Editing Flood basalt (section)

==Formation==
[[File:Plume 2.jpg|thumb|Plume model of flood basalt eruption]]
Theories of the formation of flood basalts must explain how such vast amounts of magma could be generated and erupted as lava in such short intervals of time. They must also explain the similar compositions and tectonic settings of flood basalts erupted across geologic time and the ability of flood basalt lava to travel such great distances from the eruptive fissures before solidifying.

===Generation of melt===
A tremendous amount of heat is required for so much magma to be generated in so short a time.{{sfn|Philpotts|Ague|2009|p=380}} This is widely believed to have been supplied by a [[mantle plume]] impinging on the base of the Earth's [[lithosphere]], its rigid outermost shell.<ref>{{cite journal |last1=White |first1=Robert |last2=McKenzie |first2=Dan |title=Magmatism at rift zones: The generation of volcanic continental margins and flood basalts |journal=Journal of Geophysical Research |date=1989 |volume=94 |issue=B6 |pages=7685 |doi=10.1029/JB094iB06p07685|bibcode=1989JGR....94.7685W }}</ref><ref name="Saunders2005">{{cite journal |last1=Saunders |first1=A. D. |title=Large Igneous Provinces: Origin and Environmental Consequences |journal=Elements |date=1 December 2005 |volume=1 |issue=5 |pages=259–263 |doi=10.2113/gselements.1.5.259|bibcode=2005Eleme...1..259S }}</ref>{{sfn|Philpotts|Ague|2009|p=52}} The plume consists of unusually hot mantle rock of the [[asthenosphere]], the ductile layer just below the lithosphere, that creeps upwards from deeper in the Earth's interior.{{sfn|Schmincke|2003|p=111}} The hot asthenosphere [[Rifting|rifts]] the lithosphere above the plume, allowing magma produced by decompressional melting of the plume head to find pathways to the surface.{{sfn|Schmincke|2003|pp=110-111}}{{sfn|Philpotts|Ague|2009|p=57}}

The swarms of parallel dikes exposed by deep erosion of flood basalts show that considerable [[crustal extension]] has taken place. The dike swarms of west Scotland and Iceland show extension of up to 5%. Many flood basalts are associated with rift valleys, are located on passive continental plate margins, or extend into [[aulacogen]]s (failed arms of [[triple junction]]s where continental rifting begins.) Flood basalts on continents are often aligned with [[hotspot (geology)|hotspot]] volcanism in ocean basins.{{sfn|Philpotts|Ague|2009|pp=57, 380}} The [[Paraná and Etendeka traps]], located in South America and Africa on opposite sides of the Atlantic Ocean, formed around 125 million years ago as the South Atlantic opened, while a second set of smaller flood basalts formed near the Triassic-Jurassic boundary in eastern North America as the North Atlantic opened.{{sfn|Philpotts|Ague|2009|p=52}}{{sfn|Schmincke|2003|p=108}} However, the North Atlantic flood basalts are not connected with any hot spot traces, but seem to have been evenly distributed along the entire divergent boundary.{{sfn|Philpotts|Ague|2009|p=381}}

Flood basalts are often interbedded with sediments, typically [[red beds]]. The deposition of sediments begins before the first flood basalt eruptions, so that subsidence and crustal thinning are precursors to flood basalt activity.{{sfn|Philpotts|Ague|2009|p=380}} The surface continues to subside as basalt erupt, so that the older beds are often found below sea level.{{sfn|Philpotts|Ague|2009|p=57}} Basalt strata at depth (''dipping reflectors'') have been found by [[reflection seismology]] along passive continental margins.{{sfn|Schmincke|2003|p=111}}

=== Ascent to the surface ===
The composition of flood basalts may reflect the mechanisms by which the magma reaches the surface. The original melt formed in the upper mantle (the ''primitive melt'') cannot have the composition of quartz tholeiite, the most common and typically least evolved volcanic rock of flood basalts, because quartz tholeiites are too rich in iron relative to magnesium to have formed in equilibrium with typical mantle rock. The primitive melt may have had the composition of [[picrite basalt]], but picrite basalt is uncommon in flood basalt provinces. One possibility is that a primitive melt ''stagnates'' when it reaches the mantle-crust boundary, where it is not buoyant enough to penetrate the lower-density crust rock. As a tholeiitic magma differentiates (changes in composition as high-temperature minerals crystallize and settle out of the magma) its density reaches a minimum at a magnesium number of about 60, similar to that of flood basalts. This restores buoyancy and permits the magma to complete its journey to the surface, and also explains why flood basalts are predominantly quartz tholeiites. Over half the original magma remains in the lower crust as [[cumulates]] in a system of dikes and sills.<ref name=Cox1980>{{cite journal |last1=Cox |first1=K. G. |title=A Model for Flood Basalt Vulcanism |journal=Journal of Petrology |date=1 November 1980 |volume=21 |issue=4 |pages=629–650 |doi=10.1093/petrology/21.4.629}}</ref>{{sfn|Philpotts|Ague|2009|p=383}}

As the magma rises, the drop in pressure also lowers the [[liquidus]], the temperature at which the magma is fully liquid. This likely explains the lack of phenocrysts in erupted flood basalt. The ''resorption'' (dissolution back into the melt) of a mixture of solid olivine, augite, and plagioclase—the high-temperature minerals likely to form as phenocrysts—may also tend to drive the composition closer to quartz tholeiite and help maintain buoyancy.{{sfn|Philpotts|Ague|2009|p=382}}{{sfn|Philpotts|Ague|2009|p=383}}

=== Eruption ===
Once the magma reaches the surface, it flows rapidly across the landscape, literally flooding the local topography. This is possible in part because of the rapid rate of extrusion (over a cubic km per day per km of fissure length{{sfn|Schmincke|2003|p=108}}) and the relatively low viscosity of basaltic lava. However, the lateral extent of individual flood basalt flows is astonishing even for so fluid a lava in such quantities.{{sfn|Philpotts|Ague|2009|pp=52-53}} It is likely that the lava spreads by a process of ''inflation'' in which the lava moves beneath a solid insulating crust, which keeps it hot and mobile.<ref name="SelfEtal1996">{{cite journal |last1=Self |first1=S. |last2=Thordarson |first2=Th. |last3=Keszthelyi |first3=L. |last4=Walker |first4=G. P. L. |last5=Hon |first5=K. |last6=Murphy |first6=M. T. |last7=Long |first7=P. |last8=Finnemore |first8=S. |title=A new model for the emplacement of Columbia River basalts as large, inflated Pahoehoe Lava Flow Fields |journal=Geophysical Research Letters |date=15 September 1996 |volume=23 |issue=19 |pages=2689–2692 |doi=10.1029/96GL02450|bibcode=1996GeoRL..23.2689S }}</ref> Studies of the Ginkgo flow of the Columbia River Plateau, which is {{convert|30 to 70|m||sp=us}} thick, show that the temperature of the lava dropped by just {{convert|20|C||sp=us}} over a distance of {{convert|500|km||sp=us}}. This demonstrates that the lava must have been insulated by a surface crust and that the flow was [[laminar flow|laminar]], reducing heat exchange with the upper crust and base of the flow.<ref name="HoCashman1997">{{cite journal |last1=Ho |first1=Anita M. |last2=Cashman |first2=Katharine V. |title=Temperature constraints on the Ginkgo flow of the Columbia River Basalt Group |journal=Geology |date=1 May 1997 |volume=25 |issue=5 |pages=403–406 |doi=10.1130/0091-7613(1997)025<0403:TCOTGF>2.3.CO;2|bibcode=1997Geo....25..403H }}</ref>{{sfn|Philpotts|Ague|2009|pp=53-54}} It has been estimated that the Ginkgo flow advanced 500&nbsp;km in six days (a rate of advance of about 3.5&nbsp;km per hour).<ref name="HoCashman1997"/>

The lateral extent of a flood basalt flow is roughly proportional to the cube of the thickness of the flow near its source. Thus, a flow that is double in thickness at its source can travel roughly eight times as far.{{sfn|Philpotts|Ague|2009|p=53}}

Flood basalt flows are predominantly [[pāhoehoe]] flows, with [[ʻaʻā]] flows much less common.<ref>{{cite book |last1=Self |first1=S. |last2=Thordarson |first2=T. |last3=Keszthelyi |first3=L. |title=Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism |chapter=Emplacement of Continental Flood Basalt Lava Flows |year=1997 |series=Geophysical Monograph Series |volume=100 |pages=381–410 |doi=10.1029/GM100p0381 |bibcode=1997GMS...100..381S |isbn=9781118664346 |chapter-url=https://books.google.com/books?id=5H8vobp2x3AC&dq=flood+basalt+aa&pg=PA381 |access-date=17 January 2022}}</ref>

Eruption in flood basalt provinces is episodic, and each episode has its own chemical signature. There is some tendency for lava within a single eruptive episode to become more silica-rich with time, but there is no consistent trend across episodes.{{sfn|Philpotts|Ague|2009|p=382}}