A carbon, nitrogen, and multi-isotope study of basalt glasses near 14N on the Mid-Atlantic Ridge. Part B: Mantle source heterogeneities

Geochemical variations along mid-ocean ridges reveal the heterogeneous nature of the convecting upper mantle and geodynamic evolution of our planet (e.g., Hofmann, 2007, Parai et al., 2012). Although the occurrence of incompatible element and isotopically enriched mid-ocean ridge basalts (E-MORB) commonly arises from interaction with nearby mantle plumes, the source of E-MORBs far from known hotspots is debated. A well-known example of an enigmatic geochemical enrichment is found at 14 degrees N on the Mid-Atlantic ridge (MAR). This is also one of the few locations worldwide where volatile-saturated E-MORBs, often referred to as "popping rocks" (PR), have been recovered. Although the mechanism(s) involved in popping rock generation remain elusive, compressional regimes associated with the exhumation and formation of oceanic core complexes (OCC) may be required to produce popping rocks via protracted volatile accumulation. However, the geochemical signature of OCC samples and their potential relationship to local geochemical enrichments associated with popping rock-affiliated MORBs remain unknown. Here, we present a comprehensive volatile characterization of popping rocks and associated MORBs (n = 19) sampled at 14 degrees N on the MAR, including (n = 2) normal MORBs (N-MORB) from an oceanic core complex (OCC) and (n = 17) E-MORBs (Bekaert et al., Part A). We use isotopic and abundance data for volatile (carbon, nitrogen, noble gases) and radiogenic (Pb, Sr, Nd) elements, as well as the abundances of major and trace elements, to elucidate on the source(s) of E-MORBs at 14 degrees N and discuss the potential origin(s) of geochemical heterogeneities within the upper mantle. We observe co-variations of helium and radiogenic element isotopes, suggesting potential contributions from a young HIMU-type (high mu = time-integrated U-238/Pb-204) component in the OCC mantle source. The mantle source of the OCC samples is clearly distinct from that of PR-affiliated samples, suggesting no genetic relationship between OCC samples and E-MORBs. In addition, elevated Dy/Yb in OCC samples likely point to the incorporation of a subducted crustal component that is not observed in the mantle source of other MORB samples analyzed in this study. We report mantle source Ar-40/Ar-36 variations at 14 degrees N, which, in line with previous studies, are interpreted as primarily reflecting variations in the amounts of recycled atmospheric Ar. Despite extensive evidence for drastic geochemical heterogeneities at 14 degrees N on the MAR, we observe no significant delta N-15 variation (average delta N-15 = -4.49 +/- 1.40 parts per thousand) across N-MORB and E-MORB samples. This is a fundamental constraint, as the absence of significant N isotope variations across the upper mantle may imply that sedimentary N (with a typical delta N-15 similar to +6 parts per thousand) is not extensively introduced within this reservoir during subduction. We investigate several scenarios that could explain this observation, including a significant contribution of altered oceanic crust to the overall budget of subducting slabs, quantitative return of subducting sedimentary N to the Earth's surface by arc volcanism, and/or preferential transport/preservation of sedimentary N in the lower mantle source of oceanic island basalts.