WORKFLOW AND PROBLEM SOLVING IN SEISMIC INTERPRETATION

Over the past fifty years, reflection seismology has become an integral tool for visualizing the Earth’s subsurface, and it is a key workforce skill in industries and academic pursuits that use this tool to image subsurface structures to locate resources, such as water, fossil fuels, and ores. Seismic data are often sparse and incomplete, making it necessary for geoscientists to make predictions and interpretations which are strongly influenced by experience, training and expertise. While the techniques and data quality in reflection seismology have been refined over the course of decades, the process of human interaction and successful problem solving approaches with seismic data remain poorly documented and understood.

Ten graduate students participated in this study, and all have seismic interpretation experience and are working toward a career in the petroleum industry. The participants are a combination of geologists and geophysicists at varying levels of ability and experience with their graduate programs. Participants are asked to geologically interpret two intersecting seismic lines over the course of an hour, and had both paper seismic lines as well as digital images of the same lines to work with as they saw fit. The entire exercise was video-recorded from multiple angles to allow detailed observations of workflow, gesture, and annotations made while the participants were engaged in interpretation. Immediately upon completion of the individual exercise, interviews were conducted with each participant to record their narrative of the process.

Interviews and video data were coded for qualitative analysis of actions and patterns in workflow. Preliminary analysis shows divergent workflows between students who identified as primarily geologists or geophysicists. Geologists tended to be drawn to interpretive issues immediately, rather than focus on processing history or artifacts. Geophysicists reported more problems with scale and reaching coherent geological interpretations. Lower-experience participants in both categories tended to fixate on faults observed in the section rather than sedimentary or other potentially relevant features. Additional analysis will focus on workflow timing and process, and on extracting lessons useful for improving instruction in this critical skill area.