Slabs, hotspots, cratons and mantle convection revealed from residual seismic tomography in the upper mantle

To further investigate the fate of slabs, the structure of hotspots and the style of mantle convection, we calculate residual tomographic maps in the upper mantle by excluding from the tomography the first-order effects of conductive cooling of oceanic plates, deep cratonic 'roots', and cooling or partial melting associated with subducted lithosphere. No correlation is found between residual tomography in the upper mantle and the 30-130 Ma subduction. The good correlations between residual tomography in the transition zone (400-650 km) and 0-30 Ma subduction, at spherical harmonic degree l = 2 can be explained either by slab accumulation in the transition zone beneath some subduction zones or by the poor lateral resolution of seismic tomography. Hotspots correlate with the residual tomography in the shallow (100-400 km) and the lowermost mantle at degree l = 2. However, correlations of hotspots and seismic velocities in the middle mantle, and correlations between residual tomography at shallow depths and lowermost-mantle tomography are poor. Therefore the connection between residual tomography at shallow depths and seismic tomography in the lowermost mantle is not clear. Craton distribution correlates with distribution of hotspots and residual topography very well at degree l = 6. This might indicate that cratonic 'roots' may affect the locations of upwellings and thus modulate upper-mantle convection, or that the degree l = 6 could be intrinsic in upper-mantle convection, with cratons and and other near-surface features establishing the pattern.