Dynamic processes in the Earth's interior are the dominant driving forces behind the continuous deformation-related reworking of its surface. The characterization of deformation caused by past tectonic events near the Earth's surface as well as mapping of ongoing dynamic-driven processes deep inside the Earth are therefore major objectives to understand the dynamics of our planet. Seismic anisotropy, the direction-dependence of seismic wave speed, is directly related to deformation processes and can be "felt" by passing seismic waves. Although seismic anisotropy is a well-known phenomenon, the individual contributions from different depth ranges are still debated. However, recordings available from
dense and large-aperture seismic station networks, provide the opportunity for resolving both, small-scale variations relatively close to the surface as well as so far unknown structures at greater depth.
In 2012, an international seismological field experiment, called ScanArray, was initiated. The combination of 72 temporary broadband stations with long-running national permanent stations and arrays resulted in a recording ... mehrnetwork consisting of 266 seismic stations in total that were distributed across the Fennoscandian peninsula in northern Europe. Fennoscandia opens the opportunity to study the (past) geodynamical evolution of crustal and upper mantle structures far away from currently active plate tectonics.
The main goal of this study is to characterize the anisotropic structure beneath the Fennoscandian peninsula as well as in the Earth's lowermost mantle based on a uniformly processed data set provided by the ScanArray network. For this purpose single-event shear wave splitting analysis was performed using core-refracted shear waves (SKS, SKKS, PKS) of around 3000 globally distributed teleseismic earthquakes (1998-2017). In order to improve the data coverage at a recording station, a new plugin (StackSplit) for a widely applied analysis software (SplitLab) is introduced allowing efficient and flexible handling of multi-event splitting measurements.
Based on the massive seismic data set, this study provides a comprehensive characterization of the distinct lateral and backazimuthal variations of the shear wave splitting pattern at individual stations and across the ScanArray network. These variations partly correlate well with different tectonic regimes related to past large-scale lithospheric deformation due to ancient collision events. Detailed forward modeling allowed to explore different anisotropic structural geometries including anisotropy with a dipping axis of symmetry. Although, the majority of the shear wave splitting observations can be explained with high reliability, for a small number of recording stations non-unique anisotropy models were found which fit the observed data equally well.
Furthermore, this study sheds light on lowermost mantle anisotropy located in the so-called D" layer just atop the core-mantle boundary in ~ 2900 km depth. This unexpected discovery is based on observations of distinct splitting discrepancies between teleseismic SKS and SKKS phases for the same source-receiver configuration. The lowermost mantle anisotropy can be associated with two large-scale seismic velocity anomalies beneath the North Atlantic and northwestern Siberia. Even though the exact geometry and mechanism of the anisotropic fabrics in D" cannot be fully constrained by ScanArray recordings alone, these new observations provide important and much-needed boundary conditions for improved future geodynamic mantle modeling.