UBC Theses and Dissertations
Shallow crustal structure beneath the Juan de Fuca ridge from 2[sup D] seismic refraction tomography White, Donald John
The formation of oceanic lithosphere along ocean ridges, and the role that crustal magma chambers play in the accretionary process, continue to be fundamental issues in plate tectonics. To address these issues, a multi-receiver airgun/ocean bottom seismograph refraction line, designed to allow definition of lateral velocity and attenuation variations within the shallow crust, was shot across the Endeavour segment of the Juan de Fuca Ridge near 48° N, 129° W. A tomographic inversion procedure has been developed to invert the first arrival travel times and amplitudes from this profile for 2[sup D] velocity and attenuation structure. The inversion method is suited to multi-source, multi-receiver refraction profiles where source/receiver spacings are denser than for conventional profiles. The travel time-velocity inversion scheme is based on an iterative solution of the linearized problem and allows for determination of continuous velocity variations as well as geometry of subhorizontal interfaces. The iterative procedure requires a good initial estimate of the velocity model. In each iteration, two-point ray tracing is performed to construct a linear system relating travel time residuals to velocity perturbations. A damped least-squares algorithm is used to solve this system for a velocity perturbation which is used to update the current velocity estimate. Once the final velocity structure of the model has been determined, amplitudes can be inverted directly for attenuation. Tests to ascertain resolution of the method reveal horizontal smearing of the solution due to ray geometry, drop-off in resolution with depth, and the effects of source-receiver geometry and velocity structure on resolution. Parameter weighting is important in removing streaking effects (caused by inhomogeneous ray coverage) from the solution. For the purposes of ray tracing, the model is parameterized in terms of constant gradient (velocity and attenuation) cells, which allow use of analytic expressions for kinematic and dynamic ray properties, attenuation and inversion quantities. This parameterization causes scatter in the amplitudes calculated using zero-order asymptotic ray theory, a problem which is remedied by smoothing the velocity models before amplitude calculation. Application of this 2[sup D] tomographic inversion scheme to first arrival travel times and amplitudes for the cross-ridge refraction line produced a 4-layer model for the shallow crust. Layer 1 is 250 — 650 m thick, with v₁ = 2.5 km/s and [Nabla, sub z]v₁ = 0.5 s⁻¹. Layer 2 is ~800 m thick, v₂ = 4.8 km/s and [Nabla, sub z]v₂ — 1.0 s⁻¹. Layer 1 and layer 2 likely represent the sequence of extrusives whereas layer 3 (~800 m thick, v₃=5.8 km/s, [Nabla, sub z]v₃=0.5 s-1) and layer 4 (v₄=6.3 km/s, [Nabla, sub z]v₄=0.3 s⁻¹) are associated with the dike complex and massive gabbro sequence, respectively. An abrupt velocity transition between layer 1 and layer 2 may be a metamorphic front within the pillow basalts. A low velocity-high attenuation anomaly (velocities decreased by < 0.4 km/s and Q ~20-100), which is interpreted as a zone of increased fracture porosity and/or permeability associated with axial hydrothermal circulation, exists beneath the ridge in layer 2 and upper layer 3. Smaller low velocity-attenuative zones in layer 2, located 8 km to either side of the ridge may be loci of off-axis hydrothermal circulation. No evidence is found for the existence of a crustal magma chamber in the depth range of 1.5 — 3.0 km below the seafloor. Tests indicate that a 1 X 1 km zone of partial melt represents the minimum dimension of such a feature that would be clearly detected by this refraction experiment. These results suggest that Endeavour Ridge may be experiencing a period of diminished magma supply with the magma chamber reduced or eliminated by hydrothermal circulation. Asymmetry of the velocity anomalies observed in layer 3 and layer 4 suggest that crustal temperatures are elevated by 125 — 200° C beneath the ridge and to the east relative to temperatures west of the ridge, indicating that a deep crustal or upper mantle melting anomaly may exist east of the ridge.
Item Citations and Data