Skarn xenolith record crustal CO₂ liberation during Pompeii and Pollena eruptions, Vesuvius volcanic system, central Italy

Highlights

• New stable isotope data from igneous rocks and contact-metamorphic skarn-type xenoliths from the Vesuvius volcanic system.

• Different mechanisms of magma–crust interaction, such as carbonate dissolution and skarn recycling, are relevant processes.

• A formulation to approximate the potential volumes of magmatic versus non-magmatic CO₂ is presented.

Abstract

Limestone assimilation and skarn formation are important processes in magmatic systems emplaced within carbonate-rich crust and can affect the composition of the magma and that of associated volcanic gas. In this study we focus on marble and calc-silicate (skarn) xenoliths from contact reactions between magma and carbonate wall-rock of the Vesuvius volcanic system. We present new elemental and C-O isotope data for marble and skarn xenoliths as well as for igneous rocks collected from the AD 79 (Pompeii) and AD 472 (Pollena) eruptions. The igneous samples have consistently high δ¹⁸O values (9.3 to 10.8‰), but low H₂O contents (≤ 1.5%), indicating that magma–crust interaction prior to eruption took place. The marble xenoliths, in turn, record initial decarbonation reactions and fluid-mass exchange in their textures and δ¹³C and δ¹⁸O ranges, while the skarn xenoliths reflect prolonged magma–carbonate interaction and intense contact metamorphism. Skarn-xenoliths record Ca and Mg release from the original carbonate and uptake of Al and Si and span the full δ¹⁸O data range from un-metamorphosed carbonate (> 18‰) to values typical for Vesuvius magmatic rocks (~ 7.5‰), which implies that skarn xenoliths comprise carbonate and magmatic components. Textural and chemical evidence suggest that direct carbonate dissolution into the host magmas occurred as well as post-metamorphic skarn recycling, resulting in progressive Ca and Mg liberation from the skarn xenoliths into the magma. Magma–carbonate interaction is an additional source of CO₂ during carbonate break-down and assimilation and we calculate the amount of extra volatile components likely liberated by contact metamorphic reactions before and during the investigated eruptions. We find that the extra CO₂ added into the volcanic system could have outweighed the magmatic CO₂ component by ≥ factor seven and thus likely increased the intensity of both the Pompeii and the Pollena eruptive events.

Introduction

The geochemical signatures of continental magmas frequently indicate involvement of crustal components in their genesis and evolution (e.g., Hildreth and Moorbath, 1988, Davidson et al., 1990, Davidson et al., 2005, Troll et al., 2005, Walker et al., 2007). In particular, magma–carbonate interaction has been demonstrated for a series of volcanic and plutonic systems which, like the Vesuvius volcanic system are emplaced into carbonate-rich crust (e.g., Goff et al., 2001, Wenzel et al., 2002, Dallai et al., 2004, Dallai et al., 2011, Barnes et al., 2005, Schaaf et al., 2005, Piochi et al., 2006, Chadwick et al., 2007, Di Renzo et al., 2007, Iacono-Marziano et al., 2007, Iacono-Marziano et al., 2008, Freda et al., 2008, Gaeta et al., 2009, Deegan et al., 2010, Troll et al., 2012, Troll et al., 2013, Di Rocco et al., 2012, Borisova et al., 2013, Jeffery et al., 2013). A large variety of xenolith types is present in the Somma-Vesuvius eruptive products, including abundant weakly-to-highly metamorphosed carbonate and calc-silicate (skarn) type compositions. These provide unequivocal evidence for magma–carbonate interaction during magma differentiation (e.g., Fulignati et al., 2000, Fulignati et al., 2001, Gilg et al., 2001, Del Moro et al., 2001, Iacono-Marziano et al., 2008, Iacono-Marziano et al., 2009). Skarn-type calc-silicate compositions typically develop in contact-metamorphic aureoles in response to interaction between limestone or dolostone with a silicate magma. Skarn formation is known to progressively modify the composition of both the magma and the carbonate wall-rock (e.g., Gilg et al., 2001, Fulignati et al., 2004, Gaeta et al., 2009, Mollo et al., 2010, Di Rocco et al., 2012). Therefore, assessment of the textural and compositional diversity of calc-silicate xenolith assemblages can provide an opportunity to shed light on the processes at work during sub-volcanic magma–carbonate interaction in such crustal settings (Fulignati et al., 2000, Fulignati et al., 2001, Fulignati et al., 2004, Barnes et al., 2005, Chadwick et al., 2007, Gaeta et al., 2009, Di Rocco et al., 2012, Troll et al., 2013).

Previous mineralogical and geochemical studies on skarn xenoliths from the Vesuvius volcanic system characterised the pressure-temperature conditions of formation, the nature of fluids involved, and identified the principal isotopic changes associated with conversion from limestone to skarn (Fulignati et al., 2000, Fulignati et al., 2001, Fulignati et al., 2004, Fulignati et al., 2005a, Fulignati et al., 2005b, Gilg et al., 2001, Del Moro et al., 2001, Lima et al., 2007). Skarn xenoliths in the Vesuvius eruptive deposits can be linked to relatively shallow magma reservoirs (≤ 4–10 km; Auger et al., 2001, Civetta et al., 2004, Scaillet et al., 2008 and references therein) that were located within the thick limestone and dolostone carbonates under the volcano and which extend from approximately ≥ 2 km to at least 8 km depth (Zollo et al., 1996, Bruno et al., 1998, Auger et al., 2001, Pappalardo and Mastrolorenzo, 2010).

Conversely, variations of Sr-Nd-Pb-O isotopic compositions in central Italian magmas, including those from Vesuvius, have indeed been attributed by a number of workers to crustal assimilation of carbonates during storage in the upper few kilometres of the crust (e.g., Cioni et al., 1998, Fulignati et al., 2000, Auger et al., 2001, Di Renzo et al., 2007, Iacono-Marziano et al., 2008, Scaillet et al., 2008, Dallai et al., 2011, Pichavant et al., 2014), while other researchers favour isotopic variability of associated mantle sources (e.g., Cortini and Hermes, 1981, Ayuso et al., 1998, Peccerillo, 2005, Martin et al., 2012, Moretti et al., 2013). To evaluate the relative roles of source-enrichment vs. interaction between magma and crustal carbonate rocks, it appears necessary to explore the mechanism and processes of magma–crust interaction in more detail. In this respect, the skarn xenoliths in the AD 79 and AD 472 eruptions have previously been considered to reflect decarbonation of the carbonate wall-rock, but with limited direct effects on the resident magmas (Fulignati et al., 2004, Fulignati et al., 2005b, Fulignati et al., 2011, Fulignati et al., 2013). On the other hand, magma–carbonate interaction has recently been suggested to have affected even the most primitive of Vesuvius's volcanic products (e.g., Dallai et al., 2011, Pichavant et al., 2014) and is thought to control the chemical and isotopic composition of present-day Vesuvius fumarole gas (Iacono-Marziano et al., 2009). Here we address this topic through a detailed study of the petrography, mineralogy, major and trace element geochemistry, and in particular the C and O isotope ratios of i) skarn-type calc-silicate xenoliths, ii) marble xenoliths and iii) igneous products (pyroclastic fragments and plutonic xenoliths) from the AD 79 and the AD 472 eruptive events. Using this integrated approach, we present a magma–carbonate interaction model that reconciles the geochemical features of the igneous samples with those of the crustal xenoliths. Moreover, we propose an easy to use formulation to quantify the potential volume of CO₂ liberated by contact metamorphic process during skarn formation relative to the volatile fraction released from the associated magma volumes, which allows us to consider consequences for resulting eruptive behaviour and style.