Search for temporal changes in seismic velocities using large explosions in southern California

Abstract

For many years, large explosions within the Caltech seismic network have occurred periodically at a quarry near Corona (75 km southeast of Pasadena) and at the Eagle Mountain iron mine (240 km east of Pasadena). Explosions have taken place almost daily since 1948 at Eagle Mountain. The long-established station at Hayfield, very close to the mine, permits accurate determinations of apparent velocities to distant stations such as Barrett, and this particular path crosses the epicentral region of the 1968 Borrego Mountain earthquake (M = 6.4). The first arrival at Barrett, at a distance of 172 km, is a refracted wave, but a distinct second arrival presumably represents a crustal phase. There has been no systematic change in the apparent P-wave velocity of either phase since 1962; maximum observed variation from the mean velocities is less than 2% for 32 events distributed throughout the 1962-1973 period. Fewer but larger events have taken place at Corona since 1949. Nine of these events at roughly three-year intervals were recorded throughout the network as well as onsite for timing purposes. The observed variation in P-wave velocity is less than 3-1/2% for all paths, and all variations can probably be explained by instrumental and reading inaccuracies. Because of the timing of events and their locations, the results from Eagle Mountain and Corona do not completely rule out dilatancy effects prior to the Borrego Mountain and San Fernando earthquakes -- the two major shocks of this period. But the observations do put certain restrictions on the possible size and nature of any proposed dilatancy regions, and there are no discernible systematic trends in the 24 years of data along any paths.