Winning at digital twinning: integrating AI and numerical modelling on the Great Barrier Reef

The ACCE+ DLA is committed to recruiting extraordinary future scientists regardless of age, disability, ethnicity, gender, gender identity, sexual orientation, faith or religious belief, pregnancy or maternity, parental or caring responsibilities or career pathway to date. We understand that a student’s potential can be shown in many ways and we strive to recruit students from all backgrounds, and support them on their scientific journey.

We have designed our application systems to identify candidates who are likely to be successful in research regardless of what opportunities may have been available to them prior to their application.

Project Overview

Oceans constitute a fundamental component of Earth’s life support system, providing a number of ecosystem services indispensable to human well-being. Shelf seas, seas that are on continental crust, are found around the globe and are the interface between the coastline and the deep ocean. Though covering less than 10% of the oceans, they contribute disproportionately to global marine production and support 90% of the global fisheries. However, anthropogenic pressures have precipitated a decline in shelf sea health. Pollution, habitat destruction, over-exploitation, and climate change are driving marine ecosystem degradation. Coral reefs, biodiversity hotspots in tropical shelf seas, are experiencing catastrophic loss; 10% are lost already with a further 60% at risk.

The Great Barrier Reef (GBR) is the world’s largest reef system. Ocean warming is accelerating on the GBR and 2024 was the warmest year on record to date. Data measurements are sparse on the reef, so using numerical models to understand the physical and biological processes is key to ensuring effective management. These models can also be used to simulate the future under various climate and sea-level rise scenarios to aid management. To make the most of these data and models a digital twin of the shelf sea environment and ecosystem is needed: a virtual copy of the real world, automatically coupling together the latest numerical modelling technology with the wealth of public data available via artificial intelligence.  

This project will extend the capabilities of Thetis, an open-source unstructured grid model that can simulate the complex bathymetry in the GBR, to include wave processes. This new model will then be validated against data collected around One Tree Reef in the GBR’s Capricorn Bunker and uncertainties in the model will be explored via construction of an emulator (surrogate model). The final task will be to build a machine learning model to extract flow information from satellite images, using the emulator and numerical model as training material. We can then use this machine learning model to examine oceanographic processes across the GBR.

The project will be based in the Environment and Geography Department at the University of York, which offers an outstanding, dynamic and multidisciplinary environment in which to carry out PhD research. Our current PhD students come from many countries around the world and are well supported by a comprehensive programme of training and an inclusive supervision network. We welcome applicants from all backgrounds, particularly those underrepresented in science, who have curiosity, creativity and a drive to learn new skills.

We are looking for an enthusiastic person to join a growing research team in York that uses numerical methods to tackle environmental problems. You should have a background in mathematics, physical sciences or computer science with a passion to develop physical oceanography and ecological knowledge. There is the opportunity of travelling to the Great Barrier Reef to collect data to calibrate the model with partners at the University of Sydney, which will take place in the second year of the project. There is also the possibility of placements in industry in year 3 (NOC Liverpool). 

Project supervisor details and information:

• Jon Hill (University of York)
• Ana Vila Concejo (University of Sydney)
• Jill Johnson (University of Sheffield)

Professor Ana Vila Conceio and Dr Jill Johnson will also be co-supervisors of this project.

Part-Time Study Options

All ACCE+ PhDs are available as part time or full time, with part time being a minimum of 50% of full time. Please discuss potential part time arrangements with the primary supervisor before applying to the programme. 

Project CASE Status – This project is not a CASE project.

Candidate webinar

The project primary supervisor will hold a Candidate Information Session via Zoom on 10th December 2024 at 11:00 am – 12 noon UK time to discuss the project with interested candidates. This will be an opportunity for you to meet the supervisor(s), find out more about the project and to ask any questions. To register for the Information Session, you should complete this Google Form, noting the important information provided about how your data will be used.

How to Apply

To apply for this project, you need to complete and submit a Google Form. This is instead of a cover letter.The proforma is designed to standardise this part of the application to minimise the difference between those who are given support and those who are not. As part of this form, you will need to include a personal statement.You will also need to upload a CV, your undergraduate transcript, and if applicable, your Masters transcript.You will also need to provide the contact details of two referees.

Please see the ACCE website for all details of how to apply to the programme at each ACCE+ institution: https://accedtp.ac.uk/how-to-apply/.