When Iceland's Eyjafjallajökull volcano began erupting on 14 April, European aviation safety authorities were faced with an unprecedented dilemma and took drastic action: to avoid a loss-of-power disaster caused by engine damage, much of Europe's airspace was shut down the next day. Six days and $1.7 billion in airline losses later, flight bans were lifted.
But was the shutdown a massive overreaction or a necessary precaution?
Ostensibly all that the European authorities had to do on 20 April to get Europe flying again was to determine an atmospheric ash contamination density, below which the risk of airframe and engine damage was sufficiently small to be acceptable, and to clear aircraft to fly in the parts of the sky that met that criterion.
On 15 April, however, the authorities received a salutary warning: five Finnish air force Boeing F-18 Hornets had taken off from their base north of Helsinki and an inspection made after their return showed what appears to have been severe ash damage to some of their engines.
The only existing International Civil Aviation Organisation guidance is that the aviation community had been unable to agree a level of atmospheric volcanic ash contamination below which the risk to aircraft and their engines is acceptable. ICAO advice is emphatic: operators should avoid any level of ash contamination.
A tacit agreement from the manufacturers existed, saying that engines could tolerate contamination levels of less than 10-17g/m3, but this was still not considered advisable and had not been adopted. The policy accepted worldwide was that the nine global Volcanic Ash Advisory Centres would track the location of volcanic ash and report it so aircraft could avoid it.
The UK Civil Aviation Authority makes this clear in its statement of principles agreed in the last few days between all the European authorities: "Current international procedures recommend avoiding volcano ash at all times. In this case owing to the magnitude of the ash cloud, its position over Europe and the static weather conditions, most of the EU airspace had to close and aircraft could not be physically routed around the problem area as there was no [uncontaminated] space to do so."
But, having submitted the 10-17g/m³ figure to industry-wide review in the light of the European emergency, the CAA reduced the acceptable contamination level to a lower figure of 10-16g/m³, and declared that they could open sky sectors with contamination levels at that level or below, and that enabled the authorities to declare most of the sky over Europe safe that day.
Volcanoes erupt frequently all over the world, but normally the ash they produce affects areas in which aviation activity is light. This European experience was unique in aviation history because ash contaminated the sky over an entire continent that happens to have the densest air traffic in the world.
The previous tactic for dealing with ash clouds - avoiding them - was not available either for operators based there or for those for which Europe is a scheduled destination.
UK air navigation service provider NATS explains its remit - one which it and all its counterparts are required to follow: they are "not to direct flights into a known flight hazard". If it were to do so, its position would be morally and legally parlous.
The question was, is this a "known" flight hazard, or just a suspected one? On Monday 19 all the European aviation agencies, with industry participants, convened a massive videoconference and thrashed out an agreement that would enable an increased amount of flying to take place within an ash-contaminated region, but only when and where the contamination was below the agreed figure.
This is easier said than measured. Weather satellites can identify visually the core volcanic ash plume downwind of the vent, but cannot detect the type of widely dispersed ash particles that filled the skies over Europe.
Their progress is estimated by computer modelling and atmospheric sampling. There are greater and lesser concentrations of particles at different vertical levels, their vertical and lateral distribution determined respectively by the strength of the individual eruption that projected the ash skyward, then the weather system.
Meanwhile, below these layers, the dust is constantly drifting down from the upper air to the ground, so nowhere below the ash layer is absolutely ash-free.
At the Monday videoconference, consideration was given to the results of test flights by aircraft of KLM, Lufthansa, Air France, British Airways and others including Airbus. Their experience had been universally benign.
But the experience of the Finnish F-18s was also on record, and a fully-instrumented Met Office Dornier 228 that flew multiple sampling sorties found levels of ash in some sectors that would represent a risk to flight safety.
Michael Fabian, a professor of mechanical engineering at Embry-Riddle Aeronautical University, explains: "It will be difficult to know the extent to which jet engines can tolerate mild to moderate intakes of ash. The damage might be cumulative and is tough to detect."
The UK CAA explains the problem the authorities faced: "We had to ensure, in a situation without precedent, that decisions made were based on a thorough gathering of data and analysis by experts. This evidence-based approach helped to validate a new standard that is now being adopted across Europe. Our way forward is based on international data and evidence from previous volcanic ash incidents, new data collected from test flights and additional analysis from manufacturers over the past few days."
In addition the CAA has told airlines they must conduct their own risk assessment, develop operational procedures to address any remaining risks, and has made it mandatory to carry out ash damage inspection before and after each flight and to report any ash incidents.
The UK's NATS predicts that Europe's experience is one from which significant lessons will be learned, and that the new standards adopted in the last few days will not be the only ones to emerge from the experience. For example, further tests are proceeding to examine whether the standard of applying a 60nm (111km) buffer zone around a contaminated area could safely be halved.
But although NATS' general manager for strategy and investment Alex Bristol reckons that Europe's decision on changes might, in a perfect world, have been delivered about 24h earlier, the process of determining an acceptable level of risk could not have been circumvented.