Acoustic-wave detection from man-made sources like explosions and artillery is of interest both for civilian and military purposes. Infrasound propagation from surface sources is controlled by a complex interplay between source location, winds, atmospheric attenuation, and topography. The seasonal and stochastic variability of stratospheric and tropospheric winds is known to play an important role in the detectability of infrasound on the ground. In particular, large wind-intensity variations occur between summer and winter months. However, the lack of high-quality observational datasets with good temporal coverage throughout the year limits our understanding of the correlations between source characteristics, range-dependent atmospheric properties, and topography. Here, we take advantage of an extensive set of artillery exercises, conducted by the Norwegian Armed Forces in southern Norway throughout 2020, to constrain the detectability and wave properties at local distances. Up to 70 km distance, signals are generally observed when the atmospheric models include stronger lower-tropospheric winds (1-5km altitude) blowing in the direction of propagation. When cross winds dominate the wind field, low-amplitude infrasound arrivals are still observed in the acoustic shadow zone while not predicted by ray-tracing simulations, highlighting both model and propagation uncertainties introduced by small-scale wind heterogeneities and diffraction effects.
Artillery shots excite acoustic waves travelling over at large distances. Characterizing infrasound sources is crucial for civilian and military purposes. We perform an extensive investigation of military exercises in Norway to derive the main property of artillery infrasound.