The differentiation and rates of generation of the continental crust

Abstract

A new approach is developed to evaluate the rates of crust generation and hence the quantities of incompatible elements processed through the continental crust over the last 4 Ga. This relies more on minor and trace elements, and residence times in the upper crust and less on radiogenic isotopes since the latter constrain the stabilisation of continental crust rather than the rates of crust generation. In this model, the composition of new material added to the continental crust is similar to estimates of the average lower continental crust. The median composition of granitic magmas with Eu/Eu* = 0.7 is strikingly similar to that of the average upper crust and, in the simplest model, this represents ∼ 14% melting or 86% fractional crystallisation of new crust. For an upper crust of 12.5 km thickness, there would be 77 km of complementary residue, for which there is scant geological evidence. It is therefore inferred that the residence times of elements in the lower crust is much less than in the upper crust. The annual flux of material into the upper crust can be inferred from its volume and the residence times of elements in the upper crust. A maximum value of the latter is provided by the model Nd age of the upper crust of 2 Ga, indicating that the average rates of crust generation are in excess of six times those in the recent geologic past and two to three times greater than the rates inferred from radiogenic isotopes. Over 4 Ga more than half the K, and one quarter of the Li, in the silicate Earth may therefore have been processed through the continental crust.