Seeing the Unseen with Seismic Imaging

• Activity Number: 146
• Type: Invited
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Terry Young*+
• Companies: Colorado School of Mines
• Keywords: seismic imaging ; 4-D imaging ; inverse problems ; uncertainty quantification

Abstract

Inferences made from earthquake seismograms helped form our early concepts of Earth's interior. Today, we continue to learn from tomography and inversion based on passive recording of earthquakes, while we also investigate the shallow (a few meters) and deep (> 10 km) subsurface with active seismic methods motivated by archaeological studies, environmental concerns, engineering projects, and resource exploration. We use a blend of deterministic and statistical methodologies to acquire, process, and interpret seismic data. Imaging targets at 10 km depth to explore deep hydrocarbon reservoirs can generate enormous volumes of data, requiring processing on the world's largest computers. Large marine surveys can generate a terabyte of data a day and 100s of terabytes for the final data volume. In that context, research topics include 4-D ("time-lapse") imaging, full-azimuth imaging, full-waveform inversion, and joint inversion of multiple geophysical data types. Ideally, the end game consists of quantifying uncertainty and producing images estimating the probability of occurrence of desirable subsurface properties.