A Reanalysis of Anomalous Acoustic Signals Recorded by the CERI Seismic Network on 28 January 2004

Abstract

We reanalyzed the arrival-time data from an unknown source recorded on 28 January 2004 (UTC) by 20 stations of the Center for Earthquake Research and Information (CERI) seismic network. As discussed by Lin and Langston (2006), the recorded signals were probably generated by an acoustic source, which they modeled as an explosion. As shown here, the data are better explained by a fireball source. For our analysis we used the inversion software developed by Pujol et al. (2005). Assuming that the speed of sound ( c ) in air is constant and known and the fireball trajectory is a straight line, this software allows the determination of the velocity of the fireball ( v ), assumed constant, and the following parameters used to describe the trajectory: the horizontal coordinates ( x , y ) of the end point ( P ) and the corresponding time ( t ), the azimuth of its horizontal projection (\batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \({\varphi}\) \end{document}), and the angle with the vertical (\batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \({\vartheta}\) \end{document}) (figure 1). The inversion software was applied to a fireball recorded in northeast Arkansas in 2003, to two fireballs recorded in Japan, and to another one recorded in the Czech Republic, and comparison of the theoretical and observed isochrones shows that the software performs very well in spite of several simplifying assumptions made in the model for the fireball (Pujol et al. 2005, 2006). It was noted however, that the fireball's velocity cannot be determined uniquely, as it trades off with t . The other parameters, however, are more robust. Moreover, under appropriate conditions the angles \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \({\vartheta}\) \end{document} and \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \({\varphi}\) \end{document} can be determined from the isochrones. If they have a clear axis of symmetry, then it determines \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \({\varphi}\) \end{document}, while the separation between isochrones is related to \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \({\vartheta}\) \end{document} by the following relation: \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \[ \mathrm{cos}{ }{\vartheta}{\approx}c\frac{{\Delta}t}{{\Delta}x^{{^\prime}}}\] \end{document}(1) where Δ x ′ is the distance between two given isochrones along the projection of the …