Accurate models of subsurface Earth structure are fundamental for finding and producing hydrocarbons. The key data come from seismic surveys and interpretation of these images employs thousands of geoscientists globally. Analytical strategies and methodologies used to interpret seismic images are diverse, commonly relying on know-how held by experts. Existing studies (Bond et al. 2007, 2012) show that the same seismic datasets can be interpreted to yield different models of subsurface structure. This project is designed to investigate methods for identifying areas of high uncertainty and complexity in structural models resulting from interpretation of seismic image data, to evaluate possible models and their consistency to improve interpretation workflows.
The PhD research will address this challenge in three parts.
1) Identification of areas of high uncertainty and high risk in structural models
Comparison of existing workflows for identification of areas of high model uncertainty and risk. Multiple methods exist e.g. use of information in the seismic data and analysis of continuity and contrast in the seismic image to determine areas of poor seismic response (e.g. Botter et al. 2014; Alcalde et al., 2017), or analysis of multiple interpretations of the seismic dataset to determine areas of model divergence (e.g. Pellerin et al. 2013, Corbel and Wellmann, 2015, Richards et al., 2015). Using a 3D seismic dataset the different methods will be tested at multiple locations within the seismic cube. Comparison of different approaches will allow the relative efficacies of the approaches, in identifying areas of likely model complexity to be assessed.
2) Analysis of the effectiveness of stochastic modelling in high risk areas.
Stochastic modelling can be used in areas of high interpretation uncertainty to provide a range of possible interpretations (Wellmann et al., 2010; Cherpeau and Caumon, 2015; Julio et al., 2015). Using the 3D seismic dataset in 1) the areas of high interpretation uncertainty will be targeted for stochastic modelling to create multiple models. These models will be compared to models created by seismic interpreters using normal interpretation workflows. The researcher will interview interpreters to understand the decisions made in creating their modelling. The comparison will assess the potential for stochastic modelling to create likely interpretations, as defined by expert seismic interpreters, and determine which models are most consistent with known information and expert knowledge.
3) Determine if a priori information can be used to condition stochastic model outputs effectively.
Model conditioning by incorporation of a priori information, can be used to minimize and refine modelling efforts (Wood and Curtis, 2004; Zhang, 2008; Cherpeau and Caumon, 2015; de la Varga and Wellmann, 2016). The researcher will use implicit models of the expert interpreters and information from the interviews conducted in 2) to determine rules that can be employed in the stochastic modelling. Balancing may be used to assess model likelihood a posteriori on various stochastic structural models. Results from the PhD will be fed into an existing framework for machine learning, developed at Aachen University, to test the use of expert input to condition future model suggestions.
The successful applicant should have a UK Honours Degree (or equivalent) at 2.1 or above in Geology/Geological Sciences. Numerically literate candidates with experience in programming are encouraged to apply. We are looking for a numerical geoscientist who is interested in working across the interface of geocognition and computer modelling. Previous experience in coding or mathematical modelling and an appreciation for structural geology techniques and analysis are desired.
The student will need to spend significant periods in Aachen and Nancy. The other supervisors on this project are Florian Wellman, RWTH Aachen University, Germany
Guillaume Caumon, University of Lorraine, France
Florent Lallier, Total Geoscience Research, Centre, UK
The start date of the project is to be agreed with the supervisors but will be as soon as possible.
This project is advertised in relation to the research areas of the discipline of Geology. Formal applications can be completed online http://www.abdn.ac.uk/postgraduate/apply. You should apply for Degree of Doctor of Philosophy in Geology, to ensure that your application is passed to the correct Discipline for processing. NOTE CLEARLY THE NAME OF THE SUPERVISOR and EXACT PROJECT TITLE ON THE APPLICATION FORM.
Informal inquiries can be made to Dr C Bond ([Email Address Removed]) with a copy of your curriculum vitae and cover letter. All general enquiries should be directed to the Graduate School Admissions Unit ([Email Address Removed]).
Alcalde, J., Bond, CE., Johnson, G., Ellis, J. & Butler, RWH. (2017). 'Impact of seismic image quality on fault interpretation uncertainty'. GSA Today, vol 27, no. 2, pp. 4-10.
Bond, CE., Gibbs, AD., Shipton, ZK. & Jones, S. (2007). 'What do you think this is?: "Conceptual uncertainty" In geoscience interpretation'. GSA Today, vol 17, no. 11, pp. 4-10.
Bond, CE., Lunn, RJ., Shipton, ZK. & Lunn, AD. (2012). 'What makes an expert effective at interpreting seismic images?'. Geology, vol 40, no. 1, pp. 75-78 Data Repository item 2012026.
Botter C., Cardozo N., Hardy S., Lecomte I., Escalona, A. (2014). From mechanical modeling to seismic imaging of faults: A synthetic workflow to study the impact of faults on seismic. Marine and Petroleum Geology 57, pp. 187-207.
Cherpeau, N. and Caumon, G., 2015. Stochastic structural modelling in sparse data situations. Petroleum Geoscience, 21(4), pp.233-247.
Corbel, S. and Wellmann, J.F., 2015. Framework for multiple hypothesis testing improves the use of legacy data in structural geological modeling. GeoResJ, 6, pp.202-212.
De la Varga, M. and Wellmann, J.F., 2016. Structural geologic modeling as an inference problem: A Bayesian perspective. Interpretation, 4(3), pp. 1–16.
Julio, C., Caumon, G. and Ford, M. (2015). Sampling the uncertainty associated with segmented normal fault interpretation using a stochastic downscaling method. Tectonophysics 639, pp. 56–67.
Pellerin, J., Caumon, G., Julio, C., Mejia-Herrera, P., Botella, A. 2013. Elements for measuring the complexity of 3D structural models: Connectivity and geometry. Computers and Geosciences, 2015, 76, pp.130-140.
Richards, FL., Richardson, NJ., Bond, CE. & Cowgill, M. (2015). 'Interpretational variability of structural traps: implications for exploration risk and volume uncertainty'. Special Publication - Geological Society of London, vol 421, pp. 7-27.
Wood, R. and Curtis, A., 2004. Geological prior information and its applications to geoscientific problems. Geological Society, London, Special Publications, 239(1), pp.1-14.
Wellmann, J.F., Horowitz, F.G., Schill, E. and Regenauer-Lieb, K., 2010. Towards incorporating uncertainty of structural data in 3D geological inversion. Tectonophysics, 490(3-4), pp. 141–151.
Zhang, T., 2008. Incorporating geological conceptual models and interpretations into reservoir modeling using multiple-point geostatistics. Earth Science Frontiers, 15(1), pp.26-35.