Conclusions

1

• Conclusions
• Challenge
• Seismic waves are affected by variations in
  temperature, pressure, composition, mineralogy,
  structure (layering, scales and distribution of
  multiphase materials, texture, fabric, grain
  size, etc.) and water content.

2

• Conclusions
• Challenge
• Seismic waves are affected by variations in
  temperature, pressure, composition, mineralogy,
  structure (layering, scales and distribution of
  multiphase materials, texture, fabric, grain
  size, etc.) and water content.
• 2) There are differences depending upon whether
  the water is in the form of hydrous melts,
  hydrous phases, or incorporated into the crystal
  structure of nominally anhydrous minerals.

3
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase
4
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase (Amounts vary GREATLY depending upon
compositions and values of temperature and
pressure)
5
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase (Water increase in ringwoodite of 1
would lower S velocities by about 5.4 and P
velocities by about 1.5 Jacobsen et al., 2004
Jacobsen and Smyth, 2006 Karato, 2006 ----gt
Water increases P-to-S velocity ratio)
6
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase (Temperature affects on attenuation
are greater at higher temperatures and lower
pressures)
7
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase 410 Height elevate depress 660
Height depress elevate TZ Width increase dec
rease
8
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase 410 Height elevate depress 660
Height depress elevate TZ Width increase dec
rease (Water increase could elevate 410 by 10-30
km, depress the 660 by up to 4 km, and so
increase TZ Width Smyth and Frost, 2002
Hirschmann et al., 2005 Higo et al., 2001)
9
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase 410 Height elevate depress 660
Height depress elevate TZ Width increase dec
rease (Temperature increase of 400ºC could
depress 410 by 30-50 km, elevate the 660 by 7-40
km, so greatly reduce TZ Width Litasov et al.,
2006)
10
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase 410 Height elevate depress 660
Height depress elevate TZ Width increase dec
rease 410 Thickness broaden sharpen 660
Thickness broaden sharpen
11
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase 410 Height elevate depress 660
Height depress elevate TZ Width increase dec
rease 410 Thickness broaden sharpen 660
Thickness broaden sharpen (Wet mantle --gt
broaden 410 by up to 40 km, broaden 660 by up to
8 km Smyth and Frost, 2002 Hirschmann et al.,
2005 Higo et al., 2001)
12
Solution There are ways to distinguish between
them Ex/ Increase in Water vs.
Temperature S Velocity DECREASE DECREASE P
Velocity decrease DECREASE Attenuation INCREA
SE increase 410 Height elevate depress 660
Height depress elevate TZ Width increase dec
rease 410 Thickness broaden sharpen 660
Thickness broaden sharpen (Hot mantle --gt
sharpen 410 and 660 by by around 5 km (Helffrich
and Bina, 1994)
13
Most Seismically-Observed Mantle Water is in
Subduction Zones (not surprisingly!)
14
Petro-thermo-mechanical models can now predict
what kinds of features we would expect to see
seismically
Gerya et al. 2006
15
Flow and Temperature
Gerya et al. 2006
16
P Velocity
Gerya et al. 2006
17
S Velocity Presence of cold, wet plumes predict
gt20 Poisson ratio variations, as opposed to 2
variations due to only temperatures
Gerya et al. 2006
18
Inability of thermal models to explain seismic
parameters is seen in subduction zone tomography
observations
Wiens et al. 2008
19
Thermal Model Predictions Tonga
Observations
Wiens et al. 2008
20
Thermal Model Predictions Tonga
Observations
Wiens et al. 2008
21
Velocity Tomography away from subduction zones
also show features that may be associated with
water.
van der Lee et al. 2008
22
van der Lee et al. 2008
23
1 2 3
4 5
van der Lee et al. 2008
24
Reciever Functions of P-to-S converted phases
find LZVs in forearc mantles interpreted as
serpentinization from slab-expelled water
Tibi et al. 2008
25
LVZ S-velocities as low as 3.6 km/s suggest
serpentinization of 30-50, corresponding to
chemically bound water contents of 4-6 wt
Tibi et al. 2008
26
Kawakatsu and Watada 2008
27
Kawakatsu and Watada 2008
28
Percentages show S-velocity reductions relative
to slab velocities parentheses show suggested
water content in wt)
Kawakatsu and Watada 2008
29
The presence of water can determine the magnitude
and orientation of seismic anisotropy in olivine
30
Seismic shear-wave splitting results from
Nakajima and Hasegawa 2004
31
Geodynamic models suggest the possibility of
water just above the 410 discontinuity.
Leahy and Bercovici 2007
32
Leahy and Bercovici 2007
33
Mantle reverberations show discontinuity depth
and impedance
Courier and Revenaugh 2007
34
Discontinuity depths..
Courier and Revenaugh 2007
35
Reflection Coefficients Results show a LVZ
above the 410 with reduced 410 impedance
attributed to partial melt from volatile-induced
melting.
Courier and Revenaugh 2007
36
Seismic Arrays (in this case RISTRA) can identify
layer velocities from traveltime moveouts.
Gao et al. 2007
37
Triplication patterns reveal vertical velocity
structures
Gao et al. 2007
38
LVZ above 410 is interpreted as partial melting
due to water released from the Transition Zone
Water on Top of Transition Zone?
Gao et al. 2007
39
Water causes increased Seismic Attenuation
Stein and Wysession 2003
40
Large High-Attenuation region 700-1200 km deep (Q
lt 100 !!!) with only slightly negative velocities
Lawrence and Wysession 2006
41
Depth 1000 km
A Vertical Cross-Section through Earths mantle
at 1000 km depth shows the high-attenuation
region above circum-Pacific subducted lithosphere
Lawrence and Wysession 2006
42
Shieh et al. 1998
43
Lawrence and Wysession 2006