Tracking American eels on the open sea to crack the mystery of their migration

It’s been one of biology’s greatest enigmas. How do American eels travel many hundreds of miles from the rivers and estuaries they live in as juveniles to their open-ocean spawning grounds?

Their rearing habitat stretches from Southern Greenland to the Gulf of Mexico and the Caribbean Sea, but somehow they make it all the way to the Sargasso Sea, south of Bermuda, to reproduce. It’s quite an open-sea journey for a fish that ranges in size from just 20 to 60 inches long (50-150 cm) and only 2 to 9 pounds (1-4 kg). Female eels from Canada’s St Lawrence river might be up to 20 years old when they migrate south on a trip they’ll make only once; they die after spawning.

Even a century ago, people had identified eel larvae in the Sargasso Sea. So their spawning areas have long been inferred. But no adult eel has ever been observed migrating in the open sea or on the presumed spawning grounds. We’ve had no idea about the routes or modes of navigation they must use to complete one of the animal kingdom’s most spectacular migrations.

The American eel in the northern and central portions of its geographic range is declining in both overall abundance and in the number of young fish that successfully make it to the juvenile freshwater habitat. So it’s now critical to gain a complete understanding of the species life history.

To throw light on this enduring mystery, we tracked the movements of adult American eels with satellite tags, allowing us to recreate their migratory routes. This is the first observation of American eels migrating in the open ocean to their spawning grounds and represents an important step forward in understanding the eels' migratory trajectories and the orientation mechanisms.

Operating on an eel to insert the satellite tag.Shiliang Shan, CC BY-ND

Tagging our slippery targets

Our study subjects were 38 wild-caught eels equipped with satellite tags, 28 of which successfully transmitted data to satellites. (The tags don’t harm the animals, but do increase drag forces while they’re swimming.) We released the fish along the coast of Nova Scotia, Canada. The tags continuously recorded water temperature, depth, and light levels; we programmed them to detach from the eels during migration, float to the surface and transmit their data – including location – to orbiting satellites.

Releasing the tagged eels off Nova Scotia.

Based on the tags' dispatches, we reconstructed the migratory routes of individual eels, one of which migrated about 1,500 miles (2,400 km) over 45 days to the northern limit of the spawning site in the Sargasso Sea.

What the tags told us

We identified two distinct migratory phases. Following their release over the Scotian Shelf, all eels immediately headed south and slightly to the east toward the edge of the continental shelf. During this first stage of the migration, eels generally exhibited daily vertical migrations: occupying shallow waters at night (about 50m deep) and bottom waters during the day (maximum of 240m).

During this stage, the eels appear to rely on gradients and fronts associated with temperature and salinity. Both of these physical variables increase from the coast to open waters and may have guided eels to the edge of the continental shelf away from the coast immediately after their release. The eels then moved mainly eastward along the edge of the Scotian shelf, departing the shelf at the exit of the Laurentian channel where the waters of the Gulf of St Lawrence flow into the Atlantic Ocean.

Reconstructed route of tagged eel #28, the only one that made it all the way to the edge of the Sargasso Sea after release off Nova Scotia.Béguer-Pon, et al, Nature Communications, CC BY-ND

The second phase of the marine migration occurs in deep (more than 2,000 meters in depth), saline oceanic waters. The eel, tracked all the way to the Sargasso Sea, exhibited a relatively sudden change in direction, heading south to the northern limit of the spawning site in a generally straight line from the area adjacent to the exit of the Laurentian Channel. It performed marked daily vertical migrations, with an average depth of 140 meters at night and 620 meters during the day (with a maximum depth of 700 meters).

We don’t know the orientation and navigation cues involved in this phase. Gradients in salinity and temperature occurring from the continental shelf to the southern edge of the Gulf Stream might play a role. But the horizontal gradients of salinity and temperature in the ocean are weak.

And these eels have never before experienced this migratory trajectory. Though they hatch in the Sargasso Sea, the larvae are shaped like leaves and drift north with the currents on the Gulf Stream before eventually peeling off to search for estuaries and fresh waters, as far as Greenland. So, although they’ve been to the Sargasso Sea before, it wasn’t as adults and they certainly weren’t actively navigating their route.

Considering the speed and directionality of the last portion of the eels' track to the Sargasso Sea, it seems likely that they use some kind of inherited bidimensional map based on the earth’s geomagnetic field. The sensitivity of eels to the geomagnetic field has been known for a long time, and previous studies have shown that they possess a magnetic compass that they can use for orientation. We suspect eels do possess a magnetic map and true navigation abilities, but this remains to be conclusively demonstrated.

Free-swimming tagged eel.José Benchetrit, CC BY-ND

Documented proof that raises more questions

The extensive vertical migrations we observed in the North Atlantic Ocean are part of the behavioral repertoire of many eel species. The reason for the daily nature of these migrations is associated with a compromise between two biological functions: the use of deep, dark waters to avoid eel predators and the need to occupy warmer, shallow waters to improve metabolic efficiency during the long migration.

There is no doubt that eels must deal with formidable predators during their migration. Our research team has previously reported heavy predation on eels by Porbeagle sharks in the Gulf of St Lawrence and, to a lesser degree, in the North Atlantic Ocean. Such predation is readily detected by satellite tags because Porbeagle sharks have warm guts; when our tagged eels get eaten, the tags record the warm temperature inside these predators – a noticeable jump over the ambient ocean temperature. Bluefin tuna, another warm-gutted fish, is also an eel predator.

For the moment, we can only speculate about the orientation mechanisms used by eels. Is the Olympic-style performance of one eel typical of all eels entering deep oceanic waters? Do variations in the earth’s magnetic field influence the migratory trajectories of eels? How does eel behavior change once the spawning grounds have been attained and spawning begins?

The American eel still has many secrets to reveal. We’re continuing to satellite-tag open-ocean-migrating eels to unravel more of their mysteries.

Julian Dodson receives funding from the Natural Sciences and Research Council of Canada and the Canadian Foundation for Innovation.

Martin Castonguay receives funding from the Natural Sciences and Research Council of Canada and the Canadian Foundation for Innovation.