Surface waves

1
Surface waves

• Guy Masters
• Cider 2008

2
(No Transcript)
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
Synthetic seismograms
Distance (degrees)
Time (seconds)
8
(No Transcript)
9
(No Transcript)
10
Phase or group velocity?
11
(No Transcript)
12
(No Transcript)
13
Transverse isotropy

• Rayleigh waves sensitive to Vsv and Love waves
  are dominantly sensitive to Vsh
• Differences in dispersion lead to the development
  of a TI 1d model (PREM) -- now have 3d TI models

14
(No Transcript)
15
(No Transcript)
16
Measuring phase

• (stay away from caustics!)

17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
Measuring group arrival times
22
Undispersed 50 s Rayleigh waves
Note large number of cycle shifts possible
23
(No Transcript)
24
After computation of envelope functions
25
50 sec Rayleigh waves -- a typical example
26
After clustering.
27
A more complicated example
28
After clustering.
29
Phase and group velocity maps
30
(No Transcript)
31
Rayleigh phase at 100 sec
32
Rayleigh phase at 50 sec
33
(No Transcript)
34
(No Transcript)
35
Rayleigh group at 100 seconds
36
Rayleigh group at 50 seconds
37
Rayleigh phase at 50 sec
38
(No Transcript)
39
100s
71, 239000
50s
93, 215000
33s
98, 125000
data
residual
40
Hit count, 2 degree blocks
41
Resolution of 1000km structure is global
42
Resolution of finer structure is more variable
43
Error map for Rayleigh waves at 50 sec
44
Finite frequency kernels
45
(No Transcript)
46
(No Transcript)
47
Should really use 3d kernels
48
(No Transcript)
49
(No Transcript)
50
Finite frequency kernels

• Expect effect to be greatest at longest periods
• Difficult to make direct comparison with great
  circle result since different damping is implied
• Comparison with analyses of SEM synthetics are
  ambiguous but 2D kernels do not seem to be
  effective -- need 3D kernels?
• What do you find?

51
What about ff kernels for group velocity?
52
Oops! times depend on phase AND group velocity!!!
53
(No Transcript)
54
Inversion for crustal thickness

• Correct maps for the CRUST2 model
• Invert corrected maps with additional point
  constraints from receiver function studies for
  perturbation to CRUST2 moho depth
• Include only maps for periods shorter than 40
  seconds to avoid mantle signal.

55
Depth to moho in CRUST2.0
56
Observationally constrained locations
57
27.5
Crust-corrected
Crust2
Observed
58
32.5
Crust corrected
Crust2
Observed
59
Depth to moho (km)
New
Crust2
60
(No Transcript)
61
(No Transcript)
62
Topo
Group vel
63
Topo
Group vel
64
Topo
Group vel.