The region of northern Borneo in South East Asia sits within a post-subduction setting formed by the recent termination of two sequential but opposed subduction systems. In this study we use seismic data from a recent temporary array deployment to image the crustal velocity structure beneath northern Borneo using a two-stage Bayesian trans-dimensional tomography scheme, in which period dependent phase velocity maps are first generated, and then used to build a 3-D shear wave model through a series of 1-D inversions. In the second stage, we also apply an Artificial Neural Network to solve the 1D inverse problem, which results in a smoother 3-D model compared to the TransD approach without sacrificing data fit. Our shear wave velocity model reveals a complex crustal structure. Under the Crocker Range, a heterogeneous velocity structure likely represents remnants of early Miocene subduction, including underthrust continental crust from subsequent continent-continent collision. In the east we observe high velocities that are interpreted to be igneous rocks in the crust generated by melting due to mid Miocene Celebes Sea subduction and later decompression melting as well as a low velocity zone that could represent underthrust sediment or duplexes from Celebes Sea subduction. A low velocity zone in the lower crust is present in a region of apparent crustal thinning. Our preferred explanation for this anomaly is remnant thermal upwelling within a failed rift that represents the on-shore continuation of the extension of the Sulu Sea, most likely caused by rollback of the Celebes Sea slab.

Understanding deep crustal structure can provide us with insights into tectonic processes and how they affect the geological record. The deep crustal structure can be studied using a variety of seismological techniques such as receiver function analysis, and surface and body wave tomography. Using models of crustal structure derived from these methods, it is possible to delineate tectonic boundaries and regions that have been affected by similar processes. However, often velocity models are grouped in a somewhat subjective manner, potentially meaning that some geological insight may be missed. Cluster analysis, based on unsupervised machine learning, can be used to more objectively group together similar velocity profiles and, thus, put additional constraints on the deep crustal structure. In this study, we apply hierarchical agglomerative clustering to the shear wave velocity profiles obtained by Gilligan et. al. (2016) from the joint inversion of receiver functions and surface wave dispersion data at 59 sites surrounding Hudson Bay. This location provides an ideal natural laboratory to study Precambrian tectonic processes, including the 1.8Ga Trans-Hudson Orogen. We use Ward linkage to define the distance between clusters, as it gives the most physically realistic results, and after testing the number of clusters from 2 to 10, we find there are 5 main stable clusters of velocity models. We then compare our results with different inversion parameters, clustering schemes (K-means and GMM), as well as results obtained for profiles from receiver functions in different azimuths and found that, overall, the clustering results are consistent. The clusters that form correlate well with the surface geology, crustal thickness, regional tectonics, and previous geophysical studies concentrated on specific regions. The profiles in the Archean domains (Rae, Hearne, and Superior) were clearly distinguished from the profiles in the Proterozoic domains (Southern Baffin Island and Ungava Peninsula). Further, the crust of Melville Peninsula is found to be in the same cluster as the crust of the western coast of Ungava Peninsula, suggesting a similar crustal structure. Our study shows the promising use of unsupervised machine learning in interpreting deep crustal structures to gain new geological insights.

The post-subduction tectonic evolution of northern Borneo, which experienced two sequential subduction episodes of opposite polarity in the Neogene, is still widely debated with first-order questions such as whether the region has been in a state of compression or extension remaining unresolved. We use waveform data recorded from a dense seismic network in northern Borneo to investigate crustal thickness variations through the application of Virtual Deep Seismic Sounding (VDSS). The new results reveal an extensive area of thin crust in central and southeastern Sabah that appears to extend northeast into the Sulu Sea, where rifting initiated. We also compute local earthquake focal mechanisms, which suggest that extension is ongoing, though now dominated by orogen collapse in the NW. Together these results point to the pervasiveness of regional extension tectonics over the last 15-20 Myr and its role in the post-subduction cycle of plate tectonics in SE Asia.