Collaborative Research: Investigating Structure and Seismicity Within the Southern M9.2 1964 Great Alaska Earthquake Rupture Area Using a Dense Node Array

俯冲带是大洋板块沿着浅倾斜的断层滑动到大陆板块下方的区域。一个例子是阿拉斯加-阿留申俯冲带，它比世界上任何其他俯冲带发生的8级或更大的地震都多，其中包括1964年的阿拉斯加9.2级大地震。科迪亚克岛位于阿拉斯加-阿留申俯冲带断层的一部分上方，该断层是1964年地震的罪魁祸首，2019年，作为阿拉斯加两栖社区地震实验(AACSE)的一部分，400台地震仪器被部署在那里25天。利用机器学习和其他技术，沃辛顿和她的团队将重新分析这些仪器收集的数据，以识别许多非常小的地震，并精确定位它们的位置。这将提供一个前所未有的快照，了解地震是如何沿俯冲带断层及其附近分布的。这些地震和其他地震的记录将被用来产生地下图像，包括俯冲带断层本身，并估计断层及其周围的物理性质。对浅俯冲弧前区内地震活动和结构的详细描述，有助于洞察沿孕震大断裂的板块界面性质，以及活动和过去变形的分布、位置和运动学。该项目利用2019年5月阿拉斯加两栖社区地震实验(AACSE)中沿约50公里横断面部署的密集地震节点阵列的现有数据，描述了奇克塔克(科迪亚克岛)下的地震活动和结构。科迪亚克节点阵列位于美国历史上最大的地震-1964年阿拉斯加大地震Mw9.2的南部凹凸处，这场地震破裂了阿拉斯加俯冲带600-800公里的一段。该项目的主要目标是确定：1)通过增加微震活动性(M1.0)的探测和重新定位，确定两个地震旋回期间上板块、下板块和界面地震活动的分布；2)1964年南部地震破裂凹凸区浅前弧内板块界面的性质，其性质控制地震的破裂、传播和引发；3)科迪亚克地区内的上板块速度结构，对前弧向海部分的材料强度和弹性性质提供限制。使用机器学习和基于合并的自动检测算法，该项目将在25天的部署期间编制一个增强的地震活动目录。将改进现有和新探测的位置，以提供大地震后两次地震间隔期间地震活动分布的快照。来自活动震源海洋炮击和当地地震(包括新探测到的事件)的旅行时间将用于光线追踪和反演，以建立板块界面上方浅层弧前的高分辨率速度模型。这个速度模型又将被用来进一步确定事件发生的地点。来自当地地震的能量将用于利用散射波和反射相位对节点阵列下方的地壳和板块界面进行成像，以进行垂直反射剖面和尾波自相关。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

A subduction zone is a region where an oceanic plate slips below a continental plate, along a shallowly inclined fault. One example is the Alaska-Aleutian subduction zone, which is responsible for more Magnitude 8 or greater earthquakes than any other subduction zone in the world, among them the 1964 Magnitude 9.2 Great Alaska Earthquake. Kodiak Island lies just above part of the Alaska-Aleutian subduction zone fault that was responsible for the 1964 earthquake, and in 2019, 400 seismic instruments were deployed there for 25 days as part of the Alaska Amphibious Community Seismic Experiment (AACSE). Using machine learning and other techniques, Worthington and her team will re-analyze data collected by these instruments to identify numerous very small earthquakes, and pinpoint their locations. This will provide an unprecedented snapshot of how earthquakes are distributed along the subduction zone fault and its vicinity. Recordings of these and other earthquakes will be used together to generate images of the subsurface, including the subduction zone fault itself, and to estimate physical properties of the fault and its surroundings.Detailed characterization of seismicity and structure within shallow subduction forearc regions offers insight into plate interface properties along the seismogenic megathrust, and the distribution, location and kinematics of active and past deformation. This project characterizes seismicity and structure beneath Qikertaq (Kodiak Island) using existing data from a densely-spaced seismic node array deployed along a ~50 km transect as part of the Alaska Amphibious Community Seismic Experiment (AACSE) in May, 2019. The Kodiak node array is located within the southern asperity of the largest historical US earthquake, the Mw9.2 1964 Great Alaska event, which ruptured a 600-800 km section of the Alaskan subduction zone. The main goals of the project are to determine: 1) The distribution of upper plate vs lower plate vs interface seismicity during the interseismic cycle through increased microseismicity (M1.0) detection and relocation; 2) The nature of the plate interface in the shallow forearc of the southern 1964 earthquake rupture asperity, properties of which control earthquake rupture, propagation and initiation; 3) Upper plate velocity structure within the Kodiak region, providing constraints on material strength and elastic properties of the seaward portion of the forearc. Using both machine-learning and coalescence-based automatic detection algorithms, this project will produce an enhanced seismicity catalog for the duration of the 25-day deployment. The locations of existing and new detections will be refined to provide a snapshot of the distribution of seismicity during the interseismic interval following great megathrust earthquakes. Travel-times from active-source marine shots and local earthquakes (including newly detected events) will be used for raytracing and inversion to develop a high-resolution velocity model of the shallow forearc above the plate interface. This velocity model will, in turn, be used to further refine event locations. Energy from local earthquakes will be used for imaging the crust and plate interface below the node array using scattered waves and reflected phases for vertical reflection profiling and coda auto-correlation.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.