Shallow structure of oceanic crust in the western North Atlantic from seismic waveform inversion and modeling

Abstract

Seismic reflection data from Mesozoic oceanic crust in the western North Atlantic, in the vicinity of the Blake Spur Fracture Zone, have imaged a number of short, subhorizontal events in the upper 1 km of the crystalline crust. Wide‐angle expanding spread profile (ESP) data from the same region indicate the presence of first‐order velocity discontinuities in this depth interval. Using ESP data reprocessed to enhance precritical reflection events, we investigate the relationship between the precritical and postcritical reflections with three different approaches: amplitude analysis, forward modeling, and waveform inversion. The precritical reflections have high apparent reflection coefficients (∼ 0.1) and may be divided into two types: those which can be traced out to critical ranges (5–6 km), representing velocity steps, and those which fade out at ranges of 3–4 km, representing shorter‐wavelength velocity variations. Synthetic seismogram modeling indicates that the velocity steps occur within a 50 m depth interval, while the shorter‐wavelength features may correspond to high‐ or low‐velocity zones with a thickness of 20 m (10% of the dominant seismic wavelength) or less. We investigate the velocity structure in detail by full waveform inversion of a selected portion of ESP data, transformed into the intercept time‐slowness domain. The inversion method minimizes the sample‐by‐sample misfit between the data and reflectivity synthetics in a least squares sense. A conjugate gradient algorithm was used, starting from a velocity structure derived from forward modeling of wide‐angle ESP data. The inversion tightly constrains short‐wavelength components of the velocity structure, and indicates that intracrustal reflections result both from major velocity steps and from zones of alternating high and low velocity. Synthetic seismograms generated from sonic log data from Deep Sea Drilling Project hole 418A show reflections of similar amplitude and character to those seen in the ESP data. These reflections probably originate from rather subtle features of the velocity‐depth profile, related primarily to changes in porosity in the pillow lava sequence. A deeper reflector, 800 ms below oceanic basement in the Blake Spur Fracture Zone, may correspond to the top of a zone of serpentinized upper mantle.

Evidence for asymmetric nonvolcanic rifting and slow incipient oceanic accretion from seismic reflection data on the Newfoundland margin

“…Most basement along the seaward part of line 2 is very reflective in its upper 0.5 km and relatively transparent beneath 0.5 km. Strong reflectivity in upper oceanic crust is often ascribed to variable porosity and hydrothermal alteration within basalt flows and sills [ Purdy , 1987; Wilkens et al , 1991; Minshull and Singh , 1993]. However, at ODP Site 1277, only 30 m of basalt and mixed debris containing clasts of basalt, gabbro and peridotite were recovered above a serpentinized peridotite basement [ Shipboard Scientific Party , 2004b].…”

Velocity structure near IODP Hole U1309D, Atlantis Massif, from waveform inversion of streamer data and borehole measurements

“…The inference of low attenuation for the deeper gabbros of the Central Dome is consistent with the general inference of high Q , often >400, for seismic layer 3 [e.g., Wilcock et al ., ; Minshull and Singh , ], even though it is at odds with Hole 735B measurements [ Swift and Stephen , ]. The presence of higher attenuation within the low‐velocity surface layer while compatible with the line A4 data is not required as there are other potential sources of the model inconsistencies.…”

Origin of dipping structures in fast-spreading oceanic lower crust offshore Alaska imaged by multichannel seismic data

“…These sections of crust were accreted at a half rate of ∼20 mm/yr, which is slow but sufficient for the lower crust to be ductile (Chen and Morgan, 1990;Phipps Morgan and Chen, 1993) allowing shear zones to develop near the ridge axis. These sections also have unusually large crustal thicknesses (∼7-8 km) (Morris et al, 1993), anomalously smooth basement topography (Minshull, 1999) and brighter than normal Moho reflections. These observations are similar to those reported in our study and elsewhere where reflections are observed, which could support a common origin for the dipping events and imply that spreading rate, alone, does not control where they form.…”