Description: Oceanic crust is formed at mid-ocean spreading centres by a combination of magmatic, tectonic and hydrothermal processes. The crust formed by magmatic process consists of an upper crust generally composed of basaltic dikes and lava flows and a lower crust presumed to mainly contain homogeneous gabbro whereas that by tectonic process can be very heterogeneous and may even contain mantle rocks. Although the formation and evolution of the upper crust are well known from geophysical and drilling results, those for the lower crust remain a matter of debate. Using a full waveform inversion method applied to wide-angle seismic data, here we report the presence of layering in the lower oceanic crust formed at the slow spreading Mid-Atlantic Ridge, ~7-12 Ma in age, revealing that the lower crust is formed mainly by in situ cooling and crystallisation of melt sills at different depths by the injection of magma from the mantle. These layers are 400-600 m thick with alternate high and low velocities, with ± 100-200 m/s velocity variation, and cover over a million-year old crust, suggesting that the crustal accretion by melt sill intrusions beneath the ridge axis is a stable process. We also find that the upper crust is ~400 m thinner than that from conventional travel-time analysis. Taken together, these discoveries suggest that the magmatism plays more important roles in the crustal accretion process at slow spreading ridges than previously realised, and that in-situ lower crustal accretion is the main process for the formation of lower oceanic crust.

Description: Oceanic crust forms at mid-ocean spreading centres through a combination of magmatic and tectonic processes, with the magmatic processes creating two distinct layers: the upper and the lower crust. While the upper crust is known to form from lava flows and basaltic dykes based on geophysical and drilling results, the formation of the gabbroic lower crust is still debated. Here we perform a full waveform inversion of wide-angle seismic data from relatively young (7–12-Myr-old) crust formed at the slow-spreading Mid-Atlantic Ridge. The seismic velocity model reveals alternating, 400–500 m thick, high- and low-velocity layers with ±200 m s−1 velocity variations, below ~2 km from the oceanic basement. The uppermost low-velocity layer is consistent with hydrothermal alteration, defining the base of extensive hydrothermal circulation near the ridge axis. The underlying layering supports that the lower crust is formed through the intrusion of melt as sills at different depths, which cool and crystallize in situ. The layering extends up to 5–15 km distance along the seismic profile, covering 300,000–800,000 years, suggesting that this form of lower crustal accretion is a stable process.

Description: Seismic full waveform inversion (FWI) is a powerful method for estimating quantitative subsurface physical parameters from seismic data. As the full waveform inversion is a non-linear problem, the linearized approach updates model iteratively from an initial model, which can get trapped in local minima. In the presence of a high velocity contrast, such as at Moho, the reflection coefficient and recorded waveforms from wide-aperture seismic acquisition are extremely non-linear around critical angles. The problem at the Moho is further complicated by the interference of lower crustal (Pg) and upper mantle (Pn) turning ray arrivals with the critically reflected Moho arrivals (PmP). In order to determine velocity structure near Moho, a non-linear method should be used. We propose to solve this strong non-linear FWI problem at Moho using a trans-dimensional Markov chain Monte Carlo (MCMC) method, where the earth model between lower crust and upper mantle is idealy parameterized with a 1-D assumption using a variable number of velocity interfaces. Different from common MCMC methods that require determining the number of unknown as a fixed prior before inversion, trans-dimensional MCMC allows the flexibility for an automatic estimation of both the model complexity (e.g. the number of velocity interfaces) and the velocity-depth structure from the data. We first test the algorithm on synthetic data using four representative Moho models and then apply to an ocean bottom seismometer (OBS) data from the Mid-Atlantic Ocean. A 2-D finite-difference solution of an acoustic wave equation is used for data simulation at each iteration of MCMC search, for taking into account the lateral heterogeneities in the upper crust, which is constrained from travel time tomography and is kept unchanged during inversion; the 1-D model parameterization near Moho enables an efficient search of the trans-dimensional model space. Inversion results indicate that, with very little prior and the wide-aperture seismograms, the trans-dimensional FWI method is able to infer the posterior distribution of both the number of velocity interfaces and the velocity-depth model for a strong nonlinear problem, making the inversion a complete data-driven process. The distribution of interface matches the velocity discontinuities. We find that the Moho in the study area is a transition zone of 0.7 km, or a sharp boundary with velocities from around 7 km/s in the lower crust to 8 km/s of the upper mantle; both provide nearly identical waveform match for the field data. The ambiguity comes from the resolution limit of the band-limited seismic data and limited offset range for PmP arrivals.