Seismic boundary 2A/2B: 1) lithologic boundary between lavas and dykes or 2) alteration boundary

Our understanding of oceanic crust is derived from exposures on land (termed 'ophiolites'), dredging of its surface, geophysical surveys that map bathymetry and changes in physical properties with depth, and through a small number of deep drill holes. Oceanic crust is typically around 6 km thick and was initially divided into sediments (layer 1), an extrusive layer (layer 2) and an intrusive layer (layer 3). In seismic-velocity data we observe a rapid increase in velocity from around 2 to 5 km/s in the upper oceanic crust. It is widely believed that this change in velocity is related to a decrease in crack porosity, and that this is either related to a geological transition from lavas to dykes (hypothesis 1: lave/dyke), or to a change in the intensity of alteration with depth, in which hydrothermal mineralisation is associated with crack closure (hypothesis 2: alteration). Although conventions vary, here we will refer to: layer 2A as the layer with a velocity of < 3 km/s, the high velocity-gradient zone as the layer 2A/2B boundary or transition, and layer 2B as the layer with a low velocity-gradient in which velocities are > 4-5 km/s. On seismic reflection data acquired across young oceanic crust, we observe wide-angle arrivals that turn within the 2A/2B transition. These arrivals can be processed to form a bright reflector on a conventional reflection stack, and converted to depth using velocity data. The crux here is that velocities obtained from refraction and reflection travel-times and/or stacking velocities are neither accurate nor well resolved, and this leads on to poor velocity and depth control for the 2A/2B transition. Two seismic datasets were acquired by the project partner across oceanic crust in the equatorial Pacific ocean. One is located at the Hess Deep Rift (HDR) which has a fast spreading rate, and the other is located at the adjacent Blanco Transform Fault (BTF) which has an intermediate spreading rate. The profiles were acquired on plateaus immediately adjacent to the walls of HDR and BTF fault scarps, and the purpose of the experiment was to compare directly the seismic layer 2A/2B boundary with mapped geological units. Initial results show that the seismic transition from 2A to 2B is close to the top of the mapped dykes at the HDR, but within the lava flows above the dykes along the BTF, and these combined observations suggest that hypothesis 2 (alteration) is more likely. However, there remains uncertainty in the precise depth of the 2A/2B seismic transition, as the current velocity model lacks good spatial resolution. The purpose of our proposed study is to use an inversion code that models the full seismic wavefield, to provide a more accurate and well-resolved velocity model. This will allow us to constrain better the depth to the 2A/2B boundary, and hence decide more confidently between the two hypotheses. We will also be able to observe the fine-scale structure of the 2A/2B transition as well as its lateral variability, and use these data to infer changes in crack porosity across and along the transition zone. This project has wider applications, including providing data that will allow us to use world wide observations of layer 2A thickness to better resolve oceanic crustal architecture, and will help us constrain the thermal budget of oceanic crust. Our results will be used to plan future broader projects on oceanic crust, as well as identify future applications of the waveform inversion codes.