416 



82 OCEANOGRAPHY IN THE NEXT DECADE 



ping of the mid-ocean ridge, for example, have been confined to 

 only a small fraction of the total ridge length. The diversity of 

 volcanic and tectonic processes manifested along the ridge axis, 

 as a consequence, has not yet been fully defined. More funda- 

 mentally, the complex and linked processes of magmatism, hydro- 

 thermal circulation, development of vent ecosystems, and lithos- 

 pheric evolution are only dimly understood. The dynamics of 

 these processes have not yet been elucidated because of the lack 

 of in situ observations of sufficient duration and diversity to de- 

 termine the important interactions and time scales. 



For a better understanding of mid-ocean ridge processes and 

 their impacts on the chemical, physical, and biological evolution 

 of the oceanic mantle, crust, and hydrosphere, specific aspects of 

 the ridge system will require focused research efforts. Some are 

 discussed below. 



Mantle Flow, Melt Generation, and Magma Transport 

 Beneath Mid-Ocean Ridges 



Plate spreading and the generation of new oceanic crust and 

 lithosphere along oceanic spreading centers involve a variety of 

 complex and interrelated geodynamic processes: upwelling and 

 horizontal divergence of the solid mantle beneath spreading cen- 

 ters, pressure-release melting of this upwelling mantle and segre- 

 gation of the partial melt from the deforming solid matrix, the 

 emplacement and solidification of melt at shallow depths to cre- 

 ate the oceanic crust, and the cooling of the crust and mantle to 

 form the oceanic lithosphere. These processes are still among the 

 more poorly understood aspects of the seafloor spreading process. 

 Two of the most important questions are (1) the pattern of mantle 

 flow beneath mid-ocean ridges, and (2) the geometry of the melt- 

 ing region in the mantle and how melt migrates to the ridge axis. 

 Simple plate-driven flow, due to viscous coupling of the astheno- 

 sphere to the separating lithospheric plates, predicts a simple two- 

 dimensional upwelling pattern more than several hundred kilo- 

 meters in width. Pressure-release melting of this upwelling mantle 

 is thus expected to occur over a very broad region beneath mid- 

 ocean ridges. One of the first-order paradoxes in our present un- 

 derstanding of mid-ocean ridge geodynamics is how partial melt 

 formed over such broad regions beneath ridges migrates to the 

 extremely narrow (1- to 5-kilometer-wide) zone of eruption ob- 

 served at mid-ocean ridges. We also have only a very crude un- 



