¿ìè¶ÌÊÓƵ

MILITARY submarines exist because the ocean is a great place to hide.
Light fades quickly underwater, shielding objects from view. Radar is nearly
useless, because its waves do not penetrate water. The result is that naval
forces assigned to hunt submarines are essentially blind. Nature, however,
makes one extremely useful concession: sound travels through the water at
1500 metres per second, which is five times as fast as it travels in air.
A loud noise in the ocean can be picked up by arrays of acoustic sensors,
called hydrophones, hundreds and even thousands of kilometres away.

NATO countries, primarily Britain and the US, have invested billions
of dollars in equipment to detect the sound of Soviet submarines. Until
recently, it was a simple task. Soviet submarines were noisy; their rotors,
bearings and turbines were less well machined and insulated than those of
the American and British boats. Tales circulated about the omnipotence of
American antisubmarine warfare capabilities. Some claimed that the US had
underwater arrays of acoustic sensors that could detect a submarine anywhere
and pinpoint its location to within 100 kilometres.

Since the beginning of the 1980s, however, the Soviet Union has been
turning out submarines that are quiet enough to confound NATO’s sonar operators.
And it is not alone: West Germany, Italy and Sweden are designing quiet
submarines for export and these boats could easily fall into hostile hands
(‘The incredible shrinking submarine’, ¿ìè¶ÌÊÓƵ, 1 April 1989). ‘Over
the past twenty years it was relatively easy to find the other guy,’ says
Craig Dorman, a retired admiral who oversaw the Pentagon’s antisubmarine
efforts until last February. ‘But what now? That’s what has got people scratching
their heads.’ In March, a panel of experts in antisubmarine warfare told
the Armed Services Committee of the US House of Representatives that ‘maintenance
of our antisubmarine warfare capability is arguably the most important of
all the challenges facing the Department of Defense today’.

The result has been a surge of interest in new ways to monitor what
is going on inside the ocean. In the US, the navy’s own network of laboratories
is doing most of the work, aided by basic research at universities. Britain
uses the network of laboratories managed by the Admiralty Research Establishment
of the Ministry of Defence, which says there is ‘close collaboration’ between
the two NATO allies. But even allied governments are wary of sharing their
information; antisubmarine warfare is among the most closely guarded areas
of military research. ‘Cooperation is not 100 per cent. We all get pretty
goosey about these things,’ says Dorman. He is now director of the Woods
Hole Oceanographic Institution, an independent research centre on the coast
of Massachusetts.

Various ways of detecting submarines without acoustics have been proposed.
Research into them has remained secret and, as a result, the work has acquired
a mystique, which has led to talk of imminent breakthroughs that could render
the oceans transparent. According to Dorman, however, none of the nonacoustic
techniques promises any dramatic improvement: ‘There’s no magic. The secret
is that there isn’t any secret.’ A laser beam of blue-green light, for example,
can travel through water far better than light of other frequencies. In
theory, the laser can act as a sort of radar, detecting an object by measuring
the reflection of its signal. But fundamental physics limits its potential
as a military device. Water reflects a continuous beam of light far more
strongly than something below the surface, effectively masking the underwater
object from view. The laser needs to fire extremely short pulses of light,
less than a microsecond in duration, to distinguish between the two reflections.
The trouble is that the atmosphere lengthens pulses of light by several
microseconds, which makes the system unworkable. The only way round this
problem is to reduce the time that the light travels through the atmosphere
by firing the laser from a low-flying aircraft instead of from a traditional
high-flying reconnaissance plane or a satellite. The disadvantage is that
low-flying aircraft can cover only a small portion of the ocean at a time.

An alternative surveillance system involves using radar. Although water
reflects radar signals, radar may still be able to detect the presence of
a submarine from the waves the boat creates at the surface. The trouble
is that these waves are tiny compared with the natural motion of the high
seas. For instance, a large submarine travelling at 5 knots at a depth of
100 metres would generate ripples rising just 1 millimetre above the surface
as it passed.

The possibility that radar can detect the effect of a passing submarine
on an ocean’s pattern of internal waves, the slow oscillation of water of
varying temperature below the surface, has aroused greater interest. Some
research indicates that surveillance systems could use radar images of an
ocean’s surface to analyse the pattern of internal waves, although no one
has yet demonstrated a way of using this information to detect submarines.

Other methods that have been mentioned are even less practical. Some
organisms in an ocean give off light when they are disturbed by a passing
submarine; but their light is too faint to be noticed at the surface. Instruments
can detect variations in the natural magnetic field caused by a submarine’s
massive steel hull; but they can distinguish the disturbance only when the
vessel is less than a few hundred metres away. In reality, nonacoustic methods
offer only limited help with surveillance; none can restore the advantage
that the US and its allies had over the Soviet Union during the era of noisy
Soviet submarines.

The US Navy wants to concentrate its efforts in a field that it knows
works: acoustics. And while it insists that its own submarines are in no
danger, particularly those carrying intercontinental nuclear missiles, other
parts of the government have spread suspicion that the navy may be ignoring
significant loopholes in the defence of the US. The US Air Force, which
has an interest in promoting its own ground-based missiles, has frequently
suggested that submarines may not be as invulnerable as the navy likes to
think. Congress has accused the navy of an institutional conflict of interest:
in order to protect its submarine programmes, it is unwilling to explore
technologies that might show its submarines to be vulnerable.

Congress showed its impatience in 1984 when it set up two programmes
of research into technologies for antisubmarine warfare and put them outside
the navy’s control. One was conducted by the Defense Advanced Research Projects
Agency (DARPA) of the Department of Defense, and the other by the CIA. DARPA
has inaugurated some unusual international cooperation in its research on
nonacoustic techniques of surveillance. In May, it signed an agreement with
institutes in Britain and Norway authorising joint research on the study
of the ocean surface by radar.

For now, however, sound waves remain the primary tool for monitoring
the ocean. One idea is to use detailed data on the underwater environment
to ‘unscramble’ the confused sounds picked up by acoustic sensors. In the
past, naval oceanographers assumed that the oceans were made up of layers
of water, each layer having a slightly different temperature and pressure.
The oceans turn out to be more complicated than that, however. They have
their own weather patterns, with fronts and eddies, that can bend sound
waves crazily, in horizontal as well as in vertical directions. This creates
strange effects. Instead of heading straight for an acoustic sensor, sound
waves may bend away into oblivion, leaving a region known as a ‘shadow zone’
from where no sound is heard at all. They may then converge from different
directions at slightly different times to confuse another sensor.

Arthur Baggaroer, director of MIT’s Underwater Acoustics Laboratory,
compares listening to sound in the ocean to looking across a hot desert
at the distant horizon. In the desert, the hot air near the Earth’s surface
bends light waves. This creates mirages – bits of blue sky that appear to
lie on the ground, or trees and buildings that seem to be rooted a long
way above it. The distortion of sound waves in the ocean is even more severe.
An array of hydrophones that is set to pick up sound waves from one direction
will ignore those coming from others. If the sounds from a submarine travel
on different paths and arrive at slightly different times, the hydrophones
may not recognise them as components of the same signal. As a result, much
of the sound that a submarine emits is discarded as meaningless noise.

With a detailed, three-dimensional and up-to-date map of the ocean,
sonar operators could predict much of the bending and distortion of the
acoustic waves. They could use the information to help powerful computers
to process the raw data from an array of hydrophones. In effect, the computers
might correct the distortions that oceans cause. This would greatly increase
the array’s ability to detect submarines. ‘Essentially, you want to turn
noise into signal,’ says Ed Freeman, director of the Scripps Institution
of Oceanography, a private research centre in San Diego. The US Navy has
simulated the method in computers, but it admits that a practical demonstration
is many years away. For the moment, scientists are working on ways to collect
detailed data on the changing temperatures, currents and internal waves
of the oceans.

One possible method, called acoustic tomography, uses sound waves to
try to determine what really happens underwater. Powerful sources of acoustic
energy send pulses of low-frequency waves of between 200 and 300 cycles
per second across hundreds, sometimes thousands, of kilometres of ocean,
where they are received and analysed. In much the same way that a crisscross
pattern of X-rays can assess the internal structure of the brain, sound
waves can reveal the temperature and pressure gradients of the ocean at
any one instant.

Researchers at Woods Hole, the Scripps Institution and the Applied Physics
Laboratory of the University of Washington have begun using tomography to
map regions of the Greenland Sea and the northeast Pacific. These experiments
aim to map features of the ocean that stretch for a hundred kilometres or
more: such ‘mesoscale’ features change slowly but they might need to be
updated every few days if sonar operators are to appreciate the influence
of the features on signals from submarines.

The US Navy would like to create a snapshot of the ocean that captures
features a few kilometres wide and a hundred metres deep. Obtaining such
a detailed picture will require many sources and receivers of sound waves,
and would be quite expensive. Other researchers are exploring whether satellites
can perform a similar task, with infrared sensors mapping the temperature
of the ocean surface and altimeters measuring its swells: navy researchers
want to be able to use the data to infer how water below the surface behaves.

Another technique would use a known acoustic signal, sent from a surface
ship, to correct the distortions in other sounds coming from nearby areas
of the ocean. After analysing what the ocean had done to the known sound,
computers could better unscramble sounds from other potential targets in
the same general area. The Naval Ocean Systems Center, an R&D establishment
of the US Navy in San Diego, is developing the technique. In one simulation,
a signal from a ship 500 kilometres away from an array of hydrophones improved
the recognition of a submarine at a range of 800 kilometres.

Apart from being able to listen more carefully, the navy’s submarine
hunters also want to monitor different sounds. Currently, sonar operators
listen for the hum of machinery, because they can distinguish it from the
ocean’s background noise. The sounds of propellers, pumps, generators and
gears show up on a sonar screen as a series of narrow spikes at particular
frequencies. The combination of frequencies that a submarine’s machinery
emits is its ‘signature’. Sonar operators spend years learning to recognise
different classes of submarines on the basis of these characteristic patterns.
By most accounts, they are quite good at it.

Machinery can be muffled, however, which is what the Soviet Union has
done. Bearings and rotors can be more finely made so they fit more snugly
in position; machinery can be mounted on floating platforms that isolate
vibration from the hull; and insulation can be added. Other sounds, however,
cannot be silenced, such as the low thunder produced by driving several
thousand tonnes of submarine through the water. On a sonar screen, this
kind of acoustic energy appears as a bulge of increased sound that spreads
over a range of very low frequencies. For that reason, it is called a ‘broad-band’
signal. The problem is that such sounds are hard to recognise against the
background noise of an ocean. There are lots of things, from whales to ocean
currents, pushing through the water, and sonar operators may not be able
to pick out the sounds of a moving submarine.

Submarines also emit noises that last only a fraction of a second and
flash across the sonar screen as a quick blip. These transient noises, which
tend not to be of a single frequency, are produced when a pump starts, or
when a submarine shifts its rudder and hydroplanes to change direction.
The US Office of Naval Research has begun an accelerated programme of research
aimed at describing such signals mathematically, so that computers can detect
them. According to Edward Franchi, director of acoustics research at the
Naval Oceanic Research and Development Activity in Bay St Louis, Louisiana,
there is no reason why transient sounds cannot be recognised. ‘Bird chirps
tend to be a sweep of frequencies. Certainly the human ear can pick them
out and recognise them. That’s sort of what we’re looking for.’

If, despite these efforts, Soviet submarines remain too quiet to be
heard, the US Navy may be driven to do something that it has been trained
to avoid: it will turn its sonars on. ‘Active’ sonar sends out sound pulses
and listens for echoes bouncing back from underwater vessels. The problem
with active sonar is that it gives away its own location. Even a submarine
that is too far away to be detected by active sonar can still hear the sonar’s
signal.

Because of the problems with active sonar, the US Navy never learnt
to use the system well. It never developed adequate ways of predicting the
echoes that an active sonar receives from the ocean bottom, from the surface
or from marine organisms. All of these can play havoc with sonar. The Royal
Navy has similar problems. During the Falklands war, British warships used
their active sonar almost constantly, trying to detect a few of Argentina’s
diesel/electric-powered submarines. But the sonar systems were so confused
by echoes from whales or from the ocean bottom that the warships dumped
most of their antisubmarine weapons on false targets.

Active sonar systems can send pulses at lower frequencies, around a
few hundred cycles per second, to increase their range. At those longer
wavelengths, the pulses are also less affected by waves in the ocean, tiny
organisms and small variations on the ocean bottom. On the other hand, low-frequency
energy reflects off sediment buried in the ocean floor, which can complicate
matters. Low-frequency signals are also hard to generate. ‘You’re talking
about sources that are large, heavy and tend to require an extraordinary
amount of power to generate the sound,’ says Franchi.

Unless research bears significant results, efforts to find submarines
in the ocean depths will depend on more prosaic measures. Arrays of hydrophones
may become larger and stretch in two dimensions instead of one. Sensors
may proliferate in more areas of the ocean, so as to shorten the distance
between them. All of these measures will cost large amounts of money, however.
Some specialists believe that it is a losing battle, and that submarines
can be made more elusive faster than measures can be developed to find them.
And growing knowledge of the ocean can reveal ways to hide a submarine as
well as ways to find one. A few years ago, James Watkins, Chief of Operations
of the US Navy, said the ocean was becoming more opaque, not more transparent.
If that is true, it is good news for submariners, and bad news for the commanders
of warships on the surface.

* * *

The secrets of the oceans – safe in military hands

FEW AREAS of research are so closely guarded by the military as techniques
to locate submarines. Naval officials responsible for R&D refuse to
discuss anything other than the most basic programmes. For instance, a project
manager at the US Office of Naval Research, which funds work at universities
that is almost entirely unclassified, refused to disclose the names of people
devising mathematical techniques for processing sonar signals. In Britain,
the Admiralty Research Establishment, which comes under the auspices of
the Ministry of Defence, was just as reticent. ‘I’m afraid I can’t say anything
at all,’ said an official.

The secrecy extends to data on the oceans themselves. In 1978, the US
Navy declassified a sonar device for mapping ocean floors, known as Sea-Beam,
and made it available for anyone to use: other systems had long since superseded
the accuracy and speed with which Sea-Beam could work.

Six years later, the National Atmospheric and Oceanographic Administration
(NOAA), a government agency, began to use the instrument to map the US’s
exclusive economic zone, which extends for 320 kilometres from the mainland.
Sea-Beam’s accuracy deteriorates with depth but it remains a useful tool:
at 2000 metres, it can measure the average depth of a block of ocean floor,
100 metres across, to within 2 metres. The navy protested, claiming that
maps of this quality could help hostile submarines close to the American
coastline. The NOAA ignored the protest; it said exploration companies needed
accurate maps to exploit the zone.

The dispute was not settled until March this year, when the National
Security Council ruled that mapping could continue except in areas that
the navy regards as particularly sensitive, such as in the vicinity of military
bases. Using Sea-Beam, the NOAA has mapped about 2 per cent of the 3.4 million
square nautical miles that make up the exclusive zone: the entire task will
take decades to complete.

Dan Charles is a science and technology writer based in Boston, Massachusetts.