The Drift Of Things

I found Art Allen a few miles inland from some uninviting Connecticut beach. He was in his mid-60s at the time, a scientist with a man-of-action feel to him. He wore a Coast Guard Search and Rescue polo, a massive Fenix 3 GPS watch, and he had this snow-white Hemingway beard. Six canoes hung from hooks inside his garage, mountain bikes leaned against the wall, and all looked as if they had a lot of miles on them. So did he.

For 35 years Art Allen had been the lone oceanographer inside the U.S. Coast Guard Office of Search and Rescue. Among other pursuits, he had mastered the art of finding things and people lost at sea. At any given moment, all sorts of objects are drifting in the ocean, a surprising number of them Americans. The Coast Guard plucks 10 people a day out of the ocean, on average. Another three die before they’re found.

“I’ve only thought about one problem in my life,” said Art, “which is how to improve Coast Guard search and rescue.”

He hadn’t set out in life to save people at sea.

“Science is driven by the love of the subject,” he said.

What he had always found beautiful, and fascinating, was water. He’d grown up on Lake Champlain, and even as a kid his idea of fun was to dig tunnels to drain snow ponds. At the University of Massachusetts in Amherst he designed his own major in aquatic science and engineering, then moved into a graduate program in physical oceanography at Dalhousie University.

The question at the start of his career, back in 1984, was where to apply himself. He’d seen an ad placed by the Coast Guard for a junior researcher, but didn’t really think of himself as a government guy.

“I thought I’d give it six months,” he said.

Just then the Coast Guard was trying to figure out how to improve its ability to spot objects on the ocean surface from its planes and helicopters. It was a little shocking how hard it was to see even a small boat, let alone a human being, from 1,000 feet; and if you flew over and didn’t see it, you might never look there again. To see better, there wasn’t much the Coast Guard wasn’t willing to try. Not long before Art arrived, for instance, they’d attempted to train pigeons, riding in cages attached to Coast Guard aircraft, to respond to any orange object in the ocean by pecking at an alarm. The pigeons seemed to have natural advantages over humans as spotters of objects lost at sea. Their vision was sharper. They never got bored or distracted. The pigeons didn’t miss a thing.

“The problem was that there are orange things that aren’t survivors and things not wearing orange that are survivors,” said Art. “The pigeons drove the pilots crazy.”

By the time Art arrived, the pigeons were gone, replaced by questions that Art did his best to answer. For example, the Coast Guard wanted to know the odds of a plane flying at 500 feet over some object actually spotting that object. Art threw stuff in the water and made people fly over it and try to see it.

The Coast Guard wanted him to find better ways to measure ocean currents and ocean winds. They needed a device that might better track what was happening to ocean currents at the last known position of some boat or person. Art took the lead on inventing a new buoy to do the job.

A Coast Guard commander looking for a guy lost at sea and presumed to be floating on a life raft made by the Elliot company realized that he didn’t really know what an Elliot life raft looked like, or how fast it might travel in relation to wind and current, compared to life rafts better known to the Coast Guard. He called Art, who called a facility in Essex, Connecticut, that certified life rafts and they sent him a brochure for one.

Human survivability

In June 2002, off the southern tip of Long Island, a fishing boat swamped in a storm and threw the four men on it into 60-degree waters. The men had been competing in a shark-fishing tournament when a storm had come through. There were two Mustang survival suits on board. Before they’d capsized, they’d sent a distress signal that had been picked up by a Coast Guard station in New Jersey, but the signal was fuzzy and the Coast Guard had no idea where they were. The search lasted four days.

A human being could survive in cold water for maybe 36 hours, even inside a Mustang suit, but the Mustang company told the families that anyone wearing the suit could last for eight days. The families implored the Coast Guard to keep looking long past the point the Coast Guard thought there was any reason to do it. Three men were never found. The body of the fourth was found a week later by a fishing boat 30 miles off the New Jersey coast.

When it was over, the people who had failed to find the men called Art with a question: Who’s right, us or the Mustang company? Art looked into the hypothermia models used by the Coast Guard and found problems. They made no allowance for the clothing a person might be wearing, for instance, or his body fat. They assumed the weather was constant throughout the search, and nights in the water the same as days in the water.

Art sought out scientists who had studied hypothermia and collected what was known on the subject. Even if they’d been wearing the survival suits, he concluded, the men almost certainly had been dead within two days. These new studies suggested to Art that the old Coast Guard models had been optimistic about the ability of human beings floating in ocean water to survive. Never mind hypothermia. A person could go only three days without water, and 62 hours without sleep, before he lost his ability to keep himself alive. But what struck Art Allen about the whole incident was that “no one really knew.”

People he worked with started treating him differently, turning to him for answers to questions he’d never before been asked. They had nowhere else to go. The biggest thing that no one knew, he decided, was how various objects drifted at sea. The ocean never stopped moving. Every object was pulled and pushed along by currents and winds. So if you wanted to know where an object might be that had been spotted, say, five hours ago 30 miles east of Cape Hatteras, you needed to know the winds and currents off Cape Hatteras over the previous five hours. But you also needed to know exactly what the object was and how it interacted with the forces of nature. “Leeway” was the technical term for the particular way a particular object moved in the ocean current. Even if they started in the same place, a disabled fishing trawler and a sea kayak might soon be many miles apart.

Art Allen set out to find what was known on the subject. Shockingly little, it turned out. The history of search-and-rescue at sea is mostly the story of people being neither searched for nor rescued. For most of human history, “lost at sea” meant gone for good.

“It really only started during World War II,” said Art. “There was not much looking for people at sea until we started losing pilots in the Pacific.”

Scouring SAR literature, he found that there was really only a single good study of leeway. A Coast Guard commander, W. E. Chapline, stationed in Hawaii in the late 1950s, had grown sufficiently weary of not finding people and had done tests on the few objects on which people lost in the South Pacific tended to float: a surfboard, a sampan, a small fishing boat. “Estimating the Drift of Distressed Small Craft” was published in 1960, in the Coast Guard Academy Alumni Association Bulletin. It was two-and-a-half pages, wholly original, and inspiring to Art. It contained new insight – like the fact that a lot of boats don’t make their leeway directly downwind.

“I’m reading this and I’m saying, ‘Right on, guy!” recalled Art.

There was no reason that Art Allen, Coast Guard oceanographer, could not study every object on which people might drift upon the sea and reduce to a mathematical equation how each of those objects moved through the water. If you knew where some object had been and when it had been there, you could predict more or less exactly where it was now. If you knew the currents and the winds – which the Coast Guard usually did – all you needed was leeway.

He’d stumbled into a small but possibly important field of inquiry and found no one there. Without anyone particularly noticing or caring, he gathered every object ever studied – the ones in Chapline’s paper, some stuff the Japanese had tested, and objects that had been tossed into the ocean by the Coast Guard and observed. To these Art added the 45 or so objects that he’d studied himself, usually after the Coast Guard had failed to find someone said to be adrift upon them. When he was finished, he had a list of 95 different objects: a Tulmar four-person life raft, a 12.5-meter Korean fishing vessel, a Japanese 13-person life raft, a sea kayak, a 100-gallon ice chest, a 65-foot sailboat, a windsurf board, a Cuban refugee raft with a sail, a Cuban refugee raft without a sail, an airplane evacuation slide/raft that Art had persuaded Delta Air Lines to lend him, and so on. Art’s list ultimately reduced itself to 63 classes of objects.

In 1999, Art published everything he knew in a 351-page treatise called “Review of Leeway,” which became required reading for anyone going through the Coast Guard’s National SAR School. It made Art Allen slightly famous in his small world. SAR people in other countries began calling him for help with specific problems.

How to apply it across the country?

When I got to the Coast Guard, their question was: How well do you see, and how do we make you see better?” he said. “My question was: Where do you look? It doesn’t matter how well you see. If you’re looking in the wrong place, you aren’t going to find it.”

The SAR people could now, in theory, use Art’s equations in their searches. But they didn’t have a simple computer program that did the work for them, so it was unclear just how the stuff he had learned was being applied. The truth was that, 15 years into his career, Art still didn’t know exactly what happened in the heat of a rescue, because he’d never been on the scene during a search. In May 2001, a commander in a field office in Portsmouth, Virginia, asked if he’d like to see what they did.

“That was the first time I got out of the office,” said Art. The idea was that Art would spend the first weekend in May with the so-called SAR operator, the person who coordinated the search and rescues in that district. The forecast was sunny, warm, and unthreatening. His host had said, “Come in the evening, ’cause that’s when things happen.”

Art had turned up at 4 in the afternoon. Not long after he sat down with the SAR guy, all hell broke loose, all calls for help, one after the other. One boat had run aground. Another had caught fire. Yet another had capsized, and several people had gone overboard and needed to be rescued from Chesapeake Bay. Across the water people were coming to grief.

“A dry cold front had come through and no one had seen it,” said Art.

The SAR operator was dispatching helicopters and cutters as fast as he could and saving one person after the next. He had only a crude computer tool and had to make a lot of calculations by hand. Art was impressed, but after six hours of drama he could see the guy tiring.

“And after all of this,” said Art, “someone calls at the end of the day and says, ‘We have an overdue sailboat.’”

The sailboat’s last-known position was the beach it had left from that morning. Its intended destination had been the Chesapeake Bay Bridge. It was 20 feet long, equipped with life vests. On board were a man and two women in their 40s and a 9-year-old girl. Just how weary the SAR guy had become revealed itself when he tried to use some satellite picture to zoom in on Chesapeake Bay, only to realize he was staring at the mouth of the Yangtze River in China.

Art watched this overwhelmed young man plug crude information into his crude computer tool. He had a fair description of these people’s plans, but his program didn’t allow him to input a voyage. Nor could he find good data on the winds over the water. He finally just pulled something from an anemometer at a local state park. He didn’t even have a good reading of Chesapeake tides. He had no drift equations for a swamped skiff, as Art hadn’t studied it. The equations the guy was using came from one study of drifting sailboats, made by W.E. Chapline 41 years before.

“I could see he couldn’t adequately plan the search,” said Art. “The tool he’d been given could not help him do what he needed to do.”

So the Coast Guard went looking for something without any real idea of where it was. Helicopters and the 87-foot cutter searched through the night and found nothing. Not until the following morning did the sailboat appear, upside down, a long way from where the Coast Guard had been searching. A 300-pound man was in the water beside it, alive, along with his heavyset wife. A 42-year-old mother and her 9-year-old daughter, both wearing life vests and hypothermic, were taken off the hull. A few hours later, at a local hospital, both were pronounced dead. Art had stayed late into the night and seen all this unfold in real time.

“I watched this happen,” he said. “These two were the same age as my wife and daughter,” said Art, and suddenly he was fighting back tears. One moment this 66-year-old man with the Hemingway beard was a dispassionate scientist; the next, he was a welter of emotion.

The envy of the world

Up until May 5, 2001, Art had been more scientist than engineer. On that day he saw that the Coast Guard – the country – needed him to be both. Every day there was more and more data that might be used to find people lost at sea that either wasn’t available or was hard to use. What was needed was a new computer tool, as simple to use as TurboTax, that instantly grabbed all the relevant data and turned it into a prediction. Four years later the Coast Guard had a prototype, soon to be the envy of the world of search-and-rescue. SAROPS, it was called – Search and Rescue Optimal Planning System.

Art hadn’t built it by himself, of course, but he’d taken the lead on most of it and made sure that research was now embedded in algorithms at the fingertips of every SAR person. They could enter a detailed description of the search – four officers, say, peering down from a C-150 flying at 1,000 feet – and SAROPS could calculate the probability of the officers’ having seen what they were looking for, so they could decide if it made sense to fly over the same patch of ocean again. They could enter the height and weight, outerwear, and other details about a person floating in the ocean and, along with the water temperature, figure out how long he had to live. They could enter the last known location of the object and, because Art had almost certainly studied its leeway, predict how it would move in relation to the winds and currents.

Often the Coast Guard was unsure exactly what it was looking for. Upright sailboat or overturned one? A disabled fishing trawler or five fishermen in the water? Now they could plug multiple objects into their tool and visualize several searches at once.

Art’s curious science had yielded information that people could act on. For instance, his investigation of the way various sailboats drifted. Chapline had shown that sailboats drifted at an angle to the wind, rather than directly with it; his study ended there. It was impossible to know the initial direction a sailboat might take, but the angle, in relation to the wind, was predictable. Art had shown how the angles, and thus the search areas, varied with the type of sailboat.

The U.S. Coast Guard rolled out its new search tool in early 2007. Art spent two days at each of the nine districts teaching people in the field how to use it, what it was, and what it wasn’t. The field people, for their part, couldn’t quite believe how much more quickly and accurately the new tool allowed them to figure out where in the ocean to look for whatever had gone missing.

The United States had always been a SAR leader; our country has made more of a priority than any other of saving its citizens at sea. If you were lost at sea, there was never much of a question which country you wanted to have looking for you. Now the United States was in a class by itself. Even as Art ran around the country unveiling the new tool, stuff happened that astonished SAR people. For instance, less than an hour past midnight on March 16, 2007, an obese 35-year-old man who’d had too much to drink fell off the balcony of his cabin on a Carnival Cruise ship and into the Atlantic.

“There’s really kind of a distressing number of people who fall off cruise ships,” says Art.

In the long run, on land, fat will kill you, but at sea it can save your life, and not just because it keeps you warm.

“Everyone floats,” explained Art, “but the fatter you are, the farther your mouth is from the waterline.”

Someone on the cruise ship had seen the man go into the dark water. The captain notified the Coast Guard, so they knew roughly where and when the man had splashed down. Still. Spotting a person without a life jacket in the ocean was, as Art liked to put it, “like looking for a soccer ball in Connecticut.” But now the Coast Guard knew the currents and how the man’s body would move in relation to them: his leeway.

Interestingly, before Art came along, the assumption was that a person in the water had no leeway. A body was assumed to simply drift with the current. Art had proved that wasn’t true. By the time the man had fallen off the ship, Art had studied five cases: a person with a life jacket, a person without a life jacket, a person in a scuba suit, a person in a survival suit, and a dead person.

The people running the rescue plugged in Art’s equations for a person with no life jacket. And on the day of the incident, the following item appeared on a CruiseJunkie.com blog post: “A 53-year-old man was rescued approximately eight hours after jumping or falling overboard from the ship when it was 30 miles east of Fort Lauderdale. A witness said that the man, who was intoxicated, ran through a window and fell 60 feet into the ocean – it is not clear whether the window was open at the time.”

They’d found the man floating 15 miles from his point of entry. Had he gone overboard at any previous moment in human history, up to about two months earlier, he likely never would have been found. Now the pictures of him flopping down naked but alive upon the deck of the Coast Guard cutter were on the front page of the local newspapers in Florida. Newspapers mostly told the story of the guy’s miraculous survival and asked the question: Exactly how was he found?

The answer, at least to the people using SAROPS, was obvious: Art Allen. Without his work that man would have stayed lost. The response of the people in the field to the new tool was almost shockingly enthusiastic.

“Everywhere I went people said, ‘Thank you. This is going to save lives.’”

Art’s mind hadn’t quite moved on from the lives they hadn’t saved. One evening, not long after he had the first working version of the new tool, he sat down and re-created the search for the sailboat that had gone missing in the Chesapeake back in 2001. The new tool automatically pulled in better wind data, as well as the currents and tides, from the day the boat was lost. It knew when the front swept across the bay so could estimate when, and where, the boat capsized and tossed the child and her mother into the chilly waters. It used Art’s equations for a drifting sailboat. Up on the screen popped the most likely track the boat had taken.

“It took you right to where they’d been,” said Art.

Deferring retirement to save lives

Every weekday for 35 years, Art had made the same drive up the Connecticut coast to his office inside the Coast Guard building. Along with a bunch of people in his building, during the government shutdowns of 2018 to 2019, Art had been designated as inessential and sent home without pay. For 35 days it was illegal for him to lift a finger during a Coast Guard search and rescue.

“It was the longest time I’d ever gone without working,” he said.

In the fall of 2016 Art had alerted the Coast Guard of his intention to retire in 18 months. At the end of 2018, 18% of the civilians employed by the United States government became eligible for retirement, and you had to wonder how many Art Allens were walking out the door and not being replaced. He’d delayed his own retirement simply because there wasn’t anyone in SAR who knew what he knew.

It’s curious how knowledge is at once so hard to create and so easily taken for granted. Just before the president had shut down the government, for instance, the Miami district had been looking for a missing scuba diver off the Florida coast. Two divers, one in his 50s, the other in his 20s, had emerged from a dive to find their boat and driver specks in the distance, and strong currents pulling them even farther apart. They were 18 miles from shore. The younger one decided to swim toward the boat and get help. On the way, he was rescued by the Coast Guard. The sun set. The other diver was alone in the dark, drifting, as described by one of Art’s equations.

Art got the call from the SAR people that night, after they’d searched the area SAROPS told them to search and found nothing. A thought occurred to Art that hadn’t occurred to the rescue squad: The guy was trying to swim ashore, 18 miles away! Art said, Broaden the search. Look closer to shore. The Coast Guard found the scuba diver 9 miles from shore.

The episode raised a question: Could you better predict the movements of the people who decided to swim for it? Art talked a friend into jumping into Long Island Sound in various conditions: with and without flippers, with and without a wet suit. Art thought the future of his research would be to explore not just the leeway of floating objects but the movement of willful ones. It was one thing to predict the drift of a sailboat that had lost its mast, another to predict the path of one still sailing.

He tried to stay as long as he could

Art took two-and-a-half years to retire. Before his departure, the Coast Guard hadn’t hired his replacement or given him some junior scientist to train. His immediate superior interviewed him on video for the Library of Congress and asked what the hardest part of his job had been. “Retiring,” he’d said, and began to weep. He’d done what he’d done without asking for much. He hadn’t expected the attention of others, outside his small circle of SAR people. It was nice that Taiwan’s Coast Guard wrote poems about him. But that sort of thing never happened in the United States. The Partnership for Public Service shocked him by nominating him for a Sammie Award. [Samuel J. Heyman Service to America medals honor the accomplishments of outstanding American public servants who’ve made a tremendous impact on our lives.]

There’d been a moment, a few years earlier, that captured the spirit of Art Allen’s relationship to the society he’d tried to save. He’d flown out to Long Beach, California, to help the Coast Guard people there upgrade their SAR tool. Purely by accident, he’d arrived the same day a ceremony was being held to honor the heroes of a recent rescue. A few months earlier, a Los Angeles man had fallen off the back of his brother’s fishing boat without anyone noticing. He’d floated in the Pacific Ocean for seven hours. The Coast Guard had plucked him from the water in the middle of the night. On the day of Art Allen’s visit, the guy had returned to the Long Beach Coast Guard station to thank his rescuers. A bunch of TV news reporters were there to witness the moment as the guy broke down and confessed that the experience of floating for hours in the darkness, and then by some miracle being saved, had left him a changed man.

Art stood off to one side respectfully and watched as the television cameras turned their attention from the man to the brave young Coast Guard search patrol that had saved him, as they stepped forward to receive their medals. Just then an older Coast Guard guy who had worked with Art Allen for many years leaned over and whispered to him, but to him alone: “Nice job.”