Heat loss and hydrothermal circulation due to sea-floor spreading

Abstract

Lithospheric cooling along the Galapagos Spreading Center at 86°W longitude, as determined by surface heat-flow measurements, appears dominated by hydrothermal circulation. This same phenomena apparently exists on the Mid-Atlantic Ridge at 36°N and presumably, in some form on all active oceanic ridges. It is responsible for removing the majority of the heat (>80%) lost through young (few m.y. old) oceanic crust. This component of heat has been ignored in previous calculations of the total rate of heat loss by the Earth. A theoretical expression is used to estimate the heat released by sea-floor spreading, since current technology does not provide any means for direct measurement. The revised value of lO.2 x 1Ol2 cal/sec (±15%) represents a 32% increase over previous estimates. More than 20% of this heat apparently escapes through hydrothermal vents near sea-floor spreading centers. The previously accepted equality of oceanic and continental heat flux is invalid. The revised analysis indicates the oceanic heat flux is 2.2 x 1O-6 cal/cm2-sec (HFU) versus l.5 HFU for the continents . The average for the Earth is then approximately 2.0 HFU. The horizontal wavelength of inferred hydrothermal convection at the Galapagos Spreading Center, in the one dimension measured, is 6 il km. The systematic modulation suggests cellular convection. If the system is dominated by cellular convection, the depth of penetration, based on laboratory modeling experiments should be 3 to 4 kilometers. The data from the Galapagos Spreading Center and laboratory experiments both suggest that the position of the cells in a cellular convection system can be a strong function of the local topography, the rising limbs of flow being located beneath topographic highs and the descending limbs beneath topographic lows. The addition of topography enhances the heat transfer efficiency of a convection system. Lateral variation in permeability or the systems bottom boundary condition will also influence the position of cells. Even if the circulation system were strongly influenced by some combination of variations in the strength of the heat source, topography or discrete zones of high permeability, it would probably still be cellular in nature, and similar deep penetration is indicated. If the Galapagos Spreading Center is typical, there are presumably numerous hydrothermal springs and fissures in each square kilometer of near-ridge sea floor and sediment thicknesses of at least 50 meters are apparently penetrable to the flow of water. As the sea floor ages the surface of the hydrothermal system becomes less permeable and eventually both the surface and the deep system are completely clogged and sealed. The age at which this occurs varies from ridge to ridge but there is evidence that suggests it may not be complete until the crust is at least 8 m.y. old and possibly as much as 40-50 m.y. old. Most of the surface is apparently sealed long before hydrothermal circulation stops, although some vents do persist. This behavior of the hydrothermal system has a dramatic effect on conductive heat-flow measurements and is largely responsible for the variations observed in conductive heat flow near active spreading ridges. The results of this study show the difficulties in resolving systematic patterns in the heat-flow distribution on spreading ridges. Numerous, closely-spaced measurements with precise navigation combined with a relatively uniform sediment cover, appear to be necessary ingredients for recognition of the heat-flow pattern near active sea-floor spreading centers.