December 2014 LIP of the Month

Printer-friendly versionPrinter-friendly version

December 2014 LIP of the Month

Magmatic pathways in Large Igneous Provinces: Initial thoughts on a research framework

Richard E. Ernst

Department of Earth Sciences, Carleton University, Ottawa, Ontario, Canada, K1T 3Y2;


Based on Ernst (2014a, b)



With our increasing understanding of Large Igneous Provinces (LIPs) and their expression through time, it is timely for system-wide approach to LIP studies.  This is enabled by an integration of the Phanerozoic LIP record (in which the flood basalt component is dominant), with the Proterozoic LIPs (in which erosion has exposed the plumbing system of dykes, sills and layered intrusions), with the deeper components (subcrustal magmatic underplating and sub-lithospheric channelling), and with parallel studies of often-affiliated silicic, carbonatite and kimberlite magmatism. Such a LIP-system integration will have obvious benefits for understanding all aspects of LIPs including  their origin, and will also have a practical benefit in providing constraints for the location of the minute ‘sweet spots’ of Ni-Cu-PGE mineralization within the broad LIP events that are the most significant host of such ore deposits (e.g. Ernst and Jowitt, 2013).

A framework for such system-LIP investigation is considered below and includes the following themes (Ernst, 2014a,b):  1) characterizing the pattern of magma ingress into the lithosphere from sublithospheric sources. 2)  tracing magma pathways (and interactions with host rocks) within the crust, and also  the links with surface flows. 3) integrating the implications from all parts of the LIP system: mafic-ultramafic, carbonatite/kimberlite and silicic, and  4)  techniques for ‘mapping’ the magma pathways.


Pattern of magma ingress into the lithosphere from underlying sources

LIP magmas can be generated in the underlying asthenosphere due to a thermal or compositional anomaly in an intraplate setting, and melting of suitably metasomatised lithosphere can also be important. A mantle plume model is frequently applied as the ultimate driver, but there is controversy on this point and non-plume aspects may be important in some cases (e.g., Ernst, 2014a; Foulger, 2010).  The anomalous mantle (e.g. mantle plume) can move and spread substantially along the base the lithosphere leading to potential ingress into the lithosphere over a large area and potentially far separated from any plume center (Fig. 1).

Figure 1: Spreading of a mantle plume beneath the lithospheric. a)  Lateral flow of plume material beneath a continent. It arrives beneath a region of thick lithosphere and then moves upwards across a passive margin, locally ponding beneath thin lithosphere of an old rift, and then flowing  toward the ridge axis.(after Sleep, 2006).  B) Spreading of a plume beneath Laurentia at c. 1100 Ma and potentially linked to both the Keweenawan LIP of the Mid-Continent region and the Southwest Laurentia LIP. After Bright et al. (2014).

Lithospheric entry can occur in a number of ways (Fig. 2), such as from: (a) widespread sources above the thermal anomaly (e.g. mantle plume head;  cf. White and McKenzie, 1989); (b) centrally located conduits near the plume center from which magma is distributed vertically and laterally in the lithosphere (Ernst, 2014a); (c) vertical access along planar zones of triple-junction rifting; (d) initial concentration at the plume center, followed by distribution along sublithospheric channels (e.g., Oyarzun et al., 1997; Ebinger and Sleep, 1998; Duggen et al., 2009; Fig. 1); (e) vertical access along translithospheric fracture zones (e.g. Begg et al., 2009, 2010) and (f) widespread points of ingress from a diapirically destabilized asthenospheric thermal boundary layer (Geoffroy et al., 2007). The contribution of each of these styles of injection into the lithosphere may vary from one LIP to another.

Figure 2: Patterns by which magma enters the lithosphere from below. Details in the text. After Fig. 5.26 in Ernst (2014a).

Pattern of magma distribution within the crust

An important research frontier is how LIP magma, after entering the lithosphere, is then distributed within the lithosphere. Transport within the mantle portion of the lithosphere is very poorly understood, and presumably involves some component of movement as dykes, and sills. Transport in narrow vertical conduits is also possible (Geoffroy et al., 2007).  The pattern of LIP distribution within the crustal portion of the lithosphere is better understood (e.g. Ernst, 2014a), and starts with significant magmatic underplating at the base of the crust. This is important as a source for magma being emplaced into the crust and also a source of heat for partial melting of lower crust to produce Silicic LIPs (e.g. Bryan, 2007; Bryan and Ferrari, 2013).


Lateral distribution of mafic magmatism

Upon ascent in the lithosphere, magma is distributed vertically and laterally within the crust as dykes and sills, with the involvement of staging magma chambers, and potentially also as vertical conduits (e.g. Ernst, 2014a,b; Geoffroy et al., 2007). Most dramatically, giant radiating and circumferential dyke swarms can have radii of >1000 km and >250 km, respectively, centered on the plume center (Figs. 3 and 4). Dolerite sills may also be laterally emplaced for great distances. Extrusive magmatism can be edifice-fed, or more typically, fissure- (dyke-) fed, and individual basalt flows can extend for 100s (e.g. Hooper, 1997) to perhaps even up to 1000 km (Self et al., 2008). Additional constraints on the plumbing system pathways are provided by modelling of magma transport in dykes (e.g. Rivalta et al., 2015) and the capacity of magma chambers to produce dykes and sills (e.g. Grosfils et al., 2014).

Figure 3: Giant Dyke Swarms:  a) Radiating: example 1270 Ma Mackenzie swarm of Canadian Shield, b) Circumferential example: Lake Victoria c.1370 Ma Circumferential swarm. Modified after Makitie et al. (2014).

Figure 4: Role of lateral emplacement of dykes in feeding distal sills and volcanics. After Figure 5.27 in Ernst (2014a).

As a consequence of lateral flow in dykes, lava flows and sills can be emplaced at great distances (i.e., >1000 km) from a mantle plume center (e.g. Baragar et al., 1996; Ernst, 2014a), and up to hundreds of km from local magmatic centres (e.g. Geoffroy et al., 2007):  For instance, laterally propagating dykes that intersect a sedimentary basin can continue into the basin as sills. A prominent example is the Nipissing sills of the Huronian sedimentary basin of southern Canada which are inferred to have been laterally fed via a radiating dyke swarm whose focus (and inferred plume centre) is located 1300 km to the northeast on the edge of the Superior craton (Fig. 5; Buchan et al., 1998). Similarly, local lava flows can be formed when the upper edge of a dyke intersects the paleosurface; for example, 200 Ma CAMP dykes fed lava flows located in the Newark rift basins of the eastern USA (e.g. Hill, 1991).

Figure 5:  Nipissing sill province of the 2220–2210 Ma Ungava-Nipissing LIP. (a) Regional map showing radiating dyke swarm of Ungava-Nipissing LIP. (b) Distribution of Nipissing sills and host Huronian sediments. After Fig. 5.11 in Ernst (2014a)

The main locus of magmatism can shift during a LIP event, due to plate movement over an underlying plume (e.g. as observed in the Deccan LIP owing to northward movement of India over the Deccan plume (e.g. Saunders et al., 2007). High-Mg magmas are preferentially concentrated above the plume-center region (e.g. Campbell, 2001).


Units emplaced vertically and integrating as a system

While lateral redistribution of mafic magma is important, at the same time it is useful to recognize that some components (units) associated with a LIP are essentially vertically emplaced. This is certainly true for carbonatites and kimberlites, and so their mantle source area lies directly underneath (Ernst and Bell, 2010; Ernst, 2014a). For instance, kimberlites associated with the Deccan LIP are found at the east end of the Indian subcontinent under the Bastar craton, confirming the presence of the Deccan plume extending further east than previously recognized (Chalapathi Rao et al., 2011). Similarly, silicic magmatism (given its buoyancy) should also essentially rise vertically and so the lower crust that is involved in melting should directly underlie the observed silicic magmatism. Furthermore, the distribution of silicic magmatism can be used to identify regions of underlying fusible lower crust (possibly linked with a prior subduction event).


Linking of ocean basin flood basalts with oceanic plateaus

Given their poor accessibility (except potentially via drilling and geophysics) the plumbing system of oceanic LIPs is poorly understand; one speculative aspect is considered here. Oceanic plateaus (e.g. Ontong Java, Kerguelen) form in the deep-ocean basins as broad, more or less flat-topped plateaus lying 2000 m or more above the surrounding seafloor, whereas the ocean-basin flood basalts (e.g. Nauru) are extensive submarine lava flows accumulating at abyssal depths in the ocean (e.g. Coffin and Eldholm, 1994, 2001). In some cases oceanic plateaus and nearby ocean basin flood basalts are the same age and therefore likely genetically-related. As an example, the ocean-basin flood basalts (Nauru, East Mariana, and possibly Pigafetta) in the vicinity of the Ontong Java oceanic plateau (OJP) have ages similar to the OJP. This suggested to Ingle and Coffin (2004) that these oceanic flood basalts are linked to the OJP, and the term “Greater Ontong Java” LIP was applied to this broader region of magmatism. The mechanism for feeding these Nauru basin flood basalts from the presumed locus of mantle-plume activity centered under the OJP has not been identified, but must have involved flow of lava from the OJP or via laterally emplaced dykes, sills, or sublithospheric channeling.


Methods of determining pathways

Here I consider some elements in the ‘toolkit’ for defining magma pathways in LIP systems: geochemistry, paleomagnetism, anisotropy of magnetic susceptibility (AMS) and geochronology:



At a broad scale, the magmatism of a LIP can be divided into distinct compositional groups, such as high-Ti vs. low-Ti (e.g. Ernst, 2014a) and representing fundamentally different magmatic source areas and subsequent pathways. There is also more subtle geochemical differences that can allow tracking of magmatic batches that have experienced more limited differentiation.   For instance, data from the Golden Valley Sill Complex (of the Karoo LIP of southern Africa) shows that the various adjacent sills do not necessarily belong to the same magma batch, but can be grouped into sub-batches  (Fig. 6) each representing different aliquots of magma expelled from a  magma chamber(s) that is progressively becoming more crustally contaminated.

Figure 6: Variations in initial Sr–Nd isotope compositions for c. 183 Ma Karoo dykes and sills in the Golden Valley area. Geochemistry data can allow tracking of the different magma pulses through the plumbing system. GVS = Golden Valley sill; GS = Glen sill; HS = Harmony sill; MS = Morning Sun sill; L1 =L1 sill; GV dyke = Golden Valley dyke. The c. 100-km-long, up to 30-m-wide Cradock dyke is located 70 km west of the Golden Valley Sill Complex. Sills that were not part of this study are shown outlined with a white interior. Tick marks on the model curve are 5% increments of crustal contamination. Modified from Neumann et al. (2011).

For the Karoo province as a whole, Neumann et al. (2011) provide an illustration of the complexity of the plumbing system that allows magma to evolve in stages at various depths (Fig. 7). Neumann et al. (2011) propose that the primary melts  for the Karoo were derived from an asthenospheric source mantle and had acquired a weak subduction signature (relative depletion in Nb–Ta, mildly enriched Sr–Nd isotopic ratios) through interaction with metasomatized lithospheric mantle. In the deep crust, the melt has to assimilate about 10% of granulites with strong arc-type geochemical signatures. During and/or after intrusion into the sedimentary rocks of the Karoo basin, the magmas underwent a second stage of fractional crystallization (50–60%) and local contamination by sedimentary host rocks.

Figure 7:  Schematic presentation of the ascent through the lithosphere of the melts that gave rise to the dolerites in the GVSC (Golden Valley Sill Complex) and drill cores, and the processes believed to have taken place at different depths and in different areas. Shading variations between the intrusions in the deep crust indicate different degrees of contamination and fractional crystallization (Stage 2); vertical shading variations in the shallow sills indicate a new stage of fractional crystallization locally combined with contamination that reflects the local sedimentary country-rocks (Stage 3). E = Ecca Group; B = Beaufort Group; S = Stormberg Group. Modified after Neumann et al. (2011).

Paleomagnetism and Geochemistry

LIP-related swarms tend to consist of dykes with average widths of >10 m and long lateral extents (up to > 2000 km), and these can exhibit remarkable along-dyke compositional and paleomagnetic consistency (e.g. Ernst, 2014a). For swarms in which the individual dykes are widely spaced and easily tracked, it has been shown that each dyke has consistent ratios of incompatible elements along strike for up to many hundreds of kilometers, and which can be distinct from the ratios in an adjacent co-extensive dyke. The inference is that each dyke represents a single magmatic pulse expelled from a differentiating magmatic chamber located along strike or at depth. After a period of time, a second dyke could be injected from this chamber. During the time gap between the two dyke pulses, the magma in the chamber will have evolved and so the second dyke pulse can have a composition distinct from the first pulse

Additional support for this model comes from complementary paleomagnetic studies (Fig. 8) that show that each such individual dyke can have a paleomagnetic direction which is consistent along its strike length, but which can be different from the paleomagnetic direction that characterizes an adjacent dyke. Such paleomagnetic differences are explained by a time gap between the emplacement of each dyke, consistent with secular magnetic variation on the scale of at least hundreds of years.

Figure 8:  Consistency in paleomagnetic data observed along individual dykes of the 1.07 Ga Lac Esprit swarm of the Superior craton  indicating that individual dykes each represent a distinct magma pulse and that there is a time gap between the emplacement of different dykes during  which paleomagnetic secular variation occurs. After Fig. 10.20 in Ernst (2014a).

Measurements of flow direction

Other types of data can reveal the flow direction in individual dolerite sills, dykes and volcanics. For instance, anisotropy of magnetic susceptibility (AMS ) data (Polteau et al., 2008) and seismic data (e.g. Hansen and Cartwright, 2006; Thomson, 2007) suggest that the direction of magma flow is outward and upward in saucer-shaped sill complexes (e.g., the Golden Valley area of the Karoo system) and that the outer parts of the saucers are fed from the inner sills.  

However, in their AMS study of the Nipissing sills (Fig. 5), Palmer et al. (2007) identified consistent north–northwest to south–southeast flow trajectories on the east and west sides of a saucer-shaped sill. In this Nipissing sill case, and in contrast to the Karoo example above, the AMS pattern was not consistent with outward radial injection; instead, it was more consistent with lateral feeding of the sills from the side (north or south side).  This apparent difference between emplacement styles in these two sill provinces (Karoo and Nipissing) illustrates the need for evaluation of flow patterns on a case by case basis. The same AMS technique was applied the Mackenzie dyke swarm (Fig. 3a) to  infer vertical emplacement within a few hundred km of the plume center and lateral emplacement at all distances beyond up to 2000 km away from the plume centre, and consistent with geochemistry variation in the swarm (e.g. Baragar et al., 1996).



With improvement in U-Pb dating precision < 1 million years for mafic-ultramafic magmatism, there is the potential for using such super-precise U-Pb dating for tracking magma batches. For instance, the high precision dating of layered intrusions such as the Bushveld and Stillwater by Scoates and Wall (2014) and Wall et al. (2014) reveals the relative timing of the different pulses that compose these layered intrusions.  Similarly, high precision dating of the Siberian Trap LIP (Burgess and Bowring, 2014) is showing that most of this LIP was emplaced in less than 600,000, an age resolution that is close to allowing dating of the geochemical distinct sequences and their correlation between different regions of the Siberian Trap LIP. 



This LIP of the Month represents some initial thoughts on a framework toward a systematic study of LIP events to decipher magma pathways: from sublithospheric and lithospheric source areas and throughout the lithospheric mantle and crustal profile.



Baragar, W.R.A., Ernst, R.E., Hulbert, L., & Peterson, T. (1996). Longitudinal petrochemical variation in the Mackenzie dyke swarm, northwestern Canadian Shield. Journal of Petrology, 37: 317–359.

Begg, G.C., Griffin, W.L., Natapov, L.M., et al. (2009). The lithospheric architecture of Africa: seismic tomography, mantle petrology, and tectonic evolution. Geosphere, 5: 23–50.

Begg, G.C., Hronsky, J.A.M., Arndt, N.T., et al. (2010). Lithospheric, cratonic, and geodynamic setting of Ni–Cu–PGE sulfide deposits. Economic Geology, 105: 1057–1070.

Bright, R.M., Amato, J.M., Denyszyn, S.W., & Ernst, R.E. (2014). U–Pb geochronology of 1.1 Ga diabase in the southwestern United States: testing models for the origin of a post-Grenville Large Igneous Province. Lithosphere, 6, 135–156.

Bryan, S. (2007). Silicic large igneous provinces. Episodes, 30: 20–31.

Bryan, S.E. & Ferrari, L. (2013). Large Igneous Provinces and Silicic Large IgneousProvinces: progress in our understanding over the last 25 years. Geological Society of America Bulletin, 125: 1053–1078.

Buchan, K.L., Mortensen, J.K., Card, K.D., & Percival, J.A. (1998). Paleomagnetism and U–Pb geochronology of diabase dyke swarms of Minto Block, Superior Province, Quebec, Canada. Canadian Journal of Earth Sciences, 35: 1054–1069.

Burgess, S.D. and Bowring, S.A. (2014). Using U/Pb geochronology to determine the timing and tempo of the Siberian traps Large Igneous Province magmatism. Abstract 260-8, Geological Society of American Annual Meeting October 19-22, Vancouver, Canada.

Campbell, I.H. (2001). Identification of ancient mantle plumes. In Ernst, R.E., Buchan, K.L. (eds.), Mantle Plumes: Their Identification through Time. Boulder, CO: Geological Society of America, Special Paper 352, pp. 5–21.

Chalapathi Rao, N.V. Lehmann, B., Mainkar, D., & Belyatsky, B. (2011). Petrogenesis of the end-Cretaceous diamondiferous Behradih orangeite pipe: implication for mantle plume–lithosphere interaction in the Bastar craton, Central India. Contributions to Mineralogy and Petrology, 161: 721–742.

Coffin, M.F. & Eldholm, O. (1994). Large igneous provinces: crustal structure, dimensions, and external consequences. Review of Geophysics, 32: 1–36.

Coffin, M.F. & Eldholm, O. (2001). Large igneous provinces: progenitors of some ophiolites? In Ernst, R.E. & Buchan, K.L. (eds.), Mantle Plumes: Their Identification through Time. Boulder, CO: Geological Society of America Special Paper 352, pp. 59–70.

Duggen, S., Hoernle, K.A., Hauff, F., et al. (2009). Flow of Canary mantle plume material through a subcontinental lithospheric corridor beneath Africa to the Mediterranean. Geology, 37: 283–286.

Ebinger, C.J. & Sleep, N.H. (1998). Cenozoic magmatism throughout East Africa resulting from impact of a single plume. Nature, 395: 788–791.

Ernst, R.E. (2014a) Large Igneous Provinces. Cambridge University Press, 653 p.

Ernst, R.E. (2014b) Magma pathways in Large Igneous Provinces.   Abstract V54A-02, American Geophysical Union (AGU) Fall Meeting, San Francisco, December 15-19, 2014.

Ernst, R.E. & Bell, K. (2010). Large Igneous Provinces (LIPs) and carbonatites. Mineralogy and Petrology, 98: 55–76.

Ernst, R.E. & Jowitt, S.M. (2013). Large Igneous Provinces (LIPs) and Metallogeny. Society of Economic Geologists Special Publication, 17: 17–51.

Ernst, R.E., Buchan, K.L., & Campbell, I.H. (2005). Frontiers in large igneous province research. Lithos, 79: 271–297.

Foulger, G.R. (2010). Plates vs. Plumes: A Geological Controversy. Chichester: Wiley-Blackwell.

Geoffroy, L., Aubourg, C., Callot, J.-P., & Barrat, J.-A. (2007). Mechanisms of crustal growth in large igneous provinces: the North Atlantic province as a case study. In Foulger, G.R. & Jurdy, D.M. (eds.), Plates, Plumes, and Planetary Processes. Boulder, CO: Geological Society of America, Special Paper 430, pp. 747–774.

Grosfils, E.B., McGovern, P.J., Gregg, P.M., et al. (2014). Elastic models of magma reservoir mechanics: a key tool for investigating planetary volcanism. In Platz, T., Massironi, M., Byrne, P.K., & Hiesinger, H. (eds.), Volcanism and Tectonism across the Inner Solar System. London: Geological Society, Special Publication 401, doi:10.1144/SP401.2.

Hansen, D.M., & Cartwright, J.A. (2006). Saucer shaped sill with lobate morphology revealed by 3-D seismic data: implications for resolving a shallow-level sill emplacement mechanism. Geological Society of London Journal, 163: 509–523.

Hill, R.I. (1991). Starting plumes and continental break-up. Earth and Planetary Science Letters, 104: 398–416.

Hooper, P.R. (1997). The Columbia River Flood Basalt Province: current status. AGU Geophysical Monograph, 10: 1–27.

Ingle, S. & Coffin, M.F. (2004). Impact origin for the greater Ontong Java Plateau? Earth and Planetary Science Letters, 218: 123–134.

Mäkitie, H., Data, G., Isabirye, E., et al. (2014). Petrology, geochronology and emplacement model of the giant 1.37 Ga arcuate Lake Victoria Dyke Swarm on the margin of a large igneous province in eastern Africa. Journal of African Earth Sciences, 97: 273–296.

Neumann, E.-R., Svensen, H., Galerne, C.Y., & Planke, S. (2011). Multistage evolution of dolerites in the Karoo Large Igneous Province, Central South Africa. Journal of Petrology, 52: 959–984.

Oyarzun, R., Doblas, M., López-Ruiz, J., & Cebra, J.M. (1997). Opening of the central Atlantic and asymmetric mantle upwelling phenomena: implications for long-lived magmatism in western North Africa and Europe.  Geology, 25: 727–730.

Polteau, S., Ferre´ , E.C., Planke, S., Neumann, E.-R., & Chevallier, L. (2008a). How are saucer-shaped sills emplaced? Constraints from the Golden Valley Sill, South Africa. Journal of Geophysical Research, 113: B12104.

Rivalta, E., Taisne, B., Bunger, A.P., Katz, R.F. (2015). A review of mechanical models of dike propagation: Schools of thought, results and future directions. Tectonophysics, 638: 1-42.

Saunders, A.D., Jones, S.M., Morgan, L.A., et al. (2007). Regional uplift associated with continental large igneous provinces: the role of mantle plumes and the lithosphere. Chemical Geology, 241: 282–318.

Scoates, J.S. & Wall, C.J. (2014) Dating layered intrusions. Abstract 260-1, Geological Society of American Annual Meeting October 19-22, Vancouver, Canada

Self, S., Jay, A.E., Widdowson, M., & Keszthelyi, L.P. (2008). Correlation of the Deccan and Rajahmundry Trap lavas: are these the longest and largest lava flows on Earth? Journal of Volcanology and Geothermal Research, 172: 2–19.

Sleep, N.H. (2006) Mantle plumes from top to bottom. Earth-Science Reviews, 77: 231–271.

Thomson, K. (2007). Determining magma flow in sills, dykes and laccoliths and their implications for sill emplacement mechanisms. Bulletin of Volcanology, 70: 183–201.

Wall, C.J. , Scoates, J.S., Friedman, R.M., Vantongeren, J., & Mathez, E.A. (2014) Ages of the Bushveld Complex. Abstract 260-2, Geological Society of American Annual Meeting October 19-22, Vancouver, Canada

White, R.S. & McKenzie, D. (1989). Magmatism at rift zones – the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research: Solid Earth and Planets, 94: 7685–7729.