Three-dimensional crustal model of Shanxi graben from 3D joint inversion of ambient noise surface wave and Bouguer gravity anomalies

Guo Zhen,Chen Yong-Shun,Yin Wei-Wei
DOI: https://doi.org/10.6038/cjg20150312
2015-01-01
Abstract:The Main objective of this study is to obtain a model that can explain the surface wave data and Bouguer gravity anomaly simultaneously and also to alleviate the non-uniqueness of surface wave inversion. We applied a novel non-linear joint inversion technique using both surface wave and gravity anomalies to obtain high resolution 3D shear-wave velocity crustal model of Central North China Craton including Shanxi graben. The crustal model helps us to understand the crust and mantle dynamics and the evolution history of the Shanxi rift. We bandpass filtered WGM2012 global Bouguer gravity anomaly data to remove the data with wavelength less than 50 km and greater than 200 km. We then projected the gravity data and surface wave data from ambient noise tomography to a Cartesian coordinate system with grid spacing 50 km x 50 km. The joint inversion was performed in this coordinate system. The inversion volume is much larger than the target region. This parameterization strategy was chosen to minimize the likelihood edge effects from gravity modelling. We used Crust 2.0 global model as our initial model in the inversion. An approximate non-linear P velocity and density relationship is derived from combing Birch's law and Nafe-Drake's curve. For each iteration of inversion, the sensitivity kernel of surface wave and gravity data were recalculated and updated until the inversion converged. We did several experiments to obtain an optimized weighting parameter to balance the influences of both datasets. A surface-wave-only inversion was also performed to compare the results of joint inversion. Both shear wave velocity models from joint inversion and surface-wave-only inversion can fit surface wave data well though only the joint inversion provides an acceptable fit to the gravity data. Gravity misfits decrease from >30 mGal to similar to 3.4 mGal after joint inversion. The velocity anomalies from both inversions all show a remarkable correlation with surface geology and upper crustal structures beneath Shanxi rift. The Luliang mountain region and Datong volcanic region all show relatively high shear wave velocity in the upper crust. Meanwhile, the basins in the southern rift show slow velocities. The southern part of the Shanxi rift is slower than the northern part in the upper crust. The Cenozoic extension started from the northern rift and then passed on to the southern rift. However, in the mid-and-lower crust beneath the Datong volcanic region, the surface-wave-only inversion resolves much slower shear wave velocity compared to the joint inversion model. The most pronounced difference occurs in the mid-and-lower crust beneath the Luliang mountain region. The joint inversion reveals a slow velocity anomaly with amplitude about 2%, while the surface-wave-only inversion model shows a fast velocity. The Datong volcanic region, north part of the Shanxi rift, might have experienced mantle upwelling since the Cenozoic time. The hot and buoyant mantle heated the upper most mantle and the crust, which caused the large-scale mid and lower crust slow velocity and negative gravity anomaly in the Datong volcanic region. The slow velocity anomaly beneath the Luliang mountain region is also related to the mantle upwelling beneath the Datong volcanic region. The Cenozoic mantle upwelling might control the formation and evolution of the Shanxi rift.
What problem does this paper attempt to address?