¿ìè¶ÌÊÓƵ

Detecting a bomb in a suitcase makes finding a needle in a haystack
look simple. In the four years since Pan Am flight 103 was destroyed over
Lockerbie, killing 270, five more bombs have exploded in the holds of commercial
airliners. But an estimated 2 billion pieces of luggage are carried on airliners
each year, making the probability that any one bag will contain a bomb less
than one in a billion.

These sort of odds do not impress airlines. The risk is so small, they
argue, that the cost of setting up systems to detect them is not justified.
But people dislike playing Russian roulette – even with those odds – each
time they board a plane, as the dip in travellers through fear of terrorist
action around the Gulf War shows.

Screening of all passengers and their hand luggage for weapons or explosives
by hand searches, metal detectors and X-ray machines has drastically cut
the number of hijackings since it was recommended by the International Civil
Aviation Organization (ICAO), to which almost every country belongs, in
1974 (see Graph). Such screening has since become mandatory, unlike another
recommendation, to screen all baggage checked into the hold. But if screening
passengers for weapons can virtually eliminate hijackings, why do airlines
not screen checked-in bags for bombs?

In fact after Lockerbie the British Airports Authority, which owns seven
of Britain’s major airports including Heathrow and Gatwick, carried out
a study to see if a combination of the most modern X-ray machines and hand
searching could be used to screen all checked-in bags. They found that the
delays, the cost of extra staff and the extra space needed for hand searching
would have halved the capacity of airports. ‘It would have devastated the
industry,’ says Bill Oatey, BAA’s director of safety and security. British
Airways says it carries out 100 per cent screening on ‘selected’ flights
(chosen on the basis of threat assessment). ‘Of course the industry is
moving towards comprehensive screening,’ says a spokesperson. ‘But it should
be done properly. If it were done now, the industry would grind to a halt.’

The problem is that terrorists have become so ingenious at disguising
bombs. Modern plastic explosives can, as the name implies, be moulded, yet
are very stable – Semtex, the one most commonly used by terrorists, will
not explode even if hit repeatedly with a hammer. The bomb that Nezar Hindawi
gave his unwitting girlfriend to carry onto an El Al flight to Israel in
1986 consisted of 3.5 kilograms of explosive, flattened and sewn into the
base of the suitcase, with a detonator hidden in a pocket calculator that
still worked. It was found by a hand search. The terrorists who constructed
the Lockerbie bomb hid a smaller charge of explosives in the power pack
of a stereo cassette player, with a barometric fuse (set off by a drop in
pressure) and timer hidden behind the tape deck. These were primed to go
off at a certain time and altitude. Had anyone checked, the tape deck would
have blared out music.

That bomb also exposed the gaping hole in international airline security.
It was hidden in a ‘rogue’ bag that somehow slipped into the air transport
system at Frankfurt, possibly passing undetected through an X-ray machine,
and then travelled on unaccompanied by a passenger. Yet the ICAO has mandatory
standards covering ‘baggage reconciliation’ to ensure that passengers board
the flight they check in for, and do not leave it without their luggage.
(It is a safe assumption that modern terrorists are not suicidal, and will
not be on board a bombed plane.) The baggage of any ‘no-shows’ must be found
and screened. But still the Lockerbie one slipped through.

Later this year the Department of Transport will stipulate to airlines
using British airports that every bag must be accounted for, and that any
unaccompanied luggage must be authorised for travel, probably by screening.
But doing this manually would be impossibly complicated except in a very
small airport, so the DoT is awaiting suitable technology to be developed.

Such systems would probably rely on tags with bar codes, like those
on supermarket goods. At check-in, each bag would be tagged with a bar code
specifying its owner, flight and destination, and also including a random
number so terrorists cannot make their own tags. The code would be added
to a computer list of the bags on the flight, and baggage handlers would
use bar code readers to check off each bag against it, removing any unaccompanied
ones for extra checks.

But reconciliation, like all forms of security, can never be 100 per
cent reliable. An airport like Heathrow handles 68 million bags every year.
Terrorists are sure to seek out chinks in its armour.

The other, blanket approach is to scan every checked-in bag with a bomb
detector. However, X-ray machines used to scan cabin baggage were designed
to spot dense metallic objects such as guns and knives. The metallic parts
of a bomb can be easily disguised as an everyday electrical gadget, and
spotting the plastic explosives is trickier. The key is that they are denser
than other organic materials, and contain between two and six times more
nitrogen than other plastics.

In 1987, the heightened political tension between the US and states
believed to have sponsored terrorists prompted the Federal Aviation Administration
to begin a concerted effort to find techniques for scanning checked-in bags.
After Lockerbie the search accelerated: between 1989 and 1991, the FAA’s
research and development budget tripled to $30.3 million per year. It issued
specifications of the type of machine that would be required. Although many
of the details are classified, it is known the machine would have to automatically
scan at least 600 bags an hour (allowing just six seconds per bag).

The only single technique that then appeared up to the job was called
thermal neutron analysis (TNA). The FAA poured $45 million into the development
of TNA machines, which were being produced by Science Applications International
Corporation (SAIC) of Santa Clara, California, and bought six prototypes
costing around $1 million each. Some of these were tested between 1989
and 1992 at three major airports in the US, as well as Gatwick, where it
is still in use.

TNA works by moving bags through a beam of thermal, or ‘slow’, neutrons
generated from radioactive californium-252. Nitrogen atoms capture thermal
neutrons, after which they emit a photon of gamma radiation at a certain
energy. The more of those gamma rays detected, the higher the nitrogen density
in the case. For a given sensitivity, TNA produces a simple, yes/no decision.

But the TNA machines proved very unpopular in the airport trials. Each
machine weighs 9 tonnes because of the lead shielding needed to protect
operators from radiation, and if combined with an X-ray machine can be up
to 12 metres long. Its effectiveness has also been questioned. Before Lockerbie,
experts had assumed that it would take a sizeable bomb to bring down a large
airliner such as a 747. TNA was designed to detect bombs of 2.5 kilograms
or larger. Some estimates of the size of the Lockerbie bomb are as small
as 300 grams.

‘TNA is dead,’ says Lee Grodzins, an expert on explosives screening
at the Massachusetts Institute of Technology. ‘It could possibly be resurrected,
but the SAIC design is not acceptable after Lockerbie.’ Adjusting a TNA
machine to detect bombs of 300 grams dramatically increases the proportion
of bags that it falsely identifies as suspicious. This is unacceptable to
airlines because suspect bags usually need to be searched by hand. At an
airport like John F. Kennedy in New York, which handles 4000 bags per hour,
a false alarm rate of 20 per cent would mean hand-searching 800 bags per
hour. The problems with TNA have shifted attention back to X-rays. In fact,
with some additional technology, X-ray machines can spot plastic explosives.

X-ray machines simply shine a beam of X-rays through the bag. Different
materials absorb or scatter different amounts; an image of the bag’s contents
can be picked up by a detector. The attenuation of the beam depends on its
energy, the thickness of any objects in its path and their effective atomic
number. The problem with detecting explosives is that they consist of elements
such as nitrogen, oxygen, carbon and hydrogen, all with low atomic numbers.
Objects with high atomic numbers attenuate more X-rays, masking the explosives:
a metal baking sheet in a suitcase easily fools a simple X-ray machine.

Several methods have evolved to get round this. The first is to use
X-rays of two different energies simultaneously. Each material attenuates
the two sets of X-rays differently. By comparing the attenuated beams, the
system can estimate the material’s atomic number. In this way the machine
can distinguish between dense organic material – bombs – and dense inorganic
materials such as metals, and between bombs and other less dense organic
material.

The machines made by EG&G Astrophysics, based in Long Beach, California,
the world’s largest manufacturer of X-ray security equipment, use this
technique. The company has been making such machines for several years but
they are not automatic; the scan’s results are presented on a colour screen
for a trained operator to interpret. But human operators get bored, and
their attention wanders.

So EG&G is now testing an automatic machine which scans the bags
with two beams, each with two different energies, from the side and below.
Information from both scans is combined in a purpose-built high-speed computer
that calculates a three-dimensional density map of the bag. If any region
shows the high-density organic material characteristic of a bomb, the machine
automatically diverts the bag from the conveyor for further screening. One
such machine is now on trial at San Francisco airport, and another is expected
to begin trials at a British airport this month. EG&G hopes to have
them in production by the summer at a cost of around $120 000.

A slice of danger

Meanwhile, InVision Technologies of San Francisco is in the final stages
of developing a quicker, toughened version of a computerised tomography
scanner adapted from hospital designs, which takes X-ray scans as it rotates
around an object and combines the views on a computer to give a ‘slice’
image.

InVision’s machine works automatically and can do a slice in 1 second.
Screening a bag takes between 9 and 12 slices, depending on its contents.
Bags that do not arouse suspicion pass automatically. For the others, the
machine presents the operator with two images taken from different directions,
to help decide what to do with it. The company is considering adapting the
machines so that such bags can be scanned again, letting the computer focus
on particular areas of the bag and even spot booby traps – a technique the
company calls ‘electronic unpacking’. At $800 000, it is more expensive
than other X-ray machines but InVision says its accuracy makes it particularly
suitable for high-risk airports.

Several techniques are also being developed for detecting the vapour
that explosives give off. However, such ‘sniffers’ face a daunting task:
modern plastic explosives are not very volatile, and wrapping reduces the
amount of vapour that escapes. Estimates suggest that sniffers would need
to detect vapour diluted to 1 part in 1015 in air.

Parallel research being pursued on both sides of the Atlantic is ‘hardening’
aircraft and luggage containers. If aircraft holds can withstand larger
internal explosions, leaving the passengers unhurt and the pilots in control,
the detectors in airports can look for larger bombs.

Coverups for shortcomings

Lyle Malotky, the FAA’s science adviser for aviation security, says
that since the late 1980s, when TNA was the favoured technology, there has
been a change of thinking at the FAA. This change may in part have been
brought about by a report on explosives detection drawn up by the National
Academy of Sciences in 1990. John Baldeschwieler of the California Institute
of Technology in Pasadena, who was a member of the NAS panel, says the report
concluded that there should be no one ‘magic device’ to solve the problem.
Airports should instead set up a multilayered system of detectors using
different technologies so that shortcomings of one method would be covered
by the strengths of another.

The BAA also came to this conclusion and has developed a system which
it plans to test at Glasgow and Heathrow from June. The first layer, which
must work quickly because it has to screen every bag, will be an automatic
advanced X-ray system. This should clear 80 per cent of the bags to go straight
to the plane. The other 20 per cent will go to a second advanced X-ray machine,
which will produce a high-quality image for an operator, who will look for
bomb components such as a detonator and fuses. This should leave only about
1 per cent of bags as ‘suspicious’, which will go finally to a vapour and
particle analyser to ‘sniff’ for explosives. The final 0.1 per cent of bags
will then be opened in the presence of their owners. (At Heathrow, that
would mean an average of fewer than 10 bags per hour.)

Grodzins says that if governments wanted, such systems could be working
in airports within a year. But there are conflicting pressures. Groups such
as the Lockerbie families are campaigning hard for better security; but
the airlines have powerful political and commercial lobbies. Recession
has pushed some airlines into bankruptcy, so enforcing strict security measures
would push up ticket prices. ‘There is a tremendous amount of politics between
the airlines and the FAA,’ says Grodzins.

Baldeschwieler says, ‘Regulation and legislation are needed to get things
moving. Federal government must require that, in such a competitive environment,
it is done by everyone.’ But one expert describes the FAA’s movement towards
mandatory screening as ‘glacial’. Malotky says the FAA does not expect to
make screening mandatory before the end of 1995.

The situation is similar in Britain. One expert says the airlines and
the DoT are ‘dancing a minuet’ over compulsory screening. The DoT has asked
the airlines and airport operators to submit their plans for screening
all checked-in bags. It is not stipulating multilayered or automatic systems.
Screening will be phased in over the next four or five years at all British
airports.

Four years after Lockerbie, governments seem to have lost a sense of
urgency about the problem, despite the progress in developing the necessary
technologies. But as Grodzins says: ‘it’s easy to do nothing until another
Lockerbie comes along.’