Kenza Tazi

University of Cambridge

I am a final year PhD student at the University of Cambridge and the British Antarctic Survey as well as a member of the ‘AI for Environmental Risk’ doctoral programme. My thesis focuses on applying novel probabilistic machine learning methods to better understand and predict precipitation over High Mountain Asia. Outside of my doctoral studies, I have developed machine learning models for other environmental applications such as cloud identification and wildfire forecasting. I am also interested in climate policy and bridging the gap between scientific knowledge and decision-making. Before moving to Cambridge in 2019, I completed an integrated master’s in Physics at Imperial College London.

Talks

Precipitation prediction from large-scale climatic features over the Upper Indus Basin using Gaussian Processes

23-Apr-24

Water resources from the Indus Basin sustain over 268 million people. However, water security in this region is threatened by climate change. This is especially the case for the Upper Indus Basin where most solid water reserves are expected to disappear and precipitation to become the main driver of river flow. Yet future precipitation estimates for this region are uncertain. This paper explores the feasibility of using large-scale atmospheric features, which are better predicted by global climate models, to predict local precipitation and assess the probability of extreme precipitation events in the future. More specifically, Gaussian Processes are trained to predict monthly ERA5 precipitation data over a 15-year horizon. This paper also explores different model configurations, including non-stationary covariance functions, and shows that a well-designed Gaussian Process model can be effectively used for extrapolation.