Nd, Sr and Os isotope systematics in young, fertile spinel peridotite xenoliths from northern Queensland, Australia: A unique view of depleted MORB mantle?

Northeastern Queensland, a part of the Phanerozoic composite Tasman Fold Belt of eastern Australia, has a Paleozoic to Mesozoic history dominated by subduction zone processes. A suite of 13 peridotite xenoliths from the <3 Ma Atherton Tablelands Volcanic Province, predominantly from Mount Quincan, comprise fertile (1.8-3.4 wt.% Al₂O₃ and 38.7-41.9 wt.% MgO) spinel lherzolites free from secondary volatile-bearing phases and with only weak metasomatic enrichment of incompatible trace elements (Sm_N/Yb_N = 0.23-1.1; La_N/Yb_N = 0.11-4.9). The suite is isotopically heterogeneous, with measured Sr (⁸⁷Sr/⁸⁶Sr = 0.7027-07047), Nd (¹⁴³Nd/¹⁴⁴Nd = 0.51249-0.51362), and to a lesser extent, Os (¹⁸⁷Os/¹⁸⁸Os = 0.1228-0.1292) compositions broadly overlapping MORB source mantle (DMM) and extending to more depleted compositions, reflecting evolution in a time-integrated depleted reservoir. Major and rare earth element systematics are consistent with mantle that is residual after low to moderate degrees of melt extraction predominantly in the spinel facies, but with a few samples requiring partial melting at greater pressures in the garnet field or near the garnet-spinel transition. In contrast to most previously studied suites of continental lithospheric mantle samples, the incompatible trace element contents and Sr and Nd isotopic systematics of these samples suggest only minimal modification of the sampled lithosphere by metasomatic processes. Five of six Mount Quincan xenoliths preserving depleted middle to heavy REE patterns form a whole rock Sm-Nd isochron with an age of ∼275 Ma (ε_Ndi = +9), coincident with widespread granitoid emplacement in the overlying region. This isochron is interpreted to indicate the timing of partial melting of a DMM-like source. Xenoliths from other Atherton localities scatter about the isochron, suggesting that the sampled mantle represents addition of DMM mantle to the lithosphere in the Permian, when the region may have broadly been within a subduction zone setting. A sixth middle to heavy REE-depleted Mount Quincan xenolith has a distinct Nd and Os isotopic composition consistent either with an earlier, possibly Precambrian melt extraction event, or with Permian derivation from a mantle source with a less depleted (time-averaged lower Sm/Nd) Nd isotopic composition, but a more depleted (low Re/Os) Os isotopic composition. The range in measured whole rock Os isotopic compositions cannot solely be the result of time-integrated effects of variable melt extraction, especially considering the coherent Sm-Nd systematics of the suite. The Os heterogeneity more likely reflects either a heterogeneous ∼275 Ma DMM source that would have a present-day Os composition (¹⁸⁷Os/¹⁸⁸Os ∼ 0.1265-0.1287) overlapping both abyssal peridotites and chondrites, or significant and variable enrichment within the lithospheric mantle by secondary sulfides carrying radiogenic Os in a cryptic chalcophile enrichment event. Regardless of the origin of the Os isotopic variability, these data highlight the mantle Re-Os isotopic heterogeneity that may be present over small length scales where the lithophile Sm-Nd system may be relatively homogeneous.