The impact hazard.

Abstract

This overview of the impact hazard characterizes the consequences of impacts as a function of meteoroid energy, assesses the probability of death from impacts, compares these risks with those of other natural disasters, and reports on preliminary studies of the public perception of impact risks. Below 10 MT (equivalent TNT) energy there is virtually no risk since few meteoroids penetrate the atmosphere. Between 10 MT and the threshold for global catastrophe impacts are a moderate source of risk, but substantially less so than more common natural disasters such as earthquakes, severe storms, or volcanic eruptions. The greatest hazard is associated with impacts at or a little above a threshold for global catastrophe, where we define a global catastrophe as one that leads to the death of >25% of the Earth's human population. Following the analysis of environmental effects of impacts by Toon et aL (this book) we conclude that this threshold lies between 5x104 and 5x106 MT, with a nominal value of 5x105 MT, corresponding to an average interval between events of about half a million years. Above this threshold the entire world population is at risk from impacts, which are the only known natural disasters capable of killing a substantial fraction of the population or, at still larger energies associated with mass extinction, of threatening the survival of the species. As an extreme example of low-probability but high-consequence disasters, large impacts are without precedent in terms of public perception, but with increasing public awareness, demands may grow for action to deal with this newly-identified hazard. Over the entire range of impact energies from the atmospheric cut-off to above the global threshold, the risk level increases with the energy of the impact. Therefore the most prudent strategies deal with detection and protection against the largest projectiles; any program to mitigate the impact hazard should begin with a comprehensive NEO census such as the proposed Spaceguard Survey.