Abstract Many Paleoarchean cratons display a gradual change from early sodic tonalite–trondhjemite–granodiorite magmatism to late K-rich granitoid magmatism; the geodynamic significance of this change is debatable though. This contribution presents field, geochemical and zircon U–Pb age and Hf isotope results of four different 3.32–3.25 Ga granitoid bodies from the northern part of Singhbhum Craton to investigate their petrogenesis and role in crustal evolution. The granitoids range in composition from tonalites to trondhjemites, derived from intracrustal melting at low- to medium-pressure conditions. The source was mainly low-K mafic rock. The granitoids show intrasuite fractional crystallization. These sodic granitoids represent the last stage of granitoid magmatism in the Singhbhum Craton which formed contemporaneously with K-rich granitoids occurring in other parts of the craton. This fact suggests that, contrary to the popular notion (of only potassic granitoids), both sodic and potassic granitoids could form at the terminal phase of cratonization, implying reworking of heterogeneous (mafic to tonalite) crust. A combination of evidence from geochemical data, secular change in granitoid composition, structural pattern and rock association of the Singhbhum Craton reflects that recurring mantle plume-related mafic–ultramafic magma emplacement in an oceanic plateau setting and attendant crustal melting can explain the Paleoarchean crustal evolution pattern.

Abstract Field and geochemical studies combined with laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) zircon U–Pb dating set important constraints on the timing and petrogenesis of volcanic rocks of the Neoarchaean Kadiri greenstone belt and the mechanism of crust formation in the eastern Dharwar craton (EDC). The volcanic rocks are divided into three suites: tholeiitic basalts, calc-alkaline high-Mg# andesites and dominant dacites–rhyolites. The basalts (pillowed in places) show flat rare earth element (REE) and primordial mantle-normalized trace element patterns, but have minor negative Nb and Ta anomalies. They are interpreted as mantle plume-related oceanic plateau basalts whose source contained minor continental crustal input. The andesites are characterized by high Mg# (0.66–0.52), Cr and Ni, with depletion of high-field strength elements (HFSE) and enrichment of light REE (LREE) and large-ion lithophile elements (LILE). They were probably derived from a metasomatized mantle wedge overlying a subducted slab in a continental margin subduction zone. The dacites–rhyolites are silicic rocks (SiO 2 =61–72 wt%) with low Cr and Ni, K 2 O/Na 2 O mostly 0.5–1.1, highly fractionated REE patterns, enrichments of LILE and distinctly negative HFSE anomalies. One rhyolite sample yielded a zircon U–Pb age of 2353±32 Ma. This suite is similar to potassic adakites and is explained as the product of deep melting of thickened crust in the arc with a significant older crustal component. Collision between a continental margin arc with an oceanic plateau followed by slab break-off, upwelling of hot asthenosphere and extensive crustal reworking in a sustained compressional regime is proposed for the geodynamic evolution of the area. This is in corroboration with the scenario of EDC as a Neoarchaean hot orogen as suggested recently by some workers. Supplementary material: Details of whole-rock major and trace element determination, Nd isotope analysis and zircon U–Pb dating and trace element analysis, the geographical coordinates of the samples and the values of the international rock standards analysed are available at http://www.geolsoc.org.uk/SUP18660