Figure 5A (left). Telseismic locations made by inverting for model velocity and hypocenters simultaneously, adapted from Engdahl and Gubbins (1987). The depths for events of h < 40 km were fixed by near-epicenter reflections seen at teleseismic distances, or by data of the local network, and the map locations were determined thereafter from times of teleseismic P waves by the multiple inversion technique. The local P and S data were not used in the inversion for the map positions of these shallow events.

Figure 5B (right). Comparison of the teleseismic locations with those made by the Adak network (Fig. 3E). The misfit is puzzling as the general Adak locations near the ridge-terrace border were confirmed by the ocean-bottom experiment, Figure 4, however, temporal variations in the patterns can not be excluded. The teleseismic locations near the trench appear correct, as a 1979 ocean-bottom experiment (Frohlich et al., 1982), with seismometers deployed at the trench and oceanward, located events there as well.

Comments and questions should be directed to James N. Murdock, 611 Green Valley Dr. SE, Albuquerque, NM 87123, USA, (phone: 505.299.5254).