Initial investigation of the Landers, California, Earthquake of 28 June 1992 using TERRAscope

• Published in: Geophysical research letters Vol. 19; no. 22; pp. 2267 - 2270
• Main Authors: Kanamori, Hiroo, Thio, Hong-Kie, Dreger, Doug, Hauksson, Egill, Heaton, Tom
• Format: Journal Article
• Language: English
• Published: Blackwell Publishing Ltd 20.11.1992
• Subjects: Directivity; Faults; Inversions; Rupture; Seismic phenomena; Slip; Stresses; Surface waves
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1029/92GL02320

The 1992 Landers earthquake (Ms=7.5, Mw=7.3) was recorded at six TERRAscope stations in southern California. Peak accelerations ranged from 0.16 g at SVD (Δ=63 km) to 0.0092 g at ISA (Δ=245 km), decreasing with distance away from the fault zone. The peak velocity showed a different pattern reflecting the rupture directivity from south to north. The largest peak velocity, 19 cm/sec, was observed at GSC (Δ=125 km). Moment tensor inversion of long‐period surface waves yielded a mechanism with M0=1.1×1027 dyne‐cm (Mw=7.3), dip=74°, rake=−176°, and strike=340°. Inversion of teleseismic P and S waves revealed two distinct sub‐events of 6 and 8 sec duration and about 10 sec apart. The source parameters for the first and second events are: M0=1.9×1026 dyne‐cm, dip=83°, rake=179°, strike=359°; and M0=6.1×1026 dyne‐cm, dip=87°, rake=178°, strike=333°, respectively. The radiated wave energy, ES, was estimated as 4.3×1023 ergs. The ratio Es/M0=3.9×10−4 corresponds to a stress drop of 280 bars, and suggests that the Landers earthquake belongs to the group of high stress drop earthquakes, and occurred on a fault with a long recurrence time. The rupture directivity can be seen clearly in the records from PFO (Δ=68 km) located to the south and GSC located to the north of the epicenter. The maximum displacement at PFO is only 13% of that at GSC despite the shorter epicentral distance to PFO than to GSC. The slip distribution determined with the empirical Green's function method indicates that the Landers earthquake consists of two distinct sub‐events about 30 km apart, with the second sub‐event to the north being about twice as large as the first one. This slip distribution is consistent with the teleseismic data and the surface offsets mapped in the field.