March/April 2005

Identifying and Saving Seismology's Heirlooms

By the time this Opinion appears in print, readers of SRL will have received an invitation to join the SSA Legacy Data E-Community. Many of you have an interest in old seismological data--seismograms, bulletins, macroseismic observations, and the like--and you appreciate the value of these data to seismology. Joining the Legacy Data E-Community is easy, and your membership is critical to its success. The more subscribers to the list server, the more likely an article will connect with someone who will act on it. So even if you can't imagine yourself submitting an article in the near future, join the e-community as a list-server subscriber and read the articles as they are posted. Simply navigate to the "Members" page from the SSA home page (http://www.seismosoc.org/) and follow the link to the "Legacy Data E-Community." We hope, of course, that at some point you will feel inspired to actively contribute your own special knowledge to the e-community or to submit an inquiry that might bring out the still-hidden knowledge of others. However you feel about the idea of submitting an article to the e-community, your presence in the audience will be important.

The fact that a data collection is considered a "legacy" suggests that it has in some sense been left behind by the person or program that assembled it. The bequeathing of data, or a data set's obsolescence with respect to the seismological goals that motivated its original acquisition, leads to a host of problems. Many of these problems are fundamentally communication problems. As formerly "hot new" data mature into "legacy" data, knowledge of the data fades in the broader seismological community and remains strong only in special-interest groups that may be unaware of one another. Those who may have exciting new uses for these data, undreamed of at the time of data collection, may not even know that the data exist or how they can be accessed. The seismological community's skill at interpreting the old data and the community's knowledge of the strengths and limitations of the old data inevitably decrease as the community understandably focuses on newly collected data. Institutions possessing important data may no longer employ researchers who understand the importance of the data and can advocate on the data's behalf. An effective electronic community of those with interest in some aspect of legacy data should go a long way toward solving these problems.

You can access the current archive of list-server articles through the Legacy Data E-Community Web page. When these words were written, the archive had been "seeded" with seven articles resulting from beta-testing of the list server by members of SSA's Executive Committee, Board of Directors, and Ad Hoc Committee on Digital Archiving of Historical Seismological Data. Paul Richards described an archive of digital seismograms for 711 underground nuclear explosions recorded from 1966 to 1996 at the Borovoye Geophysical Observatory (BRV) in Kazakhstan. Antonio Villaseñor provided a link to his new electronic catalog of hypocenters listed in the International Seismological Summary and in bulletins of the British Association for the Advancement of Science for the period 1914-1963. Diane Doser gave an overview of her collection of digitized waveforms and provided references to journal articles that effectively catalog the collection. John Ebel summarized holdings of original seismograms at Weston Observatory. Roger Musson reported on the IASPEI Working Group on Seismological Archives, which is in the process of documenting the existence of station bulletins and original seismograms from seismological observatories that were operational in 1920 and earlier. We (Bolton and Dewey) submitted articles on USGS collections of World-Wide Standardized Seismograph Network (WWSSN) film chips and microfilms of historical seismograms. Our hunch is that even readers with long-standing interests in reanalysis of premodern earthquakes will learn about new, potentially useful, data sources from a scan of the list-server archive as it currently exists.

The identification and publicizing of legacy data that might not be known to potential users will continue to be a focus of the Legacy Data E-Community. Many important data sets are not yet covered by entries in the list-server archive, and we hope that those knowledgeable about these data sets will come forward through the e-community and describe them. The potential to "rediscover" largely forgotten data sets is real and exciting. The list server, in addition to being a tool for tracking down and publicizing real data sets, should also aid in dispelling legends about the existence of data sets that do not exist, or that exist in much humbler forms than as depicted in legend. We see other important functions for the e-community besides documenting existing holdings of legacy data. For example, it would be a fine forum for sharing strategies for optimally preserving analog data in digital form. We also expect that the list server will attract information helpful for interpreting legacy data, such as information on seismograph response and lore about individual scientists' conventions that were used in calculating magnitudes or assigning intensities.

A journal paper that makes use of seismographic legacy data commonly does not have as its principal object the dissemination of information about the data. Nothing in the title, keywords, or abstract associated with such a paper may indicate that the paper has several paragraphs dealing with a particular interpretational problem associated with the legacy data. Conceivably, a paper's authors will have faced particular problems peculiar to old data and devised practical solutions that they do not describe in detail in the paper. We therefore encourage authors who have published papers using legacy data to submit articles to the list server that direct attention to the legacy data aspects of their papers and that elaborate on sections of their papers that deal with practical aspects of acquiring and using the data.

In closing, we wish to thank those who have already joined the Legacy Data E-Community as subscribers. We are especially grateful to those who have contributed articles to the list server. And we again appeal to the procrastinators among those interested in legacy data. If you have not yet done so, it is time now to join the SSA Legacy Data E-Community and participate in this effort to identify, save, and further utilize seismology's precious heirlooms.

James W. Dewey and Harold Bolton
U.S. Geological Survey
Federal Center MS 966
Denver, CO 80225

To send a letter to the editor regarding this opinion or to write your own opinion, contact the SRL editor by e-mail.



Posted: 23 June 2005