Science, Tech & Environment

For more accurate weather forecasts, follow the fish


Marine biologist Jerry Ault tracks tarpon, seen here being observed by a diver, with satellite tags, something his colleague Nick Shay, a meteorologist and oceanographer, found surprisingly useful.

Jerry Ault and Nick Shay were golfing on Key Biscayne when they had an epiphany.

It was a humid Saturday in 2011, and the men, colleagues at the University of Miami’s Rosenstiel School of Marine and Atmospheric Sciences, had each had a few beers. Somewhere around the 9th hole, they started talking about the 26° C isotherm.

Every day, Shay, a professor of meteorology and physical oceanography, was posting the depth and location of the 26° C isotherm — a zone of warm water equal to 78.8° F — on his website. Meteorologists find the data useful, since hurricanes are known to intensify after passing over pockets of warm water in the ocean. The deeper the warm water goes, the more intense a storm becomes. During Hurricane Katrina, the 26° C isotherm extended far below 100 feet. What was a mild Category 1 storm in South Florida rapidly intensified to a Category 5 storm in the Gulf of Mexico, eventually slamming into Louisiana and causing $81 billion in property damage and 1,833 deaths.

As it happened, Ault, a professor of marine biology and fisheries, was using satellite tags to track tarpon, the legendary game fish that generates $6 billion for the U.S sport fishing industry. Based on data from the tags, he’d discovered that tarpon travel along the 26° C isotherm during their annual migration through the Gulf of Mexico and the Caribbean Sea.

The problem with Shay’s satellite model was that it had trouble defining the boundaries of the isotherm. But according to Ault, the boundaries were where tarpon liked to feed. By using Ault’s fish, they could essentially fill in the gaps of Shay’s temperature and depth data. The fish, in other words, could help meteorologists make more accurate hurricane forecasts.

“It was one of those mutual jumps in scientific inquiry where you both go, ‘Holy s***!’ ” Ault recalled inside his cluttered office in Miami, where a mounted 220-pound tarpon looms on the wall. “Tagged tarpon are basically living observational platforms. In addition to telling us where they go, they can tell us crucial things about the ocean environment.”

He paused to savor the idea. “I’m a marine biologist. I never would have imagined that a network of animals attached to sensors could inform predictions of large-scale tropical storms,” he added

The First Animal Oceanographers

Biologists have been tracking birds, fish and other marine and land animals with satellite transmitters for more than three decades. As a result, they have received unprecedented insights into the mysteries of migration and other animal behavior. But recently, scientists in other fields have discovered the value of this animal-derived data for their own research. Nowhere is this more visible than among those who study the ocean and its interactions with the climate.

What has emerged is a global movement of marine scientists interested in using “animals as oceanographers,” as they often put it. On a steadily warming planet, these researchers believe that data provided by animals can help them predict and monitor hurricanes, the melting of Arctic sea ice, the formation of ocean dead zones and other climate-related issues. As the technology of satellite tags improves, marine animals may become one of science’s biggest allies in monitoring and combating the effects of climate change.

Some trace the origins of the animals-as-oceanographers era to the First Symposium on Biologging Science in Tokyo in 2003. It was here that Michael Fedak, a biology professor at the University of St. Andrews in Scotland, described what he called the first large-scale use of “animal platforms” to collect ocean data: the Southern Elephant Seals as Oceanographic Samplers project, or SEaOS.


Southern elephant seals were the first animals recruited for a large-scale oceanography study.

SEaOS deployed 80 southern elephant seals, each with a single ocean-profiling tag glued to their forehead, into remote areas of the southern ocean. Whenever the seals surfaced, data containing conductivity, temperature and depth profiles, known collectively as CTD data, as well as information about location and dive behavior, were transmitted to passing satellites. Much of the CTD data the seals returned came from areas where no oceanographic sampling had ever occurred. These data were later incorporated into forecasting models that map sea-ice formation in the southern ocean, allowing oceanographers to better predict and measure how fast the ice was melting. Seals, in other words, were filling in the blind spots on oceanographer’s maps. And they were doing it relatively cheaply — at least compared to the cost of ships or Argo floats, the international network of probes that transmit real-time measurements of the Earth’s oceans.

Fedak hoped oceanographers would rush up to him after his speech and propose a series of long-term collaborations. Instead, they stared at him in silence. But he was used to such reactions. At a conference in London a few years prior, he’d presented data gathered from a satellite tag he’d glued to the back of a beluga whale. All winter long, the whale had sent valuable CTD data from beneath the pack ice 1,000 miles north of Norway. During his lecture, Fedak noted that he’d recently tagged a hooded seal in the North Atlantic. Considering that oceanographers were hungry for data from that region, perhaps they’d like to use his hooded seal as a biological ocean-data collector?

“They bloody laughed at me,” said Fedak, a New Jersey native who has picked up some Britishisms during his 30 years at St. Andrews. “One oceanographer actually stood up in the back and yelled, ‘You just want our money!’ ”

Fedak spent the next decade advocating for animal platforms as a cost-effective and complementary way of obtaining oceanographic data. The idea gradually caught on. By the late-2000s, he’d received funding from the Sloan Foundation and the National Oceanographic Partnership Program, which supports oceanographic research and exploration.

Today, Fedak’s prediction that marine mammals could provide a rich new source of ocean data has been realized. As he noted in a recent paper, “The Impact of Animal Platforms on Polar Observation,” marine mammals have now delivered about 70 percent of all oceanographic profiles available for the southern ocean south of 60 degrees. That data, once incorporated into global and regional models, has resulted in dozens of publications on physical ocean processes.

Fedak stresses he didn't invent the idea of using animals as oceanographers.

“The U.S. Navy had wanted to use marine mammals for this purpose as early as 1972,” he said. In fact, he sees the Navy Marine Mammal Program — a controversial initiative launched in the 1960s that taught dolphins and sea lions how to locate mines and protect harbors — as a kind of military precursor to the oceanographic community’s interest in animal use.

It’s fitting, then, that Fedak is now working with the Navy. In collaboration with the Office of Naval Research, he and the members of St. Andrews’s Sea Mammal Research Unit are developing tags that last longer and provide higher resolution depth and temperature profiles, down to .0001° C. (By comparison, Ault’s tags measure temperature to .01° C.) “The Navy is as interested in this stuff as we are,” he said. “Elephant seals don’t respect international borders, after all, so you don’t have to ask Russia’s permission to use one — at least not yet.”

An Arsenal of Animals

In the wake of SEaOS, other marine mammals began proving useful to oceanographers. In 2010, for example, researchers at the University of Washington’s Polar Science Center published a paper in the Journal of Geophysical Research titled “Narwhals document continued warming of southern Baffin Bay.” As part of the study, 14 narwhals — medium-sized whales with a single, unicorn-like tusk — were equipped with pop-up satellite tags and released off the coast of Greenland. They revealed that the water in Baffin Bay was, on average, nearly a degree Celsius warmer than the existing climatology data claimed. The narwhal data also suggested that the thickness of the winter surface isothermal layer, a layer of constant temperature, was 50 to 80 meters thinner than previously thought. NOAA, which sponsored the study, reported that data of this kind is “critical for understanding the impacts of a changing Arctic.”

In the early days of satellite tracking, marine biologists used marine mammals for two main reasons: their large bodies accommodated the sizable tags, and they surface fairly often, allowing data stored in the tags on their fins and foreheads to be transmitted to satellites overhead. Monty Priede, head of the Oceanlab at the University of Aberdeen in Scotland, became the first researcher to use a satellite tag to track a fish in 1982. The fish happened to be a basking shark, the second-largest fish species in the world. Priede attached the bulky tag to it via harpoon. This so angered the fish that it swatted the researcher’s boat with its tail, nearly capsizing it.


Sensors attached to narwhals helped researchers more accurately plot the temperature of Baffin Bay.

The recent development of smaller, more sophisticated tags, however, has made it easier to gather oceanographic data with smaller ocean-dwelling animals. The Tagging of Pacific Predators Project — an ambitious collaboration among 75 scientists from five nations that concluded in 2011—attached 4,306 tags to 23 species of marine animals, including fishes, seabirds, turtles and squid. Last year, researchers at the University of Laval in Quebec City, seeking to understand how American eels make it to the Sargasso Sea in the mid-Atlantic, attached satellite tags to the fishes’ slender, snake-like bodies. Carl Wunsch, an emeritus professor of physical oceanography at MIT, has said he envisions a future “in which ever more species are used to obtain a truly global observation system.”

International partnerships between biologists and oceanographers — facilitated, in a sense, by marine animals — have already led to some astonishing discoveries. The work of Eric Prince, a NOAA fisheries biologist, and Lothar Stramma, a German oceanographer, is one example. In the mid-2000s, Prince started putting GPS tags on blue marlin to figure out where they went. One of the fish, he was stunned to learn, swam from the coast of Delaware to the coast of Mauritius, in the Indian Ocean.

“If you want an animal to tell you about the ocean, the blue marlin is a good one,” Prince said. And yet he couldn’t understand why the marlins dove below 500 meters in parts of the Caribbean, when they rarely dipped below 100 meters in the eastern Atlantic.

By chance, Prince came across a paper by Stramma, who’d been studying the expansion of oxygen minimum zones—areas of low dissolved oxygen that make it difficult for marine life to thrive — in the northeastern Atlantic. Blue marlins are high-energy fish. They require a lot of dissolved oxygen in the water. The two men guessed the fish were encountering low oxygen around 100 meters, and refusing to dip below.

To test the hypothesis, Prince deployed 47 blue marlin equipped with pop-up satellite tags into the eastern Atlantic. By comparing their movements with Stramma’s oceanographic data, they found that blue marlin were, in effect, being pushed toward the surface by the low-oxygen areas below. They concluded that oxygen minimum zones in this part of the ocean, likely caused by climate change, have reduced the habitat of billfish and tuna, making the fish increasingly vulnerable to overfishing and threatening the sustainability of eastern Atlantic fisheries. The resulting paper, published in Nature Climate Change in 2012, was hailed by one expert as a “masterpiece of fisheries oceanography.”

Fishing for Hurricanes

To tag an animal, of course, you first have to catch it. Ault and his colleague Jiangang Luo, also a professor of marine biology and fisheries at the University of Miami, have been catching tarpon for scientific purposes since 2001. They often take part in fishing tournaments during the tarpon migration through the Florida Keys, which lasts from mid-February to July. For Ault, who grew up fishing off the coast of San Diego, this is a dream come true. “I first got interested in fisheries management because I thought it’d make me a better fisherman,” he said. “I love to fish!”

After hooking a tarpon on a line, the men angle it beside the boat and measure its length and girth to determine its weight. If it weighs more than 70 pounds, it’s big enough to tag. They then remove a single scale from the fish and insert a stainless steel spear into its musculature, anchoring the tag in place. In his office, Ault demonstrated the basic technique on his wall-mounted tarpon.

For years, Ault and Luo have gathered location and CTD data — the same conductivity, temperature and depth data collected by elephant seals in the SEaOS program — using cigar-sized devices, known as PAT tags, that record for a few months and then separate from the fish and float to the surface. At that point, Ault’s team retrieves the tag and transfers the data it contains onto their computers. More recently, the men have begun experimenting with SPOT tags, short for “smart position or temperature transmitting.” SPOT tags can track a fish’s position with greater accuracy and in real-time. “They’re like crack for scientists,” Luo said. “You get instant gratification!”


Jerry Ault holds up a SPOT tag in his office.

The device’s GPS sensor is incredibly accurate, a fact that was reinforced one day this summer. While monitoring the signal’s location, a member of Ault’s team noticed that it appeared to be traveling along Route 98. It finally came to rest in Pensecola. Someone had evidently found the tag on the beach and brought it home.

Ault dispatched two of his lab technicians to recover it. “They pinpointed the signal to some guy’s house,” Ault recalled with glee. “One of our tech guys knocked on his door and asked for it back. I’m sure he thought we were from the FBI.” The man, as it happened, had placed the tag in a tree in his backyard. The technician found it hanging beside blown glass lanterns and empty bottles of rum.

The potential of these tags held serious appeal for Shay as an oceanographer. “A single PAT tag on a tarpon can provide 50,000 conductivity, temperature, and depth profiles per year,” Shay said. “If you tag 20 fish a year, you’d get a million CTD profiles. Integrating that many data points with our satellite framework could significantly improve the way we forecast hurricane intensity.”

Shay sees tagged tarpon as a welcome addition to the hurricane researcher’s toolkit. Today, before a hurricane arrives, he occasionally flies into its projected pathway in a P-3 Orion airplane with NOAA’s Hurricane Hunters. From the air, he deploys dozens of ocean profilers through a sonabuoy tube in the fuselage. The profilers return CTD data and descend as far as 4,500 feet beneath the surface. But the range of deployment is limited to around 20 kilometers, Shay says, whereas an isotherm can stretch for 200 kilometers. And the profilers, at $1,400 each, aren't retrievable.

Undersea gliders are another popular device. Autonomous, battery-powered vehicles that resemble torpedoes with wings, gliders move in a sine wave pattern through the ocean, transmitting real-time data for up to six months at a time. In the strong currents off the Gulf Coast, however, gliders may be less effective. “The Gulf current moves at around 6 or 7 miles per hour, whereas gliders move at about half a knot, or .6 miles per hour,” Shay said. “Tarpon, on the other hand, travel at sustained speeds of 35 miles per hour. They can go up to 50. They’re not going to get washed out to sea by the Gulf Stream. And at $4,000 per tag, they’re a lot less expensive than a $200,000 glider.”

Fish Forward

Using fish as oceanographers has its own limitations. You can program an undersea glider to cover certain areas, but you can’t tell a fish where to swim. Nor can you guarantee that a fish will be in the 26° C isotherm before a storm. As Carl Wunsch at MIT notes, “Most fish don’t cover an entire water column in the ocean, which limits the amount of sampling that can occur.” For this reason, Ault and Shay have proposed employing blue marlin and tiger sharks — both coastal game fish that feed on tarpon, and thus enjoy the 26° C isotherm as well — for data-gathering purposes.

Ault and Shay presented a poster of their preliminary research to the 2012 Ocean Sciences Meeting in Salt Lake City. “A clear connection is emerging between the behaviors of fish and physical ocean processes,” the summary read, in bold-faced type. The positive response at the conference persuaded them to submit a grant proposal to a federal agency, though they declined to say to say which, for $600,000 over three years. The proposal accounted for the cost of tags, data processing, boat time and salaries. The agency turned them down.

But Ault and Shay are undeterred. “We faced the same resistance to airborne oceanography,” Shay said of the now-common practice of dropping profilers into the path of storms. “Ultimately, this kind of research helps John and Jane Q. Public, so they can protect their homes and belongings and get out of harm’s way. When you consider the $20 billion cost in damages of a landfalling hurricane, $600,000 over three years is not that much.”

Ault currently receives substantial funding from through the Bonefish and Tarpon Trust, a non-profit dedicated to conserving bonefish, tarpon and permit fish. “In Florida, the sport fishing industry rivals the citrus industry in terms of economic impact and job creations,” Ault said. “Tarpon is a big deal down here.” Still, he and Shay are applying for support from the Office of Naval Research and the U.S. Integrated Ocean Observing System, a division of NOAA. According to Ault, both groups understand the value of animal-borne sensors in collecting data on a variety of physical and biological ocean processes. “We’re holding hands and hoping for a cash cow,” he said.

Since their 2011 golf outing, the idea of using fish to refine hurricane intensity forecasts has taken on quest-like proportions for both men. “I could have stayed in my secular world, but the potential impact of this fish data is huge,” Ault said. “By letting these fish talk to us even more deeply, we can eventually reduce human suffering caused by catastrophic storm encounters around the world. That’s the holy grail.”

Ideally, Ault would like to see 1,000 tagged fish of various species swimming in the Gulf of Mexico, mapping the entire ocean column for meteorologists in real-time. Given the rapid advancements of the technology, satellite tags the size of ballpoint pens are only a few years away, opening the field for a wider category of fish.

In November, Ault and Shay plan to present their findings at the 5th International Billfish Symposium in Taiwan. They’re also considering a trip to St. Andrews University to confer with the Sea Mammal Research Unit, whose advanced sensors might be applied to fish. “We’re going to bring our golf clubs when we go,” Shay said. “The courses in Scotland are supposed to be unbelievable.”

This story by Jed Lipinski was originally published on NOVA Next. Click for more from NOVA Next.