Geology Reference
In-Depth Information
an alternative to plate tectonics, as has sometimes been suggested for Venus or
for the early Earth. If plate tectonics did not operate, the mantle would have to
dispose of heat through the top thermal boundary layer in some other way. If
the lithosphere were intact, forming a 'one-plate' planet, the heat would have to
conduct through the lithosphere. If the top thermal boundary layer adopted some
other form of motion - for example, if the lithosphere were too weak to form rigid
plates - then that motion would accomplish the heat removal. The activity of plumes
is determined by the temperature difference across the bottom thermal boundary
layer, and that would not be much affected by a change in the behaviour of the top
thermal boundary layer. Thus plumes would continue their activity regardless of
how the top thermal boundary layer was behaving.
The second implication of the independence of plates and plumes is that plumes
are not the return flow of plates. If there were no heat coming from the core there
would be no plumes, but plate tectonics could operate as usual. The return flow
of the plates is a broad upwelling between subduction zones that focuses into the
mid-ocean ridges at the surface (see Figure 6.3). The return flow of plumes is a
broad and very slow downwelling between plumes. Both return flows are passive.
The idea that plumes are the return flow of plates probably comes from textbook
examples of convection, in which the flows driven by the two thermal boundary
layers are tightly coupled, so that the active upwellings superimpose on the return
flow from the active downwellings, and vice versa, as can be seen in the last panel
of Figure 6.2.
Plates do seem to influence the plume mode, even though many plumes rise under
the interiors of plates. The main evidence for this is the correlation of hotspots with
geoid highs (Figure 7.2). The geoid reflects deep variations in the density of the
mantle, and geoid lows correlate with the locations of subduction during the past
100 Myr or so, plausibly because of accumulations of old subducted lithosphere
in the deep mantle [87]. By implication, regions between subduction zones, under
the geoid highs, are regions where the deep mantle is flowing upwards in the broad
return flow from plate subduction. Plumes occur preferentially in these mantle
upwelling regions.
This correlation is understandable, because instabilities in the bottom thermal
boundary layer are most likely to occur where it is being thickened by converging
deep flow driven by subduction zones (see Figure 8.1). Equation (7.16) shows that
the growth of the Rayleigh-Taylor instability in the thermal boundary layer is faster
(smaller τ RT ) if the layer is thicker.
Plumes may also affect the plate mode of convection. Morgan originally pointed
out that the beginning of the opening of the Atlantic seemed to correlate with the
start of new plumes [44], and he actually argued that plume tails were a major
source of plate driving force. Our analysis of slab and plume buoyancy forces does
not support this possibility. However, some cases of rifting do correlate plausibly
Search WWH ::




Custom Search