TRAVERSE CITY, Mich. — For more than a decade, people walking along Great Lakes beaches have come upon a heartrending sight: dozens, or even hundreds, of dead loons, gulls and other waterfowl — victims of food poisoning that paralyzed their muscles and eventually caused them to drown.

Scientists have long known the primary cause is Type E botulism, which the U.S. Geological Survey says may have killed 100,000 birds in the region since 2000. They have ideas, but no proof, about how the toxin works its way up the food chain.

Now, using time-tested methods and new technologies, they’re coming closer to solving the mystery — a crucial step toward determining whether anything can be done to prevent future die-offs.

Florida Atlantic University recently reported progress in a first-of-its-kind effort to determine the paths of birds that washed onto beaches after dying in open water. Experts with the university’s Institute for Ocean Systems Engineering placed stuffed bird carcasses into a laboratory tank and took water resistance measurements. The information will be combined with current and wind data in computer models that attempt to retrace the birds’ floating routes.

Meanwhile, several USGS labs are studying waterfowl distribution and sampling sediments collected from Great Lakes bottomlands, hoping to pinpoint where the toxin is produced. Initial findings suggest loons and other species that plunge into the water to catch fish may be getting infected at deeper levels than previously thought.

“It’s kind of like a detective story,” said David Blehert, a microbiologist with the USGS National Wildlife Health Center in Madison, Wis. “You find a body somewhere. You want to find out where the incident took place. You look for clues on the body, you find a piece of hair, a piece of fiber, and trace it back to the location and hopefully find your culprit.”

Scientists first documented a Great Lakes Type E botulism outbreak in 1963. But they’ve become more frequent and intense since 2000. Some areas are hot spots, such as Sleeping Bear Dunes National Lakeshore in northern Michigan, where 600 dead loons washed ashore in 2012. The previous year, about 6,000 bird carcasses were beached along Lake Huron’s Georgian Bay.

Research into where, why and how the die-offs happen has picked up in recent years, supported by more than $2 million in grants through the Great Lakes Restoration Initiative, an Obama administration program designed to deal with some of the region’s most pressing ecological threats.

“We haven’t got there yet, but we’re getting closer,” said Stephen Riley, a fishery biologist with the USGS Great Lakes Science Center in Ann Arbor.

What scientists consider the most plausible theory involves several environmental villains, including invasive species, climate change and nutrient runoff from farms and sewers.

Zebra and quagga mussels, ferried to the lakes from Europe in cargo ship ballast tanks in the 1980s, filter the water, allowing sunlight to penetrate deeper and stimulate growth of a green algae known as cladophora. Phosphorus runoff from land and warmer water temperatures promote cladophora.

As thick mats of the algae die, they sink to the bottom and decompose, which sucks up oxygen — an ideal condition for Type E botulism bacteria. Invertebrates such as fly larvae and worms consume the bacteria and in turn are eaten by fish including the round goby, another invader that’s popular prey for waterfowl.

If correct, the theory explains why so many diving birds such as loons, cormorants and merganser ducks are dying of botulism. They go far beneath the surface to gobble fish — up to 150 feet deep in the case of loons, said Kevin Kenow, a wildlife biologist with the USGS’s Upper Midwest Environmental Sciences Center in La Crosse, Wis. Kenow has fitted them with radio tags to record their movements.

He also conducts aerial surveys over Lake Michigan to observe bird distribution. During those flights, he saw bird carcasses onshore and got the idea of developing a computer model that could simulate the path they took. Florida Atlantic won the contract to produce the water resistance measurements after doing similar work predicting drift patterns of floating items in oceans.

The experiments thus far have been conducted in calm water, said Karl von Ellenrieder, an associate professor of ocean and mechanical engineering. The next step: creating waves in the tank and taking new measurements to further refine the data.

If researchers can nail down where and how the poisonings are happening, they could look for ways to prevent the toxin from being produced or to keep birds away from danger zones. Regulators could step up efforts to reduce phosphorus runoff near botulism hot spots, Blehert said. Attempts might be made to remove rotting cladophora from the water, although Riley admitted that might just spread the poisons more widely.

“There may be very little we can actually do,” he said.