Geochemical evolution of granitic continental crust: A comparative study among Paleoproterozoic and Phanerozoic representatives

概要

The formation of the Earth’s continental crust and its evolution through time is elusive. Throughout the Earth’s history, Paleoproterozoic and Phanerozoic eras are of the im- portance in terms times of enormous production of granitoids. In order to understand the growth pattern of the two representative eras, the author investigated the spacial- temporal impacts of mantle-derived melt adding to the preexisting continental crust based on comparative case studies. As a Paleoproterozoic representative, the Ubendian Belt– Bangweulu Block of East Africa was chosen for geochemical and geochronological recon- naissances. The area together with the Archean Tanzania Craton represents a part of the Central African Shield, which is one of the key areas of the crustal growth during the oldest supercontinent amalgamation. Although the paucity of reliable radiogenic ages and geochemical data in the continental basement terranes hampered understanding the amalgamation history of the shield, new zircon U–Pb geochronology and whole-rock trace- element geochemistry of granitoids revealed that an extensive magmatic flare-up occurred between 1.89 and 1.85 Ga in a slab-failure tectonic setting. Moreover zircons from two Paleoproterozoic metagranitoids yielded Neoproterozoic (Ediacaran) overprints at ∼570 Ma, suggesting a crustal reworking during a Neoproterozoic Pan-African orogeny. The ∼1.9 Ga metagranitoids contain zircons with negative Hf isotope compositions, εHf (t), in- dicating the existence of the enriched mantle source. Our geochemical model postulated that the Bangweulu Block and the Tanzania Craton were separated continental blocks during the amalgamation of the Central African Shield. Slab-failure type enormous con- tinental arc magmatism, which would have occurred after every collisional event, can be responsible for the each peak of global zircon U–Pb age distribution since at least Paleoproterozoic. On the other hand, the Central Asian Orogenic Belt (CAOB), as a Phanerozoic representative, shows a different growth pattern. Although the CAOB has long been considered as a venue of the high production of the newly-formed radiogenic continental crust, recent intensive studies have argued that the crustal growth rate in the CAOB is likely overestimated. In order to evaluate this issue, metagranitoids of the Mongol–Okhtosk Belt—one of the youngest segments of the CAOB— were also in-vestigated. Newly obtained U–Pb zircon ages identified three magmatic events at ∼296, ∼280, and ∼230 Ma. Geochemical features inferred that those three granitic plutons were generated along the Andean-type active continental margin where the Mongol–Okhotsk oceanic plate subducted. In addition, both S- and I-type granitoids with their volcanic equivalents, and gabbro-diorite plutons were emplaced between ∼265 and ∼250 Ma in an intra-oceanic arc setting. The precursor of the S-, and I-type granitoids were mostly likely metagraywacke, and metabasalt of old accretionary complex, respectively. Over- all the Mongol–Okhotsk Belt granitoids are characterized by positive zircon εHf (t) and whole-rock εNd(t), marking their radiogenic source. To understand the crustal growth pat- tern on a larger scale, the author made a comprehensive comparisons of the zircon εHf (t) data in literature from the Central African Shield and CAOB. The African zircon dataset clearly suggested that the radiogenic crust was added to the initial Archean cratonic core at ∼2.8 Ga. The author proposed that an intensive Paleoproterozoic crustal reworking occurred within the mobile belts surrounding Archean cratons, with the minor addition of the radiogenic newly-formed continental crust. In contrast, zircons from major orogenic segments of the CAOB show an isotopic evolution shifting from the crustal values to the depleted mantle values in descending order of time. This isotopic evolution pattern might have been attributed to continuous intra-oceanic subduction that removes the lower con- tinental crust and lithospheric mantle. Estimation on radiogenic versus crustal portions of the CAOB inferred that the Early–Middle Paleozoic time-slice consists of 20% of the reworked crust, 60% of the mixed crust, and 20% of the radiogenic crust. Moreover Late Paleozoic to Early Mesozoic time-slice was composed 20% of the reworked crust, 50% of the mixed crust, and 30% of the radiogenic newly-formed crust, which yield a 10% in- crease of the radiogenic crust. The crustal integral modeling curve in the Central African Shield suggested that the majority of the new crust formed before ∼2.8 Ga, and the growth rate is gradually decreased. During the Paleoproterozoic time, most Neoarchean crusts were largely reworked, whereas during the Phanerozoic, radiogenic crust–recently separated from the mantle recycled considerably faster than the Paleoproterozoic time, as the case of the CAOB. In short, the comparative study among Paleoproterozoic and Phanerozoic crustal growth demonstrated different evolution patterns in terms of their Hf isotopic arrays. In the case of Phanerozoic orogeny, CAOB is characterized by a step- wise growth due to repetition of relatively short-lived subduction systems, which resulted in continuous addition to accumulate new-formed continental crust to the orogenic seg- ment. By contrast, the Paleoproterozoic orogenic system in SE Tanzania incorporated protracted subduction zone resulted in internal crustal reworking or enriched lithospheric mantle. The recognition of these disparate Hf isotopic characters in two global-scale oro- genic systems of different eras offers a new insight into feedback between magma source and tectonic activity and a difference in the melt generation models during these times.