 ft (335 m) 42.7 ft (13 m) 140.4 ft (42.8 m) 27 82
Golden Ray Ro/Ro Marshall Islands 656 ft (200 m) 30.9 ft (9.4 m) 116 ft (35.4 m) 24 16
Höegh Xiamen Ro/Ro Norway 599.7 ft (182.8 m) 26 ft (8 m) 81 ft (31.5 m) 21 94
Nomadic Milde General cargo ship Marshall Islands 453 ft (138.1 m) 26.4 ft (8.1 m) 68.9 ft (21 m) 16 14
CARGO, LIQUID BULK
21 (plus 5 on
Atina Oil tanker Malta 898 ft (273.7 m) Aft 28.9 ft (8.8 m) 157.5 ft (48 m) 10
oil platform)
Bow Fortune Chemical/product tanker Norway 600.7 ft (183.1 m) 28.9 ft (8.8 m) 105.6 ft (32.2 m) 29 14
Genesis River Liquefied gas carrier Panama 754 ft (229.9 m) 36.8 ft (11.2 m) 122 ft (37.2 m) 30 38
Levant Liquefied gas carrier Marshall Islands 741.5 ft (226 m) 34.1 ft (10.4 m) 120 ft (36.6 m) 21 56
FISHING
Lucky Angel Fishing vessel United States 75 ft (22.9 m) 11.2 ft (3.4 m) 22.4 ft (6.8 m) 3 46
Master Dylan Fishing vessel United States 85.2 ft (26 m) 12.5 ft (3.8 m) 29.5 ft (9 m) 4 22
Miss Annie Fishing vessel United States 78.2 ft (23.8 m) 6.5 ft (2 m) 22 ft (6.7 m) 3 78
Pappy's Pride Fishing vessel United States 81.7 ft (24.9 m) 9 ft (2.75 m) 24 ft (7.3 m) 4 96
Rebecca Mary Fishing vessel United States 74 ft (22.6 m) 11.2 ft (3.4 m) 22 ft (6.7 m) 4 54
Scandies Rose Fishing vessel United States 130 ft (39.6 m) 11.3 ft (3.4 m) 34 ft (10.4 m) 7 6
OFFSHORE
Cheramie Bo Truc No 22 Offshore supply vessel United States 167 ft (50.8 m) 10.5 ft (3.2 m ) 38 ft (11.6 m) 5 34
Ocean Intervention Offshore supply vessel United States 243 ft (74.1 m) 13.3 ft (4.1 m) 53.5 ft (16.3 m) 16 10
PERSONS PAGE
VESSEL NAME VESSEL TYPE FLAG LENGTH DRAFT BEAM/WIDTH ON BOARD NO.
SPECIALTY/OTHER
Iron Maiden Dive support vessel United States 163.6 ft (49.6 m) 15 ft (4.5 m) 44 ft (13.4 m) 0 90
Waymon Boyd Dredge United States 151.3 ft (46.1 m) 5.5 ft (1.7 m) 33.8 ft (10.3 m) 18 56
TOWING/BARGE
Alton St. Amant Towing vessel United States 83.5 ft (25.4 m) 7 ft (2.1 m) 30 ft (9.1 m) 0 18
Barge breakaway Various towing vessels and tank barges United States 52–200 ft (31–61 m) 7–13 ft (2.7–4m) 20–48 ft (10.4–14.6 m) 9 68
BH 2903 Hopper barge United States 295 ft (89.9 m) 9.6 ft (2.9 m) 54 ft (16.5 m) 0 32
City of Cleveland Towing vessel United States 140 ft (46.2 m) 11 ft (3.4 m) 42 ft (13 m) 9 84
Cole Hopper barge United States 230 ft (70.1 m) 7 ft (2.1 m) 45 ft (13.7 m) 0 70
Cooperative Spirit Towing vessel United States 200 ft (61 m) 10 ft (3 m) 54 ft (16.5 m) 9 24
George C Towing vessel United States 67.5 ft (20.6 m) 6 ft (1.8 m) 26 ft (7.9 m) 4 22
Ho'omaka Hou Deck cargo barge United States 340 ft (103.6 m) 11.9 ft (3.6 m) 90 ft (27.4 m) 0 42
Hoku Loa Towing vessel United States 108 ft (32. 9 m) 16.9 ft (5.2 m) 34 ft (10.4 m) 6 18
Hoku Loa Towing vessel United States 108 ft (32. 9 m) 16.9 ft (5.2 m) 34 ft (10.4 m) 6 18
Island Lookout Towing vessel United States 65 ft (19.8 m) 7 ft (2.1 m) 26 ft (7.9 m) 4 44
Mariya Moran (Tug of ATB–Texas) Towing/Barge ATB United States 121 ft (36.9 m) 17 ft (5.1 m ) 36 ft (11 m) 9 14
Old Glory Towing vessel United States 51 ft (15.7 m) 7 ft (2.1 m) 20 ft (6 m) 4 48
PBL 3422 Tank barge United States 297 ft (90.7 m) 9.5 ft (2.9 m) 54 ft (16.5 m) 0 30
RC Creppel Towing vessel United States 69 ft (21 m) 10 ft (3 m) 30 ft (9.1 m) 4 52
Savage Voyager Towing vessel United States 83.5 ft (25.5 m) 10 ft (3.1 m) 32 ft (9.8 m) 6 22
Susan Lynn Towing vessel United States 119 ft (36.3 m) 6 ft (1.8 m) 28 ft (8.5 m) 1 30
Texas (Barge of ATB–Mariya Moran) Towing/Barge ATB United States 463 ft (141.2 m) 16 ft (4.9 m) 78 ft (23.8 m) 1 84
Trent Joseph Towing vessel United States 67 ft (20.4 m) 10 ft (3 m) 24 ft (7.3 m) 4 12
Voyager Towing vessel United States 68.9 ft (21 m) 8.5 ft (2.6 m) 26.1 ft (8 m) 4 82
YACHT/BOAT
Andiamo Private yacht Marshall Islands 120 ft (36.6 m) 12.9 ft (3.9 m) 25.8 ft (7.9 m) 5 66
Table and Map of Accident Locations

PAGE VESSEL NAME VESSEL GROUP LOCATION
CAPSIZING/LISTING
4 Golden Ray CARGO, GENERAL • Ro/Ro St. Simons Sound, Brunswick River, near Brunswick, Georgia
8 Scandies Rose FISHING • Fishing vessel Pacific Ocean, near Sutwick Island, Alaska
COLLISION
12 Bow Fortune / Pappy's Pride CARGO, LIQUID BULK • Chemical/product tanker / FISHING • Fishing vessel Outer Bar Channel, Galveston, Texas
14 Cheramie Bo Truc No 22 / Mariya Moran–Texas OFFSHORE • Offshore supply vessel / TOWING/BARGE • Towing vessel / Barge Sabine Pass Jetty Channel, Port Arthur, Texas
16 Cooperative Spirit / RC Creppel / Glory First TOWING/BARGE • Towing vessel / Towing vessel / CARGO, DRY BULK • Dry bulk carrier Lower Mississippi River, mile 123, Destrehan, Louisiana
18 Genesis River / Voyager CARGO, LIQUID BULK • Liquefied gas carrier / TOWING/BARGE • Towing vessel Houston Ship Channel, Upper Galveston Bay, Texas
22 Nomadic Milde / Atlantic Venus CARGO, GENERAL • Cargo vessel / CARGO, DRY BULK • Dry bulk carrier Lower Mississippi River, mile 114.5, South Kenner, Louisiana
CONTACT
24 Atina CARGO, LIQUID BULK • Oil tanker Southwest Pass Fairway Anchorage, Gulf of Mexico, 21.5 miles south-southwest of Pilottown, Louisiana
26 Atlantic Huron CARGO, DRY BULK • Self-unloading bulk carrier Soo Locks, Saint Mary's River, Sault Sainte Marie, Michigan
28 Barge Breakaway TOWING/BARGE • Towing vessels / Tank barges San Jacinto River Fleet, San Jacinto River, Channelview, Texas
30 Cooperative Spirit TOWING/BARGE • Towing vessel Lower Mississippi River, mile 121.6, near Luling, Louisiana
32 GH Storm Cat CARGO, DRY BULK • Dry bulk carrier Zen-Noh Grain Facility, Lower Mississippi River, mile 163.8, Convent, Louisiana
34 Island Lookout / BH 2903 TOWING/BARGE • Towing vessel / Hopper barge Albemarle and Chesapeake Canal, mile 15.2 of southern branch of Elizabeth River, to
Albemarle and Chesapeake Canal section of Atlantic Intracoastal Waterway, Chesapeake, Virginia
36 Levant CARGO, LIQUID BULK • Liquefied gas tanker Petrogas Ferndale Wharf, near Ferndale, Washington
38 Old Glory / Cole TOWING/BARGE • Towing vessel / Hopper barge Intracoastal Waterway, Indian River, mile 965, Fort Pierce, Florida
40 Savage Voyager / PBL 3422 TOWING/BARGE • Towing vessel / Tank barge Jamie Whitten Lock & Dam, Tennessee-Tombigbee Waterway, mile 411.9, near Dennis, Mississippi
42 Trent Joseph / George C TOWING/BARGE • Towing vessel / Towing vessel Barataria Waterway, Barataria, Louisiana
FIRE/EXPLOSION
44 Andiamo YACHT/BOAT • Private yacht Island Gardens Deep Harbour Marina, Miami, Florida
46 City of Cleveland TOWING/BARGE • Towing vessel Lower Mississippi River, mile 348, near Natchez, Mississippi
48 Höegh Xiamen CARGO, GENERAL • Ro/Ro Pier 20, Blount Island, Jacksonville, Florida
52 Iron Maiden SPECIALTY/OTHER • Dive support vessel Gulf Intracoastal Waterway, mile 36, Larose, Louisiana
54 Lucky Angel FISHING • Fishing vessel Gulf of Mexico, 20 miles south-southwest of Pascagoula, Mississippi
56 Master Dylan FISHING • Fishing vessel Gulf of Mexico, about 32 miles west-southwest of Port Fourchon, Louisiana
58 Ocean Intervention OFFSHORE • Offshore supply vessel Anchorage B, Mamala Bay, Honolulu, Hawaii
60 Susan Lynn TOWING/BARGE • Towing vessel Barataria Waterway, Lafitte, Louisiana
62 Waymon Boyd SPECIALTY/OTHER • Dredge Epic Dock, near Inner Harbor, Corpus Christi, Texas
FLOODING/HULL FAILURE
68 Alton St. Amant TOWING/BARGE • Towing vessel Harvey Canal, New Orleans, Louisiana
70 Rebecca Mary FISHING • Fishing vessel Atlantic Ocean, about 40 miles south of Martha's Vineyard, Massachusetts
GROUNDING/STRANDING
72 Miss Annie FISHING • Fishing vessel Calibogue Sound, Hilton Head Island, South Carolina
SHIP/EQUIPMENT/CARGO DAMAGE
74 CMA CGM Bianca CARGO, GENERAL • Containership Napoleon Avenue Container Terminal, Lower Mississippi River, mile 100, New Orleans, Louisiana
76 Hoku Loa / Ho'omaka Hou TOWING/BARGE • Towing vessel / Deck cargo barge Pacific Ocean, 6.9 miles north-northwest of Hilo, Hawaii
Acknowledgment
For each marine accident the NTSB investigated, investigators from the Office of Marine Safety
worked closely with the Coast Guard Office of Investigations and Casualty Analysis in
Washington, DC, and with the following Coast Guard units:
ACCIDENT VESSEL COAST GUARD UNIT

Alton St. Amant Coast Guard Sector New Orleans
Andiamo Coast Guard Sector Miami
Atina Coast Guard Sector New Orleans Figure 159. Coast Guard personnel observe the
Atlantic Huron Coast Guard Sector Sault Sainte Marie Waymon Boyd engulfed in flames (above) and the
Barge breakaway (various barges) Coast Guard Sector Houston Golden Ray after it capsized (below and opposite page).
Bow Fortune / Pappy's Pride Coast Guard Sector Ohio Valley Louisville and Marine Safety Unit Texas City SOURCE: COAST GUARD
Cheramie Bo Truc No 22 / Mariya Moran–Texas Coast Guard Marine Safety Unit Port Arthur
City of Cleveland Coast Guard Marine Safety Detachment Vicksburg
CMA CGM Bianca Coast Guard Sector New Orleans
Cooperative Spirit Coast Guard Sector New Orleans
Cooperative Spirit / RC Creppel / Glory First Coast Guard Sector New Orleans
Genesis River / Voyager Coast Guard Sector Houston-Galveston
GH Storm Cat Coast Guard Sector New Orleans
Golden Ray Coast Guard Marine Safety Unit Savannah
Höegh Xiamen Coast Guard Sector Jacksonville
Hoku Loa / Ho'omaka Hou Coast Guard Sector Honolulu
Iron Maiden Coast Guard Marine Safety Unit Houma
Island Lookout / BH 2903 Coast Guard Sector Virginia
Levant Coast Guard Sector Puget Sound
Lucky Angel Coast Guard Sector New Orleans
Master Dylan Coast Guard Marine Safety Unit Houma
Miss Annie Coast Guard Sector Charleston
Nomadic Milde / Atlantic Venus Coast Guard Sector New Orleans
Ocean Intervention Coast Guard Sector Honolulu
Old Glory / Cole Coast Guard Marine Safety Detachment Lake Worth
Rebecca Mary Coast Guard Sector Southeastern New England
Savage Voyager / PBL 3422 Coast Guard Marine Safety Detachment Nashville
Scandies Rose Coast Guard Sector Anchorage
Susan Lynn Coast Guard Sector New Orleans
Trent Joseph / George C Coast Guard Sector New Orleans
Waymon Boyd Coast Guard Sector Corpus Christi
Who Has the Lead:

USCG or NTSB?
In a memorandum of understanding (MOU) signed June 17, 2021, the NTSB and the
US Coast Guard agreed that when both agencies investigate a marine casualty, one agency will
serve as the lead federal agency for the investigation. The NTSB Chair and the Coast Guard
Commandant, or their designees, will determine which agency will lead the investigation.
The NTSB may lead the investigation of

major marine casualties, defined in the MOU
as involving another transportation mode;
serious threat of, or presumed loss of six or Figure 160. The NTSB Chair meets with Coast Guard
Sector New York personnel to discuss marine safety
more lives on a passenger vessel; serious issues and cooperation between the two agencies.
threat of, or presumed loss of 12 or more
lives on a commercial vessel; serious threat
of, or presumed high loss of life beyond the
vessel(s) involved; significant safety issues
relating to the infrastructure of the maritime
transportation system or the environment by
hazardous materials; safety issues of a recurring
character; or significant safety issues relating
to Coast Guard statutory missions, specifically
aids to navigation, search and rescue, and
marine safety.
Table of Figures
1. Ro/Ro vehicle carrier Golden Ray. . . . . . . . . . . . . . . . . . . . . 4 31. Voyager following the accident. . . . . . . . . . . . . . . . . . . . . 18 59. GH Storm Cat moored in Cork, Ireland.. . . . . . . . . . . . . . . 32
2. Golden Ray heeled to its port side.. . . . . . . . . . . . . . . . . . . . 4 32. Screen capture from wheelhouse video on board the 60. The ZGC shoreside facility before the accident. . . . . . . . 32
3. Trackline of the Golden Ray’s transit.. . . . . . . . . . . . . . . . . . 5 Voyager.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 61. ZGC-owned and -operated payloader.. . . . . . . . . . . . . . . . 32
4. The Golden Ray departing Jacksonville, Florida.. . . . . . . . . 5 33. Genesis River under way.. . . . . . . . . . . . . . . . . . . . . . . . . . 19 62. Video footage still image of the GH Storm Cat’s crane. . 33
5. Golden Ray after the heeling event.. . . . . . . . . . . . . . . . . . . .6 34. The accident location.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 63. Point of contact of GH Storm Cat’s crane boom and ZGC
6. Simplified profile of the Golden Ray and 3-D diagram.. . . . . 6 35. Typical Lower Houston Ship Channel profile.. . . . . . . . . . 19 runway no. 3.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7. Responders drilled holes into the hull to access the 36. Hydrodynamic bank effects acting on the Genesis River..20 64. Simple representation of swing span lighting. . . . . . . . . 34
engineers.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 37. Genesis River bridge control panel and EOT lever.. . . . . . 20 65. Island Lookout pushing a barge.. . . . . . . . . . . . . . . . . . . . 34
8. Graphs from the Golden Ray performance study by the 38. Barges from the Voyager postaccident.. . . . . . . . . . . . . . 21 66. Centerville Turnpike Bridge in the closed position. . . . . . 35
NTSB.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 39. Typical steering sequence during head-on meeting in a 67. Centerville Turnpike Bridge at the time of the accident.. 35
9. Emergency responders attempt to rescue the trapped narrow, symmetrical channel.. . . . . . . . . . . . . . . . . . . . . 21 68. Petrogas Wharf with the catwalk and south mooring
engineers and cadet.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 40. Image from video taken from the bridge of the Atlantic dolphin missing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
10. The Scandies Rose wreck.. . . . . . . . . . . . . . . . . . . . . . . . . . 8 Venus.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 69. Petrogas Wharf with a tanker of similar size to the
11. Scandies Rose arriving in Kodiak, Alaska. . . . . . . . . . . . . . 8 41. Nomadic Milde on the right descending bank at the Levant docking starboard side to the wharf.. . . . . . . . . 36
12. Timeline of the Scandies Rose voyage. . . . . . . . . . . . . . . . 9 Cornerstone Dock.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 70. The Levant’s positions as it moved from the anchorage
13. Scandies Rose’s positions before mayday call. . . . . . . . . . 9 42. The Nomadic Milde loading cargo.. . . . . . . . . . . . . . . . . . 23 to the Petrogas Wharf, based on AIS data. . . . . . . . . . . 37
14. Bridge of the Scandies Rose. . . . . . . . . . . . . . . . . . . . . . . 10 43. GPS positions of the Nomadic Milde from the vessel’s 71. AIS tracks of all liquid petroleum gas carrier approaches
15. Diagram of ice and wind acting on the Scandies Rose.. . 10 ECDIS.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 to the Petrogas Wharf. . . . . . . . . . . . . . . . . . . . . . . . . . . 37
16. Coast Guard Cutter Mellon crew breaking ice.. . . . . . . . . 10 44. SP-57B preaccident. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 72. Old Glory under way. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
17. Crewmembers aboard Coast Guard Cutter Polar Star.. . 11 45. The Atina's damaged accommodation ladder.. . . . . . . . . 24 73. The ICW approach to the Peter P. Cobb Memorial Bridge
46. SP-57B's fractured horizontal and damaged leg.. . . . . . . 25 from the south. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
18. Bow Fortune docked after the casualty.. . . . . . . . . . . . . . 12
47. Atina’s trackline taken from VDR data.. . . . . . . . . . . . . . . 25 74. Postaccident damage to the eastern fendering of the
19. The Pappy’s Pride before the accident. . . . . . . . . . . . . . . 12
48. The Atlantic Huron alongside the west center pier.. . . . . 26 bridge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
20. Pappy’s Pride outbound track and Bow Fortune inbound
49. Atlantic Huron under way before the accident.. . . . . . . . 26 75. AIS trackline of the Old Glory and Cole.. . . . . . . . . . . . . . 39
track just before the collision.. . . . . . . . . . . . . . . . . . . . . 13
50. Postaccident photo of the west center pier.. . . . . . . . . . . 27 76. The approximate position of the tow after striking the
21. Pappy’s Pride postaccident damage.. . . . . . . . . . . . . . . . 13
[bridge]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
22. Mariya Moran–Texas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 51. Postaccident photo of vessel’s OD box and control valve
assembly.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 77. Preaccident photo of Savage Voyager.. . . . . . . . . . . . . . . 40
23. Cheramie Bo Truc No 22.. . . . . . . . . . . . . . . . . . . . . . . . . . 14
52. Barges resting against I-10 bridge pilings.. . . . . . . . . . . . 28 78. A tow with a similar arrangement to the
24. Damaged stem of the barge Texas. . . . . . . . . . . . . . . . . . 14 Savage Voyager. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
25. Port side of the Cheramie Bo Truc No 22, post-collision..15 53. Multibeam sonar images overlayed with above-surface
photograph of the I-10 bridge. . . . . . . . . . . . . . . . . . . . . 28 79. Depiction of Savage Voyager tow prior to locking down..41
26. Sabine Pass chart showing path of Cheramie Bo Truc No 80. Depiction of the PBL 3422 in lock chamber.. . . . . . . . . . . 41
22 and Mariya Moran–Texas.. . . . . . . . . . . . . . . . . . . . . 15 54. Layout of San Jacinto River Fleet’s tiers.. . . . . . . . . . . . . 29
55. Cooperative Spirit tow configuration at the time of the 81. Trent Joseph and George C under way before the
27. RC Creppel under way before the accident.. . . . . . . . . . . 16 accident.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
accident.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
28. Cooperative Spirit moored after the accident. . . . . . . . . 16 82. Postaccident damage to Barataria Bridge.. . . . . . . . . . . . 42
56. Cooperative Spirit moored before the accident. . . . . . . . 30
29. Glory First anchored after the accident.. . . . . . . . . . . . . . 16 83. Position of the Trent Joseph tow as it passed through
57. The Hale Boggs Memorial Bridge. . . . . . . . . . . . . . . . . . . 31
30. Rose Point ECS display screenshot from the Cooperative the Barataria Bridge.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Spirit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 58. Track of Cooperative Spirit tow. . . . . . . . . . . . . . . . . . . . . 31
84. Representation of fendering and swing span lighting; 111. Engine room layout and area where captain observed 145. Inverted and mostly submerged Rebecca Mary. . . . . . . 71
postaccident damage to the corner of barge sparking.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 146. Working deck of Rebecca Mary. . . . . . . . . . . . . . . . . . . . 71
JMSS Mobile.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 112. The Master Dylan aground after being towed by the 147. Miss Annie shortly after all three crew were rescued.. . 72
85. Andiamo listing to starboard.. . . . . . . . . . . . . . . . . . . . . . 44 Master Dustin II. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 148. Miss Annie under way.. . . . . . . . . . . . . . . . . . . . . . . . . . . 72
86. Andiamo before the accident.. . . . . . . . . . . . . . . . . . . . . . 44 113. The Master Dylan before the accident.. . . . . . . . . . . . . . 57 149. Image of wreck from NOAA Danger to Navigation
87. Plan view of the lower deck.. . . . . . . . . . . . . . . . . . . . . . . 44 114. Ocean Intervention under way before the accident. . . . 58 Report.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
88. Promotional and postaccident photos indicate location 115. The damaged section of the engine. . . . . . . . . . . . . . . . 58 150. Miss Annie top of wheelhouse and debris washed
of candles.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 116. Cavitation erosion in upper connecting rod bearings.. . 59 ashore.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
89. Flames coming from the starboard rudder room door.. . 46 117. The damaged section of the engine. . . . . . . . . . . . . . . . 59 151. Simplified diagram of gantry crane no. 6.. . . . . . . . . . . .74
90. City of Cleveland under way before the accident.. . . . . . 46 118. Susan Lynn under previous ownership. . . . . . . . . . . . . . 60 152. CMA CGM Bianca before the accident. . . . . . . . . . . . . . 74
91. Semi-portable extinguisher in starboard engine room 119. The Susan Lynn at its berth following the fire.. . . . . . . . 60 153. Security camera footage of Napoleon Avenue Container
following the fire.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 120. The Susan Lynn port generator and lower engine room Terminal pier. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
92. City of Cleveland main deck arrangement.. . . . . . . . . . . . 47 following the fire.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 154. Simplified diagram of CMA CGM Bianca's mooring
93. Holed crankcase at no. 4 left cylinder inspection cover, 121. Ruptured Susan Lynn port generator oil reservoir.. . . . . 61 line.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
left and right connecting rods, and connecting rod 122. Crew quarters on board the Susan Lynn.. . . . . . . . . . . . 61 155. Locking cone and stacking cone similar to ones used
clamp.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 on the Ho’omaka Hou.. . . . . . . . . . . . . . . . . . . . . . . . . . . 76
123. Simplified inboard profile of the Susan Lynn.. . . . . . . . . 61
94. Höegh Xiamen under way.. . . . . . . . . . . . . . . . . . . . . . . . . 48 156. The collapsed row of containers, from the starboard
124. Waymon Boyd preaccident.. . . . . . . . . . . . . . . . . . . . . . . 62
95. Thermally damaged vehicles after removal from the quarter of the Ho’omaka Hou.. . . . . . . . . . . . . . . . . . . . . 76
125. Post-fire photo of the Waymon Boyd, before sinking. . . 62
Höegh Xiamen.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48 157. Portion of stow plan showing container weights of the
126. Dredge Waymon Boyd before the accident. . . . . . . . . . . 62
96. Firefighters conducting exterior boundary-cooling.. . . . . 48 toppled container row.. . . . . . . . . . . . . . . . . . . . . . . . . . . 77
127. Dredge template and pipeline locations. . . . . . . . . . . . . 63
97. Arrangement of portside aft cargo ventilation trunks on 158. The Ho’omaka Hou, loaded for a previous voyage, as
deck 12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 128. EPIC dock project area.. . . . . . . . . . . . . . . . . . . . . . . . . . 63 viewed from the stern.. . . . . . . . . . . . . . . . . . . . . . . . . . . 77
98. Höegh Xiamen fire zones.. . . . . . . . . . . . . . . . . . . . . . . . . 49 129. Typical dredging operations. . . . . . . . . . . . . . . . . . . . . . 64 159. Coast Guard personnel observe the Waymon Boyd
99. Damaged aft ventilation housings after the reported 130. Waymon Boyd general arrangements. . . . . . . . . . . . . . . 64 engulfed in flames and the Golden Ray after it
explosion.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 131. The Waymon Boyd at the EPIC Marine Terminal.. . . . . . 64 capsized. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
100. Engine 48 arriving on scene; the white paint on 132. Waymon Boyd lever room postaccident.. . . . . . . . . . . . 65 160. The NTSB Chair meets with Coast Guard Sector
the starboard side of the vessel exhibits thermal 133. Cutterhead assembly after underwater recovery. . . . . . 65 New York personnel.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
discoloration. Vents exhibit smoke flow.. . . . . . . . . . . . 50 134. Screen captures from security camera.. . . . . . . . . . . . . 65 161. An NTSB investigator during the Scandies Rose
101. Battery in towed vehicle removed from deck 7; battery 135. Tugboats Ted C Litton and Evelena.. . . . . . . . . . . . . . . . 65 investigation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
in forklift vehicle removed from deck 5.. . . . . . . . . . . . . 50 136. Waymon Boyd AIS data on the accident date.. . . . . . . . 66
102. Completed “Vehicle Lashing Inspection Procedure.”. . . 51 137. Postaccident pipeline marker survey locations, accident
103. Excerpt from Grimaldi battery disconnect procedure.. . 51 location, and pipeline TX219; preaccident photo of
104. The Iron Maiden before the accident.. . . . . . . . . . . . . . . 52 marking cane pole. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
105. Starboard exhaust trunk fan.. . . . . . . . . . . . . . . . . . . . . . 52 138. Post-salvage images of the Waymon Boyd.. . . . . . . . . . 67
106. Fire damage to living quarters and generator room.. . . 52 139. Alton St. Amant under way before the accident.. . . . . . 68
107. General layout of the Iron Maiden generator room and 140. Alton St. Amant, partially submerged.. . . . . . . . . . . . . . 68
surrounding main deck spaces.. . . . . . . . . . . . . . . . . . . 53 141. Simple profile of the Alton St. Amant. . . . . . . . . . . . . . . 69
108. Lucky Angel on fire. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 142. Open access hatches. . . . . . . . . . . . . . . . . . . . . . . . . . . 69
109. Lucky Angel before the accident.. . . . . . . . . . . . . . . . . . 55 143. Rebecca Mary before the accident. . . . . . . . . . . . . . . . . 70
110. Area of accident where the Lucky Angel fire started. . . 55 144. Rebecca Mary before the accident. . . . . . . . . . . . . . . . . 70
90
NTSB 2021–2022
Most Wanted List of Improve Passenger and Fishing Vessel Safety

Passenger and fishing vessels present distinct safety challenges within the marine transportation industry.
Transportation Safety Improvements The US Coast Guard can improve safety on both passenger and fishing vessels by implementing our
recommendations.
T
he NTSB’s Most Wanted List (MWL) highlights
transportation safety improvements needed
now toprevent accidents, reduce injuries, and save
PASSENGER VESSELS 79 feet long. Many fishing crews aren't trained in
Passenger vessels range in size from small charter stability management techniques or emergency
lives. We use the list to focus our advocacy efforts response, and we have found that many vessels do not
vessels, such as dive boats and amphibious passenger
during the current MWL cycle. The NTSB urges have proper life-saving equipment, such as flotation
vessels (DUKW boats or “duck boats") to large cruise
lawmakers, industry, advocacy and community devices and search-and-rescue locator devices.
ships operating in international waters. The number of
organizations, and every American to learn
passengers and crew on these types of vessels may vary. OUR SOLUTIONS . . .
more about what they can do to implement and
champion the 2021–2022 MWL. Adopting NTSB Fires pose a catastrophic threat to passenger vessels, TAKE ACTION NOW!
safety recommendations associated with these as we saw in the 2019 Conception dive boat accident The Coast Guard needs to act on our
safety items will save lives. off the coast of California in which 34 people died. Our recommendations. Although many of our
investigations have revealed that crew training and recommendations call for regulatory action, passenger
AVIATION
safety regulations for these vessels vary, increasing and fishing vessel associations, training centers, and
• Require and Verify the Effectiveness of the risk to passengers and crew.
Safety Management Systems in all Revenue marine safety advocacy groups should also promote
To prevent needless deaths and mitigate injuries, awareness and encourage operators to take voluntary
Passenger-Carrying Aviation Operations
passenger vessels should have safety management measures to improve safety on their vessels.
• Install Crash-Resistant Recorders and Establish systems, use voyage data recorders, and provide On passenger vessel safety, the Coast Guard should:
Flight Data Monitoring Programs adequate fire-detection and extinguishing systems
and enhanced emergency egress options. Operators • Require all operators of domestic passenger vessels
HIGHWAY
need to ensure their crews have enhanced training that to implement safety management systems.
• Implement a Comprehensive Strategy to
includes fire drills and firefighting techniques. We also • Develop a US voyage data recorder standard for
Eliminate Speeding-Related Crashes
need to see more roving patrols on our waterways to ferry vessels that meets the International Maritime
• P
rotect Vulnerable Road Users through a ensure passengers are being transported safely. Organization's performance standards and require the
Safe System Approach installation of such equipment on new and existing
• Prevent Alcohol- and Other Drug-Impaired Driving COMMERCIAL FISHING ferry vessels.
• Require Collision-Avoidance and The commercial fishing industry, which remains • Require companies operating domestic passenger
Connected-Vehicle Technologies on all Vehicles largely uninspected, is another marine sector of vessels to develop and implement a preventive
concern. Fishing consistently tops the list of most maintenance program for all systems affecting the
• Eliminate Distracted Driving
deadly occupations, due, in large part, to challenging safe operation of their vessels.
MARINE work environments, such as poor weather and rough • Evaluate the feasibility of creating a passenger vessel
• Improve Passenger and Fishing Vessel Safety waters. These conditions threaten vessel stability and safety specialist billet and staff sector-level billets at
RAIL, PIPELINE, AND HAZARDOUS MATERIALS integrity—an issue we have seen in our investigations. each sector that has the potential for a search and
More than 800 fatalities have occurred on fishing rescue activity.
• Improve Pipeline Leak Detection and Mitigation vessels in the past two decades.
• Improve Rail Worker Safety • Require fire-detection systems in unoccupied spaces
We need new standards to address—and periodically with machinery or other potential heat sources on
To find out how to take action, and for a reassess—intact stability, subdivision, and watertight board small passenger vessels.
complete history of action or inaction on these integrity in commercial fishing vessels up to
recommendations, visit www.ntsb.gov/mwl.
91
• Require newly constructed vessels and those On passenger vessel safety, operators and • Re-evaluate emergency procedures regarding
currently in service with overnight accommodations organizations representing small passenger vessel lifejacket donning aboard modified DUKW amphibious
to have interconnected smoke detectors in all operators should: passenger vessels when equipped with fixed
accommodation spaces. • Implement safety management systems to improve canopies.
• Develop and implement an inspection procedure to safety practices and minimize risk. • Share the circumstances of the Conception accident
verify that small passenger vessel owners, operators, • Develop and/or improve procedures to manage and and encourage members to voluntarily install
and charterers are conducting roving patrols. account for all persons aboard in the event of a mass interconnected smoke and fire detectors in all
• Require newly constructed small passenger vessels evacuation of a ship while in port. accommodation spaces and a secondary means of
and those constructed prior to 1996 with overnight escape into a different space than the primary exit.
• Perform a worst-case scenario risk assessment for
accommodations to provide a secondary means of all active water-based fire-suppression systems to On fishing vessel safety, the Coast Guard should:
escape into a different space than the primary exit. evaluate whether the existing freshwater supply is • Establish standards for new and existing commercial
• Review the suitability of regulations regarding means sufficient. fishing industry vessels of 79 feet or less in length that
of escape to ensure there are no obstructions to • Review lifesaving appliance training programs, address intact stability, subdivision, and watertight
egress on small passenger vessels constructed prior including recordkeeping procedures, and revise the integrity and include periodic reassessment of the
to 1996 and modify regulations accordingly. programs to ensure that crewmembers are proficient vessels' stability and watertight integrity.
• Ensure that amphibious passenger vehicle operators with onboard systems. • Require all owners, masters, and chief engineers
tell passengers that seat belts must not be worn while • Provide formal and recurrent training to shoreside of commercial fishing industry vessels to receive
the vessel/vehicle is operated in the water and visually management and senior shipboard officers to ensure training and demonstrate competency in vessel
check that each passenger has unbuckled his or her that all senior leaders are fully knowledgeable about stability, watertight integrity, subdivision, and use of
seat belt. the policies and procedures in the safety management vessel stability information, including preventing and
• Require DUKW amphibious passenger vessels to system. properly responding to emergency situations as well
have sufficient reserve buoyancy through passive as the actual use of emergency equipment.
• Develop and apply an oversight system to
means, and for those that don't, require the removal of ensure that maintenance programs comply with • Require that all personnel employed on vessels in
canopies, side curtains, and their associated framing the manufacturer's recommended preventive coastal, Great Lakes, and ocean service be provided
during waterborne operations. maintenance program. with a personal locator beacon.
• Require that amphibious passenger vessels equipped • Revise marine firefighting and job training programs,
with forward hatches enable operators to securely including documenting, both onboard and ashore, that STATS TO KNOW
close them during waterborne operations. all crewmembers are qualified and can continually
• Review the circumstances of the Stretch Duck 7
sinking and other amphibious passenger vessel
demonstrate proficiency in their duties.
• Review and revise current operating policy to provide
58,000
Commercial fishing
805 fatalities
In commercial fishing
accidents, and revise Navigation and Vessel specific guidance on vessel operations when adverse vessel accidents in
Inspection Circular 1-01 to address the issues found vessels in service in the US between 2000
conditions could be encountered during any part of the US in 2020.
in these accidents. the waterborne tour by implementing a go/no-go and 2020.
SOURCE: COAST GUARD SOURCE: COAST GUARD
• Examine existing training and knowledge policy.
36
requirements for understanding and applying • Modify spring-loaded forward hatches of modified NTSB recommendations pertaining
fundamental weather principles to waterborne DUKW amphibious passenger vessels to enable their to fishing or passenger vessel safety
operations for Coast Guard-credentialed masters who closure during waterborne operations. issued since 2002.
operate small passenger vessels; and, if warranted,
require additional training.
To see our full list of recommendations, visit ntsb.gov, select Search CAROL on the home page, then Published Searches.
NTSB Office of Marine Safety
T
he Office of Marine Safety investigates and determines the probable cause of major marine casualties
on or under US territorial waters, major marine casualties involving US-flagged vessels worldwide, and
accidents involving both US public (federal) and nonpublic vessels in the same casualty. In addition, the office
investigates select catastrophic marine accidents and those of a recurring nature.
The US Coast Guard conducts preliminary investigations of all marine accidents and notifies the NTSB when an
accident qualifies as a major marine casualty.
For select major marine casualties, the office launches a full investigative team and presents the investigative
report to the Board. For all other major marine casualties, the office launches a field team of marine investigators
to the scene to gather information to develop either an investigation report or brief. Most briefs are issued by the
office director through delegated authority; those involving public or nonpublic marine accidents and those that
contain safety recommendations are adopted and issued by the Board.
The office also participates with the Coast Guard as a substantially interested State in investigations of serious
marine casualties involving foreign-flagged vessels in international waters. Additionally, as part of the NTSB's
international marine safety program, the office coordinates with other US and foreign agencies to ensure
consistency with IMO conventions and cooperates with other accident investigation organizations worldwide at
annual meetings, which track developments related to marine accident investigations and prevention.
The NTSB is the only federal organization that performs independent, comprehensive, and transparent
multidisciplinary investigations to determine the probable cause of marine accidents, with the goal of making
safety recommendations to prevent similar events from occurring in the future.
The thoroughness and independence of these investigations maintain public confidence in marine transportation
systems and provide policymakers with unbiased analysis.
Figure 161. An NTSB investigator measures ice accretion

on a crab pot during the Scandies Rose investigation.
www.twitter.com/ntsb
www.instagram.com/ntsbgov
www.facebook.com/ntsbgov
National Transportation Safety Board www.youtube.com/user/ntsbgov
490 L’Enfant Plaza, SW • Washington, DC 20594 • (202) 314-6000 www.flickr.com/ntsb
www.ntsb.gov www.linkedin.com/company/ntsb
ntsb.gov

NTSB Safer Seas Digest 2021 2022 10

Uploaded by

Copyright:

Available Formats

NTSB Safer Seas Digest 2021 2022 10

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NTSB Safer Seas Digest 2021 2022 10

Uploaded by

Copyright:

Available Formats

N A T I O N A L T R A N S P O R T A T I O N S A F E T Y B O A R D

SAFER SEAS Digest

Lessons Learned from

The NTSB is an independent federal agency charged by Congress with investigating

SAFER SEAS Digest

Vessel Group Key CARGO, GENERAL COMBATANT/MILITARY PASSENGER TOWING/BARGE

30.9 feet (9.4 meters). The Golden Ray arrived outside

moved the rudder to comply. At 0136:15, with the

vessel’s speed at 12.9 knots, the helmsman informed the

SAFETY ISSUES SAFETY RECOMMENDATIONS

Capsizing and Sinking

and deployed a C-130 aircraft. The first helicopter arrived

at the captain’s mayday coordinates around 0200 and

At 1537:32, the vessels collided just outside the Outer

The probable cause of the collision of the inbound

OFFSHORE • TOWING/BARGE offshore supply vessel Cheramie Bo Truc No 22

Tug and Barge

Figure 25. The port side of the Cheramie Bo Truc No 22,

Once in sight of one another, there was minimal time

CARGO, LIQUID BULK • TOWING/BARGE

Figure 33. Genesis River under way in Bolivar Roads near

At 1446, Pilot 2 ordered the Genesis River’s engine to

The probable cause of the collision between

CARGO, GENERAL • CARGO, DRY BULK

At 1655, the port side of the Nomadic Milde collided

had decreased or if the ship had moved. Had this been

CARGO, LIQUID BULK

Figure 45. The Atina's damaged accommodation ladder. SOURCE: COAST GUARD

CARGO, DRY BULK

Contact of Bulk Carrier

Technicians discovered that a bearing within the CPP’s

TOWING/BARGE historic flood waters and high river current,

Towboat crews corralled the block of six barges as they

The probable cause of the barge breakaway and

The probable cause of the contact of the

CARGO, DRY BULK

Contact of Bulk Carrier

stopping the lift and establishing communications with

The GH Storm Cat’s company quality, health, safety, and

Vessel Crane Operations

When the swing bridge was about halfway open, the

CARGO, LIQUID BULK

The probable cause of the contact of the liquid

Passage Planning for Short Transits

Contact of Old Glory

Figure 75. AIS trackline of the Old Glory and Cole outside

Figure 76. The approximate position of the tow after

Contact Fire aboard

Engine Room Fire

The engine room was not equipped with a fixed

Figure 97. Arrangement of portside aft cargo ventilation

Beginning at 1549, the master made several calls for

that battery securement crews were unable to gain

access to the engine compartments. If they had followed

Figure 102. Completed “Vehicle operator) “Cargo Safety Awareness” procedure, to

significant damage to the passageway outside the

Engine Room Fire

close the two access doors or two exhaust fan vents

The probable cause of the engine room fire