Instrumentation and Station Installation

Standard field procedure was modified somewhat, due both to the exigencies of working in an urban area, and our desire to co-site GEOS with permanent strong-motion accelerographs operated by the U. S. Geological Survey (Etheredge and Porcella, 1987, 1988), the California Division of Mines and Geology (Shakal and others, 1987), and the University of Southern California (Anderson and others, 1981). For reasons of security, most instruments were emplaced either within structures or on private property adjacent to structures. Outdoors, sensors were buried in soil. indoors, sensors were initially set on, but not otherwise attached to, building floors. The large aftershock at 277 10:59 was recorded this way, and preliminary examination of the recorded motions suggested that at least one indoor sensor (the fba at station ELM) might have slid during the earthquake. Between 279 19:00 and 280 00:00 all indoor sensors were affixed to floors using window glazing putty - a glob of putty was placed under each sensor foot, and sensors were pressed down until firm contact with the floor was established. Good sensor-floor contact appeared to be maintained in each case for the duration of the experiment. (We visually examined each indoor site after the large aftershock, and found no direct evidence of sliding. We have not examined the pre-putty data in detail for other evidence of sensor sliding, which might be detectable by comparing integrated fba records with response-corrected geophone records.)

Clock Corrections

We encountered several problems in the field at Whittier Narrows that made it difficult to sort out timing relationships. First, WWVB reception was generally poor during the Whittier Narrows study, presumably because most of the recorders were installed in or near buildings. A tendency to compensate for poor WWVB reception by taking more measurenients (say, by decreasing the programmed measurement interval from 12 to 3 hours), and a tendency of GEOS to accept rather than reject marginal WWVB data, conspired at Whittier Narrows to create a large, but largely useless, set of WWVB clock corrections. Second, during most of the experiment we used two master clocks (called TMC and CMC), but, unfortunately, neither clock ran uninterupted from beginning to end. TMC lost power twice during the experiment, and CMC was turned off during a period when field personnel were withdrawn from the field area. In these cases, both clocks had to be reset to other clocks (in one case, TMC was synched to one of the GEOS clocks). Finally, because we could not maintain personnel in the field during the entire experiment, there were several periods during which clock errors could not be measured manually.

These circumstances led to a complicated set of interrelated clock measurements, and we decided to solve for the clock corrections as a least-squares adjustment. Mathematically, GEOS clocks and master clocks were treated equally in the model. Each GEOS clock was assumed to drift at a constant rate between resynchs at a given site, but the rate could be different if the GEOS was moved to a different site. Each master clock was assumed to drift at an unknown constant rate throughout the whole experiment. Thus, a clock correction at time t has the form:

( cc_i ± o_cci ) + ( r ± o_r ) × ( t - t^r_i )   for  t^r_i < t < t^r_i+1

This system of equations was solved by least squares for the unknowns cc_i , o_cci , r, and or for all stations. The results of this computation generally supported the assumption that each GEOS clock had a linear drift that varied when the GEOS was moved. Stations ARC, ECP, and ROS, however, showed obvious nonlinear drift rates. Thus, the clock model for each of these stations was artificially divided into two linear segments, with the clock correction constrained to be continuous at the dividing time. For example, the clock at station ARC was allowed to drift at different linear rates before and after day 285.680, where the time 285.680 was selected by visual inspection of the clock-drift data.

The system was solved again, and the resulting clock corrections are listed in Table 4. The complete system, including measured and predicted clock corrections and residuals, is tabulated in (Appendix B). Clock corrections at most stations are well determined, with discrepancies between measured and predicted clock corrections (residuals) less than 5 ms. The largest residuals are 20 ms or greater; they occur near day 303, between two long periods when field personnel were withdrawn from the field, and may result from possible nonlinear clock drifts or accidental unrecorded clock resets that are not included in the model.

Data Highlights

While it is impossible to give a complete overview in this report, we would like to point out several features of special interest in the Whittier Narrows dataset. The first GEOS station, MOT, was installed in Whittier approximately 15 hours after the mainshock, and recorded eight aftershocks early in the sequence (Table 1). (No clock corrections are available for MOT, however.) The largest aftershock (277 10:59 GMT, M_L = 5.3) was recorded by 10 GEOS stations; acceleration seismograms are plotted in Figure 3. Preliminary analysis of a subset of well recorded aftershocks shows that ground motions recorded around the L. A. basin during Deployment 1 generally appear to mimic the intensity distribution for the mainshock. Ground motions recorded in the local Whittier area (Deployment 3) showed strong site-to-site variations that seemed related to observed intensity patterns (P. Thenhaus, written communication, 1987).

---

Playback and Processing

Data tapes were returned to Menlo Park for final playback after preliminary analysis in the field. Complete and partial (records from events recorded on four-or-more stations) datasets have been archived on nine-track tape in a compact block-binary format (Mueller et al, 1988). These datasets are available from:

ES&G Data Project
US Geological Survey MS 977
345 Middlefield Rd.
Menlo Park, CA 94025.

[ Note from Gary Glassmoyer (23 July 1997): The binary data are now available online at ftp://nsmp.wr.usgs.gov/geos/WNE ]

---

Acknowledgements

We are grateful to Prof. M. Trifunac of the University of Southern California and to Dr. A. Shakal of the California Division of Mines and Geology for permission to co-site GEOS with strong-motion instruments, and to Dr. J. Beck of the California Institute of Technology for assistance in instrumenting Millikan Library. J. Coakley of USGS provided a daily seismicity summary in the form of smoked-drum records in the field. R. Kaderabek of USGS was very helpful during the recording of the calibration explosion. Dr. P. Thenhaus of USGS gave us valuable guidance on selection of GEOS sites for Deployment 3. C. Ramseyer and P. Cuneo of USGS provided valuable logistical support during the experiment, both in the field and in the office. We especially wish to thank the citizens of Los Angeles who opened their garages, storage closets, and backyards to us for several weeks. Without their cooperation, this work would not have been possible. This report was much improved after thoughtful reviews by M. Andrews and J. Fletcher.

---

References

Anderson, J., M. Trifunac, T. Teng, A. Amini, and K. Moslem (1981). Los Angeles vicinity strong motion accelerograph network, Report CE81-04, Dept. of Civil Engineering, Univ. of Southern Calif., Los Angeles.

Borcherdt, R. D., J. P. Fletcher, E. G. Jensen, G. L. Maxwell, J. R. VanSchaak, R. E. Warrick, E. Cranswick, M. J. S. Johnston, and R. McClearn (1985). A General Earthquake Observation System (GEOS), Bull. Seism. Soc. Am. 75, 1783-1825.

Etheredge, E. and R. Porcella (1987). Strong motion data from the October 1, 1987 Whittier Narrows earthquake, U. S. Geol. Surv., Open-File Rept. 87-616.

Etheredge, E. and R. Porcella (1988). Strong motion data from the Whittier Narrows aftershock of October 4, 1987, U. S. Geol. Surv., Open-File Rept. 88-38.

Mueller, C., G. Brady, A. Converse, and J. Watson (1988). Near-source high-frequency seismic waveform data available from the U. S. Geological Survey, U. S. Geol. Surv., Open-File Rept. 88-596, 12 pp.

Perkins, G. (1988). Data report for the 1987 seismic calibration/refraction survey, Whittier, California, U. S. Geol. Surv., Open-File Rept. 88-?, 29 pp.

Shakal, A., M. Huang, C. Ventura, D. Parke, T. Cao, R. Sherburne, and R. Blazquez (1987). CSMIP strong motion records from the Whittier, California, earthquake of October 1, 1987, Report No. OSMS 87-05, California Strong Motion Instrumentation Program, Division of Mines and Geology, Sacramento, California.

---

Figures

Figure 1: Map of the Whittier Narrows meizoseismal area

Figure 2: GEOS deployment summary

Figure 3: Acceleration seismograms from large aftershock at 277 10:59

Figure 3a: UP component

Figure 3b: H000 component

Figure 3c: H090 component

---

Tables

Table 1: Records listed by time and station (all events)

Table 1a: Deployment 1

Table 1b: Deployment 2

Table 1c: Deployment 3

Table 2: Nominal GEOS parameters

Table 3: Station histories

Table 4: GEOS clock corrections

---

Appendix

Appendix A: Maps showing instrument location details at selected stations

Figure A1: Bechtel Building (BEC) - instrument locations

Figure A2: Whittier Lutheran Towers (LTB,LTH) - instrument locations

Figure A3: Millikan Library basement (MIL,BNE,BNW,BSE) - instrument locations

Figure A4: Millikan Library roof (RNE,RSE,RWC) - instrument locations

Figure A5: Millikan Library steam tunnels (STE,STW) - instrument locations

Appendix B: Measured and predicted clock drifts

---

The USGS Home Page is at http://www.usgs.gov
The USGS Geologic Division Home Page is at http://geology.usgs.gov
The USGS National Strong Motion Program Home Page is at http://earthquake.usgs.gov/monitoring/nsmp

The GEOS Home Page is at /GEOS/geos.html
The URL of this page is /GEOS/WNE/OFR_88-688/Whittier_Narrows.html
e-mail: glassmoyer@usgs.gov
or address:
Gary Glassmoyer
345 Middlefield Road M/S 977
Menlo Park, California 94025