Death Knell.

30th Oct 2002 Story: society.guardian.co.uk

Last month, 17 whales beached and died in the Canary Islands. Nobody knows for sure what killed them, but the government of the Canaries made an educated guess. Four mass whale strandings have occurred in the islands in the last 20 years, and each one of them coincided with Nato naval exercises.

This time was no exception. Neo Tapon 2002, a Nato exercise hosted by the Spanish government, had just left the area. Nato argues that there is no proof of a causal link. However, the local authorities politely have asked them not to return.

In the summer of 2000, 17 whales of various species beached in the Bahamas. Necropsies revealed bleeding in the ears and brains. Just an hour before the first whale was found, a squadron of American destroyers had passed close by.

In 1996, 12 Cuvier's beaked whales died among the Greek islands, immediately following Nato naval exercises. Nato was testing a new system of sonar that many marine biologists fear could prove the final straw for the harassed leviathans' low frequency active sonar.

Sonar, the acoustic system that ships use to find submarines, is the prime suspect in all these cetacean deaths. Everyone is familiar with the pinging sound that punctuates old submarine movies. This is active sonar, a sonic equivalent of radar. The submarine hunter emits pings that bounce off the hull of the lurking sub and are picked up by microphones.

Sonar has changed dramatically in the last few years. Cold war technology relied on passive sonar, which emits no sound but merely listens for the engine noises of the underwater enemy. Passive sonar is no longer good enough, according to military planners, as modern subs are just too quiet. So navies are turning back to active systems, but these new sonars emit pings that are millions of times louder than the vintage models.

The US navy's new sonar, called Surtass LFA, carries 18 loudspeakers that generate 235 decibels each. In crude terms, that is equivalent to standing next to an Apollo moonrocket. At a range of 160 kilometres, the signal can still reach 160 decibels. That is 50 times louder than the US navy's official safety limit for human divers.

When it emerged in 1996 that the US navy had conducted at least 22 secret tests of Surtass LFA around the US coastline, marine biologists and environmental groups were enraged. The National Resources Defense Council, one of America's biggest environmental groups, accused the navy of breaching the Endangered Species Act, the Marine Mammal Protection Act, and the Environmental Policy Act, which demands that an environmental impact analysis be performed before undertaking any potentially harmful activity.

Hoping to avoid a lawsuit, the navy promised to perform environmental studies. These consisted of observing the reactions of humpback, right, grey and fin whales to reduced-power signals from the LFA array. The tests produced clear reactions from most whales exposed to the sonar, with the majority fleeing the scene.

At the end of it all, the US navy announced that the observed reactions did not add up to a "significant biological impact" that might interfere with mating, feeding or migration. That, say conservation groups, is a completely arbitrary judgment.

What worries scientists most is that none of those whales were exposed to much more than 150 decibels, yet the navy used those results to set a safety limit of 180 decibels, a level 1,000 times louder.

The US navy is under fire precisely because it is more open and accountable than European navies, whose LFA programmes are developing away from public scrutiny. Anthony Watts, the editor of Jane's Underwater Warfare, says France, Germany and Britain are all building similar systems. "The French system is about 220 decibels; the German model, intended for export, is about the same. The British system's power is classified."

The Ministry of Defence says revealing the sound level