Transcript Seismic tomography

Seismic tomography
Tomography attempts to determine anomalous
structures within the Earth as revealed by
deviations from “average” seismic
properties at depth.
Average is usually determined by one of the
simple “radial” structural models of the
Earth. PREM (Anderson and Dziewonski,
1981) is the most commonly used reference
Earth model.
Thanks and/or apologies to Barbara Romanowicz, UC Berketey whose slides have
been used liberally in this presentation.
PREM
Anderson and Dziewonski (1981) determined a
spherical shell model of the Earth that was most
consistent with the observed travel times from
seismic sources to seismic stations that had
been accumulated in the previous 80 years of
seismology.
Note that the model is “layered” and laterally
averaged over the whole Earth within layers and
so no lateral variations in structure are
modelled.
P-wave velocity
S-wave velocity
Preliminary Reference
Earth Model
IASPEI91
Kennett et al. (1991) obtained another 1-D Earth
model that is often used in reference.
http://ds.iris.edu/spud/earthmodel
What are we missing?
What are we seeking?
Crust and Mantle Structures
What we would like to see...
Body and surface waves
Seismic waves integrate the seismic velocities
experienced point-by-point along their paths:

T() = x/v(x) dx
along path
Seismic wave paths
A tomographic slice
Over a diametrical slice through the Earth, we look
for regions that are anomalously slow or fast
compared to the PREM average for that depth within
Basic concept
Each of these 3
paths is the same
distance.
S1-A: no variation
S2-B: encounters a
fast region
S1-C: encounters a
slow region
Earthquakes and seismographs
Earthquakes and
seismographs are not
uniformly or even with
uniform randomness
distributed over the
world. We only have
biased data sets.
Sample bias -- 1
Density of paths transiting a region of
between 660 and 870km
Sample bias -- 2
Density of paths transiting a region of
between 2670km and CMB
Vasco and Johnson, 1998
Example results
Van der Hilst, et al., 1998
The Scripps SB4L18 model
Seismic tomography is very
fashionable, now, and most major
seismic laboratories are
presenting mantle velocity
anomaly models. This one, the
Scripps SB4L18 model, presents
results as depth layers over the
Earth. What is plotted is slow Swave and fast S-wave velocities.
Laske, Masters and Reif, AGU 2001
Berkeley vs. CalTech
Laske, Masters and Reif, AGU 2001
3D DNA09
Regional tomographic model of
structure beneath the Western
US.
Note the P-wave and S-wave velocity
anomalies, especially below Yellowstone. Is
the Yellowstone “plume” so restricted to the
upper mantle.
http://ds.iris.edu/spud/earthmodel
PNW10-S
Cascadia subduction zone
Figure 1. Transect S-N along the
subduction interface.
Figure 2. Transect following the spine
of the Cascades mountains chain and
heat flow profile.
http://ds.iris.edu/spud/earthmodel
So, what do we really know?
Layered Structure
Seismology finds global radial variations in physical
properties α and β at many nearly fixed depths within the
Earth.
~5-70 km
410 km
660 km
~2700 km ±
2890 km
5150 km
Rheology from Tectonics
Isostatic recovery from glacial loading and lateral tectonic
plate motions suggest a different layering, especially of the
outer regions.
0 – 200 km (~elastic)
to ~400 km (~low viscosity fluid)
to ~2900 km (~higher viscosity)
to ~5100 km (very low viscosity)
to 6375 km (~soft elastic)
Rheological hack:
e = (t/2η + 1/2μ)σ
shear
shear
Seismic provisos
Seismic technologies are not as powerful as we
might like you to believe... geologists and mantle
petrologists want to believe in seismic tomography
What Lies Deep in the Mantle Below?
“Teleseismic tomography, although
sensitive to horizontal variations in
seismic wave speed, has virtually no
ability to determine depth variations.”
...but still
“Were it not for seismology, we would
be almost totally ignorant of the
structure of the deep interior of Earth.”