Sebastian Hickman



Enrichment Student

Cohort year


Partner Institution


Sebastian Hickman is a PhD student at the University of Cambridge, where he is part of the AI for Environmental Risks Centre for Doctoral Training and the Centre for Atmospheric Science. He works primarily on modelling air pollution with machine learning, with a particular focus on extreme ozone air pollution episodes. More generally, he is interested in applying machine learning to environmental problems to help mitigate the effects of climate change. In particular, he has previously worked on using remote sensing and computer vision to analyse changes in tropical forests over time, and on flood forecasting. 

Research interests

Sebastian's research is focussed on using machine learning and statistical techniques to better predict and understand air pollution. The WHO estimates that air pollution causes approximately 7 million premature deaths annually worldwide. Part of Sebastian's work looks at deploying state of art machine learning methods to make short-term future predictions of air pollution in Europe, with a particular focus on ozone air pollution. This is a tricky problem - there are both temporal and spatial aspects to air pollution which have to be addressed in models. The hope is that with more accurate predictions, policy could be put in place to reduce the risk to humans. However, these models often struggle to predict extreme pollution effectively, and furthermore they do not give insight into what factors cause extreme pollution events. To address this, Sebastian is currently applying physically-informed machine learning models which look at the causal drivers of air pollution, in order to determine the drivers of high air pollution, and hence isolate possible governmental interventions to reduce air pollution in future climates.