Mid-ocean Ridge Basalts (MORB)

Basalts are volumetrically the most significant rock type sampled, and to date have provided the bulk of the compositional information we have about the processes of melt generation, transport and crystallization that create the oceanic crust. Progress in the study of basalts over the past 4 years has included new insights into and refinement of our knowledge of the global variability of the chemical compositions of MORB [ Langmuir et al., 1992]), further characterization of the extent of compositional variability exhibited by basalts sampled from individual segments of the global ridge system, links between compositional variability and observed temporal and spatial relationships, determination of chemical signatures of basalts that convey information about the nature of the mantle source, as well as characteristics of the melt transport processes, and the development of new methods for determining the relative ages of these very young basalts.

Sinton et al. [1991] interpreted the scales and kinds of magmatic variations observed at the superfast spreading southern East Pacific Rise (EPR) to indicate that the regional temperatures of the upwelling asthenosphere, magma supply to the axis, and crustal magmatic temperatures reflect independent, regionally decoupled processes. In contrast, Klein et al. [1991] found a strong correlation between upwelling temperature and variations in major and moderately incompatible trace elements observed in MORB from Australian-Antarctic Discordance (AAD). These authors also found that the highly incompatible trace elements and isotopes appear to be decoupled from the major and moderately compatible trace elements. Pyle et al. [1992] presented new Sr, Nd, and Pb isotopic data that confirmed the presence of a boundary between `Indian' type and `Pacific' type MORB mantle beneath the AAD, and used the isotopic variations to constrain the westward migration of the Pacific mantle into the AAD. Large variations in abundance ratios of incompatible elements at similar MgO contents observed in basalts from the AMAR (ALVIN mid-Atlantic Ridge) axial valleys on the mid-Atlantic Ridge (MAR) are interpreted as evidence for mixing of magmas that formed by very different extents of melting in the mantle [ Frey et al., 1993]. This interpretation requires that parental magmas of widely different extents of melting are sampled from the same melting regime directly beneath a given ridge segment and thus, that a large, long-lived magma chamber was not present during eruption of the AMAR lavas. Batiza and Niu [1992] conducted detailed sampling of MORB from the EPR between 9N and 951N. They found that major and trace element data favored derivation of the magma compositions from a single parental composition by low pressure crystallization of the minerals olivine, plagioclase and clinopyroxene, although the lavas were erupted with only plagioclase as a phenocryst phase. They resolved the compositional and petrographic observations by suggesting that the mafic phases settled due to gravitational forces and the plagioclase floated. In addition they inferred that a single parental magma seemed to supply melts to the axial magma chamber along the entire 60 km segment of the EPR, suggesting that central supply injection sites are widely spaced, and that the axial magma chamber was continuous along the ridge.

Mahoney et al. [1992] suggested that the origin of the Pb isotopic signatures of MORB throughout the Indian Ocean could be related to the initiation of a more than 4400-km-long band of juxtaposed plume heads beneath the nearly stationary lithosphere of prebreakup Gondwana. White [1993] linked the variation in the U/Pb ratio observed in MORB sampled from different ocean basins to open system evolution of the depleted upper mantle, suggesting that >10% of the observed mantle plume flux is required to mix with the depleted MORB mantle to supply the required flux of Pb. Mahoney et al. [1994] found a decoupling in the spatial patterns of trace element and isotopic enrichment for the superfast southern EPR that appears to be unique among the ocean ridge system. Salters and Hart [1991] significantly expanded the Lu-Hf isotopic data set for MORB and proposed that the Hf and Nd isotopic systematics of many MORB could only be explained if garnet was a residual phase during melting. As garnet is only stable in the mantle at pressures greater than 25 kbar, melting in the presence of garnet has to occur at depths greater than 80 km. This controversial hypothesis has become a central point of debate in current discussions of MORB genesis.

Geochemical variations in MORB sampled from tectonically complex ridge settings have been reported for the Gulf of Aden, the Tuzo Wilson Volcanic Field (TWVF), and the southern end of the Pacific-East Pacific Rise. Schilling et al. [1992] use the rare earth element and Nd-Sr-Pb isotopic compositional profile of MORB sampled from the Sheba Ridge axis in the Gulf of Aden to constrain the interaction of the head of a starting mantle plume with continental lithosphere and an ocean basin in an early stage of development. Allan et al. [1993] identified alkaline volcanics in the TWVF which are distinctly different from MORB in their major and trace element characteristics. These authors suggested that in contrast to current wisdom which states that the TWFV is either a site of seafloor spreading or a hotspot/mantle plume, their data are consistent with other geophysical data in suggesting that the TWVF represents `leaky transform' volcanism in an oceanic setting. Lonsdale et al. [1992] characterized the triple junction at 2N on the EPR as a ridge-ridge-ridge junction, and found that the MORB erupted there, like those erupted at the similar Pacific-Nazca-Galapagos junction, are slightly enriched in incompatible elements relative to lavas on adjacent segments of the EPR axis.

Exciting progress has been made in the study of temporal variations of the magmatic processes that create the oceanic crust. In contrast to the view that the generation of the oceanic crust is a steady state process producing MORB of essentially constant composition through time, detailed studies conducted at the segment lengthscale have linked patterns of compositional variation in MORB with spatial and temporal variability at the EPR [ Reynolds et al,. 1992], and at the southern Juan de Fuca Ridge [ Smith et al., 1994]. In addition, significant progress has been made in the dating of the very young basalts that erupt at ocean ridges. Goldstein et al. [1991] used U-Th systematics to date axial and off-axis MORB from the Juan de Fuca and Gorda ridges, and showed that the spreading rates inferred from the U-Th dates are consistent with paleomagnetic spreading rates. Preliminary measurements of Ra-Th disequilibrium in MORB glasses from the same region were shown to be magmatic in origin and used by Volpe and Goldstein [1993] to quantify volcanic episodicity at ocean ridges. Goldstein et al. [1994] used U-Th and U-Pa ages (the technique is described in Goldstein et al. [1993]) for basalts to quantify the spatial extent of young volcanism and crustal accretion at 931N on the EPR. They found anomalously young ages for MORB relative to ages inferred from spreading rates and distance from the axis, and inferred from the young ages that most of the dated basalts were actually erupted up to 4 km off-axis. Rubin et al. [1994] present a new chronometer based on Po-Pb radioactive disequilibrium, which allows the dating of glassy eruption products within a few years of their eruption.