OBSTransformer: a deep-learning seismic phase picker for OBS data using automated labelling and transfer learning

Abstract

SUMMARY Accurate seismic phase detection and onset picking are fundamental to seismological studies. Supervised deep-learning phase pickers have shown promise with excellent performance on land seismic data. Although it may be acceptable to apply them to Ocean Bottom Seismometer (OBS) data that are indispensable for studying ocean regions, they suffer from a significant performance drop. In this study, we develop a generalized transfer-learned OBS phase picker—OBSTransformer, based on automated labelling and transfer learning. First, we compile a comprehensive data set of catalogued earthquakes recorded by 423 OBSs from 11 temporary deployments worldwide. Through automated processes, we label the P and S phases of these earthquakes by analysing the consistency of at least three arrivals from four widely used machine learning pickers (EQTransformer, PhaseNet, Generalized Phase Detection and PickNet), as well as the Akaike Information Criterion (AIC) picker. This results in an inclusive OBS data set containing ∼36 000 earthquake samples. Subsequently, we use this data set for transfer learning and utilize a well-trained land machine learning model—EQTransformer as our base model. Moreover, we extract 25 000 OBS noise samples from the same OBS networks using the Kurtosis method, which are then used for model training alongside the labelled earthquake samples. Using three groups of test data sets at subglobal, regional and local scales, we demonstrate that OBSTransformer outperforms EQTransformer. Particularly, the P and S recall rates at large distances (>200 km) are increased by 68 and 76 per cent, respectively. Our extensive tests and comparisons demonstrate that OBSTransformer is less dependent on the detection/picking thresholds and is more robust to noise levels.