1. Université Joseph Fourier Grenoble, France
The nature and promise of broadband surface-wave measurements from the random wavefield
Nikolai Shapiro
Michael Ritzwoller
Michel Campillo
Laurent Stehly
Université Joseph Fourier Grenoble, France
University of Colorado at Boulder

2. Main goal: to optimize the surface-wave imaging in context of USArray
earthquakes with M>4.5 occurred during 18 months (01/01/ /01/2002)
Main problem: low regional seismicity for most of USA

3. Diffuse field vs. ballistic waves
localized lateral sensitivity
samples all directions
source independent
may allow many short-period
measurements
Alternative solution:
making measurement from
random wavefield
(ambient seismic noise)
May improve resolution
traditional approach:
using
teleseismic surface waves
source
extended lateral sensitivity
sample only certain directions
source dependent
difficult to make short-period
measurements
Consequence: limited resolution

4. Extracting Green functions from the random wavefield by field-to-filed correlation: theoretical background
seismic noise is excited by randomly distributed ambient sources (oceanic microseisms and atmospheric loads)
modal representation of the random field:
- eigenfunctions
- eigenfrequencies
- modal excitations, uncorrelated random variables:
- spectral energy density
cross-correlation between points x and y :
differs only by an amplitude factor F() from an actual Green function between x and y

5. Cross-correlations from ambient seismic noise: ANMO - CCM
(from Shapiro and Campillo, GRL, 2004)
cross-correlations from 30 days of continuous vertical component records (2002/01/ /02/08)
prediction from global group velocity maps of Ritzwoller et al. (2002)
frequency-time analysis of the broadband cross-correlation

6. Cross-correlations from ambient seismic noise at US stations
(from Shapiro and Campillo, GRL, 2004)
frequency-time analysis of broadband cross-correlations computed from 30 days of continuous vertical component records

7. Cross-correlation from ambient seismic noise in North-Western Pacific
(from Shapiro and Campillo, GRL, 2004)
broadband cross-correlation computed from 30 days of continuous vertical component records

8. Cross-correlation from ambient seismic noise in North-Western Pacific
(from Shapiro and Campillo, GRL, 2004)
broadband cross-correlation computed from 30 days of continuous vertical component records

9. Cross-correlations from ambient seismic noise in California
(from Shapiro and Campillo, GRL, 2004)
cross-correlations of vertical component continuous records (1996/02/ /03/10) Hz
3 km/s - Rayleigh wave

10. correlations computed over four different three-week periods
PHL - MLAC 290 km
band-
passed s
repetitive measurements provide uncertainty estimations
band-
passed s

11. correlations computed over four different three-week periods
PHL - MLAC 290 km
band-
passed s
band-
passed s
repetitive measurements provide uncertainty estimations

12. Conclusions
Cross-correlations computed from the ambient seismic noise can provide new surface-wave dispersion measurements that have numerous advantages relative to traditional measurements made from teleseismic waves:
Measurements possible for every pair of stations
No source related errors
Localized sensitivity zones
Measurements can be extended to shorter periods
Repetitive measurements provide uncertainty estimations
Those measurements may be particularly useful in context of dense arrays of seismometers, such as USArray