Fully-funded PhD Scholarship in Deep Learning Applications for Geological Science: Automatic Seismic Data Interpretation using Deep Learning

Fully-funded PhD Scholarship in Deep Learning Applications for Geological Science:
Automatic Seismic Data Interpretation using Deep Learning



for UK/EU students


The Department of Computer Science and the Department of Earth Sciences at Durham University are pleased to offer a PhD studentship for a collaborative project with industry, funded by European Regional Development Funding, Durham University and GeoTeric Ltd. The project is ideal for someone who wishes to apply research to problems with a strong application to industry.
The fully-funded studentship will start in October 2019 and the successful applicant will receive a scholarship for three years (subject to satisfactory progression).


The studentship includes:
- A stipend of £19,000 per year (tax-free)
- Research costs to cover publications/travel expenses/consumables.
- Domestic fees supported by the business. Note that international students from outside the EU will need to pay the balance of their fees.



Project Description

In close collaboration with GeoTeric Ltd., the project involves the development of novel Deep Learning-based solutions to advance the interpretation and understanding of seismic data in two main areas:

Identification of Geobodies
To gain a better understanding of the subsurface, geological structures are segmented into geobodies; and reviewed in either isolation or together to examine possible interplays between features. Segmenting complex geological structures is both time consuming and challenging. The prospective student will build new deep learning algorithms to segment complex geological structures such as salt domes, gas chimneys or channel systems in order to improve efficiency and reduce risk during interpretation. This is particularly motivated by the unprecedented success of deep learning in image segmentation.

Analysis of Outcrop Analogues
Seismic interpreters often use outcrop images or synthetic 3D models to help build a conceptual model of the subsurface. The successful candidate will develop deep learning algorithms to accurately match geological features within highly variable seismic data to outcrop images or synthetic 3D models which could increase the efficiency with which complex geological structures are understood.


The collaborative nature of the PhD means that you will spend time in both Durham University and GeoTeric’s offices in Newcastle upon Tyne, just 12 minutes on the train from Durham.


Entry Requirements:

You need a 2:1 Bachelor’s degree or equivalent in an appropriate subject such as Physics, Maths, Computer Science or Geophysics, from a recognised university.
EU students would also require a minimum overall IELTS 6.5 score of which no element of less than 6.0 (or equivalent).



How to apply

Please send a CV and covering letter to
Dr Noura Al Moubayed (Computer Science): [Email Address Removed],


Further information on the departments and GeoTeric can be found at:
https://www.dur.ac.uk/computer.science/
https://www.dur.ac.uk/earth.sciences/
https://www.geoteric.com/