Scientists researching the severe decline in Alaska's Steller sea lion population have a fingered a new possible suspect for the drop: Pacific sleeper sharks, a species previously thought of as a scavenger and fish-eater.

A study led by Oregon State University researcher Markus Horning concludes that three juvenile sea lions tracked with implanted data transmitters were killed by a cold-blooded predator, and the likely culprits were sleeper sharks, which can grow to 20 feet long.

Transmitters surgically implanted in the sea lions that died indicated a sudden change in temperature of warm to cold but no exposure to water or light, indicating they were eaten by something cold-blooded.

The recorded temperatures ruled out a kill by a killer whale, Horning said by phone. "It would have been at the same temperature as the Steller sea lion," he said.

Likewise, known sea lion predators such as great white sharks and salmon sharks have body types that would have reflected higher temperatures, he said, if they had consumed a sea lion.

The western population of Steller sea lions, which live in Alaska from Cordova to the Aleutian Islands, has declined to about 20 percent of the levels they were before 1975.

The cause is unknown. Among the theories are a change in fish population because of global warming or competition from commercial fisherman. The theory of mortality from predators, Horning said, has "fallen by the wayside."

Federal fisheries managers in late 2010 as a precaution restricted commercial fishing in the western Aleutians to limit competition for sea lions. The state of Alaska and commercial fishing interests unsuccessfully sued to overturn the restrictions, which are under review.

Julie Speegle, spokeswoman for the National Marine Fisheries Service, said Wednesday that the agency next month will publish revised protection measures that will lighten restrictions for fishing of Alaska pollock, Pacific cod and Atka mackerel in some areas and tighten others to spatially and temporally disperse fishing and protect sea lions.

Horning and colleagues from 2005 to 2011 surgically implanted "life-history transmitters" into the abdomens of 36 juvenile sea lions released in Resurrection Bay off Seward. The tags record temperature, light and other information. The buoyant tags do not transmit information until the animals die and tags float to the surface.

Researchers recorded 17 deaths. Three tags recorded abrupt temperature drops without detecting light and while the tags were still surrounded by tissue.

Reviewers of the study, Horning said, questioned whether cold-blooded sharks may have ingested the transmitters by feeding off sea lion carcasses. Unless a shark was nearby to start eating when a sea lion died of another cause, Horning said, recorded temperatures should have shown a gradual temperature drop in the carcass or some indication of illness, such as a fever spike.

Horning acknowledges that the observations do not constitute proof that sleeper sharks are killing live Steller sea lions — only that they must be considered a possible source of mortality.

The number of sharks killed in Alaska by fishermen seeking other species ranges from 3,000 to 15,000 annually. If sleeper sharks are found to kill sea lions, and commercial fishing continues to be limited, it could mean more sleeper sharks available to kill juvenile sea lions, he said. A management decision to protect sea lions might be having the opposite effect, he said.

"Their numbers may go up," he said of sleeper sharks. "That may be future trouble for sea lions."

The study was published in the quarterly journal Fishery Bulletin.