We present an inversion methodology where acoustic observations of infrasound waves are used to update an atmospheric model. We sought a flexible parameterization that permits to incorporate physical and numerical constraints without the need to reformulate the inversion. On the other hand, the optimization conveys an explicit search over the solution space, making the solver computationally expensive. Nevertheless, through a parallel implementation and the use of tight constraints we demonstrate that the methodology is computationally tractable. Constraints to the solution space are derived from the spread (variance) of ERA5 ensemble reanalysis members, which summarize the best current knowledge of the atmosphere from assimilated measurements and physical models. Similarly, the initial model temperature and winds for the inversion are chosen to be the average of these parameters in the ensemble members. The performance of the inversion is demonstrated with the application to infrasound observations from an explosion generated by the destruction of ammunition at Hukkakero, Finland. The acoustic signals are recorded at an array station located at 178 km range, which is within the classical shadow zone distance. The observed returns are assumed to come from stratospheric reflections. Thus, the reflection altitude is also an inverted parameter.

Modelling the spatial distribution of infrasound attenuation (or transmission loss, TL) is key to understanding and interpreting microbarometer data and observations. Such predictions enable the reliable assessment of infrasound source characteristics such as ground pressure levels associated with earthquakes, man-made or volcanic explosion properties, and ocean-generated microbarom wavefields. However, the computational cost inherent in full-waveform modelling tools, such as Parabolic Equation (PE) codes, often prevents the exploration of a large parameter space, i.e., variations in wind models, source frequency, and source location, when deriving reliable estimates of source or atmospheric properties – in particular for real-time and near-real-time applications. Therefore, many studies rely on analytical regression-based heuristic TL equations that neglect complex vertical wind variations and the range-dependent variation in the atmospheric properties. This introduces significant uncertainties in the predicted TL. In the current contribution, we propose a deep learning approach trained on a large set of simulated wavefields generated using PE simulations and realistic atmospheric winds to predict infrasound ground-level amplitudes up to 1000 km from a ground-based source. Realistic range dependent atmospheric winds are constructed by combining ERA5, NRLMSISE-00, and HWM-14 atmospheric models, and small-scale gravity-wave perturbations computed using the Gardner model. Given a set of wind profiles as input, our new modelling framework provides a fast (0.05 s runtime) and reliable (~5 dB error on average, compared to PE simulations) estimate of the infrasound TL.