Bio
Fiona holds an MSc in Mathematical Physics (University of Edinburgh) and an MSc in Environmental Change and Management (University of Oxford). She is currently a PhD student at the University of Reading, working with Prof Marlene Kretschmer and Prof Ted Shepherd on developing causal machine learning methods to study teleconnections in the climate system. Fiona co-developed and maintains the open-source software package ibicus for the comparison and evaluation of statistical bias adjustment of climate models. Prior to starting her PhD, she worked for two years at a not-for-profit organisation on climate finance and was part of the international Mercator Fellowship programme, working on finance for climate resilience and loss and damage.
Research interests
Long-range interactions in the climate system, so-called teleconnections, are important sources of predictability for sub-seasonal to seasonal (S2S) forecasting and contribute to our understanding of near-term climate variability and change. While these teleconnections are often misrepresented in physical models of the climate system, existing statistical and machine learning methods struggle to identify teleconnection signals from reanalysis data that are robust and extrapolate under climate change. Causal models can address the former issue, but so far have limited ability to jointly identify optimal and causally related reduced representation of the relevant high-dimensional processes. Fiona's research so far has explored the use of generative machine learning methods to identify physically robust circulation patterns that improve predictability of extreme precipitation. To further extend the predictive skill of these methods, her research at the Turing Institute aims to further develop causal representation learning methods that are capable of identifying interpretable, robust and causally related latent representations of high-dimensional climatic processes.