skip to content

AI for the study of Environmental Risks (AI4ER)

UKRI Centre for Doctoral Training

2019 Cohort

Edward Brown

PhD project - AI for state-of-the-art Space Weather Forecasting

My PhD project will focus on the creation of state-of-the-art artificial intelligence/machine learning techniques to forecast space weather conditions on Earth. Space weather is a potent natural hazard capable of massive economic damage to the satellite industry, power outages and harm to human health. Robust forecasting of this phenomena is of massive importance. Research will be conducted into providing forecasts using solar wind data and solar imaging.

Petr Dolezal

PhD project - Complementarity of Renewables across Super-grids

I'm studying the possibility of using multiple renewable energy sources, distributed spatially across large areas, to complement each other and ultimately reduce the uncertainty of the total renewable energy supply in an electric grid.

These sources, such as solar panels and wind turbines, have their ability to deliver power limited by multiple meteorological variables, primarily cloud cover and wind speed, respectively. If the variability of these sources was independent of each other, we could expect the total supply to be more stable. However, it is often heavily correlated, especially across short distances. I'm studying this correlation both in meteorological data and in climate models, where the distribution will likely shift due to Climate Change.

The goal is to quantify the stability gained by connecting existing grids into a single large super-grid. Another goal is to figure out how to use anti-correlated regions to supplement each other during an abnormally low power supply, instead of fossil fuel sources of electricity. Finally, I analyze what is the necessary grid connectivity for harnessing this level of complementarity.

Marc Girona

PhD project - tbc

Omer Nivron

PhD project - Climate Model Bias Correction – the Unfathomable Neural Network

For my PhD project I plan to focus on the development of a new Bias Correction framework that would match climate model outputs and real-world observations. Additionally, I aim to introduce a new method that would allow us to read aleatoric and epistemic uncertainty estimates associated with the bias correction. Both goals would be applied on the daily mean temperature for latitudes between 0 ̊ and 15 ̊ North and the year 2050.

Raghul Parthipan

PhD project - Probabilistic machine learning, using GANs and Gaussian Processes, to reduce climate model computational costs, improve parameterisations and improve accuracy

Global Climate Models (GCMs) are important for our understanding of Earth's past, present and future climates. They are based on fundamental physical processes, therefore accurate cloud modelling is important, with clouds strongly influencing the transfer of radiant energy and spatial distribution of latent heat in the atmosphere, influencing weather and climate. The representation of clouds is a major source of uncertainty in climate models and the primary reason behind the inter-model variance in climate response. Raghul will be bringing probabilistic machine learning approaches, such as Gaussian Processes and Generative Adversarial Networks, to this problem. These will be used to reduce computational requirements of GCMs, reduce errors through better understanding of the distribution and uncertainty associated with cloud processes, and improve generalisability for future climate scenarios. 

Tudor Suciu

PhD project - tbc

Kenza Tazi

PhD project - tbc

Anna Vaughan

PhD project - Bayesian physics-informed learning for severe weather forecasting

Anna's PhD research focuses on applying recent advances in Bayesian learning and physics-informed neural networks to forecast extreme weather events. Specifically, her interests include short-range forecasting of tornadoes and extreme convection using remote sensing data and exploring the application of physics informed learning in medium-range weather prediction. This work aims to develop a global statistical forecasting model leveraging remote sensing observations capable of out-performing existing operational numerical weather prediction systems. 

Prior to her PhD Anna completed her undergraduate studies in pure mathematics and theoretical physics, followed by a MSc in meteorology from the University of Melbourne. Her MRes project in the CDT developed a new methodology for downscaling climate model output using Bayesian learning. 

Mala Virdee

PhD project - tbc

Michelle Wan

PhD project

Michelle’s research explores the applications of machine learning in the predictive modelling of outdoor air pollutants and their association with human lung cancer. The first stages of the project consist of investigation into air pollution simulation and the utility of real-world data. This includes the evaluation of modelling approaches such as land-use regression and dispersion models, and the application of machine learning techniques to incorporate real-world data into modelling approaches. The project then explores a pollution/disease case study using simulated pollutant data to predict lung cancer incidence. Air pollution epidemiology studies typically analyse association using odds or hazard ratio statistics, which have limited utility for incidence trend prediction, thus motivating the application of predictive machine learning models. The project takes a long-term view (beyond the project itself) of predictive model development for applications in fields such as healthcare and policy-making.

Before joining the CDT, Michelle completed her BA and MSci in Natural Sciences (Biological) at the University of Cambridge, specialising in Chemistry and Atmospheric Chemistry for her MSci year. During her MRes year with the CDT, Michelle studied the association between NO2 air pollution exposure and human breast cancer incidence in London, using LSTM networks.

Aligned Students


Edgar Cifuentes

Edgar's research aims to develop microclimatic models to understand how forest composition affects local climates in Colombia and how its degradation has repercussions for biodiversity and global warming. He will track the changes in forest cover and microclimates by combining climatic data from field sensors with a diverse range of satellite data. By using AI, he will be able to build statistical models that will enable the discovery of microclimatic patterns and predicting these in the future (or past), and in larger areas. Because Colombia has a diverse set of climates and landscapes, these models will be useful for predicting microclimates at regional and global scale in the tropics. Datasets and models will be the basis for showcasing detailed distributions of tropical species and the quantification and localization of future changes in temperature and moisture. Through this work, it would be possible then to prove the hypothesis that climatic changes in the tropics are predominantly driven by land-use changes (deforestation) and to a lesser extent, by global warming through greenhouse gas emissions.

Ramit Debnath

Ramit’s research focuses on the distributional justice implications of built environment design and energy demand. He is trying to understand the injustices associated with energy service demand in slum rehabilitation housing of the Global South. He is employing data-driven approaches to inform evidence-based policymaking. Recently, he collaboratively developed a novel methodology for discovering hidden topics in the narratives of energy and climate justice of low-income communities (1). This methodology was called ‘deep-narrative’ analysis that had its basis in social sciences (grounded theory), and machine learning (ML) and AI (topic modelling using Latent Dirichlet Allocation). He is currently expanding its application to discover policy focus points from occupants’ narratives of energy and climate injustices in the slum-built environment of Mumbai (India), Rio de Janeiro (Brazil) and Abuja (Nigeria). 

Lately, Ramit has used a similar approach in exploring India’s policy nudges to contain COVID-19 pandemic (2), and market shifts in global food systems due to the pandemic using ML-driven web-scraping. He is also investigating the effect of lockdowns on residential energy demand in India using ML-based clustering. For the latest update on his research, please visit 

twitter: @RamitDebnath

Jess Fleming

Further details to follow, however Jess's project relates to atmospheric chemistry and will involve the use of portable sensors to measure air quality in indoor and outdoor settings and analysis of the associated data.

Joseph Lillington

Joseph's research focusses on nuclear waste disposal aiming to better understand the behaviour of waste, were it to be stored deep underground within a geological disposal facility. Half of this research is on predicting radioactive waste glass dissolution behaviour and glass material properties by applying machine learning algorithms to large-scale industrial and lab derived experimental data. The other half of this research is on developing Monte-Carlo models to predict radionuclide transport behaviour in the near-field environment of a geological disposal facility.

Jason Zhe Sun

Jason's research looks at improving estimates of the global distribution of surface ozone (O3) and accordingly improving our estimates of the health impacts of surface O3. This will enable an improved population exposure risk assessment for the associated chronic diseases (like COPD).

His focus so far has been on the analysis of multi-model CMIP6 (1990-2100) global surface O3 simulations. The model evaluation, together with integrated analyses on other assisting data source (like underlying land-surface types, population etc) are being investigated to determine the sources of simulation-observation biases and malposition, so as to provide suggestions on making improvements in further process based modelling. The prediction enhancement using Machine Learning (ML) could provide more credible global O3 concentrations, based on which the all-cause premature death and COPD-related mortality will be estimated with multiple Shared Socioeconomic Pathways (SSPs) into 2100, for public health awareness rising and policy-making suggestions. 

Aaron Wienkers

Aaron is studying the dynamics of ocean fronts using theory and a suite of idealized numerical simulations.  Although Aaron’s work doesn’t involve machine learning directly, he has acquired expertise in theory of ocean and atmospheric circulation and high performance computing.


James Ball

James' PhD will help us discover how resilient tropical rainforest are to anthropogenic climate change. Tropical forests (TFs) store a vast quantity of carbon and can act as a substantial carbon sink, playing an important role in regulating the global climate system. However, the interlinked threats of deforestation and increased likelihood of drought from climate change put their continued existence in doubt. The aim is to apply innovative deep learning methods (2D/3D-CNNs, RNNs) to satellite/geospatial data to develop a system that forecasts the locations of future deforestation events. Through collaboration with an international NGO it will facilitate pre-emptive action that protects valuable ecosystems. To better understand how TFs will respond to future climate change, James will be join a team to conduct novel repeat surveys of an Amazon forest plot with airborne LiDAR and other ground based techniques. He will investigate what drives changes in the forest canopy (climatic conditions, plant phenology etc.), including at the individual/species level, and use satellite data to scale up the findings and understand the implications of a changing climate on the wider region. The results will lead to improved design and parameterization dynamic global vegetation models, and thereby advance understanding of the resilience of TFs to climate change.

Matt Ball

Matt is working on the use of a Zeiss Secondary Ion Mass Spectrometry (SIMS) instrument for Earth and environmental science applications. As part of this work, Matt works extensively with Machine Learning methods for the analysis of hyper-spectral data, which will be of direct relevance to many of the applications worked on by students in the CDT. He is familiar with the "Hyperspy" toolkit that is used extensively by people in this field for the extraction of quantitative data from hyper-spectral images.

Xiaolei Feng

Xiaolei's work involves the prediction of mineral structure from quantum mechanical first principles calculations. She is currently investigating the cycling of volatile elements (H, O, N, C) in the silicate portion of the Earth. Her work involves creation of large data sets and search for low energy structures, with implications for geochemical cycling.

Rachel Furner

Rachel's research focuses on using Machine Learning and data-driven methods to improve the efficiency of climate models. In particular her work focuses on learning from model data, to enable faster algorithms for specific applications, such as finding the equilibrium state of particular scenarios. Future plans include developing and applying these data-driven techniques to modelling of the ocean carbon cycle - a computationally expensive part of a climate simulation, which is also key to understanding and predicting the future climate. Prior to her PhD Rachel completed an undergraduate degree in Mathematics at the University of Oxford, she then worked for 8 years as a researcher at the UK Met Office, developing physically-based ocean models for short term forecasting. Following this she worked coordinating data-focused research institutes, before returning to research to pursue her PhD.

Martin Rogers

Martin's PhD research project involves exploiting satellite technology and machine learning to describe and predict shoreline change for hazard regulation. His main focus is on East Anglia, UK, which is particularly vulnerable to sea level rise, increases in storminess, coastal erosion, and coastal flooding. Critical national infrastructure (including Sizewell’s nuclear power stations and the Bacton gas terminals), population centres close to the coastal zone (> 600,000 in Norfolk and Suffolk) and iconic natural habitats (the Broads, attracting 7 million visitors a year) are under threat. Shoreline change, driven by complex interactions between environmental forcing factors and human shoreline modifications, is a key determinant of total vulnerability: its prediction is imperative for future coastal risk adaptation. This project will train Artificial Neural Networks (ANN) to predict shoreline evolution until 2040. High resolution satellite imagery, databases of sea defences, beach modification, wave and tide gauge data, bathymetry and meteorological fields, combined with land cover, population and infrastructure data will act as inputs.

Jonathan Rosser

Jon will apply novel data analysis techniques and machine learning algorithms such as dimensionality reduction, clustering and graphical networks to output produced from the latest group of coupled climate models. Unguided learning techniques in particular offer considerable potential to uncover presently unknown or misunderstood relationships within models. Challenges will include adapting existing algorithms to deal with the very high dimensional nature of the data and its non-stationary, non-isotropic, highly-correlated nature. Jon will contrast the state space parameters between models and work to associate such parameters with future model states (e.g why some models convect while others don’t) and constrain climate projections using the dynamical knowledge uncovered.

Robert Edwin Rouse

Robert is applying Bayesian machine learning methods to provide informative and actionable predictions on the future risk of urban flooding. Urban flooding may be caused by a number of factors including local geology and soil type, urban structural design and drainage capacity, extreme weather, and previous rainfall over daily to seasonal timescales. The challenge here is to intelligently combine the required information in a computationally efficient manner while providing robust estimates of uncertainty as required by decision makers. Robert works closely with industry partners Mott MacDonald who are a world-leading environmental engineering company, focusing on a range of structural projects including urban drainage, dams and reservoirs. Over the course of the PhD Robert will work with their engineers to integrate his new algorithms into Mott MacDonald’s data analysis pipelines.
Before joining Cambridge, Robert completed an MA in Royal College of Art, and an MEng at Imperial College London. He is also a Co-founder and the Chief Technology Officer at Ichthion Limited.

Rebecca Self

Rebecca's masters title is "Taskforce on Climate-related Financial Disclosures (TCFD) : Using geospatial data at Financial Institutions to analyse the physical climate risks associated with mortgage portfolios in London". Her research will seek to analyse how these data could be used to identify, analyse and report physical climate risks related to mortgages at large UK and Dutch banks. Working closely with risk managers at banks, the aim is to investigate the current application of geospacial data, possible barriers and any unintended consequences.

Matthew Shin

Matthew’s research is using ML to understand how we can make more accurate projections of the exposure of people to air pollution in the future using global chemistry climate models. Air pollution and climate change are coupled problems with some very non-linear feedbacks. To understand them we use expensive simulators, models. These models are run at a low resolution and the initial focus has been on using Gaussian Mixture Models (GMMs) to unpick rural and urban air pollutant levels simulated by the global models.

Will Tebbutt

Will Tebbutt holds an M.Phil. in Machine Learning, Speech and Language Techology from the University of Cambridge and an M.Eng. in Engineering Mathematics from the University of Bristol. He previously worked as a researcher in machine learning at Invenia Labs in Cambridge, and is a member of Darwin College. His research interests include Gaussian processes, automating Bayesian inference, the application machine learning to climate science, and probabilistic numerics.

Risa Ueno

Risa is developing state-of-the-art probabilistic machine learning methods to predict future changes in heatwaves and their associated increases for energy demand driven by air conditioning usage. Initially the aim is to focus on megacities, where we have a wealth of relevant climate and non-climate data, before applying these models to more data-sparse cities around the world to improve resilience to future climate extremes. Risa’s working closely with AI engineers at the Alan Turing Institute, and industry partners Max Fordham who are an award-winning architectural engineering firm with a specialism for designing buildings accounting for future environmental change, withstanding weather extremes, and maintaining human comfort through natural or low-energy solutions.

Before joining BAS and the University Risa completed an MSc in Computational statistics and Machine Learning from UCL and an MPhys from Imperial College London. Between her two masters degrees Risa worked as a Technology Consultant for IBM.

Ran Xiao

Ran is a PhD student in the Department of Architecture. In his previous career as an architect, he became fascinated about how design knowledge is acquired through experience. This inspired his research topic, which explores using machine learning to acquire existing spatial knowledge to better inform architects of design decisions. Spatial design is a fundamental issue to sustainability of the built environment. Orientation, shape and size all affects a building’s exposure to the elements. Existing practice, however, relies on modelling, which predicts performances at final stages of design. It means that fundamental changes are rarely made as the spatial design has already been committed. Ran’s research aims to produce a prototype for learning from real world data, coupling spatial design and sustainability performances. It aims to generate sustainable designs at an early stage for similar context, based on a database of best examples of sustainable design.

Yue Zhu

Due to rapid urbanisation and climate change, a large number of cities across the globe are now facing increased flood risks. Recent advances in data and machine learning (ML) methods have provided new opportunities to investigate the effects of spatial-temporal changes in land-cover and land-use (LCLU) on flood risk in and around cities. This research project exploits new machine learning methods to investigate urban flooding risks. The hypothesis underlying the research is that ML methods, which hitherto have rarely been tested in this field, would bring significant new insights because of their abilities to model the diversity and variability of LCLU. This research aims to facilitate more informed decision-making for urban planners, economists, ecologists, and policy-makers.