Soapbox Post

Everything we do has some element of risk.  The real issue is whether the costs of minimizing or eliminating the risks outweigh the benefit. The recent closure of airspace over Europe dramatically illustrates this point.  Rather than perform an analysis of what concentration of volcanic ash will cause a flame out of an engine on a commercial jet aircraft, the authorities assumed that arbitrarily small concentrations of ash would lead to a significant hazard to air travel. 

The same attempt to eliminate low probability events has also characterized the UK (and other) government responses to recent terrorist incidents.  According to ICAO there have been 11.5 x10⁹ commercial passenger airplane departures in the 10 year period from 1999 to 2009.  About 500 of these passengers have lost their lives from an airplane terrorist incident.  That gives a probability of dying from a terrorist incident on any given flight as about 1 in 20 x 10⁶.  The same low probability could have been deduced from data spanning the previous 2 decades. 

As a response to the failed Christmas day airplane bombing the authorities are now introducing Compton Backscatter (body scanner) screening systems at airports.  Although the manufacturers claim that the exposure from these machines is less than 0.25 microSv, my calculations from the contrast in published images and estimates of the Inter-Agency Committee on Radiation Safety would suggest that the exposure could be as high as a few microSv. According to the linear no threshold theory (LNT) model used in radiation protection published by the ICRP, the probability of dying from cancer caused by a Compton Backscatter screening is about 1 in 20 x 10⁶ per microSv.

The probability of dying from a terrorist attack and from radiation from the screening are therefore of the same magnitude. The only difference is that in one case the cause of death is obvious, and in the other case it cannot be proved since, according to LNT, the probability of dying from background radiation at an average level, approximately 1 mSv, is 1000 times higher, not to mention the radiation from the flight which could be between 3 and 6 microSv per hour. 

There is also the possibility of receiving a very large dose should the mechanical scanning systems in these machines malfunction.  According to the standard governing these machines, NS 43.17, they should shut down if the dose were to reach the administrative limit of 0.25 mSv.  That means the system has less than 1 msec to respond. If the fail-safe mechanism doesn’t work as advertised (as we have seen recently with a Blow Out Preventer in the Gulf of Mexico) it might take about 10-20 seconds to manually shut down the system.  The person being screened will then receive a very large dose, in the range of a few Sv.  It is hard to believe that any mechanical scanning system used for universal screening in almost continuous operation would have a failure rate so low that this would not be a concern.

The real question is whether it is worth putting any effort into preventing these low probability events.  It is interesting to note that many of the recent airport screening measures were put in place as a response to incidents, such as the shoe bomber or the underwear bomber, where no one was hurt or killed.  It is highly unlikely that there would have been mass casualties even if the explosives in these two incidents had fully detonated. There might have been some injuries to nearby passengers and depressurization.  Although not pleasant, depressurization of a commercial aircraft does not lead to fatalities.

Absolute safety can never be assured, so is the cost, the inconvenience and the possibility of very high exposures really worth the benefit?  The risk of being killed in a lightning strike in the United States in any year is 1 in 5x10⁶, about 4 times greater than the risk of being killed by a terrorist in an airplane, yet the Federal authorities have not yet mandated everyone walk around in a Faraday cage, even during a storm!