Direct Inversion of S‐P Differential Arrival Times for Ratio in SE Asia

Abstract

Southeast Asia lies within one of the most complex tectonic settings on Earth and exhibits a range of features, including strongly curved subduction zones, arc‐continent collision, and slab break‐off, which are not well understood. To help gain insight into mantle structure and processes beneath this region, we perform an inversion for variations in Vp, Vs, and VpVs structure using arrival time information from the ISC‐EHB catalog. The oceanic lithosphere subducting beneath Java is imaged as a positive dVp and negative d(Vp/Vs) anomaly. At 200 km depth, the forearc mantle beneath Sumatra and Java is revealed by positive dVp and d(Vp/Vs) anomalies which cease at Sumba island, where negative d(Vp/Vs) anomalies mark the presence of cold Australian lithosphere (down to 200–250 km depth) which is colliding with Sundaland. These negative d(Vp/Vs) anomalies depict a ∼WE trending structure that appears to correspond with the underthrusting of Australian continental crust. One notable salient has a location and shape which appears to coincide with those of ancient terranes or a Gondwana‐related microcontinent reconstructed by paleogeographic studies and may have been entrained in the subduction process. The velocity and d(Vp/Vs) patterns beneath the Banda Arc support the existence of a single curved subducting slab associated with rollback. The extreme extensional strike‐slip setting in Seram produces the highest positive d(Vp/Vs) anomalies in the model which may be due to one or more of widespread serpentinization, high concentrations of intraslab fluid‐filled faulting, and mantle upwelling.

Tectonic Fabric in the Banda Arc‐Australian Continent Collisional Zone Imaged by Teleseismic Receiver Functions

“…This array (coded as the https://doi.org/10.7914/SN/YS_2014) includes the first broadband seismometer deployment in Timor‐Leste and on the islands of Lomblen, Savu, Rote, and Alor in eastern Indonesia (Figure 1b; Miller et al., 2016). Using this data set, new imaging of crustal and mantle structure suggests continent subduction/underthrust is the dominant lithospheric structure beneath the region (Miller et al., 2021; Porritt et al., 2018; Zhang & Miller, 2021), in line with conclusions of other regional seismic tomography studies (e.g., Fichtner et al., 2010; Harris et al., 2020; Zenonos et al., 2019, 2020). The pronounced along‐strike structural variations at shallow lithospheric depths may either result from the diachronous (progressive) collision as a result of the oblique convergence (Porritt et al., 2018; Zhang & Miller, 2021) or from inherent structural heterogeneities of the incoming and colliding Australian (lower) plate (Miller et al., 2021), or both.…”

Seismic Imaging of the Subducted Australian Continental Margin Beneath Timor and the Banda Arc Collision Zone

“…Most early studies focused on large-scale tectonic evolution by imaging the slab morphology along the entire Sunda-Banda Arc system with global datasets (e.g., Hafkenscheid et al, 2001;Spakman & Hall, 2010;Widiyantoro & van der Hilst, 1996, 1997Widiyantoro et al, 2011). Some more regionally focused studies addressed possible hypotheses that dominate the collisional evolution in Banda such as deep subduction of the Australian continent, the presence of ancient continental terranes and slab tears, but they only resolve mantle structure at depth greater than ∼60-100 km (Fichtner et al, 2010;Harris et al, 2020;Zenonos et al, 2019Zenonos et al, , 2020. Most recently, Supendi et al (2020) developed a new P wave velocity (Vp) model at depths above 120 km using local earthquakes for which they propose that a seismically fast forearc sliver of continental origin is underthrusted beneath Flores and accounts for the arc volcanism contamination.…”

Detailed 3D Seismic Velocity Structure of the Prague, Oklahoma Fault Zone and the Implications for Induced Seismicity

“…We selected a subset of 1508 well-recorded earthquakes that were recorded by at least 5 stations from the template-matching derived catalog by McMahon et al (2017), and manually picked P and S wave arrivals. We only used earthquakes with both P and S phase picks at five or more stations to facilitate a better comparison between Vp and Vs, and to ensure the reliability of the Vp/Vs (Zenonos et al, 2020). To further filter out low-quality data, we evaluated the distribution of the P and S wave travel times versus distance and fitted the data using least squares linear regression (Figure S1).…”

“…We adopt the Fast Marching tomography algorithm (FMTOMO, de Kool et al, 2006;Rawlinson et al, 2006, Text S2 in Supporting Information S1) to invert for both the Vp and Vs structures. For the determination of the Vp/Vs model, instead of simply taking the ratio of Vp and Vs models, we apply a modified version of the FMTOMO code to directly invert for Vp/Vs using S-P times following Zenonos et al (2020), which has been demonstrated to produce more robust Vp/Vs model. The region of interest in this study is approximately 40 km (EW direction) by 30 km (NS direction), and extends from the ground surface to 16 km depth.…”