Calcium isotopic compositions of oceanic crust at various spreading rates

Abstract

The oceanic crust consists mainly of a lower layer of cumulate gabbroic rocks and an upper layer of differentiated basalts. The thicknesses and proportions of the gabbroic and basaltic layers in different oceans are largely controlled by spreading rate, magma supply, and magmatic differentiation processes. Evaluating the effects of complex magmatic differentiation as a function of spreading rate on Ca isotope composition is critical to understanding whether the Ca isotope compositions of oceanic crust from different oceans are homogeneous and thus whether the observed considerable variation of δ44/40Ca in basalts (up to 0.4‰) results from magmatic differentiation or mantle source heterogeneity. To address the question, we present δ44/40Ca measurements of a series of gabbroic rocks (n = 38) and mineral separates from the 810-m-long U1473A hole drilled into the gabbroic lower crust at the ultraslow-spreading Southwest Indian Ridge (SWIR), along with 12 mid-ocean ridge basalts (MORBs) from the slow-spreading South Mid-Atlantic Ridge (SMAR) and the fast-spreading East Pacific Rise (EPR). Although the gabbroic rocks of the SWIR reflect several events of magma supply and strong magmatic differentiation (bulk rock Mg# of 64–79 for each event), their δ44/40Ca values (0.85 ± 0.09‰, 2sd, n = 37) are uniform. The results are consistent with limited inter-mineral Ca isotope fractionation between plagioclase (Pl) and co-existing clinopyroxene (Cpx) in the accumulated gabbros (average Δ44/40CaPl-Cpx = −0.10‰, n = 5). This indicates that no measurable Ca isotope fractionation occurs during formation of ultraslow-spreading oceanic crust. The MORBs from the SMAR and EPR show consistent δ44/40Ca values (0.82 ± 0.08‰ (2sd, n = 4) and 0.86 ± 0.09‰ (2sd, n = 8), respectively), regardless of the degree of fractional crystallization. On the whole, the ultraslow-, slow- and fast-spreading gabbroic cumulates and MORBs display indistinguishable δ44/40Ca within analytical uncertainty, suggesting a homogenous Ca isotope composition for the global igneous oceanic crust (δ44/40Ca = 0.85 ± 0.09‰, 2sd, n = 49) even if they experience complex magmatic differentiation. Comparison with values for fertile mantle rocks (δ44/40Ca = 0.94 ± 0.10‰) reveals that partial melting triggers only slight Ca isotope fractionation (0.09 ± 0.02‰, 2se). In this light, the considerable variation of previously reported δ44/40Ca values for basalts may result from their different mantle sources, and is probably attributable to the recycling of crustal materials.