Parkfield Monitoring: Discontinued Monitoring
For a variety of reasons, several monitoring networks have been discontinued. Reasons include changes in personnel, poorly designed networks or instruments, or natural hazards (floods) that caused the instrument to fail.
A network of four, closely spaced shallow borehole tiltmeters were located near Parkfield at Gold Hill. The instruments were located down-hole below the surface. Data were recorded on-site and transmitted every 10 minutes with digital telemetry via satellite to Menlo Park. Although the observed tilts due to solid-earth tides are consistent with other measurements, the longer term tilts recorded by these instruments are believed to be contaminated by instabilities in the near surface materials. The tilt resolution is approximately 0.1 to 1 microradian at periods of days, and 0.01 to 0.1 microradian at periods of hours.
Ultra-low Frequency Magnetic Fields (ULF)
An attempt to measure ultra low frequency (ULF) magnetic signals prior to further earthquakes was made in Parkfield, California. Independent ULF measurement systems have been installed at the Varian and Haliburton ranches near Parkfield and at Piñon Flat and Table Mountain in Southern California.
A network of leveling lines in the Parkfield region was repeatedly surveyed beginning in 1977, in a cooperative effort with the University of California, Santa Barbara. The networks consisted of 7 lines, and included: a 10-km-long line perpendicular to the San Andreas fault at Parkfield; a 32-km-long line near Middle Mountain; a 17-km-long line perpendicular to the fault at the southern end of the 1966 rupture zone; and a 24-km-long line parallel to the fault line. Short (less than 1 km) sections of these long lines were surveyed 3 to 4 times per year.
Small Aperture Networks
Three small aperture trilateration networks spanned the Parkfield section of the San Andrea fault. Standard errors for individual measurement are 4 mm. Thirty-one near fault lines were surveyed quarterly.
Liquefaction is said to occur when, as consequence of high pore pressure within a water saturated sand, the sand mass undergoes a significant decrease in shear strength and large deformation may occur under the existing level of vibrational stress. To understand the process of liquefaction the USGS and EPRI selected a site, about 15 km southeast of Parkfield and 0.7 km from the San Andreas fault, to install liquefaction sensing instrumentation, that included accelerometers and piezometers. A piezometer instrument is used to measure hydrostatic pressure, such as within saturated sand. There were also 8 bench marks installed with calibrated elevations to measure any surface subsidence which may occur during liquefaction The goal of the experiment was to monitor pore pressure in the sand deposit as it undergoes liquefaction and simultaneously record the acceleration levels within the sand layer. This data lead to a better understanding of ground failure due to liquefaction and enabled engineers and scientists to improve their capability to predict the magnitude and extent of failure due to liquefaction. This study focused on measuring ground vibrations and are designed to improve the capability of predicting the response of engineered structures to ground shaking. For this purpose, instrumented experiments were being conducted on lifeline facilities, unreinforced masonry structures, and school buildings. The data collected is being used to design structures that are more resistant to shaking damage.
One-hundred-sixteen seismometers and accelerators were deployed by the University of California, Berkeley and Santa Barbara, in a 5,150-foot deep well (the Varian 1-A well near the San Andreas fault). A surface vibrator (VIBROSEIS) was used as a source of seismic energy to measure subtle changes in seismic velocities in the crustal rocks along the fault near Parkfield. The Parkfield network is described by Karageorgi et al.
Fluid pressure transducers
Pore-pressure fluctuations and variation in temperature in the over-pressured zone at the bottom of the Varian 1-A well are monitored. The proximity of the well not only to the fault, but also to the expected epicenter of the previously predicted magnitude 6 shock, suggests that these observations might be critical in understanding the details of the process leading to the next Parkfield earthquake.
Groundwater Radon and Soil Hydrogen
Although geotechnical data recorded at properly located stations has been found to show anomalous changes before earthquakes, the mechanism for these changes is not understood. Radon, hydrogen and carbon dioxide gas, and temperature in ground water were maintained in wells near Parkfield. Water samples were collected periodically from these and several other wells for chemical analysis. In addition continuous soil-gas radon and hydrogen monitors were installed near some of the groundwater monitoring sites.