← Back to Sessions

Date: 4/19/2021

Session Time: 5:00 PM to 6:15 PM Pacific

Back to the Future: Innovative New Research with Legacy Seismic Data

There has been much discussion in recent years about Big Data and, within the seismological community, how to cope with its ever-expanding volume of digital data. But there exists a source of yet Bigger Data: historical seismic records. With more than a century of seismic waveform data, there is opportunity to resolve intimate details of, and potentially revolutionize, our understanding of Earth dynamics, including phenomena associated with tectonic and geologic processes, seismic sources, climate change and seismic hazard. The challenge: much of the waveform data is tucked away on analog media such as paper, tape or film, or archaic and arcane digital media in holdings that are at risk of being lost forever. These data sets are not only more difficult to physically access and read than their digital counterparts, but often demand innovative approaches to perform any type of modern seismic analysis.

We invite presentations that highlight the discovery, preservation and/or use of seismic datasets spanning multiple decades. Such presentations would include those that address the problems of restoration, digitization and storage of the vast archives of legacy data. We encourage contributions that illustrate the on-going value of legacy data in the general fields of study for which seismographic data have been used and the value of legacy seismographic data in other geophysical disciplines. A few examples include studies of regional or local seismicity, earthquake recurrence and prediction, seismic hazard, climate signatures, inner core rotation and growth and 4D seismic tomography. We also seek contributions that feature efforts in standardizing metadata and image data formats, improving accessibility through rapid scanning, advances in vectorization software and tuned data compression algorithms, efforts in compiling calibrations of seismometers and application of machine learning techniques to directly extract geophysical information from the legacy data.

Conveners

Garrett Euler, Los Alamos National Laboratory (ggeuler@lanl.gov)
Brian Young, Sandia National Laboratories (byoung@sandia.gov)
Ana Aguiar Moya, Lawrence Livermore National Laboratory (aguiarmoya1@llnl.gov)
Thomas Lee, Harvard University (thomasandrewlee@g.harvard.edu)
Qi Ou, University of Oxford (qi.ou@earth.ox.ac.uk)
Richard Lewis, Defense Threat Reduction Agency (richard.d.lewis1.civ@mail.mil)
James Dewey, U.S. Geological Survey (jdewey@usgs.gov)

Oral Presentations