Probing depth origin of gravity anomalies in Taiwan through 3-D coherent velocity model

  • The velocity structures are most reliable in the depth range from 5 to 35 km
  • The Puli gravity low arose from the mismatch of shallow and deep gravity effects
  • The calculated gravity of coherent model shows better fitting with observed gravity
Abstract

We constructed a 3-D coherent velocity model by extracting the common features from recent tomographic models and utilized the forward gravity modeling and wavelength filtering method to explore the depth origin of the gravity anomalies, such as the long-puzzling Puli Bouguer gravity low in west central Taiwan. The similarity analysis among the tomographic models shows the velocity structure most reliable in the depth range from 5 to 35 km. Using empirical velocity-to-density relationships, we show that the calculated Bouguer gravity anomalies from the 3-D coherent velocity model could well reproduce the gravity observations in general. Through filtering process, we separated the Bouguer gravity anomalies into shallow (0 - 17 km) and deep gravity effects (17 - 60 km) for structural exploration. The results show that the shallow gravity effect mainly reflects the surface geological features with low gravity values for unconsolidated sediments in plain and foothill areas to the west and high values for exhumed metamorphic rocks in the mountainous areas to the east. The deep gravity effect reveals the lowest gravity anomalies located in central Taiwan beneath the western Central Range. The Puli Bouguer gravity low is likely a coincident feature arising from the mismatch of the shallow and deep gravity effects due to the N-S trending undulation of crustal thickness beneath the western plain/foothill. This crustal thickness undulation may be related to the rifting activity of Taihsi and Tainan basins in the past. Better discriminating the gravity effect from shallow and deep crust could help better understand the architecture and tectonic process of the Taiwan orogenic belt.

Read 281 times