MANTLE PLUMES plumes and plumes do we need

MANTLE PLUMES, plumes and ‘plumes’: do we need all of them? Alexei V. Ivanov Institute of the Earth’s Crust SB RAS, Irkutsk Russia www. mantleplumes. org 1. History of plumes and definitions 2. Plumes are thought to be required by evidence: a. Temperature b. Age progressive volcanism c. Heat flow d. Trace elements e. Radiogenic isotopes f. Seismic tomography g. Uplift shape and history h. Volume of magma Alexei Ivanov: a lecture at the Paleomagnetism and geochronology laboratory in Beijing (PGLB), China, January 11, 2006. The lecture was given on the occasion of his visit sponsored by a PGLB visitor grant. Special thanks to Prof. R. Zhu and Dr. H. He. 3. Does the evidence require plumes? 4. New models for melting anomalies: case studies a. Permo-Triassic Siberian Traps b. Late Cenozoic volcanism of Central Asia 5. Conclusions

History of plumes and definitions Modified after (Anderson and Natland, 2005; Ivanov, 2006 in press) Morgan, 1971 Why current skepticism?

History of plumes and definitions Morgan’s evidence and assumptions

History of plumes and definitions 1. Oceanic island chains (e. g. Hawaii) do not associate with lithospheric plate boundaries and hence cannot be explained by plate tectonics; 2. Age progressive volcanism requires stable source of heat (hot spots) (after Wilson, 1963); 3. Upper mantle is vigorously convecting and hence the source of heat is somewhere below (source is most likely in the D’’ layer); 4. Oceanic island basalts are enriched in incompatible elements compared with depleted mid-oceanic ridge basalts, hence source is primitive in composition (Earth’s lower mantle); 5. Hot spots lie atop of localized thermal upwellings referred to as plumes; 6. Plumes represent the major part of upraising convective flow; 7. Plumes give the force for plate tectonics.

History of plumes and definitions In fluid mechanics literature, "plumes" refer to upwelling or downwelling driven by self-buoyancy (Korenaga, 2005) A mantle plume is a localised, roughly axisymmetric upwelling of buoyant rock, originating from a boundary layer deep within the Earth (Saunders, 2004). . . surface hotspots on Earth may have three distinct origins: (a) <primary plumes> would originate from the deepest part of the lower mantle (. . . ‘Morganian’ <plumes>) (b) <Secondary> plumes originate from. . . the top of the large transient domes that correspond to the superswells. (c) The remainder could be upper mantle features, and in that sense ‘Andersonian’ (Courtillot et al. , 2003).

History of plumes and definitions Plume, hotspot, wetspot, melting anomaly - a region of active or ancient volcanism not associated with MOR- and IA-types of volcanism. Superplume, superswell, megaplume, reheated slab - thermal and/or thermochemical broad upwelling structure either traversing through the whole mantle or located in the lower mantle on a scale of >108 yr. Lower mantle plume, primary plume, Morganian plume thermal and/or thermochemical localized upwelling structure traversing through the whole mantle to the subsurface. Upper mantle plume, secondary plume - thermal and/or thermochemical localized upwelling structure originating in the transitional zone. Andersonian plume - asthenospheric structure, which is not a plume in any sense.

History of plumes and definitions The classical Morganian plume probably does not exist Arguments pro and contra

Plumes are thought to be required by evidence Temperature

Does evidence require plumes? Temperature

Plumes are thought to be required by evidence Age progressive volcanism Age incr ease Plumes are fixed - plates are moving No

Does evidence require plumes? Age progressive volcanism Raymond et al. , 2000 Fixed plumes gone with the <mantle? > wind

Does evidence require plumes? Age progressive volcanism Stuart, 2005 Hawaiian chain is an effect of thermoelastic propagating crack?

Does evidence require plumes? Age progressive volcanism Rasskazov et al. , 1997; 2000 Dobretsov et al (2001) use as example of moving plume Example of complex stress conditions in intracontinental setting

Plumes are thought to be required by evidence Heat flow Hofmeister and Criss, 2004 Total output is 44 TW >than radiogenic output of 20 TW, hence core must contribute heat to the surface 1. Measured heat flow value for oceanic crust is underestimated 2. Half space cooling model used instead of measurements -> 44 TW 3. BSE is made of CI chondrite -> 20 TW

Does evidence require plumes? Heat flow Hofmeister and Criss, 2004 Is the half space cooling model is better than direct measurements ? ? ? Total output 31 TW ~ radiogenic output 31 TW if BSE is made of EH

Plumes are thought to be required by evidence Trace elements MORB and IAB are from depleted upper mantle. OIB must come from somewhere else (lower mantle? )

Does evidence require plumes? Trace elements Ontong-Java There are many examples when studied basalts do not resemble OIB-trace element component, but they are still considered to be of lower mantle plume origin

Plumes are thought to be required by evidence Radiogenic isotopes Iceland plume Elam and Stuart, 2004 Iceland plume

Does evidence require plumes? Radiogenic isotopes (Helium) Iceland plume Elam and Stuart, 2004 Iceland plume Class et al. , 2005

Does evidence require plumes? Radiogenic isotopes (Helium) Comoro plume Iceland plume Anderson, 2000, 2001

Plumes are thought to be required by evidence Seismic tomography Montelli et al. , 2004 Plumes are imaged via seismic tomography

Does evidence require plumes? Seismic tomography Plume from the core/mantle boundary or lowermost mantle Ritsema and Allen, 2002 RA 2002 Afar yes Bowie yes Hawaii yes Iceland yes Reunion no Tristan no Louisville yes Samoa yes Tahiti no Ascension no Azores no Canary no Easter yes Yellowstone no M 2004 no no yes yes yes no

Plumes are thought to be required by evidence Uplift shape and history Campbell, 2005 Plumes support broad surface uplifts

Plumes are thought to be required by evidence Uplift shape and history Burov and Guillou-Frottier, 2005

Does evidence require plumes? Uplift shape and history Korenaga, 2005 Two most voluminous LIPs; Siberian Traps and Ontong-Java did not experienced uplift before the eruptions !

Plumes are thought to be required by evidence Large volume of magma requires high temperature and hence lower mantle plumes

Does evidence require plumes? Large volume of magma Actually, it does not Siberian Traps example

New models for the melting anomalies: case studies Siberian Traps

New models for the melting anomalies: case studies Siberian Traps Ivanov et al. , 2005

New models for the melting anomalies: case studies Siberian Traps

New models for the melting anomalies: case studies Siberian Traps Modified after Ivanov and Balyshev, 2005

New models for the melting anomalies: case studies Siberian Traps Ivanov et al. , 2004; in preparation

New models for the melting anomalies: case studies Siberian Traps Ivanov et al. , 2004; in preparation

New models for the melting anomalies: case studies Siberian Traps Ivanov et al. , 2004; in preparation

New models for the melting anomalies: case studies Siberian Traps Ivanov et al. , 2004; in preparation Upper Mantle Transitional zone Lower Mantle

New models for the melting anomalies: case studies Siberian Traps Ivanov et al. , 2004; in preparation

New models for the melting anomalies: case studies Siberian Traps Data do not require lower mantle plume origin of the Siberian Traps Data do require water in the mantle source of melting Water is brought into the mantle by plate tectonic processes

New models for the melting anomalies: case studies Central Asia Enriched isotopic signatures require long conservation of recycled material in the lower mantle and hence lower mantle plumes

New models for the melting anomalies: case studies Central Asia Actually, it does not Central Asian example

New models for the melting anomalies: case studies Central Asia A number of papers refer Cenozoic volcanism of the Central Asia to be related with lower mantle plume on basis of geochemical data (e. g. Grachev, 2000; Yarmolyuk et al. , 2003) Zorin et al. (2004) revealed 7 upper mantle plumes on basis of gravity data

New models for the melting anomalies: case studies Central Asia Zorin et al. , 2004

New models for the melting anomalies: case studies Central Asia Zorin et al. , 2006 (2 papers in press) Seismic tomography after model of Zhao, 2004

New models for the melting anomalies: case studies Demonterova et al. , 2005 Central Asia

New models for the melting anomalies: case studies Demonterova et al. , 2005 Central Asia

New models for the melting anomalies: case studies Demonterova et al. , 2005 Central Asia S=Do/Co I=[Rbo/Co][1 -Po]

New models for the melting anomalies: case studies Central Asia Modified after Zorin et al. , 2006 (in press)

New models for the melting anomalies: case studies Central Asia Data do not require lower mantle plume origin of the Central Asian basalts Data do require fertile fusible source and upper mantle plumes (in terms of fluid dynamics) Fertile source in the mantle is created by plate tectonic processes

Conclusions 1. The term ‘plume’ is poorly defined and hence should not be used in publications without special explanation of its meaning. 2. Evidence cited to prove plumes from the core and lower mantle in many instances do not require such interpretation. 3. Many alternative models can better explain melting anomalies than lower mantle plumes do.

Thanks!