Hotspot tracks are age-progressive chains of volcanoes or volcanic islands and seamounts that are believed to form as tectonic plates drift over more-or-less stationary plumes of upwelling mantle rocks. Partial melting of the upwelling mantle peridotite occurs as it approaches the surface due to the reduction of pressure on the hot rocks.
- Are hotspot tracks caused by mantle plumes or lines of fractures in oceanic lithosphere?
J. Tuzo Wilson (1963) recognized that oceanic islands were generally younger the closer they were to midocean ridges. He proposed that age-progressive island chains, hotspot tracks, were produced as tectonic plates moved over stationary mantle plumes, cylindrical upwellings of hot mantle rocks. An alternative explanation (Turcotte and Oxburgh, 1978) was that hotspot tracks could be the result of fractures in the lithosphere that allowed magma to rise to the surface. However, why should stresses in the lithosphere produce fractures in the same same orientation across large plates, and why do they all develop/propogate in the same direction and distance over time? The mantle plume model does a better job of explaining the observations.
- How do mantle plumes behave - laboratory and numerical modeling of plumes
Laboratory models of plumes are created by introducing a trickle of lower density fluid into the bottom of a tank of a slightly more dense fluid. The rising lower density fluid forms a mushroom-shaped head and a narrow tail. After the mushroom head has reached the surface the trickled fluid will continue to rise through the conduit formed by the tail.
Numerical models of the mantle also show that mantle rock (solid but ductile - fluid-like) at the core mantle boundary (CMB), heated and made less dense, will rise, forming a mushroom head with a tail. When the plume head reaches the base of the lithosphere it would spread out to cover a region perhaps as large as 2000 km across. As the hot lower mantle rocks approach the surface, partial melting will occur by decompression melting, thereby producing widespread basaltic volcanic activity. After the heat dissipates from the spread mushroom head, material continues to rise through the conduit producing a steady state hotspot (a localized volcanic point).
- Large Igneous Provinces (LIPs) and hotspot tracks
Many hotspot tracks on the Earth originate in large igneous provinces. Tristan da Cunha island in the south Atlantic lies over the hotspot that produced the Parana-Etendeka flood basalts which erupted around 125 m.y. ago along the line of separation of South America and Africa. An age-progressive trail of volcanic islands and seamounts runs from the Etendeka basalts in southwestern Africa across the south Atlantic to Tristan da Cunha. Reunion Island in the southern Indian Ocean lies at the young end of a hotspot track that originates in the Deccan Traps flood basalt province of eastern India which erupted 65 m.y. ago. While most hotpots are found on ocean crust, some are also found on the continents. Yellowstone National Park (geysers & hot springs) lies at the young end of a hotspot track that formed the Snake River plain following the origination of the hotspot with the eruption of the Columbia flood basalts around 17 m.y. ago. Numerous other examples of hotspot tracks and associatiated large igneous provinces are found around the world.
- How deep is the source region for mantle plumes - upper mantle or lower mantle?
The 670 km seismic discontinuity is the seismic signature of the boundary between less dense rock of the upper mantle from the denser perovskite of the lower mantle. The old conventional model (Don Anderson and others) held that this density difference is great enough to prevent any material from the upper mantle from sinking into the lower mantle or any material from the lower mantle from rising into the upper mantle. The upper and lower mantle must cool via separate upper mantle and lower mantle convection systems. So plumes (if they exist) must originate in the upper mantle.
But a number of observations argue that plumes originate in the lower mantle.
incompatible elements: Many of the minor elements will substitute into the crystal lattice structures of the common silicate minerals in the Earth's mantle. Certain elements with high ionic charges and/or very large or very small ionic sizes have difficulty substituting into the common silicates. They are called incompatible elements (e.g. Rb, Ba, Th , Ta, Nb). When partial melting occurs in the mantle these elements prefer to go into the liquid which then rises toward the surface. The upper mantle has become depleteded in these incompatible elements as the continental crust has grown by separation of melts from the mantle. However, the lower mantle, which doesn't take part in the formation of melts for midocean ridges or volcanic arcs at subduction zones should retain all or most of its incompatible elements. Hotspot volcanoes are enriched in incompatible elements, especially the most incompatible ones, whereas midocean ridge basalts and ultramafic xenoliths known to have come from the shallow upper mantle are depleted in incompatible elements, especially the most incompatible ones. So the hotspot volcanoes tap a source of mantle rocks that have not been depleted - presumably the lower mantle.
Os isotope ratios: The platinum group elements, platinum (Pt), Rhenium, (Re), and Osmium (Os) have a strong affinity for iron and most of the Pt, Re, and Os in the Earth was concentrated in core when the Earth formed. Some isotopes of Pt and Re decay into isotopes of Os (these are called radiogenic isotopes of Os because they are generated by radioactive decay). Hotspot basalts typically have higher ratios of radiogenic to non-radiogenic Os compared to midocean ridge basalts. This could be the result of the inclusion of up to 1% core material in a mantle plume derived from the D" layer. In fact, some recent lab work indicates that iron from the core may react with mantle silicates near the core mantle boundary (CMB).
shear wave anisotropy at the CMB: Minerals with elongate crystal shape can be aligned by flow of the solid but ductile mantle. Shear waves will travel slightly faster if their oscillation direction is aligned with the flow-aligned crystal orientation. Mantle tomography studies look at seismic velocities of waves that travel through particular points in the mantle passing from all different directions (using the waves from many different earthquaks). Studies like these have been interpreted to indicate widespread horizontal alignment in the D" layer except vertical alignment beneath the central Pacific. This would indicate lateral flow of mantle material in the D" layer feeding into the vertically rising plume conduit beneath Hawaii.
mantle tomography beneath Hawaii and Iceland shows low velocity (high temperature/low density) roots extending beneath Hawaii and Iceland deep into lower mantle.
Much of the evidence is being interpreted as supporting a lower mantle, probably D" layer, origin for at least many of the plumes feeding hotspot volcanoes. But the argument continues.
- Are mantle plumes stationary with respect to one another, thereby forming a mantle plume reference frame for determining "absolute" plate motions?