Ship traffic, seismic tests and sonar pings can make navigating the seas tricky for whales. But with the help of the heat and sound naturally generated by the warm-blooded creatures, researchers are hoping to make their waters safe again.

"They are basically deer on the highway," says Christopher Clark, director of the Bioacoustics Research Program at Cornell University. "And they don't get a chance to learn—they only get one trial."

If whales are unable to avoid humans and their machinery, then it may be up to humans to avoid the whales. The first crucial step is deciphering their whereabouts. That information then has to be relayed in time for appropriate avoidance procedures—such as flipping off sonar or a seismic air gun, or pulling back on a throttle.

Collisions with ships continue to be the leading cause of human-induced casualties for the North Atlantic right whale, an endangered population that is estimated to have declined to around 350. And naval sonar and other ocean-acoustic signals can cause strandings—marine mammals apparently become disoriented by the sounds and sometimes "beach" themselves on shore. A new naval warfare training range off the coast of Florida has further enlivened the sonar debate, which culminated last year in a U.S. Supreme Court decision in a case brought by the Natural Resources Defense Council against the U.S. Navy. (The court decided in favor of national security over whale welfare.) Meanwhile, U.S. Navy, academic and industry researchers continue the hunt for the best mitigation methods.

Traditionally, whale detection was almost solely the job of human observers. "But a person can only concentrate on searching for clues on the sea's surface for a certain amount of time," says Olaf Boebel, head of the Oceanic Acoustics research group at the Alfred Wegener Institute for Polar and Marine Research in Germany. "And you can look in only one direction—you can't look behind you." And that's in the best of conditions. After the sun sets, or during windy or rainy days, whales become especially difficult to see.

Boebel and his colleagues went in search of a solution starting about five years ago. Their result, a new thermal-imaging camera that recognizes the heat of the whale's spout, has just successfully completed its first field test on board the research icebreaker Polarstern. In contrast to cold ocean water, Boebel says, "the blow of the whale will stand out very clearly." On a black-and-white computer screen on board a ship, it will appear a bright white.

Although the earlier versions of the technology had a limited field of view and suffered some instability—a little wave action at sea and it would no longer be pointed at the surface—the technology has evolved so that the camera remains stable against roll, pitch and vibrations. Mounted about 28 meters above the water on the boat’s crow’s nest, it can scan 360 degrees. An automatic recognition algorithm then looks for evidence of a whale's presence in the half a million pictures it produces daily.

The big challenge now, Boebel says, is to "eliminate false positives," noting other features that may produce a whalelike temperature anomaly, including breaking waves, circling birds and the sun reflecting off the water. This task becomes even more difficult in warmer, tropical seas. The end goal, which he predicts to be about five years out, is a working real-time, automatic system that will allow ships to either maneuver around whales or turn off their seismic or sonar equipment in the mammals' presence.

Some scientists take an acoustic approach to whale detection, although not strictly with their own ears. Depending on the species, marine mammals click and whistle at frequencies far higher than humans can detect.

Acoustic sensors developed at Woods Hole Oceanographic Institution (WHOI) and Clark's group at Cornell University can pick up these sounds. Researchers helped place buoys in shipping channels off the coast of New England last year to do just that. The connected underwater acoustic system is designed to detect a common call of the right whale, which is known to traffic these waters.

"There's a brain on the buoy," says Cornell's Clark, who helped conceive the project. He explains how the buoy system automatically sends each suspected whale call to his lab, where it can be scrutinized by human brains. If they decide it's a right whale, a phone call is then made to any approaching ships instructing them to take the proper measures—which almost always means to slow down. (Clark's team is currently working with the U.S. Coast Guard and the National Oceanic & Atmospheric Administration to make this process more sophisticated—alerting mariners on their computerized navigational maps with a blinking buoy.)

More recently, researchers from Analysis, Design & Diagnostics, Inc., of Jacksonville, Fla., joined the mission. Their aim is to expand the range of whale call frequencies that can be identified, allowing them to locate, and potentially save, more species of marine mammals. Two ships currently cruise Florida's Gulf Coast with processing systems that are linked to a hydrophone lying on the ocean floor. More sensors could be deployed from small boats, even dropped out of airplanes.

Last week, company president Gary Donoher submitted a proposal to the U.S. Navy, which is seeking help from an organization to do marine mammal monitoring at ranges throughout the world. In two or three years, he hopes his sensors will be working in real-time, 24/7.

"We are using the sound they make to help protect them," Donoher says, noting the difficulty of sorting out whale calls from the ocean's "cacophony of noise" as well as the problem of detecting whales when they aren't vocalizing. He points out a similar problem for heat sensing, because many species of whales can stay submerged for long periods of time.

Most experts agree that no single approach can do the entire job of keeping a lookout for whales. Donoher advocates combining "complementary technologies" to look, listen and feel for them. "Whether navy or commercial, if you are trying to protect marine mammals, you need multisensor capability," he says. "That is the ideal."