Two-component mantle melting-mixing model for the generation of mid-ocean ridge basalts: Implications for the volatile content of the Pacific upper mantle

We review the different scenarios for the origin of light noble gases (He, Ne, and Ar) on Earth. Several sources could have contributed to the Earth's noble gas budget: implanted solar wind, solar nebula gas, chondrites, and comets. Although there is evidence for “solar‐like” neon in the Earth's mantle, questions remain as to its origin. A new compilation of noble gas data in lunar soils, interplanetary dust particles, micrometeorites, and solar wind allows examination of the implanted solar wind composition, which is key to understanding the “solar‐like” mantle neon isotope composition. We show that lunar soils that reflect this solar‐wind‐implanted signature have a 20Ne/22Ne ratio very close to that of ocean island basalts. New data and calculations illustrate that the measured plume source 20Ne/22Ne ratio is close to the primitive mantle ratio, when taking into account mixing with the upper mantle (that has lower 20Ne/22Ne ratio). This favors early solar wind implantation to account for the origin of light volatiles (He, Ne, and possibly H) in the Earth's mantle: they were incorporated by solar wind irradiation into the Earth's precursor grains during the first few Myr of the solar system's formation. These grains must have partially survived accretion processes (only a few percent are needed to satisfy the Earth's budget of light volatiles). As for the atmosphere, the neon isotope composition can be explained by mixing 36% of mantle gases having this solar‐wind‐implanted signature and 64% of chondritic gases delivered in a late veneer phase.

Section: Discussionmentioning

confidence: 99%

Origin of Light Noble Gases (He, Ne, and Ar) on Earth: A Review

Péron

Moreira

Agranier

2018

“…The composition of mid‐ocean ridge basalts (MORB) reflects a variety of subsurface processes [e.g., Sinton and Detrick , ; Perfit and Chadwick , ; Klein , ; White and Klein , ]. In addition to fractional crystallization, reactive crystallization in dunite channels [e.g., Kelemen et al ., ; Collier and Kelemen , ; Liang et al ., ], melt rock reactions in the lithosphere [e.g., Drouin et al ., ; Sanfilippo et al ., ; Paquet et al ., ], or the gabbroic crust [e.g., Lissenberg and Dick , ; MacLeod et al ., ], mantle source heterogeneity [e.g., Sims et al ., ; Donnelly et al ., ; Shimizu et al ., ], variations in depths or extents of melting [e.g., Klein and Langmuir , ; Langmuir et al ., ; Perfit et al ., ], and crustal assimilation [e.g., Michael and Schilling , ; Wanless et al ., ] can all modify trace element contents and, depending on the phases involved, alter the CaO, Al 2 O 3 , and MgO content of erupted basalts [e.g., Michael and Shilling , 1989; Lissenberg and Dick , ; Paquet et al ., ]. For example, melt rock reaction in gabbroic rocks may involve the dissolution of olivine and crystallization of clinopyroxene [ Paquet et al ., ] changing the CaO/Al 2 O 3 of the melt.…”

Section: Basalt Crystallization Pressuresmentioning

confidence: 99%

“…Finally, we suggest that the remarkably similar maximum VSPs recorded in melt inclusions across all spreading rates (3 kbars) reflects a relatively uniform CO 2 content of 50-85 ppm for the depleted upper mantle feeding the global MOR system. Kelemen, 2010;Liang et al, 2010], melt rock reactions in the lithosphere [e.g., Drouin et al, 2009;Sanfilippo et al, 2014;Paquet et al, 2016], or the gabbroic crust [e.g., Lissenberg and Dick, 2008;MacLeod et al, 2013], mantle source heterogeneity [e.g., Sims et al, 2002;Donnelly et al, 2004;Shimizu et al, 2016], variations in depths or extents of melting [e.g., Klein and Langmuir, 1987;Langmuir et al, 1992;Perfit et al, 1994], and crustal assimilation [e.g., Michael and Schilling, 1989;Wanless et al, 2010] can all modify trace element contents and, depending on the phases involved, alter the CaO, Al 2 O 3 , and MgO content of erupted basalts [e.g., Michael and Shilling, 1989;Lissenberg and Dick, 2008;Paquet et al, 2016]. For example, melt rock reaction in gabbroic rocks may involve the dissolution of olivine and crystallization of clinopyroxene [Paquet et al, 2016] changing the CaO/Al 2 O 3 of the melt.…”

Section: Introductionmentioning

confidence: 99%

Spreading rate‐dependent variations in crystallization along the global mid‐ocean ridge system

Wanless

Behn

2017

We investigate crustal accretion at mid‐ocean ridges by combining crystallization pressures calculated from major element contents in mid‐ocean ridge basalt (MORB) glasses and vapor‐saturation pressures from melt inclusions and MORB glasses. Specifically, we use established major element barometers and pressures estimated from 192 fractional crystallization trends to calculate crystallization pressures from >9000 MORB glasses across the global range of mid‐ocean ridge spreading rates. Additionally, we estimate vapor‐saturation pressures from >400 MORB glasses from PETDB and >400 olivine‐hosted melt inclusions compiled from five ridges with variable spreading rates. Both major element and vapor‐saturation pressures increase and become more variable with decreasing spreading rate. Vapor saturation pressures indicate that crystallization occurs in the lower crust and upper mantle at all ridges, even when a melt lens is present. We suggest that the broad peaks in major element crystallization pressures at all spreading rates reflects significant crystallization of on and off‐axis magmas along the base of a sloping lithosphere. Combining our observations with ridge thermal models we show that crystallization occurs over a range of pressures at all ridges, but it is enhanced at thermal/rheologic boundaries, such as the melt lens and the base of the lithosphere. Finally, we suggest that the remarkable similarity in the maximum vapor‐saturation pressures (∼3 kbars) recorded in melt inclusions from a wide range of spreading rates reflects a relatively uniform CO2 content of 50–85 ppm for the depleted upper mantle feeding the global mid‐ocean ridge system.

“…This is similar to other estimates for the formation of tholeiitic and alkalic magmas in the Galápagos (Naumann et al, ). In particular, the high total alkali contents, F/Nd ratios (Figure S4; Shimizu et al, ), and [Sm/Yb] N , which is not modified by fractional crystallization (Figure b) can be explained by variations in the extent of melting of the same mantle source. To reproduce the observed trace element ratios in the alkalic DWF lavas compared to the tholeiitic ones, extents of melting must be modulated by a 500‐MPa increase to the top of the melting column (for [Sm/Yb] N ) and decrease in mantle potential temperature by 25–50 °C.…”

Section: Discussionmentioning

confidence: 95%

Submarine Deep‐Water Lava Flows at the Base of the Western Galápagos Platform

Anderson

Wanless

Schwartz

et al. 2018

To investigate the initial phases of magmatism at the leading edge of the upwelling mantle plume, we mapped, photographed, and collected samples from two long, deep‐water lava flows located at the western base of the Galápagos Platform using the remotely operated vehicle Hercules. Lavas were recovered from four areas on the seafloor west of Fernandina volcano, including the western flow fronts of two deep‐water flows, heavily sedimented terrain between the two flows, and the eastern, shallower end of one flow. The sediment cover and morphologies are distinct between the western flow fronts and the eastern region based on seafloor imagery, suggesting that the long lava flows are not a single eruptive unit. Major and trace element concentrations reveal both tholeiitic and alkalic compositions and support the interpretation that multiple eruptive units comprise the deep‐water flows. Alkalic lavas have higher [La/Sm]N ratios (2.05–2.12) and total alkali contents (5.18–5.40) compared to tholeiitic lavas, which have [La/Sm]N ratios ranging from 1.64 to 1.68 and total alkali contents ranging from 3.07 to 4.08 wt%. Radiogenic isotope ratios are relatively homogeneous, suggesting a similar mantle source. We use petrologic models to assess three alternative mechanisms for the formation of the alkalic magmas: (1) high‐pressure crystallization of clinopyroxene, (2) mixing of high silica and mafic magmas, and (3) variable extents of melting of the same mantle source. Our modeling indicates that the alkalic samples form from lower extents of melting compared to the tholeiitic lavas and suggests that the deep‐water alkalic lavas are analogous to the initial, preshield building phase observed south of Hawaii and at the base of Loihi Seamount.