VOL. 84, NO. B9 JOURNAL OF GEOPHYSICAL RESEARCH AUGUST 10, 1979 Intraplate Seismicity on Bathymetric Features: The 1968EmperorTrough Earthquake SETH STEIN Department of Geophysics, Stanford University, Stanford, California 94305 TheApril 28, 1968 (mo = 5.5,Ms = 5.2),earthquake in theEmperor Trough, a long linear deep east of theEmperor Seamounts in theNorth Pacific, istheonlyrecorded seismic event in this region. The focal mechanism, determined using bodyand surface waves, shows faulting striking along the trough with the eastsidethrustupward. This mechanism is consistent with bathymetry showing the eastflank 500 m higher thanthewest and suggests that the earthquake represents part of continuing deformation at the trough. This earthquake is one of a largenumber of intraoceanic plate events that occur on major bathymetric features. The focal mechanisms of earthquakes on such features probably represent the reactivation of already present faults or weak zones. The orientations of intraplate stresses inferred from the focal mechanisms may differ significantly from the true stress directions. INTRODUCTION Earthquake focalmechanisms in the interiorof lithospheric platesprovide a methodof studying intraplatestresses and thusyieldsome insight on the driving forces of platetectonics. Mendiguren [1971] analyzed an earthquake within the Nazca plate; Forsyth [1973] studied one between thePacific-Antarctic andChile ridges. A global compilation of intraPlate earth- quakemechanisms is given by Sykes and Sbar [1974]. Both Mendiguren [1971]and Forsyth [1973]studied isolated earthquakes, not associated with major bathymetric features, whose mechanisms are most likely to represent average intra- plateratherthan local stresses. Many oceanic intraplate earth- quakes, on the other hand, are concentrated on bathymetric features.The intraplate seismicity of the Indian Ocean, for example,occurs along the Ninetyeast Ridge [Steinand Okal, 1978]and on the Chagos-Laccadive Ridge [Stein, 1978]. The purpose of thispaper is to examine an earthquake in the Emperor Trough,a major feature of theNorth Pacific Ocean, and then discuss the overall phenomenon of earthquakes on preexisting bathymetric structures. A large number of oceanic intraplate events, including the largest everrecorded (the mag- nitude 7 eventson the NinetyeastRidge) occur on suchfea- tures, whichseem active tectonically today.The fact that these earthquakes probably occur by reactivation of older weak zones has major implications for their usein studies of intra- plate stresses. EMPEROR TROUGH The Emperor Troughis a longlineardeep east of the Em- peror Seamounts. The trough trends northwest from 37øN, 175øW to about 48øN, 172øE, with its axis 1000-1500m below the sea floor on its flanks [Erickson et al., 1970].Larson and Chase [1972], reconstructing the late Mesozoic evolution of the western Pacific, proposed that the troughdates from the Late Cretaceous. In their model the trough was originally an activespreading center, part of the Pacific-Farallon ridge, which generated a portion of theHawaiian Mesozoic lineation set. At approximately 100 m.y.B.P. thePacific-Farallon-Kula triple junction jumped 2000 km south, converting theEmperor Troughto a transform fault between the Kula and Pacific plates. Hilde et al. [1976] also interpret thetrough asa Kula- Pacific transform. Copyright ¸ 1979 by the American Geophysical Union. In thesemodels the Emperor Trough predates the Emperor Seamount chain to the west and was formed through an en- tirely differentprocess. The Emperor Seamounts, interpreted asthe traceof the Hawaiian hot spot [Morgan, 1972; Gordon et al., •978],were formed much later, since theoldest one, Meiji SeamoUnt, dates at 67-70m.y.B.P. [Clague and Jarrard, 1973] and extends further north, toward the cusp formed at the intersectionof the K urile and Aleutian trenches[Vogt et al., 1976]. The final phase of the subduction of the Kula-Pacificridge east of the trough into the Aleutian trench beganin the mid- Tertiary (about 30 m.y.B.P.) [Growand Atwater, 1970]. Since that time the present plate motion (direction shownin Figure 1) has continued, with no major tectoniceventsin the area. The Emperor Trough is in an extremelyseismically quiet region. During the period 1960-1977, in the region 165øE - 170øW and 37øN-48øN, only one earthquake(and an after- shock) was recorded. This curiousearthquakeoccurred in the Emperor Trough and is the subject of this investigation. FOCAL MECHANISM The April 28, 1968, earthquake(origin time 0418:15 UT) was located by the International SeismologicalCenter at 44.8øN, 174.6øE. It was followed about 2 hours later (0631:54 UT) by a small aftershock,located at 44.4øN, 174.7øE.The location, shownon the bathymetry of Figure I [Chase et al., 1970], is almost exactly in the Emperor Trough. The body wave magnitude reportedby the preliminaryde- termination of epicenters was 5.5. The surfacewave magni- tude, determined from Rayleigh waves on the long-period vertical component at 14 World-Wide StandardSeismograph Network (WWSSN) stations (MAT, ALQ, MSH, LON, QUE, ANP, SEO, COL, DAV, CAR, OGD, LPB, HNR, and AFI) is 5.2. The earthquakeis extremelysmall for study by teleseismic methods.Only at a few very close stations is the body wave arrival visible on the long-periodrecords.The short-period vertical records, however, often show clear first arrivals. First motions, from WWSSN and Canadian network stations, al- most all from the short-period vertical components, are plotted on the lower focal hemisphere in Figure 2. It is clear that the stationgeometry is unfavorable, both in distance and azimuthal coverage. The first motions, as is common with the short-period instrument, are somewhat scattered.The com- pressional arrivals in the center of the focal sphereshow a Paper number 9B0209. 0148-0227/79/009B-0209501.00 4763