ABSTRACT Shear‐wave splitting analysis is commonly used to infer deep anisotropic structure. For simple cases, obtained delay times and fast‐axis orientations are averaged from reliable results to define anisotropy beneath… Click to show full abstract
ABSTRACT Shear‐wave splitting analysis is commonly used to infer deep anisotropic structure. For simple cases, obtained delay times and fast‐axis orientations are averaged from reliable results to define anisotropy beneath recording seismic stations. However, splitting parameters show systematic variations with back azimuth in the presence of complex anisotropy and cannot be represented by average time delay and fast‐axis orientation. Previous researchers had identified anisotropic complexities at different tectonic settings and applied various approaches to model them. Most commonly, such complexities are modeled using multiple anisotropic layers with priori constraints from geologic data. In this study, a graphical user interface called M‐Split is developed to easily process and model multilayered anisotropy with capabilities to properly address the inherited nonuniqueness. M‐Split program runs user‐defined grid searches through the model parameter space for two‐layered anisotropy using formulation of Silver and Savage (1994) and creates sensitivity contour plots to locate the local maxima and analyze all possible models with parameter trade‐offs. To minimize model ambiguity and identify the robust model parameters, various misfit calculation procedures are also developed and embedded in M‐Split and can be used depending on the quality of the observations and their back azimuthal coverage. M‐Split is an open‐source program and can be extended by users for additional capabilities or for other applications.
               
Click one of the above tabs to view related content.