Bay Area Regional Deformation Network

The last year has been a transition period for BSL, with the departure of Mark Murray in January 2006 and that of his assistant Cedric dela Beaujardiere in November 2005. We have been lucky to hire Nicolas Houlie on a post-doctoral position to assume the responsibility for the routine operations and development of the BARD network and related data acquisition and processing. Some of the field related tasks that Cedric performed have been transferred to the BSL's field engineering group. This reorganization of the tasks has caused some temporary disruption in the processing of the GPS data, but we are now back on track on all of the processing and archiving tasks. A significant effort in the last three years has been to upgrade thoses stations where it is possible to 1Hz continuous data acquisition, to respond to a growing interest in the community for this type of data, for the estimation of earthquake ground motions and real time earthquake quantification.

BARD overview

Description of the network

Of the 30 sites, ten (BRIB, CMBB, FARB, HOPB, MHCB, ORVB, PKDB, SAOB, SUTB, YBHB) are co-located with broadband seismic stations of the Berkeley Digital Seismic Network (BDSN) with which they share continuous frame-relay telemetry to UC Berkeley. These sites use the Quanterra data loggers to store and retrieve the GPS data converted to MiniSEED format (Perin et al., 1998). The MiniSEED approach provides more robust data recovery from onsite backup on the Quanterra disks following telemetry outages.

Another five stations (SVIN, MHDL, OHLN, OXMT and SBRN) have been installed in the last 3 years in the SFBA and along the Hayward fault as the Berkeley part of a multi-institutional effort funded by the NSF/MRI program to improve strain monitoring in the SFBA using an integrated approach, with significant participation of the USGS/MP (Murray et al., 2002a). These stations include borehole tensor strainmeters, three-component borehole seismic velocity sensors, downhole pore pressure and tilt sensors and GPS receivers. This project served as a prototype for the strainmeter installations planned for PBO, which faces many of the same station installation, configuration, and data retrieval issues we have addressed. Consequently, these 5 stations have received the nickname mini-PBO. From July 2001 to August 2002, five boreholes were drilled to about 200-m depth and equipped with tensor strainmeters recently developed by CIW and 3-component L22 (velocity) seismometers. For this project, we developed a self-centering GPS antenna mount for the top of the borehole casings, which are mechanically isolated from the upper few meters of the ground, to provide a stable, compact monument that allows access to the top of the borehole casing for downhole maintenance. The 5 GPS receivers were progressively installed and connected to Quanterra 4120 data loggers, which provide backup and telemetry capabilities. The completion of the last station (MHDL), located in the Marin Headlands, took longer because it required AC power, which PG $\&$ E installed in December 2005. The site is operational as of Sept 1, 2006. In addition, 10-minute interval data, which are retrieved from all the sites by the USGS via a backup GOES satellite system, show that all the sites are successfully measuring strains due to tidal effects and to local and teleseismic earthquakes (Murray et al., 2002b).

**Figure 42.1:** Operational BARD stations (dark triangles) in northern California (top) and in the San Francisco Bay area (bottom), including an 18-station network near the Long Valley Caldera (LVC) and a 14-station network near Parkfield (PKFD). In the oblique Mercator projection expected Pacific-North America relative plate motion is parallel to the horizontal. Circled stations use continuous real-time telemetry. The small black triangles near BRIB are the experimental L1 stations. Light triangles are PBO and Nucleus (previously existing continuous stations now part of PBO) stations operating in July 2005.
$\begin{figure}\begin{center} \epsfig{file=map_bard.eps, width=15cm}\end{center}\end{figure}$

There is growing interest in collecting higher rates of data for a variety of applications. For example, GPS measurements can accurately track the propagation of earthquake dynamic motions both on the ground (e.g., Larson et al., 2003) and in the atmosphere (e.g., Artru et al., 2001, Ducic et al., 2003), providing complementary information to seismic observations (calibration of integrated acceleration and velocity sensor data) and estimates of earth structure (direct observation of surface wave propagation over the oceans). We started collecting 1 Hz observations at 2 stations (DIAB and MONB) in 2003. In the last year, we have progressively upgraded the telemetry to continuous 1 Hz telemetry at 9 additional stations (CMBB, MHCB, OHLN, OXMT, SBRN, SVIN, TIBB and two new stations , SRB1, see below, and MHDL), where the bandwidth of the existing telemetry system allowed it. At stations collocated with broadband seismic sensors, the seismic data has priority for telemetry, because it is used in the Northern California real-time earthquake notification system (see http://www.cisn.org/ncmc/) making this upgrade more difficult and in general not feasible with the current Z12 receivers because of insufficient data compression. All data collected from BARD/BSL, including the 1Hz data are publicly available at the Northern California Earthquake Data Center (NCEDC; http://www.ncedc.org/bard/).

Table 42.1: List of the BARD maintained by the BSL. Five models of receiver are operating now: Trimble 4000 SSE (T-SSE), Trimble 4000 SSI (T-SSI), Trimble NETRS, (T-NETRS), Ashtech Z12 and Ashtech Micro Z (A-UZ12). The replacement of the Ashtech Z12 by Trimble NETRS will make the receiver park more homogeneous. The telemetry types are listed in column 6. FR = Frame Relay, R = Radio, Mi= Microwave, WEB = DSL line. Some sites are transmiting data over several legs with different telemetry.

	Sites	Lat.	Lon.	Receiver	Telem.	Sampling	Collocated	Location
		(deg.)	(deg)		.	rate	Network
1	BRIB	37.91	237.84	A-Z12	FR	30 s	BDSN	Briones Reservation, Orinda
2	CMBB	38.03	239.61	A-UZ12	FR	1Hz	BDSN	Columbia College, Columbia
3	DIAB	37.87	238.08	A-Z12	FR	1Hz		Mt. Diablo
4	FARB	37.69	236.99	A-Z12	R-FR/R	15 s	BDSN	Farallon Island
5	EBMD	37.81	237.71	T-SSI	WEB	1Hz		East Bay Mud Headquarter
6	HOPB	38.99	236.92	A-Z12	FR	15 s	BDSN	Hopland Field Stat., Hopland
7	LUTZ	37.28	238.13	A-Z12	FR	30 s		SCC Comm., Santa Clara
8	MHCB	37.34	238.35	A-Z12	FR	1Hz	BDSN	Lick Obs., Mt. Hamilton
9	MHDL	37.84	237.50	T-NETRS	FR	1Hz	mini-PBO	Marin Headland
10	MODB	41.90	239.69	A-UZ12	NSN	15 s		Modoc Plateau
11	MONB	37.48	238.13	A-Z12	FR	1Hz		Monument Peak, Milpitas
12	MUSB	37.16	240.69	A-Z12	R-Mi-FR	30 s		Musick Mt.
13	OHLN	38.00	237.72	A-UZ12	FR	1Hz	mini-PBO	Ohlone Park, Hercules
14	ORVB	39.55	238.49	A-Z12	FR	15 s	BDSN	Oroville
15	OXMT	37.49	237.57	A-UZ12	FR	1Hz	mini-PBO	Ox Mountain
16	PKDB	35.94	239.45	A-Z12	FR	30 s	BDSN	Bear Valley Ranch, Parkfield
17	PTRB	37.99	236.98	A-Z12	R-FR	15 s		Point Reyes Lighthouse
18	SAOB	36.76	238.55	A-Z12	FR	30 s	BDSN	San Andreas Obs., Hollister
19	SBRN	37.68	237.58	A-Z12	FR	1Hz	mini-PBO	San Bruno
20	SODB	37.16	238.07	A-Z12	R-FR	30 s		Soda Springs, Los Gatos
21	SRB1	37.87	237.73	T-SSE	FR	1Hz		SRB building, Berkeley
22	SUTB	39.20	238.17	A-Z12	R-FR	30 s	BDSN	Sutter Buttes
23	SVIN	38.03	237.47	A-UZ12	R-FR	1Hz	mini-PBO	St Vincents
24	TIBB	37.89	237.55	A-UZ12	R	1Hz		Tiburon
25	UCD1	38.53	238.248	T-SSE	WEB	1Hz		UC - Davis
26	YBHB	41.73	237.289	A-Z12	FR	15 s	BDSN	Yreka Blue Horn Mine, Yreka

Between 1993 and 2001, the BSL acquired 29 Ashtech Z-12 and Micro-Z receivers from a variety of funding sources, including from federal (NSF and USGS), state (CLC), and private (EPRI) agencies. The network enhances continuous strain measurements in the Bay Area and includes several profiles between the Farallon Islands and the Sierra Nevada in order to better characterize the larger scale deformation field in northern California (Figure 42.1).

The number of continuous GPS stations in northern California is significantly increasing with over 250 new site installations planned by 2008 as part of the Plate Boundary Observatory (PBO) component of the NSF-funded Earthscope project. UNAVCO and researchers from BARD and the other regional networks, such as SCIGN, BARGEN, and PANGA, are funded by NSF to fold operation and maintenance of about 200 existing stations, which constitute the PBO Nucleus network, into the PBO array by 2008. Two BSL-maintained stations (SUTB and MUSB) are included in the PBO Nucleus network. The other BSL stations are either collocated with seismic instrumentation or are located near the San Andreas Fault where real-time processing of the GPS data for earthquake notification is a high priority. Another 23 northern California stations, including most of the Parkfield network, will be included in the PBO Nucleus, and we are working with UNAVCO to facilitate their transition to UNAVCO control.

BARD Stations

The majority of the BSL BARD stations use a low-multipath choke-ring antenna, most of which are mounted to a reinforced concrete pillar approximately 0.5-1.0 meter above local ground level. The reinforcing steel bars of the pillar are drilled and cemented into rock outcrop to improve long-term monument stability. A low-loss antenna cable is used to minimize signal degradation on the longer cable setups that normally would require signal amplification. Low-voltage cutoff devices are installed to improve receiver performance following power outages. Most use Ashtech Z-12 receivers that are programmed to record data once every 30 seconds and observe up to 12 satellites simultaneously at elevations down to the horizon. The antennas are equipped with SCIGN antenna adapters and hemispherical domes, designed to provide security and protection from weather and other natural phenomena, and to minimize differential radio propagation delays.

Data from most BSL-maintained stations are collected at 15 or 30-second intervals and transmitted continuously over serial connections (Table 2.1). Station TIBB uses a direct radio link to Berkeley, and MODB uses VSAT satellite telemetry. Most stations use frame relay technology, either alone or in combination with radio telemetry. Fourteen GPS stations are collocated with broadband seismometers and Quanterra data loggers (Table 38.2). With the support of IRIS we developed software that converts continuous GPS data to MiniSEED opaque blockettes that are stored and retrieved from the Quanterra data loggers (Perin et al., 1998), providing more robust data recovery from onsite disks following telemetry outages.

Data from DIAB, MHCB, MONB, OHLN, OXMT, SBRN, SRB1, SVIN, and TIBB in the Bay Area, and 13 stations in the Parkfield region (all but PKDB), are now being collected at 1-second intervals. All high-rate observations collected by these stations are currently available from the NCEDC. Collecting at such high-frequency (for GPS) allows dynamic displacements due to large earthquakes to be better measured; however, this 30-fold increase in data can pose telemetry bandwidth limitations. We are planning to convert additional stations to 1-second sampling where possible during the next year. The acquisition of the 5 NETRS bundles will help to complete this project (see subsection 3.1). In the Bay Area, we have converted stations that have sufficient bandwidth and are currently assessing bandwidth issues at other stations. Prior to the September 28, 2004 M6 Parkfield earthquake, data from the Parkfield stations were collected on an on-site computer, written to removable disk once per month, and sent to SOPAC for long-term archiving (decimated 30-sec data is acquired daily via the BSL frame relay circuit). In response to the earthquake, we modified the procedures to download 1-second data converted to compact RINEX format at hourly intervals, which does not significantly impact the telemetry bandwidth.

Data archival

As part of the activities funded by the USGS through the BARD network, the NCEDC has established an archive of the 7000+ survey-mode occupations collected by the USGS since 1992. The NCEDC continues to archive non-continuous survey GPS data. The initial dataset archived is the survey GPS data collected by the USGS Menlo Park for northern California and other locations. The NCEDC is the principal archive for this dataset. Significant quality control efforts were implemented by the NCEDC (Romanowicz et al., 1994) to ensure that the raw data, scanned site log sheets, and RINEX data are archived for each survey. All of the USGS MP GPS data has been transferred to the NCEDC and virtually all of the data from 1992 to the present has been archived and is available for distribution. We are also archiving additional high-precision GPS data sets from northern California (mainly Parkfield measurements). Together with students in the department who are now using the GAMIT software to process survey-mode data in the San Francisco Bay area, we are working to combine the survey-mode and C-GPS solutions into a self-consistent velocity field for northern California. The campaign velocity field computed from campaign measurements by university and USGS groups has been published by dÕAlessio et al., (2005).

We also participate in the UNAVCO-sponsored GPS Seamless Archive Center (GSAC) project, which provides access to survey-mode and continuous GPS data distributed over many archives. We helped to define database schema and file formats for the GSAC project, and produce monumentation and data holdings records for the data archived at the NCEDC to provide GSAC with up-to-date information about our holdings. Currently, the NCEDC is the primary provider for over 74,000 data files from over 1400 continuous and survey-mode monuments. The records for these data have been incorporated into the retailer system that became publicly available in early 2003.

Data from five of our sites (HOPB, MHCB, CMBB, OHLN, YBHB) are sent to the National Geodetic Survey (NGS) in the framework of the CORS (Continuous Operating Reference Stations) project (http://www.ngs.noaa.gov/ CORS/). The data from these five sites are also distributed to the public through the CORS ftp site.

2005-2006 Activities

New stations and upgrades

We have also assisted collaborators in installations and upgrades of several continuous stations, including at Thales Navigation (THAL), Hat Creek Radio Observatory (HCRO), and EBMD in downtown Oakland, which is being upgraded to real-time 1 Hz telemetry with a direct radio link to the BSL. This site, which belongs to East Bay Municipal Utilities District (EBMUD) has been renamed EBM1 was operated for part of 2004-2005 at 30s sampling rate. It was upgraded to 1Hz using a new DSL line provided by EBMUD. It is not currently operating because the PC failed and EBMUD is looking for ways to restore the data flow. EBM1 will be the reference station for the East Bay Park surveyor teams for real time kinematic positioning (RTK) applications. A similar collaboration has been engaged with East Bay Parks in collaboration with Mr. Jim Swanson in order to get, archive and process data from the PBO-East Bay Mud GPS sites in real time (Figure 2.1). The site P222 located south of the bay will be equipped to become a RTK reference base during the next year. This site will help define the standard used by East Bay Parks to upgrade (to real time and 1Hz) the PBO sites in the SFBA.

In the last year, we also performed routine maintenance at several sites Ðreplacement of a failing receiver at CMBB (lightning), MODB, TIBB (lightning), YBHB and repair of the failing receivers at mini-PBO stations (SBRN, SVIN). Cylink radios used for FARB were replaced by a Wi-Lan radio as part of a reconfiguration of the telemetry paths through UC San Francisco rather than Mount Tamalpais (2005). As part of this reconfiguration and coincident with the replacement of a badly rusted antenna mast, the Freewave radio path at PTRB was adjusted to be routed through UCSF. In February 2003, the BSL assumed responsibility for data telemetry from a 14-station GPS network in the Parkfield region, in addition to the BSL station PKDB. Most of these stations were constructed using Mini-PBO funding with contributions from the USGS and SCIGN. We replaced a computer at Carr Hill with the appropriate scripts to sequentially download the stations over radio connections, and then retrieve the data over our existing frame relay circuit. All the Parkfield GPS sites are being transferred to the PBO network except the PKDB site, which remains at BSL/BARD site, as it is collocated with a long-term BDSN station.

Five Ashtech receivers (Ashtech $\mu Z$ ) purchased by the mini PBO project have been repaired during the Fall 2005. The whole series was delivered with factory defects by Ashtech corp. However, the replacement of the capacitor and shipment fees have been covered by Ashtech corp. These receivers have been installed mainly at the mini-PBO sites. The first site affected was SVIN in September 2005. The receiver was able to see fewer and fewer satellites every day at the same time. After a few hours the number of visible satellites increased again until the next day. This problem increased dramatically with time and the state of the capacitor.

BSL acquired five NETRS Bundle pack that will be dispatched in the next year over sites where a internet DSL connection access will be available and where telemetry bandwidth does not allow to increase the sampling rate without changing the receiver (BRIB is one the priority site).

Data Analysis and Results

CALREF, a stable reference frame for northern California

The BARD dataset have been processed in the ITRF2000 (Altamimi et al., 2002). The solutions (Houlié and Romanowicz, in prep) are in good agreement with previous campaign solutions (BAVU and USGS) previously released (d'Alessio et al., 2005). The new coordinates release of the BARD network includes the presently operating site and the velocity of the sites already transferred from BSL to PBO during the last two years.

**Figure 42.2:** Comparison of the BARD solutions (white arrows) with the Bay Area Velocity Unification (BAVU) solutions (black arrows). All the data available at the BSL between 1994 and 2006 have been reprocessed (From *Houlié and Romanowicz,* in prep).BAVU website: http://seismo.berkeley.edu/ burgmann/RESEARCH/BAVU/
$\begin{figure}\begin{center} \epsfig{file=bard_velo.epsi, width=13cm}\end{center}\end{figure}$

All the BARD sites have been processed jointly with IGS sites in California. No a priori constraints have been assumed during the processing. A specific troposphere study is under progress to estimate the troposphere gradient over the bay area. The velocities computed for a small selection of sites (Table 42.2) are accurate and well compatible with ITRF2000 solutions (Altamimi et al., 2002). All the velocities included in the first release of California Reference Frame (CALREF) are given in Table 42.3. The CALREF will provide velocities and coordinates of sites located in the bay area at specific epochs. Each solution will be associated to discussed error estimations (formal and real). Every surveyor will be able to control the reference site coordinates for a given survey.

Table 42.3: CALREF 2006 official velocities. All velocities and estimated errors ( $\sigma$ )are indicated in mm per year. For each site, the relevant time-span and the network are specified. The sites with a star are the sites for which the velocities have been used during the combination of the daily solutions.

Site	Lon.	Lat	Ve	Vn	$\sigma_e$	$\sigma_n$	Start
			(mm/y)	(mm/y)	(mm/y)	(mm/y)
BAY1*	197.29	55.19	-6.3	-25.5	0.0000	0.0000	1996.08
BAY2	197.29	55.19	-5.6	-25.2	0.0400	0.0300	1996.08
BRIB	237.85	37.92	-24.8	5.6	0.0100	0.0100	1993.58
CMBB	239.61	38.03	-22.9	-2.8	0.0100	0.0100	1993.92
CNDR	238.72	37.90	-24.4	-5.5	0.0200	0.0200	1999.27
DIAB	238.08	37.88	-23.7	-2.2	0.0100	0.0100	1998.33
FARB	237.00	37.70	-39.8	23.3	0.0100	0.0100	1994.00
GOLD*	243.11	35.43	-18.2	-5.4	0.0000	0.0000	1989.95
HCRO	238.53	40.82	-18.0	-8.7	0.1400	0.1500	2003.50
HOPB	236.93	39.00	-31.1	6.8	0.0100	0.0100	1995.58
JPLM*	241.83	34.21	-36.6	11.8	0.0000	0.0000	1989.44
LUTZ	238.14	37.29	-31.7	9.5	0.0100	0.0100	1996.33
MHCB	238.36	37.34	-24.2	-2.4	0.0100	0.0100	1996.33
MODB	239.70	41.90	-16.9	-9.1	0.0200	0.0200	1999.83
MOLA	237.58	37.95	-30.5	9.7	0.0100	0.0100	1993.75-2002.22
MONB	238.13	37.49	-27.5	2.7	0.0100	0.0100	1998.50
MUSB	240.69	37.17	-22.3	-4.0	0.0100	0.0100	1997.83
OHLN	237.73	38.01	-26.4	4.4	0.0200	0.0200	2001.83
ORVB	238.50	39.56	-22.7	-6.6	0.0100	0.0100	1996.83
OXMT	237.58	37.50	-36.9	18.0	0.0600	0.0600	2004.12
P181(PBO)	237.62	37.92	-29.0	9.6	0.3800	0.4000	2005.09
P198 (PBO)	237.39	38.26	-29.2	7.9	0.0900	0.1000	2004.77
P200 (PBO)	237.55	38.24	-24.3	4.7	0.2000	0.2200	2005.73
P222 (PBO)	237.92	37.54	-31.5	10.0	0.1100	0.1200	2005.26
P224 (PBO)	237.78	37.86	-26.9	5.5	0.1000	0.1100	2005.25
P225 (PBO)	237.94	37.71	-25.2	2.7	0.0900	0.1000	2005.14
P227 (PBO)	238.21	37.53	-28.6	-0.4	0.5800	0.6300	2006.20
P228 (PBO)	238.31	37.60	-23.5	1.0	0.4300	0.4700	2005.93
P229 (PBO)	238.02	37.75	-26.8	1.6	0.1100	0.1200	2005.29
P230 (PBO)	238.21	37.82	-22.5	-3.1	0.1100	0.1200	2005.15
P261 (PBO)	237.78	38.15	-21.0	-0.5	0.0900	0.1000	2004.50
P262 (PBO)	237.90	38.03	-24.2	1.2	0.1100	0.1200	2005.32
PKDB	239.46	35.95	-43.0	18.7	0.0100	0.0100	1996.67
PPT1*	237.61	37.19	-40.7	22.1	0.0000	0.0000	1996.14
PTRB	236.98	38.00	-37.7	22.2	0.0100	0.0100	1998.58
S300	238.44	37.67	-22.9	-4.4	0.0200	0.0200	1998.48
SAOB	238.55	36.77	-41.4	22.0	0.0100	0.0100	1997.58
SBRN	237.59	37.69	-32.0	14.2	0.0300	0.0300	2003.18
SODA	26.39	67.42	18.7	34.1	0.1400	0.1600	1994.70
SODB	238.07	37.17	-33.1	11.7	0.0100	0.0100	1996.33
SUAA	237.83	37.43	-33.7	12.4	0.0100	0.0100	1994.30
SUTB	238.18	39.21	-23.1	-6.7	0.0100	0.0100	1997.33
SVIN	237.47	38.03	-30.5	10.3	0.0400	0.0400	2003.89
THAL	238.07	37.35	-32.0	9.5	0.2000	0.2200	2003.00
TIBB	237.55	37.89	-30.8	11.2	0.0100	0.0100	1994.42
UCD1	238.25	38.54	-23.1	-6.0	0.0100	0.0100	1996.38
VNDP*	239.38	34.56	-42.2	20.9	0.0000	0.0000	1992.48
YBHB	237.29	41.73	-15.8	-6.7	0.0100	0.0100	1996.75

Acknowledgements

Since the departure of Mark Murray at the end of 2005, Barbara Romanowicz oversees the BARD program. Rich Clymer, Cedric de La Beaujardiere, Bill Karavas, Rick Lellinger, John Friday, Nicolas Houlié and Doug Neuhauser contributed to the operation of the BARD network. The operation of the BARD network is partially supported by funding from the USGS/NEHRP program and funding from the NSF/UNAVCO PBO nucleus grant

References

Altamimi, Z., P. Sillard, and C. Boucher , ITRF2000: A new release of the International Terrestrial Reference Frame for earth science applications, J. Geophys. Res., 107(B10), 2214, doi:10.1029/2001JB000561, 2002

Bürgmann, R., D. Schmidt, R.M. Nadeau, M. d'Alessio, E. Fielding, D. Manaker, T. V. McEvilly, and M. H. Murray, Earthquake potential along the northern Hayward fault, California, Science, 289, 1178-1182, 2000.

d'Alessio, M. A., I. A. Johanson, R. Bürgmann, D. A. Schmidt, and M. H. Murray, Slicing up the San Francisco Bay Area: Block kinematics from GPS-derived surface velocities, J. Geophys. Res., 110, B06403, doi:10.1029/2004JB003496, 2005.

Dreger, D. S., L. Gee, P. Lombard, M. H. Murray, and B. Romanowicz, Rapid finite-souce analysis and near-fault strong ground motions: Application to the 2003 Mw 6.5 San Simeon and 2004 Mw 6.0 Parkfield earthquakes, Seismol. Res. Lett., 76(1), 40-48, 2005.

Houlié, N. and P. Briole, Laterally heterogeneous troposphere: a potential explanation for seasonality in GPS coordinates time series, J. Geophys. Res., submitted.

Houlié, N., G. Occhipinti, P. Lognonné, and M. Murakami (2006), Rayleigh seismic waves detected by GPS. Application to the 2003 September $25^{th}$ , Hokkaido earthquake, Geophys. Res. Lett., submitted.

Houlié, N. and Romanowicz, B., CALREF, a stable reference frame for the Northern California, in prep.

King, N. E., J. L. Svarc, E. B. Fogleman, W. K. Gross, K. W. Clark, G. D. Hamilton, C. H. Stiffler, and J. M. Sutton, Continuous GPS observations across the Hayward fault, California, 1991-1994, J. Geophys. Res., 100, 20,271-20,283, 1995.

Langbein et al., Preliminary report on the 28 September 2004, M 6.0 Parkfield, California earthquake, Seismol. Res. Lett., 76(1), 10-26, 2005.

Murray, M. H., and P. Segall, Continuous GPS measurement of Pacific-North America plate boundary deformation in northern California and Nevada, Geophys. Res. Lett., 28, 4315-4318, 2001.

Murray, M. H., R. Bürgmann, W. H. Prescott, B. Romanowicz, S. Schwartz, P. Segall, and E. Silver, The Bay Area Regional Deformation (BARD) permanent GPS network in northern California, EOS Trans. AGU, 79(45), Fall Meeting Suppl., F206, 1998.

Murray, M., Neuhauser D., Gee, L., Dreger, D., Basset, A., and Romanowicz, B., Combining real-time seismic and geodetic data to improve rapid earthquake information , EOS. Trans. AGU, 83(47), G52A-0957, 2002.

Murray, M.H., D.C. Agnew, R. Bürgmann, K. Hurst, R.W. King, F. Rolandone, J. Svarc, GPS Deformation measurements of the 2003 San Simeon earthquake, Seism. Res. Lett., 75, 295, 2004.

Perin, B. J., C. M. Meertens, D. S. Neuhauser, D. R. Baxter, M. H. Murray, and R. Butler, Institutional collaborations for joint seismic and GPS measurements, Seismol. Res. Lett., 69, 159, 1998.

Rhie, J., D. Dreger, and M. H. Murray, A prediction of strong ground motions from geodetic data for PGV ShakeMaps, Geophys. Res. Lett., in prep., 2005.

Rolandone, F., I. Johanson, and R. Bürgmann, Geodetic observations of the M6.5 San Simeon earthquake with focus on the response of the creeping segment of the San Andreas fault, Seism. Res. Lett., 75, 293, 2004.

Romanowicz, B., B. Bogaert, D. Neuhauser, and D. Oppenheimer, Accessing northern California earthquake data via Internet, EOS Trans. AGU, 75, 257-260, 1994.

Uhrhammer, R., L. S. Gee, M. Murray, D. Dreger, and B. Romanowicz, The $M_{w}$ 5.1 San Juan Bautista, California earthquake of 12 August 1998, Seismol. Res. Lett., 70, 10-18, 1999.