Physics-informed deep learning for crustal stress
Integrating observations and physical models to estimate plate-boundary stress at high resolution.
Research Overview
Associate Professor at the Earthquake and Volcano Research Center, Graduate School of Environmental Studies, Nagoya University. Research on plate-boundary crustal deformation, earthquake potential, and volcanic activity, combining observations and modelling.
Crustal deformation observation & modeling
Earth tides
Ionospheric effects (GNSS)
Volcano monitoring
InSAR / GNSS time series
Machine learning × geophysics
International collaboration
Approach
- Integrate satellite (InSAR) and GNSS data to estimate deformation and stress fields.
- Combine physics-based models and observations to infer plate-boundary behavior and inland earthquake potential.
- Monitor short-term volcanic activity and feed findings into risk assessment.
- Quantify subsidence in urban/coastal areas and assess impacts on infrastructure.
Recent Topics
Slow-slip vs. tremor moment rates
Comparing moment rates of short-term SSEs and tectonic tremors to understand slow-slip speed characteristics.
Jakarta subsidence mapping
Quantifying subsidence from satellite time series to inform future risk.