Targeting Temporal Seismicity with a Neural Point Process

Sam Stockman, Dan Lawson, & Maximilian J. Werner

Submitted September 11, 2022, SCEC Contribution #11932, 2022 SCEC Annual Meeting Poster #TBD

Point processes have been the dominant way of describing the evolution of seismicity for a while, with the Epidemic Type Aftershock Sequence (ETAS) model being the most successful. Recent advances in the field of machine learning have constructed highly flexible point process models using neural networks to seek improvements on existing parametric models. We conduct an investigation into the whether these flexible point process models can be applied to short term seismicity forecasting by extending an existing temporal neural model to the magnitude domain. Furthermore, we show how this model can be used to forecast earthquakes above a target magnitude threshold. We first find that the neural model can learn to capture synthetic ETAS data, however, requiring less computational time due to the fact that it is not dependent on the full history of the sequence. On data from a new high-resolution earthquake sequence in the Central Apennines we investigate the sensitivity of both ETAS and the neural model with respect to the lowest magnitude passed into each. We find that both models perform similarly at previously explored thresholds but as the magnitude is lowered the performance of ETAS reduces unlike that of the neural model. We argue that some of the gains made by the neural model are due to its ability to handle incomplete data. Since neither model has an overall forecasting advantage over the other, our work suggests that there is not enough signal in 2-dimensional data to give a neural model an advantage over ETAS, but does offer a more efficient way to model data incompleteness. Our work also stresses the need for considering temporal incompleteness when using new machine learning generated earthquake catalogs.

Citation
Stockman, S., Lawson, D., & Werner, M. J. (2022, 09). Targeting Temporal Seismicity with a Neural Point Process. Poster Presentation at 2022 SCEC Annual Meeting.


Related Projects & Working Groups
Earthquake Forecasting and Predictability (EFP)