Old seismograms, which are records of the earth’s vibrations, hold a wealth of information that could be beneficial to today’s work in nuclear nonproliferation, earthquake modeling, oil and gas extraction, and even climate-change modeling.
But there’s one main problem: these decades-old seismograms are film-based. As such, the data are rarely exploited because of the difficulty in converting the single snapshot images of the analog archives to more useful digital data, in which each seismogram line is made up of individual dots with a value and a time stamp.
Enter Retriever Technology, LP, a company that provides imaging software and hardware for scientific and industrial users. Andy Bartlett of Retriever Technology wants to develop a process to digitize seismograms so it’s possible to elicit usable information from the historical data. To tackle the problem, Bartlett approached NMSBA, which set him up with Bill O’Rourke of Sandia National Laboratories.
Tapping into Sandia’s signal and image processing expertise as well as geophysical proficiencies, O’Rourke consulted with Retriever Technology on suitable image and signal processing techniques to convert seismograms into usable digitized formats. His guidance provided Bartlett with an understanding of how image processing could be used to begin to solve conversion problems from film-based seismograms into digital data.
With Sandia’s assistance, Retriever Technology secured a $1 million Small Business Innovation Research (SBIR) Phase II award for future research into possible solutions.