Annual Report, l997

Evidence of Fault Zone Healing after the 1992 M7.S Landers,
California, Earthquake

Yong-Gang Li, John E. Vidale and Keiiti Aki

Our earlier studies of fault zone trapped waves from Landers aftershocks allowed us to delineate the Johnson Valley fault, to a depth of at least 10 km, marked by a lowvelocity waveguide 100 to 200 m wide where shear velocity is reduced by 30 to 50 percent and Q~30 (Li et al., 1994a,b). From the view point of fracture mechanics, the distinct low-velocity waveguide on the ruptured faults may represent a process zone of inelastic deformation around the propagating crack tip (e.g. Ida, 1973; Rice, 1980; Papageorgiou and Aki, 1983; Scholz et al., 1993; Marone and Kilgore, 1993). The in-site fault healing has been found at the Calaveras, California, fault by Marone et al. (1995) when they analyzed the data collected from repeated earthquakes occurring on the same fault patch (Vidale et al., 1995). It is likely that the waveguide on the Johnson Valley fault could be softened by the dynamic rupture in the Landers earthquake, and is regaining the strength before the future event. Our recent observations in repeated surveys support this brokenthen-healing cycle of an active fault.

Repeated seismic surveys revealed an increase in.seismic velocity with time. P, S, and fault zone trapped waves were excited by near-surface explosions in two locations on November of 1994 and August of 1996, and recorded on 2 linear, three-component seismic arrays deployed across the Johnson Valley fault trace. The travel times of P, S, and trapped waves for identical shot-receiver pairs decreased by 0.5 to 1.5 percent from 1994 to 1996, with the larger changes at stations located within the fault zone. These observations indicate that the shallow Johnson Valley fault is strengthening after the mainshock, most likely due to the closure of cracks that were opened by the 1992 earthquake. The increase in velocity is consistent with the partially fluid-filled cracks, and with a reduction in the apparent crack density in the fault zone by ~1.0 percent from 1994 to 1996.

In the third repetition of the experiment at the Johnson Valley fault in October of 1997, we obtained the data from one explosion located in the northern part of the JVF. The 1997 data are consistent with similar amounts of healing on the northern part of the JVF, and show slower healing from 1996-1997 than from 1994-1996. However, the operational difficulties have prevented us from acquiring much data that samples the same fault segment during all three experiments. We need further data to study if this healing process is linear or logarithmic with time in a longer time scale.

Fig. 1 shows locations of explosions and seismic arrays in repeated experiments conducted in 19941997. Fig. 2 illustrates the P, S and fault zone trapped waves recorded at line 1 for shot 4 in 1994 and 1996. Fig. 3 illustrates the data recorded at line 1 for shot 3 in 1994 and 1997. Trapped waves with relative large amplitudes, long period and slight dispersion appeared only at stations located within the fault zone. The explosion-excited trapped waves are similar to those generated by microearthquakes occurring within the fault zone at Landers, but have lower frequencies and travel more slowly, suggesting that the fault zone is slower and possibly wider as it approaches the surface. P, S and trapped waves had the similar waveforms between 1994 and 1996 (or 1997), but traveled faster in 1996 and 1997 than in 1994, For example, Fig. 4a shows more dearly the traveltime decreases of these waves from 1994 to 1996 at station W1, and Fig. 4b shows the traveltime advance from 1994 to 1997 at station E1 within the fault zone. We extracted P,
S and trapped waves from 4 time windows and cross-correlated each pair of recordings for the same shot and seismometer to obtain time differences between 1994 and 1996 (or 1997) recordings. All waves arrived faster in 1996 and 1997 than in 1994. Note that trapped waves showed larger time advances than P and S waves because they had longer times. Fig. 5a shows the traveltime decreases from 1994 to 1996 for 2 shots and 2 arrays, including 112 measurements from shot-receiver pairs. The arrivals in the middle of the linear arrays, near the fault trace, have a greater time advance from 1994 to 1996 than the arrivals at stations toward the edges of the arrays. The decrease in the travel time of these waves is largest within the fault zone. The ratio of decrease in travel time for P to S waves

(Atp/AtS) between 1994 and 1996 is 0.77 with a standard deviation of 0.07. This value is valid for all P and S arrivals both within and outside the fault zone. Fig. 5b shows traveltime decreases from 1994-1997 for the pair of SP3 and line 1 with Atp/AtS) of ~0.65.

The increase in the velocity of P and S waves with time is most likely due to the closure of dry cracks as the crust heals after the earthquake. Estimates of the change in velocity due to the change in the density of cracks may be calculated (Garbin and Knopoff, 1973, 1975a,b). We assumed randomly oriented cracks, although there may be some alignment, whose coherence is not simple to predict and may change with time. Dry cracks in a Poisson solid are predicted to cause Atp/AtS ~1.22. Water-filled cracks, on the other

hand, cause Atp/AtS ~0.27 only. So the observed Atp/AtS ~0.65-0.77 suggests that cracks are partially fluid-filled. Also, the apparent density of saturated cracks may decrease more slowly than the density of dry cracks immediately after an earthquake. A 1% increase in the seismic velocities is expected from roughly 1% decrease in the apparent crack density.

We conclude that some cracks which had opened during the mainshock closed soon thereafter. The closure of cracks 2 to 5 years after the earthquake is consistent with our tentative interpretation of the strong low-velocity Landers fault-zone waveguide as being at least partially created during the mainshock (Li et al., 1994a,b). Also, such a pattern of healing fault zones may help explain observations of increasing stress drop with increasing recurrence intervals (Kanamori and Allen, 1986; Scholz et al., 1986; Houston, 1990).

Publications Related to this Project:

Li, Y.-G., J. E. Vidale, K. Aki, F. Xu, and T. Burdette, Evidence of shaUow fault zone strengthening after the 1992 M7.5 Landers, California, earthquake, submitted to Science, 1997.

Li, Y. G., K. Aki, and a field team (D. Adams, D. Bowman, T. Burdette, G. Ely, M. Forrest, A. Martin, J. Wedberg and A. Wei), Detecting the healing of the fault zone ruptured in the M7.5 Landers, California, earthquake of 1992 using fault-zone trapped waves, Seism. Res. Lett., 68,334, 1997.

Li, Y. G., W. L. EUsworth, C. H. Thurber, P. E. Malin, and K. Aki, Observations of fault-zone trapped waves excited by explosions at the San Andreas fault, central California, Bull. Seism. Soc. Am., 87, 210-221, 1997.

r i, Y. G. and J. E. Vidale, Low-velocity fault zone guided waves; numerical investigations of trapping
efficiency, Bull. Seism. Soc. Am., 86, 371-378, 1996.

Li, Y. G., and K. Aki, Monitoring post-seismic changes of the fault zone of the 1992 M7.5 Landers,

California, earthquake using explosion-excited fault-zone trapped waves, EOS, Trans. Am. Geophys. Union, 77, No.46, 508, 1996.

Li, Y. G., K., Aki, B.-H. Chin, D. Adams, B. Periklis, J. B. Chen, Explosion-excited fault-zone guided waves at Landers, California, Seism. Res. Lett., 66, l99S.

Li, Y. G., K. Aki, D. Adams, A. Hasemi, W. H. K. Lee, Seismic guided waves trapped in the fault zone of the Landers, California, earthquake of 1992, J. Geophys. Res., v99, B6, 11,705-11,722, 1994a.

Li, Y. G., J. E. Vidale, K. Aki, C. J. Marone, W. H. K. Lee, Fine structure of the Landers fault zone; segmentation and the rupture process, Science, 256, 367-370, 1994b.

Li, Y. G., K. Aki, J. E. Vidale, P. Malin, Observations and implications of fault zone guided waves at Landers and Parkfield, invited, EOS, Trans. Am. Geophys. Union, 75, No.44, 469, 1994.