查看更多>>摘要：Indian-type mantle is extensively distributed in the marginal basins of the western Pacific; however, die origin of diis mantle remains unclear. The Celebes Sea is located between the Pacific and Indian oceans, and its igneous crustal rocks provide crucial constraints on if and on how this anomalous mantle developed in the region. However, limited knowledge has been obtained on the origin of mantle source of the Celebes Sea because only normal-type mid-ocean ridge basalts (N-MORBs) at ODP Site U770 in this basin have been examined. In this study, bodi N-MORBs and enriched (E)-MORBs were identified at Site U770, which were analyzed for whole-rock major and trace elements, Sr-Nd-Pb-Hf isotopes, and in situ olivine geochemistry. Site U770 basalt samples are Indian-type (i.e., plot above the Northern Hemisphere reference line [NHRL] on plots of 208Pb/204Pb and 207Pb/204Pb versus 206Pb/204Pb) MORBs, based on their Sr-Nd-Pb-Hf isotopic compositions. An enriched mantle end-member exists in the MORB-type depleted mantle source, as evidenced by correlations between eNd and SmN/YbN, Sr/Sr*, and Eu/Eu*. Recycled igneous oceanic crust that experienced plagioclase fractionation can account for die geochemically enriched end-member in the mantle source, based on the positive correlation of eNd vs. Sr/Sr* and Eu/Eu*. The recycled igneous oceanic crust plus minor amounts of pelagic sediments can also explain the origins of Indian-type mantle in the Celebes Sea. The E-MORBs have higher Zn/Mn and Zn/Fe ratios, and similar olivine Ni, Ca, and Mn contents as compared to global MORBs, which is also consistent witii a recycled oceanic crust-derived, eclogite-rich, mantle source. This suggests recycled components contribute to die lidiological and geochemical heterogeneity in the mantle beneath the Celebes Sea.
查看更多>>摘要：Constraining the growth and evolution of die continental crust provides insight into the secular change of tectono-magmatic processes dirough time. Radiogenic isotope pairs such as ~(176)Lu/~(176)Hf are commonly used to track the growth rates of the continental crust. Previous studies utilizing global compilations of Lu—Hf isotopes of zircon to query continental growth through time. However, we show that sample clustering and spatial bias lead to erroneous interpretations and casts doubt as to the representativeness of global zircon compilations. To I ameliorate this issue, we apply a geographic declustering algorithm to remove the spatial bias in the nonrep-resentative dataset. A more representative declustered dataset indicates that the continental crust grew much faster tiian previous assessments have proposed. We further speculate that the transition to modern-style tec-I tonics (as heralded by the appearance of low temperature-high pressure metamorphism) was accompanied by little to no net continental growth and propose tiiat continental growtii may be a feature of pre-plate tectonic magmatic processes.
NSTL