This short paper provides a summary of postcruise scientific results from Ocean Drilling Program (ODP) Leg 209 available to date, building upon shipboard observations and synthesis summarized in Kelemen, Kikawa, Miller, et al. (2004). Emphasis is placed upon published and submitted papers, although in a few cases these are supplemented by data and interpretations published only in abstract form. The paper is organized by discipline, beginning with those most central to the primary goals of the leg. Discussion and conclusions are offered within each disciplinary group rather than in a comprehensive section at the end of the paper.
In addition to this summary, a Leg 209 scientific synthesis paper was submitted to both Science, where it was not reviewed, and to Nature. Efforts continue to publish this paper, which is currently in revision as a Letter to Nature (Kelemen et al., submitted [N1]). Meanwhile, the reader is directed to the syntheses available in Kelemen, Kikawa, Miller, et al. (2004) for a more thorough treatment of the overall results of Leg 209.
Understanding the creation of oceanic plates at spreading ridges provides the simplest template for understanding creation and evolution of Earth's crust and shallow mantle in all tectonic settings. Some aspects of ridge processes are clear; for example, it is established that plate spreading drives mantle upwelling and leads to decompression melting, which in turn forms igneous crust. However, fundamental differences exist between oceanic plates formed at "fast-spreading" ridges (>2 cm/yr half-rate) versus those formed at "slow-spreading" ridges (<2 cm/yr half-rate) that are not well understood. Seafloor topography, fractionation-corrected lava compositions, seismic crustal thickness, and topographically corrected gravity anomalies are all more variable at slow-spreading ridges compared to fast-spreading ridges, but the reasons for these differences remain unclear. Unresolved debate over the past two decades has focused on the pattern of mantle upwelling (two-dimensional, plate-driven corner flow versus melt-buoyancy driven radial diapirs), the causes of fractionation-corrected variation in lava compositions (variation in mantle source temperature versus variation in the depth of crystal fractionation), and the reasons for variable crustal thickness along strike (a short review of these topics is provided in Kelemen, Kikawa, Miller, et al., 2004). ODP Leg 209 was proposed to address these problems, providing constraints via drilling of mantle peridotite intruded by gabbroic plutons and exposed on both sides of the slow-spreading Mid-Atlantic Ridge between 14°43´ and 15°44´N.
During Leg 209, 19 holes were drilled at 8 sites (Fig. F1). Sites were previously surveyed by submersible and were <200 m from peridotite or dunite exposed on the seafloor; outcrops of gabbroic rock were also near some sites. Primary goals of Leg 209 were to constrain deformation associated with mantle upwelling and corner flow and mechanisms of melt migration and igneous petrogenesis.
In six holes at Sites 1269 and 1273, we penetrated a total of 112 m of basaltic rubble; recovery was poor (3.7 m total; recovery = 3.3%) and these holes were unstable, so drilling was terminated. Lavas at these sites probably form nearly horizontal surfaces overlying cliffs exposing peridotite and gabbro. These sites will not be discussed further here.
In 13 holes at 6 other sites, we drilled a mixture of residual peridotite and gabbroic rocks intrusive into peridotite. We penetrated a total of 1075 m at these six sites and recovered 354 m of core (recovery = ~33%). This paper reports on postcruise scientific results from these six sites.
Simplified core logs (Fig. F2) show that drilling at Sites 1268, 1270, 1271, and 1272 recovered ~25% gabbroic rocks and ~75% residual mantle peridotite (detailed lithologic sections are available in Shipboard Scientific Party, 2004a: figs. F7, F21, F31, F37). Core from Site 1274 is mainly residual peridotite with a few meter-scale gabbroic intrusions (Shipboard Scientific Party, 2004a: fig. F42 and associated text). Core from Site 1275 is mainly gabbroic but contains 24% poikilitic peridotite (also known as "troctolite") interpreted as residual peridotite that was "impregnated" by plagioclase and pyroxene crystallized from melt migrating along olivine grain boundaries. These impregnated peridotites were later intruded by evolved gabbros (Shipboard Scientific Party, 2004a: fig. F50 and associated text). Impregnated peridotites form the majority of core from Site 1271 and are present at Sites 1268 and 1270. Thus, we infer that the entire area may be underlain by mantle peridotite with ~20%–40% gabbroic intrusions and impregnations. The overall proportion of gabbroic rocks versus residual peridotites from these six sites is similar to previous dredging and submersible sampling in the area (compiled data and references in Kelemen, Kikawa, Miller, et al., 2004).