The May 7, 1986, Andreanof Islands earthquake (Mw 8.0) is the largest event to have occurred in that section of the Aleutian arc since the March 9, 1957, Aleutian Islands earthquake (Mw 8.6). Teleseismically well-reported earthquakes in the region of the 1986 earthquake are relocated with a plate model and with careful attention to the focal depths. The data set is nearly complete for mb≥4.7 between longitudes 172 ¿W and 179 ¿W for the period 1964 through April 1987 and provides a detailed description of the space-time history of moderate-size earthquake in the region for that period. Additional insight is provided by source parameters which have been systematically determined for Mw≥5 earthquakes that occurred in the region since 1977 and by a modeling study of the spatial distribution of moment release on the mainshock fault plane. A tectonically significant component of oblique convergence in the central Aleutian arc results in its breakup into clockwise-rotating and westward translating blocks. The western part of the Andreanof block is distinct from and stronger than flanking regions. The greater strength of this block segment and strong coupling along the main thrust zone result in the accumulation of high levels of shear stress, which give rise to great earthquakes near its eastern boundary. The occurrence of the 1986 rupture only 29 years after the 1957 earthquake may indicate that in the central Aleutians M7+earthquakes ordinarily do represent the predominant mode of strain release. Segmentation of the main thrust zone into upper and lower planes is supported by spatial and temporal patterns of seismicity and by focal mechanism data. This and other lines of evidence indicate a downdip increase in fault strength and possibly in heterogeneity within the main thrust zone in subduction zone environments. Aftershocks of the 1986 earthquake were bounded to the west by the Andreanof block boundary (Adak Canyon) and to the east by an aseismic segment of the main thrust zone near the subducted extension of the Amlia fracture zone. The aftershock distribution was bounded to the south by a forearc shear zone and to the north by the base of the main thrust zone in the Hawley Ridge segment and by the downdip edge of the upper plane of the main thrust zone in the eastern segment. Aftershocks which occurred near the volcanic line at shallow crustal depths in the upper plate were triggered by the mainshock and manifest a partial decoupling of oblique slip in this region along a west-striking right-lateral fault with low shear strength. Aftershock clustering along the main thrust zone was very similar to the distribution of prior seismicity, suggesting a continuation of long-term processes and the existence of areas with distinct mechanical properties. Interconnecting regions of low seismicity during both the premainshock and aftershock periods coincided with areas of major moment release during the mainshock. Seismicity data prior to the 1986 mainshock gave few clues about the location of the mainshock nucleation point, the mainshock size, and its time of occurrence. However, a large part of the mainshock moment release did coincide with a zone of seismic quiescence monitored by the Adak local seismograph network. ¿ American Geophysical Union 1989 |