Oceanic crust composition and structure.
Oceanic crust is created at mid-ocean ridges where magma is continuously supplied from the mantle below, generated by the rise of hot, solid material from deep in the earth, followed by its partial melting at shallow depths. Three main crustal formations result from different rates of magma cooling and crystallization: fossil magma chambers, sheeted dikes, and pillow laves. Fossil magma chambers are composed of gabbroic rocks with large crystals (1 to 10 millimeters in diameter) that form by slow cooling of magma within the crust. The crust acts like a Thermos bottle, insulating magma from cold seawater, and allowing crystallization and solidification to proceed over tens of thousands of years. Sheeted dikes are "frozen" channels where magma once flowed up toward the seafloor. When flow in these channels ceased, magma crystallized rapidly, perhaps within hours, to form basalts with small crystals (most less than 1 millimeter in diameter and many too small to see without the aid of a microscope). Pillow laves form by the eruption and "quenching" of magma on the seafloor--cooling is so fast that volcanic glass forms on the rims of pillows Slightly slower cooling within pillows produces crystalline basalt.
The amount of magma generated and the proportion of it that erupts varies along mid-ocean ridges, leading to significant variations in total crustal thickness and in the relative proportions of gabbroic rocks, sheeted dikes, and pillow lavas. High magma supply, high eruption rates, and a thick crust are typical of rapid spreading rates at mid-ocean ridges such as the East Pacific Rise. Low magma supply, low eruption rates, and a thin crust are typical of mid-ocean ridges where spreading rates are low, such as the Southwest Indian Ridge. Theoretically, high magma-supply rates should also result in slower cooling rates and a higher proportion of gabbroic rocks in the crust, but this remains to be proven.
To fully characterize the oceanic crust's composition and to understand how its composition is influenced by magmatic and hydrothermal processes requires a scale of sampling that can only be achieved by drilling. DSDP and ODP have successfully recovered most parts of the crust by drilling deep at a few sites and by taking advantage of lower crustal rocks exposed on the seafloor. A complete section of upper crust, including the lava and sheeted dike complex, has been sampled in a hole about 2,200 meters deep drilled during seven legs in the eastern equatorial Pacific at Site 504. Leg 118 recovered 500 meters of gabbroic rocks that formed in a magma chamber beneath the very slow-spreading Southwest Indian Ridge. Leg 147 recovered a sequence of gabbros that formed at the magma-rich East Pacific Rise, as well as the complex transition zone between the crust and upper mantle, revealing the trapping and crystallization of magma within a previously melted piece of mantle. Investigation of these and other cores has significantly changed our view of how oceanic crust is built.
Variations in magma supply imply variations in the average degree of melting in the mantle, which affects the composition of primary magmas coming out of the mantle and therefore the average composition of the oceanic crust. At one extreme, low magma-supply rates result in infrequent intrusion of magma into the crust and "freezing" of the magma to form dikes whose basaltic composition is nearly the same as the melt initially generated in the mantle. With greater magma supply, more magma is intruded into the crust, its cooling rate decreases, and it is subject to the process of fractional crystallization prior to solidification. Just as evaporation of seawater leads to removal of pure water and concentration of salt in the water, the fractional crystallization of magma leads to the removal of some elements in crystal form and the concentration of others in the residual liquid. Dikes and lavas formed after fractional crystallization are significantly different in chemical composition than the melts originally generated in the mantle. This is because the first crystals to form in a basaltic magma, olivine and plagioclase, are chemically very different from the initial magma. Extensive crystallization and the addition of iron-titanium oxide minerals to the crystallizing assemblage may lead to the generation of melts that are very rich in silica (trondhjemite in the table opposite).
As magmas cool, crystals may accumulate on the floors, walls, and roofs of magma chambers and form crystal mushes that initially contain 40 percent melt, but prior to solidification contain less than 15 percent trapped melt. Melt may be expelled from a mush by such processes as compositional convection, compaction, and deformation. Solidification of mushes produces cumulate gabbros (troctolite and iron-titanium oxide gabbro in the table) with compositions that are significantly different from magma compositions (basalts). Troctolites are primitive cumulates, assemblages of olivine and plagioclase crystals together with a small fraction of crystallized trapped liquid, that formed during the early stages of magma crystallization. Iron-titanium oxide gabbros, on the other hand, are evolved cumulates that formed after extensive crystallization of basaltic magma at mid-ocean ridges.
Crystallization models and magmatic intrusions exposed on land suggest a simple crustal stratigraphy for the lower ocean crust, with primitive gabbros at the base displaying well-developed crystal layering and evolved gabbros toward the top characterized by the absence of layering. So far, we have yet to observe well-developed layering in drilled sequences of oceanic gabbros, and at Site 735 we found evolved gabbros interdigitated with primitive olivine gabbros and troctolites. Detailed chemical mapping of contacts between iron-titanium oxide gabbros and olivine gabbros at Site 735B indicates that evolved melts are sometimes mobilized in response to crustal deformation, and that melt flow may be either diffused through intergranular networks or focused along centimeter-scale channels. Depending on magma supply and cooling rates, crystal mushes may be invaded with new magma prior to solidification, modifying the bulk composition of the mush in addition to the composition of the invading magma.
The composition of the oceanic crust that results from magmatic processes is the starting point for a complex history of chemical exchange with seawater that leads to the formation of ore deposits and influences TABULAR DATA OMITTED the composition of the world's oceans, arc volcanics, and the mantle. When erupted on the seafloor, volcanics immediately begin to react with the surrounding seawater. Within a hydrothermal system, seawater flows downward through cracks, fissures, and faults and may penetrate to depths as great as 5 or 6 kilometers. It reacts with rocks all along its path, resulting in exchanges of elements and the chemical modification of both seawater and crust. The extent of this exchange depends primarily on temperature and the abundance of seawater passing through a volume of rock, or, in other words, the water/rock ratio. Seawater heats as it migrates down into the crust toward an active magma chamber, and resulting hydration reactions enrich crustal rocks in magnesium and sodium and depletes them of calcium. By the time seawater makes it to just above the magma chamber, its composition has been significantly modified and low water/rock ratios and high temperatures (about 400 |degrees~ C) lead to leaching of metals from the rocks. These buoyant, hot, metal-enriched fluids rise to the seafloor where they mix with the ambient seawater and precipitate sulfides that accumulate in chimneylike structures and mounds. Leg 139 investigated the structure and composition of a hydrothermal system that extends from the basaltic basement into an overlying sequence of marine sediments. In fall, 1994, Leg 158 will drill into an unsedimented deposit. Chemical exchange within the deep root-zones of hydrothermal systems has been documented at Hole 504B where the base of a sheeted dike complex was found to be depleted in copper and zinc.
Although it is not known whether or not seawater-derived fluids actually enter into active magma chambers, the gabbroic core recovered at Hole 735B demonstrates that seawater did penetrate the lower crust very early in its history. Alteration of gabbroic rocks was initiated at temperatures greater than 600 |degrees~ C and focused within zones of ductile deformation. These zones show very little change in composition because the fluids had become strongly enriched in basaltic components by the time they reached this depth (3 to 4 kilometers). Lower-crust hydration may not be a significant process at ridges where there is a high rate of magma supply, simply because the crust is too hot to deform in a way that provides pathways for fluids to flow deep into it. In fact, East Pacific Rise gabbroic rocks recovered during Leg 147 show that ductile deformation is not prevalent at this magma-rich ridge.
Modification of oceanic crust composition does not stop when a section of crust moves away from a mid-ocean ridge and off-axis. Seawater continues to circulate in and out of the crust until fluid pathways are sealed by the precipitation of minerals, or until sediment accumulation prevents penetration into the crust. As a crustal section ages and moves away from the mid-ocean ridge, the most significant compositional change occurs in the upper volcanic carapace, as fluid pathways deeper in the crust are thought to become sealed by the time it leaves the ridge. Within 5 to 10 million years, the volcanics are enriched in elements such as magnesium and potassium, and much of the basaltic iron has been oxidized. Although isotopic age dating of carbonates shows that mineral precipitation ceases within 20 million years, heat-flow data indicate that seawater may well continue to circulate beyond this time frame. The chemical consequences of such prolonged seawater circulation are not known.
Most of our knowledge of crustal aging processes comes from the recovery of shallow oceanic crust that ranges from essentially zero age (such as at Site 649) to as old as the Jurassic, about 160 million years ago (Hole 801C). Core from more than 150 basement sites demonstrates that interaction between seawater or seawater-derived fluids and rock has a significant impact on crustal composition. Downhole compositional trends at Hole 504B show that different elements are mobile in different parts of the crust. Differences in chemical fluxes found in cores from Sites 417 and 418, drilled only 500 meters apart, show that the composition of the uppermost crust may be quite heterogeneous. Comparing cores of varying ages from all ocean basins suggests that the rate of chemical exchange is not simply a function of age, and that the greatest change in composition may occur in young crust. Chemical exchange within the oceanic crust plays an important role in world-ocean water composition by contributing to the delicate balance of sources and sinks that include the continents (through river input), the atmosphere, ocean sediments, and the ocean itself. Drilling the oceanic crust has proven to be an essential step in furthering our understanding of global geochemical cycles.
Peter S. Meyer is an Associate Professor at Rhode Island College and an Adjunct Scientist in the Department of Geology and Geophysics at the Woods Hole Oceanographic Institution. He developed an interest in geology while writing a career notebook in the eighth grade and visiting marble quarries in Vermont, then became hooked while scrambling up volcanos in Central America as an undergraduate at Dartmouth College. His current research interests include magma chamber dynamics, crystal-melt equilibria, and the evolution of the lower oceanic crust.
Kathryn M. Gillis is an Associate Scientist in the Department of Geology and Geophysics at the Woods Hole Oceanographic Institution. She developed an interest in geology during a family vacation across the US and Canada where she encountered a thoughtful observer of the earth, her cousin Jack. Over the years this interest became linked with her roots in eastern Canada and she eventually developed into a marine geologist. Her research interests revolve around the processes that shape the seafloor and the interaction between fluids and rocks.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||25 Years of Ocean Drilling|
|Author:||Meyer, Peter S.; Gillis, Kathryn M.|
|Date:||Dec 22, 1993|
|Previous Article:||Oceanic crust and mantle structure.|
|Next Article:||Exploring large subsea igneous provinces.|