Brown, D., Alvarez-Marron, J., Camanni, G., Biete, C., Kuo-Chen, H., & Wu, Y.-M. (2022). Structure of the south-central Taiwan fold-and-thrust belt: Testing the viability of the model. Earth-Science Reviews, 231, 104094. https://doi.org/https://doi.org/10.1016/j.earscirev.2022.104094
A structural model is developed for the south-central Taiwan fold-and-thrust belt that shows consistency across an array of data types and marks an important step forward in the consideration of geological hazards and risks. Although there is general agreement about the regional scale geology of Taiwan, there are considerable differences in the structural interpretations of its western fold-and-thrust belt. In this paper, we bring together results of our previous studies and add new data and data analyses to develop a consistent 3D structural model for the south-central Taiwan fold-and-thrust belt that can explain key aspects of the seismicity, GPS, and topography data. We interpret the fold-and-thrust belt to be a west-verging, imbricate thrust system developed above a single basal thrust that is breached by ENE-striking faults that are inherited from the continental margin. These breaching faults are associated with ENE-striking transverse zones in the fold-and-thrust belt that are marked by changes in stratigraphy, structural style, strike and dip of the basal thrust, and uplift of the stratigraphic contacts. Along the eastern flank of the fold-and-thrust belt, metamorphic basement rocks are involved in the thrusting. Shortening estimates range from 15 km to >25 km. Much of the seismicity is taking place beneath the basal thrust, in the basement, along the flanks of basement highs and lows where strike-slip and transpressive fault types are common. There are systematic changes in GPS displacement vectors and strain rates across the transverse zones. Topography is higher in areas where basement is involved in the thrusting. The proposed structural model has depth and along-strike consistency, and can explain aspects of the distribution of seismicity, faults types, GPS displacement vectors and strain rates, and topography of the study area, and can therefore be considered a viable model.