Networks of low-frequency acoustic detectors, originally built for other purposes, are now finding a new and unexpected use in detecting objects that enter our atmosphere. The orbit of Earth through the solar system passes through much solid particle debris, including pieces of material from both comets and asteroids. We call these arriving particles "meteoroids." The asteroidal and cometary sources have a wide variety of properties, so meteoroids can arrive from very different orbits and belong to one of several types of observed materials. They can be iron, rocky stones, very weak stones (Carbonaceous chondrites) or there are two brands of very weak cometary material as well. This debris can be either very small or very large or have a large range of possible sizes, depending on the source and how long the material has been orbiting in space free from its source and other factors. This material can also have a large range of possible entry speeds and densities.

As Earth moves about the Sun, it acts as a tiny dust mop sweeping up this material which can strongly interact with the atmosphere at very great heights (above 60 miles). On occasion these larger and brighter meteors and fireballs or bolides (the name of the atmospheric phenomena) travel at high speeds to collide with Earth's surface and possibly even produce an extensive crater. This delivery of meteorite samples (the ponderable pieces that reach the Earth intact), originating on other worlds beyond our own, provides a means of studying our own origins as well.

The interaction of these meteoroids with the atmosphere is very strong partly due to the very high speed at entry and partly due to the compressibility of the atmosphere. The entry speed compared to the speed at which sound waves travel, called the Mach number, typically can range from 50 - 300. For comparison, a typical Mach number of a commercial or military supersonic jet is less than 3. As a direct consequence of this high speed, an explosion is generated along a cylindrical path about the entry trajectory. This deposition of energy along the path constitutes an explosion whose characteristic scale is called the "blast wave radius", which delineates the size of the region in which an explosion has occurred. For large meteoroids capable of penetrating the atmosphere down to heights
where a shock wave is formed, this scale can range from a minimum of 10 meters (in order to be recorded at ground level) to many kilometers in length. For comparison, the typical size scale of the sound source in ordinary thunder is about 2 or 3 m.

Sounds that emanate from such sources in the atmosphere can have very large amplitudes, even great enough to break glass windows at close range. Frequencies of such sources can be low enough so that the peak energy is below the range of audible sound waves, which we call "infrasound." As the
blast wave radius increases we find that these frequencies become progressively lower. For the famous Siberian meteorite explosion (Tunguska) of 1908, ultra-low sound frequencies of 1/60 Hertz (corresponding to a period of about 1 minute) were observed at great distances from the entry trajectory. As these signals propagate through the atmosphere, the ambient
temperature and winds aloft can bend the signals away from straight-line paths, i.e., refraction. They can also be diffracted and scattered as well since this is a wave phenomenon. We now know empirically how to relate the period at maximum amplitude of the sound waves to the source energy. For the blast wave radius values quoted above, source energies range from 0.00001 kt (1/100 of a ton of TNT) to 10 Mt (megaton) of TNT equivalent (1 kt = 4.186×1012 Joules). For comparison, the nuclear weapons dropped in Japan in WWII produced explosions of about 15 kt.

Over the past few years we have observed a number of these very large bolides over a very large energy range. From these data we have been able to locate the sources and calculate the frequency of occurrence of these large bodies at the Earth in a year. The observations at arrays of sensors on the ground using low frequency microphones, separated horizontally by distances of a few hundred meters to a few km typically, can be used to determine both the angular great circle distance of the arrival as well as the elevation angle of the signals. This allows us to uniquely locate these sources in three-dimensional space within the atmosphere within certain errors. For example at a range of 3350 km, a bolide of about 0.2 kt was readily recorded infrasonically even as long ago as 1965. Also, from such data we can estimate that the frequency of occurrence of rocky type meteoroids for a energy of 15 kt is about once per year over the globe. At the energy of Tunguska (10 Mt), it is about once every 120 years and this event last occurred about 93 years ago. Corresponding to an energy of
0.1 kt for example, we find a value and associated uncertainty of about 30 ± 9 large bolides/year and this value continues to increase as the source energy decreases and vice versa.

Douglas ReVelle is at the Los Alamos National Laboratory. This paper is based on the lay-language version of paper 2pPAa4 at the 141st ASA Meeting, Chicago, IL