Forecasting Seismic Waveforms: A Deep Learning Approach for Einstein Telescope
This work addresses seismic forecasting for gravitational wave observatories, potentially aiding in noise mitigation and control, but it is incremental as it applies a known deep learning paradigm to a specific domain.
The paper tackles seismic waveform forecasting for gravitational wave detectors like the Einstein Telescope by introducing SeismoGPT, a transformer-based model that learns temporal and spatial dependencies from data, achieving accurate short-term forecasts with performance degrading over longer horizons as expected.
We introduce \textit{SeismoGPT}, a transformer-based model for forecasting three-component seismic waveforms in the context of future gravitational wave detectors like the Einstein Telescope. The model is trained in an autoregressive setting and can operate on both single-station and array-based inputs. By learning temporal and spatial dependencies directly from waveform data, SeismoGPT captures realistic ground motion patterns and provides accurate short-term forecasts. Our results show that the model performs well within the immediate prediction window and gradually degrades further ahead, as expected in autoregressive systems. This approach lays the groundwork for data-driven seismic forecasting that could support Newtonian noise mitigation and real-time observatory control.