Forecasting the 2016–2017 Central Apennines Earthquake Sequence With a Neural Point Process
Point processes have been dominant in modeling the evolution of seismicity for decades, with the epidemic‐type aftershock sequence (ETAS) model being most popular. Recent advances in machine learning have constructed highly flexible point process models using neural networks to improve upon existing...
Saved in:
| Published in: | Earth's future 2023-09, Vol.11 (9), p.n/a |
|---|---|
| Main Authors: | , , |
| Format: | Article |
| Language: | eng |
| Subjects: | |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Point processes have been dominant in modeling the evolution of seismicity for decades, with the epidemic‐type aftershock sequence (ETAS) model being most popular. Recent advances in machine learning have constructed highly flexible point process models using neural networks to improve upon existing parametric models. We investigate 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 and we show how this model can forecast earthquakes above a target magnitude threshold. We first demonstrate that the neural model can fit synthetic ETAS data, however, requiring less computational time because it is not dependent on the full history of the sequence. By artificially emulating short‐term aftershock incompleteness in the synthetic data set, we find that the neural model outperforms ETAS. Using a new enhanced catalog from the 2016–2017 Central Apennines earthquake sequence, we investigate the predictive skill of ETAS and the neural model with respect to the lowest input magnitude. Constructing multiple forecasting experiments using the Visso, Norcia and Campotosto earthquakes to partition training and testing data, we target M3+ events. We find both models perform similarly at previously explored thresholds (e.g., above M3), but lowering the threshold to M1.2 reduces the performance of ETAS unlike the neural model. We argue that some of these gains are due to the neural model's ability to handle incomplete data. The robustness to missing data and speed to train the neural model present it as an encouraging competitor in earthquake forecasting.
Plain Language Summary
For decades, the Epidemic‐Type Aftershock Sequence (ETAS) model has been the most popular way of forecasting earthquakes over short time spans (days/weeks). It is formulated mathematically as a point process, a general class of statistical model describing the random occurrence of points in time. Recently the machine learning community have used neural networks to make point processes more expressive and titled them neural point processes. In this study we investigate whether a neural point process can compete with the ETAS model. We find that the two models perform similarly on computer simulated data; however, the neural model is much faster with large data sets and is not hindered if there is missing data for smaller earthquakes. Most earthquake catalogs contain missing data due to var |
|---|---|
| ISSN: | 2328-4277 2328-4277 |