Identification of Rare Multiple Core‐Mantle Boundary Reflections PmKP Up To P7KP With Deep Learning.

The core‐mantle boundary (CMB) marks the most dramatic changes in physical properties within the Earth, and plays a critical role in the understanding of the Earth's dynamics. PmKP waves are seismic phases that reflect (m − 1) times under the CMB and are useful for studying the complex CMB structure. We present an automated workflow for detecting PmKP phases using multi‐station records from global seismic stations. We employ a novel sampling method to extract PmKP waveforms into a 2‐D matrix. Two deep neural networks are then utilized for initial phase detections and subsequent slowness validations. Numerous PmKPab (3 ≤ m ≤ 7) and their CMB diffracted signals were identified for deep earthquakes (magnitude >6) occurred from 2000 to 2020, including diffracted P7KPab waves with diffraction lengths of nearly 20°. Our approach significantly improves the efficiency of PmKP phase identification and holds the capability to detect other weak core phases, such as PKiKP. Plain Language Summary: Seismic waves provide invaluable information about the Earth's core when sampling the core‐mantle boundary (CMB) complex structures. However, adequate sampling of this region remains a challenge due to the uneven distribution of earthquakes and seismic stations. PmKP is a seismic phase that reflects multiple times (m − 1) within the CMB, and even though it has been considered elusive, it offers an opportunity to overcome the sampling limitation caused by the earthquake‐station distribution and enhance the CMB seismic coverage. Here, we propose a new automated approach to detect unprecedent high reverberations of PmKP. Our method utilizes a combination of two deep neural networks to analyze seismic records from stations distributed globally. By applying this approach to seismic records between 2000 and 2020, we successfully detected up to six underside reflections of PmKP phases and their CMB diffracted waves. Our method enhances the efficiency of identifying PmKP phases and has the potential to be applied to other core seismic phases. Key Points: We propose a workflow based on neural networks to detect rarely observed core seismic phases and their diffractionsOur method successfully detects PmKP up to six underside reflections on the core‐mantle boundary (CMB), increasing the seismic sampling of the CMBThe new workflow can be used for future automated identification of other rare or weak seismic phases [ABSTRACT FROM AUTHOR]