The Mechanics of Subduction in Japan from High-Precision Earthquake Location and Tomography

从高精度地震定位和层析成像研究日本俯冲机制I. Gregory C. Beroza俯冲带板块边界是已知最大地震的所在地。 它们也是破坏性的上地壳和中深度地震的发生地，这些地震本身就构成巨大的地震危险。 在这项资助下，我们正在提高对日本俯冲带浅层和中深层地震机制的理解。 日本是我们学习的理想之地，有几个原因。日本的俯冲带表现出多种多样的样式，包括：北海道和本州北部的古老洋壳快速俯冲，本州南部和四国的年轻的菲律宾海板块的近平俯冲，以及九州南部的相对年轻的洋壳的陡俯冲。 日本是地震高发区。 自1994年以来，日本地震台网记录的地震超过50万次，并出现在JMA目录中。 真正使日本的俯冲带与地球上其他地方的俯冲带区别开来的是地震仪器的水平。 日本的俯冲带有1500多台高增益短周期地震监测仪器，是迄今为止世界上最好的仪器。 此外，新的Hi-Net地震网络目前有近700个三分量钻孔台站，提供了其他地方没有的独特地震监测能力。 在我们的研究中，我们正在应用和改进最近开发的高精度，多事件地震重新定位和高精度层析成像技术在日本的地震。 这些方法揭示了世界上其他地震活跃地区断层活动过程的显著细节，为地震的发生机制提供了新的见解，将其应用于日本也有类似的前景。我们的研究有两个主要目标。 首先是试图了解板块边界界面的几何形状和力学，那里发生了最大的地震，以及限制其发生深度范围的因素。 精确的地震位置和它们相对于深度速度结构的位置是这种分析的关键。 第二，我们试图将精确定位的地震与下行板块的详细地震速度结构联系起来。 我们特别感兴趣的是限制可能的作用，由于变质作用在下行板块的脱水反应中发挥使中深度地震。

THE MECHANICS OF SUBDUCTION IN JAPAN FROM HIGH-PRECISION EARTHQUAKE LOCATION AND TOMOGRAPHYP.I. Gregory C. BerozaSubduction zone plate boundaries are host the largest known earthquakes. They are also the sites of damaging upper crustal and intermediate-depth earthquakes that in themselves pose substantial seismic hazard. Under this grant we are developing an improved understanding of the mechanics of shallow and intermediate-depth earthquakes in the subduction zone in Japan. Japan is the ideal place for our study for several reasons. Subduction zones in Japan exhibit a diversity of styles including: rapid subduction of old oceanic crust in Hokkaido and North Honshu, nearly flat subduction of much younger Philippine Sea plate in South Honshu and Shikoku, and much steeper subduction of relatively young oceanic crust under the southern island of Kyushu. Japan is highly seismic. Since 1994 there have been over half a million earthquakes recorded by seismic networks in Japan that appear in the JMA catalog. What really sets the subduction zones in Japan apart from those elsewhere on Earth is the level of seismic instrumentation. With over 1500 high-gain short period instruments available for earthquake monitoring, the subduction zones in Japan are by far the best instrumented in the world. Moreover, the new Hi-Net seismic network, which now numbers nearly 700 3-component borehole stations, offers a unique capability for earthquake monitoring found nowhere else. In our research we are applying and improving recently developed high-precision, multiple-event earthquake relocation and high-precision tomography techniques to earthquakes in Japan. These methods have revealed remarkable details of the faulting process in other seismically active regions of the world that have lead to new insights into how earthquakes work and their application to Japan holds similar promise.Our research has two main goals. First are trying to understand the geometry and mechanics of the plate-boundary interface, where the very largest earthquakes occur, and the factors that limit the depth extent of their occurrence. The combination of precise earthquake locations and how they are situated relative to the velocity structure at depth is the key to this analysis. Second, we are trying to relate precisely located earthquakes to the detailed seismic velocity structure of the downgoing plate. We are particularly interested in constraining the possible role that dehydration reactions due to metamorphism in the downgoing plate play in enabling intermediate-depth earthquakes.