Abstract Seismometers have been shown to record ground motion caused by acoustic signals from aircraft. These signals have characteristic time–frequency signatures because of the Doppler effect of the aircraft’s motion and the mechanical features that propel the aircraft. We use a set of 303 seismometers deployed in central Alaska in February and March 2019 to estimate a set of flight parameters for a set of 1216 known flights, from 48 aircraft types. The flight parameters include the closest time, closest distance, and speed of the aircraft, as well as the acoustic velocity (sound speed) and a set of frequencies characteristic of the aircraft and its mode of flight. We estimate flight parameters for a subset of 1193 seismic recordings to reveal the frequency signatures of different aircraft types. We demonstrate how modes of flight, such as a propeller’s rotational rate, alter these signatures. These data demonstrate the potential of passive seismometers and acoustic sensors to classify the type of aircraft, in addition to its timing, speed, and distance from the sensor. Our method could be applied in a straightforward manner to other data sets, augmented with machine-learning approaches, and integrated within operational settings.
Seppi et al. (Wed,) studied this question.