Massive miscalculation makes LHC safety assurances invalid

It just gets worse for CERN and its attempts to reassure us that the Large Hadron Collider won’t make mincemeat of the planet.

It’s beginning to look as if a massive miscalculation in the safety reckonings means that CERN scientists cannot offer any assurances about the work they’re doing.

In a truly frightening study, Toby Ord and pals at the University of Oxford say that “while the arguments for the safety of the LHC are commendable for their thoroughness, they are not infallible.”

When physicists give a risk assessment, their figure is only correct if their argument is valid. So an important questions is then: what are the chances that the reasoning is flawed?

Ord and co say that roughly one in a 1000 scientific papers have to be withdrawn because of errors. And errors are by no means unknown among particle physicists, even those doing safety calculations

The Oxford team points to a 1999 study of the risk of a “dangerous event” at the Relativistic Heavy Ion Collider at the Brookhaven National Laboratory in New York state. This study calculated that the chance of such an event was tiny: 2 x 10^-9.

On the strength of this study, the Collider was deemed safe and operated for five years before a serious error was found in the calculations. (It turned out that the actual risk was closer to 10^-12 but the important point is that there was an error).

The problem is compounded when the chances of a planet-destroying event are deemed to be tiny. In that case, these chances are dwarfed by the chances of an error in the argument. “If the probability estimate given by an argument is dwarfed by the chance that the argument itself is flawed, then the estimate is suspect,” say Ord and co.

Nobody at CERN has put a figure on the chances of the LHC destroying the planet. One study simply said: “there is no risk of any significance whatsoever from such black holes”.

Which means we are left with the possibility that their argument is wrong which Ord reckons conservatively to be about 10^-4, meaning that out of a sample of 10,000 independent arguments of similar apparent merit, one would have a serious error.

Of course, this doesn’t mean that the LHC is dangerous, only that there is no reasonable assurance of safety which, as Mark Buchanan writing in New Scientist this week says, is not the same thing at all.

Even still, when it comes to the lives of 6.5 billion people, we need to have better assurances than this.

Ref: arxiv.org/abs/0810.5515: Probing the Improbable: Methodological Challenges Risks with Low Probabilities and High Stakes

[Via New Scientist]

Update: as an aid to the discussion here is a link to CERN’s analysis of the safety of LHC collisions, which was published last year.