Complementary to the regional broadband network, a deployment of borehole-installed, wide-dynamic range seismographic stations is being established along the Hayward Fault and throughout the San Francisco Bay toll bridges network. This network is a cooperative development of the BSL and the USGS, with support from USGS, Caltrans, EPRI, the University of California Campus/Laboratory Collaboration (CLC) program, LLNL, and LBNL (Figure 40.1 and Table 40.1). Efforts at ongoing development of the network have also recently been enhanced through coordinated efforts with the Mini-PBO project which is partially funded by NSF and by the member institutions of that project.
The purpose of the network is threefold: 1) to lower substantially the threshold of microearthquake detection, 2) to increase the recorded bandwidth for events along the Hayward fault, and 3) to obtain bedrock ground motion signals at the bridges from small earthquakes for investigating bridge responses to stronger ground motions. A lower detection threshold increases the resolution of the fault-zone seismic structure; allows seismologists to monitor the spatial and temporal evolution of seismicity at magnitudes down to , where earthquake rates are many times higher than those captured by surface sites; allows researchers to look for pathologies in seismicity patterns that may be indicative of the nucleation of large damaging earthquakes; and allows scientists to investigate fault and earthquake scaling, physics and processes in the San Francisco Bay Area. This new data collection will also contribute to improved working models for the Hayward fault. The bedrock ground motion recordings are also being used to provide input for estimating the likely responses of the bridges to large, potentially damaging earthquakes. Combined with the improved Hayward fault models, source-specific response calculations can be made as well.
The Hayward Fault Network (HFN) consists of two parts. The Northern Hayward Fault Network (NHFN) is operated by the BSL and currently consists of 28 stations with various operational status, including those located on Bay Area bridges and at borehole sites of the Mini-PBO (MPBO) project. This network is considered part of the BDSN and uses the network code BK. The Southern Hayward Fault Network (SHFN) is operated by the USGS and currently consists of 5 stations. This network is considered part of the NCSN and uses the network code NC. This chapter is primarily focused on the NHFN and activities associated with the BSL operations.
The five MPBO sites have 3-component borehole geophone packages. All the remaining HFN sites have six-component borehole sensor packages. The packages were designed and fabricated at LBNL's Geophysical Measurement Facility by Don Lippert and Ray Solbau, with the exception of site SFAB. For the HFN sites, three channels of acceleration are provided by Wilcoxon 731A piezoelectric accelerometers, and three channels of velocity are provided by Oyo HS-1 4.5 Hz geophones. Velocity measurements for the MPBO sites are provided by Mark Products L-22 2 Hz geophones (Table 40.2). The 0.1-400 Hz Wilcoxon accelerometers have lower self-noise than the geophones above about 25-30 Hz, and remain on scale and linear to 0.5 g. In tests performed in the Byerly vault at UC Berkeley, the Wilcoxon is considerably quieter than the FBA-23 at all periods, and is almost as quiet as the STS-2 between 1 and 50 Hz.
Sensors are generally installed at depths of about 100 m, but several sites have sensors emplaced at depths of over 200 m, and the Dumbarton bridge sites have sensors at multiple depths (Table 40.1). During initial stages of the project, the NHFN sensors provided signals to on-site Quanterra Q730 and RefTek 72A-07 dataloggers.
Today, 14 of the NHFN sites have Quanterra dataloggers with continuous telemetry to the BSL. Similar to BDSN sites, these stations are capable of on-site recording and local storage of all data for more than one day and have batteries to provide backup power. Signals from these stations are digitized at a variety of data rates up to 500 Hz at 24-bit resolution (Table 40.3).
The NHFN dataloggers employ casual FIR filters at high data rates and acausal FIR filters at lower data rates. Because of limitations in telemetry bandwidth and disk storage, 9 of these sites transmit one channel of 500 sps continuous data and 90 sec., 500 sps triggered data snippets for the remaining channels. The Murdock, Hutt, and Halbert (MHH) event detection algorithm (Murdock and Hutt, 1983) is operated independently at each station on 500 sps data for trigger determinations. Continuous data for all channels at reduced rates (20 and 1 sps) are also transmitted to and archived at the BSL. The five MPBO sites transmit continuous 100, 20 and 1 sps 3 component data streams that are also archived at BSL.
The remaining 14 sites of the NHFN have in the past recorded data using RefTek dataloggers. These sites do not have continuous telemetry for acquisition and in the past required visits from BSL staff for data recovery. Collection of data from these sites has been discontinued, but efforts are underway to upgrade them with Quanterra Q4120, Q730 or Q330 dataloggers and continuous telemetry.
Signals from the 5 SHFN stations are digitized by Nanometrics data loggers at 100 sps and transmit continuous data to Menlo Park by radio. These digital data streams are processed by the Earthworm system with the NCSN data and waveforms are saved when the Earthworm detects an event.
Data from both the NHFN and SHFN are archived at the NCEDC (Northern California Earthquake Data Center). At this time, the tools are not in place to archive the Hayward fault data together. The NHFN data are archived with the BDSN data, while the SHFN are archived with the NCSN data.
The NHFN station hardware has proven to be relatively reliable. Nonetheless, numerous maintenance and performance enhancement measures are still required. Below is a synopsis of maintenance efforts performed recently for several NHFN stations that serves to illustrate some of the ongoing maintenance and enhancement measures that are typically performed.
BBEB: Ran radio tests on Wilan link to Space Sciences Lab at 18 dBm and at maximum power (23 dBm) to ascertain effect on dropped packets. At 24 dBm power, the throughput was 6 times higher than at 18 dBm power and the number of dropped packets reduced from 4.6 BRIB: Numerous frame relay telemetry problems were encountered during August and September, and the station was visited several times to troubleshoot and correct the problem.
CMSB: Quanterra hung after 8/17 reboot. The power was manually recycled, and the Quanterra came back up.
CRQB: Quanterra hung after 8/17 reboot. The power was manually recycled, and the Quanterra came back up and was functioning normally.
HERB: Velocity channel was found in September to not be responsive to events. The problem was traced to a blown fuse in the power system, although it is unclear as to how that problem effected the responsiveness of the velocity channel.
RFSB: Visited station several times to repair frame relay and power supply problems.
SMCB: Quanterra hung after 8/17 reboot. The power was manually recycled, and the Quanterra came back up.
W02B: Telemetry link went down in October and again in December due to an antenna problem.
A commonly used check on the calibration of the borehole installed network, is to compare the bandpass filtered (0.3-2 Hz) ground velocity data recorded at NHFN and MPBO stations for large teleseismic earthquakes. As an example, a M 7.5 intermediate focus teleseism that occurred in Peru at a depth of 115 km is shown in Figure 40.3.
Another practise for quality control is the assessment of power spectral density (PSD) distributions for the network stations. Shown in Figure 40.2 are power spectral density distributions of background noise for a sample of 13 NHFN land and bridge site stations. In general, background noise levels of the borehole HFN stations is more variable and generally higher than that of the Parkfield HRSN borehole stations (see Parkfield Borehole Network chapter). This is due in large part to the significantly greater level of cultural noise in the Bay Area, and to the fact that noise reduction efforts on the much more recently installed NHFN stations are still underway. For example the two noisiest stations (i.e. BBEB and W02B) are located on the Bay Bridge, which is currently undergoing earthquake retrofit and east span reconstruction. These stations have also only recently come back on-line with upgraded infrastructure and instrumentation, so the full complement of noise reduction modifications have not yet been implemented.
On average the MPBO component of the NHFN sites is more consistent and somewhat quieter. This is due in large part to the greater depth of the MPBO sensors, the locations of MPBO stations in regions of generally less industrial and other cultural noise sources, and possibly to the absence of powered sensors (i.e. accelerometers) in their borehole sensor packages.
One of the most pervasive problems at NHFN stations equipped with the Q4120 dataloggers is power line noise (60 Hz and its harmonics at 120, 180, and 240 Hz). This noise reduces the sensitivity of the MHH detectors. Whenever a NHFN station is visited, the engineer at the site and a seismologist at the BSL work together to expedite the testing process, especially when attempting to identify and correct ground-loop faults which generally induce significant 60, 120, 180, and 240 Hz seismic signal contamination due to stray power line signal pickup, generally inductively coupled and aggravated by the presence of ground loops.
Generally speaking, the accelerometers, being an active device, are more accurate and also more stable than the geophones, so it is reasonable to assume that the most likely reason for the difference is that the assumed generator constants for the geophones are inaccurate. Rodgers et al. (1995) describe a way to absolutely calibrate the geophones in situ and to determine their generator constant, free period and fraction of critical damping. The only external parameter that is required is the value of the geophones inertial mass.
We have built a calibration test box which allows us to routinely perform the testing described by Rodgers et al. whenever site visits are made. The box drives the signal coil with a known current step and rapidly switches the signal coil between the current source and the datalogger input. From this information, expected and actual sensor response characteristics can be compared and corrections applied. Also, changes in the sensor response over time can be evaluated so that adjustments can be made, and pathologies arising in the sensors due to age can be identified. Once a geophone is absolutely calibrated, we can also check the response of the corresponding accelerometer.
We are still performing the initial calibration tests and response adjustments for all NHFN stations as sites are visited for routine maintenance. We also plan to schedule routine re-tests of all sites to monitor for sensor responses changes through time.
As originally conceived, the Hayward Fault Network was to consist of 24 to 30 stations, 12-15 each north and south of San Leandro, managed respectively by UCB and USGS. Due to funding limitations, however, progress has been slow and the original plan has been significantly modified. Fortunately and with additional Caltrans support continued development of the NHFN component of the project has been possible and is ongoing. This important contribution to the Hayward Fault Network has more than doubled the number of sites with instrumentation that would otherwise not have existed. Caltrans continues to provide holes of opportunity (e.g., recently SMCB, PETB, VALB), so we have plans for additional stations that will bring the network geometry to a more effective state for imaging and real-time monitoring of the Hayward fault. Below are short summaries of activities over the past year related to the preparation, installation and activation of new NHFN stations.
Current support is allowing 4 Bay-Bridge stations to be included in the compliment of NHFN stations and two of these stations are already on-line (i.e., BBEB and W02B). Telemetry issues dictate that these two site also serve as rely sites for data coming from the remaining two sites (i.e., W05B and E07B). Because of their critical roles as data relay sites, robust telemetry from these sites is needed. This year various adjustments have been made to optimize telemetry performance of these two stations including their upgrade to Wilan radio telemetry. The infrastructure of the W05B and E07B sites has been upgraded with the installation of weatherproof boxes, power, and telemetry in anticipation of installing Q4120 dataloggers and telemetering their data back to Berkeley.
These and continuing efforts to bring W05B and E07B on-line have been significantly hampered by ongoing Bay-Bridge retrofit work. As a particularly poignant example of this occurred in the Spring of 2006 when retrofit work crews severed the sensor cable to the mid-western span station W05B, losing the cabling into deep bay waters. Fortunately, with Caltrans assistance, a deep diving crew was dispatched and the sensor cable was recovered in good condition.
On the Richmond-San Rafael Bridge, similar problems associated with Bridge retrofit work have been encountered. In addition to the complete loss of mid-span station RSRB a few years ago, our toll plaza site RB2B has had to be relocated. Drilling and installation of the sensors down hole for the site was finished last year and this year installation of the new site infrastructure and electronics has been largely completed. Coordination with Caltrans for power and telephone hook-ups are currently underway.
Caltrans has also provided funding and support for drilling and sensor installation at 2 other land sites (VALB and PETB). The Napa River Bridge site in Vallejo, CA (VALB) is now operating and has been on-line since November of 2005. The PETB (Petaluma River Bridge) site has been drilled and instrumented. Infrastructure construction on the site continues and should be completed within weeks. Routine data flow and archival at the NCEDC is expected after telemetry and power hook-ups are completed.
Currently we are also considering three other sites (in Pt. Pinole regional Park, Mt. Diablo Regional Park and at Wildcat Mtn.) as candidates for two additional holes-of-opportunity in the North//East Bay. We are in the process of obtaining permission from the East Bay Regional Park District (EBRPD) to site the Pt. Pinole station at the Point Isabel Regional Shoreline and have completed field inspection of the two other sites which appear to be suitable.
In order to monitor and capture the source spectrum of moderate down to micro-scale earthquakes, it is essential that the NHFN instruments operate at high precision and in an extremely low noise environment. Therefore, the stations record at high sample rate and their sensors are emplaced in deep boreholes to reduce noise contamination originating in the near surface weathered zone and from cultural noise sources. In addition, the reduction of noise at these stations through vigilant monitoring of actual seismic events plays a central part of our quality control effort.
In Figure 40.4, we show a profile of the NHFN stations for the recent (August 3, 2006) 4.6 earthquake located at Glen Ellen, California, about 60 km NW of Berkeley. Figures such as this are helpful for evaluating network health, and analysis of the waveforms and spectra assist in troubleshooting problems. As Figure 40.4 shows the network is performing very well, but there are some stations exhibiting problems. For example, the Bay Bridge site W02B shows high frequency noise, and OLNH and OXMT show sensitivity and dropout problems.
As mentioned, a key aspect of quality control of the NHFN data is the analysis of actual seismic events. Seismic events of larger magnitude are relatively rare and generally provide more energy at lower frequencies. Hence in order to provide more frequent real events and quality control in the higher frequency band of the NHFN stations, analysis of recordings from the much more frequent microearthquakes are needed. Because real event analyses are relatively labor intensive and because of inadequate insufficient funding, traditional methods of event analysis have proven financially infeasible. To help circumvent these problems, efforts to develop new and improved analysis techniques are ongoing. We have developed and are currently testing some promising techniques that are particularly well suited to the analysis of similar and repeating microearthquakes. The advantages of similar and repeating event analyses for both quality control and scientific purposes are numerous, and the nature of the seismograms from these types of events make automated, rapid and robust analysis possible.
Towards this end, we are currently testing three new
algorithms which we have developed: 1) a phase onset time detector
with sub-sample timing
resolution for improved absolute pick time accuracy, 2) a pattern
scanning recognition scheme to detect, pick, locate and determine
magnitudes for small and very small similar events recorded either
continuously or from among large volumes of noisy triggered data
snippets, and 3) a phase coherency method for identification of
characteristically repeating events sequences from among groups
of similar event multiplets.
Phase Onset Time Detection:
The phase onset time detector makes use of the concept that the complex
spectral phase data over the bandwidth of interest (i.e., where the SNR
is sufficiently high) will sum to a minimum at the onset of an
impulsive P-wave. The algorithm searches for the minimum phase time
via phase shifting in the complex frequency domain over the bandwidth
where the SNR is above 30 dB to identify the onset time of the
seismic phase. The algorithm requires that the recorded waveforms be
deconvolved to absolute ground displacement. This implicitly requires
that any acausality in the anti-aliasing filtration chain, such as the
FIR filters used in the BDSN Quanterra dataloggers, be removed. The
algorithm typically resolves P-wave onset times to one-fiftieth of the
sample interval or better.
Pattern Scanning Recognition: The Murdock-Hutt detection algorithms used by MultiSHEAR, which basically flags an event whenever the short-term average exceeds a longer-term average by some threshold ratio, is neither appropriate for nor capable of detecting the smallest seismic events where signal to noise levels approach those of spurious cultural and earth noise signals. This is because the increased sensitivity parameters needed for small event detection also result in a large fraction of false event triggers. The use of multiple station association filters to reduce the false trigger rates are also of limited value since many of the smaller events are only recorded with enough signal to noise to trigger at a few stations and noise triggers at high sensitivity also often appear to associate temporally at several stations. Added to this is this the exponential increase in the frequency of events with decreasing magnitude, which quickly makes analyst time requirements for comprehensive review and processing of the smallest events financially infeasible.
The approach we have been working on this year to help overcome these problems has been to enhance the effective signal to noise and to focus on identification and processing of some of the more scientifically significant events through the use of a cross-correlation based scanning approach, which scans known waveform patterns through either continuous or collections triggered event snippets (regardless of the triggered event noise levels). With this approach continuous or triggered waveform data that does not match selected patterns are ignored while waveforms that approximately match selected reference event patterns are flagged as newly identified earthquakes.
This approach is less comprehensive in that it only detects events that are somewhat similar in waveform character to the reference patterns. However, it can be generalized significantly by increasing the number of event patterns scanned or by using fairly low maximum cross-correlation thresholds for event flagging. Preliminary tests of our scanning code show that scans of 100 distinct event patterns can be scanned through a days worth of waveform data in 75 minutes on one 900Mhz SPARC cpu when continuous seismic data is used. Scanning through collections of all triggered snippets is substantially faster, in proportion to the inverse fraction of total time spanned by the snippet data.
The approach also provides automated cross-correlation pick alignments that can be used for high precision relative locations and for automated low-frequency spectral ratio determinations for magnitude estimates. Clearly the method has potential for automatically cataloging a large fraction of the more numerous microearthquakes, and in conjunction with the special attributes of similar event groups, updates of the catalogs in an automated monitoring mode can provide near-real-time microearthquake information that can be a powerful tool for monitoring network performance of real event data. Future plans include development and implementation of an automated similar event scanning and cataloging scheme that will provide real-event data from similar small magnitude events for assessment of network health on a much more frequent basis (every few days).
Perhaps more significantly, the approach can also capture and
rapidly catalog some of the most scientifically relevant events
(e.g. repeats of characteristically repeating microearthquakes used
for deep slip rate monitoring and swarms of similar events typically
associated with foreshocks and aftershocks).
The approach is also surprisingly good at detecting events over a wide
magnitude range. Hence there is clear potential for using patterns
from larger aftershocks (e.g. flagged by REDI) to rapidly and
automatically develop a
high-resolution picture of foreshock and aftershock activity associated
with large mainshocks. Tests so far using waveform patterns from an
aftershock from the Parkfield magnitude 6 event (2.2Ml) have been able
to detected and fully process similar events as low as Ml - 1.2 (a
range of 3.4 magnitude units). Testing in this regard is continuing,
but clearly the 3.4 magnitude range is a lower bound on the potential
magnitude range attainable.
Phase Coherency: A spectral phase coherency algorithm was developed to facilitate high resolution quantification of the similarities and differences between highly similar Hayward fault events that occur months to years apart. The resolution of the complex spectral phase coherency methodology is an order of magnitude better than the cross correlation method, which is commonly used to identify highly similar events with resolution of order of a few 10's of meters. This method, originally developed using NHFN borehole data, is now also being tested and refined using data from another borehole network (the HRSN). The goal of the testing and refinement is ultimately to develop a scheme for rapid and objective discrimination and identification of characteristically repeating microearthquakes sequences down to the lowest magnitude possible (where recurrence times are short and hence temporal resolutions are higher) at both Parkfield, and in the Bay Area of California.
Under Bob Nadeau's, Bob Uhrhammer's and Doug Dreger's general supervision, Rich Clymer, Doug Neuhauser, Bill Karavas, John Friday, and Rick Lellinger all contribute to the operation of the NHFN. Bob Nadeau and Bob Uhrhammer contributed to the preparation of this chapter.
Partial support for the NHFN is provided by the USGS through the NEHRP external grant program (grant no. 04HQGR0104). Expansion of the NHFN has been made possible through generous funding from Caltrans (grant no. 59A0245), with the assistance of Pat Hipley. Larry Hutchings and William Foxall of LLNL have also been important collaborators on the project in years past.
Rogers, P.W., A.J. Martin, M.C. Robertson, M.M. Hsu, and D.B. Harris, Signal-Coil Calibration of Electromagnetic Seismometers, Bull. Seism. Soc. Am., 85(3), 845-850, 1995.
Murdock, J., and C. Hutt, A new event detector designed for the Seismic Research Observatories, USGS Open-File-Report 83-0785, 39 pp., 1983.
Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail: firstname.lastname@example.org
© 2006, The Regents of the University of California