PROBLEMS WITH IMAGING CELLULAR HYDROTHERMAL CONVECTION IN OCEANIC-CRUST

Heat-flow variations are commonly used to infer the existence of hydro thermal convection in young oceanic crust and, in special cases, the g eometry and vigor of convection. In a small area on the eastern flank of the Juan de Fuca Ridge where the seafloor and seismic basement reli ef are only about 10 m, a coherent pattern of heat-now variation havin g 40 mW m(-2) peak-to-peak amplitude and half-wavelength of about 600 m has been previously observed and interpreted to be caused by cellula r convection within the extrusive layer 2a of the oceanic crust. In Au gust, 1995, new seismic reflection and heat-flow data were collected t o define better the structure of basement in this area, and to determi ne the planform of the heat-flow variations revealed along the single transect of the earlier observational study. The new, higher resolutio n seismic results show considerable intra-crustal structure; in partic ular, the uppermost crust is reverberant beneath areas of lower heat f low, and a diffractive character extends close to the basement surface beneath areas of higher heat flow. This relationship suggests that th e heat-flow variations may be caused by variations in depth to the top of a high-permeability intracrustal zone that is not coincident with acoustic basement, where relatively uniform temperatures are maintaine d by vigorous hydrothermal circulation of unresolved planform. If this is the case, the search that began in the 1970's for heat-flow variat ions that might reveal the scale of cellular convection, and hence by inference the penetration depth of hydrothermal circulation in the oce anic crust, may have been frustrated once again.