Machine learning and marsquakes: a tool to predict atmospheric-seismic noise for the NASA InSight mission

The SEIS experiment on the NASA InSight mission has catalogued hundreds of marsquakes so far. However, the detectability of these events is controlled by the weather which generates seismic noise. This affects the catalogue on both diurnal and seasonal scales. We propose to use machine learning methods to fit the wind, pressure and temperature data to the seismic energy recorded in the 0.4-1 Hz and 2.2-2.6 Hz bandwidths to examine low and high frequency event categories respectively. We implement gaussian process regression and neural network models for this task. This approach provides the relationship between the atmospheric state and seismic energy. The obtained seismic energy estimate is used to calculate signal to noise ratios (SNR) of marsquakes for multiple bandwidths. We can then demonstrate the presence of low frequency energy separate to the noise level during several events predominantly categorised as high frequency, suggesting a continuum in event spectra distribution. A method to detect marsquakes is implemented whereby a variable threshold is calculated for the subtraction of the predicted noise from the observed signal. This algorithm finds 32 previously undetected marsquakes, with another 34 possible candidates. Furthermore, an analysis of the detection algorithm's variable threshold provides an empirical estimate of marsquake detectivity. This suggests that the largest events would be seen almost all the time, the median size event 45-50% of the time and smallest events 5-20% of the time.