arXiv:2606.02912v1 Announce Type: cross Abstract: Forecasting seismic waveforms beyond observed data remains challenging due to the nonlinear, dispersive, and multi-scale nature of seismic wave propagation. In this work, we introduce \textsc{SeismoGPT}, a transformer-based autoregressive model designed to forecast three-component seismic waveforms directly in the time domain. Forecasting is formulated as a physically constrained continuation problem in which the model receives waveform context beginning at the P-wave arrival and extending a defined time beyond the S-wave arrival, after which f
Advances in transformer architectures and computational capabilities enable the application of sophisticated AI models to complex time-series data like seismic waveforms.
Accurate seismic forecasting can significantly improve earthquake prediction, natural disaster preparedness, and geological exploration, impacting infrastructure and resource planning.
The ability to forecast three-component seismograms directly in the time domain using AI means more precise and earlier warnings for seismic events could become feasible.
- · Geophysical research institutions
- · Insurance companies (reducing risk)
- · Mining and energy sectors
- · Disaster preparedness agencies
- · Predictive modeling relying solely on traditional methods
- · Regions without access to advanced AI infrastructure
Improved early warning systems for earthquakes and tsunamis.
Reduced economic losses and increased human safety in seismically active regions.
Potential for new regulations and urban planning based on enhanced seismic predictability, altering real estate values and development patterns.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG