Virtual Seminar-The Structure of Oceanic Plates using Machine Learning on Seafloor Vibrations

Join EPS virtually for a seminar. 

Seismic imaging of ocean plate structure provides fundamental constraints on plate formation, evolution, and composition, and is necessary for understanding the role that temperature, hydration, or melts plays in controlling the strength and buoyancy of 70% of Earth’s tectonic plates. While it is easy to decipher vibrations on land and use the data to build maps of the seismic structure of continental plates, on the seafloor, it is harder to detect and isolate vibrations generated from short-period earthquake waves that convert within or underneath the oceanic plate. This is because these signals are often buried within the loud singing of sediments. In this talk, I describe a technique for silencing the singing of sediments and how machine learning (ML) assists in the rapid separation of the noisy, yet predictable ringing from ocean plate conversions. ML can also facilitate the rapid extraction of waves buried within continuous ground vibration i.e., ambient noise using probabilistic inverse theory which produces auto-adaptive maps. Sharper images of ocean plate structure will advance our understanding of plate tectonic theory.

Dr. Tolulope Olugboji is an Assistant Professor at the University of Rochester (UofR). Olugboji is a Furth award recipient (given to promising early-career scientists at the UofR) and a NEF fellow -a select program that recognizes Africa’s best young scientists and technologists. At UR he leads the computational seismology and geophysics group and teaches courses in nature’s fury, earth science data analysis, and seismology (@URSeismo). He is actively engaged in outreach and service to the geophysics research community. He teaches a yearly introductory seismology course for the IRIS-NSF sponsored summer research internships for undergraduates and has been nominated as a 2021 CIG distinguished speaker. His group uses recordings of ground vibrations from seismic sensors across the globe, combined with high-performance computing, to build high resolution images of Earth’s interior.

Tolulope Olugboji (TOE-LOO OH-LOO-BOW-JI)
Office: Hutchison 224B (329) RC Box 270221 Rochester, NY 14627-0221
+1 (585) 276-6609
www.ees.rochester.edu/urseismo
twitter @UofRSeismo

Join the ZOOM https://mcgill.zoom.us/j/87575996613?pwd=c1VvVVI0Y1ZxQ2Y0Tml3MlZWeUtDQT09 Meeting ID: 875 7599 6613 Passcode: 215159