Deepening of the axial magma chamber on the southern East Pacific Rise toward the Garrett Fracture Zone

Abstract
A wide‐aperture profile along the ridge axis from 14°29′S to 13°39′S, 120 km to 30 km south of the Garrett Fracture Zone, is analyzed to constrain the thickness of layer 2a and the depth to the axial magma chamber reflector. Five areas along the 90 km line are examined in detail, with several consecutive gathers being analyzed for each area to establish the degree of consistency within each area. A genetic algorithm code is used to find a best fit model from a comparison of the data and WKBJ synthetic seismograms. One hundred starting models are generated using a predefined set of velocity nodes, with a fixed window of allowable depth variations between nodes. An evolutionary process favors the better fitting models in each generation, and a satisfactory misfit is usually obtained within 40 generations. Within individual areas the models were in good agreement with the depth of a given velocity node, generally varying by not more than 20 m, the depth discretization interval for the models. A consistent deepening trend of the axial magma chamber (AMC) is observed across the five areas as the Garrett Fracture Zone is approached. The depth varies from 0.99 km at area 1, which is approximately 100 km south of the Garrett, to 1.23 km at area 5, which is approximately 40 km south of the Garrett. The depth to the axial magma chamber is highly sensitive to any ship wander off axis since layer 2a thickens rapidly off axis with age. For the areas examined here, layer 2a is observed to be relatively constant in thickness along the axis, although it is about 40 m thicker over area 5, where the axial magma chamber is deepest. This variation is within the scatter of previously detailed layer 2a measurements at 13°N on the East Pacific Rise, where an effectively constant thickness is observed. This implies that layer 2a thickening is not a significant factor along this profile and that the AMC deepening is real rather than apparent. Theoretical modeling suggests that the depth to the lid of the axial magma chamber is related to the rate of heat supply at a given location. Thus the gradual consistent deepening of the axial magma chamber can be taken as an indication of a slightly reduced magma supply toward the Garrett Fracture Zone, which marks a major interruption of hundreds of kilometers of continuous ridge axis. The deepening may also be interpreted as a downward limb from a central injection point; however, there is no indication of a similar downward trend in the other (southern) direction. Furthermore, there is no accompanying systematic variation in axial depth or axial volume, both of which are proposed to be indicators of central injection and along‐axis flow.