Journal of Petrology Advance Access originally published online on January 21, 2009
Journal of Petrology 2009 50(2):361-385; doi:10.1093/petrology/egp002
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Carbonate Assimilation in Open Magmatic Systems: the Role of Melt-bearing Skarns and Cumulate-forming Processes
1Dipartimento Di Scienze Della Terra, Sapienza Università Di Roma, P.Le Aldo Moro 5, 00185 Rome, Italy
2Istituto Nazionale Di Geofisica E Vulcanologia, Via Di Vigna Murata 605, 00143 Rome, Italy
RECEIVED APRIL 14, 2008; ACCEPTED JANUARY 1, 2009
 |
ABSTRACT |
|---|
The geochemical characteristics of volcanic products in a variety of tectonic settings demonstrate that incorporation of crustal material into magmas is a relatively common process. Contamination of magmas by crustal components, in turn, can have a significant effect on magma composition and rheology. Despite this, the mechanism by which contamination occurs is still not well established and its efficacy is denied by some. In this study we focus on magma–carbonate interaction and on the rock shells (cumulates and skarns) formed at the contact between a magma chamber and its wall-rocks. We deduce that previous, unsuccessful attempts at carbonate assimilation–fractional crystallization (AFC) modelling can be related to the paucity of information about the cumulate zone in contact with skarns. We use one of the best examples of a magmatic plumbing system emplaced within a thick carbonate substratum (the Colli Albani Volcanic District in Central Italy) to demonstrate that a ‘skarn environment’ can act as a source of CaO-rich silicate melts, and that the assimilation of these melts into the primitive magma is the main process responsible for magma contamination, rather than the ingestion of solid carbonate wall-rocks. In particular, by means of microtextural observations, mineral chemistry, whole-rock geochemical data and MELTS simulations we highlight the effect of high Ca-Tschermaks (CaAl2SiO6) activity in the melt on the stability of Cr-spinel, olivine, and clinopyroxene in cumulate rocks, define a reaction-cumulate zone where clinopyroxene crystallization is favoured, and model the magmatic differentiation processes active in this zone.
KEY WORDS: cumulate; skarn; carbonate assimilation; open system; Colli Albani
|
INTRODUCTION |
|---|
The geochemical signatures of various volcanic products clearly indicate that crustal components are commonly involved in the genesis and evolution of magmatic systems (e.g. Kamenetsky & Gurenko, 2007
; Walker et al., 2007). Magma–carbonate interaction, in particular, is not a rare process and its occurrence has been suggested and demonstrated for several volcanic and plutonic systems (e.g. Wenzel et al., 2002; Barnes et al., 2005; Piochi et al., 2006; Chadwick et al., 2007; Freda et al., 2008). Thermodynamic modelling of AFC processes involving sialic crust (Bohrson & Spera, 2001; Spera & Bohrson, 2001) has been significantly improved in the past decade thanks to the latest version of the MELTS code (Ghiorso & Sack, 1995). However, modelling based on the thermodynamics of silicate melts (Ghiorso & Kelemen, 1987) is inapplicable to carbonate assimilation–fractional crystallization (AFC) processes whose description still remains semi-quantitative. In fact, because of low CO2 solubility in the melt, carbonate assimilation results in a three-phase (solid + melt + gas) process characterized by a reduction of aH2O and by an increase of fO2 in the silicate melt (Wenzel et al., 2002). Moreover, Freda et al. (2008) demonstrated that the decrease of SiO2 activity that occurs during carbonate assimilation cannot be simply explained as a dilution effect, but it is also related to the stability of CaO-free mineral phases (i.e. oxide, leucite, and mica) with respect to clinopyroxene.
The Colli Albani Volcanic District (hereafter CAVD) represents one of the best examples of a magmatic plumbing system emplaced within a thick carbonate sequence. Mass-balance calculations, together with geochemical and experimental data (Dallai et al., 2004; Gaeta et al., 2006; Iacono Marziano et al., 2007; Freda et al., 2008), indicate that the CAVD magmatic system experienced strong interaction with carbonate rocks during magma differentiation. Notwithstanding, the only evidence of carbonate entrained within the magma is the presence of carbonate xenoliths in differentiated pyroclastic rocks (carbonate xenoliths are never found in lava flow deposits). We believe that these xenoliths were entrained during the fragmentation process and played a role in enhancing the explosivity of the eruptions (Freda et al., 1997) but cannot be considered responsible for primitive magma contamination and differentiation. The aim of this study is to provide insights into the large-scale magma–carbonate interaction process occurring in the magma chamber during magmatic contamination and differentiation.
The most primitive rocks that crop out in the CAVD are leucite-bearing phonotephrite lava flows (Fornaseri et al., 1963; Peccerillo et al., 1984; Trigila et al., 1995). Considering their aphyric microtexture and their very high liquidus temperature at 1 atm (Thompson, 1977), these lava flows were believed to represent mantle-derived parental magma compositions (Ferrara et al., 1985). Nevertheless, Freda et al. (2008), on the basis of crystallization experiments, demonstrated that olivine is not a stable phase in CAVD phonotephritic melts. This observation contrasts strongly with experimental petrology data on primitive ultrapotassic melts (Melzer & Foley, 2000), which show olivine on the liquidus over a wide temperature range. The lack of olivine phenocrysts in primitive rocks suggests that carbonate assimilation processes affect the stability of liquidus phases in primitive magmas (Gaeta et al., 2006). Experimental data on limestone assimilation by hydrated basaltic magmas indicate that carbonate incorporation favours the crystallization of clinopyroxene whereas olivine is consumed (Iacono Marziano et al., 2008). Skarn rocks (Einaudi et al., 1981) forming at magma–carbonate interfaces are indeed characterized by the occurrence of abundant clinopyroxene (Tilley, 1952; Joesten, 1977; Baker & Black, 1980; Joesten et al., 1994; Cioni et al., 1995; Owens, 2000; Wenzel et al., 2002; Fulignati et al., 2004). At the CAVD, in particular, skarns contain mineral phases, such as Ca-Tschermaks (CaTs)-rich clinopyroxene, Mg-rich phlogopite, Al-rich spinel, and Fo-rich olivine (Trigila et al., 1995), not found in the juvenile volcanic products and characterized by 
18O values much higher (>10
SMOW, Barbieri et al., 1975) than those measured in minerals from the volcanic rocks (Dallai et al., 2004). This could suggest that the skarn rock-shell acts as a barrier between the carbonate wall-rocks and the magma, preventing further assimilation. Gaeta et al. (2006) and Freda et al. (2008), however, have demonstrated that in the CAVD magmatic system, carbonate assimilation plays an important role in each step of the liquid line of descent. Thus, the rock-shell should represent the endoskarn (or magmatic skarn, Kerrick, 1977; Fulignati et al., 2004) grading into a cumulate reaction zone (Beard et al., 2005), in turn affected by high CaTs-clinopyroxene activity. Previous unsuccessful attempts at modelling carbonate AFC (Fowler et al., 2007) can be related to the scarcity of information on this peculiar cumulate reaction zone, located far from the carbonate wall-rocks but characterized by high CaTs-clinopyroxene activity.
In this study, by using microtextural observations, mineral chemistry, and bulk-rock geochemical data combined with MELTS simulations, we highlight the effect of high CaTs activity on the stability of Cr-spinel, olivine and clinopyroxene, and we define a reaction-cumulate zone where clinopyroxene crystallization is favoured. Finally, we explain how magmatic processes occurring in this reaction-cumulate zone can be responsible for the distinctive CAVD liquid line of descent.
|
GEOLOGICAL BACKGROUND |
|---|
The CAVD is one of the main volcanic districts of the ultrapotassic Roman Province (Peccerillo, 2005
), developed along the Tyrrhenian margin of Italy during Quaternary times (Fig. 1). The district is rooted in an overthickened carbonate sequence (Funiciello & Parotto, 1978; Chiarabba et al., 1997) whose top was located at about 2 km depth and thickness estimated up to about 5 km (Bianchi et al., 2008). Consistent with a magma reservoir located within carbonate wall-rocks, petrological and experimental data indicate a pre-eruptive crystallization pressure of about 0·2–0·3 GPa and temperatures of about 1050–1150°C (Freda et al., 1997, 2008). The volcanic history (see Marra et al., 2003, and references therein) of the CAVD (Fig. 1) is volumetrically dominated by the explosive activity of the Tuscolano–Artemisio phase (Fornaseri et al., 1963; De Rita et al., 1988). This phase started at 561 ka and ended at 366 ka, with the formation of the Tuscolano–Artemisio caldera. A second phase, the Faete phase (De Rita et al., 1988), is characterized by mainly effusive activity, subordinately accompanied by Strombolian activity, and spanned the interval 308–250 ka. Finally, a hydromagmatic phase characterized by pyroclastic surge eruptions from multiple tuff rings and by the lack of effusive events started in several centres. The Albano Maar represents the most recent (< 70 ka) and voluminous activity of the hydromagmatic phase (Freda et al., 2006). This activity is represented by an 80 m thick succession of pyroclastic deposits, resting above an 
1·5 m thick mature paleosol developed on earlier Ariccia Maar hydromagmatic deposits. Coarse breccia levels, ranging from matrix-supported, lithic-rich zones within laminated ash levels to matrix-poor beds, occur sporadically throughout the Albano Maar pyroclastic succession. These breccia levels are characterized by the occurrence of abundant granular, mafic, lithic clasts and loose clinopyroxene megacrysts (Freda et al., 2006). On the basis of literature data (Federico et al., 1994; Trigila et al., 1995; Federico & Peccerillo, 2000; Gaeta et al., 2000, 2006; Dallai et al., 2004), the clasts can be divided in two populations: CaTs-rich skarns and phlogopite-bearing orthocumulates. In addition to these two populations, a new type of granular, mafic, lithic clast, occurring in the breccia levels and not previously described, is presented in this study.
|
|
SAMPLE SELECTION AND ANALYTICAL METHODS |
|---|
 TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND SAMPLE SELECTION AND ANALYTICAL... PETROGRAPHYMINERAL CHEMISTRYGEOCHEMISTRYRECALCULATED MELT INCLUSIONSDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
More than 100, centimetre-sized, granular, mafic, lithic clasts and a few loose centimetre-sized clinopyroxenes were sampled from a breccia level of ‘Unit a’ of the Albano Maar succession (Fig. 1; Freda et al., 2006
). Scoria clasts, occurring in the distal equivalent (DU1 of Giaccio et al., 2007) of ‘Unit a’, were also collected. Microscope and X-ray diffraction analyses of the collected granular, mafic, lithic clasts show a mineralogical assemblage made up of abundant clinopyroxene and variable amounts of olivine, mica, oxide and glass. As stated above, these clasts have been studied previously and classified as CaTs-rich skarns (CaTsSK) and phlogopite-bearing orthocumulates (Federico et al., 1994; Trigila et al., 1995; Federico & Peccerillo, 2000; Gaeta et al., 2000, 2006; Dallai et al., 2004). However, the new sampling performed during this study allowed us to identify a third type of granular, mafic, lithic clast. This new type is an olivine–clinopyroxene cumulate (OCC) characterized by (1) the occurrence of olivine + dark brown spinel + colourless clinopyroxene and (2) the absence of mosaic textures, thermometamorphic phases (i.e. green spinel), and phlogopite. This study is focused on the OCC samples and on CaTsSK samples that contain interstitial glass. We selected nine OCC and two CaTsSK samples for detailed geochemical analysis. Additionally, we analysed one centimetre-sized clinopyroxene megacryst, sampled as a loose crystal in the same breccia level, scoria clasts from the distal equivalent of the breccia level, and melt inclusions in olivines from both OCC and scoria clasts.
Analyses of glasses and crystals were performed at the CNR-Istituto di Geologia Ambientale e Geoingegneria (Rome, Italy) with a Cameca SX50 electron microprobe (EMP) equipped with five wavelength-dispersive spectrometers using 15 kV accelerating voltage, 15 nA beam current, 10 µm beam diameter, and 20 s counting time. The following standards were used: wollastonite (Si and Ca), corundum (Al), diopside (Mg), magnetite (Fe), rutile (Ti), orthoclase (K), jadeite (Na), barite (Ba and S), celestine (Sr), F-phlogopite (F), sylvite (Cl), apatite (P) and metals (Cr, Ni and Mn). Ti and Ba contents were corrected for the overlap of the TiK
and BaK peaks. The precision of the analytical totals of the glasses was estimated as follows: we doped a natural basaltic andesite powder with a known amount of water and fused it to a glass in an internally heated pressure vessel; then we analysed the glass by using both the Karl Fischer titration (KFT) technique [details have been given by Behrens (1995)] and the EMP. The KFT analysis gives a H2O content of 2·23 wt % (±0·04 wt %) whereas the EMP analyses give (100 – Total) = 1·96 wt % with a 1
= 0·28. By assuming the KFT value as the true value, the relative EMP error closure is 12%. The complete dataset of EMP analyses is available as an Electronic Appendix (Tables 1EA–4EA), which can be downloaded from http://www.petrology.oxfordjournals.org.
|
Trace elements in glasses and crystals were measured at the CNR-Istituto di Geoscienze e Georisorse (Pavia, Italy) using a pulsed Nd:YAG laser working at 213 nm coupled to an inductively coupled plasma sector field mass spectrometer (Element I by ThermoFinnigan). Instrumental and analytical details have been reported by Tiepolo et al. (2003
). For this study the laser was operated at a repetition rate of 10 Hz, and the spot diameter was set at 40 µm with a pulse energy of about 0·1 mJ. Ablation signal integration intervals were selected by carefully inspecting the time-resolved analysis to ensure that no inclusions were present in the analysed volume. Data reduction was performed using the software package ‘Glitter’ (van Achterbergh et al., 2001). NIST SRM 612 was used as an external standard and 44Ca was used as an internal standard. Reproducibility and accuracy of trace element concentrations were evaluated for a control sample BCR2-glass [MUN, inductively coupled plasma mass spectrometry (ICP-MS) unpublished data]. For this sample the error on reproducibility and accuracy is < 7% and < 10%, respectively.
Stable isotopes were measured on olivines and clinopyroxenes, separated by handpicking under the microscope, by laser fluorination (Sharp, 1995) at the CNR-Istituto di Geoscienze e Georisorse (Pisa, Italy) using a 15 W CO2 laser operating at a wavelength of 10·6 µm. Pure fluorine, desorbed at 290–310°C from hexafluoropotassium-nickelate salt (Asprey, 1976) and stored in an F2 reservoir, was used as the reagent. The O2 produced during laser fluorination was purified of excess fluorine by means of a KCl trap, and Cl was trapped cryogenically. The gas was then transferred to a 13 Å molecular sieve-filled cold finger. Further oxygen gas purification was achieved by desorbing oxygen from the molecular sieve at about –110°C using a liquid nitrogen–ethanol mixture. In this way NFx compounds were retained on the 13 Å zeolites (Clayton & Mayeda, 1983). The gas was then analysed for oxygen isotope composition using a Finnigan Delta Plus XP mass spectrometer. Four to seven aliquots of laboratory quartz standards (QMS, 18O = 14·05; L1, 18O = 18·15) were normally analysed, with an average reproducibility of ± 0·12 (1), during each set of analyses. During the period of study seven aliquots of NBS 28 standard (accepted 18O = 9·60) were also measured, with an average 18O = 9·56 ± 0·16 (1). No data correction was necessary, and results are reported in the standard per mil notation. All 18O values are relative to SMOW.
|
PETROGRAPHY |
|---|
Olivine–clinopyroxene cumulates (OCC)
Based on their texture, OCCs can be subdivided into three types, heteradcumulate, adcumulate (Irvine, 1982
; McBirney & Hunter, 1995; Mitchell et al. 1998), and ‘open’ cumulate (Gaffney, 2002). The OCC modal mineralogy ranges from olivine clinopyroxenite to wehrlite (Table 1). Interstitial glass is scarce, and is completely transformed into a cryptocrystalline aggregate of zeolites. Because of their lack of cohesion, these aggregates are easily removed from the rocks (e.g. during thin-section preparation) leaving empty spaces that make it difficult to estimate the amount of pristine glass in the sample (for this reason this is not reported in Table 1). Kink-bands are rare whereas crystal-free melt inclusions are abundant; both features are indicative of subsolidus fast cooling, in agreement with the sub-volcanic nature of the OCC rocks. In association with the OCCs, we sometimes find clinopyroxene megacrysts, characterized by rare, small (
100 µm) olivine inclusions.
Heteradcumulates comprise a touching framework of poikilitic, millimetre- to centimetre-sized, clinopyroxene crystals including olivine (Fig. 2a and Photo 1EA provided as an Electronic Appendix). The olivine inclusions range from rounded, very small (
100 µm) grains to sub-euhedral, larger grains (Fig. 2b and Photo 2EA provided as an Electronic Appendix). They are generally randomly distributed in the host clinopyroxene but the larger grains, sometimes forming grain clusters, often occur in the central part of the poikilitic clinopyroxene (Fig. 2b). In turn, the olivines host rounded spinel crystals and ‘dusty’ clouds made up of spinel ± clinopyroxene ± mica ± glass (Fig. 2c and d).
|
Adcumulates (Fig. 2e) comprise a touching framework of subhedral, millimetre- to submillimetre-sized, olivine and clinopyroxene grains. A very small amount of interstitial, zeolitized glass is also present. Spinels are rare and, unlike those in the heteradcumulates, they are euhedral (Fig. 2e).
‘Open’ cumulates are characterized by euhedral to subhedral, millimetre- to submillimetre-sized olivine and clinopyroxene grains. The ‘open’ texture, caused by the touching framework of crystals with well-developed habits, means that these cumulates have a higher porosity and, consequently, the highest amount of interstitial zeolitized glass (Fig. 2f).
Interpretation of OCC texture
Heteradcumulate textures can result from a number of different crystallization processes. This texture, for example, can be simply related to the different rates of nucleation and growth (Vernon, 2004) of olivine and clinopyroxene during cotectic crystallization (OlN > CpxN and CpxG > OlG, where N and G are the velocity of nucleation and growth, respectively). Accordingly, olivines enclosed in poikilitic clinopyroxenes should be euhedral and smaller in the core, becoming larger towards the rim. There is not evidence of such an olivine size distribution in the OCC heteradcumulates (Fig. 2b). On the other hand, Irvine (1982) suggested that heteradcumulate textures could be the result of a peritectic replacement reaction between cumulus phases. Accordingly, as observed in poikilitic orthopyroxene (Gaffney, 2002), olivines enclosed in pyroxenes should be rounded in shape. OCC heteradcumulates are characterized by small rounded olivines enclosed in poikilitic clinopyroxene (Fig. 2a). Given this, we believe that the OCC heteradcumulates formed as a result of a reaction between olivine and clinopyroxene.
Adcumulate textures can be simply explained as the result of the aggregation of olivine and clinopyroxene during their cotectic crystallization. The same mechanism can explain the formation of the ‘open’ cumulates, but in a zone of the magma chamber richer in melt.
Ca-Tschermak-rich skarn (CaTsSK)
Ca-Tschermak-rich skarn samples are made up of abundant, submillimetre-sized, xenomorphic clinopyroxene, olivine, phlogopite, and green spinel (Fig. 2g). Inclusions of calcite are common in both clinopyroxene and spinel, whereas phlogopite inclusions are common in olivine. These skarns also contain, mostly between clinopyroxene crystals, pools of glass enclosing euhedral green spinel (Fig. 2h and Photo 3EA provided as an Electronic Appendix). Such a feature has never been reported previously (Trigila et al., 1995). The presence of glass in textural equilibrium with spinel suggests high-temperature formation; thus these rocks can be classified as endoskarns or magmatic skarns (Kerrick, 1977; Fulignati et al., 2004).
Melt inclusions
The studied melt inclusions occur in variable amounts, either isolated or in clusters, in olivines from the OCC samples. They mostly show primary features (Roedder, 1979), with rounded or negative olivine shapes (Fig. 3a and b), and usually consist of silicate glass containing one or more shrinkage bubbles and, rarely, daughter minerals (generally phlogopite). By means of an image analysis program, we calculated the volume of bubbles (Vb) and melt inclusions (VMI) by assuming a spherical shape for bubbles and rounded melt inclusions and a parallelopiped shape for negative-shaped melt inclusions. For the latter we assumed the average of the two measurable dimensions in thin section as the third dimension. In Fig. 3c we plot Vb vs VMI (for both rounded and negative-shaped melt inclusions), and obtain a positive linear correlation that according to Frezzotti (2001) is indicative of in situ vapor immiscibility after homogeneous trapping. The melt inclusions we selected for geochemical analysis occur in sub-euhedral olivines, show rounded or negative olivine shapes, are up to 250 µm in size, colourless to pale brown in colour, and free of daughter minerals.
|
|
MINERAL CHEMISTRY |
|---|
Olivines in the OCC samples are the most magnesian (Table 2) so far reported among the CAVD cumulates (Trigila et al., 1995
; Dallai et al., 2004; Gaeta et al., 2006). In particular, two olivine populations can be distinguished (Table 2, Fig. 4): olivines with Fo contents around 90 and with Fo contents of about 88. The majority of the studied OCC samples are characterized by the occurrence of Fo90 olivines and we will refer to them as O90CC, whereas O88CC will be used for those samples characterized by Fo88 olivines. Olivine inclusions in the clinopyroxene megacryst are Fo90 in composition. Olivines in CaTsSK have very high Fo contents (up to 96), high CaO, and very low NiO contents (Table 2, Fig. 4). Notably, all olivines occurring in juvenile volcanic products (i.e. scoria clasts associated with OCC and CaTsSK) have Fo contents around 88 (Fig. 4).
|
|
Clinopyroxenes in OCC samples plot on the Di–Hd joint of the Diopside–Hedenbergite–Ferrosilite–Enstatite quadrilateral and are characterized by very high CaO contents (Table 3). In particular, the O90CC clinopyroxenes show the widest range of CaO and Cr2O3 contents, whereas the O88CC clinopyroxenes cluster around relatively high CaO and low Cr2O3 values (Fig. 5 a and b). O90CC and O88CC clinopyroxenes have comparable amounts of Al2O3 (Fig. 5c). Notably, poikilitic clinopyroxenes from O90CC samples and the megacryst are characterized by an Al2O3 content higher than the cumulus cpx (Fig. 5 c and d); this Al2O3 enrichment is often coupled with Cr2O3 enrichment in the rim (Table 3). Clinopyroxenes in CaTsSK are characterized by lower Cr2O3 and higher Al2O3 contents than the OCC clinopyroxenes (Fig. 5b and c).
|
|
Spinels in OCC are rare. They are all Cr-bearing and show variable habit and composition (Table 4, Fig. 6), with the hercynite–spinelss molecule ranging from 26 to 85 mol %. In particular, Al-poor Cr-spinels are euhedral and occur only in Fo90 olivines associated with cumulus clinopyroxene (adcumulate texture in Fig. 6). These spinels are similar to those enclosed in olivines from phonotephritic lava flows (Freda et al., 2008
) and from Albano Maar scoria clasts (Fig. 6). In contrast, Al-rich Cr-spinels are rounded in shape and are mostly found in Fo90 olivines associated with poikilitic clinopyroxene (heteradcumulate texture in Fig. 6). Rounded spinels can occur either as isolated crystals or (the smallest) associated with Al-rich clinopyroxene, phlogopite, and glass (Fig. 2d). As shown in the Cr-number vs Fe2+/(Fe2+ + Mg) diagram (Fig. 6), spinel compositions form a Cr–Al trend with the Al-richest spinels pointing towards the composition of spinels occurring in the CaTsSK. Notably, OCC spinels are significantly different (Fig. 6) from spinels occurring in the phlogopite-bearing orthocumulates (Trigila et al., 1995).
|
|
|
GEOCHEMISTRY |
|---|
Major elements
The OCC bulk compositions (Table 5) are in agreement with olivine and clinopyroxene modal abundances (Table 1). The scarcity of interstitial glass and absence of intercumulus phlogopite is confirmed by the low Al2O3 and K2O content. CaTsSK bulk compositions (Table 5) are also in agreement with clinopyroxene and olivine modal abundances and with the occurrence of phlogopite and Al-rich spinel in these rocks.
|
Because the interstitial glass occurring in the OCC samples is always replaced by zeolites (Fig. 2f), the only glass suitable for geochemical analysis is that entrapped in olivine. Melt inclusions in O90CC have a trachybasaltic composition whereas those in O88CC have a composition ranging from tephrite to phonotephrite (Table 6, Fig. 7a). All compositions plot in the Group III field of the Foley et al. (1987
) classification for ultrapotassic rocks; moreover, most of the trachybasaltic melt inclusions, according to the same classification, can be considered primitive (MgO >3 wt %, Table 6). It is worth pointing out that this is the first report of primitive trachybasalt compositions in the CAVD stratigraphic record (Fornaseri et al., 1963; Trigila et al., 1995; Gaeta et al., 2006).
|
|
Interstitial glasses in the CaTsSK samples have a silica-poor, CaO-rich chemical composition (Table 6, Fig. 7) comparable with that of interstitial glasses usually found in skarn rocks (e.g. Fulignati et al., 2004
, and reference therein); these compositions are comparable with those of melilitite bulk-rocks (e.g. Di Battistini et al., 2001; D’Orazio et al., 2007).
Trace elements
Olivines in OCC have very low trace element contents (Table 5EA provided as an Electronic Appendix), with the exception of Co (125 ppm) and Cr (200 ppm). Clinopyroxenes in O90CC and O88CC show parallel, convex-upward, chondrite-normalized rare earth element (REE) patterns (Fig. 8), similar to those observed in the orthocumulates studied by Dallai et al. (2004). Notably, the O90CC clinopyroxenes are depleted in 
REE (Table 7, Fig. 8) and enriched in Cr with respect to O88CC and orthocumulates.
|
|
A trace element traverse performed across the clinopyroxene megacryst shows an interesting Cr zoning pattern, with higher values in the core and the rim and lower values between (Fig. 9a and Table 6EA provided as an Electronic Appendix).
|
Melt inclusions in olivines from OCC have high large ion lithophile element to high field strength element ratios (LILE/HFSE;
LILE/HFSE = 2–3) and light REE to heavy REE ratios (LREE/HREE; La/Yb = 30–40), and negative Nb and Ta and positive Pb spikes in primitive mantle-normalized trace element diagrams (Fig. 10a, Table 8). The O90CC inclusions have compatible trace element (e.g. Cr) contents higher than those in O88CC (Fig. 10b). Moreover, the melt inclusions in O90CC olivines, when compared with CAVD primitive lava flows, have higher Cr and lower LREE, Sr, and Ba contents (Fig. 10c), confirming their more primitive composition among the CAVD products. Interestingly, the REE partition coefficients between clinopyroxene and glass, calculated by using data from poikilitic clinopyroxene and trachybasaltic melt inclusions (sample AH3AX25 in Tables 7 and 8), are lower than those typical of ultrapotassic compositions (Foley & Jenner, 2004). We interpret this feature as indicative of ‘disequilibrium conditions’ between the O90CC poikilitic clinopyroxenes and the trachybasaltic melt inclusions in olivine. This interpretation is also confirmed by the textural relationships between olivine and clinopyroxene in the heteradcumulates (see Fig. 2a and b). However, the differences between the partition coefficients might simply indicate the slow enrichment of diffusing elements (such as REE) in the melt inclusions relative to the bulk melt (i.e. a pile-up effect; see Baker, 2008). For this reason we prefer not to use trace element abundances in melt inclusions to define the OCC parental magma, but to constrain it by using trace element abundances in clinopyroxene.
|
|
Interstitial glasses in the two analysed CaTsSK are characterized by high contents of LREE, Nb, U and Th, high LREE/HREE, and low contents of Rb, Cs and Hf. These glasses have Nb/Ta and Zr/Nb ratios ranging from those typical of calciocarbonatites (Nb/Ta = 38, Zr/Nb = 0·82; Chakhmouradian, 2006
) to those of melilitites (Nb/Ta = 21, Zr/Nb = 4; D’Orazio et al., 2007). This suggests that calciocarbonatite and melilitite-like melts can also form in a skarn environment.
Oxygen isotope data
18O values measured in olivines and clinopyroxenes in OCC show a wide 
18OCpx–Ol range (Table 9; Fig. 9b). Such a 18OCpx–Ol variation suggests that the OCC olivines and clinopyroxenes moved from 18O equilibrium (18OCpx–Ol = 0·4 SMOW; magma temperature 1200°C, Chiba et al., 1989) to disequilibrium conditions. In particular, olivines from the O90CC heteradcumulates and O88CC have 18O values higher than those measured in the associated clinopyroxenes. Moreover, the 18OOl and Cpx values are higher than those of uncontaminated primary magmas or mantle rocks (Mattey et al., 1994). The 18O values measured in the clinopyroxene megacryst increase from core to rim (from 5·86 ± 0·07 to 6·46 ± 0·06 SMOW, Table 9 and Fig. 9c) confirming a feature acquired during magmatic differentiation (see Dallai et al., 2004) and not inherited from high 18O parental magmas (Perkins et al., 2006).
|
Olivine and clinopyroxene occurring in the glass-bearing CaTsSK analysed in this study have
18O values (Table 9) ranging from those of volcanic rocks (< 8 SMOW; Dallai et al., 2004) to those of glass-free skarns (>10 SMOW; Barbieri et al., 1975). |
RECALCULATED MELT INCLUSIONS |
|---|
The original composition of melt inclusions is usually modified after trapping and prior to eruption by crystallization of the host phase (Kamenetsky et al., 1995
; Danyushevsky et al., 2000; Frezzotti, 2001). In particular, in the OCC samples, the decrease in Fo content at melt inclusion–host phase contacts (Fig. 3d), is indicative of post-entrapment crystallization (Danyushevsky et al., 2000, 2002). The composition of melt inclusions trapped in olivines from OCC and scoria clasts has therefore been corrected by adding olivine to the original glass composition until equilibrium was reached, according to Kd°l/melt = 0·27 (Kamenetsky et al., 1995). The calculation was performed by means of the ‘Fractional Crystallization Reverse’ option of the program ‘Petrolog’ (Danyushevsky et al., 2000). The Fe2+/Fe3+ ratio was defined by the equation of Maurel & Maurel (1982), log10 (Fe2+/Fe3+)Spl = 0.764 x log10(Fe2+/Fe3+)melt – 0.343, by using the Fe2+ and Fe3+ contents of euhedral Cr-rich spinels (Table 4). According to the model of Kress & Carmichael (1988), the corresponding oxidation state of the trapped melt is close to the nickel–nickel oxide (NNO) buffer. Similar oxidation values were obtained by Kamenetsky et al. (1995) for the most primitive magma of the Roman Province (Mt. Vulsini Volcanic District). Calculations indicate that 9–12 wt % of olivine (T = 100–200°C) must have crystallized to obtain the analysed melt inclusions composition. If we estimate the volume of the Fo <90 zone around the melt inclusions (e.g. the zone between the first and the second points of traverse AB in Fig 3d) we obtain a value of about 11%.
Because of the significant increase in MgO, after correction the compositions of the melt inclusions are clearly more primitive than before. However, their position in the TAS diagram (where only the uncorrected compositions are reported; Fig. 7a), does not change significantly and thus their classification does not need to be changed. In the following discussion we will always refer to the corrected, more primitive, melt composition. Moreover, we estimated the average volatile (H2O + CO2) content in the melt inclusions by using the 100 – EMPAtotal equation (Devine et al., 1995), corrected for Fe3+ and Fe2+ partition, trace element contents (0·3 wt %), and olivine crystallized on the inclusion wall (Table 10). H2O and CO2 concentrations determined by means of Fourier transform IR (FTIR) spectroscopy are uncertain because of the unknown effective thickness of the melt inclusions and the occurrence of bubbles. However, preliminary spectra (Table 7EA provided as an Electronic Appendix) show rather constant H2O and CO2 relative absorbances. By using the obtained CO2/H2O ratio (0·075, 1 = 0·007), we estimated the relative amount of H2O and CO2 wt % in the melt inclusions (Table 10). These volatile amounts, according to experimental solubility data on the CAVD primitive ultrapotassic melts (Behrens et al., 2009), allowed us to estimate a minimum pressure for melt inclusions entrapment of about 200–300 MPa.
|
|
DISCUSSION |
|---|
Here we address: (1) the origin of the trachybasaltic melt inclusions in terms of primary vs differentiated magmas; (2) the olivine–clinopyroxene cumulate-forming process during carbonate assimilation, focusing on the stability of Cr-spinel, olivine, and clinopyroxene during the crystallization of parental magma in a zone close to, but not in contact with the carbonate wall-rocks; (3) the effect of this cumulate zone on magmatic differentiation.
Origin of trachybasaltic melts
The Quaternary ultrapotassic volcanic districts of Italy are commonly characterized by the occurrence of Fo-olivine, Cr-spinel-bearing basanitic, trachybasaltic, and tephritic volcanic rocks (Civetta et al., 1981; Rogers et al., 1985; Kamenetsky et al., 1995; Perini et al., 2004). These products, generally considered as representative of the parental mantle-derived magmas (Peccerillo, 2005), are absent in the CAVD stratigraphic record (Fornaseri et al., 1963; Peccerillo et al., 1984; Trigila et al., 1995). Nevertheless, the occurrence of very primitive parental magmas in the CAVD plumbing system is recorded by the mineral chemistry of olivines, clinopyroxenes, and spinels from O90CC. In addition, as confirmed by trace element concentrations (Fig. 10), trachybasaltic melt inclusions in O90CC olivines have (Fig. 7a) the most primitive compositions so far analysed among the CAVD products. Notably, these trachybasaltic glasses exhibit a CaO/Al2O3 ratio (Fig. 7b) higher than that typical of Roman Province primitive rocks with similar MgO contents (> 4 wt %). Such a high CaO/Al2O3 ratio in the parental melts can be explained by assuming either a large amount of clinopyroxene in the mantle source (Foley, 1992) or a carbonate-bearing peridotitic mantle source (e.g. Italian melilitites from Mt. Vulture and Mt. Vulsini; Di Battistini et al., 2001; D’Orazio et al., 2007). Partial melting of mantle sources with a high clinopyroxene/olivine ratio would produce melts with a high CaO/Al2O3 ratio, as well as high SiO2 contents. In contrast, the O90CC trachybasaltic melt inclusions have silica contents similar to or lower than those of Roman Province primitive rocks (Fig. 11). On the other hand, the O90CC melt inclusions have silica contents that are not compatible with melts formed from a carbonate-bearing mantle source (e.g. melilitites, carbonatites, Lee et al., 2000; Fig. 11). Both features rule out a mantle origin for the CAVD melt inclusions of trachybasaltic composition.
|
Assuming that the most primitive CAVD melt composition is a differentiated melt, the best candidate as a parental magma is the Mt. Vulsini leucite basanite lava flow (Rogers et al., 1985
; Kamenetsky et al., 1995) whose 87Sr/86Sr signature is comparable with that of the CAVD clinopyroxenes (Di Battistini et al., 1998; Gaeta et al., 2006). This leucite basanite, assumed to be representative of the Roman Province primary magma composition, is characterized by phenocrysts of Fo90–91 olivine, Di-rich clinopyroxene, and Cr-rich spinel; the CAVD O90CC adcumulates are characterized by the same mineral assemblage. Moreover, according to trace element partition coefficient data (Foley & Jenner, 2004), the O90CC clinopyroxenes could be in equilibrium with the leucite basanite composition. For these reasons, the latter can be assumed as the parental magma of the CAVD trachybasaltic melts. Nevertheless, mass-balance calculations fail (least-squares residuals, r2 = 1· 8) to produce the trachybasaltic melt inclusions composition (AH3AX25 in Table 10) from the leucite basanitic parental melt (LBSN in Table 10 and Fig. 11) by means of clinopyroxene + olivine ± spinel fractionation (AH3AX25 in Tables 3, 2 and 4, respectively) in a closed system. Dallai et al. (2004) demonstrated the need for a contribution from an external source of 18O during the CAVD magmatic differentiation, confirmed here by the 18O core–rim variation in the clinopyroxene megacryst. In detail, the 18O core–rim variation measured in the megacryst (from 5·86 to 6·46 SMOW, Fig. 9c) can be obtained by assuming the assimilation of about 3 wt % of calcite according to the simple mass-balance equation 1 – [(18Olimestone – 18OCpx)/18Olimestone] = assimilated limestone (where 18Olimestone = 25 SMOW and 18OCpx = 6·46–5·86 SMOW). Despite the contribution of carbonate during differentiation, the mass-balance calculations fail again if we use as the fractionated solid the mineral phases mentioned above. To obtain low values of the least-squares residuals (r2 0·1), we must assume not only the assimilation of about 3 wt % of calcite but also a solid fraction made of CaTs-rich clinopyroxene + Fo-rich olivine + Al-rich spinel (AH3AX65 in Table 3, Table 2, and Table 4, respectively). Low values of the least-squares residuals are also obtained by using a combination of O90CC and CaTs clinopyroxene (CaAl2SiO6) (e.g. AH3AX25Cpx + CaTs Cpx + AH3AX25Ol). Because CaTs-rich clinopyroxene and Al-rich spinel phases are absent in the volcanic rocks but common in the CAVD endoskarns, a significant contribution of the latter during the differentiation of the leucite basanite parental melt should be taken into account.
Olivine–clinopyroxene cumulate (OCC) zone
Most of the OCC samples are characterized by the occurrence of Fo90 olivines, which are not present in the juvenile volcanic component (see scoria clasts in Fig. 4) of the studied deposits. For this reason, we infer that O90CC could represent the ‘non-eruptible’ magmatic zone (i.e. a mush zone forming close to the wall-rocks; Marsh, 1995) of the Albano maar pre-eruptive system. Commonly, the rocks in direct contact with limestone wall-rocks are skarns (e.g. CaTsSK); we therefore assume that the O90CC zone is located between the skarns and the main body of magma and that the petrological processes occurring in this zone have to be strongly controlled by the CaTsSK rocks. This hypothesis is supported by the microtextural characteristics and the mineral chemistry of the O90CC rocks.
For example, the O90CC olivines associated with poikilitic clinopyroxene exhibit microtextural features (Fig. 2 a and c) indicative of phase instability (Shelley, 1993; Vernon, 2004), also confirmed by the Fo90–Fo88 gap observed in the olivine compositions (Fig. 4). Primitive ultrapotassic magmas show olivine as a liquidus phase over a wide temperature range (Melzer & Foley, 2000), and MELTS simulations (Ghiorso & Sack, 1995), performed at 300 MPa and fO2 = NNO (inferred from the Cr-rich spinel composition) using the leucite basanite melt composition (LBSN in Table 10 and Fig. 11), indicate Fo90 olivine as the liquidus phase as well as olivine + spinel + clinopyroxene co-saturation (1180°C, Fig. 11). The O90CC samples could thus represent a cumulate zone in which, because of carbonate assimilation, the increase of fO2 reduces the olivine stability field (Wenzel et al., 2002). Following this hypothesis, we performed MELTS simulations on the leucite basanite composition at 300 MPa and higher fO2 (NNO + 2) and actually found clinopyroxene as the liquidus phase. However, the latter clinopyroxene composition exhibits Fe3+ enrichment (Fe3+/Fe2+ >1) never observed in the O90CC clinopyroxenes.
In summary, we believe that the olivine instability observed in the O90CC rocks can be explained by means of an increase in the CaTs clinopyroxene activity in the parental magma (i.e. leucite basanite) according to the reaction
|
| (1) |
Further insights into the influence of CaTs-clinopyroxene activity on the parental magma come from the geochemical characteristics of the Cr-spinels in O90CC. In particular, heteradcumulate O90CCs with Al-rich poikilitic clinopyroxenes (Fig. 5) are characterized by Cr-spinels that are rounded in shape and exhibit the so-called ‘Cr–Al trend’ (Fig. 6; Irvine, 1967). Various mechanisms have been suggested to explain this trend. Barnes & Roeder (2001), for example, explained it as resulting from the re-equilibration of Mg–Al-rich spinel and Al-bearing pyroxene during magma differentiation, whereas Holness et al. (2007) suggested that it is caused by the reaction of spinel with olivine and plagioclase. Alternatively, the Cr–Al trend of spinel observed in O90CC (i.e. the enrichment of Al in the Cr-spinels; Fig. 6) could reasonably be related to the increase of CaTs-clinopyroxene activity in the parental magma because of its proximity to CaTsSK according to the reaction
|
| (2) |
The effectiveness of reaction (2) is supported by the non-equilibrium shape of the Al-rich Cr-spinels and by the evidence that they are often associated with Al-rich clinopyroxenes (Fig. 2d). Moreover, the high–low–high zoning of Cr coupled with the increase, from core to rim, of 18O values in the clinopyroxene megacryst (Fig. 9 a and c), support a petrological process involving the dissolution of both Cr-rich phases (i.e. Cr-spinel) and 18O-rich phases (i.e. CaTs-rich clinopyroxene).
Finally, combining reactions (1) and (2) we obtain
|
| (3) |
On the basis of the above reactions, we suggest that the O90CC rocks form in a ‘reactive assimilation zone’ (Beard et al., 2005) where the parental magma interacts with CaTsSK. Despite this, one could argue that assimilation processes in the O90CC zone, as well as in the whole magmatic system, might be prevented by the occurrence of the skarn rocks themselves. The latter, crystallizing at the magma–carbonate contact zone during the very early stage of magma emplacement, might indeed create a barrier between the magma and the carbonate wall-rocks. However, magma–carbonate interaction processes have been inferred for many magmatic systems (e.g. Wenzel et al., 2002; Barnes et al., 2005; Piochi et al., 2006; Chadwick et al., 2007), and for the CAVD in particular Gaeta et al. (2006) and Freda et al. (2008) have demonstrated that carbonate assimilation is an effective process during each step of the liquid line of descent. To overcome the ‘skarn barrier’ problem, Wenzel et al. (2002) suggested that contamination of the magma could occur through calcite melts originating in the skarn zone itself. Nevertheless, neither Wenzel et al. (2002) nor we have any evidence for the existence of calcite melt (sensu strictu) in the skarns. Actually, the CAVD CaTsSK are characterized by the occurrence of CaO-rich silicate glasses (Table 6) with major element concentrations comparable with those of glasses occurring in Vesuvius skarns (Fulignati et al., 2004) and trace element and 18O values comparable with those of the calciocarbonatites occurring at Mt. Vulture (D’Orazio et al., 2007). The absence of negative Nb and Ta spikes in the mantle-normalized trace element pattern of the CaO-rich silicate glasses (Fig. 10a) suggests an origin different from that inferred for the mantle-derived CAVD magmas (Peccerillo, 2005). Preliminary geochemical and experimental data (Gaeta et al., 2007), indeed, suggest that the endoskarn CaO-rich silicate melts represent one of the first products of the ultrapotassic magma–carbonate interaction.
The occurrence of CaO-rich silicate glasses in the CaTsSK rocks allows us to infer that the CaTs-rich component acting as a contaminant in the reactive assimilation process summarized in reaction (3) (CaTs Cpx on the left side of the reaction) could be a melt. The interaction between such a CaO-rich silicate melt and the parental magma can be described by the reaction
|
| (4) |
1180°C), the addition of such a CaO-rich silicate melt to the parental magma, even in small amounts, causes a significant increase in Cpx/Ol and Cpx/Spl mass ratios, a decrease of Cr-number in spinel (Fig. 12), and the formation of a silica-depleted differentiated melt [i.e. Melt3 in reaction (4)]. The formation of silica-depleted melt compositions can be explained by a silica dilution effect (parental magma mixed with Melt2), but also by the increase of clinopyroxene stability with respect to olivine and spinel (Cpx effect, Fig. 12d). Moreover, Ussler & Glazner (1992) and Fowler et al. (2007) have demonstrated that a mixing process with a high parental magma to CaO-rich silicate melt ratio, as modelled in Fig. 12, should not be highly dependent upon the initial CaO-rich silicate melt temperature.
|
Effect of OCC zone on magmatic differentiation
Our model of a reactive assimilation process involving the CaO-rich silicate melt (AH3AX65 in Table 6) and the leucite basanite parental magma (LBSN in Table 10) in the OCC zone is illustrated schematically in Fig. 13 [using nomenclature according to Marsh (1995
)].
|
According to this model at time t0 a parental magma [PM in reaction (4), corresponding to leucite basanite in Fig. 13] is emplaced into limestone wall-rocks. At time t1 an endoskarn zone forms and the magmatic system is characterized by a ‘suspension’ of olivine crystals [Fo90 in reaction (4) and Fig. 13] and a ‘free’ magma. At time t1 + dt the magma zone close to the wall-rock reaches the clinopyroxene saturation temperature (1180°C in Fig. 13); under these conditions, both olivine and clinopyroxene are stable. At this stage, the magmatic system is made up of three zones: free magma, Fo90 suspension, and olivine + clinopyroxene mush [Fo90 + Di and CpxR in Fig. 13]. In particular, the CpxR zone forms when the intercumulus melt (in the olivine + clinopyroxene mush) becomes olivine-undersaturated as a consequence of CaO-rich silicate melt assimilation. This results in the crystallization of poikilitic clinopyroxene and dissolution of olivine, according to reaction (4) and the O90CC heteradcumulate texture. At time t2 the intercumulus melt in the olivine + clinopyroxene mush has a composition intermediate between trachybasalt and phonotephrite, and Fo90 olivine is no longer stable. At this stage the O90CC mush stops forming and Fo88 olivine starts to crystallize. Between time t2 and t3, the O88CC forms and eventually (at t3), the magmatic system becomes unstable and the phonotephritic Fo88-bearing magma erupts. If we assume a short time interval between t1 and t2, trachybasaltic and/or less differentiated melts remain as free, ‘eruptible’ magmas for a very short period. This assumption could explain the absence in the CAVD stratigraphic record of primitive volcanic products (i.e. from leucite basanitic to trachybasaltic in composition).
|
CONCLUSIONS |
|---|
Mafic magma chambers emplaced in carbonate wall-rocks are generally characterized by the association of CaTs-rich clinopyroxene-bearing endoskarns and cumulate rocks. The latter, by recording features of the magmatic differentiation process, such as magma contamination, can give insights into the nature of the contaminant, the contamination mechanism, and the contaminated products, as follows.
- Nature of the magma contaminant: because of their liquid state, the CaO-rich silicate melts forming at the beginning of the assimilation process between the primitive ultrapotassic magma and the carbonate wall-rock can be considered as the best candidate for the contaminant.
- Contamination mechanism: the CaO-rich silicate melts contaminate the olivine–diopside cumulate zone, inducing the crystallization of poikilitic clinopyroxene and the dissolution of olivine (± Cr-spinel).
- Contaminated products: the contamination process results in the formation of heteradcumulate rocks and differentiated melts (SiO2-poor).
|
SUPPLEMENTARY DATA |
|---|
Supplementary data for this paper are available at Journal of Petrology online.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to L. Dallai, M. Serracino, and M. Tiepolo for help during
18O, EMP, and LAM-ICP-MS analyses, respectively. H. Behrens is thanked for providing the H2O and CO2 FTIR data we used for the volatile content estimation in melt inclusions, and for the synthesis and analyses of the basaltic andesitic glass. Constructive reviews by C. G. Barnes, M. C. S. Humphreys, and an anonymous referee improved the manuscript. We are indebted to W. A. Bohrson for her thoughtful and useful comments. This research is part of the PhD program of T.D.R. and was funded by INGV-DPC Project V3_1 Colli Albani and MIUR ‘Caratteristiche vulcanologiche e geomeccaniche dei prodotti recenti dei Colli Albani’.
*Corresponding author. E-mail: mario.gaeta{at}uniroma1.it
|
REFERENCES |
|---|
Asprey LB. The preparation of very pure F2 gas. Journal of Fluorine Chemistry (1976) 7:359–361.[CrossRef][Web of Science]
Baker CK, Black PM. Assimilation and metamorphism at a basalt–limestone contact, Tokatoka, New Zealand. Mineralogical Magazine (1980) 43:797–807.[Web of Science]
Baker DR. The fidelity of melt inclusions as records of melt compositions. In: Contributions to Mineralogy and Petrology (2008) doi:10.1007/s00410-008-0291-3.
Barbieri M, Penta A, Turi B. Oxygen and strontium isotope ratios in some ejecta from the Alban Hills volcanic area, Roman Comagmatic Region. Contributions to Mineralogy and Petrology (1975) 51:127–133.[CrossRef][Web of Science]
Barnes SJ, Roeder PL. The range of spinel composition in terrestrial mafic and ultramafic rocks. Journal of Petrology (2001) 42:2279–2301.
Barnes C, Prestvik T, Sundvoll B, Surratt D. Pervasive assimilation of carbonate and silicate rocks in the Hortavær igneous complex, north–central Norway. Lithos (2005) 80:179–199.[CrossRef][Web of Science]
Beard JS, Ragland PC, Crawford ML. Reactive bulk assimilation: A model for crustal–mantle mixing in silicic magmas. Geology (2005) 33:681–684.
Behrens H. Determination of water solubilities in high viscosity melts: an experimental study of NaAlSi3O8 and KAlSi3O8 melts. European Journal of Mineralogy (1995) 7:905–920.
Behrens H, Misiti V, Freda C, Vetere F, Botcharnikov RE, Scarlato P. Solubility of H2O and CO2 in potassic melts at 1200 and 1250°C and pressure from 50 to 500 MPa. American Mineralogist (2009) 94:105–120.
Bianchi I, Piana Agostinetti N, De Gori P, Chiarabba C. Deep structure of the Colli Albani volcanic district (central Italy) from receiver functions analysis. Journal of Geophysical Research (2008) 113. B09313, doi:10.1029/2007JB005548.
Bohrson WA, Spera FG. Energy-constrained open-system magmatic processes II: Application of energy-constrained assimilation–fractional crystallization (EC-AFC) model to magmatic systems. Journal of Petrology (2001) 42:1019–1041.
Chadwick JP, Troll VR, Ginibre C, Morga D, Gertisser R, Waight TE, Davidson JP. Carbonate assimilation at Merapi Volcano, Java, Indonesia: insights from crystal isotope stratigraphy. Journal of Petrology (2007) 48:1793–1812.
Chakhmouradian AR. High-field-strength elements in carbonatite rocks: geochemistry, crystal chemistry and significance for constraining the sources of carbonatites. Chemical Geology (2006) 235:138–160.[CrossRef][Web of Science]
Chiarabba C, Amato A, Delaney PT. Crustal structure, evolution, and volcanic unrest of the Alban Hills, Central Italy. Bullettin of Volcanology (1997) 59:161–170.[CrossRef]
Chiba H, Chacko T, Clayton RN, Goldsmith JR. Oxygen isotope fractionations involving diopside, forsterite, magnetite, and calcite; application to geothermometry. Geochimica et Cosmochimica Acta (1989) 53:2985–2995.[CrossRef][Web of Science]
Cioni R, Civetta L, Marianelli P, Metrich N, Santacroce R, Sbrana A. Compositional layering and syn-eruptive mixing of a periodically refilled shallow magma chamber: the AD 79 plinian eruption of Vesuvius. Journal of Petrology (1995) 36:739–776.
Civetta L, Innocenti F, Manetti P, Peccerillo A, Poli G. Geochemical characteristics of potassic volcanics from Mts. Ernici. Contributions to Mineralogy and Petrology (1981) 78:37–47.[CrossRef][Web of Science]
Clayton RN, Mayeda TK. Oxygen isotopes in eucrites, shergottites, nakhilites, and chassignites. Earth and Planetary Science Letters (1983) 62:1–6.[CrossRef][Web of Science]
Conticelli S, Francalanci L, Manetti P, Cioni R, Sbrana A. Petrology and geochemistry of the ultrapotassic rocks from the Sabatini Volcanic District. Central Italy: the role of evolutionary processes in the genesis of variable enriched alkaline magmas. Journal of Volcanology and Geothermal Research (1997) 75:107–136.[CrossRef][Web of Science]
Dallai L, Freda C, Gaeta M. Oxygen isotope geochemistry of pyroclastic clinopyroxene monitors carbonate contributions to Roman-type ultrapotassic magma. Contributions to Mineralogy and Petrology (2004) 148:247–263.[CrossRef][Web of Science]
Danyushevsky LV, Della Pasqua FN, Sokolov S. Re-equilibration of melt inclusions trapped by magnesian olivine phenocrists from subduction-related magmas: petrological implications. Contributions to Mineralogy and Petrology (2000) 138:68–83.[CrossRef][Web of Science]
Danyushevsky LV, McNeill AW, Sobolev AV. Experimental and petrological studies of melt inclusions in phenocrysts from mantle-derived magmas: an overview of techniques, advantages and complications. Chemical Geology (2002) 183:5–24.[CrossRef][Web of Science]
De Rita D, Funiciello R, Parotto M. Carta Geologica del Complesso Vulcanico dei Colli Albani (scala 1:50 000) (1988) Rome: Progetto Finalizzato Geodinamica CNR.
Devine JD, Gardner JE, Brack HP, Layne GD, Rutherford MJ. Comparison of microanalytical methods for estimating H2O content in silicic glasses. American Mineralogist (1995) 80:319–328.[Abstract]
Di Battistini G, Montanini A, Vernia L, Bargossi GM, Castorina F. Petrology and geochemistry of ultrapotassic rocks from the Montefiascone Volcanic Complex (central Italy): magmatic evolution and petrogenesis. Lithos (1998) 43:169–195.[CrossRef][Web of Science]
Di Battistini G, Montanini A, Vernia L, Venturelli G, Tonarini S. Petrology of melilite-bearing rocks from the Montefiascone Volcanic Complex (Roman Magmatic Province): new insights into the ultrapotassic volcanism of Central Italy. Lithos (2001) 59:1–24.[CrossRef][Web of Science]
D’Orazio M, Innocenti F, Tonarini S, Doglioni C. Carbonatites in a subduction system: the Pleistocene alvikites from Mt. Vulture (southern Italy). Lithos (2007) 98:313–334.[CrossRef][Web of Science]
Einaudi MT, Meinert LD, Newberry RJ. Skarn deposits. Economic Geology (1981) 75:317–391.[CrossRef][Web of Science]
Federico M, Peccerillo A. Mineral chemistry and petrogenesis of granular ejecta from the Alban Hills volcano (Central Italy). Mineralogy and Petrology (2000) 74:223–252.[CrossRef]
Federico M, Peccerillo A, Barbieri M, Wu TW. Mineralogical and geochemical study of granular xenoliths from the Alban Hills volcano, Central Italy: bearing on evolutionary process in potassic magma chambers. Contributions to Mineralogy and Petrology (1994) 115:384–401.[CrossRef][Web of Science]
Ferrara G, Laurenzi MA, Taylor HP Jr, Tonarini S, Turi B. Oxygen and strontium isotope studies of K-rich volcanics rocks from the Alban Hills, Italy. Earth and Planetary Science Letters (1985) 75:13–28.[CrossRef][Web of Science]
Foley SF. Petrological characterization of the source components of potassic magmas: geochemical and experimental constraints. Lithos (1992) 28:187–204.[CrossRef][Web of Science]
Foley FS, Jenner GA. Trace element partitioning in lamproitic magmas—the Gaussberg olivine leucite. Lithos (2004) 75:19–38.[CrossRef][Web of Science]
Foley SF, Venturelli G, Green DH, Toscani L. The ultrapotassic rocks: characteristics, classifications and constraints for petrogenetic models. Earth-Science Reviews (1987) 24:81–134.
Fornaseri M, Scherillo A, Ventriglia U. La regione vulcanica dei Colli Albani. Vulcano Laziale (1963) Rome: Consiglio Nazionale delle Ricerche. 561 pp.
Fowler SJ, Spera FJ, Bohrson WA, Belkin HE, De Vivo B. Phase equilibria constraints on the chemical and physical evolution of the Campanian Ignimbrite. Journal of Petrology (2007) 48:459–493.
Freda C, Gaeta M, Palladino DM, Trigila R. The Villa Senni Eruption (Alban Hills, Central Italy): the role of H2O and CO2 on the magma chamber evolution and on the eruptive scenario. Journal of Volcanology and Geothermal Research (1997) 78:103–120.[CrossRef][Web of Science]
Freda C, Gaeta M, Karner DB, Marra F, Renne PR, Taddeucci J, Scarlato P, Christensen JN, Dallai L. Eruptive history and petrologic evolution of the Albano multiple Maar (Alban Hills, Central Italy). Bulletin of Volcanology (2006) 68:567–591.[CrossRef][Web of Science]
Freda C, Gaeta M, Misiti V, Mollo S, Dolfi D, Scarlato P. Magma–carbonate interaction: an experimental study on ultrapotassic rocks from Alban Hills (Central Italy). Lithos (2008) 101:397–415.[CrossRef][Web of Science]
Frezzotti ML. Silicate-melt inclusions in magmatic rocks: applications to petrology. Lithos (2001) 55:273–299.[CrossRef][Web of Science]
Fulignati P, Marianelli P, Santacroce R, Sbrana A. Probing the Vesuvius magma chamber–host rock interface through xenoliths. Geological Magazine (2004) 141:417–428.
Funiciello R, Parotto M. Il substrato sedimentario nell'area dei Colli Albani: considerazioni geodinamiche e paleogeografiche sul margine tirrenico dell'Appennino centrale. Geologica Romana (1978) 17:233–288.
Gaeta M, Fabrizio G, Cavarretta G. F-phlogopite in the Alban Hills Volcanic District (Central Italy): indications regarding the role of volatiles in magmatic crystallization. Journal of Volcanology and Geothermal Research (2000) 99:179–193.[CrossRef][Web of Science]
Gaeta M, Freda C, Christensen JN, Dallai L, Marra F, Karner DB, Scarlato P. Time-dependent geochemistry of clinopyroxene from the Alban Hills (Central Italy): clues to the source and evolution of ultrapotassic magmas. Lithos (2006) 86:330–346.[CrossRef][Web of Science]
Gaeta M, Di Rocco T, Freda C, Dallai L, Tiepolo M, Perinelli C, Scarlato P. The origin of ‘hyper-calcic’ silicate melts and their role as crustal contaminant. Cities on Volcanoes 5 Conference, 2007 Japan, Abstracts Volume (2007) 11-P-129.
Gaffney AM. Environments of crystallization and compositional diversity of Mauna Loa xenoliths. Journal of Petrology (2002) 43:963–980.
Ghiorso MS, Kelemen PB. Evaluating reaction stoichiometry in magmatic systems evolving under generalized thermodynamic constraints: examples comparing isothermal and isenthalpic assimilation. In: Magmatic Processes: Physicochemical Principles, Volume 1—Mysen BO, ed. (1987) University Park, PA: Lancaster. 319–336.
Ghiorso MS, Sack RO. Chemical mass transfer in magmatic processes. IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid–solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology (1995) 119:197–212.[Web of Science]
Giaccio B, Sposato A, Gaeta M, Marra F, Palladino DM, Taddeucci J, Barbieri M, Messina P, Rolfo MF. Mid-distal occurrences of the Albano Maar pyroclastic deposits and their relevance for reassessing the eruptive scenarios of the most recent activity at the Colli Albani Volcanic District, Central Italy. Quaternary International (2007) 171–172:160–168.
Hofmann AW. Chemical differentiation of the Earth: the relationship between mantle, continental crust and oceanic crust. Earth and Planetary Science Letters (1988) 90:297–314.[CrossRef][Web of Science]
Holness MB, Hallworth MA, Woods A, Sides RE. Infiltration metasomatism of cumulates by intrusive magma replenishment: the wavy horizon, Isle of Rum, Scotland. Journal of Petrology (2007) 48:563–587.
Iacono Marziano G, Gaillard F, Pichavant M. Limestone assimilation and the origin of CO2 emission at the Alban Hills (Central Italy): Constraints from experimental petrology. Journal of Volcanology and Geothermal Research (2007) 166:91–105.[CrossRef][Web of Science]
Iacono Marziano G, Gaillard F, Pichavant M. Limestone assimilation by basaltic magmas: an experimental re-assessment and application to Italian volcanoes. Contributions to Mineralogy and Petrology (2008) doi:10.1007/s00410-007-0267-8.
Irvine TN. Chromian spinel as a petrogenetic indicator. Part 2. Canadian Journal of Earth Sciences (1967) 4:71–103.
Irvine TN. Terminology for layered intrusions. Journal of Petrology (1982) 23:127–162.
Joesten R. Mineralogical and chemical evolution of contaminated igneous rocks at a gabbro–limestone contact, Christmas Mountains, Big Bend region, Texas. Geological Society of America Bulletin (1977) 88:1515–1529.
Joesten R, Hill J, Van-Horn SR. Limestone assimilation and clinopyroxene production along the contacts of a 9 meter alkali olivine basalt dike. Geological Society of America, Abstracts with Programs (1994) 26:476.
Kamenetsky VS, Gurenko AA. Cryptic crustal contamination of MORB primitive melts recorded in olivine-hosted glass and mineral inclusions. Contributions to Mineralogy and Petrology (2007) 153:465–481.[CrossRef][Web of Science]
Kamenetsky V, Métrich N, Cioni R. Potassic primary melts of Vulsini (Roman Province): evidence for mineralogy and melt inclusion. Contributions to Mineralogy and Petrology (1995) 120:186–196.[Web of Science]
Kerrick DM. The genesis of zoned skarns in the Sierra Nevada, California. Journal of Petrology (1977) 18:144–181.
Kress VC, Carmichael ISE. Stoichiometry of the iron oxidation reaction in silicate melts. American Mineralogist (1988) 73:1267–1274.[Abstract]
Lee WJ, Huang WL, Wyllie P. Melts in the mantle modeled in the system CaO–MgO–SiO2–CO2 at 2·7 GPa. Contributions to Mineralogy and Petrology (2000) 138:199–213.[CrossRef][Web of Science]
Marra F, Freda C, Scarlato P, Taddeucci J, Karner DB, Renne PR, Gaeta M, Palladino DM, Trigila R, Cavarretta G. Post caldera activity in the Alban Hills volcanic district (Italy): 40Ar/39Ar geochronology and insight into magma evolution. Bulletin of Volcanology (2003) 65:227–247.[Web of Science]
Marsh BD. Solidification fronts and magmatic evolution. Mineralogical Magazine (1995) 60:5–40.[CrossRef][Web of Science]
Mattey D, Lowry D, Macpherson C. Oxygen isotope composition of mantle peridotite. Earth and Planetary Science Letters (1994) 128:231–241.[CrossRef][Web of Science]
Maurel C, Maurel P. Etude expérimentale de la distribution de l'aluminium entre bain silicate basique et spinelle chromifère. Implications pétrogenetiques: teneur en chrome des spinelles. Bulletin Minéralogique (1982) 105:197–202.
McBirney AR, Hunter RH. The cumulate paradigm reconsidered. Journal of Geology (1995) 103:114–122.[Web of Science]
Melzer S, Foley SF. Phase relations and fractionation sequences in potassic magma series modelled in the system CaMgSi2O6–KAlSiO4–Mg2SiO4–SiO2–F2O–1 at 1 bar and 18 kbar. Contributions to Mineralogy and Petrology (2000) 138:186–197.[CrossRef][Web of Science]
Mitchell AA, Eales HV, Kruger FJ. Magma replenishment, and the significance of poikilitic textures, in the Lower Main Zone of the western Bushveld Complex, South Africa. Mineralogical Magazine (1998) 62:435–450.[Abstract]
Owens BE. High-temperature contact metamorphism of calc-silicate xenoliths in the Kiglapait Intrusion, Labrador. American Mineralogist (2000) 85:1595–1605.
Peccerillo A. Plio-Quaternary Volcanism in Italy. In: Petrology, Geochemistry, Geodynamics (2005) Heidelberg: Springer.
Peccerillo A, Poli G, Tolomeo L. Genesis, evolution and tectonic significance of K-rich volcanics from the Alban Hills (Roman Comagmatic Region) as inferred from trace element geochemistry. Contributions to Mineralogy and Petrology (1984) 86:230–240.[CrossRef][Web of Science]
Perini G, Francalanci L, Davidson JP, Conticelli S. Evolution and genesis of magmas from Vico Volcano, Central Italy: multiple differentiation pathways and variable parental magmas. Journal of Petrology (2004) 45:139–182.
Perkins GB, Sharp ZP, Selverstone J. Oxygen isotope evidence for subduction and rift-related mantle metasomatism beneath the Colorado Plateau–Rio Grande rift transition. Contributions to Mineralogy and Petrology (2006) 151:633–650.[CrossRef][Web of Science]
Piochi M, Ayuso RA, De Vivo B, Somma R. Crustal contamination and crystal entrapment during polybaric magma evolution at Mt. Somma–Vesuvius volcano, Italy; geochemical and Sr isotope evidence. Lithos (2006) 86:303–329.[CrossRef][Web of Science]
Roedder E. Origin and significance of magmatic inclusions. Bulletin Minéralogique (1979) 102:467–510.
Rogers NW, Hawkesworth CJ, Parker RJ, March JS. The geochemistry of potassic lavas from Vulsini, central Italy and implications for mantle enrichment processes beneath the Roman region. Contributions to Mineralogy and Petrology (1985) 90:244–257.[CrossRef][Web of Science]
Sharp ZD. Oxygen isotope geochemistry of the Al2SiO5 polymorphs. American Journal of Science (1995) 295:1058–1076.
Shelley D. Igneous and Metamorphic Rocks under the Microscope (1993) London: Chapman & Hall. 445 pp.
Spera FJ, Bohrson WA. Energy-constrained open-system magmatic processes I: General model and energy-constrained assimilation and fractional crystallization (EC-AFC) formulation. Journal of Petrology (2001) 42:999–1018.
Sun SS, McDonough WF. Chemical and isotopic systematic of oceanic basalts: implication for mantle composition and processes. Magmatism in the Ocean Basins. Geological Society, London, Special Publications—Saunders AD, Norry MG, eds. (1989) 42:313–345.
Thompson RN. Primary basalts and magma genesis. III. Alban Hills, Roman comagmatic Province, Central Italy. Contributions to Mineralogy and Petrology (1977) 60:91–108.[CrossRef][Web of Science]
Tiepolo M, Bottazzi P, Palenzona M, Vannucci R. A laser probe coupled with ICP-double-focusing sector-field mass spectrometer for in situ analysis of geological samples and U–Pb dating of zircon. Canadian Mineralogist (2003) 41:259–272.[CrossRef][Web of Science]
Tilley CE. Some trends of basaltic magma in limestone syntexis. American Journal of Science, Bowen volume (1952) 529–545.
Trigila R, Agosta E, Currado C, De Benedetti AA, Freda C, Gaeta M, Palladino DM, Rosa C. Petrology. In: The Volcano of the Alban Hills—Trigila R, ed. (1995) Rome: Università degli Studi di Roma ‘La Sapienza’. 95–165.
Ussler W III, Glazner AF. Graphical analysis of enthalpy–composition relationships in mixed magmas. Journal of Volcanology and Geothermal Research (1992) 51:23–40.[CrossRef][Web of Science]
van Achterbergh E, Ryan CG, Jackson SE, Griffin WL. Data reducing software for LA-ICP-MS. Mineralogical Association of Canada, Short Course Series (2001) 29:239–243.
Vernon RH. A Practical Guide to Rock Microstructure (2004) Cambridge: Cambridge University Press. 594 pp.
Walker JA, Mickelson JE, Thomas RB, Patino LC, Cameron B, Carr MJ, Feigenson MD, Edwards RL. U-series disequilibria in Guatemalan lavas, crustal contamination, and implications for magma genesis along the Central American subduction zone. Journal of Geophysical Research (2007) 112. B06205, doi:10.1029/2006JB004589.
Wenzel T, Baumgartner LP, Brügmann GE, Konnikov EG, Kislov EV. Partial melting and assimilation of dolomitic xenoliths by mafic magma: the Ioko–Dovyren intrusion (North Baikal Region, Russia). Journal of Petrology (2002) 43:2049–2074.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||