Abstract. We present results from a combined multichannel seismic reflection (MCS) and wideangle onshore/offshore seismic experiment conducted in 1996 across the southeast Greenland continental margin. A new seismic tomographic method is developed to jointly invert refraction and reflection travel times for a two-dimensional velocity structure. We employ a hybrid raytracing scheme based on the graph method and the local ray-bending refinement to efficiently obtain an accurate forward solution, and we employ smoothing and optional damping constraints to regularize an iterative inversion. We invert 2318 Pg and 2078 PmP travel times to construct a compressional velocity model for the 350-km-long transect, and a long-wavelength structure with strong lateral heterogeneity is recovered, including (1) -30-km-thick, undeformed continental crust with a velocity of 6.0 to 7.0 km/s near the landward end, (2) 30-to 15-km-thick igneous crust within a 150-km-wide continent-ocean transition zone, and (3) 15-to 9-km-thick oceanic crust toward the seaward end. The thickness of the igneous upper crust characterized by a highvelocity gradient also varies from 6 km within the transition zone to -3 km seaward. The bottom half of the lower crust generally has a velocity higher than 7.0 km/s, reaching a maximum of 7.2 to 7.5 km/s at the Moho. A nonlinear Monte Carlo uncertainty analysis is performed to estimate the a posteriori model variance, showing that most velocity and depth nodes are well determined with one standard deviation of 0.05-0.10 km/s and 0.25-1.5 km, respectively. Despite significant variation in crustal thickness, the mean velocity of the igneous crust, which serves as a proxy for the bulk crustal composition, is surprisingly constant (-7.0 km/s) along the transect. On the basis of a mantle melting model incorporating the effect of active mantle upwelling, this velocitythickness relationship is used to constrain the mantle melting process during the breakup of Greenland and Europe. Our result is consistent with a nearly constant mantle potential temperature of 1270-1340øC throughout the rifting but with a rapid transition in the style of mantle upwelling, from vigorous active upwelling during the initial rifting phase to passive upwelling in the later phase.

No abstract

No abstract

Reprocessed multichannel seismic profiles from the 9°N segment of the East Pacific Rise reveal prominent shallow subbasement events. These events are identified as wide‐angle reflections from the base of seismic layer 2A, based upon modeling of expanding spread profile data and velocity functions. The layer 2A reflections typically increase from 0.15 s after the seafloor reflection at the rise axis to 0.3–0.45 s within 1–2 km of the axis, corresponding to an increase in layer thickness of 200–600 m. No further systematic increase in layer thickness is observed, although lateral variability of the order of a few hundred meters in thickness is observed at greater offsets from the rise axis. However, the intermittent character of the imaged layer 2A reflection is attributed to focusing and defocusing of energy by the seafloor bathymetry rather than necessarily to intrinsic lateral variability at the base of the layer. The base of layer 2A is interpreted as corresponding to the transition between the extrusive section, pillow basalts and sheet flows, and a sheeted dike complex. The rapid thickening of the layer near the rise axis is attributed to successive lava flows burying the initially shallow top of the sheeted dike complex as the layer passes through the neovolcanic zone. Lateral variability of layer 2A can significantly affect the imaging of the underlying axial magma chambers as average velocities within layer 2A are approximately half that of layer 2B. For an along‐axis profile, apparent along‐axis variability in the depth of the axial magma chamber is traced to variability in the thickness of layer 2A caused by wandering of the profile relative to axis. Within the resolution of the data, the time delay of the magma chamber reflection relative to the base of layer 2A is constant.