Abstract

Seismic data is essential for studying planetary interiors and dynamics, yet acquiring high-quality recordings in harsh extraterrestrial environments is far more challenging than on Earth. Despite being collected over five decades ago (1969-1977), the Apollo seismic data is the only available source of lunar seismic data, continuing to provide valuable insights into the Moon’s interior. When AI meets the Apollo seismic dataset, what secrets will be unlocked? Here, I present some fascinating results from applying deep learning to this legacy dataset, including the discovery of Lunar’s long-period signal, the search for the planetary free oscillation, and a tool for automatic disturbance detection and mitigation. These findings demonstrate how modern techniques can extract new knowledge from historic recordings, opening a new window into understanding planetary interiors and revolutionising our approach to lunar seismology.