Clare E. Bond

What do you think this is? "Conceptual uncertainty" in geoscience interpretation

Interpretations of seismic images are used to analyze sub-surface geology and form the basis for many exploration and extraction decisions, but the uncertainty that arises from human bias in seismic data interpretation has not previously been quantified. All geological data sets are spatially limited and have limited resolution. Geoscientists who interpret such data sets must, therefore, rely upon their previous experience and apply a limited set of geological concepts. We have documented the range of interpretations to a single data set, and in doing so have quantified the �conceptual uncertainty� inherent in seismic interpretation. In this experiment, 412 interpretations of a synthetic seismic image were analyzed. Only 21% of the participants interpreted the �correct� tectonic setting of the original model, and only 23% highlighted the three main fault strands in the image. These results illustrate that conceptual uncertainty exists, which in turn explains the large range of interpretations that can result from a single data set. We consider the role of prior knowledge in biasing individuals in their interpretation of the synthetic seismic section, and our results demonstrate that conceptual uncertainty has a critical influence on resource exploration and other areas of geoscience. Practices should be developed to minimize the effects of conceptual uncertainty, and it should be accounted for in risk analysis.

Structural models: Optimizing risk analysis by understanding conceptual uncertainty

Abstract Geoscience may be regarded as an uncertain science, as it is often based on the interpretation of equivocal data. Analysis of multiple interpretations of a single dataset has shown that conceptual uncertainty can result in a wide range of interpretational outcomes. Many geological models based on a wide variety of concepts were developed by different geoscientists for the same dataset. In this paper we suggest methods to improve the effectiveness of interpretation workflows based on understanding of how geoscientists apply concepts to equivocal datasets, the processes they use, the effects of their previous experience, and their use of broader contextual information. We argue that understanding the influence of conceptual uncertainty on interpretation of equivocal data and modification of current workflow practices can improve risk management.

Co-axial horizontal stretching within extending orogens: the exhumation of HP rocks on Syros (Cyclades) revisited

Abstract Although the role of extensional tectonics in the exhumation of high-pressure metamorphic terranes is widely established, the kinematics of such deformation remains ambiguous. This paper outlines new field data from the Attic-Cycladic blueschist belt that suggest that distributed ductile strain plays a significant role in the extension and that, consequently, the role of major detachment faults may have been over-emphasized in previous studies. The high-pressure blueschist terrane (Ermoupolis Unit) of Syros shows abundant evidence of subhorizontal extension, manifest as layer boudinage and ductile thinning without the development of significant internal detachments. The deformation approximates to pure shear stretching that was heterogeneously distributed in space and time. Minor zones of asymmetric shear are interpreted not as through-going extensional shear zones but as structures that maintain compatibility between zones of differential stretching. The progression of deformation is charted through the systematic development of increasingly lower-pressure metamorphic assemblages. However, most of the decompression (potentially from 20 kbar to 6 kbar) occurred within the blueschist stability field, as the rocks were actively extending. Heterogeneous retrogression and concomitant deformation are believed to relate to the local chemistry and availability of hydrous fluids.

Structural model creation: the impact of data type and creative space on geological reasoning and interpretation

Abstract Interpretation of sparse or incomplete datasets is a fundamental part of geology, particularly when building models of the subsurface. Available geological data are often remotely sensed (seismic data) or very limited in spatial extent (borehole data). Understanding how different datasets are interpreted and what makes an interpreter effective is critical if accurate geological models are to be created. A comparison of the interpretation outcome and techniques used by two cohorts interpreting different geological datasets of the same model, an inversion structure, was made. The first cohort consists of interpreters of the synthetic seismic image data in Bond et al. (‘What do you think this is?: “Conceptual uncertainty” in geoscience interpretation’, GSA Today, 2007, 17, 4–10, http://dx.doi.org/10.1130/GSAT01711A.1); the second cohort is new and interpreted borehole data. The outcomes of the borehole interpretation dataset support earlier findings that technique use, specifically evidence of geological evolution thought processes, results in more effective interpretation. The results also show that the borehole interpreters were more effective at arriving at the correct interpretation. Analysis of their final interpretations in the context of psychological and medical image analysis research suggests that the clarity of the original dataset, the amount of noise and white space may play a role in interpretation outcome, through enforced geological reasoning during data interpretation.

Impact of seismic image quality on fault interpretation uncertainty

Uncertainty in the geological interpretation of a seismic image is affected by image quality. Using quantitative image analysis techniques, we have mapped differences in image contrast and reflection continuity for two different representations of the same grayscale seismic image, one in two-way-time (TWT) and one in depth. The contrast and reflection continuity of the depth image is lower than that of the TWT image. We compare the results of 196 interpretations of a single fault with the quality of the seismic image. Low contrast and continuity areas correspond to a greater range of interpreted fault geometries, resulting in a broader spread of fault interpretations in the depth image. Subtle differences in interpreted fault geometries introduce changes in fault characteristics (e.g., throw, heave) that are critical for understanding crustal and lithospheric processes. Seismic image quality impacts interpretation certainty, as evidenced by the increased range in fault interpretations. Quantitative assessments of image quality could inform: (1) whether model-based interpretation (e.g., fault geometry prediction at depth) is more robust than a subjective interpretation; and (2) uncertainty assessments of fault interpretations used to predict tectonic processes such as crustal extension.

Increasing the quality of seismic interpretation

AbstractGeologic models are based on the interpretation of spatially sparse and limited resolution data sets. Nonunique interpretations often exist, resulting in commercial, safety, and environmental risks. We surveyed 444 experienced geoscientists to assess the validity of their interpretations of a seismic section for which multiple concepts honor the data. The most statistically influential factor in improving interpretation was writing about geologic time. A randomized controlled trial identified for the first time a significant causal link between being explicitly requested to describe the temporal geologic evolution of an interpretation and increased interpretation quality. These results have important implications for interpreting geologic data and communicating uncertainty in models.

Interpretational variability of structural traps: implications for exploration risk and volume uncertainty

Abstract Defining the size and shape of hydrocarbon traps is a critical component in estimating the economic value of potential and existing oil and gas fields and is, therefore, a key business risk. Structural traps, defined by fault and fold geometries, form the most common type of hydrocarbon trap, the size estimates of which are based on interpretation of subsurface data, most notably seismic imagery. Interpretation of seismic image data is uncertain, as the subsurface images have limited resolution and quality; in 2D datasets the imagery is spatially limited and the interpretation requires interpolation between images. Here we present data from top reservoir maps created by eight interpretation teams, each of which interpreted a grid of 2D seismic sections at a regular spacing of 1 km, over a 220 km2 area. The resultant maps are compared for interpretation variability. Fault statistics have been generated for each map and compared with analogue datasets to aid in the identification of anomalous interpretations, and to create a likelihood rank for each map. The structural traps identified by each team are compared, and the two largest traps are assessed for their potential trapped hydrocarbon volume. An initial volume and a corrected volume, accounting for potential fault seal breach by reservoir–reservoir juxtaposition across the trap-defining faults, are calculated. The integrated analysis of the multiple interpretations: (a) captures the interpretational uncertainty, (b) determines the likeliness (or risk) of each interpretation being valid, when compared with analogue datasets and (c) assesses the impact of each interpretation on the economic viability of potential prospects (defined by structural traps).

Estimating geological CO2 storage security to deliver on climate mitigation

Carbon capture and storage (CCS) can help nations meet their Paris CO2 reduction commitments cost-effectively. However, lack of confidence in geologic CO2 storage security remains a barrier to CCS implementation. Here we present a numerical program that calculates CO2 storage security and leakage to the atmosphere over 10,000 years. This combines quantitative estimates of geological subsurface CO2 retention, and of surface CO2 leakage. We calculate that realistically well-regulated storage in regions with moderate well densities has a 50% probability that leakage remains below 0.0008% per year, with over 98% of the injected CO2 retained in the subsurface over 10,000 years. An unrealistic scenario, where CO2 storage is inadequately regulated, estimates that more than 78% will be retained over 10,000 years. Our modelling results suggest that geological storage of CO2 can be a secure climate change mitigation option, but we note that long-term behaviour of CO2 in the subsurface remains a key uncertainty.Carbon capture and storage can help reduce CO2 emissions but the confidence in geologic CO2 storage security is uncertain. Here the authors present a numerical programme to estimate leakage from wells and find that under appropriate regulation 98% of injected CO2 will be retained over 10,000 years.

Framing bias: The effect of figure presentation on seismic interpretation

AbstractInterpreters of reflection seismic data generally use images to disseminate the outcomes of their geologic interpretation work. The presentation of such interpretation images can generate unwanted biases in the perception of the observers, an effect known as “framing bias.” These framing biases can enhance or reduce the confidence of the observer in the presented interpretation, independently of the quality of the seismic data or the geologic interpretation. We have tested the effect of presentation on confidence in interpretation of 761 participants of an online experiment. Experiment participants were presented with seismic images and interpretations, deliberately modified in different aspects to introduce potential framing biases. Statistical analysis of the results indicates that the image presentation had a subdued effect on participants’ confidence compared with the quality of the seismic data and interpretation. The results allow us to propose recommendations to minimize biases in the obser...

Can uncertainty in geological cross-section interpretations be quantified and predicted?

This work was undertaken while C.H. Randle held a joint British Geological Survey University Funding Initiative (BUFI) and University of Aberdeen, College of Physical Sciences Ph.D. Studentship at Aberdeen University. The contributions by C.H. Randle, R.M. Lark, and A.A. Monaghan are published with the permission of the Executive Director of the British Geological Survey Natural Environment Research Council. We would also like to thank all those who took part in both experiments as well as the many people who have given input on our results.