Geology Reference
In-Depth Information
Some of the observed complications may turn out to be explicable through the
effects of compositional buoyancy within plumes. Plumes have long been thought
to have a different composition than ambient mantle, and in particular to have a
higher proportion of basaltic component, because their source is inferred to contain
an accumulation of subducted oceanic crust [60, 74]. Because basalt transforms to
an eclogitic assemblage in the upper mantle, and is probably denser than ambient
mantle through much of the depth of the mantle, some compositional buoyancy
(usually negative) is to be expected [80]. Thus the thermal buoyancy of a plume
may be opposed, through much of the plume's ascent, by negative compositional
buoyancy. Indeed, there must be a limit to how much basaltic component a plume
may contain and still have a net positive buoyancy, and there may be some self-
regulation of plumes, only those below the limit being able to rise.
Recent modelling has confirmed that compositional buoyancy may considerably
complicate the dynamics of plumes. Lin and van Keken [81, 82] have shown that
entrainment of compositionally denser material in axisymmetric numerical plumes
can cause plume tail flows to vary erratically and potentially to yield multiple
major eruption episodes. Numerical and laboratory studies of three-dimensional
thermochemical plumes show plumes that are more irregular in shape and behaviour
than the classic thermal plume heads. The laboratory models of Kumagai
et al.
[83] document a spectrum of behaviour. Plume heads with relatively less heavy
component may drop some of it as they flatten under the lithosphere. Plumes with
more heavy component may stop ascending part way up the mantle, and material
with even more may remain in piles near the base of the mantle. In the numerical
examples of Farnetani and Samuel [84], much of the plume head stalls under the
transition zone but narrow upwellings break through and rise through the upper
mantle (Figure 7.18). Such models could provide an explanation for 'headless'
plume tracks and a mechanism for small plumes that some have inferred to have
arisen within the upper mantle.
These results take plume models into a new realm, with potentially a much richer
range of behaviour that may account for a greater proportion of non-plate volcanism.
However, the behaviour of thermochemical plumes is clearly complicated and its
exploration is still in its early stages, so clear conclusions either admitting or
excluding plume sources may not emerge for some time.
7.6 Summary of the plume mode
The existence of volcanic island and seamount chains terminating in isolated, active
volcanic hotspots, such as Hawaii, and surrounded by broad topographic swells
implies the existence of narrow, long-lived columns of buoyant, rising mantle
material. Morgan called these columns
mantle plumes.
The buoyancy and excess
