Potential control of hydrothermal circulation on the structure and segmentation of axial magma chambers along fast-spreading ridges

J.A. Olive¹*, F.J. Fontaine², & J. Escartin²

Corresponding author: jaolive@mit.edu
¹Massachusetts Institute of Technology / Woods Hole Oceanographic Institution Joint Program, Department of Geology and Geophysics, Woods Hole, MA 02543
²Institut de Physique du Globe, Equipe de Géosciences Marines, Paris, cedex 05, France

Abstract:
We use 2D porous convection models to investigate the thermal structure of the transition zone between a hydrothermal convection system and an impermeable layer representing an axial magma chamber. We use temperature-dependent permeability with a sharp permeability drop at 600°C, temperature above which micro cracks are assumed to be poorly connected or closed. Reciprocally, this assumes that cooling results in instantaneous permeabilizing. Within an appropriate range of Rayleigh numbers, convection systematically develops above the 600°C isotherm and deepens at a rate that scales with the permeability structure. The bottom of the hydrothermal system is a thermal boundary layer showing a significantly higher heat extraction rate below down-flow zones than below up-flow zones (conductive heat flux may vary by a factor reaching 5 to 10 between both areas). We show that this laterally heterogeneous boundary condition applied at the top of an axial magma chamber (AMC) could cause AMC height variations ranging from 50 to 100 m, assuming no significant melt convection within the AMC. This may result in along-axis transitions from melt to mush (0 to 40% melt) zones on the length scale of one hydrothermal convection cell (several kilometers, Rayleigh-dependent) within the AMC. If no magma replenishment occurs, this structure evolves into an along-axis segmented AMC with discontinuities lying below hydrothermal recharge areas. The AMC then fully crystallizes in tens of years as the hydrothermal system keeps migrating downward (for a characteristic permeability of 10-14 m²).

Keywords:
Hydrothermal convection, axial magma chamber, segmentation, oceanic crust permeability

Contributions to Integration and Synthesis:
This work provides new ideas to investigate the relationship between hydrothermal and magmatic systems, particularly: heat transfer through the dike-gabbro transition zone and AMC lifespan, which are critical to understand how lower plutonic crust is accreted at fast spreading ridges.
More specifically, along-axis variations in melt content within axial magma chambers have been documented on fast-spreading ridges at the segment scale (Singh et al. [1998]). Here we suggest that hydrothermal convection may induce along-axis heterogeneities in melt content at the scale of one hydrothermal convection cell (several kilometers), which unfortunately is probably below the current resolution of reflection seismology. However, recent micro-seismicity data on the EPR at 9°N seem to image a hydrothermal recharge area sitting above an AMC discontinuity (Tolstoy et al. [2009]), which is consistent with our model results. We also suggest that (sub)kilometer-scale observations of alteration patterns along the dike-gabbro transition in ophiolites may also help constrain our model results.