SUMMARY
For the groundwater level observed at the Haibara well, Shizuoka Prefecture, central Japan, time series analysis using state-space modelling is applied to extract hydrological anomalies related to earthquakes. This method can decompose observed groundwater level time series into five components: atmospheric pressure, tidal, and precipitation responses, observation noise, and residual water level. The decomposed responses to atmospheric pressure and precipitation are independently determined and are consistent with the expected response to surface loading. In the groundwater level at the Haibara well, 28 coseismic changes can be discerned during the period from 1981 April to 1997 December. There is a threshold in the relationship between earthquake magnitude and the well–hypocentre distance, above which earthquakes cause coseismic changes in the residual water level. All of the coseismic water level changes at the Haibara well are decreases, although 33 per cent of the estimated coseismic volumetric strain steps are contraction, which would be expected to cause water level increases. The coseismic changes in groundwater level are more closely proportional to the estimated ground motion than to coseismic volumetric strain steps, suggesting that ground motion due to earthquakes is the major cause of the coseismic water level drops and that the contribution from static strain is rather small. Possible pre- or inter–earthquake water level changes have occurred at the Haibara well and may have been caused by local aseismic crustal deformation.
Abstract The Plate Boundary Observatory (PBO), the geodetic component of the U.S. National Science Foundation–funded Earthscope program, includes 75 borehole and 6 laser strainmeters ( http://pbo.unavco.org ). The strainmeters are installed at several locations: on the Cascadia forearc in Washington state and on Vancouver Island, Canada; in arrays of two to nine instruments along the North American–Pacific plate boundary in California; at Mount St. Helens; and in Yellowstone National Park. For deformation signals seconds to weeks in duration, strainmeters have a resolution and a signal‐to‐noise ratio superior to those of seismometers and GPS. However, this high sensitivity can introduce nontectonic signals into strain data, presenting data interpretation challenges, especially for borehole strainmeters.
This report compiles hydrologic observations in southern California and elsewhere associated with the 1992 ML = 7.3 Landers, California earthquake sequence.In southern California, the largest ground-water-level changes were a rise of 3 meters at Lucerne Valley and a drop of 5 m at Pinon Flat.Most of the steplike water-level changes recorded in the hours following the Landers and Big Bear earthquakes agreed in direction with the sign of the calculated coseismic volume strain field.In the Pinon Flat area, however, two wells measured on June 28, after both these earthquakes, displayed water-level rises of 9 cm above the reading made two days before.A spring discharge increase in Millard Canyon was reported to have preceded the earthquake by several days.Outside of southern California, water-level changes were also observed, but are not consistent in sign or size with the static strain field of the earthquake sequence.At Parkfield, California, water-level changes took place in three wells at the time of the earthquake, and recovered over periods as long as 30 days.At Long Valley, California, observed water-level changes generally returned to normal after minutes to hours, consistent with their having been caused by the passage of surface waves.Water levels in one well at Long Valley and in a well near Grants Pass, Oregon, remained low for at least two days following the earthquake.Water-level oscillations took place in two wells in eastern Nevada.Phenomena accompanying the Landers earthquake that were of practical significance include the Tapo Canyon oil seep, which polluted part of the Santa Clara River; gas bubbles in San Bernardino city water supply wells, which clogged filters; and a coseismic discharge increase in Millard Creek, which added to the water supply.1. Includes response to Big Bear earthquake.
