An Arc Origin for Archean High MgO "Eclogite" Xenoliths?

The origin and evolution of Archean continental crust is an important topic in the Earth sciences. By understanding how Archean crust forms, we are better able to track how chemical differentiation and geodynamic processes have evolved on the Earth over billion-year timescales. Suggested mechanisms responsible for Archean crust formation include melting of subducted slabs, melting of orogenically thickened basaltic crust, and generation of large oceanic plateaus (proto-continents). In all these scenarios, high temperatures are required. Since it is reasonable to assume that the mantle in the Archean was probably hotter, these mechanisms may have been dominant at that time. The likely colder temperatures in the Phanerozoic, however, would prevent these mechanisms from operating extensively today. Instead, most continental crust formation in the Phanerozoic tends to be associated with arc magmas formed not by slab melting but by hydrous melting of the peridotitic mantle wedge. The question we wish to address is the extent to which Phanerozoic-like arc processes might also have operated in the mid- to late-Archean. Answering this question may help us better understand Earth's secular thermal evolution. Towards these ends, we have been focusing on understanding the origins of high MgO "eclogite" xenoliths found in Archean cratons. While the origin of low MgO "eclogites" is understood to be partially melted subducted oceanic crust, the origin of the high MgO "eclogites" is still debated. Here we show that high MgO Archean "eclogite" xenoliths have major element systematics remarkably similar to high MgO garnet pyroxenite ("eclogite") xenoliths originating from the lithospheric root underlying the Phanerozoic Sierra Nevada batholith in California, the remnant of a Mesozoic continental arc. Both groups have similarly high MgO contents, high Mg/(Mg+Fe) ratios, and relatively high SiO2 contents. Such compositions are not represented by typical frozen melts. In the case of the Sierra Nevada, the mafic end of the Sierran plutonic differentiation series requires a complementary reservoir having a composition that is matched by the high MgO Sierran "eclogites". The uniformly high Mg#s of the Sierran high MgO "eclogites" prevent them from being melt residues of a basaltic protolith. Were they to be melt residues, we might also expect to see a continuous compositional spectrum reflecting varying degrees of melt depletion of a common protolith, but no such continuity is observed. This suggests instead that the Sierran high MgO "eclogites" are high pressure cumulates formed during juvenile arc magmatism on the edge of a continent. Given the remarkable major-element similarities between the Archean and Sierran high MgO "eclogites", we speculate that they may have similar origins. The similarly high SiO2 contents, reflecting higher pyroxene to garnet ratios than seen in low MgO eclogites, is likely to have been enhanced by the presence of water. We note that although Archean and Sierran high MgO "eclogites" both have low Al contents, the Archean ones have slightly higher Al than the Sierran ones, suggesting greater garnet involvement and hence higher pressures of crystallization in the Archean. If these compositional similarities indeed reflect genetic similarities, the implication is that Archean high MgO "eclogites" might be cumulates formed in continental arcs. Seismic velocities calculated for high MgO "eclogites" reveal considerable overlap with peridotite velocities, unlike low MgO eclogite velocities, which tend to be high. This suggests that it may be potentially difficult to seismically detect high MgO "eclogites" in the cratonic roots.