Double Segmentation of the 2011 Tohoku-Oki Earthquake: Uniqueness and Similarity to Other Megathrust Earthquakes

We present some singular characteristics of the 2011 Tohoku-Oki earthquake in comparison with other megathrust earthquakes, such as the 1960 Chilean and the 2004 Sumatra-Andaman earthquakes. In addition to conventional along-strike segmentation, along-dip segmentation of the fault area or subduction zone (we call it double segmentation while the conventional one called by single segmentation) is required, as indicted by the difference in background seismicity: virtually no seismicity in shallow segments but active with large events (M=7 class) repeating in deep segments. The interaction between along-dip segments (i.e., deep and shallow segments) led to the great 2011 Tohoku-Oki earthquake. The along-dip or double segmentation results in plane or two-dimensional rupture propagation on a coseismic fault. Significant along-strike variability is also important for the 2011 Tohoku-Oki earthquake, with segments of both weak (e.g., slow or tsunami earthquakes) and strong plate couplings located adjacent to each other. In contrast, every segment appears to be of strong plate coupling for other megathrust earthquakes of single segmentation. One exception is the 1964 Alaska earthquake that shares a certain degree of common characteristics with the 2011 Tohoku-Oki earthquake: two distinct seismogenic zones along the dip direction of the trench, that is, the double segmentation is noticeable. Dramatic along-strike variability also characterized activities in and around the subduction zone of the 1964 Alaska earthquake, including a creeping segment and a tsunami-earthquake segment. The classification of single versus double segmentation helps us to understand an apparent variety of megathrust earthquakes.