Radiated Seismic Energy from Very Small and Very Large Earthquakes

短波长携带了非常小的地震的最多信息，但由于地震波在地壳中传播的复杂性，很难辨别震源的影响。多重经验格林函数分析（Multiple empirical Green’s function analysis）是一种利用极小的地震来校正较大地震的复杂波传播效应的方法，当用于这些非常小的事件时，即使在深钻孔记录中也存在意想不到的波传播效应。这可能导致在过去的研究中对非常小的地震（小于2.0级）的能量学得出不正确的结论。这项技术正在应用于大量的数据集，以便观察这些结论是否适用于不同环境下的地震：即南加州的圣安德烈亚斯断层系统和日本变形的地壳。这是一项特别及时的研究，因为井眼记录中的传播效应可能会影响日本和加利福尼亚正在进行和计划进行的大规模井下实验的测量结果。这项技术由出井聪（Satoshi Ide）（东京大学）开发，用于大地震，从地震的强震模型中确定辐射地震能量的空间分布。对包括1995年日本神户地震在内的三次地震的研究表明，大部分地震能量是在震源附近辐射的，这表明断裂过程的其余部分在很大程度上是耗散的。这种方法正被应用到一个更大的扩展源模型的可用目录中，试图提高对断层期间能量平衡如何控制地震大小的理解。

For short wavelengths, which carry the most information for very small earthquakes, the complexity of wave propagation in the Earth's crust makes it difficult to discern effects due to the earthquake source. Multiple empirical Green's function analysis, a method that uses extremely small earthquakes to correct for complicated wave propagation effects for somewhat larger earthquakes, when used for these very small events shows that unanticipated wave propagation effects are present even in deep borehole recordings. This may have led to incorrect conclusions about the energetics of very small (magnitude less than 2.0) earthquakes in past studies. This technique is being applied to an extensive data set in order to see if these conclusions generalize to earthquakes in diverse environments: viz. the San Andreas fault system in southern California and the deforming crust of Japan. This is a particularly timely study because propagation effects in borehole recordings may affect measurements obtained from ongoing and planned large-scale downhole experiments in Japan and in California. The technique developed by Satoshi Ide (University of Tokyo) is being used for large earthquakes, to determine the spatial distribution of radiated seismic energy from strong motion models of the earthquake. Studies of a set of three earthquakes, including the 1995 Kobe, Japan earthquake, shows that most of the seismic energy was radiated near the hypocenter, suggesting that the rest of the faulting process was largely dissipative. This method is being applied to a much larger available catalog of extended-source models in an attempt to improve understanding how the energy balance during faulting controls the size of an earthquake.