Slab Morphology Beneath Northern Sumatra Revealed by Regional and Teleseismic Traveltime Tomography

Abstract

An eikonal equation-based tomography method is used to invert the high-quality regional and teleseismic traveltime data recorded by 26 broadband seismic stations in northern Sumatra, following which we obtain the P and S wave velocity structures of the crust and mantle down to a depth of 800 km. The results of both P and S wave tomography clearly show that the Indo-Australian oceanic plate continuously penetrates downward beneath northern Sumatra and the maximum penetration depth varies approximately from 400 km at the northern tip of Sumatra to about 800 km around the southern boundary of our study area. Significant slab folding or bending reported in the literature as the main feature of the subducted slab beneath northern Sumatra is not found in our results. Instead, our tomographic images demonstrate only a less curved slab that mimics the shape of the Sunda Trench and volcanic arc and generally extends over depths from 120 to 450 km. P wave tomography shows broad and pronounced low-velocity anomalies beneath the island of Sumatra in the lower crust and uppermost mantle. Our model also reveals a slab tear approximately at 120-km depth, which has been documented in previous studies and considered to be related to the eruption of the Toba supervolcano.

The Geometry of the Subducted Slab Beneath Sumatra Revealed by Regional and Teleseismic Traveltime Tomography

“…One plausible interpretation is that the slab-related high-velocity anomalies beneath northern Sumatra are lumped together with the high-velocity anomalies beneath Peninsular Malaysia, which are revealed in both Richard et al (2007) and Pesicek et al (2008); as a result, the imaged slab exhibits significant curvature. The high-velocity anomalies in the upper mantle beneath Peninsular Malaysia have been repeatedly confirmed by recent regional seismic tomography studies (Huang et al, 2015;Liu et al, 2019), which provided the necessary facts to support this interpretation. Figure 12 depicts two subvertical slab tears.…”

“…Thus, our tomographic results do not support slab detachment beneath this area. Slab detachment was also not supported by a recent tomographic study of northern Sumatra using both regional and teleseismic traveltime data (Liu et al., 2019).…”

“…(2018), and the shallow structure was well defined by the inversion of regional seismic data, and therefore the imaging resolution is much higher in Liu et al. (2019) than that in Liu et al. (2018).…”

“…Therefore, it is appropriate to set the number of iterations in the interval [5, 20] and choose the prescribed value in the interval (0, 0.2] for the inversion of velocity. We further transform the origin time of the earthquake and its spatial position into relative perturbations for earthquake relocation (Liu et al., 2019). The same intervals for the iteration number and prescribed value are appropriate for the relocation of earthquakes.…”

Slab Models Beneath Central Myanmar Revealed by a Joint Inversion of Regional and Teleseismic Traveltime Data

“…In their forward waveform modeling, they also obtained a P-wave velocity model by scaling the shear-wave velocity through an empirical relation (Equation 9 in Brocher, 2005). Considering that the locations of regional earthquakes are only roughly estimated using the ak135 model (Figures 3a-3b), the initial three dimensional P-wave velocity model is first used to relocate all the regional earthquakes (S. Liu et al, 2019). After that, we use a cascade scheme to perform the joint inversion: regional traveltime tomography is first conducted to build an accurate shallow velocity model, and then teleseismic traveltime tomography is performed to refine the deep part of the initial model ( Figures S5 and S6; the discussion about different inversion strategies can be found in Text S3 in the supporting information).…”