CAN SEISMOLOGY TELL US ANYTHING ABOUT CONVECTION IN THE MANTLE? Jean-Paul Montagner Laboratoire de Sismologie !nstitutde Physique du Globe Paris, France Abstract. The understandingof mantle convection is one of the mostpuzzling problemsof moderngeophys- ics. Among the differentapproaches usedby geophys- icists to investigate mantle convection, seismic tomog- raphy is the only one able to visualize, at the same time, temperature, petrological anomalies, and flow directions from seismic velocity and anisotropy heter~ ogeneities. In order to enable a comparison with other geophysical observables, tomographicmodels are ex- panded into spherical harmonics. Most tomographic models agreethat down to 300-400 km, deep structure is closely related to plate tectonics and continental distribution. Its corresponding spectral content regu- larly decreases with decreasing wavelength. At greater depthsin the transition zone, degree 2 and (to a lesser extent) degree6 distributions becomepredominant. A degree2 pattern is also present in the lower mantle and is strongly correlated with the geoid but offset with respect to the degree 2 pattern of the transition zone. A simple flow pattern with two upgoing and two down- goinglarge-scaleflows can be invoked to simply ex- plain the predominanceof degrees2 and 6 for seismic velocity and degree 4 for radial anisotropy. Therefore below the apparent complexity of plate tectonics, it turnsout that mantle convectionis surprisingly simply organized in the transition zone. Between 400 and 1000 km, these large-scale flows are not independent from the circulation in the first 400 km but are connected to some of the most tectonically active zones (fast ridges and slabs). It is alsosuggested that the degree 6 which seems to be a marker of the hotspotdistribution is not independent of the deep degree 2 but might be the consequence of this simple flow pattern. The good correlation between seismictomographydegree6 and hotspot degree 6 favors an origin at depth of hotspots in the transition zone. Generally, the mantle cannotbe divided into independentconvecting cells but is char- acterized by imbricated convection, where different scales coexistand where exchangeof matter is possi- ble. Therefore seismic tomographyis able to provide very strong constraints on possible models of mantle convection, but many features are still unexplained. Only very long spatial wavelengths are well resolved sofar, and a complete understanding of mantle dynam- ics necessitates relating the different scalespresentin convective processes. 1. INTRODUCTION Thermal convectionin the Earth has been recog- nized for a long time [Holmes, 1928;Pekeris, 1935]as the cause of dynamic processes such as mountain building, volcanic and seismicactivity, and, more re- cently, of seafloor spreading. With the adventof plate tectonics, it has becomepossible to describequantita- tively the motions at the surface of the Earth. The plates, considered rigid bodies, are the surface mani- festationof deep-seated movements related to mantle convection [Turcotte and Oxburgh, 1967]. This evi- dence constitutes the first manifestation of pecularities of mantleconvection, for it demonstrates largevaria- tions in the rheologicalproperties of Earth material between rigid plates and underlying mantle. Mean- while, hotspots were identified [Wilson, 1963] and their fixity with respect to plate motions demonstrated [Wilson,1965; Morgan, 1971]. There is a good consen- sus on the origin of plumes in a thermal boundary layer, but the depthof the origin of hotspots is still an openquestion. The rheology of the lithosphere and the existence of hotspots are two simpleexamples which illustrate the fact that mantle convection differs in many respects from "classical" convection (such as Benard convection) observed in usual viscous fluids. It involvesa large rangeof scales,from the large scaleof the Pacific Ocean to the small one of hotspots. Different approacheshave been used to character- ize and understand mantle convection. So far, none of them can provide a general and consistent explanation of simplefeatures such as the shapeand geometryof plates, the origin of hotspots, the number of layers in the mantle, etc. In that context the determination of three-dimensional modelsof seismic parametersof the mantle (also coined tomographicmodels) can be con- sideredas a pragmatic or naturalist approachwhich consists of imaging the structure of convection in or- Copyright 1994 by the American Geophysical Union. 8755-1209/94/94 R G-00099$15.00 ß 115 ß Reviews of Geophysics, 32, 2 / May 1994 pages115-137 Papernumber 94RG00099