What question did this study set out to answer?

This research aims to estimate the yield of surface explosions using regional infrasound signals and machine learning techniques.

Estimating Surface Explosion Yield From Single‐Channel Infrasound Signals at Regional Distances With Random‐Forest Regression

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial estimates explosion yield from infrasound signals, suggesting improved monitoring methods.

This research aims to estimate the yield of surface explosions using regional infrasound signals and machine learning techniques.
Generated full-waveform regional infrasound data simulating sources between 20 and 500 km away.
Used scaled-yield source models and normal-mode wave propagation for realistic simulations.
Applied a random forest regressor on ML features from the infrasound data to estimate explosion yields from 1 to 1,500 tonnes of TNT.
The random forest model performed well on five surface shots from the Humming Roadrunner experiment.
Yield estimates matched favorably with existing methods despite the training on synthetic data.
The approach enhances the effectiveness of explosion monitoring using infrasound signals.

Awender et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark