Constraints on radial anisotropy in the central Pacific upper mantle - - PowerPoint PPT Presentation

Constraints on radial anisotropy in the central Pacific upper mantle from the NoMelt OBS array

Joshua B. Russell1, James B. Gaherty1, Peiying (Patty) Lin2, Molly Zebker1

1Lamont-Doherty Earth Observatory, Columbia University 2Taiwan Ocean Research Institute, Kaohsiung, Taiwan

• S14A-05

600x400 km footprint – 16 broad-band OBS – Short-period OBS – Magnetotelluric array

NoMelt Experiment

160˚W 160˚W 150˚W 150˚W 140˚W 140˚W 0˚ 0˚ 5˚N 5˚N 10˚N 10˚N 15˚N 15˚N 20˚N 20˚N 25˚N 25˚N

70 Ma 8 M a 60 Ma

Situated on relatively pristine oceanic lithosphere (~70 Ma) One year of continuous data collected in 2012

• Azimuthal and radial anisotropy

constrain flow patterns within the mantle

• Previous observations of

anisotropy in the lithosphere beneath ocean basins are consistent with horizontally aligned olivine fabric associated with seafloor spreading

• Inconsistencies remain between

recent regional and global models

• f radial anisotropy in the

lithosphere

Motivation

Nishimura & Forsyth, 1989 VSH > VSV

NoMelt provides new constraints on Pacific mantle anisotropy, measured at a local scale.

Radial Anisotropy Azimuthal Anisotropy

NoMelt azimuthal anisotropy

Lin et al., Nature (2016)

Azimuthal variation of Rayleigh wave phase velocities oriented parallel to the fossil spreading direction (~78º) in the lithosphere

• Strong anisotropy in lid
• riented parallel to FSD
• Weak anisotropy in the low

velocity zone (100-150 km)

• Stronger anisotropy below

the LVZ associated with asthenospheric flow How does strength of radial anisotropy beneath the central Pacific compare with azimuthal anisotropy?

VSV G Amplitude Fast Direction

PeiYing (Patty) Lin: [S21C-05] Tues. 9-9:15 Room 307

3.5 4 4.5 5 50 100 150 200 250 300

VSV (km/s) Depth (km)

Beghein 2014 NF 52−110 Ma S362WMANI+M Pa5 NoMelt SV

0.9 1 1.1 50 100 150 200 250 300

ξ = (VSH/VSV)2

Motivation: Radial anisotropy

Central Pacific Models

• Radial anisotropy ξ is also important

for constraining upper-mantle circulation and evolution of the lithosphere-asthenosphere system

• Radial anisotropy may reflect

processes controlling the G discontinuity

• Discrepancies exist between current

models of radial anisotropy in the central Pacific upper-mantle

• Requires constraints from both Love

and Rayleigh waves

Method

0.1 0.15 0.2 0.25 −0.02 −0.01 0.01 0.02 SNR : 2.3127 Distance : 540.6619km Frequency (Hz) 4 5 6 7 8 9 10 3.8 4 4.2 4.4 4.6 B01−B24 Period (s) Phase Velocity (km/s)

Data Synthetic Starting Model Fit

240 cross-correlation functions Ambient noise provides constraints from both Rayleigh and Love wave fields Phase velocities derived from waveform fitting of ambient-noise cross spectra [Menke & Jin, BSSA 2015]

cross spectra:

1st overtone Rayleigh waves (4-10 s)

−200 −100 100 200 100 200 300 400 500 600

lag time (s) Distance (km)

Vertical 4-10 s

water column 1st overtone

10

1

10

2

−140 −120 −100 −80

Period (s) Power All stations Z

mean

All stations Z

Cross spectral power

−100 100 −4 −2 2 4

10 s c/c (%)

−100 100 −4 −2 2 4

8.5714 s

−100 100 −4 −2 2 4

7.5 s

−100 100 −4 −2 2 4

6.6667 s

−100 100 −4 −2 2 4

6 s

−100 100 −4 −2 2 4

5.4545 s Azimuth (degrees)

−100 100 −4 −2 2 4

5 s

−100 100 −4 −2 2 4

4.6154 s −2 2 4

δc/c (%)

−100 100

Azimuth (degrees) 4 s

10 s 8.6 s 7.5 s 6.7 s 6 s 5.5 s 5 s 4.6 s

• Sensitivity of 1st overtone Rayleigh comparable

to fundamental mode Love wave

• Strong 2θ azimuthal signal

– Rayleigh fast direction parallel to fossil spreading (78º)

2θ Anisotropy

Fundamental mode Love waves (4-10 s)

−200 −100 100 200 100 200 300 400 500 600

lag time (s) Distance (km) T : All non−repeated pairs, filtered at 4 −10(s)

Transverse 4-10 s

10

1

10

2

−140 −120 −100 −80

Period (s) Power All stations T

mean

All stations T

Cross spectral power

−100 100 −4 −2 2 4

10 s c/c (%)

−100 100 −4 −2 2 4

8.5714 s

−100 100 −4 −2 2 4

7.5 s

−100 100 −4 −2 2 4

6.6667 s

−100 100 −4 −2 2 4

6 s

−100 100 −4 −2 2 4

5.4545 s Azimuth (degrees)

−100 100 −4 −2 2 4

5 s

−100 100 −4 −2 2 4

4.6154 s −2 2 4

δc/c (%)

10 s 8.6 s 7.5 s 6.7 s 6 s 5.5 s 5 s 4.6 s

• Clear 4θ azimuthal signal

– Love slow direction parallel to fossil spreading (78º) – Consistent with predictions of olivine fabric

4θ Anisotropy

−100 100

Azimuth (degrees) 4 s

Measurements

Strong azimuthal anisotropy suggests significant mantle sensitivity

1 2 x 10

−7

5 10 15 20 25 Depth (km) SV 1 2 x 10

−7

5 10 15 20 25 Depth (km) SH

Rayleigh S1 Love T0

Fréchet Kernels (5-7.5 s)

• Fund. mode Love

1st overtone Rayleigh

10

1

2.5 3 3.5 4 4.5 Phase Velocity (km/s) 10

1

1 2 3 4 5 Peak−to−peak amp (%) 10

1

50 100 150 Fast Direction Period (s)

Rayleigh S1 Rayleigh S0 Love T0 NoMelt

FSD ~78º

• Fund. mode Love

1st overtone Rayleigh

• Fund. mode Rayleigh

NoMelt SV

Isotropic phase veloc. Azimuthal anisotropy (%) Azimuth (º)

4θ Love 2θ Rayleigh 4θ slow 2θ fast

Inversion

Starting model

VP

0-35km: NoMelt refraction model accounting for ~8% P-azimuthal anisotropy in the mantle

(D. Lizarralde personal communications)

VS

Sediments: Seafloor compliance

[Ruan et al., JGR 2014]

Crust: VP/VS = ~1.85 [Brocher, BSSA 2005] Mantle: NoMelt SV [Lin et al., Nature 2016]

Solving for horizontal and vertical VP & VS VP

2 4 6 4 5 6 7 8 9 10 11 12 13

Depth (km) VP (km/s)

PV

2 4 4 5 6 7 8 9 10 11 12 13

Depth (km) VS (km/s)

SV

NoMelt crust1.0 Starting model

Crustal Model

NoMelt SV

VS

Inversion results

VSH > VSV required in the mantle lithosphere and cannot be ruled out in the crust

1 1.1 1.2 4 6 8 10 12 14 16 18 20 ξ = (VSH/VSV)2 Anisotropy 3 4 5 4 6 8 10 12 14 16 18 20 Depth (km) V (km/s) VSV 5 6 7 3.6 3.8 4 4.2 Rayleigh Phase velocity (km/s) 5 6 7 −1 1 Rayleigh Residual δc % Periods (s) 5 6 7 4 4.1 4.2 4.3 4.4 Love 5 6 7 −1 1 Love Residual Periods (s)

starting Observed

Starting Model

Fit to data Anisotropy VSV

2σ error

~ 2% ~ 5%

Summary & Interpretation

Both azimuthal and radial anisotropy required in the lithospheric mantle

Mantle Ø Radial anisotropy: VSH > VSV (~ 3-7%) Ø Clear 2θ and 4θ azimuthal anisotropy

• Consistent with petrologic models of olivine

with orthorhombic or hexagonal symmetry

• Horizontal preferred alignment of olivine a-

axis associated with fossil spreading Crust Ø VSH > VSV (0-5%)

• Horizontal crustal fabric?

– Layering processes? Cracks? Fluids?

FSD VSH > VSV 3 - 7% VSH > VSV 0 - 5 % ? ¡

4θ slow 2θ fast

To Ridge ? ¡ a-axis

Additional inversions: Good fits

1 1.1 1.2 4 6 8 10 12 14 16 18 20 Depth (km) ξ = (VSH/VSV)2 Anisotropy 3 4 5 4 6 8 10 12 14 16 18 20 V (km/s) VSV 5 6 7 8 3.6 3.8 4 4.2 Rayleigh Phase velocity (km/s) 5 6 7 8 −1 1 Rayleigh Residual δc % Periods (s) 5 6 7 8 4 4.1 4.2 4.3 4.4 Love 5 6 7 8 −1 1 Love Residual Periods (s)

1 1.1 1.2 4 6 8 10 12 14 16 18 20 Depth (km) ξ = (VSH/VSV)2 Anisotropy 3 4 5 4 6 8 10 12 14 16 18 20 V (km/s) VSV 5 6 7 8 3.6 3.8 4 4.2 Rayleigh Phase velocity (km/s) 5 6 7 8 −1 1 Rayleigh Residual δc % Periods (s) 5 6 7 8 4 4.1 4.2 4.3 4.4 Love 5 6 7 8 −1 1 Love Residual Periods (s)

starting iso mantle iso crust + mantle fixed crust

Additional inversions: Poor fits

Kernel Nonlinearity: Love waves (20-100 s)

900 1000 1100 1200 1300 1400 1500 1600 51 52 53 54 55 56 57 201202021334 20.5km Seconds Distance (degrees)

Love Waves (20-100 s)

Kernel Nonlinearity: 4-30 s

3 4 5 50 100 150 200 250 300 Velocity Model VSH (km/s) Depth (km) 2 4 x 10

−8

50 100 150 200 250 300

5 s 5.4545 s 6 s 6.6667 s 7.5 s 8.5714 s 10 s 12 s 12.8571 s 13.8462 s 15 s 16.3636 s 18 s 20 s 22.5 s 25.7143 s 30 s

2 4 x 10

−8

50 100 150 200 250 300

5 S 5.4545 S 6 S 6.6667 S 7.5 S 8.5714 S 10 S 12 S 12.8571 S 13.8462 S 15 S 16.3636 S 18 S 20 S 22.5 S 25.7143 S 30 S

Rayleigh 1st overtone Love Fundamental Mode