For the first time in the year since a 656-foot cargo vessel capsized off the coast of Georgia, crew members and the ship’s owners are speaking publicly about the moments before the Golden Ray listed, overturned, and triggered a multiday rescue effort to save the 24 people on board.
The vessel has remained on its side in the St. Simons Sound — with its cargo of more than 4,000 vehicles still inside — since September 8, 2019. The plan to have cut it into pieces and hauled away is currently paused until early October because of weather and Covid-19 related challenges.
A public hearing, part of the investigation into why the vessel capsized, wraps up on Tuesday. Here’s what we know so far:
Crew didn’t sense anything wrong
Capt. Jonathan Tennant, the pilot who steered the ship out of the port at 1 a.m. on Sept. 8, 2019, recalled calm winds and good visibility. He said typically there are warning signs when something is wrong — alarms, the sound of crew members speaking rapidly in a native language, the engine room calling up. But none of that happened.
“Everything was just as normal as it could be, until it capsized,” Tennant said.
Tennant told investigators the ship started leaning in a turn and started to over-rotate to starboard. He applied what’s known as counter-rudder to reduce the rate of the turn, but it wasn’t enough. He said the accident happened within seconds.
“When I felt like I was losing control of the vessel, I reached behind me where I propped up the ship’s radio and said to Jamie on the inbound, ‘watch out, Jamie. I’m losing her,’ in which time she capsized, I dropped the radio, held onto the gyro, the ship capsized, I tried to ease the rudder, still trying to drive the ship.”
Tennant said he had no context that the ship’s rudder and propeller were already out of the water.
“I was still like an airline pilot trying to drive the plane, trying to solve the problem until it flew into the ground,” he said.
He said he saw fear in the faces around him.
The other crew members were South Korean nationals. Twenty people were rescued within a few hours. Four remaining crew members stayed trapped in the engine room.
The last person aboard, Junyong Kim, was rescued about about 36 hours after the accident.
He gave a written testimony for the hearing this week. Kim, the first engineer on the ship, also said nothing seemed out of the ordinary until the ship turned sideways.
In his statement, Kim said that he was in the engine room, trying to protect himself and the second engineer.
“I grabbed him not to fall down to the end of the engine room because the engine room was … a bit long. And if he falls down to end, he could be injured. So, I try to grab him to not fall down to the end of this.”
Kim said the generator went out, the engine shut down, and the standby generator also went out.
He described water rising, and the room becoming incredibly hot. Capt. John Reed of the US Coast Guard testified separately that the inside reached more than 150 degrees, with passageways becoming a “death trap.”
Kim said he and others eventually went into the water that had risen in the engine room to stay cool.
Smoke was visible throughout the rescue mission, indicating fires on board. Rescuers had to bore through fireproof glass and navigate passageways that had become vertical drops to rescue the people in the engine room.
“I keep thinking, why it happens to me. I think, if I have water, I could survive two days more,” Kim said in his statement.
Investigators asked about how the cars were loaded
The Golden Ray was originally supposed to travel from Mexico to several ports in the US, ending in Wilmington, North Carolina. Hurricane Dorian slightly altered its plans.
There was a delay leaving the port in Texas, after which the route changed to go to Jacksonville, Florida, before Brunswick, Georgia.
Investigators asked Mike Mavrinac, operations manager for ship owner Hyundai Glovis, about those changes and any issues with loading or unloading cargo.
Mavrinac said at the Port of Brunswick, “we did have some tighter space, and we added some additional cargo downstairs on the main deck, deck five.”
The Korean Maritime Safety Tribunal, one of the investigating entities, challenged Mavrinac about not having a final load plan provided to the vessel before departure. Mavrinac pushed back, saying the stevedore usually hands a handwritten plan to the crew, and that the crew uses that to account for the cargo present.
The man in charge of creating a plan for loading and unloading, Sammy Maataki of Norton Lilly, said there was eventually less cargo loaded onto deck 12 and more vehicles loaded on deck five out of Brunswick.
“Guessing lack of space. They ran out of space, so they put the remaining balance on five,” Maataki said.
A National Transportation Safety Board investigator asked whether Maataki considers difference in weight between the cars that are unloaded and the new ones brought on board.
Maataki answered, “No, like I said, we receive a plan from Mexico reflecting the cargo to be discharged between the Gulf and the East Coast. And based off the space that’s available at the port of load, we just place it in a way where it’s an efficient load and discharge operation.”
A cause won’t be determined for a while
Jason Neiman, public affairs officer with the US Coast Guard, said the investigation is complex and, therefore, may take many months to complete.
The process of removing the stranded vessel has also been prolonged. Ten members of the salvage crew tested positive for Covid-19 and 50 were quarantined. Some of the people who tested positive are critical to the project, including salvage and support personnel. Currently, no one remains in isolation or quarantine.
The removal process was paused in late July, with a restart expected in early October.
Ongoing support operations have continued during the pause, including teams ensuring that any releases of oil are quickly identified and remediated to the maximum extent possible.
When the project restarts, part of the Covid-19 mitigation plan involves having salvage crew members sequester for two weeks before being housed in a facility throughout the duration of the removal process, which is expected to be about eight weeks.