A Geochemical Approach to Allochthonous Terranes: A Pan-African Case Study [and Discussion]

The recognition of Mesozoic and Cenozoic terranes can best be made from palaeomagnetic, structural and palaeontological studies, but older regions of continental crust require geochemical constraints to evaluate crustal growth through terrane accretion. For Precambrian shields, the pattern of Pb and Nd isotopic provinces may reveal the mechanism of crustal growth. The Afro-Arabian Shield was generated by calc-alkaline magmatism between 900 and 600 Ma ago. This example of Pan-African crustal growth underlies an area of at least 1.2 $\times $ 10$^{6}$ km$^{2}$, which may extend to 3.5 $\times $ 10$^{6}$ km$^{2}$ beneath Phanerozoic sediments and Tertiary volcanic cover. Field evidence and trace element geochemistry suggest that Pan-African tectonics began as a series of intra-oceanic island arcs that were accreted to form continental lithosphere over a period of 300 Ma. The great majority of Nd and Pb isotope ratios obtained for igneous rocks from the shield are indicative of a mantle magma source. Although many of the dismembered ophiolites cannot be identified with inter-terrane sutures in their present location, the eastern margin of the Nabitah orogenic belt is a major tectonic break that coincides with a critical boundary between Nd and Pb isotopic provinces and is marked by a linear array of ophiolite fragments across the length of the shield. Other terrane boundaries have not been identified conclusively, both because coeval island arcs can not be distinguished readily on isotopic grounds and because many ophiolites are allochthonous. However, the calculated rates of crustal growth (measured as volume of magma, extracted from the mantle per unit time) between 900 and 600 Ma are similar to those calculated for Phanerozoic terranes from the Canadian Cordillera. Such high rates in the Afro-Arabian Shield suggest that island arc terranes have accreted along a continental margin now exposed in NE Africa, together with minor continental fragments. If crustal growth rates during this time were no greater than contemporary rates, ca. 4000 km of arc length are required, which is considerably less than that responsible for crustal growth in the SW Pacific.