High Resolution Density Interface Inversion In Spherical Coordinates With Application To The South China Moho

Density interface inversion is an important means of gravity data interpretation,the fluctuation information of density interface obtained from gravity anomaly can be used for crust-mantle interface structure detection,regional structural research,oil and gas exploration.When the research area is a large area or even the whole world,the actual observation surface is a curved surface,in which the conventional Cartesian coordinate system density interface inversion method is no longer fit,and the spherical coordinate system density interface inversion method is needed.However,there is no method around the world to invert the density under spherical coordinate system for the observation surface on the ground.In this paper,an improved Gauss-Legendre integral algorithm based on Tesseroid model and gravity forward method is delivered to suit for high precision density interface forward of the observation surface on the ground in the spherical coordinates.Furthermore,we combine Cordell iterative inversion algorithm to make it suitable for the density interface inversion of the observation surface on the ground in spherical coordinates system,improving the computation efficiency.This method provides technical support for the high precision inversion of regional crustmantle interface structure.According to model test,when the observation interface is on the earth's surface,we verify that the effectiveness of the method in this paper,shows that the forward precision of this method is much higher than that of conventional Taylor series expansion algorithm and Gauss-Legendre integral algorithm in spherical coordinates and is also higher than that of conventional algorithm in Cartesian coordinate system.The forward result is the closest to the theoretical value.The inversion precision of the method in this paper is higher than that of the conventional frequency domain algorithm in Cartesian coordinate system and the Taylor series expansion algorithm in spherical coordinate system.The inversion result is the closest to the theoretical value.The computational efficiency of the proposed method is higher than that of the conventional Taylor series algorithm and Gauss-Legendre algorithm under the condition of equal computational accuracy.The Moho depth in the South China obtained by using the method in this paper was in the range of 30km-57 km.The overall fluctuation trend is deep in the West and shallow in the east.The depth of Moho surface in the west of South China fluctuates greatly,ranging from 35 km to 57 km.The depth of Moho surface in the east changes gently,and the depth ranges between 30 and 35 km,and is the shallowest in the coastal area.Along the Wuling mountain and Guizhou-Guangxi junction,there is an obvious NNE Moho gradient belt,which rises 5km from west to east.The undulation characteristics of that the Moho surface of South China continent is deep in the west and shallow in the east reflect the thinning of the eastern crust in a large area,which is speculated to be closely caused by the subduction of the Mesozoic ancient Pacific plate.