How to defuse Earth impact threat announcements

Abstract: Summary: In the past decade both scientists and laymen have probably heard at least once through the newspapers, TV and Internet that a new asteroid has been discovered with non-zero (sometimes "high") probability of collision with the Earth in the near future. Since early 2000's, such probabilities are routinely calculated by two impact monitoring systems (one in the US, the other in Europe) on preliminary orbits of newly discovered Near Earth Asteroids (NEAs), and are regularly updated whenever additional new astrometric observations for each threatening object become available. A typical pattern is that as the orbit becomes more precisely determined, impact probability often increases initially, but then turns around and decreases until it falls to zero, or some very low number. In the present study we define a probability measure which provides a simple tool to evaluate from the very beginning the chance that the impact probability calculated by monitoring systems for a threatening object will reach unity at the end of the overall process of orbit refinement. We stress that this chance is independent of the specific (fluctuating) value of the probability calculated by the monitoring systems. In a precise sense, which should be clear throughout the paper, the concrete impact probability of a newly discovered asteroid is not that given by the monitoring systems.