by Rachel Slade
Tom Roth-Roffy was on that mission. He boarded the 226-foot vessel in Virginia and rode it nearly one thousand miles down to the warm waters off the Bahamas. Once the boat arrived in the search area three days later, the technical team aboard the Apache towed a seventy-pound, bright yellow stingray-shaped pinger locator back and forth along the search pattern, hoping to pick up a signal from El Faro’s beacon.
The nerve center of the operation was set up in a narrow, windowless room in the Apache where an array of outboard gear and servers stacked in racks along one wall fed nearly a dozen computer screens of various sizes.
The pinger locator had to be within a certain distance of the beacon, so the technicians towed it far below the Apache on a long cable. Though it was sensitive, the locator couldn’t be rushed. The initial two-hundred-square-mile search went at about two miles per hour. When the Apache reached the edge of its search field, the boat above had to reverse direction, which meant waiting for the locator to swing back and settle into position before continuing. After four days of dragging the locator twenty-four hours a day, the technical crew aboard the Apache heard nothing from El Faro. The massive container ship had seemingly vanished.
The Apache team was confident that they were close enough to El Faro to pick up the pinger’s signal if it was working. Clearly, something was wrong.
They switched to using a side-scan sonar called Orion, a thirty-six-hundred-pound truck-size machine that sends pulses down into the deep to produce contour images of the seafloor on the technicians’ computers. Orion had been strapped to the Apache’s deck, and the crew used the boat’s crane to carefully lift it up and over the side, then gently release it into the sea, its fiber-optic umbilical cord trailing behind it as it traveled down into the void.
Again, with every direction change following the search field, they had to winch Orion almost to the surface, turn the boat, then send Orion back down and wait for it to stabilize before proceeding. It was tedious work, projected to take fourteen days to complete.
The Apache’s crew was divided into six-hour watches to monitor the computer screens for anything that looked unusual.
After five days of dragging, at 1:35 on October 31, a technician saw something new: right angles on the otherwise indeterminate seafloor. Then he saw shadows. He jacked up the resolution of Orion’s imagery to reveal what looked like a tiny object the size and shape of the missing container ship. It was unmistakable.
Tom got the wake-up call as soon as the vessel was discovered. He hurried over to the control center to watch the recording of the initial sonar echoes. On the computer screen, Tom clearly saw a container ship resting upright on the ocean floor. He could make out what appeared to be the ship’s house and the exhaust stack. Finer resolution revealed that El Faro had landed with such force that the sandy ocean floor around it was frozen in the shape of a permanent splash.
El Faro and her secrets lay directly below him.
Tom felt a great sense of relief wash over him. The investigation, one of the most important of his career, now had an accident site that he could study. They could scan the hull to determine whether it had been breached. They could examine patterns on the seafloor to get a better understanding of how it went down. Even better, the ship’s VDR might be accessible somewhere near the wreck. That black box was the holy grail of any investigation. Tom was one step closer to solving a case. He was elated.
After locating the hull, the techs slowly winched up Orion, and then sent down a CURV-21 to document the site. The eight-foot-high, sixty-five-hundred-pound fiber-optic cable-controlled vehicle was capable of withstanding three tons per square inch, the pressure at El Faro, three thousand feet deeper than the Titanic. The force down there was so great that it caused the heavy steel containers that were once lashed to the ship’s decks to crush inward as they sunk.
A primitive version of the CURV-21 had been used to help the NTSB learn more about another famous shipwreck, the SS Edmund Fitzgerald, a 729-foot freighter that disappeared in 1975 while sailing in a severe storm on Lake Superior. All twenty-nine crewmen aboard perished; like El Faro, there were no eyewitnesses.
The great tragedy of the Fitzgerald—the worst peacetime accident on the Great Lakes—led to major industry changes. It was a benchmark case for the Marine Board of Investigation, one that Captain Neubauer and Tom Roth-Roffy thought a lot about when El Faro disappeared.
Ultimately, the cause of the Fitzgerald’s sinking was inconclusive, and the event continues to draw armchair theorists, but many of the contributing factors debated during that case came up again in the El Faro investigation.
It’s possible that opened or damaged hatch covers may have caused flooding in the thirty-five-foot waves that ultimately overwhelmed the ship, the 1970s US Coast Guard report theorized. This flooding could have been slowed by the installation of watertight doors between the vessel’s holds, but the shipping industry had fought against such protective measures for years, claiming that such modifications would be prohibitively expensive. Without watertight separation between her holds, the Fitzgerald probably went down very fast. The captain didn’t even have time to send a distress signal.
The Fitzgerald was also overloaded—her Plimsoll line had been raised three times in a few short years leading up to the accident, allowing her freeboard to run three feet closer to the water than specified by her original designers. That made her very vulnerable to flooding in foul weather.
Another problem revealed in the case: the National Weather Service had inaccurately forecasted the conditions on the Lakes during the storm that sunk her. In fact, the seas were much higher than predicted and may have caused a few massive rogue waves.
There were human factors involved as well. The Fitzgerald’s captain was known for pushing his vessel hard and flouting storm warnings. When weather kicked up, he’d thrill to it. He boasted about the strength of his ship and the bravado of his crew—they could take on anything Mother Nature could throw at them. In his book Deep Survival, Laurence Gonzales writes, “The word ‘experienced’ often refers to someone who has gotten away with doing the wrong thing more frequently than you have.” This captain was experienced in exactly in that way.
The sinking of the Fitzgerald showed the coast guard that shipping companies weren’t capable of regulating themselves; the organization needed to step up to protect the men and women who crewed these vessels. Among the many new regulations implemented following the sinking: shippers had to provide survival suits for all hands sailing on the Great Lakes to stave off hypothermia and drowning; Great Lakes navigation charts were improved; vessels above a certain tonnage had to be equipped with depth finders and EPIRBs; and the coast guard committed to boarding and inspecting all Great Lakes ships for watertight integrity each year before the fall storms kicked in.
These new regulations served the industry well for decades after the Fitzgerald catastrophe, with the exception of the sinking of the Marine Electric in 1983. American commercial shipping enjoyed a nearly spotless safety record, with no major losses after that.
But almost forty years to the day, the specter of the Fitzgerald returned when El Faro went down. Had regulators and industry slacked off in the ensuing time? Had we all become too complacent?
That’s how Tom Roth-Roffy found himself in the dark, chilly control room on the Apache, transfixed by the images being sent back to his computer screen from the deepest Atlantic by the CURV-21.
The CURV ran tethered to the Apache by a thirty-six-thousand-foot-long electro-optical-magnetic cable that wrapped around an enormous winch on the vessel’s deck. Using the CURV’s thrusters, the techs could fly it like a drone, but very slowly.
When the CURV was in position close to the sunken hull’s location, the techs switched on its high-resolution cameras and bright lights to illuminate the abyss. A murky gray-blue landscape appeared on their screens. If the CURV got too close to the sea bottom, its thrusters kicked up fine particles that got caught in its bright spotlight
, temporarily blinding them.
The first thing the CURV revealed was the miles-long and -wide path of debris and containers. As the loaded freighter went down, everything she held peeled off, leaving a trail for the CURV to follow, like breadcrumbs in the forest. Some containers opened as they made their rapid journey three miles to the bottom, scattering their contents onto the soft ocean floor. Cars spilled out of the ship’s hull in slow motion, their components breaking free and raining down. The site looked like a murky gray sprawling city made entirely of junk.
Tom saw a bicycle. Car batteries. Crumpled containers everywhere. Flung about was the eerily familiar detritus of modern life—a printer, a microwave, the top of a car, abstractly rendered in the foreign dusty desert of the deep Atlantic.
Eventually, they came upon the hull—a steep wall of solid steel.
In the quiet calm of the ocean’s depths, the CURV glided along the port side of the ship, recording the frozen remains of a calamitous event. Then the CURV reached the stern. It turned the corner and slowly panned up. Emerging from the blue gloom, white letters stood proud on the twisted hull: EL _ARO, San Juan PR. The steel where the F had been was crushed inward.
Tom and the tech team gazed at those words in silence.
After lingering for a few minutes, the CURV moved up to the vast top deck, now empty of the cargo boxes that once were stacked three-high. It ran down the length of the deck like a plane taxiing on a runway until it reached the multistory wheelhouse where the ship’s crew once lived and worked. The CURV’s operator pivoted its camera and lights upward and began puttering up to the top of the house, where the VDR was expected to be. As the vehicle rose, those watching aboard the Apache shuddered.
El Faro’s top two stories, including the navigation bridge, mast, and VDR, were gone.
From the way she’d landed, the team theorized that after El Faro sank below the surface, she spiraled down at about 45 miles per hour stern-first at a steep angle to her final resting place on the ocean floor. She descended so quickly that at some point, the navigation bridge must have torn away. Later, another theory would be put forth: that when the cooler waters hit the ship’s boilers’ superheaters, there was an explosion that blew off the house.
After weeks of searching, the NTSB had found the ship, an important first step. But Tom’s mood was solemn. The VDR was attached to the roof of the navigation bridge, and that piece of the ship was nowhere to be seen. They hadn’t located the one thing that would unlock his investigation. If they couldn’t find the VDR, the NTSB and the coast guard would have to rely on conjecture. But how would they find something the size and shape of a coffee can in miles and miles of wreckage?
Chapter 25
How to Sink a Ship
When El Faro disappeared, John Glanfield, the retired shipbuilder, then seventy-five years old, pulled out the stack of ancient Sun Ship blueprints that he kept in his basement. He used the drawings to begin looking for flaws in the ship’s design that could have sunk her. Driven by curiosity and anguish, he needed to know. He needed to understand how one of his beloved children could end up three miles down, taking thirty-three people with her.
John sat at his kitchen table making a paper model of the ship’s midsection, trying to reconstruct what might have happened to her on the morning of October 1, 2015. Building models came naturally to John. Throughout his life, he’d nurtured his artistic tendencies by drawing cartoons on Sun Ship drafting paper and crafting models to amuse his shipyard colleagues.
When his kids were little, John would put them to bed and disappear to the basement, working for hours on tiny versions of the ships he’d built or admired. He once created a five-foot-long Titanic for a young man who believed he was the reincarnation of someone who’d died when the great ship went down. When John went to deliver the model to his client’s Philadelphia apartment, he was amazed to see that the rooms had been done just like a cabin of the ill-fated passenger ship, complete with heavy drapes, brass, and oriental rugs.
John developed a yen for crafting ships in a bottle, too. He made a model of the Hughes Glomar Explorer, the vessel that recovered the Soviet sub. His model was built into the oversize lightbulb that once sat atop Sun Ship’s tallest crane to warn away aircraft. To complete the model, John had to imagine what the mysterious CIA-commissioned claw looked like because he’d never seen it; nor had anyone he knew. The miniature “drilling rig” rests atop a sliver of blue sea. Below it, a thin cable hoists a yellow five-armed claw; in its clutches is a black-painted sub, carved out of wood.
When John heard about the sinking of El Faro, he built several El Faros in a bottle for families of the deceased. Red, blue, and gray hand-lettered “Sea Star” boxes are stacked on the deck. A tiny plume of cotton trails from the ship’s red and blue stack.
John had a hard time getting addresses of the families to send them his ships. So he turned to Facebook where he found Deb Roberts, who’d lost her son, Michael Holland. Deb helped connect John with others, and the little models went out across the country—Maine, Florida, Massachusetts. One went to Arizona to Captain Earl Loftfield, master of El Yunque.
John developed a theory about El Faro’s fate and wanted to share his thoughts with me after reading my article about El Faro published in Yankee magazine in October 2016.
When we first spoke on the phone, John kept referencing something called the “downflooding angle.” After hanging up, I looked up the term in the Code of Federal Regulations—a multivolume compilation of all U.S. rules covering every conceivable industry from education, to energy, to agriculture, to foreign relations, to shipping. The downflooding angle refers to how far you’d have to tip a boat in calm conditions for water to penetrate the boat’s first nonweathertight opening. In a simple sailboat, that would be the moment when the boat tips enough that its gunwale dips below the waterline. Water then floods into the hull.
I once capsized a small boat this way. On a very windy day, I accidentally exceeded the boat’s downflooding angle while executing a jibe. Wind caught my sail and pulled the mast toward the water. I’d always popped up before. But when the gunwale dipped below the water, the river came pouring in, pulling the boat into an unrecoverable roll. In an instant, I found myself clinging to its side, looking for a rescue launch. Once it tipped, there was literally nothing I could do but slide into the river and wait for help.
Now that I understood the term, it seemed like such a critical part of sailing that I assumed every ship’s officer would know his or her vessel’s downflooding angle.
So what did this have to do with El Faro?
One early April afternoon, I visited John and his wife, Dot, at their immaculate suburban Philadelphia house. After lunch, we cleared the kitchen table to make way for John’s forty-year-old blueprints. After years of working with power tools in resonant steel hulls, he can’t hear that well; he spoke for three hours in a nearly unbroken stream, walking me through every aspect of the ship that he thought might have come into play that October morning.
He’d been thinking about this for a long time and was excited to share.
“When you look at El Faro, you see the big blisters on the side, forward of the house,” he says. “There are three of them on each side. They stick out from the hull. And those were the ends of the ventilation shafts. I’ve never, ever seen that on another ship—air shafts external to the hull.”
Inside those blisters—metal boxes hanging off the ship—were big fans designed to blow air down into the deepest part of the vessel, or allow air to escape out of the vessel. If you didn’t ventilate the cargo areas, poisonous and flammable gases from the cars and trucks stowed below could build up, risking fire or an explosion.
John continued: “When you look at how the ship was designed, you say, oh my gosh. Because you see these big ducts through here?”—he points to some ovals on his blueprints, just below the deck line—“They go all the way to the bottom of the cargo hold. There are special rules that allow this vent to
be [only] this far above the waterline and still maintain watertight integrity. And that’s all calculated into what’s called the downflooding angle.”
In modern vessels, ventilation stacks rise up from the main deck. But the designers of the Ponce-class ships wanted to keep the decks as clear as possible. Otherwise, trucks and trailers could hit the stacks during loading and unloading. The designers were also working with less stringent codes.
So they hung the ventilation boxes off the side of the ship below the main deck line. Inside each of these boxes was a sizable duct that penetrated through the ship’s thick steel hull, then turned straight down, running deep along the inside wall of the ship and into the cavernous holds.
It was like having your nostrils somewhere near your belly button.
John picks up the flimsy paper model of El Faro’s midsection that he made a few months after the ship disappeared. He built it so that he could study the vessel’s inherent vulnerabilities, and it takes me a while to get oriented because it’s missing context—no bow, no stern—more like a fragile box with some dollhouse-size ramps and a few big holes in the sides.
As John points out various things to me, the model flops around in his large, rough shipbuilder’s hands. The paper floor of the second deck isn’t attached to the paper hull, adding to my confusion. But in moments of clarity, I see the ship, and I see how a serious list would position these unprotected openings dangerously close to the waterline. El Faro was an accident waiting to happen.
“So here we are with the ship that’s all intact and everything is fine as long as the ship is sailing,” John says. “But then all of a sudden, the downflooding angle comes into play. When? At what point in time?”
He rolls his model to the starboard side. And now I notice that he’s drawn a blue line indicating where a calm sea would be on the ship at a 15-degree list. The line—marking the point of no return, just like on my sailboat—wraps around the starboard side. Then I feel my gut tighten. The waterline goes right over the ventilation openings in the hull at the second deck.