When will the fireball fall on me?: Air crashes inspire an illogical terror. However small the risk, it will never be small enough, says Christian Wolmar

EACH TIME there is an air disaster such as the El Al crash in Amsterdam last Sunday, the anti- aviation pressure groups suddenly become newsworthy. In Amsterdam, local environmentalists responded immediately to the disaster by saying that they had warned of such a catastrophe. While that is true, the residents of the Bijlmermeer estate were desperately unlucky. Their estate is indeed near one of the flight paths, but 10 miles from Schiphol airport. Residents had protested about flights passing over the estate but if demands such as theirs were met, there would be no civil aviation industry at all. Many city- dwellers in the Western world are at similar risk of being killed by a falling aeroplane. Across the world, next to virtually every airport, there are countless pressure groups whose dark forebodings have not come true.

Nevertheless, the aviation industry cannot ignore the protesters. Every year some 15 to 20 jets fall out of the sky killing some or all of the passengers. If present safety levels are maintained and traffic increases at the predicted rate of 6.5 per cent a year, by the early 21st century there will be 40 such accidents annually - in other words, one every 10 days. The industry is unlikely to be able to withstand the protests at such a level of carnage. People will stop flying, local groups will stop airport developments and extensions, and politicians will be forced to side with the protesters.

Looked at dispassionately, the statistics show that air travel is safe and getting safer. However, as one pilot described it recently, the inherent concept of flying is mad: 'You take a machine which is 70 metres long, 65 metres wide and 20 metres high, you pour 200,000 litres of flammable liquid into it, load 450 human beings into the tube and seal it tightly. You ignite the fuel at several locations, using the energy to accelerate this Molotov cocktail to 330 kilometres an hour. Even worse, you climb up to 12,000 metres where the temperature is -60C and the air so thin people can survive for only 20 seconds. In addition you have increased your speed to 1,000 kilometres an hour . . .' And so on.

The insanity of getting into one of these winged Molotov cocktails precludes rational debate. However good the safety record, it may not be good enough if aircraft regularly fall out of the sky, but it is undoubtedly impressive. In 1991, according to Lufthansa's survey of accidents, there were 13,345,000 commercial jet flights in the world, excluding the former Eastern bloc. There were 19 losses, 10 of them crashes involving fatalities, which killed 673 people. Looked at another way, you would have to fly an average of 1,300,000 hours or 150 years to be in a fatal crash. The likelihood has decreased over the last 30 years: in the 1960s you would have had to fly a notional 450,000 hours.

But the public is not interested in the cold logic of these relative figures. The man or woman on the Glasgow shuttle will increasingly let the train take the strain if their complimentary in-flight newspapers are filled with graphic accounts of big plane crashes.

Airline safety will have to be improved further still for it to remain an acceptable form of transport. Hard decisions need to be made. Every day, 500 jets fly into Heathrow. There are two main flight paths and one is through the centre of London. While Heathrow has not had a major crash since the Staines disaster 20 years ago (a Trident stalled after take-off and all 118 people on board died) it is almost bound to have one, say, over the next decade - in the next two million landings. Most accidents occur at, or near, airports. With good luck the next Heathrow crash will result in the plane plunging into the reservoir at the end of the runway; with bad luck it will smash into the NatWest tower. If it is the latter, what politician will resist the clamour to stop the overflying of central London? French aviation authorities have already, in the wake of the Amsterdam disaster, ordered a safety study of the runway approaches at French airports, in particular Orly and Charles de Gaulle in Paris.

The airlines are going to have to take the lead in further reducing risks. An analysis of accidents suggests some simple measures can be taken. Freighters have a disproportionately high number of crashes: they make up only 4 per cent of all flights but, in 1991, out of the 19 losses, five, or 26 per cent, were freighters. This year, including Sunday's crash, six out of the 13 losses - 46 per cent - were freighters. The reasons are unclear, though one pilot told me that planes are always 'filled to the gunwales' and that accurate weights, crucial for the pilot to know how to handle the aircraft, are not taken. Sunday's disaster should prompt detailed examination of the poor safety record of freighters.

The most obvious way to reduce crashes is to improve the performance of pilots. Lufthansa's accident analysis shows that 75 per cent of crashes between 1959 and 1990 were caused by human error (the remaining quarter being split almost equally between technical and environmental factors). Clearly, improved training, particularly in emergency situations, is needed. But several trends in the industry are likely to put more, rather than less, pressure on pilots. Many newer aircraft such as Boeings 757, 767 and the new 747-400 (not the version involved in the Amsterdam crash) and the Airbus 320 have only two, rather than three, crew on the flight deck. Lengthy negotiations are already under way to standardise pilots' hours in the 12 EC countries which may result in longer hours.

In addition, safety costs money and as the deregulation of the European skies takes place, airlines will look at ways to improve economic performance. The United States pilots organisation, USALPA, has long expressed fears about safety being compromised because of cost-cutting.

The Amsterdam catastrophe has given unprecedented ammunition to the pressure groups. It was the aviation disaster we had all been dreading, because it makes everyone on the ground a potential victim.

The debate on air safety is not a rational one. Spending on safety must reduce risk to levels much lower than, say, those required of car design or road layout. That is the inevitable logic of flying in winged Molotov cocktails. In the immediate aftermath of the disaster, the plans to expand Schiphol to take 30 million passengers a year by the year 2000 (as against 18.5 million last year) will be questioned and similar schemes across the world will be re-examined. As Bert Wiedermeyer, leader of a campaign against the expansion of Schiphol airport, put it: 'The government said the risk (of an air disaster) was very low. They said that statistically it couldn't happen. We say that if it can happen, it will - just like Chernobyl.'