Mathematical Geophysics Opinion

Why Seismic Networks Need Digital Object Identifiers

In a move to give credit where it's due, the International Federation of Digital Seismograph Networks will link digital object identifiers to data from seismic networks and project deployments.

By , , , , , , , and

In science, peer-reviewed, published journal articles are the fundamental measure of output and the primary means of career advancement. As scientists, we all absorb, as part of our professional training, a nuanced understanding of what it means to publish and be cited in different journals and how to evaluate metrics like h-indices and impact factors to assess our and other scientists’ prestige and productivity.

However, not all scientists have equal access to this publication-driven reward system. In particular, data providers are often left out. Data almost always serve as crucial components of any scientific study, and transparent, reproducible research requires open, permanent access to these source data.

Producing and making available such data are legitimate and important scientific activities, yet data providers do not receive the same recognition as data users, with traceable citation information facilitating measurement of the impact of their data for their funding providers, tenure and promotion committees, and others [Data Citation Synthesis Group, 2014].  We hope that assigning seismic data networks a universal and easily cited digital identity—the digital object identifier, or DOI—will help bring data providers the recognition they deserve [International DOI Foundation, 2012]. We believe that such a step is long overdue.

Unheralded but Critical

As seismic networks become larger and more numerous, they increasingly provide the seismology community with ever-growing troves of waveform data. These data form the basis of important studies and thus need treatment as independently citable objects.

The good news is that the scientific community generally recognizes this. The bad news is that current citation and acknowledgment practices vary widely, often omitting data providers, and it is often unclear which reference, if any, is preferred for a given network.

The Community Responds

In response to this murky situation, which often prevents seismic data providers from receiving recognition, the International Federation of Digital Seismograph Networks (FDSN) has recommended the attribution of a digital object identifier to each seismic network [FDSN, 2014]. This recommendation follows discussion between the seismological data centers within the European Integrated Data Archive and Incorporated Research Institutions for Seismology Data Management Center  about possible methods for the generation, maintenance, and promotion of persistent identifiers (PIDs) for seismic networks.

Unique PIDs make consistently citing and acknowledging seismic networks easier for users of seismological data. PIDs offer network operators, data centers, and individual scientists a straightforward way to measure the scientific impact of the data they produced, archived, and distributed.

DOIs are a popular type of PID, and we considered them most suitable for acknowledging seismic networks properly at present because they are already in operation: DOIs are well known in the scientific community and widely accepted by publishers. Moreover, good DOI metadata, describing geolocation, time frames, and data types in a generic format, assist users outside seismology to discover data using search tools such as the DataCite service.

How It Will Work

Since 2014, FDSN and data centers have linked DOIs to individual temporary experiments and named permanent seismic networks having FDSN-allocated network codes. Seismic network operators can choose for their network a DOI issued by any DOI-minting agency, such as a national library, their hosting data center, or the FDSN.

The FDSN recommendation requires that author, publication year, title, and publisher information be included in DOI metadata, following the DataCite model of data sets [DataCite Metadata Working Group, 2013]. This basic information is needed for citing an object, whether it be a data set or traditional article, in a published scientific article.

The citation text should be prepared automatically from the DOI metadata in a consistent format. As examples, we are now able to cite existing networks [GEOFON Data Centre, 1993; Asch et al., 2011; Scripps Institution of Oceanography, 1986] in the American Geophysical Union’s house style with their functioning DOIs, as shown in the reference list to this article.

Additional generic metadata, including geographical information, organizations and individuals involved in the distribution chain, and links to resources such as publications and technical reports can be provided in the DOI metadata for a particular network. The best-known use of the DOI system is termed “resolution,” which takes users to a landing page on the Internet that houses metadata along with some pathway to the data themselves.

For a seismic network, FDSN recommends that a landing page include at least a station list with locations, channel information, periods of data collection, and information on how to retrieve the data (see an example here for the GEOFON network, a German global seismological broadband network). FDSN has created a public DOI registry that will assist authors in finding DOIs associated with one or more specific network codes.

Another FDSN service will prepare standard citation texts for networks. DOIs and other persistent identifiers are being included in seismological metadata (StationXML), making discovery of seismic network DOIs easier and supporting future persistent identifier usage.

A Cultural Change

We encourage network operators to consider the implications for their networks and to ensure, with the possible assistance of FDSN, that DOIs are issued with associated landing pages and helpful metadata, facilitating their discovery. We expect that citation of seismic network DOIs will gain steam as more operators provide DOIs and produce high-quality landing pages. Authors can then cite seismic data as they do traditional articles.

In practice, authors need only provide a list of network codes and years for the seismological data used to the FDSN-operated Web service. The resulting citation list can be included in the scientific paper reference list. The next steps required remain largely cultural, as scientists learn to cite data sources as standard practice.

Editors and referees should check that submitted manuscripts reference seismic networks appropriately, ideally using a data citation as outlined above, just as they presently check that literature is cited correctly. Journals are encouraging data citations and should drop arbitrary restrictions on the maximum number of allowed references. Online-only citations could help avoid unwieldy reference lists.

A Huge Step Forward

Assigning DOIs for seismic networks and project deployments is a major step toward agreement within the seismology community on common standards for acknowledging large data objects in seismology. Reproducible, verifiable research requires precise identification of the data used.

Seismic networks will be the parent objects needed for more finely grained citation of specific channels and time windows in the future. We anticipate that the growing availability of DOIs for seismic networks will assist the seismology community and encourage other geoscience groups to create citable source data sets.

Ultimately, the new practice will give data providers long-overdue credit for their scientific contributions. Such change can only encourage the generation of more high-quality data.


Asch, G., F. Tilmann, B. Schurr, and T. Ryberg (2011), Seismic network 5E: MINAS Project (2011/2013), other/seismic network, Dtsch. GeoForschungsZentrum GFZ, Potsdam, Germany, doi:10.14470/ab466166.

Data Citation Synthesis Group (2014), Joint declaration of data citation principles—Final, FORCE11, San Diego, Calif. [Available from]

DataCite Metadata Working Group (2013), DataCite metadata schema for the publication and citation of research data, DataCite, Hannover, Germany, doi:10.5438/000808.

GEOFON Data Centre (1993), GEOFON Seismic Network, other/seismic network, Dtsch. GeoForschungsZentrum GFZ, Potsdam, Germany, doi:10.14470/TR560404.

International DOI Foundation (2012), DOI handbook, version 5. [Available at]

International Federation of Digital Seismograph Networks (FDSN) (2014), FDSN recommendations for seismic network DOIs and related FDSN services [WG3 recommendation], doi:10.7914/D11596.

Scripps Institution of Oceanography (1986), IRIS/IDA Seismic Network, other/seismic network, Int. Fed. of Digital Seismograph Networks, doi:10.7914/SN/II.

—P. L. Evans and A. Strollo, Deutsches GeoForschungsZentrum GFZ, Potsdam, Germany; email: [email protected]; A. Clark, T. Ahern, and R. Newman, Data Management Center, Incorporated Research Institutions for Seismology, Seattle, Wash.; J. F. Clinton, Swiss Seismological Service, Eidgenössische Technische Hochschule (ETH) Zürich, Zurich, Switzerland; and H. Pedersen and C. Pequegnat, University of Grenoble Alpes, ISTerre, Grenoble, France; also at Centre National de la Recherche Scientifique, ISTerre, Grenoble, France

Citation: Evans, P. L., A. Strollo, A. Clark, T. Ahern, R. Newman, J. F. Clinton, H. Pedersen, and C. Pequegnat (2015), Why seismic networks need digital object identifiers, Eos, 96, doi:10.1029/2015EO036971. Published on 8 October 2015.

© 2015. The authors. CC BY-NC 3.0