COMPARING GLOBAL AND REGIONAL CRUSTAL MODELS: EXAMPLES FROM EUROPE AND SIBERIA

We use a new regional crustal database to compare regional and global crustal models for most of Eurasia. The area of interest covers about 1/4 of the globe and encompasses most of Europe, the Ural mountains, West Siberia basin, and the Siberian craton. Our new crustal database is compilation "from scratch" of all available seismic data for the region, acquired and interpreted from the late 1960-ies until present. It includes results of seismic reflection, refraction and receiver functions studies, but excludes unreliable constraints, such as based on seismic interpolations, gravity modelling or tectonic similarities. The new database comprises detailed and reliable information on the seismic structure of the crust for most of the tectonic structures of the region and allows for examination of spatial correlations with tectonic and geological structures. We shows that several widely accepted hypotheses about crustal structure of the continents are not supported by the seismic data for Europe and Siberia. (1) Contrary to the conclusions of Durrheim and Mooney (1991), the Archean crust in Europe and Siberia is similar to the Proterozoic crust both in thickness and in structure. In both types of crustal provinces, the crust is 40-45 km thick (locally >50 km) and has a thick (10-25 km thick) high-velocity lower crust. (2) Global averages for structure of the continental crust of different tectonic types (Christensen and Mooney, 1995) are not confirmed by seismic data from Europe and Siberia. Although the "averages" for Eurasia are similar to the earlier reported global averages, there is not a single tectonic structure in Europe and Siberia that matches "average" pattern neither in crustal thickness nor in average crustal Vp velocity. (3) Regional seismic data averaged on 2x2 deg or 5x5 deg grid are in striking contrast with global crustal models CRUST5.1 and CRUST2.0. While the differences in crustal thickness are 5 to 10 km everywhere, significant differences in the crustal velocities and thickness of individual crustal layers question representativeness of global crustal models for many continental regions. It is clear that an improved crustal model based on a high-quality compilation of seismic data is needed.