Semi-automatic meteoroid fragmentation modeling using genetic algorithms

Abstract

Context. Meteoroids are pieces of asteroids and comets. They serve as unique probes to the physical and chemical properties of their parent bodies. We can derive some of these properties when meteoroids collide with the atmosphere of Earth and become a meteor or a bolide. Even more information can be obtained when meteoroids are mechanically strong and slow enough to drop meteorites. Aims. Through physical modeling of bright meteors, we describe their fragmentation in the atmosphere. We also derive their mechanical strength and the mass distribution of the fragments, some of which may hit the ground as meteorites. Methods. We developed a semi-automatic program for meteoroid fragmentation modeling using parallel genetic algorithms. This allowed us to determine the most probable fragmentation cascade of the meteoroid, and also to specify its initial mass and velocity. These parameters can be used in turn to derive the heliocentric orbit of the meteoroid and to place constraints on its likely age as a separate object. Results. The program offers plausible solutions for the majority of fireballs we tested, and the quality of the solutions is comparable to that of manual solutions. The two solutions are not the same in detail, but the derived quantities, such as the fragment masses of the larger fragments and the proxy for their mechanical strength, are very similar. With this method, we would like to describe the mechanical properties and structure of both meteoroids belonging to major meteor showers and those that cause exceptional fireballs.