• journal_title：Bulletin of the Seismological Society of America
• Contributor：A. F. Shakal ; H. R. Haddadi ; M. J. Huang
• Publisher：Seismological Society of America
• Date：2006-
• Format：text/html
• Language：en
• Identifier：10.1785/0120050828
• journal_abbrev：Bulletin of the Seismological Society of America
• issn：0037-1106
• volume：96
• issue：4B
• firstpage：S119
• section：Articles

A very high acceleration recording was obtained at station Fault Zone 16, near Parkfield, California, during the M 6.0 Parkfield earthquake of 28 September 2004. This note documents key aspects of the station, the instrument, and the record. A preliminary reconstruction of the record is performed to produce an estimate of the true acceleration, which is then integrated to obtain the estimated velocity.

The strong-motion accelerograph at Fault Zone 16 is deployed at a soil site in a light fiberglass housing common in the Parkfield array. Postearthquake testing indicates that the accelerograph operated normally. Instrument shake-table tests indicate that the recorded acceleration is relatively accurate even at the high accelerations involved in this record. That is, the very high accelerations in the record are a reasonably accurate reflection of the accelerations that occurred at the site during the Parkfield earthquake, although the peak motion was not recorded.

The record itself has several unusual features. The trace was lost at the top edge of the film, where the excursion went beyond the recordable part of the film. Most unusually, the traces of two accelerometers recorded on top of one another during the positive, highest-amplitude part of the record. The accelerograph did not record the peak accelerations of the components, which occurred during this time. An additional aspect is that for the largest amplitude component the negative part of the acceleration was affected by momentary stalling of the recording film during the strongest shaking. The constraint for the mean of the acceleration over the whole record to be zero and detailed analysis of the time trace leads to a reconstruction of the record. The peak acceleration on this component is over 1.8g and most likely over 2g. The peak velocity is estimated to be about 80 cm/sec. The other horizontal component goes off the film at the same time. Combined, the horizontals yield a fault-normal acceleration of over 2.2g and most likely over 2.5g. This record is a challenge to interpret—an important practical lesson is that, given what has been learned about near-fault accelerations, any instrument deployed in the region should have a recording capacity well in excess of 2g.