Pwave residuals as a function of azimuth: 2. Undulations of the mantle low-velocity layer as an explanation

In an earlier paper Bolt and Nuttli presented evidence that the P wave delay times, relative to Berkeley, at a number of seismographic stations in central and northern California, exhibit a significant dependence on azimuth of wave approach. Numerical analysis of these data indicates that their amplitudes are too large to be accounted for by crustal structure only. The data can be explained by variations in the depth to both the top and the bottom of the mantle low-velocity channel for P waves. The derived model which is preferred has a velocity of 7.9 km/sec at the top of the mantle, 7.2 km/sec in the channel, and 8.2 km/sec below the channel; calculation then shows that the channel is thickest between the coast and the Central Valley of California (average depth to top of 45 km and to bottom of 225 km) and thins eastward (depths to top of 90 and 95 km and to bottom of 200 and 165 km at Shasta and Mineral, respectively) below the Sierra Nevada. The data also suggest that the lower surface of the channel perhaps rises in a westward direction from the coast. The model is compatible with Bouguer gravity data if one includes a crustal root extending to depths of the order of 40 km under the Sierra Nevada. The model is also consistent with existing heat flow data, which indicate that heat flow is less than normal in the Sierra Nevada and higher than normal in the regions to the west and east.