Global Distribution of Low Frequency Family Marsquakes From Deep Learning-Based Polarization Estimation

The deployment of the seismometer on Mars has recorded thousands of marsquakes. Accurately locating these events is crucial for understanding Mars' internal structure and geological evolution. With only a single station, determining the location, especially the accurate back-azimuth, is more challenging than on Earth. Deep learning, being data-driven, can learn patterns of complex noise that are difficult for traditional methods to model, making it promising for improving back-azimuth estimation of marsquakes. However, challenges arise when applying deep learning to estimate marsquake polarization due to the limited quantity and low quality of the data. In this study, we assumed the background noise remains relatively stable around the P-wave arrivals and trained a deep learning model to learn noise patterns preceding marsquakes. Then we combined the trained model with Sliding Window Inference and Featured-Training (SWIFT) to handle the high uncertainty in P phase picking to predict polarizations of low frequency family marsquakes. As a result, we have further improved the localization of marsquakes by relocating 56 events, including 7 Quality C events with epicentral distances over 90°. For two Martian impact events with ground-truth locations, S1000a and S1094b, our deviations are only ~5° and ~3°. Our results reveal a new clustered seismicity zone around compressional structures in Hesperia Planum, including 7marsquakes with magnitudes from 2.5 to 3.6. Marsquakes are also widely distributed along the northern lowlands, dichotomy boundary, and higher latitude southern highlands, suggesting a globally distributed character. Our renewed marsquake location brings new insight to the tectonic interpretation of marsquakes.