How a regional mailing list facilitated response to an international event

EUGENE S. SCHWEIG, JOAN GOMBERG, PAUL BODIN, GARY PATTERSON & SCOTT DAVIS

Here we explore how the Internet and, in particular, a regional mailing list facilitated coordination of scientific response, communication of results, and exchange of ideas.

Why rapid communication?

The need to communicate rapidly after an earthquake may seem obvious, but what are the goals of such communication? Clearly, one aim is to inform fellow scientists that the event has occurred and to provide them with as many details as possible so that the response can begin quickly and efficiently. If the data include potentially ephemeral field observations, rapid mobilization may be critical. In the case of major earthquakes, near-source seismic recording of aftershocks may provide valuable information about the main shock and its effects. Because the rate at which aftershocks occur decreases with time, recording equipment must be deployed quickly. In the case of the Bhuj earthquake, June monsoons promised to erase much of the surface evidence of the event.

When resources are limited, views about what data are most important and how to respond must be exchanged as quickly and efficiently as possible. To avoid duplication of effort and to ensure that further observations are made in the case of critical or ambiguous data, news of the response and rapid dissemination of initial observations are necessary. The case of the missing surface faulting from the Bhuj earthquake illustrates this need. Earthquakes of this magnitude normally rupture the Earth's surface, and the characteristics of the breakage constrain models of how and why the earthquake occurred. Other earthquakes of comparable size have produced fault scarps several metres high, in places where they are easily recognized. This was not the case for the Bhuj earthquake. Much of the epicentral region lies within the Great Rann of Kachchh, which is a vast, uninhabited, inhospitable, salt-covered plain that floods annually. The adjacent Banni Plains are only slightly more accessible and hospitable. Nonetheless, field parties from around the world flocked to the region, and the Internet became a key mechanism for exchanging information about working conditions, territory covered, and discoveries (or lack of). As a result, most of the area was searched and most scientists concluded that the earthquake indeed produced no surface faulting.

The advent of the Internet, including e-mail, newsgroups and web pages, has provided an unprecedented opportunity to exchange information. However, news from unqualified or crank sources can also creep into the mix. Thus it becomes important to distinguish credible sources from less reliable ones.

Characterizing communications of scientific events

Whether communicating the results or initiation of a scientific experiment, a new discovery, or the occurrence of a natural disaster, the benefits and effects of different modes of communication can be measured. These measures include the time lag between release and receipt of the information by interested parties and the effect of the information on the science and the public.

•Peer-reviewed professional journals and professional meetings
Publication of results in peer-reviewed journals typically takes months or more. Recognizing this, many scientific organizations send out newsletters or use other mechanisms to fast-track important news events or time-critical results. Professional meetings commonly require long lead times between submission of abstracts or conference papers, but many organizations set up special topical sessions for time-sensitive events. Both the American Geophysical Union and the Seismological Society of America organized special sessions on the Bhuj earthquake for their spring meetings. Such special sessions can be organized in a matter of weeks. Participants in these sessions are typically experts in the field or closely related ones and, occasionally, science journalists who monitor the journals. Peer-reviewed articles and meetings advance science by providing a forum for information to be exchanged. They also influence the public indirectly by educating scientists and by informing scientific news reporters.

• Popular science magazines
Journals such as Discover Magazine and New Scientist fall into the category of magazines aimed at the educated layperson. The lag between the time of the event and its publication may be somewhat shorter than for a peer-reviewed journal. The audience is broader than that for peer-reviewed journals, and will include scientists as well as the educated public. Exchange of ideas, however, is not the primary purpose of this medium. News is communicated from scientists to reporters and then to the public as recipients of information. Most of the information contained in the articles will already have been reported at professional meetings and in scientific journals; the primary purpose of these outlets is to educate the public.

Popular press
This category includes newspapers and the traditional electronic news media of television and radio. To report the results of a study or an experiment can take weeks or months. The press usually receives the news through press briefings at professional meetings or upon publication of the results in a peer-reviewed journal. When natural disasters occur, the press often arrives first on the scene and may inform the scientists of the event, thus facilitating a rapid scientific response. Indeed, with the exception of the few US seismologists who carry earthquake pagers, even the most interested scientists first heard about the Bhuj earthquake when they were awakened by the news on their clock radios or while drinking their morning coffee and listening to the radio or TV news. Reporting of events by the popular press makes science accountable and responsive to public needs. However, science is sometimes misreported or oversimplified, resulting in misinformation of the public and of scientists who depend on press reports for accurate information. For example, public panic ensued in the Gujarat city of Ahmedabad when an Indian scientist reported that major earthquakes were marching eastwards and that the next one would strike Ahmedabad. Undoubtedly, the scientist's comments had been taken out of context.

The advent of Internet editions of newspapers has influenced earthquake science in another way. They provide almost real-time data on earthquake effects that would otherwise take many months to compile and analyse. Maps of earthquake effects, showing the extent of the damage and where the earthquake was and was not felt, offer important clues about the earthquake process (such as its depth and the orientation of the fault that broke). Immediately after the Bhuj earthquake, a group of US scientists began combing the Internet editions of newspapers published in India and the surrounding countries for reports on the earthquake's effects. Within days they had completed a preliminary map that showed the extraordinarily large area affected by the earthquake. Previous studies suggested that this result could have been anticipated, but the reasons for it are not fully understood and have already become the subject of new studies.

• Word of mouth and standard e-mail
Rapid communication between individual scientists is conducted in the form of telephone conversations and e-mail, the latter certainly being the most common. Communication by telephone is instantaneous; e-mail communication is almost instantaneous, being slowed by Internet traffic (and hampered by those who do not read their e-mails!). Participants are usually scientists who are directly involved in responding to the event or who generated the results, although news reporters may occasionally be notified. These modes of communication facilitate a quick response to the event, but initially only by the limited group involved in the exchange. Unless the press picks up these exchanges, the public remains unaware of the situation.

• Mailing lists and newsgroups
Mailing lists and newsgroups are similar in that both are internet-based methods of general discussion among large groups. A mailing list is a system that allows people with common interests to send e-mail to one address, whereupon their message is distributed to anyone who has subscribed to the list. Mailing lists may be moderated or unmoderated. In unmoderated lists, people can submit messages to the list and they are automatically posted to all subscribers with no monitoring or editing. Moderated lists, however, allow a moderator to monitor the submissions for offensive material, ensure that the messages are on the topic of interest, or edit messages for clarity.

Newsgroups are similar to mailing lists in that users can submit messages to a group. However, the message is not e-mailed directly to the subscribers; it is posted to a newsgroup on the Usenet, a network of computers around the world that transmits the postings. Newsgroups are more like bulletin boards, in that they are public and anyone can log on to a local news server and browse the postings in any group. There is no alert that a message of interest to a user has been posted.

Generally, for an unmoderated mailing list, the lag between submission of news and its receipt by subscribers is a matter of minutes. However, if the list is moderated, messages are not posted until the moderator has read and posted the message, and in newsgroups the posting must circulate through the Usenet system, which may take hours. Participants in mailing lists may be restricted or not, and may include scientists, the press, the public and any subgroup. Newsgroups are public and thus messages can be broadly monitored. In both mailing lists and newsgroups, there are no geographical barriers to communications, and both methods provide a means for an effective, timely response.

A specific example: the Central United States Earthquake Hazard Mailing List

In the central United States, earthquake zones, most notably the New Madrid and Wabash Valley seismic zones , span several states. If a major, damaging earthquake occurs, as many as 12 states may be affected, each with its own geological survey, emergency-management agency, transportation department, and so on. To deal with this, the state emergency services formed the Central United States Earthquake Consortium (CUSEC) in 1984 and the Association of CUSEC State Geologists in 1992. Many people felt, however, that it was necessary to have an ongoing means of exchanging ideas on the reduction of earthquake risk, and of distributing news and announcements. By late 1998, people concerned with these topics had access to e-mail, thus allowing the design of an Internet-based system for exchanging news and ideas. In February 1999, we launched the Central United States Earthquake Hazard Mailing List. We opted for a moderated mailing list, having seen unmoderated earthquake newsgroups degenerate into a deluge of false earthquake predictions and uninformed postings, eventually causing professionals to tune out.

The mailing list provides a simple and effective way to disseminate ideas to a broad range of people, including those outside of our own fields — anyone can subscribe. More than one moderator monitors submissions, which are rarely delayed for more than overnight. Generally the time lag between submission and posting is no more than a few hours; postings on important events are usually made within minutes. The mailing list currently has over 570 subscribers; a breakdown is shown in. At least 38% are from educational institutions; about 22% are from the US Geological Survey and individual state geological surveys. Over 16% of the subscribers are members of groups with which scientists traditionally do not communicate, such as emergency managers, departments of transportation, financial institutions and utilities. Some of the discussions that have taken place as a result of this mailing list have reflected this diverse population to the benefit of both groups.

The Bhuj earthquake

The Bhuj earthquake occurred in the same region as the 1819 Kutch earthquake, which measured 7.7 on the Richter scale and is often cited as one of the few large, historical, intraplate — occurring away from plate boundaries — earthquakes. Earthquakes in the central United States also fall in this category, although the region has not experienced a major earthquake since 1812. The classification of the Kutch and Bhuj earthquakes is contravercial. However, the geological environment and many of the characteristics that influence the hazard (i.e. how hard and how often the ground shakes) in the region of the Bhuj and Kutch earthquakes seem to be similar to those in the central United States. Thus, the Bhuj earthquake provided a rare opportunity to study such events using modern analytical tools.

Many scientists first heard about the Bhuj earthquake through either the popular press or one of several mailing lists that notify subscribers of earthquake events. The moderators of the Central United States Earthquake Hazard Mailing List posted the occurrence of the earthquake as soon as they had been notified by the USGS National Earthquake Information Center in Golden, Colorado. Scientists immediately recognized the relevance of the Bhuj earthquake to understanding intraplate earthquakes in general, and to the central United States in particular. In the wake of this earthquake, traffic on the Central United States Earthquake Hazard Mailing List shot up immediately , and its multipurpose utility quickly became apparent.

Within a few days of the earthquake, several teams departed for the field, and the mailing list became a conduit for information about field studies. Although the Internet is not available throughout most of the epicentral area, initial field reports were submitted through a telephone modem, and these early communications provided guidance for subsequent teams. Interpretations of several field observations were debated on the mailing list (such as those related to the search for surface rupture and to the reasons for the observed damage patterns), as was the issue of whether the Bhuj earthquake was truly an intraplate event. Subsequently, the list provided a mechanism for disseminating and archiving field reports.

The mailing list also became a clearing house for background information related to the earthquake. Participants contributed references to favourite scientific publications, data sources and websites. It also allowed scientists to keep track of the interest and perceptions of the public, as participants noted discussions published in their local newspapers.

The Mailing List seemed to provide a new forum for scientific debate. Emotional exchanges flew across the wires, debating a range of fundamental scientific questions: Was the Bhuj earthquake interplate or intraplate? Should we set aside the framework of plate tectonics in order to understand this type of earthquake? Can knowledge gained from study of the Bhuj earthquake be transferred to other regions? How can we capitalize on lessons learned from an earthquake in India to educate the public in the United States and elsewhere?

Finally, although traffic certainly has tapered off with time, the mailing list still serves to inform its participants about analyses of data collected in India, continuing fieldwork, and upcoming meetings that are relevant to the Bhuj earthquake.

Conclusions

The Central United States Earthquake Hazard Mailing List has proven to be a simple and effective means of communicating scientific events, and also provides a moderated forum for real-time scientific debate. The cost and maintenance of such a list can be minimal, depending upon its level of sophistication, and the cost of data storage is currently so low that archiving of mailing-list exchanges is negligible. Given the pace of change in the cyber world, new and more effective means of communicating will undoubtedly emerge. For now, however, we encourage other groups to take advantage of this exciting information medium.

---

References

1. Frankel, A. et al. US Geological Survey Open-File Report 96-53 (1996).
2. The USGS has several such services, including BIGQUAKE.

---

Eugene S. Schweig and Joan Gomberg are at the US Geological Survey, 3876 Central Avenue, Memphis, Tennessee 38152-3050, USA. e-mails: schweig@usgs.gov; gomberg@usgs.gov.
Paul Bodin and Gary Patterson are in the Center for Earthquake Research and Information, University of Memphis, 3876 Central Avenue, Memphis, Tennessee 38152-3050, USA. e-mails: bodin@ceri.memphis.edu; patterson@ceri.memphis.edu.
Scott Davis is at Clifty Webworks, 1901 Wells Drive, Madison, Indiana 47250, USA. e-mail: scott@clifty.com

Nature © Macmillan Publishers Ltd 2001 Registered No. 785998 England.