A fast finite-element software for gravity anomaly calculation in complex geologic regions Yongen Cai Department of Geophysics Peking University, Beijing, 100871 Chi-yuen Wang Department of Earth and Planetary Science University of California, - PowerPoint PPT Presentation

1 / 15
About This Presentation
Title:

A fast finite-element software for gravity anomaly calculation in complex geologic regions Yongen Cai Department of Geophysics Peking University, Beijing, 100871 Chi-yuen Wang Department of Earth and Planetary Science University of California,

Description:

... gravity anomaly of a density model may be computationally ... Density model. The density distribution can be obtained from the velocity from seismic ... – PowerPoint PPT presentation

Number of Views:193
Avg rating:3.0/5.0

less

Transcript and Presenter's Notes

Title: A fast finite-element software for gravity anomaly calculation in complex geologic regions Yongen Cai Department of Geophysics Peking University, Beijing, 100871 Chi-yuen Wang Department of Earth and Planetary Science University of California,


1
A fast finite-element software for gravity
anomaly calculation in complex geologic
regionsYongen Cai Department of Geophysics
Peking University, Beijing, 100871 Chi-yuen
WangDepartment of Earth and Planetary
ScienceUniversity of California, Berkeley, CA
94720
2
Introduction
  • For geologically complex regions, forward
    computation of the gravity anomaly of a density
    model may be computationally demanding and the
    bottle-neck in gravity inversion.
  • We present a fast finite-element software for
    solving this problem.

3
P(x,y,z)
dv
4
GBOX(R.J.Blakely,1995)
P(0,0,0)
x
R
y
z
5
Boundary value problem
g (x, y, z) -
6
Boundary condition
(Jeffreys, 1962)
7
FEM formulation

8
Accuracy verification
Density model for verifying
(c 0.001 kg/m4 )
9
GBOX(average density)
FFEM(distributed density)
10
(No Transcript)
11
Application to Taiwan


Source elements 76,500 Source nodes
83,448 Calculated gravity points
GBOX4636 points only at ground surface FEM
285488 at all nodal points Computer PC with 2.3
GHz CPUs
Figure 7(b) (GBOX)
12
Comparison between FFEM and GBOX

mGal

FFEM used cpu time 280 s
GBOX used cpu time 6780 s
Figure 7(b) (GBOX)
13
Application to Sirrea Nevada (Cai, Zhang and
Wang, 2006)
Calculated Bouguer anomalies by FFEM
Calculated Bouguer anomalies by classical method
14
Conclusion
  • A software FFEM is provided which is more
    accurate and much faster than the classical
    integration method, if density in the material
    body is highly heterogeneous.
  • The computational efficiency for the FFEM method
    is more pronounced in regions with greater
    heterogeneities.

15
Density model
  • The density distribution can be obtained from the
    velocity from seismic tomograph.
Write a Comment
User Comments (0)
About PowerShow.com