A cool model for the Iceland hotspot

Several primary features of the Iceland region require a posteriori adaptions of the classical plume hypothesis to explain them, which erodes confidence in this model. These include the lack of a time-progressive volcanic track and the paucity of evidence for a seismic anomaly in the lower mantle. Diverse studies suggest a mantle potential temperature anomaly beneath the region of no more than 50-100 K, which is probably insufficient for a thermally buoyant plume. We suggest an alternative model that attributes the enhanced magmatism in the Iceland region to high local mantle fertility from subducted Iapetus oceanic crust trapped in the Laurasian continental mantle lithosphere within the collision zone associated with the Caledonian suture. This crust is recycled into the melt zone locally beneath the mid-Atlantic ridge where isentropic upwelling of eclogitized crust or a crust-peridotite mixture produces excess melt. The production of anomalously large volumes of melt on this part of the spreading ridge has built a zone of thick seismic crust that traverses the whole north Atlantic. A weak, downward continuation of the seismic low-velocity zone into the transition zone between the Charlie Gibbs and Jan Mayen fracture zones may correspond to a component of partial melt, too low to be extractable, that indicates the depth extent of enhanced fusibility or volatiles resulting from the recycled crust. The Iceland region separates two contrasting oceanic tectonic regions to its north and south that may behave independently to some degree. Perhaps as a result of this, it has persistently been characterized by complex and unstable tectonics involving spreading about a parallel pair of ridges, intervening microplates, ridge migrations, and local variations in the spreading direction. These tectonic complexities can explain a number of primary features observed on land in Iceland. A captured microplate that may contain oceanic crust up to similar to30 m.y. old underlies central Iceland submerged beneath younger lavas. This may account for local thickening of the seismic crust to similar to40 kin there. Eastward-widening, fan-shaped extension about a west-cast zone that traverses central Iceland culminates in northwest Vatajokull and may cause the enhanced volcanism there that is traditionally assumed to indicate the center of a plume. The general locus of spreading has not migrated east as is often suggested in support of an eastward-migrating plume model. The model suggested here attributes the Iceland melting anomaly to structures and processes related to plate tectonics that are sourced in the shallow upper mantle. Similar models may explain other hotspots. Such models suggest a simplifying view of mantle convection since they require only a single mode of convection, that associated with plate tectonics, and not an additional second independent mode associated with deep mantle plumes. (C) 2004 Elsevier B.V. All rights reserved.