Earthquakes - Geotechnical Engineering Investigation - page 926

Dilatancy Theory
The dilatancy theory (or V p / V s anomaly, or seismic velocity ratio method) was based on the
observation that the arrival times of transient P waves traveling through the Earth's crust
undergo a gradual decrease when compared with the arrival times of S waves, until just
before an earthquake. Then the arrival time difference returns to normal relatively quickly
and is followed by the shock (Scholtz et al., 1973; Whitcomb et al., 1973). The time period
from the return to normal of the P wave velocity until the actual event has been found by
examination of pre-earthquake records to be roughly onetenth the time interval during
which the decreasing velocities occurred. If the time of the initial decrease and the return
to normal are known, predictions as to when the earthquake is likely to occur and what
the anticipated magnitude is likely to be can be made.
Rock-mass behavior under stress provides an explanation for the dilatancy theory. As the
crustal pressures preceding a quake approach the failure point in rock masses under high
stress, a myriad of tiny cracks open. This causes the decrease in the velocity of the P waves,
since their velocity is reduced when they travel across air-filled openings. As groundwa-
ter seeps into the cracks, the velocity increases until all of the cracks are filled and veloc-
ity returns to normal. The presence of the water “lubricates” the cracks, reducing rock
strength, permitting failure, and producing the earthquake. Some investigators now
believe that the stress level on faults is quite low and dilatancy, if it occurs, is likely to be
confined to small areas of stress concentration and not spread over significant volumes of
rock where it can be readily measured.
Deep-well injection has apparently verified the “lubrication” effect (which is in reality
probably a pore-pressure effect). A series of shocks occurred between 1962 and 1967 near
Denver, Colorado, following the pumping of liquid wastes down a borehole into rock at a
depth of 3 mi below the Rocky Mountain arsenal, a region where earthquakes were almost
unknown. After the waste pumping was suspended, the number of events declined
sharply. A similar experiment was carried out by the USGS at Chevron Oil Company's oil
field in Rangely, Colorado, in 1972. Water was forced under high pressure into a number
of deep wells and a series of minor earthquakes occurred. Activity ended immediately
when the water was pumped from the wells (Raleigh et al., 1972).

Surface Warping
Overstresses in the Earth's crust cause surface warping, which may predate an earth-
quake. Records from the literature regarding the phenomenon appear meager. Data are
available on ground surface elevation changes for many locations in the United States
from the Vertical Division Network, National Geodetic Survey, Silver Springs, Maryland.
Niigata, Japan: Japanese geologists reported that a land area near Niigata had risen 13 cm
in 10 years before the 1964 event ( M 7.5).

Palmdale, California: Measurements by the USGS determined that an area of about 4500
km 2 around Palmdale rose by as much as 45 cm between 1959 and 1974 as shown in Figure
11.23. The area, now known as the Palmdale bulge, is centered on the San Andreas Fault.
Subsequent data indicated that between 1974 and 1977, Palmdale had dropped by 18 cm
(Hamilton, 1978). Two earthquakes of M 5.7 and 5.2, which centered on the bulge,

occurred on March 15, 1979 (ENR, 1979).

Research and Monitoring Networks
Space-based instruments image Earth movements to millimeters, measuring the slow
buildup of deformation along faults, and mapping ground deformation after an earth-
quake. The primary methods are the GPS navigation system ( Section 4.2.2) and
Interferometric Synthetic Aperture Radar (InSAR, Section 2.2.3). Numerous GPS systems