4D seismic inversion for pressure and saturation using machine learning

The interpretation and analysis of 4D or time-lapse seismic data for monitoring hydrocarbon and CO2 reservoirs is one of the most exciting and active subjects of research in the upstream energy industry. In this subject, we attempt to provide an accurate understanding of changes in the subsurface state from geo-energy activity (injection, production and storage). This is a very challenging multi-disciplinary objective which can add considerable value to the asset (whether for production or storage). A key technique that delivers the desired quantitative information is direct 4D inversion of the seismic volumes to pressure and saturation changes. Whilst successes have been reported, techniques to implement this approach remain technically underdeveloped, and more focus on application to data is required. This is particularly true when utilising the pre-stack domain and when working with full 3D volumes and anomaly geobodies. As data acquisition and processing improves, there is growing interest in the utilisation of these domains to extract more value from our quantitative interpretations.

In this project we propose to extend our existing tools and procedures for 4D inversion to pressure and saturation, and to apply these improvements to several high quality, commercially acquired, datasets with known time-lapse changes. You will work with techniques in both the post-stack and pre-stack data domains and close the loop with the fluid flow simulator. You will apply the new technique to different datasets for CO2 injection as well as those producing hydrocarbons. The approach uses supervised machine learning techniques with the application of fluid flow and geomechanical constraints. You will learn to work with our current tools for time-shift analysis and modelling to close the loop between the model and 4D seismic data. The data for the project has been donated by industry, and will consist of 3D and 4D seismic data, wireline logs, production data, a field flow simulation model, and in some cases a geological model. The project will be delivered to you in a Petrel project. You will join the vibrant community of the ETLP research group which has twenty-five years of industry-relevant experience in quantitative 4D seismic interpretation.

You will work on this PhD project supervised by a multi-disciplinary team led by Prof Colin MacBeth.

Eligibility

This project is available to ALL students, whether home, EU or overseas. The successful candidate should have a strong interest in applied research and possess at minimum a Masters AND undergraduate degree in Geophysics, Applied Mathematics, Physics, or a related field. Formally four years of university study including a minimum of one year at an advanced level are required. Programming skills, particularly in Python, are an essential requirement of this project, whilst some experience of fluid flow simulation and seismic processing is also necessary. Several years of additional experience working in industry is desirable.

Funding 

This is a full scholarship which will cover tuition fees (Home and Oversees) and provide an annual stipend (paid in line with UKRI recommended rates, £19,237 in 2024-25) for 42 months. Thereafter, candidates will be expected to pay a continuing affiliation fee (currently £130) whilst they complete writing up their thesis.

How to Apply

To apply you must complete our online application form.

Please select PhD Applied Geoscience as the programme and include the full project title, reference number (EGIS2025-CM1) and supervisor name on your application form. Ensure that all fields marked as ‘required’ are complete.

Once have entered your personal details, click submit. You will be asked to upload your supporting documents. You must complete the section marked project proposal; provide a supporting statement (1-2 A4 pages) documenting your reasons for applying to this particular project, outlining your suitability and how you would approach the project. You must also upload your CV, a copy of your degree certificate and relevant transcripts and an academic reference in the relevant section of the application form.

You must also provide proof of your ability in the English language (if English is not your mother tongue). We require an IELTS certificate showing an overall score of at least 6.5 with no component scoring less than 6.0, or a TOEFL certificate with an overall score of at least 85, including reading 20, listening 19, speaking 20 and writing 21. Alternatively, if you have received an English-taught Bachelors or Masters degree from one of the countries listed on the UK Government Guidance under ‘Who does not need to prove their knowledge of English’, and it was obtained less than five years from your intended start date, you should provide evidence of your award that clearly states it was delivered and assessed in English language.

For more information on our specific activities or project details please visit our website: https://etlp.hw.ac.uk or contact Colin MacBeth at C.MacBeth@hw.ac.uk

Please contact egis-pgr-apps@hw.ac.uk for technical support with your application.

Timeline

The closing date for applications is 15 March 2025 and we expect interviews to take place on 20 March. Applicants must be available to start in either May or September 2025.

References

Côrte, G., Amini, H. and MacBeth, C., 2023. Bayesian inversion of 4D seismic data to pressure and saturation changes: Application to a west of Shetlands field. Geophysical Prospecting, 71(2), pp.292-321.
Côrte G, Tian S, Marsden G, MacBeth C. Bayesian Inversion of 4D Seismic Data with a Machine Learning Prior: Application to the Catcher Fields. In84th EAGE Annual Conference & Exhibition 2023 Jun 5 (Vol. 2023, No. 1, pp. 1-5). European Association of Geoscientists & Engineers.
Côrte, G., Dramsch, J., Amini, H. and MacBeth, C., 2020. Deep neural network application for 4D seismic inversion to changes in pressure and saturation: Optimizing the use of synthetic training datasets. Geophysical Prospecting, 68(7), pp.2164-2185.