Abstract
The explosion at the Ingolstadt oil refinery was widely recorded at seismic and infrasound stations deployed throughout Central Europe, to distances of several hundred to a thousand kilometres. This study focuses on the wealth of data recorded at infrasound stations in Central and Eastern Europe, while from the many detecting seismic stations within 400 km range, only seismic and seismo-acoustic arrivals at the close-in Gräfenberg array are considered here. Most of the infrasound stations are acoustic arrays enabling us to apply array processing techniques to determine relevant wave field parameters, such as backazimuth and slowness (resp. trace velocity). These parameters not only confirm the source direction, but also put constraints on the observed arrivals’ propagation modes. Wave field parameters suggest that we observe tropospheric arrivals to about 150 km and stratospheric and/or thermospheric returns for longer distances. 1D, 2D and 3D ray tracing predict tropospheric arrivals to westerly directions up to distances of 100 km, beyond which only thermospheric returns are obtained azimuth-independent beyond 250–300 km. Stratospheric returns do not follow from any of the increasingly complex ray tracing models. Parabolic equation propagation modeling however suggests that in a number of cases stratospheric ducting may be possible. However, neither the tropospheric seismo-acoustic arrivals at the Gräfenberg array nor the various arrivals at IMS station IS26 could be modeled. Therefore, the Ingolstadt explosion along with the observed infrasonic phases provide an excellent test bed to investigate our ability in realistically forecasting atmospheric wave propagation with existing algorithms and available atmospheric models.
Reconstruction of Initial Wave Field of a Nonsteady-State Wave Propagation from Limited Measurements at a Specific Spatial Point Based on Stochastic Inversion