Tsunami earthquakes are a group of enigmatic earthquakes generating disproportionally large tsunamis relative to seismic magnitude. These events occur most typically near deep-sea trenches. Tsunami earthquakes occurring approximately every 10 years near Torishima on the Izu-Bonin arc are another example. Seismic and tsunami waves from the 2015 event [M_w (moment magnitude) = 5.7] were recorded by an offshore seafloor array of 10 pressure gauges, ~100 km away from the epicenter. We made an array analysis of dispersive tsunamis to locate the tsunami source within the submarine Smith Caldera. The tsunami simulation from a large caldera-floor uplift of ~1.5 m with a small peripheral depression yielded waveforms remarkably similar to the observations. The estimated central uplift, 1.5 m, is ~20 times larger than that inferred from the seismologically determined non-double-couple source. Thus, the tsunami observation is not compatible with the published seismic source model taken at face value. However, given the indeterminacy of M_zx, M_zy, and M_{tensile} of a shallow moment tensor source, it may be possible to find a source mechanism with efficient tsunami but inefficient seismic radiation that can satisfactorily explain both the tsunami and seismic observations, but this question remains unresolved.

中文翻译：

---

日本鸟岛附近的2015年火山海啸地震的机理。

海啸地震是一组神秘的地震，相对于地震震级产生过大的海啸。这些事件最典型地发生在深海海沟附近。另一个例子是在伊豆-波宁弧线上的鸟岛附近大约每10年发生一次海啸地震。2015年大地震的地震波和海啸波[ M _w（矩量级）= 5.7]由离震中点约100 km的10个压力表的近海海底阵列记录。我们对弥散性海啸进行了阵列分析，以找到海底史密斯火山口内的海啸源。由约1.5 m的大火山口地板隆起和外围凹陷引起的海啸模拟产生的波形与观测结果极为相似。估计的中心隆起1.5 m比根据地震学确定的非双偶合声源推断的隆起高约20倍。因此，海啸观测与以面值获取的已发布地震源模型不兼容。但是，鉴于M _zx，M _zy和M _{tensile}的不确定性 对于一个浅矩张量源，也许有可能找到一种能有效解释海啸和地震观测值的，具有有效海啸但无效率地震辐射的震源机制，但这个问题仍未解决。