Title:Recent multi-kiloton impact events: are they truly random?

Abstract:It is customarily assumed that Earth-striking meteoroids are completely random, and that all the impacts must be interpreted as uncorrelated events distributed according to Poisson statistics. If this is correct, their impact dates must be uniformly spread throughout the year and their impact coordinates must be evenly scattered on the surface of our planet. Here, we use a time- and yield-limited sample of Earth-impacting superbolides detected since 2000 to explore statistically this critical though frequently overlooked topic. We show that the cadence of these multi-kiloton impact events is incompatible with a random fall pattern at the 0.05 significance level or better. This result is statistically robust and consistent with the observed distribution of the longitudes of the ascending nodes of near-Earth objects (NEOs). This lack of randomness is induced by planetary perturbations, in particular Jupiter's, and suggests that some of the recent, most powerful Earth impacts may be associated with resonant groups of NEOs and/or very young meteoroid streams. An intriguing consequence of this scenario is that the impact hazard of Chelyabinsk-like objects should peak at certain times in the year.