Towards a theoretical determination of the geographical probability distribution of meteoroid impacts on Earth

TL;DR

This paper introduces a novel numerical method called Gravitational Ray Tracing to predict the impact probability distribution of meteoroids on Earth, revealing complex patterns and applying it to real impact events and fireball data.

Contribution

The paper presents a new GRT technique for calculating impact probabilities on planets, with initial testing and application to Earth's impact events and fireball data.

Findings

Impact probability peaks at 60-90° from the apex, especially at midnight.

No clear spatial or temporal pattern for Tunguska and Chelyabinsk impacts.

The method predicts impact speed distribution with reasonable accuracy.

Abstract

Tunguska and Chelyabinsk impact events occurred inside a geographical area of only 3.4\% of the Earth's surface. Although two events hardly constitute a statistically significant demonstration of a geographical pattern of impacts, their spatial coincidence is at least tantalizing. To understand if this concurrence reflects an underlying geographical and/or temporal pattern, we must aim at predicting the spatio-temporal distribution of meteoroid impacts on Earth. For this purpose we designed, implemented and tested a novel numerical technique, the "Gravitational Ray Tracing" (GRT) designed to compute the relative impact probability (RIP) on the surface of any planet. GRT is inspired by the so-called ray-casting techniques used to render realistic images of complex 3D scenes. In this paper we describe the method and the results of testing it at the time of large impact events. Our findings suggest a non-trivial pattern of impact probabilities at any given time on Earth. Locations at $60-90\deg$ from the apex are more prone to impacts, especially at midnight. Counterintuitively, sites close to apex direction have the lowest RIP, while in the antapex RIP are slightly larger than average. We present here preliminary maps of RIP at the time of Tunguska and Chelyabinsk events and found no evidence of a spatial or temporal pattern, suggesting that their coincidence was fortuitous. We apply the GRT method to compute theoretical RIP at the location and time of 394 large fireballs. Although the predicted spatio-temporal impact distribution matches marginally the observed events, we successfully predict their impact speed distribution.