To sign up for this event, please register here.
The event will take place as a hybrid event, meaning online and in-person participation will be offered. All participants will receive a Zoom link in advance, those able and willing to participate in person are welcome to join us at Eilert Sundts hus (Meeting Room 629) at the UiO Blindern Campus.
Speaker: Dr Steven Gibbons, Senior advisor, Norwegian Geotechnical Institute (NGI)
Discussant: TBD
Abstract:
The Arctic islands of Novaya Zemlya were the sites of both atmospheric and underground Soviet nuclear tests from the mid-1950s until October 1990. Following the Partial Test Ban Treaty of 1963, monitoring focus shifted to the detection and measurement of underground weapons tests and the field of forensic seismology was born. We will examine seismic signals from Novaya Zemlya nuclear tests over the years in the context of how instrumentation and global station coverage has improved. The remote Arctic location of Novaya Zemlya presents challenges for seismic monitoring of very low magnitude seismic events with over 1000 km separating the Matochkin Shar nuclear test site and the closest monitoring stations. Fortunately, the geology of the Barents Sea transmits seismic signals with exceptionally little loss of high frequency energy, meaning that the seismic networks of Northern Europe would detect and locate confidently an explosion with yield well below 1 kT. At the lowest seismic magnitudes, the detection threshold is defined by two seismic arrays of the International Monitoring System for verifying compliance with the Comprehensive Nuclear-Test-Ban Treaty. We describe how seismic array processing helps us to detect and locate seismic events and stress the importance of preserving and exploiting data from historical seismic events to help calibrate event locations. Machine Learning offers exciting new possibilities for detecting and locating small seismic events on and around Novaya Zemlya using the available historical data.
Speaker bio:
Steven J Gibbons graduated from the University of Leeds in 1999 with a PhD in computational geophysics, specializing in numerical simulations of geophysical and astrophysical magnetic field generation. Between 2002 and 2019, he worked at the seismological observatory NORSAR performing geophysical research and supporting Norway's technical contributions to the verification regime for the Comprehensive Nuclear-Test-Ban Treaty. There he specialized in signal processing and pattern recognition to detect, classify, and interpret seismic signals, with special focus on array stations and real-time signal detection and identification. He has explored the limits of matched filter detectors, that recognize weak seismic signals - often far below the noise level, and so-called precision seismology, to locate seismic events with unprecedented accuracy. Since 2019 he has worked within Natural Hazards and Offshore Energy at the Norwegian Geotechnical Institute (NGI) specializing in High Performance Computing and numerical simulations of geophysical flows in natural hazards. Areas of focus include Probabilistic Hazard Analysis, cohesive landslides, Urgent Tsunami Computation for civil protection, and machine learning for rapid hazard estimates. He still follows closely developments in forensic seismology and is passionately interested in open science and increasing public awareness and understanding of available scientific data. He sits on the editorial board of "The Seismic Record", responsible for nuclear explosion monitoring.