Much has been learned about Large Igneous Provinces (LIPs) and their database greatly expanded since their first formal categorization in the early 1990s. This progress provides an opportunity to review the key characteristics that distinguish LIP events from other melting events of the upper mantle, and to reassess and revise how we define LIPS. A precise definition is important to correctly recognize those LIP events with regional to global effects, and to aid in refining petrogenetic models of the origin of LIPs. We revise the definition of LIPs as follows: "Large Igneous Provinces are magmatic provinces with areal extents > 0. 1 Mkm(2), igneous volumes >0.1 Mkm(3) and maximum lifespans of similar to 50 Myr that have intraplate tectonic settings or geochemical affinities, and are characterised by igneous pulse(s) of short duration (similar to 1-5 Myr), during which a large proportion (>75%) of the total igneous volume has been emplaced." They are dominantly mafic, but also can have significant ultramafic and silicic components, and some are dominated by silicic magmatism. In this revision, seamounts, seamount groups, submarine ridges and anomalous seafloor crust are no longer considered as LIPs. Although many of these are spatially-related features post-dating a LIP event, they are constructed by long-lived melting anomalies in the mantle at lower emplacement rates, and contrast with the more transient, high magma emplacement rate characteristics of the LIP event. Many LIPs emplaced in both continental and oceanic realms, are split and rifted apart by new ridge spreading centres, which reinforce the link with mid-ocean ridges as a post-LIP event. Three new types of igneous provinces are now included in the LIP inventory, to accommodate the recognition of a greater diversity of igneous compositions, and preserved expressions of LIP events since the Archean: 1) giant diabase/dolerite continental dyke swarm, sill and mafic-ultramafic intrusion-dominated provinces; 2) Silicic LIPs; and 3) tholeiite-komatiite associations, which may be Archean examples of LIPS. A revised global distribution of LIPS for the Phanerozoic is presented. Establishing the full extent of LIPs requires well-constrained plate reconstructions, and at present, plate reconstructions for the Precambrian are poorly known. However, the possibility of reconstructing the LIP record back to and into the Archean and using this expanded LIP record to better constrain the origins and effects of LIPs is an exciting frontier, and our revised definition is a contribution to that effort. (C) 2007 Elsevier B.V. All rights reserved.
