Worldwide distribution of ages of the continental lithosphere derived from a global seismic tomographic model

It has long been known that continental regions of different age have different seismic properties: the older the lithosphere, the greater the velocity. Here we ask whether we can obtain a more formal relationship between seismological observations and the age of continents. The deep structure of continents has been seismically mapped for the entire Earth. S-vertical travel time delay maps were computed from velocity maps. S-delay maps (using [Shapiro, N.M., Ritzwoller, M.H., 2002. Monte-Carlo inversion of broad-band surface wave dispersion for a global shear wave velocity model of the crust and upper mantle. Geophys. J. Int. 151, 88-105]) correlate with maps show a broad correlation with the ages of surface rocks but when studied in detail, the correlation is not good. The histogram of the S delays for continents is not Gaussian but contains two maxima. From a study in North American stations, Romanowicz and Cara [Romanowicz, B.A., Cara, M., 1980. Reconsideration of the relationship between S and P station anomalies in North America. Geophys. Res. Lett. 7, 417-420] found that the ratio of S- to P-delay does not fit a linear relationship: different intercept times have to be used for positive and for negative S to P ratios. They explained this by the presence of low-velocity mantle beneath tectonically active regions but not beneath stable cratons. By imposing that the negative peak of the S-delay histogram corresponds to a major period of continent growth, a relationship between S delay and age is obtained. Using this relationship, a “seismic continental growth curve” is derived. This growth curve is similar to curves published by geochronologists in that it shows strong growth between 3 and 1 Ga. Many uncertainties remain, particularly on the exact meaning of the age of a continental region.