Thermodynamic Modelling of Mantle–Melt Interaction Evidenced by Veined Wehrlite Xenoliths from the Rockeskyllerkopf Volcanic Complex, West Eifel Volcanic Field, Germany

Journal of Petrology, Jan 2018

The oldest volcanic center of the Rockeskyllerkopf Volcanic Complex (RVC) in the West Eifel volcanic field hosts three distinct compositional groups of mantle xenolith: two groups of lherzolite and harzburgite and a group of wehrlite xenoliths that are cross-cut by phlogopite–clinopyroxene veins and variably impregnated by these same minerals. The lherzolite and harzburgite xenoliths represent mantle that was affected by metasomatism prior to the Quaternary magmatic activity below the RVC. We interpret the wehrlites to be the result of infiltration of orthopyroxene-undersaturated alkaline melt into the orthopyroxene-bearing lithospheric mantle. Reaction between mantle and melt consumed orthopyroxene and precipitated olivine and clinopyroxene as well as phlogopite. Models of the equilibration of peridotite and infiltrated melt created with alphaMELTS produce wehrlite with olivine, phlogopite and spinel compositions that are similar to those observed in the natural samples for melt:rock ratios between 0·11 and 1·11. Model clinopyroxene compositions do not match the observed range of compositions in wehrlite. The intra-sample range in clinopyroxene composition in wehrlite xenoliths suggests that clinopyroxene did not reach equilibrium, whereas olivine compositions in the same samples show much less variation, suggesting that they attained equilibrium. We calculate the time required for olivine homogenization to be around 3300 years, whereas the time needed for clinopyroxene to be homogenized is 100 times longer. We suggest that the range of clinopyroxene compositions within and between the wehrlite xenoliths is the result of varying degrees of equilibration of the initial lherzolite or harzburgite clinopyroxene with clinopyroxene formed by reaction with the melt. Melt compositions from the alphaMELTS models suggest that the differences in lava composition observed in the RVC can be produced by mixing between melts from a garnet peridotite source with secondary melts produced by peridotite–melt reaction in the spinel-facies mantle. We suggest that mixing occurred during melt transport through the lithospheric mantle as secondary melts produced during wehrlite formation were drawn into fractures that were exploited by several generations of magma rising from the garnet peridotite source.

Article PDF cannot be displayed. You can download it here:

https://academic.oup.com/petrology/article-pdf/59/1/59/24839385/egy018.pdf

Thermodynamic Modelling of Mantle–Melt Interaction Evidenced by Veined Wehrlite Xenoliths from the Rockeskyllerkopf Volcanic Complex, West Eifel Volcanic Field, Germany

JOURNAL OF PETROLOGY Journal of Petrology, 2018, Vol. 59, No. 1, 59–86 doi: 10.1093/petrology/egy018 Advance Access Publication Date: 20 February 2018 Original Article Thermodynamic Modelling of Mantle–Melt Interaction Evidenced by Veined Wehrlite Xenoliths from the Rockeskyllerkopf Volcanic Complex, West Eifel Volcanic Field, Germany Cliff S. J. Shaw1,2*, Breagh S. Lebert1 and Alan B. Woodland2 1 Department of Earth Sciences, University of New Brunswick, 2 Bailey Drive, Fredericton, NB E3B 5A3, Canada; Institut für Geowissenschaften, Physikalisch-Chemische Mineralogie, Altenhöferallee 1, Frankfurt am Main D-60438, Germany 2 *Corresponding author. E-mail: Received March 8, 2017; Accepted February 14, 2018 ABSTRACT The oldest volcanic center of the Rockeskyllerkopf Volcanic Complex (RVC) in the West Eifel volcanic field hosts three distinct compositional groups of mantle xenolith: two groups of lherzolite and harzburgite and a group of wehrlite xenoliths that are cross-cut by phlogopite–clinopyroxene veins and variably impregnated by these same minerals. The lherzolite and harzburgite xenoliths represent mantle that was affected by metasomatism prior to the Quaternary magmatic activity below the RVC. We interpret the wehrlites to be the result of infiltration of orthopyroxene-undersaturated alkaline melt into the orthopyroxene-bearing lithospheric mantle. Reaction between mantle and melt consumed orthopyroxene and precipitated olivine and clinopyroxene as well as phlogopite. Models of the equilibration of peridotite and infiltrated melt created with alphaMELTS produce wehrlite with olivine, phlogopite and spinel compositions that are similar to those observed in the natural samples for melt:rock ratios between 011 and 111. Model clinopyroxene compositions do not match the observed range of compositions in wehrlite. The intra-sample range in clinopyroxene composition in wehrlite xenoliths suggests that clinopyroxene did not reach equilibrium, whereas olivine compositions in the same samples show much less variation, suggesting that they attained equilibrium. We calculate the time required for olivine homogenization to be around 3300 years, whereas the time needed for clinopyroxene to be homogenized is 100 times longer. We suggest that the range of clinopyroxene compositions within and between the wehrlite xenoliths is the result of varying degrees of equilibration of the initial lherzolite or harzburgite clinopyroxene with clinopyroxene formed by reaction with the melt. Melt compositions from the alphaMELTS models suggest that the differences in lava composition observed in the RVC can be produced by mixing between melts from a garnet peridotite source with secondary melts produced by peridotite–melt reaction in the spinel-facies mantle. We suggest that mixing occurred during melt transport through the lithospheric mantle as secondary melts produced during wehrlite formation were drawn into fractures that were exploited by several generations of magma rising from the garnet peridotite source. Key words: Eifel; wehrlite; magma–mantle reaction; magma evolution; magma transport INTRODUCTION Magmas passing through the shallow mantle are commonly out of equilibrium with the mantle mineral assemblage as indicated by the presence of reaction zones in the peridotite sections of ophiolites (Kelemen & Dick, 1995; Dygert et al., 2016), massif peridotites (Mazzucchelli et al., 2009; Piccardo, 2010) and peridotite xenoliths (Peslier et al., 2002; Matusiak-Malek et al., 2017). C The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: V 59 60 lithospheric mantle. In their models of melt migration, Keller et al. (2013) showed that as the competence of the mantle increases, melt percolation by porous flow, which will dominate in the asthenosphere, becomes inefficient and as the magma enters the thermal boundary layer at the base of the lithosphere, new modes of melt transport have to be considered. Of particular interest is magma flow along tensile fractures that propagate rapidly and have the ability to transport magma from the lithospheric mantle to surface on very short time scales. Havlin et al. (2013) suggested that magma at the lithosphere–asthenosphere boundary is sufficiently overpressured that it is capable of generating centimetre-wide veins that can propagate a few kilometres into the lithosphere. In intraplate settings the basaltic melts in these veins are likely to be alkaline and will often be feldspathoid-bearing. They are therefore significantly more orthopyroxene undersaturated at lithospheric mantle pressures than MORB (Keshav et al., 2004; Lambart et al., 2009). This undersaturation will drive reaction with lherzolite or harzburgite as melt migrates out of the fractures and into the peridotite wall rock. Field evidence for the transport of intraplate magmas through the mantle is limited to studies of peridotite xenoliths. These xenoliths represent fragments of the lithospheric mantle (e.g. Edgar et al., 1989; Dawson, 2002). The xenoliths range from harzburgite to wehrlite and commonly contain secondary minerals such as amphibole, phlogopite and clinopyroxene. They also show evidence of magma–xenolith reaction in the form of reaction of orthopyroxene to olivine and silica-rich melt, and the secondary development of sieve texture in clinopyroxene and spinel (Zinngrebe & Foley, 1995; Shaw & Edgar, 1997; Yaxley et al., 1997; Klügel, 1998; Wulff-Pedersen et al., 1999; Klügel, 2001; Shaw et al., 2005). These reactions are similar to those observed for MORB mantle reaction (Tursack & Liang, 2012) with one important difference, clinopyroxene does not appear to be consumed as part of the reaction. Rather, it occurs as a stable phase in orthopyroxene reaction zones and as modified primary clinopyroxene that develops sieve texture as shown in the experiments of Shaw & Dingwell (2008). Discordant pyroxenite, phlogopite pyroxenite and amphibole pyroxenite veins, interpreted to result from crystallization of alkaline basaltic melts being transported through the lithospheric mantle, are also present in many massif peridotites and xenolith suites (Zanetti et al., 1996; Witt-Eickschen & Kramm, 1998; WittEickschen et al., 1998; Fabries et al., 2000; Dawson, 2002; Bodinier et al., 2004; Shaw et al., 2005). In many cases, these veins have marginal reaction zones in which the lherzolite or harzburgite assemblage is transformed to wehrlite, suggesting that outward migration of the vein-forming melt may result in the formation of wehrlite haloes around melt-filled veins in the lithospheric mantle. The most voluminous magmas on Earth, mid-ocean ridge basalts (MORB), are not in chemical equilibrium with typical mantle mineral assemblages at low pressure (Stolper, 1980), which led Spiegelman & Kenyon (1992), Kelemen & Dick (1995), Kelemen et al. (1995) and Asimow & Stolper (1999) to suggest that the erupted basalts (...truncated)


This is a preview of a remote PDF: https://academic.oup.com/petrology/article-pdf/59/1/59/24839385/egy018.pdf
Article home page: https://academic.oup.com/petrology/article/59/1/59/4893735

Shaw, Cliff S J, Lebert, Breagh S, Woodland, Alan B. Thermodynamic Modelling of Mantle–Melt Interaction Evidenced by Veined Wehrlite Xenoliths from the Rockeskyllerkopf Volcanic Complex, West Eifel Volcanic Field, Germany, Journal of Petrology, 2018, pp. 59-86, Volume 59, Issue 1, DOI: 10.1093/petrology/egy018