Real

dat

aSy

nthe

tics

10 s

PKKPab PKJKP

pPKKPab pPKJKP??

0% 2%AnisotropyPKiKPPKJKPPKKPab

!40 !35 !30 !25 !20 !15 !102.35

2.4

2.45

2.5

2.55

2.6

2.65

2.7

2.75

2.8

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

x 10!11

!40 !35 !30 !25 !20 !15 !10!10

!8

!6

!4

!2

0

2

4

6

8

10

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

x 10!11

106 108 110 112 114 116 118 120

-4

0

2

4

6

8

-2

10

PKiKP pPKiKP

1100 1105 1110 1115 1120 1125 1130

1.65

1.7

1.75

1.8

1.85

1.9

1.95

2

2.05

0

0.5

1

1.5

2

x 10!9

Absolute time (sec)

Ab

so

lute

slo

wn

ess (

se

c/d

eg

)

PKKPbc pPKKPbc

PKKPab pPKKPab

1755 1760 1765 1770 1775 1780 1785!5

!4.8

!4.6

!4.4

!4.2

!4

!3.8

!3.6

!3.4

!3.2

!3

0

0.5

1

1.5

2

2.5

3

x 10!9

106 108 110 112 114 116 118 120

-4

0

2

4

6

8

-2

10

Epicentral distance (degrees)

Time r

elativ

e to

pred

icted

PKK

Pab

(sec)

Time r

elativ

e to

pick

ed P

KKPa

b (se

c)

Epicentral distance (degrees)

Absolute time (sec)

Ab

so

lute

slo

wn

ess (

se

c/d

eg

)

PKJKP pPKJKP PKJKP pPKJKP

a b

c d

e f

Slown

ess r

elativ

e to

PKKP

ab (s

ec)

Azim

uth

devia

tion

from

GC

path

(deg

rees

)

Time relative to PKKPab (sec) Time relative to PKKPab (sec)

pPKKPabpPKKPab

PKKPabPKKPab

? ?

SEISMO-47/KLARGE-SCALE ARRAY USE FOR PHASE IDENTIFICATION AND WAVEFORM

DETERMINATIONG. Helffrich, J. Wookey

Earth Sciences, University of Bristol (UK)

ABSTRACTThe early experiments of seismic array use for discrimination purposes led, in the mid-1970s, to the view that large scale arrays were undesirable due to the loss of wavefield coherence at scales larger than a few tens of kilometers. Consequently, small aperture arrays were preferred for detection, location and discrimination, and are an element of the present IMS.

A variety of large-scale national seismic monitoring networks for seismic hazard reduction and basic research are a new, and a readily available resource usable in some seismic array roles. They comprise hundreds of stations with apertures up to a thousand km. The aperture and the station density enhance the utility of simple methods such as delay-and-sum to suppress noise and enhance waveform features. However, the large station spread and independent station time bases virtually guarantee that crustal structure and instrument timing variations will make absolute time measurements problematic. Relative time and slowness measurements, after adjustment for receiver statics, are quite feasible given this constraint. While losing an absolute time base impairs regional array use for event location, it is still beneficial for wavenumber- and waveform-based event study.

We show two examples of how regional seismic arrays benefit earth structure studies: in subduction zones and in the inner core of the earth. The regional network in the United

Kingdom (~100 stations; slowness resolution 8 × 10-4 s/km) provided data with which the seismic discontinuities at 410 and 660 km were traced through the Izu-Bonin subduction zone in the western Pacific. Array processing of nearby earthquakes illuminating the discontinuities revealed their locations by reflections and conversions of direct waves at the discontinuities. Hi-Net in

Japan (~800 stations; slowness resolution 5 × 10-4 s/km) provided data with which high frequency records of the inner core shear wave, PKJKP, were obtained by array processing methods. Compared to PKKP and PKiKP, also recorded by Hi-Net, the PKJKP waveform has extra pulses in it that suggest shear-wave splitting in the inner core.

These studies show the potential that regional array use has for discrimination work. Greater waveform fidelity enhances detectability of depth phases associated with natural events. Relative slowness estimates between direct and later arriving phases provide azimuth and range information that constrain source locations. The free availability of data also means that the analyses may be replicated by any national agency. The factors are all beneficial as an element of a global seismic event monitoring facility

References:[1]Wookey & Helffrich (2008) Nature, 454, 873-877.[2]Rawlinson & Kennett (2004) GJI, 157, 332-340.[3]Schimmel & Paulssen (1997) GJI, 130, 497–505.

Conclusions

Large-scale arrays have good azimuthal and slowness resolution due to their spatial extent, but must be corrected for local structure delays.

Can serve as valuable information sources for event discrimination.

Large-scale vs small-scale arrays

Hinet UK array

2000

km

1100

km

EKA array(small-scale)

50 k

m

Beam Functions (1.2 Hz)Hinet UK array EKA array

Central peak ~ 5 × 10-4 s/km Central peak ~ 8 × 10-4 s/km Central peak ~ 8 × 10-2 s/km

Application: PKJKP detection

Reference phase: PKKP and PKIKP by absolute slowness (a, b) Alignment on PKKP arrival by adaptive stacking to remove static delays² (c, d)

Slant-stack relative to reference arrival to detect PKJKP (e) and azimuthal stack to verify approach azimuth (f)

Processing¹ Waveforms

Aligned waveforms (top) from phase-weighted stacking³ of ~700 traces showing trace and envelope (gray), and reflectivity synthetics (bottom) of PKKP and PKJKP. Extra arrival suggests shear-wave splitting of PKJKP.