A fault model for the Niigata, Japan, earthquake of June 16, 1964.

place on a fault dipping 56° westward and striking N9°E. The motion is a pure dip-slip on this fault, the west side moving up relative to the east. The average dislocation was estimated to be 3.3m on a fault plane of 80×30kmdm2 in area. This fault model is hereafter referred to as high-angle fault model. The high-angle fault model produces surface deformation in general harmony with the results inferred from geodetic measurements both on the coast of Honshu and on an islet called Awashima, located about 8km north of the epicenter (for details, see Paper I). However, it appears that this model is not consistent with the sea-bottom deformation inferred from the echo-sounding survey which was repeated over the epicentral area both before and after the earthquake; the survey by MOGI et al. (1964) reveals a predominant uplift of the sea-bottom in a wide area, while the high-angle fault model produces both subsidence and upheaval in the epicentral area (H. Kanamori, personal communication, 1983). In this paper, we try to interpret the vertical deformation data, including the echo-sounding data, in terms of faulting on a complementary, low-angne plane which is the auxiliary plane in the double-couple fault representation. A method similar to that used inl Paper I is employed. In Paper I, the high-angle fault plane is placed so as to intersect the central part of the aftershock area, by reason that small submarine faults were found there by MOGI et al. (1964)after the earthquake. The motion on the low-angle fault