Broadband study of the 1989 Loma Prieta Earthquake

Publisher Website
Google Scholar
Abstract
We determined the source characteristics of the Oct. 18, 1989 Loma Prieta, California, earthquake using teleseismic data. The solution from body waves gives a mechanism with dip=70°, rake=138°, and strike=128°, and a seismic moment of 3×1026dyne‐cm (Mw=6.9). This solution is similar to those obtained from long‐period Rayleigh and Love waves, PnLwaves, and first‐motion data. The body wave form suggests a depth of about 15 km. The effective duration of the source is 6 sec, which suggests a fault length of 30 km and 15 km for bilateral and unilateral faulting, respectively. Considering the extent of the aftershock area, we used 35 km as an estimate of the total rupture length. The strike‐slip and the thrust components of the coseismic slip are 177 and 159 cm, respectively. The large thrust component raises an important question regarding the recurrence pattern. If the Loma Prieta earthquake is a characteristic earthquake with a recurrence interval of about 100 years, the 159 cm displacement implies a long term uplift rate of about 1 cm/year, which appears too large for this region. Three hypotheses for reconciling this apparent conflict are: (1) the geometry of plate motion along the Santa Cruz Mountain segment changes over a time scale of several thousand years so that the coseismic displacement has not accumulated enough to produce high topographic relief; (2) the coseismic sHp direction varies from event to event; (3) the slip plane of the 1989 Loma Prieta earthquake is different from the major boundary between the Pacific and the North American plates; if this is the case, the Loma Prieta earthquake is a rather rare non‐characteristic event. The surface slip of about l m for the 1906 San Francisco earthquake is one of the key data in long‐term forecasting. No surface slip was observed for the 1989 Loma Prieta earthquake, even if the horizontal slip at depth is as large as 1.8 m. This points to a risk of relying too heavily on surface observations for long‐term seismic risk analysis.