Executive Committee Meets in Thatcham

Executive Committee members Jens Havskov and Akio Yoshida with ISC staff Dmitry Storchak, Mamy Andrianirinna and Maureen Aspinwall.

The ISC's Executive Committee held its annual meeting this year at the ISC in early June. The Committee was at "full strength", with the participation of all elected and ex-officio members (http://www.gxjydl.com/execcomm.html). Dr Chris Browitt of the BGS joined in his first full meeting, after being elected at the 1999 Governing Council meeting.

The Committee noted the increasing use of data collection thresholds that is described in the 1999 Director's Report (http://www.gxjydl.com/doc/report/1999DR.pdf). They supported plans to use in preference the new data management system to select the most important events for re-analysis after collection. The Committee urged the Centre to eliminate collection thresholds as soon as possible, to gather all the data it possibly can and make its entire collection available on-line.

The Committee was pleased by the continued strong demand for ISC data in computer-readable form and noted that CD sales have outnumbered sales of the printed Catalogue in each of the last few years. They encouraged the Centre to continue seeking the best ways to deliver data to as many users as possible.

Turning to the future, the Committee noted progress in using S and PKP times, as well as 3-D earth models, to compute locations. They urged the Centre to plan to update the methods and data used for locations in the Bulletin, perhaps in stages and for earthquakes as soon as January 2000. They suggested that the ISC might encourage users to contribute more secondary arrival data, even if only for a few, selected events.

They also discussed the advantages of reducing the fraction of events without an ISC magnitude by computing an ISC ML. There was some concern, however, that adequate accounting for variability of attenuation would be difficult.

For the long term, the Committee noted that the Centre should be able to take advantage of waveform data to support its core mission of producing an accurate earthquake Bulletin. In the future, ISC seismologists could review waveforms of particularly troublesome earthquakes or the ISC could compute depths of selected events by waveform fitting, perhaps using the best hypocentral depth from traditional analysis as a starting point for a new set of iterations.

ISC at the ESC Meeting

ISC staff will present three talks at the European Seismological Commission in September in Lisbon on the theme of computing better locations. Dmitry Storchak and Mamy Andrianirina will talk about improving locations by calculating them using arrival times from a wider range of phases, while Qi-fu Chen will discuss the effects from using modern 1- and 3-dimensional earth models. See the following links for information about their talks:

• Dmitry
• Mamy
• Qi-Fu

ISC at the EGS Meeting

Ray Willemann presented a poster at the European Geophysical Society meeting in April in Nice. The poster (available at http://www.gxjydl.com/doc/analysis/2000p03/egs00.pdf) showed the new services available from the ISC web site. Users of the on-line Bulletin (http://www.gxjydl.com/search/bulletin/) selecting "include links" now see icons in the Bulletin that can be used to retrieve specific waveform data sets from ORFEUS, the IRIS DMC and elsewhere, station data from the NEIS and FDSN station books and moment tensors, source time functions and other earthquake parameters from universities and other data centres. We welcome suggestions of further useful links.

Survey of Users

The ISC is conducting a survey to help focus the ways in which data are delivered to users. Everybody who uses ISC data is encouraged to respond, preferably using the on-line form at http://www.gxjydl.com/survey.html. For those without web access, the survey is also available by e-mail from admin@isc.ac.uk, post or fax. A complimentary copy of the 1997 Catalogue CD (i.e., without the phase data included in the Bulletin CD) is offered to each person returning a completed survey.

1997 Bulletin Completed

The ISC finished analysing events from December 1997 in April, just 10 months after December 1996. Bulletin data were posted to the web straightaway, and the Bulletin CD and printed Bulletin were distributed in May.

September-December 1997 included the now well-known Umbria-Marche, Italy, sequence. For all M=3 (ING) events in the sequence, the ISC associated readings from other networks and computed residuals with respect to the epicentre that best fit the joint data set. The ING Bulletin continues to be the definitive source for readings associated with smaller events, which were rarely recorded outside of Italy.

As the Umbria-Marche sequence tapered off, we faced a large sequence following the 5 December MW 7.7 Kamchatka earthquake. Thanks to readings and epicentres from the Kamchatka Regional Seismic Center, ISC editors could properly associate readings from stations in Alaska, the Aleutians, Japan, Europe, Australia and elsewhere with many events of this sequence.

Since 1974 the ISC has searched among unassociated phase readings to discover unreported events. In the last few years, however, a growing fraction of these events have been located in well-monitored regions, including Alaska, California, Europe, Japan, Kamchatka, Kuril Islands and New Zealand. For events from mid-1997 we stopped analysing and reporting "search" events in these regions, since they generally represent re-discovery of events known locally, but not reported to the ISC because they are too small or of non-seismic origin.

Instead, we now concentrate on areas where data from several networks are required to locate some events, including Central and South America, the Balkans, the Middle East, southeast Asia and the oceans. The map below shows the change in the distribution of search events before and after the new procedures were introduced. No amplitudes are reported for most "search" events, but of those with at least one amplitude, we computed an mb = 4.0 for nearly half.

In recent years ISC has received data from Indonesia only through NEIS, and then only for selected stations and events. As reporting for the region has grown better, the Bulletin has included an increasing number of hypocentres in Indonesia, but most were computed from teleseismic readings alone.

The situation improved dramatically when the Indonesian Meteorological and Geophysical Agency (MGA) began e-mailing bulletins directly to the ISC. MGA station data are included in the ISC Bulletin for events from February 1997. Used along with data from Australia and elsewhere, they proved to be useful for more accurate locations in and near to Indonesia. Selected MGA epicentral solutions were introduced in July 1997 with agency code DJA, which last was last seen in the ISC Bulletin in 1992. In many instances the MGA hypocentre proved to be a good starting point for the ISC location. As several users have noticed, however, DJA magnitude estimates are systematically larger than the ISC's final value.

The geographic distribution of "search" events, those found by the ISC without a reported hypocentre, changed from mid-year
when the ISC began concentrating searches on regions where genuinely new events were likely to be found.