Roadmap | The review process | Home
   LIP classification II

The distribution of LIP events, both mafic and felsic, through time.

Proposed Revision to Large Igneous Province Classification

Scott Bryan1, & Richard Ernst2

1School of Earth Sciences & Geography, Kingston University, Penrhyn Rd, Kingston Upon Thames, Surrey KT1 2EE United Kingdom

2Ernst Geosciences, 43 Margrave Ave Ottawa, Ontario K1T 3Y2 Canada


This webpage has been stimulated by the paper Sheth, H.C., ‘Large Igneous Provinces (LIPs)’: Definition, recommended terminology, and a hierarchical classification, Earth-Science Reviews 85 (2007) 117–124.

Click here for a webpage summary of this paper.

Click here to go to a general Discussion of this subject.


The term “Large Igneous Province” (LIP) was initially proposed by Coffin & Eldholm (1994) to represent a variety of mafic igneous provinces with areal extents >0.1 Mkm2 that were “massive crustal emplacements of predominantly mafic (Mg and Fe-rich) extrusive and intrusive rock, and originated via processes other than ‘normal’ seafloor spreading”. The initial database upon which the term LIP was defined relied almost exclusively on the relatively well-preserved Mesozoic and Cenozoic record that comprised continental flood basalt provinces, volcanic passive margins, oceanic plateaus, submarine ridges, seamount groups and ocean basin flood basalts (e.g., Coffin & Eldholm, 1994, 2005). Since the first categorisation of LIPs, substantial progress has been made in extending the LIP record back to the Paleozoic, Proterozoic and Archean (Ernst & Buchan, 1997; 2001, 2003; Tomlinson & Condie, 2001; Arndt et al., 2001; Isley & Abbott, 2002). For many ancient LIPs, where much or all the volcanic component of the LIP has been lost to erosion, definition has to be based on the areal extent and volume of intrusive rock (e.g., giant continental dyke swarms, sills, layered intrusions), which is the exposed plumbing system to the province. In addition, it has additionally been recognised that LIPs are also massive crustal emplacements of predominantly silicic (>65 wt% SiO2) extrusive and intrusive rock that have originated via processes other than ‘normal’ seafloor spreading or subduction (see Bryan et al., 2002). Consequently, the increasing realisation that LIPs are more varied in character, age and composition than first defined has prompted others to revise and broaden the original definition of LIP (Sheth, 2007; click here for a webpage summary of this paper).

LIP Classification

Our contribution to LIP classification (Figure 1) has been stimulated by the newly proposed classification scheme of Sheth (2007). Like Sheth (2007), we also consider LIP to be a broad category and it must encompass a greater variety of igneous provinces than previously considered by Coffin & Eldholm (1994; 2005). However, in contrast to Sheth (2007) who proposes a minimum areal extent of 0.05 Mkm2, we argue that the original size definition of at least >0.1 Mkm2 be retained; this is the minimum areal extent for flood basalt events based on the initial work of Coffin & Eldholm (1994). Studies of the areal extents of the exposed plumbing systems and intrusive provinces of LIPs are also consistent with this (e.g., Yale & Carpenter, 1998; Marzoli et al., 1999; Ernst et al., 2005). Major regional continental dyke swarms for example, are >300 km in length (Ernst & Buchan, 1997) and typically have areal extents >90,000 km2; this approaches the minimum areal extent considered for the continental flood basalt provinces. Similarly, Silicic LIPs have dimensions well in excess of 0.1M km2 (all are >0.5 Mkm2) and igneous volumes >0.25 Mkm3 (Bryan et al., 2002). The review of several classic Cenozoic-Mesozoic LIPs by Courtillot & Renne (2003) concluded that the minimum dimensions of LIPs should be >1 Mkm2. Most LIPs have areal extents and eruptive and/or intrusive volumes well in excess of 1 Mkm2 and 1 MKm3, respectively, and thus are anomalous events emplacing tremendous volumes of magma throughout the crustal profile and at the Earth’s surface. We therefore conclude that LIPs should be defined as having the minimum dimensions of >0.1Mkm2 and extrusive/intrusive volumes >0.1Mkm3.

Figure 1. Proposed classification of LIPs, based on the initial work of Coffin & Eldholm (1994), but incorporating recent advances in the recognition of ancient Mafic LIPs and Silicic LIPs. Representative examples of the various types of LIPs are given.

We subdivide LIPs into two compositional groupings: “Mafic LIPs” and “Silicic LIPs”. We note that it is possible to subdivide LIPs in other ways, such as by age (e.g., Archean, Proterozoic, Phanerozoic), size (e.g, Bleeker & Ernst, 2006), crustal setting (continental vs. oceanic) and whether primarily intrusive or extrusive (Sheth, 2007). The review of Bryan et al. (2002) concluded that there is a clear compositional demarcation for LIPs where Silicic LIPs have ≤10% by volume of mafic igneous rocks, and mafic LIPs, as represented by continental flood basalt provinces, have ≤10% by volume of silicic igneous rock. Importantly, in the current LIP database (Ernst & Buchan, 2001), there is an absence of LIPs with subequal proportions of mafic and silicic igneous rocks. Silicic igneous rocks do occur independently of crustal setting and as demonstrated by the recent ODP results from the Kerguelen oceanic plateau (Frey et al., 2003), should not be assumed to be absent from LIPs in oceanic settings. It is important to stress that Mafic and Silicic LIPs are compositionally bimodal, and may also show a spectrum of compositions from basalt to high-silica rhyolite, but the prefix modifier reflects the overwhelming volumetric dominance of that composition. Whereas the presence of silicic igneous rocks is an additional characteristic of Mafic LIPs, and vice versa for the Silicic LIPs, they are not required for categorization. We emphasise, however, that ultramafic magmatism is an important but usually minor component of Mafic LIPs.

Mafic and Silicic LIPs vary in terms of the proportional volumes of volcanic and intrusive rock preserved, which reflects the degree of exhumation and age of the LIP, and potentially, lithospheric conditions at time of LIP emplacement that may have promoted eruption or magma underplating/intrusion. The extrusive and intrusive components of LIPs are fundamentally related. For example, giant continental radiating dyke swarms are part of the intrusive architecture of a continental flood basalt province. For these reasons, we have not followed the approach of Sheth (2007) of subdividing LIPs into primarily volcanic or plutonic types.

Mafic LIPs can simply be subdivided into those occurring on the continents or in oceanic settings. Silicic LIPs are exclusively continental as they are produced by large-scale and prolonged crustal anatexis (Bryan et al., 2002). The products of continental Mafic and Silicic LIPs can, however, be deposited in oceanic settings (e.g., Ukstins Peate et al., 2003), but their primary sites of eruption and accumulation are on the continents.

Mafic LIPs are then further subdivided into different types based partly on the various provinces recognised and discussed by Coffin & Eldholm (1994), but updated and expanded to include those major ancient LIPs in which the dyke swarms and mafic-ultramafic intrusive provinces are dominant. A separate grouping is also made for greenstone belts of tholeiitic and komatiitic rocks that may be mafic LIPs of Archean age.

The subgroupings of Mafic LIPs also have temporal significance, as the character of mafic LIPs varies through time:

  • Greenstone belts of the tholeiite-komatiite association in the Archean
  • Proterozoic-Palaeozoic Mafic LIPs occurring as eroded flood basalts with exposed plumbing systems of dykes, sills, layered intrusions, and
  • Continental flood basalts, volcanic rifted margins and oceanic plateaus in the Mesozoic and Cainozoic

Silicic LIPs were defined by Bryan et al. (2002), but have not been subdivided further. The Mesozoic-Cainozoic examples are the best preserved, and their general architecture is as an extensive (>0.5 Mkm2), relatively flat-lying ignimbrite plateau of ≥1 km thickness (e.g., Sierra Madre Occidental). More ancient examples occur as continental caldera systems and major batholiths (e.g., Kennedy-Connors-Auburn province). Silicic LIPs are expected to have similarly extensive mid- to upper-crustal granitic batholith underpinnings and dyke swarms, and more mafic igneous underplating at lower crustal depths. The identification of deeply exhumed Silicic LIPs that are comparable to the giant continental dyke swarms and mafic-ultramafic intrusive provinces of the Mafic LIPs awaits further discovery. A provisional subdivision of Silicic LIPs is made here into those that occur on or form volcanic rifted margins, and those that likely represent failed rifts. The Permo-Carboniferous Kennedy-Connors-Auburn province of eastern Australia (Bain & Draper, 1997; Bryan et al., 2003) is a possible example of this latter type. Volcanic rifted margins can therefore be characterised as Mafic or Silicic LIPs.


last updated 3rd December, 2007