Joint Seismic and Magnetotelluric Imaging of the San Andreas Fault Zone

地球物理方法是探测断裂带深部结构的重要手段。 不同的地球物理技术对不同的物理性质敏感。 通常，多种技术应用于同一研究区域，每种技术分别获得结构结果，并将所得结构进行比较以进行解释。 联合反演提供了一种替代方法，其中使用单个结构模型来尝试拟合多个数据集，最常见的是将不同的物理属性（例如，地震速度和密度）联系起来的经验关系。 我们的工作将涉及地震波走时和大地电磁剖面数据的联合反演，以确定横跨加州圣安德烈亚斯断层的两个段的地震波速度和电阻率结构。 一个剖面穿过圣安德烈亚斯断层深度观测站（SAFOD），SAFOD钻孔提供了现场信息，另一个剖面是沿着靠近霍利斯特的圣安德烈亚斯蠕动段。 这项工作将提高我们对活动转换断层的结构和复杂的继承框架，他们是侵位的理解。 我们将开发的约束反演方法应该对断层带的研究具有广泛的适用性。 地下结构，地震活动性和水文约束之间的空间关系将导致更好地了解控制断层行为的主要因素（岩性，流体，断层几何形状），我们将应用一种新的联合反演方法，其中包括一个约束，鼓励模型属性的空间变化发生在模型中的相同位置。 具体地说，计算物理性质的梯度场（在我们的情况下，地震速度和电阻率），并评估模型空间中每个单元处的两个梯度的叉积。 模型因具有非零叉积而受到“惩罚”，这具有产生如下模型的效果：其中一个参数的空间变化伴随着另一个参数的变化，该另一个参数共享相同的几何形状，但可以具有相同或相反的符号（平行或反平行梯度），或者没有变化（零梯度）。

Geophysical methods are key tools for exploring the deep structure of fault zones. Different geophysical techniques are sensitive to different physical properties. Commonly, multiple techniques are applied to the same study area with structural results for each technique obtained separately, and the resulting structures compared for interpretation. Joint inversions provide an alternative in which a single structural model is used to try to fit multiple datasets, most commonly with empirical relationships linking the different physical properties (for example, seismic velocity and density). Our work will involve the joint inversion of seismic wave travel time and magnetotelluric profiling data to determine seismic wave velocity and electrical resistivity structure across two segments of the San Andreas Fault in California. One profile passes through the San Andreas Fault Observatory at Depth (SAFOD) where in-situ information is available from the SAFOD borehole, and the other is along the creeping section of the San Andreas near Hollister. This work will improve our understanding of the structure of active transform faults and the complex inherited framework in which they are emplaced. The constrained inversion methods we will develop should have broad applicability to the study of fault zones. The spatial relationship between subsurface structure, seismicity, and hydrologic constraints will result in a better understanding of the primary factors (lithology, fluids, fault geometry) controlling fault behavior.We will apply a novel variation of the joint inversion approach that involves a constraint where spatial variations in model properties are encouraged to occur at the same places in the model. Specifically, the gradient fields of the physical properties (in our case seismic velocity and electrical resistivity) are computed and the cross product of the two gradients at each cell in the model space is evaluated. Models are "penalized" for having a non-zero cross product, which has the effect of producing models for which a spatial change in one parameter is accompanied by a change in the other that shares the same geometry but can have the same or opposite sign (parallel or anti-parallel gradient), or has no change (zero gradient).