Planetary seismic data contain key information on understanding planetary interiors but are plagued by a considerable amount of transient strong disturbances related to harsh environments and spacecraft components. This issue is particularly evident in Apollo passive seismic data, which encounter a broad range of interference types and complexities. Consequently, these disturbances present substantial challenges for analyzing low‐frequency components and conducting related studies. In this study, we propose a highly automated workflow for the accurate detection of strong transient disturbances in Apollo seismic data. Our method leverages deep learning techniques to identify disturbances and uses zero‐muting or Lomb–Scargle periodogram analyses to enable robust low‐frequency signal exploration. We first evaluate the proposed method on synthetic datasets to verify its feasibility from a theoretical perspective. Then, by applying this method to real Apollo seismic recordings, we successfully identify long‐period temperature‐related signals that were previously obscured by strong transient disturbances without relying on clustering or stacking. Furthermore, we demonstrate the potential of this approach by reanalyzing large lunar seismic events and continuous datasets over extended periods, which may contain critical low‐frequency information. In addition, we introduce a graphical interface for manual annotation of various transient disturbances, facilitating the training of deep learning models and refining model predictions. Our study provides an effective and automated approach for detecting and removing strong transient disturbances in Apollo seismic data, significantly improving the practicality of their low‐frequency component analysis, thereby contributing to the search for valuable low‐frequency signals such as free oscillations and gravitational waves from the current and future planetary seismic data.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.