Surface Rupture Reevaluation of the 1951 Earthquake Sequence and Neotectonic Implication of East Taiwan

In 1951, a sequence of destructive earthquakes occurred in eastern Taiwan from October to December, including 788 felt earthquakes among total 3090. The earthquake sequence comprised of two major phases in a time span of one mouth. The first phase was characterized by three successive disastrous earthquakes (Ms7.3, 5:34am; Ms7.1, 11:29am; Ms7.1, 1:43pm) in the northeastern offshore of the Hualien city on Oct. 22nd. One month later, two large earthquakes (Ms6.0, 2:47am; Ms7.3, 2:50am) resurged 100 km farther south in Yuli on Nov. 25th characterizing the second phase. The 1951 surface ruptures were mapped as three earthquake faults: the Meilun, Yuli, and Chihshang fault from Hualien southward. For the purpose of reconstruction the spatial relationship along the 1951 surface ruptures, we remap them by literature collection, field checking, and oral interview. In general, the ruptures show oblique left-lateral slip with nearly vertical fault planes. The maximum surface offset occurred at the northern tips for each segment, probably suggesting that the directivity of the slip of the earthquake propagated toward the north. The newly presented rupture map in this study also shows that the Meilun and the Yuli fault both curve toward northeast near the northern tip. We further consider published geological map and the related geomorphic features to divide the entire Longitudinal Valley fault system into segments including the Meilun, Yuli and Chihshang fault. Each segment runs NNE-SSW attached on the western flank of Coastal Range with a bending northern end pointing into the Coastal Range. We therefore proposed a kinematic tectonic model, dividing the entire Coastal Range into two major entities. The southern one is rather intact without observable major fault cutting through, while the northern one can be further subdivided into four minor tectonic blocks. Each tectonic block is of distinct geologic and geophysical characteristics, as shown by data of lithology, gravity, GPS geodesy, paleomagnetism, and so on. Since some of the boundaries between the proposed tectonic blocks are 1951 surface ruptures, we believe the others are potential active faults that need further paleoseismology study to confirm.