Questions: Seismic Anisotropy and Shear Wave Splitting

Question 1

A seismic station records SKS shear-wave splitting with a fast polarization direction of N10°E and a delay time δt = 1.8 seconds. What is the most geophysically meaningful interpretation?

• A. The crust beneath the station has fluid-filled cracks aligned N10°E, producing the 1.8-second delay
• B. The upper mantle beneath the station has a fabric with a N10°E fast axis, likely from olivine crystals aligned by northward mantle flow; the 1.8-second delay implies strong or thick anisotropy
• C. The seismic wave was deflected by a fault striking N10°E, and 1.8 seconds is the travel-time anomaly
• D. The lower mantle has a NNE-oriented fabric; SKS phases sample the entire mantle

Answer: (B)

SKS phases convert from S to P in the liquid outer core and back to S at the core-mantle boundary, arriving at the station with a known initial polarization. Any splitting they acquire occurs in the solid mantle and crust. The dominant contribution is typically the upper mantle (100-200 km thick anisotropic layer), where olivine lattice-preferred orientation (LPO) from mantle flow creates the anisotropy. A 1.8-second delay is too large to be primarily crustal (crustal splitting is typically 0.1-0.3s). Fluid-filled cracks (option A) are a crustal mechanism inconsistent with a 1.8s delay. The lower mantle (option D) is largely isotropic or has minimal contribution to SKS splitting.

Question 2

Two seismologists compare splitting measurements. Station A shows δt = 0.5s; Station B shows δt = 1.5s. Seismologist X concludes the mantle under B is three times more anisotropic than under A. Is this conclusion justified?

• A. Yes — δt is directly proportional to anisotropy strength, so a 3× larger delay means 3× stronger anisotropy
• B. No — δt depends on both anisotropy strength and path length through the anisotropic region; Station B may have weaker anisotropy over a longer path, or stronger anisotropy over a similar path
• C. No — δt measures only the depth of the anisotropic layer, not its strength
• D. Yes — but only if the two stations use the same SKS phase, otherwise the comparison is invalid

Answer: (B)

The delay time δt is the product of the percentage anisotropy and the path length through the anisotropic medium: δt ∝ (anisotropy strength) × (path length). A larger δt could reflect stronger anisotropy over the same path length, the same anisotropy strength over a longer path, or some combination. Without independent constraints on path length (e.g., from receiver function analysis of lithospheric thickness), you cannot disentangle the two contributions. This is an important limitation of shear-wave splitting — it measures the integrated effect along the entire path, not the local anisotropy at any specific depth.

Question 3

Shear-wave splitting occurs because an S-wave entering an anisotropic medium splits into two components that travel at different velocities, producing a time delay between their arrivals at the surface.

• T. True
• F. False

Answer: True

This is the core phenomenon. In an isotropic medium, S-wave velocity is the same regardless of polarization direction. In an anisotropic medium (e.g., olivine-rich mantle with LPO, or crust with aligned fluid-filled cracks), velocity depends on polarization direction relative to the fabric. The incoming S-wave projects onto the fast and slow polarization eigenvectors of the medium, and these two components then propagate at different speeds. By the time they emerge, a delay δt has accumulated. A seismogram shows two S-wave pulses instead of one, with orthogonal polarizations. This is directly analogous to optical birefringence in calcite crystals.

Question 4

A large delay time (δt > 2 seconds) in shear-wave splitting measurements generally indicates very strong seismic anisotropy in the mantle.

• T. True
• F. False

Answer: False

δt = (fractional anisotropy) × (path length through anisotropic region) / (average S-wave velocity). A large δt can result from moderate anisotropy spread over a thick layer (e.g., 200 km of 2% anisotropy) just as easily as from strong anisotropy over a thin layer. In subduction zones, for example, large δt values sometimes reflect a thick anisotropic wedge of mantle material rather than exceptionally strong crystal alignment. Conversely, short path lengths through strongly anisotropic material (e.g., in a thin but intensely deformed shear zone) can produce small δt despite high local anisotropy. Disentangling strength from thickness requires combining splitting measurements with other constraints.

Question 5

Explain the physical process by which a single S-wave becomes two distinct arrivals after passing through an anisotropic region of the mantle.

Model answer: When an S-wave enters an anisotropic medium, its particle motion (polarization) projects onto two orthogonal eigenvectors of the medium: the fast polarization direction and the slow polarization direction. These two components propagate independently at different wave speeds — the fast component moves at higher velocity than the slow component. As both components travel through the anisotropic region, they accumulate a time difference proportional to the anisotropy strength and path length. Upon exiting, what was a single coherent pulse arrives as two separate pulses offset in time, polarized at 90° to each other. The effect is identical to optical birefringence in crystals like calcite.

The analogy to optics is exact because both phenomena arise from the same physical principle: in a medium with directional symmetry breaking, waves with different polarizations couple to different effective elastic (or optical) moduli and therefore travel at different speeds. Geophysicists exploit this by measuring the fast direction φ (which tells them about mantle flow or stress orientation) and the delay time δt (which quantifies the integrated effect of anisotropy along the path).