Lecturer (Associate Professor), Civil and Environmental Engineering
University of Canterbury, Christchurch, New Zealand
I am a geophysicist and data scientist specialising in Environmental Informatics. My research develops probabilistic and machine learning models for natural hazard forecasting of volcanic eruptions, wildfires, snow avalanches, and floods, by integrating time series analysis, Bayesian inference, and deep learning with real-time observational data. My tools are designed to generalise across hazard systems and to be deployable in under-resourced monitoring settings.
Applying machine learning to seismic time series to identify transferable eruption precursors across diverse volcanic settings. Our ergodic and transfer-learning approaches work even at data-scarce volcanoes.
Sub-hourly fire-potential forecasting using ML on weather time series. We build cost-effective, real-time tools deployable in regions with limited historical fire records via transfer learning.
Bayesian magnetotelluric inversion and multi-channel data modelling to image clay caps, quantify heat flux, and infer subsurface structure in geothermal and geyser fields.
Physics-informed LSTM models for real-time river flow forecasting and integrating ML into hydrology education, extending deep learning tools to flood analysis and avalanche detection.
Nature Communications 16, 1758 (2025)
Nature Communications 13, 2002 (2022)
International Journal of Wildland Fire 34(1) (2025)
Geophysical Research Letters 48(8) (2021)
UN Chair: Appointed Chair of the Topic Group on AI for Volcanic Eruption Forecasting within the Global Initiative on Resilience to Natural Hazards through AI Solutions (United Nations).
New paper: From forecast skill to economic value: sub-hourly wildfire potential forecasting across Australian regions published in International Journal of Wildland Fire.
NZ Geophysics Prize: Awarded the New Zealand Geophysics Prize for the most outstanding research publication in geophysics related to New Zealand (Geoscience Society of New Zealand).
Keynote at FireNZ 2025: Invited keynote on AI-powered wildfire forecasting for better decision-making.
Keynote at UFBA AGM 2025: Keynote on AI-powered wildfire forecasting.
Allianz Climate Risk Award: Recognised for innovative ML-based wildfire danger forecasting system providing real-time, sub-hourly updates using cost-effective infrastructure.
New paper in Nature Communications: Ergodic seismic precursors and transfer learning for eruption forecasting at data-scarce volcanoes.
University of Auckland, New Zealand
University of Chile
University of Chile
University of Canterbury, Christchurch, NZ
University of Canterbury, Christchurch, NZ