地震学会ニュースレター連載：強震観測の最新情報（２１）

The California Strong Motion Instrumentation Program and Recent Developments

Anthony F. Shakal (California Div. of Mines and Geology)

The damaging San Fernando earthquake of 1971 in southern California was an important event for California. It showed, for example, that ground shaking during earthquakes could be significantly stronger than expected, and that the motion of buildings during strong shaking was not well understood. One result of the earthquake was the establishment of the California Strong Motion Instrumentation Program, in the Div. of Mines and Geology of the State Department of Conservation. The objective of CSMIP is to record the shaking of the ground and structures in order to provide the essential data needed as the quantitative basis for improvements of building codes.
Construction of buildings in California occurs after the design is reviewed by building officials and certified to meet building codes. A fee is charged for the review, and an incremental fee is added as a strong motion instrumentation fee. This incremental fee was established in law when the CSMIP program was established, and provides its base funding. The law also established an Advisory Committee with members from earthquake engineering and academia. A plan has been developed with portions of the fee targeted for building instrumentation, for ground-response stations, and for instrumentation of other structures (bridges, dams, etc.).
Since 1972 CSMIP has installed over 900 stations, approximately 75% of which are free field, or ground response stations. After the 1989 Loma Prieta earthquake, a focused effort to increase utilization of data recorded from the installed stations was undertaken. A portion of the program funding was allocated to fund studies by engineers and academic researchers. Each year a seminar is held (e.g., Huang, 2000) to review the results of data interpretation studies. These seminars target practicing design engineers so that important results can more quickly be incorporated into structural designs.
Both the 1989 Loma Prieta earthquake and the 1994 Northridge earthquake caused dramatic bridge collapses. These collapses increased realization on the part of bridge designers about the need for quantitative data on the response of structures, and this led to significantly increased support for bridge instrumentation. For example, the well-known Golden Gate Bridge was instrumented with over 75 sensors in 1994. In addition, more than 50 bridges have been instrumented, supported by the Calif. Dept. of Transportation. All of the major toll bridges in the state will be instrumented with up to 100 sensors.
Although the initial and central purpose of CSMIP is to record strong shaking for the purpose of improving building codes and structure design guidelines, an additional value of the data has developed in the last five years. Advances in instrument technology and parallel developments by the program have made it possible for stations to call personal computers at headquarters in Sacramento within minutes after an earthquake and transmit the data recorded. The data then undergoes automated preliminary processing (Shakal et al., 1996). Thus, data that once was not available for days or weeks or more after an earthquake is available immediately, and can be used for guiding emergency response. This emergency response aspect, not envisioned in the original program plan, has made the program and its data more valuable to local and state governments.

Partnerships
TriNet: After the Northridge earthquake of 1994, the new capability of strong motion stations and developments at Caltech led to a cooperative project called TriNet, involving Caltech, CSMIP, and the USGS, as summarized by Mori (1999). A primary goal of this cooperative project is the generation of a map of the shaking intensity, called ShakeMap, rapidly after an earthquake. ShakeMaps are now generated after even relatively small events and available at www.trinet.org.
CISN: Although a major advance, TriNet was limited to the area of the Northridge earthquake in southern California. To extend that effort, the TriNet partners joined with UC Berkeley and USGS Menlo Park to form a statewide consortium. This consortium, known as the California Integrated Seismic Network, builds on and will supplant TriNet, and a Memorandum of Understanding was recently finalized by the agencies. Funding is projected at the state and federal government level.
One of the key elements of CISN is the Engineering Data Center, to be a cooperative effort of CSMIP and the USGS strong motion program. Records of engineering significance produced by any CISN network will be processed for use by engineers and distributed through the Engineering Data Center. The same records will be available via the seismic data centers of CISN in formats used by seismologists, as part of the dual use strategy. Early work for the Engineering Data Center is described in Shakal and Scrivner (2000), and work is scheduled for completion before the TriNet project ends at the end of calendar 2001.

Structural and Subsurface Instrumentation
CSMIP has instrumented over 250 structures, including 170 buildings, 60 bridges, and 20 dams. Each of these structures is extensively instrumented with a comprehensive plan, based on instrumentation objectives developed before instrumentation. All instrumentation in the last eight years uses digital instruments, but structures instrumented earlier have analog film recorders. The conversion of these to modern digital instruments is planned to occur over a period of several years. All new installations use 2 g accelerometers. Since observed structural motion in the Northridge earthquake exceeded 2 g, sensors with up to 4 g capacity are commonly used on most structures.
To study site amplification effects, CSMIP began a downhole instrumentation project in 1989. A method of instrument locking was developed assuring quality control and repeatable re-installation after repair. Eleven subsurface arrays are currently operational and installation of eight new arrays is planned in the near future, many installed with the support and cooperation of the Department of Transportation. Although data recordings so far are low amplitude, recordings from large distant events are proving to be very different than records from nearby events. The expected doubling of amplitudes in the surficial layers is not observed in the distant earthquake recordings, possibly due to surface wave propagation effects. In contrast, records from more local events, for which surface waves are poorly developed, show amplifications of twice or more from depth to surface. More will be learned from this new arena of measurement.
Processed strong motion data recorded at SMIP stations and arrays are available at the CSMIP website: ftp://ftp.consrv.ca.gov/pub/dmg/csmip/.

Summary and Lessons from Recent Developments
- Collaborative projects have become an important part of current approaches in networks. Technology has removed many reasons for traditional boundaries, while the expectation of the public is that networks should be more integrated. This often requires new approaches and skill by network operators.
- The Internet has increased the expectations of the user community for data availability. The availability of news and meteorological data almost immediately, 24 hours a day, leads users, both individual and government, to expect earthquake data rapidly.
- Correspondingly, the immediate usefulness of strong motion data makes it more valuable, and thus increases its likelihood of continued, or even expanded support.
- The utilization of data is important for the full value of a network's efforts to be obtained, and for the investment in the networks to be fully realized.

References

Huang, M.J. (2000).  Proceedings, SMIP2000 Seminar on Utilization of Strong Motion
     Data, Calif. Div. Mines and Geol. Report OSMS 00-02, 124pp.
Shakal, A. and C. Scrivner (2000).  TriNet engineering strong motion data center, 
     p. 115-124, in Proceedings volume above.
Shakal, A., C. Petersen, A. Cramlet, and R. Darragh (1996) Near-real-time CSMIP
     strong motion monitoring and reporting for guiding event response, 11th World 
     Conference on Earthquake Engineering, Paper No. 1566.
Mori, J. (1999)  TriNet: Seismic and strong-motion recording in southern California, 
     Newsletter of Seismological Society of Japan.
「強震観測の最新情報」目次に戻る