Faulting, flow, and the strength of the continental lithosphere

Most of our knowledge about the deformation of the continents conies from observations made at the surface or within the seismogenic layer where faulting occurs in earthquakes. Yet the bulk of the lithosphere thickness deforms by creep, which probably becomes more distributed with depth. The extent to which motions at the surface are controlled by the strength and anisotropy of the seismogenic layer, or by viscous forces on its base arising from flow beneath, is a central question in continental tectonics, affecting our approaches to modeling or visualizing the behavior of continental lithosphere. Attempts to answer this question directly, by comparing block motions at the surface with estimates of the overall velocity field, have proved difficult and inconclusive. One difficulty is in knowing whether velocity fields we observe geodetically are elastic and recoverable or truly representative of longer-term motion. Another is in understanding how finite deformation, particularly rotations about a vertical axis, is related to the instantaneous fault configurations and velocity fields we see active today. Nonetheless, new results from Greece and western North America indicate fault patterns that appear to be dominated by the relative rotation of a few rigid blocks, with rapid spatial variations in rotation rates. At least in Greece, these blocks appear to accommodate Substantial finite rotations, but the faulting on their boundaries changes with time. A recent re-assessment of continental strength profiles, based on earthquake depth distributions and gravity anomalies, suggests that the seismogenic layer may be the only significant source, of strength in the continental lithosphere, and that the upper mantle beneath the continents is relatively weak. Taken together, these observations suggest that the detailed behavior of the upper crust is indeed dominated by the strength and anisotropy of fault-bounded blocks, but that the faulting can evolve with time so as to accommodate the overall characteristics of a flow that is probably controlled by larger-scale effects, such as buoyancy forces arising from crustal thickness contrasts and forces on the edges of plates.