Sam A. Scivier (he / him)

I am a PhD student in the Department of Earth Sciences at the University of Oxford, developing probabilistic methods to improve uncertainty quantification in geophysics. My PhD is funded by the Oxford‑NERC DTP in Environmental Research. I hold a Master's in Physics from the University of Birmingham. Alongside my studies at Birmingham, I completed internships in quantum computing at D‑Wave Systems (Burnaby, Canada) and Riverlane (Cambridge, UK).

Sam Scivier

Research

I currently work at the intersection of geophysics, data science and machine learning, developing methods for uncertainty quantification, and building open‑source software for environmental applications.

  • My PhD

    I develop probabilistic methods to improve uncertainty quantification in geophysics—in particular relating to estimates of seismic wave speeds in the Earth (seismic velocity models). My core contribution is a Gaussian process-based approach for the probabilistic fusion of overlapping geospatial datasets. I am applying this method in earthquake ground motion prediction to account for inconsistencies between existing seismic velocity models, as well as in other geophysical problems of interest. Through the development of open-source software, I aim to make these methods broadly accessible and applicable to spatial datasets across the geosciences.

  • Broader and future interests

    I am interested in applying physics-based computational, data science, and machine learning methods to tackle challenges across geoscience, aerospace, sustainability, and emerging technologies. My experience spans from algorithm development in quantum computing systems at D-Wave Systems and Riverlane to uncertainty quantification in Earth systems, giving me a unique perspective on computational approaches to complex physical problems. I am particularly drawn to research opportunities that combine rigorous scientific methodology with practical applications in areas such as environmental modelling, aerospace engineering and space systems, sustainable energy systems, and other scientifically impactful domains. I aim to contribute to interdisciplinary teams working on problems with tangible societal impact.

Selected Publications

Full list on Google Scholar →

Open‑Source Software

See more @sscivier on GitHub →
Loading repositories...

Teaching & Outreach

  • Gaussian Processes for Probabilistic Earthquake Ground Motion Prediction
    Oxford Intelligent Earth CDT • November 2024 • Practical workshop for first-year PhD students
    Led a hands-on workshop on probabilistic fusion of seismic velocity models using Gaussian Processes, with application to uncertainty quantification in ground motion prediction. Created an open-source Jupyter notebook featuring interactive examples covering data fusion using Gaussian processes, finite difference seismic wave simulations, and uncertainty quantification for ground motion predictions. Students worked through exercises including: (1) engineering safety assessments using peak ground displacement predictions with different GP kernels and earthquake scenarios, (2) quantifying the impact of velocity model inconsistencies on ground motion predictions, and (3) computational optimization to determine minimum simulation requirements for robust uncertainty quantification.

Contact

If you'd like to get in touch, please email me at:

  • sam.scivier [at] earth.ox.ac.uk

Short Bio

Sam is a PhD student in the Department of Earth Sciences at the University of Oxford, developing probabilistic methods for uncertainty quantification in geophysics. He holds a Master's in Physics from the University of Birmingham and has gained industry experience through internships in quantum computing at D-Wave Systems (Canada) and Riverlane (UK). His research focuses on Gaussian process-based approaches for probabilistic fusion of geospatial datasets, with applications to earthquake ground motion prediction and seismic hazard assessment.