LISTENING for whales may provide a more accurate idea of their numbers than looking out for them from ships, say American biologists. Researchers at Cornell University in Ithaca, New York, have developed an 鈥渁coustic telescope鈥� that is better at spotting the mammals than human whale watchers.
The sound-sensitive 鈥渢elescope鈥� is an array of 16 hydrophones on a line 1.6 kilometres long. The team from Cornell towed the hydrophones behind their research ship off the coast of southern California in September. While professional whale spotters kept a lookout for whales, the hydrophones listened for their sounds. The hydrophone system did a better job, says Christopher Clark, director of Cornell鈥檚 bioacoustics research programme. For example, at one point, says Clark, the whale watchers spotted a number of blue whales near the ship 鈥渉aving a party鈥�. But the acoustic records revealed what the human spotters could not see 鈥� that there were several other whales swimming in to join the group. The acoustic system was able to identify individual whales and work out what they were doing from the sounds they were making.
鈥淭he humans are pretty much limited to what you can see from a pair of binoculars from the bridge of a boat,鈥� says Clark. 鈥淏ut what鈥檚 going on is quite different from what meets the eye.鈥�
Advertisement
The trial also showed that the acoustic technique can correctly identify the species of whale from its characteristic sounds. This means that it could come up with far more accurate figures for whale populations.
The next step is to compare the data from the experiment with a set of acoustic recordings made at the same time by a network of secret US Navy hydrophones on the Pacific seafloor. The network was intended to track Soviet submarines, but the Navy recently decided to give some researchers access to the data it collects. If the Navy鈥檚 hydrophones work as well as Cornell鈥檚, they could be used to monitor whales over many hundreds of kilometres, says Clark. 鈥淚t could be the basis for looking at large-scale trends in populations.鈥�