California Integrated Seismic Network

Introduction

Advances in technology have made it possible to integrate separate earthquake monitoring networks into a single seismic system as well as to unify earthquake monitoring instrumentation. In California, this effort was initiated under the TriNet Project in southern California, where Caltech, the then California Division of Mines and Geology, and the USGS combined their efforts to create a unified seismic system for southern California. With major funding provided by FEMA, OES, and the USGS, the TriNet project provided the opportunity to upgrade and expand the monitoring infrastructure, combining resources in federal, state, university partnership. More recently, the California Geological Survey, Caltech Seismological Laboratory, Berkeley Seismological Laboratory, USGS Menlo Park, and the USGS Pasadena have agreed to cooperate on a statewide basis, because of the obvious benefit to the state.

In the 2000-2001 Annual Report, we described the efforts to create this collaboration through the establishment of a memorandum of agreement and the development of the CISN strategic and implementation plans. Last year, we reported on the first steps toward establishing a statewide system with funding provided by the OES. This year we continued our efforts to move forward with the CISN.

CISN Background

Organization

The core CISN institutions (California Geological Survey, Berkeley Seismological Laboratory, Caltech, USGS Menlo Park, USGS Pasadena) and OES have signed a MOA (included in the 2000-2001 Annual Report) that describes the CISN organizational goals, products, management, and responsibilities of member organizations. To facilitate coordination of activities among institutions, the CISN has formed three management centers:

A goal of the CISN is for the Northern and Southern California Management Centers to operate as twin earthquake processing centers. The Engineering Management Center has the lead responsibility for producing engineering data products.

The Steering Committee oversees CISN projects and is comprised of two representatives from each core institution and a representative from OES. The position of chair rotates among the institutions; Barbara Romanowicz served as the first chair of the Steering Committee; the position has rotated to the California Geological Survey and Jim Davis served as chair from January 2003 until his retirement in the end of June. Mike Riechle is the new chair of the Steering Committee.

An external Advisory Committee, representing the interests of structural engineers, seismologists, emergency managers, industry, government, and utilities, has been formed for review and oversight. The Advisory Committee is chaired by Bruce Clark of the California Seismic Safety Commission. The Advisory Committee held its first meeting in July 2001 and met most recently in January of 2003.

The Steering Committee has formed other committees, including a Program Management Group to address planning and coordination, a Strong Motion Working Group to focus on issues related to strong-motion data, and a Standards Committee to resolve technical design and implementation issues.

In addition to the core members, several organizations contribute data that enhances the capabilities of the CISN. Contributing members of the CISN include: University of California, Santa Barbara; University of California, San Diego; University of Nevada, Reno; University of Washington; California Department of Water Resources; Lawrence Livermore National Lab; and Pacific Gas and Electric.

CISN and ANSS

The USGS Advanced National Seismic System (ANSS) is being developed along a regionalized model. 8 regions have been organized and the CISN represents the "California region". Over the last 4 years, ANSS funding in California has primarily been directed to the USGS Menlo Park to expand the strong-motion instrumentation in the San Francisco Bay Area. As a result, instruments at over 100 sites have been installed or upgraded, significantly improving the data available for ShakeMaps.

The CISN is currently developing plans for the FY03/04 ANSS program. As the ANSS moves forward, committees and working groups are being established to address issues of interest. Currently, Lind Gee and David Oppenheimer represent the CISN on an ANSS working group for business rules for earthquake reporting (developing figures of merit for selecting the "best" location, magnitude, and other earthquake products when multiple solutions are available).

CISN and OES

The California Governor's Office of Emergency Services has had a long-term interest in coordinated earthquake monitoring. The historical separation between northern and southern California and between strong-motion and weak-motion networks resulted in a complicated situation for earthquake response.

OES has been an advocate of increased coordination and collaboration in California earthquake monitoring and encouraged the development of the CISN Strategic and Implementation Plans. In FY01/02, Governor Gray Davis requested support for the CISN, to be administered through OES. Funding for the California Geological Survey, Caltech and UC Berkeley was made available in spring 2002, officially launching the statewide coordination efforts.

Despite the dire budget situation in the state of California in FY02/03, OES support led to the establishment of 3-year contracts to the BSL, Caltech, and the California Geological Survey for CISN activities. Although at a reduced level of support from the previous year, these funds are critical to continued efforts in statewide integration.

2002-2003 Activities

The CISN funding from OES facilitated a number of activities at the BSL during the past year.

Expanded Instrumentation

In 2001-2002, the BSL purchased equipment for 5 BDSN stations, including STS-2 seismometers, Episensors, and Q4120 data loggers and initiated efforts to identify potential sites, considering such factors as the current distribution of stations, private versus public property, location of power and telecommunications, and geologic materials. Two sites were permitted - McLaughlin Mine Natural Reserve and Alder Springs Conservation Camp.

In 2002-2003, the BSL permitted a 3rd site at Pacheco Peak with the California Department of Forestry and Fire. This site is located in south Santa Clara County. The BSL installed two sites during this year, at McLaughlin Reserve and Pacheco Peak, complementing the BDSN installations in the San Francisco Bay Area. The efforts for site preparation and installation are more fully described in Chapter 3.

Other areas under consideration for future installations include the Pt. Reyes area, the Santa Cruz Mountains (in collaboration with UC Santa Cruz), Placerville (in collaboration with Davey Jones and Lava Cap Winery), near Pinehurst (in collaboration with Peggy Hellweg), Hat Creek (in collaboration with UC Berkeley Department of Astronomy), and Carmel Valley (UC Berkeley Hastings Preserve). We hope to install two more sites in FY03/04.

Network Operations

As part of the CISN project, the BSL purchased a number of upgrade kits for their Q4120 data loggers with the goal of improving remote diagnostic capabilities last year. Three different kits were purchased - power board only, calibration board only, and combined power and calibration boards - in order to ensure that every Q4120 has a power board and that every 8-channel Q4120 also has a calibration board. The power boards provide the capability to monitor battery voltage, allowing staff to discriminate between power and telemetry problems remotely. The calibration boards provide the capability to monitor mass position as well as allow remote calibration of the seismic sensors. Both boards also record data logger temperature.

The boards were received in the winter of 2002. BSL staff, particularly Dave Rapkin, began to work on installing the upgrade kits. Of the 23 kits purchased, 11 have been installed as of June 30th and 9 of these dataloggers have been reinstalled in the field. In addition to the installation of these boards, the BSL staff must also prepare new cables in order to record these new channels. That effort is also underway.

Statewide Communications

One of the major accomplishments in FY01/02 was the design and initial implementation of a CISN communications infrastructure. Doug Neuhauser of the BSL took the lead in investigating options and the CISN partners decided to establish a "ring" of T1 communication links (Figure 2.1) with dual routers at each node.

The implementation of the CISN ring was completed in early 2003, when the last problem (a bad wire in building 11 at the USGS Menlo Park) was resolved. All links are now fully operational and the ring is being used to push data among the CISN partners such as waveforms, picks, and ground motions. In addition, the CISN partners have migrated the transmission of ShakeMaps to OES to the ring. Use of the ring for sending ShakeMaps to OES has been a long-standing goal for the CISN but required coordination with OES personnel to configure machines. This effort was accelerated by the failure of the computer at OES that had been the recipient of ShakeMaps via the Internet. When the computer failed, BSL staff worked with OES and CISN partners to set up an interim system where ShakeMaps were transmitted to a UC Berkeley-maintained computer at OES. This stop-gap provided a local site within OES where the ShakeMaps could be retrieved while more permanent solutions were developed. OES replaced the failed computer and set up a new recipient system on the CISN ring. As of the end of May, ShakeMaps are being pushed to two separate OES machines, one on the CISN ring and one on the public Internet.

Early in 2003, Doug Neuhauser of the BSL installed the Multi Router Traffic Grapher (MRTG) software package to monitor the CISN ring. This package collects data for graphical display of traffic on the CISN router interface. More recently, Stan Schwartz of the USGS began using the Big Brother software to establish alarming capabilities - that is, to notify personnel when a problem with the ring is detected. The alarming system is running in Pasadena and will be installed at Berkeley soon. Two separate instances of the alarming software should provide reliable notification of problems with the ring.

The remaining outstanding issues with the ring include the connection of the OES routers to the Internet and the development of a security document.

Figure 2.1: Map showing the geographical distribution of the CISN partners and centers. The communications "ring" is shown schematically with installed links (solid lines).
\begin{figure}\begin{center}
\epsfig{file=cisn_centers.ps, width=7cm, bbllx=50,bblly=125,bburx=571,bbury=711}\end{center}\end{figure}

Northern California Management Center

As part of this effort within the CISN, the BSL and the USGS Menlo Park have begun to plan for the next generation of the northern California joint notification system. Chapter 10 describes the operations of the existing Management Center and reports on design discussions.

Communications Infrastructure

In order to migrate to a design such as Figure 10.7, the BSL and the USGS Menlo Park need to enhance the communications infrastructure between their sites. Presently, data and information are shared on a dedicated connection, with fallback to the Internet.

Last year, the BSL commissioned Telecommunications Design Services, Inc. to perform a feasibility study for a microwave communication link between Berkeley and Menlo Park. The goal of the study was to evaluate options for a microwave communication link between the BSL and USGS elements of the Northern California Management Center. The report concludes that a repeater site will be required, given the length of the path and the obstructions (buildings, bridges, etc.). According to the report, the Space Sciences Laboratory at UC Berkeley will be a good site for the repeater.

Unfortunately, funding to establish this microwave link has not been identified. As a modest step to improve the communications links between the BSL and Menlo Park, the BSL installed a dedicated T1 connection this year, similar to the one established last year for the CISN ring. This dedicated T1 replaces the previous frame-relay connection as a more cost effective alternative. The second T1 provides the necessary bandwidth between the Berkeley and Menlo Park elements of the Northern California Management Center. The original frame-relay T1 will be reused by the BSL as a second link for connections to its seismic networks.

Computing Upgrade

The current data acquisition and processing computers used as part of the Northern California Management Center at the BSL are nearly 4 years old. As part of the OES project, the BSL purchased computers to replace these aging systems.

In the past year, the five Sun 280R computers have been brought online. Two of them were use to replace the data acquisition and processing computers. One is being used to generate ShakeMaps for northern California. And two are being setup as as pilot system for statewide earthquake processing. In addition to the 280R computers, the BSL purchased two Sun StorEdge RAID disk systems. The additional disk systems are required by the expanded waveform exchange among the centers.

Figure 2.2: Comparison of $M_{L}$ station adjustments estimated for BDSN stations from independent inversions of Wood Anderson amplitudes, using methodologies developed at UCB and Caltech (CIT).
\begin{figure}\begin{center}
\epsfig{file=dml.ps, width=7cm, bbllx=22,bblly=34,bburx=435,bbury=720}\end{center}\end{figure}

Statewide Integration

BSL staff are involved in many elements of the statewide integration effort. In FY02/03, the Standards Committee of the CISN addressed a number of topics critical to this effort such as the software calibration issues discussed below. The Standards Committee continues to define and prioritize projects necessary to develop a prototype system and established working groups to address them (see the minutes from meetings and conference calls at http://www.cisn.org/standards/meetings.html).

Software Calibration

The CISN partners are working together on the problem of software calibration, particularly as it pertains to automated earthquake processing. Currently, the software implemented in the Northern and Southern California Management Centers is very different. Eventually, there may be standardization of software across the management centers, but in the short term, the focus is on calibrating the software to produce the same answers, given the same input data.

In the last year, effort was continued to focus on phase pickers, the association algorithm (binder), the location algorithm (hypoinverse), and magnitude estimation (various).

The CISN continued evaluation of data from a test system that is performing statewide earthquake notification using seismic data from all CISN partners. These tests are structured so that both northern and southern California are operating a statewide system in parallel with their current "regional" system. A recent 2-week evaluation of results from an exchange of data between northern and southern California networks showed that approximately 95% of the earthquakes were identical. Of the remaining 5%, most were "noise" associations on one system that were rejected on the other. The discrepant behavior can be attributed to differences in the order in which travel-time information is received by the two systems. This type of algorithm behavior may be difficult to eliminate without significant revision to the software design. We have identified several other issues and continue to improve the behavior of the statewide associator.

In parallel, the CISN has been working on issues related to magnitude. Here, the CISN is working in several areas. Pete Lombard of the BSL and Caltech staff have been working together to resolve issues in the original TriNet software that computes magnitude related to the selection of time windows and stations used in the estimation. Bugs in the codes that compute travel time have been identified and are being corrected. Once these issues have been resolved, then the Northern California Management Center will implement the magnitude codes for computing $M_{L}$ and $M_{e}$. Similarly, the Southern California Management Center is working on the implementation of codes to estimate $M_{w}$ and the seismic moment tensor that were developed in Northern California.

Also part of the magnitude calibration effort is the computation of station adjustments for $M_{L}$ and $M_{e}$ on a statewide basis. This effort is underway by BSL and Caltech staff. Figure 2.2 shows the results from two separate inversions of Wood Anderson amplitudes from the BDSN to estimate local magnitude adjustments. Bob Uhrhammer of the BSL and Jascha Polet of Caltech have been comparing their independent results as a first step to determine a common set of adjustments. The method Jascha employs estimates the adjustments and attenuation relationship simultaneously, while Bob's approach a differential approach while fixing the attenuation relationship. The good agreement between the estimates of the adjustments provides confidence in the first step of the process. The next step is a joint inversion of BDSN and TriNet data for magnitude adjustments and a unified statewide attenuation relationship. In parallel, the BSL has worked to develop a collection of energy magnitude estimates in order to determine $M_{e}$ station adjustments.

A final component of the magnitude efforts is the designation of a magnitude reporting hierarchy. There is general agreement at the low end and at the high end, but the working group is still reviewing issues relation to transition points from one magnitude type to another. More details about the magnitude calibration effort are documented in Chapter 10.

Metadata Exchange

The CISN is also working on issues related to metadata exchange. This is an important component of CISN activities, as correct metadata are required to insure valid interpretation of data. A Standards Working Group has developed and initiated testing of a model for database replication of metadata, and is currently reviewing how much of the schema to exchange and how to address metadata from partners such as CGS, who do not currently maintain their metadata in a database.

The Metadata Working Group has compiled a list of metadata necessary for data processing and developed a model for exchanging metadata. In this model, each CISN member is responsible for the metadata of their stations and other stations that enter into CISN processing through them. For example, Menlo Park is responsible for the NSMP, Tremor, and PG&E stations and Caltech is responsible for the Anza data. The Working Group believes that metadata exchange should proceed on a timely basis, not just when data are generated, and is testing an approach using database replication.

The core of the model is to have a master database at each organization and use multi-master replication to propagate the changes between the different data centers. Within a data center, it is proposed to use an interim database as the staging area. This database would contain snapshots of the master tables and the changes would be pushed manually to the master database by using snapshot replication. The use of such a staging area is particularly important because of the way Northern and Southern California currently load their databases. The current programs usually delete some or all the metadata before repopulating the updated data. This will introduce a latency period where the users will see inconsistent information in the database. The working group believes that this period should be relatively small and acceptable in our model. However, a longer term solution will certainly include the coding of new population programs. This model is currently being tested.

Dual Station Feeds

One of the major accomplishments last year was the establishment of "dual station feeds" from 20 stations (10 in northern California and 10 in southern California). To achieve this, the BSL and Caltech both ordered the DLCIs (data link connection identifier) that allow the 2nd center to establish a PVC (permanent virtual circuit) to each station using the frame-relay network.

The Northern California Management Center is using data from the 10 Southern California stations to estimate magnitudes on a routine basis. A subset of these stations are being used for the moment tensor inversions, a computation that is more sensitive to the background noise level.

This initial set of 20 stations was expanded to 30 this year (Figure 2.3), providing a broad sampling of the state. The next step for the BSL is to test these stations in their automated processing system. This direct feed of data from the station to two processing centers is an important step to improving the robustness of earthquake monitoring statewide.

Figure 2.3: Map showing the 30 stations selected to send data directly to the Northern and Southern California processing centers, and the 5 stations that send data directly to the Engineering Data Center and the Southern California processing center.
\begin{figure}\begin{center}
\epsfig{file=cisn_dual.ps, width=7cm, bbllx=42,bblly=94,bburx=565,bbury=690}\end{center}\end{figure}

Data Exchange

Pick exchange was initiated between the Northern and Southern California Management Centers last year. Although the CISN has developed software to exchange the reduced amplitude timeseries, this aspect of data exchange has been delayed while certain problems in the codes that generate the time series are addressed. We hope to begin exchanging these timeseries in the fall of 2003.

The CISN partners completed the first stage of a system to exchange peak ground motion data this year. Using a common format, the CISN partners are exchanging observations with one another following an event or a trigger. This step increases the robustness of generating products such as ShakeMap, since all CISN partners are now exchanging data directly with one another. It also improves the quality of ShakeMaps on the boundary between northern and southern California, such as the recent events in Lompoc, by allowing all data to be combined in a single map. Finally, it is a necessary step toward the goal of generating statewide ShakeMaps.

Earthquake Information
Distribution

In response to a request from the PMG, USGS and OES management established an Ad Hoc panel to develop specifications for an earthquake information system and to review existing systems as well as systems under development. Lind Gee of the BSL and David Oppenheimer of the USGS were asked to co-chair this panel and to provide a written report within 90 days. The panel was put together in May and a meeting took place on July 9-10th at the BSL. The report is due in early September.

Outreach

The CISN hosted an outreach workshop on ShakeMaps in January 2003, targeting the media. "ShakeMap for the News Media" was held at the USGS Menlo Park and attended by a number of print, radio, and television media in northern California. The purpose of the workshop was to raise awareness of ShakeMaps as an important tool in earthquake reporting. As part of the workshop, Lind Gee gave a talk on the CISN and its activities.

The CISN Web site http://www.cisn.org is continuing to develop. As part of the Web updating of the NCEDC and BSL this year, the CISN Web site was revamped and reorganized. In March, the Web site was updated to conform with USGS usage of their logo.

One of the major changes this year was the addition of "Seismology Data Reports". Both the Northern and Southern California Management Centers have generally produced special reports following earthquakes of note. These reports have generally provided tectonic and seismological context for the event and have included detailed maps showing background seismicity and figures showing waveforms of interest. We developed a prototype for a standard CISN report, and had the opportunity to try it out during the November 24, 2002 San Ramon swarm as well as the February 2, 2003 Dublin swarm and the February 22, 2003 Big Bear earthquake.

Plans are underway to expand the capabilities and services of CISN Web site. This year, the BSL purchased two computers for Web servers, in order to improve response. This upgrade is critical to insure that the CISN Web site can respond to post-earthquake traffic. Once the upgrade is complete, the BSL will migrate the recenteqs and ShakeMap pages from the NCEDC server to the CISN server. That will make these products available directly from cisn.org. In addition, we are working with Caltech to setup a system to mirror the CISN Web site in southern California in order to distribute the load.

We have also moved forward with the development of a "myCISN". The ability to personalize earthquake content on a Web site was suggested at an early CISN Advisory Committee meeting. This year, Ionut Iordache developed a prototype for myCISN and we have just started testing this feature.

Acknowledgements

CISN activities at the BSL are supported by funding from the Governor's Office of Emergency Services.

Barbara Romanowicz and Lind Gee are members of the CISN Steering Committee. Lind Gee is a member of the CISN Program Management Committee and she leads the CISN project at the BSL. Doug Neuhauser is chair of the CISN Standards Committee, which includes Lind Gee and Pete Lombard as members.

Because of the breadth of the CISN project, many BSL staff have been involved including: John Friday, Lind Gee, Ionut Iordache, Wade Johnson, Bill Karavas, Pete Lombard, Doug Neuhauser, Charley Paffenbarger, Dave Rapkin, Cathy Thomas, and Stephane Zuzlewski. Lind Gee contributed to this chapter. Additional information about the CISN is available through reports from the Program Management Committee (Hauksson et al., 2002; 2003a; 2003b).

References

Hauksson, E., L. Gee, D. Given, D. Oppenheimer, and T. Shakal, Report to the CISN Advisory and Steering Committees, #4, http://www.cisn.org/oes/2003.05.09.pdf, 2003b.

Hauksson, E., L. Gee, D. Given, D. Oppenheimer, and T. Shakal, Report to the CISN Advisory and Steering Committees, #3, http://www.cisn.org/oes/2003.01.17.pdf, 2003a.

Hauksson, E., L. Gee, D. Given, D. Oppenheimer, and T. Shakal, Report to the CISN Advisory and Steering Committees, #2, http://www.cisn.org/oes/2002.10.25.pdf, 2002.

Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail: www@seismo.berkeley.edu
© 2004, The Regents of the University of California