An Update on the Development of the Digitseis Software
Description:
Legacy seismic data are a valuable resource for studying seismically recorded events from the past century and long-term processes such as the earthquake cycle or climate change. However, before legacy data can be processed using modern analysis techniques, they must be converted from their analog form to a digital image and then digitized into time series. Furthermore, working with the analog seismograms will become more difficult with time as they are deteriorating rapidly with age. The enormous number of analog recordings that exist around the world in combination with deterioration and increasing costs associated with storage, mean that many of these seismograms are in danger of being lost.
Efficient and accurate digitization software is crucial to efforts to save and utilize these historical seismograms. DigitSeis is one of several softwares which can take a digital image and convert it into either digital time series or a set of xy-position values of the seismic traces. Since it was first released in 2016, continuous improvements have been made to this software. These improvements have targeted functionality, efficiency, accuracy, and robustness as well as usability.
Here, we overview a sampling of records which showcase some key features and capabilities of DigitSeis. In particular, we examine several records from both the Harvard Seismographic Observatory (HRV) and the World-Wide Standardized Seismographic Network (WWSSN) that demonstrate how these new features allow for easier handling of crossing traces as well as displaying convincing replication of original records. We also present a few records showing features of paper records or their scans that can be problematic for digitization, highlighting the need for careful handling and attention to detail in preservation. Finally, we provide a look at the directions in which DigitSeis is being further developed.
Session: Legacy Seismic Data Collections: The Present State of and Future Outlook for Data from the Past
Type: Oral
Date: 4/18/2023
Presentation Time: 03:00 PM (local time)
Presenting Author: Thomas A. Lee
Student Presenter: Yes
Invited Presentation:
Authors
Thomas Lee Presenting Author Corresponding Author thomasandrewlee@g.harvard.edu Harvard University |
Miaki Ishii ishii@eps.harvard.edu Harvard University |
Hiromi Ishii ishii@eps.harvard.edu Harvard University |
|
|
|
|
|
|
An Update on the Development of the Digitseis Software
Category
Legacy Seismic Data Collections: The Present State of and Future Outlook for Data from the Past