Mantle layered convection

Overall, using the basic principles of this model, the basic relationships between the isotopic compositions of the atmosphere and upper mantle cannot be explained. However, the preservation of noble gas isotope heterogeneities in convectively isolated mantle layers remains appealing. 

Figure 23 Mantle potential temperature (°C) versus age (Ga) showing thermal evolution models for the upper mantle. The dashed line is a model for whole-mantle convection, and the solid line shows the trace of maximum upper-mantle temperatures in a model of transient layered convection with periodic mantle overturn (Davies, 1995, 1998). The large circles show estimates for average mantle lithosphere. Lithosphere labels are as in Figure 21.

Machetel P. and Weber P. (1991) Intermittant layered convection in a model mantle with an endothermic phase change at 670 km. Nature 350, 55-57. 

One of the most fundamental, contemporary debates about the nature of the Earth s mantle centers on the subject of mantle convection. There are two conflicting views which are commonly described as "layered convection" and "whole-mantle convection" (see Fig. 3.17). The layered convection model is championed by geochemists who prefer to see the mantle as two separate convecting layers. In this model the upper and lower mantle are geo-chemically isolated from each other and convect separately. Whole mantle convection is advocated by geophysicists, who that believe there is evidence for a significant exchange of mass between the upper and lower mantle. 

Temperature and pressure both increase with depth in the Earth and control the composition and properties of the material present at various depths. The Earth comprises a number of layers, the boundaries between which are marked by relatively abrupt compositional and density changes (Fig. 1.1). The inner core is an iron-nickel alloy, which is solid under the prevailing pressure and temperature ranges. In contrast, the outer core is molten and comprises an iron alloy, the convection currents within which are believed to drive the Earth s magnetic field. The core-mantle boundary lies at c. 2900km depth and marks the transition to rocky material above. The mantle can be divided into upper and lower parts, although the boundary is quite a broad transitional zone (c.lOOCM-OOkm depth). It behaves in a plastic, ductile fashion and supports convection cells. The upper mantle layer from c.100 to 400 km depth is called the asthenosphere, and its convection system carries the drifting continental plates. 

Planet Earth acquired an ocean early in its history, probably by 3.8 billion years before present. Most of the water is thought to have been released during the process of differentiation in which density-driven convection and cooling caused the still-molten planet to separate into layers of decreasing density, i.e., core, mantle, crust, and atmosphere. Once the crust had cooled sufficiently, gaseous water condensed to form a permanent ocean. 

To explain the imbalance, O Nions and Oxburgh (1983) and Oxburgh and O Nions (1987) proposed that a barrier, which is suggested to exist between the upper and the lower mantle from seismic observation, has trapped helium in the lower mantle and retarded the heat transport from the lower mantle to the upper mantle. O Nions et al. (1983) suggested, from a semiquantitative discussion, that delayed heat transfer from the lower mantle to the upper mantle with a time constant of about 2Ga would enhance the present heat flow by a factor of two. McKenzie and Richter (1981) made numerical calculation on a two-layered mantle convection and showed that heat transfer from the lower mantle to the upper mantle is considerably retarded to give rise to an enhancement of the present surface heat flow up to a factor of two. If the thermal barrier not only retards the heat transfer and hence enhances the present surface heat flow but also essentially prevents the 4He flux from the lower to the upper mantle, this would qualitatively explain the imbalance. If this indeed were the case, we would expect a large amount of 4He accumulation in the lower mantle. However, it is difficult to conclude such a large accumulation of 4He in the lower mantle from the currently available scarce noble gas data derived from mantle-derived materials. 

The mode of the mantle convection, layered or mantle-wide, is one of the most fundamental problems in current earth science. As we discussed earlier, 4He-heat systematics appears to suggest that the lower mantle (apart from the exact locale) is essentially isolated from the upper mantle by a barrier that impedes He migration between the layers. Other noble gas characteristics, for example much higher 40Ar/36Ar and 129Xe/130Xe in the upper mantle than in the lower mantle, also appear... 

With the exception of Davies, who favored whole-mantle convection all along, the above authors concluded that it was only the upper mantle above the 660 km seismic discontinuity that was needed to balance the continental crust. The corollary conclusion was that the deeper mantle must be in an essentially primitive, nearly undepleted state, and consequently convection in the mantle had to occur in two layers with only little exchange between these layers. These conclusions were strongly reinforced by noble gas data, especially He/ He ratios and, more recently, neon isotope data. These indicated that hotspots such as Hawaii are derived from a deep-mantle source with a more primordial, high He/" He ratio, whereas MORBs are derived from a more degassed, upper-mantle reservoir with lower He/ He ratios. The noble-gas aspects are treated in Chapter 2.06. In the present context, two points must be mentioned. Essentially all quantitative evolution models dealing with the noble gas evidence concluded that, although plumes carry... 

---