Chapman, Clark R. (2003), Predicting and Comprehending Asteroid Impacts, Computing Science and Statistics, 35, I2003Proceedings/ChapmanClark/ChapmanClark.presentation.pdf ,
Asteroids and comets have struck our planet, with devastating consequences, during geological history. Very improbably during our lifetimes, one may strike again with sufficient energy to destroy civilization as we know it or -- even more improbably -- erase the human species from planet Earth. This "impact hazard" has a number of unusual features that should interest experts in statistics. As an astronomer, not a mathematician, I discuss several of these issues.
First, the prediction of impacts might, at first glance, seem to be a "particle-in-a-box" theoretical problem, and then a straightforward "missile impact" problem when an asteroid is detected that is heading our way. But the reality is different. The orbital dynamical aspects of the problem, in the context of the telescopic observations that discover and track the potential impactors, are highly non-intuitive.
Second, the impact hazard challenges ordinary citizens, and more scientifically educated people as well, to grabble with the essence of very low probability but high consequence events. It is like the lottery in reverse. Whether society should take this hazard seriously, and invest in practical technological approaches to mitigating the hazard, depends on effective communication between technical experts and the lay public and decision makers who set national priorities. Especially in the context of 9/11 terrorism, an understanding of how ordinary people relate to statistics in the context of potential danger is a central issue for our society.