Quaternary International 467 (2018) 369e390
Contents lists available at ScienceDirect
Quaternary International journal homepage: www.elsevier.com/locate/quaint
Green, grey and black: A comparative study of Sierra de las Navajas (Mexico) and Lipari (Italy) obsidians P. Donato a, *, L. Barba b, R. De Rosa a, G. Niceforo a, A. Pastrana c, S. Donato d, e, G. Lanzafame f, L. Mancini f, G.M. Crisci a della Calabria, Italy DiBEST, Universita gicas, Universidad Nacional Auto noma de M Instituto de Investigaciones Antropolo exico, Mexico c n de Estudios Arqueolo gicos, Instituto Nacional de Antropología e Historia, M Direccio exico, Mexico d Department of Physics, University of Trieste, Italy e INFN Sezione di Trieste, Trieste, Italy f Elettra-Sincrotrone Trieste S.C.p.A., Basovizza, Trieste, Italy a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 May 2017 Received in revised form 10 November 2017 Accepted 12 November 2017 Available online 20 November 2017
Sierra de Las Navajas (State of Hidalgo, Mexico) and Lipari (Aeolian Islands, Italy) were among the most important sources for obsidian trade in Mesoamerica and in the Mediterranean during the Stone Age. In this paper obsidians from these two localities were compared in terms of their aspect, chemical composition, microcrystallinity and microvesiculation. In Sierra de las Navajas, the typical deep green obsidian with a golden hue has been analyzed together with a less common dark grey, porphyritic variety. Lipari obsidian is commonly black, but a light grey variety also occurs. Obsidian of both varieties was analyzed for this paper. Microvesicularity was investigated through Scanning Electron Microscope (SEM) observation, microporosimetry and, for Sierra de las Navajas green obsidian, through a preliminary X-ray computed microtomographic study. Crystallinity and micro- or nano-crystallinity were investigated through X-Ray Powder Diffraction and EDS (Energy Dispersion System) microanalyses. Finally, the chemical composition in terms of major and trace elements, including rare earth elements, was determined using X-Ray Fluorescence (XRF) and Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). The comparison of samples of different colors suggests that the characteristic green color of Sierra de las Navajas obsidians could be related to their relatively high iron content, and to the occurrence of many elongated and iso-oriented vesicles which may also be responsible for the obsidian's golden hue. Low iron and an absence of vesicles give Lipari obsidian its “normal” black color. The light grey obsidian from Lipari probably owes its color and imperfect conchoidal fracture to numerous bubbles of less than 1 mm in size and to nano-crystallinity. Sierra de las Navajas obsidians show a significant chemical variability in terms of trace elements, that can be explained by common evolutionary processes in the magma chamber. However, this variability is also internal to a single volcanic complex and this makes the trace element contents unsuitable to differentiating between the different sub-sources of the same area. On Lipari, our data do not allow us to distinguish between the two sub-sources of Vallone del Gabellotto and Canneto Dentro on the basis of major and trace elements. On the whole, our study suggests that caution should be used for both Lipari and Sierra de las Navajas when identifying obsidian sub-sources on the basis of trace element contents. © 2017 Elsevier Ltd and INQUA. All rights reserved.
Keywords: Sierra de las Navajas (Mexico) Lipari (Aeolian Islands) Obsidian source analyses Obsidian color Microtexture Microvesiculation XRF and ICP-MS analyses
1. Introduction Obsidian is a hard, dark, crystal-free glass-like volcanic product
* Corresponding author. E-mail address:
[email protected] (P. Donato). https://doi.org/10.1016/j.quaint.2017.11.021 1040-6182/© 2017 Elsevier Ltd and INQUA. All rights reserved.
formed during both explosive and effusive eruptions of viscous, silica-rich magmas, often associated with highly vesicular pumices. The rapid solidification and the high viscosity inhibit crystallization and originate an optically isotropic glass with a high mechanical strength. Due to its glassy structure, obsidian is highly attractive and relatively easy to work, as it breaks in very predictable and controlled ways via conchoidal fracturing. Moreover, the working
370
P. Donato et al. / Quaternary International 467 (2018) 369e390
edges obtained by its fracturing are sharper than other types of tool-making stones, making obsidian a highly sought after material for the production of sharp-edged implements (Glascock et al., 1998). The volcanic glass was a strategic material in ancient times before the use of metals, and it was widely used for the production of weapons and jewelry, as well as in medicine and magic-religious activities. The control of its exploitation, production and distribution, via commercial, tributary or direct control systems, was one of the most important activities of the ancient cultures of Europe and America (Hirth, 2006). Although black is the most common color of obsidians, different color varieties also occur, such as grey, mahogany (a combination of a reddish/brown and black color; see Doland and Shackley, 2017), green or “rainbow” (bands of various colors, ranging from red to purple; see Ma et al., 2001). Different varieties of color very frequently occur in the same locality (e.g. Glass Butte in Oregon, USA; see Ambroz et al., 2001). The different colors have been attributed to physical features or chemical composition. The vibrant color of “fire” obsidians has been related to thin layers of concentrated nanometric crystals of magnetite, which give rise to brilliant colors in reflection (Ma et al., 2007), while Argote-Espino et al. (2012) attributed the green color of Sierra de las Navajas obsidians to an abundance of specific trace elements. The circulation and exchange of obsidian artifacts is of great importance in the investigation of cultural, social, and economic development in ancient societies and reconstruction of ancient maritime or terrestrial trade routes (Glascock, 2002; Cortegoso et al., 2016; Pintar et al., 2016; Kuzmin, 2017; Freund, in press). This requires the possibility to ascribe archaeological artifacts to a specific source. In recent decades, a number of studies have been conducted in order to identify the sources of obsidian artifacts through the use of different analytical methodologies ranging from the Optical Emission Spectrometry adopted in the pioneering work of Cann and Renfrew (1964) for Mediterranean sources to backscattered electron images (Burton and Krinsley, 1987), 57Fe €ssbauer spectroscopy, Electron Paramagnetic Resonance and Mo magnetization properties (Poupeau et al., 2001) or fission-track dating (Bigazzi and Bonadonna, 1973; Bigazzi and Radi, 1998). Chemical fingerprinting of natural and archaeological obsidians proved to be the most successful way to correctly ascribe artifacts to a specific source in various regions: e.g. in Syria and Iraq (Khalidi et al., 2016), in Korea (Yi and Jwa, 2016), in the Mediterranean (Barca et al., 2007; De Francesco et al., 2008; Tykot, 2002, 2017; Orange et al., 2017), in Russia (Grebennikov and Kuzmin, 2017), in Central America (Cobean et al., 1991; Glascock, 2002) and in Southern America (Cortegoso et al., 2016; Escola et al., 2016; De Francesco et al., in press). The chemical composition, and particularly the trace element contents, can also help to discriminate between sub-sources in the same area - Glasse Butte, Oregon (Ambroz et al., 2001) and Monte Arci, Sardinia (Tykot, 2002, 2017) - thus, contributing to a better understanding of the exploitation and trade of obsidians in the past. This paper presents a multidisciplinary work aimed at the chemical and micro- and nano-textural characterization of macroscopically different obsidians from Sierra de las Navajas (State of Hidalgo, Mexico) and Lipari (Aeolian Islands, Italy) through different analytical techniques. The two localities were key sources for obsidian trade in Mesoamerica and in the Mediterranean during the Stone Age. Obsidians from these two sources are very different in color (green in Sierra de las Navajas and black on Lipari) and geochemical affinity (calc-alkaline on Lipari and peralkaline in Sierra de las Navajas). This makes them particularly useful for investigating the factors influencing the macroscopic differences between obsidians. Therefore, the first aim of this work is to test the
roles of chemical composition and microtexture in the color and type of fracturing, the most important features determining the use of obsidian in the past. A second aim is to verify the possibility of using the chemical data to identify sub-sources within the same main source area, so providing a more precise determination of provenance in the archaeological studies. 2. Geological setting and obsidian occurrences 2.1. Lipari With its surface area of 38 km2, Lipari is the largest island in the Aeolian Archipelago, a Quaternary volcanic arc which developed at the southern periphery of the Tyrrhenian basin (Italy). The magmatism in this area mainly relates to the subduction and southeast rollback of the Ionian slab below the Calabrian Arc (Patacca and Scandone, 1989; Faccenna et al., 2001; Gvirtzman and Nur, 2001; Chiarabba et al., 2008; Ventura, 2013). This process has, in turn, induced the upwelling and SE migration of the asthenospheric mantle (Gvirtzman and Nur, 1999; Chiarabba et al., 2008), causing extensional strain and volcanism in the last 0.7e1 Ma (Ventura, 2013). A simplified geological map of Lipari is shown in Fig. 1. The eruptive history of Lipari started about 270 ka ago and can be divided into nine Eruptive Epochs separated by periods of quiescence (Forni et al., 2013; Lucchi et al., 2013). One of these periods, occurring between 81 and 43 ka, marked an important change in the composition of erupted products, which changed from the basaltic andesites and andesites of the first six epochs to the rhyolites of the more recent activity (Crisci et al., 1991; Forni et al., 2013). Rhyolitic obsidians were emplaced both effusively, as lava flows, and explosively, in association with pumices, over the last 43 ka. During Epochs 7 and 8 three main eruptions occurred, each starting with an explosive phase and ending with the emplacement of endogenous lava domes: Punta del Perciato (Epoch 7, age unknown), Falcone (Epoch 7, 43e40 ka) and Monte Guardia- Monte Giardina (Epoch 8, 27e20 ka). Obsidian clasts are common in the explosive sequences; locally, the rhyolitic lava forming the domes has a vitreous aspect. About 20 ka ago, several minor domes, aligned in a N-S direction, formed on the eastern coast of the island (Gioncada et al., 2003). One of these, the dome of Canneto Dentro, was preceded by a low energy explosive eruption which formed obsidian-rich breccias (Forni et al., 2013). The ninth and last eruptive Epoch (8.7 ka- AD 1220) included both effusive and explosive phases, all occurring in the northern sector of Lipari and producing rhyolitic lavas and pumiceous successions that are rich in obsidian clasts. The first eruption of this Epoch was the Vallone del Gabellotto explosive eruption (8.7e8.4 ka, Zanchetta et al., 2011), leading to the formation of a widespread pumiceous succession and followed by the emplacement of the obsidian coulee of Pomiciazzo (8.6 ka; Wagner et al., 1976; Arias et al., 1986; Forni et al., 2013). After a period of quiescence, activity restarted on the northeastern side of the island in medieval times with the historical eruptions of Monte Pilato pumice cone (AD 776, Forni et al., 2013) and Forgia Vecchia lava flow (ca. AD 400, Bigazzi and Bonadonna, 1973; Forni et al., 2013). The Monte Pilato cone is mainly made up of highly vesiculated rhyolitic pumices; obsidian clasts and lava lithics are also common. Forgia Vecchia is a bilobate obsidianaceous lava flow outcropping south of Monte Pilato and overlying a pumiceous pyroclastic sequence (Forni et al., 2013). Finally, the last eruption on the island occurred in AD 1220 (Tanguy et al., 2003) and formed the Rocche Rosse obsidianaceous
P. Donato et al. / Quaternary International 467 (2018) 369e390
371
Fig. 1. Geological sketch map of Lipari from Forni et al. (2013). Red squares indicate the sampling areas. CD: Canneto Dentro; GA: Gabellotto; FB: Fiumebianco; PO: Pomiciazzo; LA: Lami; FV: Forgia Vecchia; PI: Pilato.
372
P. Donato et al. / Quaternary International 467 (2018) 369e390
lava flow, extending from the flank of Monte Pilato into the sea for a total length of 4 km (Forni et al., 2013). A pyroclastic succession, probably coheval with the Rocche Rosse lava flow (Forni et al., 2013), can be observed in outcrops in the area of Lami, at the foot of the southern flank of Monte Pilato. The juvenile fraction consists of highly vesiculated pumices and grey and black obsidians (Davì et al., 2011). 2.2. Sierra de las Navajas Sierra de las Navajas, also known as Sierra de Pachuca, is an extinct volcanic complex situated on the north-eastern edge of the Trans-Mexican Volcanic Belt (hereafter TMVB), just north of the city of Pachuca (Hidalgo, Mexico). The TMVB exhibits a great variability in the affinity and composition of erupted products, with rocks showing OIB (Oceanic Island Basalt) geochemical signatures occurring side by side with calc-alkaline, subduction-related volcanic rocks (e.g. Righter et al., 1995; Righter and Rosas-Elguera, ~ ez, 2002; Petrone et al., 2003; 2001; Siebert and Carrasco-Nun Orozco-Esquivel et al., 2007). Since the late Pliocene, modest volumes of intraplate-like lavas and rhyolites with peralkaline composition have also been erupted in several areas of the TMVB (Ferrari et al., 2012). Sierra de las Navajas is one of these rhyolitic peralkaline centers, which formed on the northern boundary of the eastern sector of the TMVB. The rhyolites of Sierra de las Navajas locally overlie a basaltic lava flow sequence which has been correlated with similar lavas dating from 2.4 to 2.6 Ma (Cantagrel and Robin, 1979; Nelson and Lighthart, 1997). Lighthart Ponomarenko (2004) recognizes three main stages in the volcanic history of Sierra de las Navajas (Fig. 2). During the first stage, three rhyolitic lava flows and associated pyroclastics were emplaced: Guajalote Flow Complex in the south-western sector, Las Minas Flow Complex to the west and south, and the Ixatla Flow Complex in the southern and central sectors. During the emplacement of the Ixatla Flow, a huge sector collapse occurred in the northern portion of the volcanic edifice (stage II), producing a catastrophic debris avalanche, the deposit of which covers an area of about 560 km2 (Nelson and Lighthart, 1997). The sector collapse was accompanied by explosive eruptions whose pyroclastic deposits directly overlie the debris avalanche breccias. Within the amphitheater depression formed by the sector collapse, volcanic activity continued (stage III) with the emplacement of the rhyolite lava domes and lava flows of n Flow Complex (Lighthart Ponomarenko, 2004). the El Horco Obsidians were erupted during the pre- and post-collapse activity. In the Las Minas complex they occur as pyroclastic fragments within the explosive sequences, and, above all, as a deposit that was extensively mined in pre-colonial times and is still exploited today. These obsidians are deep green in color with a golden/silver hue, are completely aphyric and have a perfect conchoidal fracture (Lighthart Ponomarenko, 2004; Argote-Espino et al., 2012). They occur as blocks of different sizes set in a volcanic ash matrix. This led Pastrana (1996) to consider the deposit as a lahar or a block and ash flow. However, Nelson and Lighthart (1997) and Lighthart Ponomarenko (2004) re-interpreted it as a rhyolitic lava flow with obsidianaceous portions and rhyolitic bands weathered to form a clay matrix. It is beyond the scope of this work to investigate the origin of this important obsidian source; for simplicity, we will refer to it as “block and ash flow” in the following text. 3. Use of obsidians in ancient times 3.1. Lipari obsidian Lipari obsidian has been known about in southern Italy since the
Early Neolithic at least (Bigazzi et al., 2005); the earliest recognized case of its exploitation dates back to the sixth millennium BC at Impressed Ware archaeological sites in Apulia (Bigazzi and Radi, 1998; Gratuze and Boucetta, 2009; Freund et al., 2017). During the Neolithic, Lipari was the most important source of obsidian in the Western Mediterranean. Obsidian artifacts in southern Italy are mainly related to Lipari sources and Lipari obsidian is distributed throughout the Italian Peninsula. It has also been found with relation to the middle phases of the Neolithic in southern France (Bigazzi et al., 2005). Although obsidian has been widely erupted (explosively or effusively) on Lipari over the last 43 ka, only two deposits have been identified as sources of archaeological obsidian: the Vallone del Gabellotto sequence and, subordinately, the Canneto Dentro deposit (Bigazzi et al., 2005; Tykot et al., 2006; Freund et al., 2017). These two sub-sources have been geochemically distinguished between on the basis of their Fe/Sr and Rb/Sr ratios (Tykot et al., 2013; Freund et al., 2015, 2017; Tykot, 2017). No artifact can be related to the older obsidians of the southern part of the island (Bigazzi et al., 2005). During the Neolithic, Lipari obsidian was mainly used to make blades (Tykot et al., 2013). Primary reduction was probably carried out in the source area by local populations, with obsidian being transported from Lipari to eastern Sicily and Calabria in the form of preformed cores (Freund et al., 2015). Numerous workshop sites in which the raw material was prepared for distribution around mainland Italy have been identified along the Tyrrhenian coast of Calabria (Ammerman, 1979). Transport occurred mainly by sea; only internal regions were reached by land (Bigazzi et al., 2005; Freund et al., 2015). Long-distance obsidian trade was probably as a series of small-scale movements (Tykot, 1996). The most abundant obsidian at sites in southern and central Italy is from Lipari, but it decreases in frequency with distance to just 17% of the obsidian assemblage in France (Tykot, 2002). This relates to the down-the-line trading system, in which progressively smaller amounts of obsidian were traded from one village to the other (Renfrew, 1969; Tykot, 2002). During the more recent Neolithic phases, trade in Lipari obsidian probably followed the Adriatic routes, as testified to by the occurrence of this obsidian in the eastern coastal areas of Italy. Lipari obsidian is found over a wide area, but its use appears to have been more limited at more distant sites (Bigazzi et al., 2005). The use of obsidian had declined in the Central Mediterranean by the third millennium BC (Freund, 2014; Freund et al., 2017) and only a few artifacts from the Aeneolithic era have been found in mainland Italy and France. However, on the islands of Malta and Ustica, and near the source areas in Sicily, Sardinia and Corsica, obsidian continued to be used (Freund et al., 2017). During the Aeneolithic era, Lipari obsidian was distributed around Sicily in the form of preformed cores or finished products, originally manufactured in the source area (Freund et al., 2017). Blades were still the most common artifacts; however, they were probably not used for daily activities, but had a distinct social role and were devoted to body modification (shaving, hair cutting, scarification etc.) (Freund et al., 2015). During the Metal ages the use of obsidian was limited to jewelry or other ornamental objects; medical proprieties and magical power were attributed to obsidian (Bigazzi et al., 2005). 3.2. Sierra de las Navajas obsidian Sierra de Las Navajas, or Itztepec in Nahuatl (the language of the Aztecs), meaning “the mountain of the obsidian blades”, was one of the major sources of high quality volcanic glass in Mesoamerica. This location for obsidian was intensively exploited in pre-Hispanic times, and its green obsidian was distributed throughout the
P. Donato et al. / Quaternary International 467 (2018) 369e390
373
Fig. 2. Geological sketch map of Sierra de las Navajas. Re-drawn from Lighthart Ponomarenko (2004). Topographic background obtained by shape files provided by INEGI of Pachuca (Hidalgo). LJ: La Joya quarry; P: Palmilla quarry; N: El Nopalillo quarry; MW: mine well in the “block and ash flow”; A: Alfajayucan quarry; CM: Cruz el Milagro; EH: El Horcon; IS: Ixtula Sembo; SB: San Bartolo.
374
P. Donato et al. / Quaternary International 467 (2018) 369e390
ancient cultures of Mesoamerica (Pastrana and Carballo, 2016). In its transparent, translucent, golden and opaque varieties, this green obsidian is a high quality vitreous rock used for the manufacture of artifacts through the knapping techniques of percussion, pressure and polish. The “block and ash flow” of the Las Minas complex was highly exploited in the past. As its outcrops are only visible in a few points, mining was necessary to obtain a sufficient quantity of obsidian to satisfy the constant demand of the powerful Mesoamerican civilization. The ancient miners looked for blocks, nodules and slates of vitreous obsidian, avoiding fractures, coarse cortex, and crystals or spherulites. The mining was carried out through the excavation of vertical shafts of about 1 m in diameter that connected horizontal and inclining tunnels of elliptic sections of 1.70 m in diameter at the points where the miner found obsidian of good shapes, dimensions and vitreous texture. The mines, excavated without metal tools and support structures, reached depths of 100 m and were the deepest pre-Hispanic mines in the Americas. Archaeological studies of the sequences of exploitation and occupation of the Sierra de Las Navajas obsidian source (Hirth, 2003; Pastrana and Domínguez, 2009) have identified archaeological materials in the mines for production of several kinds of artifacts, including weapons, jewelry, and magic-religious objects that are directly associated with some of the main pre-Hispanic gods (Pastrana and Athie, 2014). This important obsidian was widely used all over the Mesoamerican territory over a long period of time. The first use of this special green obsidian at its source probably coincides with the arrival of prehistoric humans to Mesoamerica around 10,000e14,000 years ago. However, the first intensive exploitation can be related to the development of the first preHispanic city, Teotihuacan (ca. AD 200e600). The exploitation sequence of the Sierra de las Navajas obsidian source was not, though, continuous as it developed in distinct steps corresponding to major changes in the political order (Pastrana and Domínguez, 2009). After a lapse of 200 years, during which the established Teotihuacan systems fell from power, the Toltec culture (AD 950e1100) concentrated the regional power and continued the exploitation of the source. Later, during the Aztec military Empire, which was constituted by the Triple Alliance: Tenochtitlan, Texcoco, and Tacuba (AD 1325e1521), the exploitation of Sierra de las Navajas obsidian reached a high level of social and technical organization (Pastrana and Domínguez, 2009). Even after the Hispanic conquest, Sierra de las Navajas obsidian was in continuous use for native agriculture and many Hispanic craft interests, like feather work, leather and textile production for clothing and transport of pre-Hispanic and European objects. In the Early Colonial Period (17th century), the extraction and use of obsidian was limited by the production and distribution of European metals (Pastrana and Fournier, 1998). The Spanish also started to prohibit the use of obsidian because of the fear of an indigenous rebellion with obsidian weapons and because the Catholic Inquisition associated the use of obsidian with witchcraft. Presently, besides the use of the obsidian in modern crafts, belief in the magic properties of obsidian still persists.
breccias mainly made up of coarse pumice blocks and rich in obsidian clasts. Three samples come from the pyroclastic deposit of the Vallone del Gabellotto Formation (EE9, ca. 8.7e8.4 ka), a thick pumiceous succession with small amounts of obsidian made up of an alternance of pyroclastic density current and fall deposit; two of these were sampled in the Vallone Fiumebianco quarry, one in Vallone del Gabellotto. Finally, we selected one specimen from the Pomiciazzo lava flow. Moreover, we collected obsidian which was erupted during more recent volcanic activity. One obsidian sample comes from the pumice quarry of Acquacalda and belongs to the fallout sequence forming the Mt. Pilato cone (Sciarra dell’Arena Formation). Two obsidian pieces are from the fallout breccias relating to the volcanic center of Lami. One sample is representative of the Forgia Vecchia lava flow. All the samples are macroscopically aphyric. With the exception of one of the Lami specimens, which is light grey, all the other obsidians are black and with perfect conchoidal fracture (Table 1; Fig. 3a). The grey obsidian from Lami has a glassy surface and an opaque inner portion; its fracture is not perfectly conchoidal (Table 1, Fig. 3b). The sampling points and macroscopic features of samples are listed in Fig. 1 and Table 1. A total of 18 obsidians were sampled in Sierra de las Navajas (Fig. 2 and Table 1). Nine green obsidians were collected from the different stratigraphic sections in the Las Minas complex outcropping outside the depression formed after the sector collapse. The deposits from which obsidians were picked are mainly pumice-rich Strombolian fall deposits and breccias underlying rhyolitic, obsidianaceous lava flows. In particular, four samples were from a quarry in the area of the obsidian mine wells (El Nopalillo quarry), where the Strombolian fall deposits are interbedded with pyroclastic density current deposits and are partly overlaid by an obsidianaceous lava flow with brecciated base, and partly by a chaotic deposit made of obsidian and rhyolite blocks in an ash matrix, very similar to the exploited “block and ash flow”. Another sample from this complex (SN13_3b) was collected from a mine well in the “block and ash flow”. One sample is from the mines on Cruz El Milagro, the highest peak in the area (3180 m above sea level). All the obsidians are deep to light green with a golden hue and perfect conchoidal fracture (Fig. 3c). No mineral phases are visible with the naked eye. Obsidians from the third period of activity (post-debris n. In both avalanche) were sampled from Ixtula Sembo and El Horco cases, no in situ outcrop was found and obsidians came from n complex is reworked deposits; their attribution to the El Horco therefore uncertain. In Ixtula Sembo, deep green obsidians, very similar to those of Las Minas complex, were sampled together with a completely different type, grey in color and with coarse (>2 mm in size) crystals of feldspar (SN13-2a, Fig. 3d). Finally, a green n peak. Four green obsidian was collected at the base of El Horco obsidians were sampled from the deposit of the debris avalanche caused by the sector collapse in San Bartolo, San Miguel Regla and ~ a El Aire (Table 1). Pen Samples were analyzed using different techniques (Table 1) at the laboratories of the University of Calabria, Italy. X-ray computed microtomography measurement on one sample of Sierra de las Navajas green obsidian was performed at Elettra synchrotron light source (Trieste, Italy).
4. Sampling and analytical techniques 4.1. Sampling and macroscopic description
4.2. Chemical analyses
On Lipari we collected 11 obsidians from pyroclastic sequences and lavas representative of the activity of the last 20 ka. Seven of these samples were from sites considered to be sources for archaeological artifacts. Three specimens were collected from the pyroclastic deposit of Canneto Dentro (Eruptive Epoch 8), fallout
All samples were crushed and milled, and pellets were prepared for the determination of their chemical composition in terms of major elements (SiO2, TiO2, Al2O3, Fetot, MnO, MgO, CaO, K2O, Na2O, P2O5) and some trace elements (Ba, Cr, Co, Nb, Ni, Rb, Sr, V, Zr, Y) by X-Rays Fluorescence (XRF). The instrument used was a Bruker S8
P. Donato et al. / Quaternary International 467 (2018) 369e390
375
Table 1 Samples list including sampling points, macroscopic description and analyses carried out for each obsidian sample. See text for analyses details. Sample
Coordinates
Sierra de las SN11_43B 19 40 21.7400 N Navajas 98 350 21.0300 W SN11_43A 20 40 21.7400 N 98 350 21.0300 W SN11_23 20 40 55.7000 N 98 330 47.2000 W SN13_3B 20 40 57.3900 N 98 340 23.9100 W SN11_18 AV SN11_18 AN SN11_17 SN11_15C SN13_4B SN11_7B SN11_8C SN13_20 SN13_2B SN13_2A SN13_8 SN13_5B SN13_13 SN12_4A Lipari
CAN 2 LIP 16 CD 1a LIP 16 CD 1b Lami 4 FV 1 Lami 5 bo LIP 16 FB 9 LIP 16 FB 10 PIL 2 Lami 2G Lami 2N
17 50 15.8400 N 98 330 55.4400 W 18 50 15.8400 N 98 330 55.4400 W 19 50 15.8400 N 98 330 55.4400 W 20 50 15.8400 N 98 330 55.4400 W 20 50 11.4900 N 98 330 12.5400 W 20 30 53.8600 N 98 330 2.2600 W 20 30 53.8600 N 98 330 2.2600 W 20 70 4.0000 N 98 330 6.9200 W 19 80 38.0000 N 98 330 15.5100 W 20 80 38.0000 N 98 330 15.5100 W 20 160 58.5700 N 98 320 30.0900 W 20 150 35.9000 98 280 12.7600 W 20 160 41.5400 N 98 310 2.2500 W 20 150 27.6200 N 98 280 7.2000 W 38 290 5.98200 N 14 570 41.70900 E 38 290 5.98200 N 14 570 41.70900 E 38 290 5.98200 N 14 570 41.70900 E 38 290 54.2500 N 14 570 17.7700 E 38 290 22.6700 N 14 570 7.1400 E 38 290 40.14600 N 14 570 27.68700 E 38 300 5.93300 N 14 550 57.26600 E 38 300 5.93300 N 14 550 57.26600 E 38 310 6.05600 N 14 560 53.28600 E 38 290 50.65800 N 14 570 3.71500 E 38 290 50.65800 N 14 570 3.71500 E
Locality (deposit)
Eruptive Epoch (EE) or Volcanic Complex (VC)
Macroscopic description
Chemical analyses
Other analyses
Palmilla (breccia)
Las Minas VC
XRF
None
Palmilla (breccia)
Las Minas VC
Green, gold hue, conchoid fracture, aphyric Green, gold hue, conchoid fracture, aphyric Green, gold hue, conchoid fracture, aphyric Green, gold hue, conchoid fracture, aphyric
XRF, ICPMS XRF, ICPMS XRF, ICPMS
None
Alfajayucan (breccia) Las Minas VC Mine well (block & ash flow)
Las Minas VC
El Nopalillo (block & Las Minas VC ash flow) El Nopalillo (block & Las Minas VC ash flow El Nopalillo (breccia) Las Minas VC El Nopalillo (fall deposit) Cruz el Milagro
Las Minas VC Las Minas VC
La Joya (fall deposit)
Las Minas VC
La Joya (fall deposit)
Las Minas VC
El Horcon base
El Horcon VC?
Ixtula Sembo
El Horcon VC?
Ixtula Sembo
El Horcon VC?
San Miguel Regla (debris avalanche) San Bartolo (debris avalanche) ~ a el Aire (debris Pen avalanche) San Bartolo (debris avalanche) Canneto Dentro (fall deposit) Canneto Dentro (fall deposit) Canneto Dentro (fall deposit) Pomiciazzo (lava flow) Forgia Vecchia (lava flow) Gabellotto (fall deposit) Fiumebianco (fall deposit) Fiumebianco (fall deposit) Pilato (fall deposit)
__
EE 9
Lami (fall deposit)
EE 9
Lami (fall deposit)
EE 9
__ __ __ EE 8 EE 8 EE 8 EE 9 EE 9 EE 9 EE 9 EE 9
Tiger. The percentage difference between certified and measured composition of standards analyzed as unknown samples was always 70 wt% and total alkalies at around 10 wt% (Table 2). However, in spite of this general similarity, the obsidians from the two localities show important chemical differences: Lipari samples have a ratio Al/ alkalies of about 1 and K2O > Na2O, while Sierra de las Navajas samples are peralkaline (total alkalies > Al) and more Na2O-rich (Fig. 4a and b). On the whole, the composition of Lipari obsidians is rather constant in terms of both major and trace elements, while Sierra de las Navajas samples show a higher variability in silica and many other elements. This heterogeneity, however, does not correlate with the time of emplacement (pre- or post-debris avalanche), and obsidians of the same volcanic complex (and even of the same volcaniclastic sequence) show different contents for many elements. This is evident, for example, in the TiO2-Fetot n obsidians diagram (Fig. 5a), where both Las Minas and El Horco show variable contents in the two elements; Lipari samples have lower contents in both elements. A more marked variation can be observed in some trace elements: Ba and Rb have a negative correlation in the Sierra de las Navajas samples (Fig. 5b), while Nb and Zr are positively correlated (Fig. 5c). In these cases, too, at least two groups of samples can be distinguished in the Sierra de las Navajas, with different incompatible elements contents. Lipari obsidians are Ba-, Nb- and Zr-poorer and Rb-richer (Fig. 5b and c). In Fig. 5 diagrams, the light grey obsidian of Lami and the dark grey, porphyritic obsidian of Sierra de las Navajas are represented by using a different symbol. As can be observed in the diagrams, their composition is consistent with that of other samples of the same group. In the spider diagram of REE (Fig. 5d) the obsidians of Sierra de las Navajas show slightly more variable contents in all the elements than those of Lipari. In particular, a certain variation in the HREE (Tb to Lu) contents is observed in the group of Las Minas obsidians. The pattern of the dark grey obsidian of Ixtula Sembo is very similar to that of “normal” green samples. Lipari obsidians have a more pronounced Eu anomaly than Sierra de las Navajas samples. Even in this case though, no appreciable difference can be observed between the Lami grey obsidian and the other black samples. 5.2. Micro- and nano-vesicularity Micro- and nano-vesicularity of the obsidians was investigated by using SEM and Hg-porosimetry. An X-ray micro-CT) measurement was also performed on one sample of Sierra de las Navajas green obsidian. Under the SEM, Lipari black obsidians do not show any evidence of incipient vesiculation, either externally or internally. On the other hand, the external surface and the inner portion of Lipari's grey obsidians (Lami volcanic center) show many small (