Abstract

Wind measurements from landed missions on Mars are vital to characterise the near surface atmospheric behaviour on Mars and improve atmospheric models. These winds are responsible for aeolian change and the mixing of dust in and out of the atmosphere, which has a significant effect on the global circulation. The NASA InSight mission successfully recorded wind data for around 750 sols. The seismometer, however, recorded nearly continuous data for around 1400 sols. The dominant source of energy in the seismic data is in fact due to the wind. To this end, we propose a machine learning model, dubbed WindSightNet, to map the seismic data to wind speed and direction. This converts the atmospheric information in the seismic data into a physically meaningful wind signal which can be used for analysis. We retrieve wind data from the entire period the seismometer was recording which enables a comparison of the year-to-year wind variations at InSight. The continuous nature of the dataset also enables the extraction of information on baroclinic activity at long periods and the periodicity of observed convective cells. A data science based metric is proposed to provide a quantification of the year-to-year differences in the wind speeds, which highlights variations linked to dust activity as well as other transient differences worthy of further study. On the whole, the seismic-derived winds confirm the dominance of the global circulation on the winds leading to highly repeatable weather patterns.