TH Thursday Seminar

A critical look at safety analyses for RHIC and ALICE

by A. KENT (DAMTP, Cambridge)

→ Europe/Zurich
TH Auditorium (CERN)

TH Auditorium

CERN

Description
Recent analyses by Busza et al. (BJSW) and Dar et al. (DDH) have estimated upper bounds on the risk involved in the Brookhaven relativistic heavy ion collider (RHIC) experiments and concluded that it is completely negligible. I examine some weaknesses in these analyses. The most obvious of these is that they estimate only the probability of catastrophe rather than the expected cost, the variance, and other relevant quantities. That these can be independently significant is well illustrated by the DDH analysis, which produces the not entirely reassuring estimate that the expected number of fatalities caused by RHIC cannot exceed 120, or on worst case assumptions 12000. It should be stressed that these estimates are conservative upper bounds. Further analyses could possibly reduce the figures by many orders of magnitude. (Indeed, some of the figures suggested by BJSW are much lower than DDH's: however, their most credible estimates are not.) It must also be stressed, though, that the figures ignore both the cost in future lives and the cost of risk itself, and thus understate the bounds on the true cost by many orders of magnitude. In summary, it seems hard to conclude that the analyses presented to date demonstrate beyond reasonable doubt that the novel safety risks of the RHIC and ALICE experiments are negligible.