Systematic study of the lithic resources’ regional base (Ericson Reference Ericson, Ericson and Purdy1984) is one of the main approaches to analyzing the landscape’s use by human populations (e.g., Hughes Reference Hughes, Ramenofsky and Steffen1998; Renfrew Reference Renfrew and Earle1977; Torrence et al. Reference Torrence, Kelloway and White2013). In particular, the location and characterization of obsidian deposits have been key to studying many dimensions of the human use of space; for example, mobility, interaction, and exchange networks at different spatial and temporal scales (Frahm Reference Frahm2025; Kuhn Reference Kuhn2004; Shackley et al. Reference Shackley, Morgan and Pyle2018; Smith and Harvey Reference Smith and Harvey2018). That is because obsidian has some special features: the long durability of artifacts; its limited number of sources; its chemical composition, which in general is homogeneous and distinctive of each source; and the ability to attribute artifacts to specific sources with a high degree of accuracy through chemical analysis (Eerkens et al. Reference Eerkens, Spurling and Gras2008; Glascock et al. Reference Glascock, Braswell, Cobean and Shackley1998; Negash and Shackley Reference Negash and Shackley2006; Rademaker et al. Reference Rademaker, Glascock, Kaiser, Gibson, Lux and Yates2013; Renfrew et al. Reference Renfrew, Dixon and Cann1965; Yacobaccio et al. Reference Yacobaccio, Escola, Lazzari, Pereyra and Glascock2002).
Although there are many volcanic glass deposits in different parts of the world, not all forms of obsidian are suitable for knapping activities (e.g., Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; Fernández and Leal Reference Fernández and Leal2014; Frahm Reference Frahm2025). First, there was a temporary constriction in the preservation of its vitreous properties, so most of the pre-Tertiary deposits cannot be considered (e.g., Glascock Reference Glascock2020). Thus, obsidian deposits of archaeological interest are limited to those located in tectonic regions with volcanism dating from the early Tertiary onward. Second, there is compositional conditioning because not all lava flows are rich enough in silicon to form glass (Glascock et al. Reference Glascock, Braswell, Cobean and Shackley1998). Together, these features limit the number of possible deposits, which makes it possible to characterize all of them and to study the relative importance of each obsidian source.
Patagonia is a large region located in the southern part of South America in Argentina and Chile (Figure 1). Even though it is a very large area, there are few obsidian sources in Patagonia compared to other volcanic regions of the world (e.g., Frahm Reference Frahm2023; Stern Reference Stern2018). However, although they are limited in number, more deposits are becoming known. It is likely that this number will continue to increase as new areas are included and prospecting work in the field expands.
Study area: Argentinian and Chilean Patagonia and their provinces and regions. (Color online)
The study of obsidians in archaeological sites in Patagonia has yielded much information (Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019, Reference Barberena, Hajduk, Gil, Neme, Durán, Glascock and Giesso2011; Bellelli et al. Reference Bellelli, Pereyra, Carballido, Maggetti and Messiga2006; Fernández et al. Reference Fernández, Barberena, Rughini, Giesso, Cortegoso, Durán and Villanueva2017; Stern Reference Stern2018) that has been integrated to some extent (e.g., Pallo and Borrero Reference Pallo and Borrero2015; Stern Reference Stern2018). We report here on the first macroregional-scale review of the total number of obsidian groups that considers not only the most prominent archaeological sources but also those that have been less intensively used or of limited geographic scope.
Following Glascock and others (Reference Glascock, Braswell, Cobean and Shackley1998), our macroregional study sought to describe all known natural obsidian deposits south of the Colorado River in Argentina and south of Region VII in Chile, thereby encompassing the entirety of Patagonia (Figure 1). It determined each spatial occurrence of obsidian known to date and the different geochemical types at each site. Obsidians with the same chemical signature can emerge or be distributed in various areas, forming different sources that would have different implications for population home range analysis (e.g., Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; see Fernández et al. [Reference Fernández, Barberena, Rughini, Giesso, Cortegoso, Durán and Villanueva2017] for the case of Maule obsidian).
This macroregional approach is relevant in the Patagonian context because its spatial scale is appropriate to the high-mobility societies that inhabited this territory. Its aims were to compile, locate, standardize, and characterize all geographic occurrences of obsidian in Patagonia and indicate their geochemical signal, if it could be determined, and its correspondence with the geochemical types already known. We recommend the development of an obsidian geoarchaeological program on a macroregional scale.
Patagonia Obsidians
Study Area
Patagonia is a geographical, cultural, and historical region in the southern end of South America that is part of both Argentina and Chile (Figure 1). Orographically, climatically, and biologically, it is divided into Andean and extra-Andean zones. The Patagonian Andes are the southern end of the Andean Mountain range. It is a strip that is about 2,000–2,500 km long and nearly 100–250 km wide in a predominantly north–south direction (Clapperton Reference Clapperton1993). The Andes function as a topographic barrier to the prevailing westerly storm tracks, affecting the patterns of circulation and imposing a steep west–east gradient on precipitation effective moisture (Garreaud et al. Reference Garreaud, Vuille, Compagnucci and Marengo2009). Thus, annual precipitation varies from about 3,000 mm close to the Chile–Argentina border to 200–150 mm in the eastern lowlands of Argentina. Vegetation distribution follows this pattern: from rainforest to the west to grass and shrub steppes toward the eastern Argentinian lowlands, with desert vegetation characteristic of the Monte phytogeographic province (Chiapella and Ezcurra Reference Chiapella and Ezcurra1999; León et al. Reference León, Bran, Collantes, Paruelo and Soriano1998).
Volcanic activity is discontinuous along this range, and discrete volcanic zones exhibit different characteristics (Muñoz and Stern Reference Muñoz and Stern1988; Stern Reference Stern, Civalero, Fernández and Guráieb2004). The type of silicic volcanic activity that is conducive to the formation of obsidian deposits is restricted to certain segments of the Andes (Cabrera et al. Reference Cabrera, Weinberg, Wright, Zlotnik and Cas2011). In this region, Andean obsidian deposits are associated with both arc and back-arc settings, presenting varying ages of formation that extend from the Miocene to the Holocene (e.g., Folguera et al. Reference Folguera, Bottesi, Zapata and Ramos2008; Hildreth et al. Reference Hildreth, Godoy, Fierstein and Singer2009). These diverse geological and chronological contexts, as well as the subsequent action of exogenic geomorphic processes, produce a complex obsidian availability landscape in which different types of obsidian deposits exist—from primary outcrops in the Andean highlands to secondary accumulations in ignimbrite deposits and fluvial landforms. In addition, some Patagonian obsidian deposits are associated with the Miocene centers that formed the back-arc lavas flow of the Patagonian plateau (Bellelli et al. Reference Bellelli, Carballido and Stern2018; Ramos and Kay Reference Ramos and Kay1992).
Obsidian Studies in Patagonia
Many studies have been conducted since the early 1990s to search for and characterize obsidian sources used by human populations in Chilean and Argentinian Patagonia (e.g., Stern and Curry Reference Stern and Curry1995; Stern and Porter Reference Stern and Porter1991; Stern and Prieto Reference Stern and Prieto1991). The geochemical uniformity of the deposits encouraged these efforts, because it facilitated studies that addressed some of the key questions on the research agenda in the region over the past 40 years (e.g., Borrero Reference Borrero1994–1995; Boschín and Massaferro Reference Boschín, Massaferro, de Lara and Rocchietti2014; Franco and Borrero Reference Franco, Borrero, Aschero, Korstanje and Vuoto1999; Méndez et al. Reference Méndez, César, Delaunay, Reyes, Gutiérrez and Mena2018). Toward the end of the 1990s and the beginning of the twenty-first century, those efforts paid off, and the obsidian deposits most used by the societies that inhabited Patagonia were identified (Stern Reference Stern, Juan, Pablo, Rafael, Ana and Mariana1999, Reference Stern and Franco2000; Stern and Franco Reference Stern and Franco2000; Stern et al. Reference Stern, Gómez Otero and Belardi2000, Reference Stern, Bellelli, de Micou, Morello, Martinic, Prieto and Bahamonde2007, among others). Interdisciplinary work with geologists was a common feature in almost all these studies. However, the geologists were primarily focused on the technical determination and interpretation of the geochemical analyzes and not on the systematic research design, evaluation, and general characterization of the deposits in the field. This in part could be related to the limited interest that they had at that time in the study of obsidians (see Frahm Reference Frahm2025:13).
With the main geochemical groups known and their composition characterized, the next stage was to increase the number of geochemical analyses performed on archaeological artifacts from different areas of Patagonia. In some areas, especially in Neuquén Province, portable XRF (pXRF) spectrometers have recently become the most-used tool for obsidian characterization studies. The advantages of pXRF are that it is affordable and transportable, yields very quick results, is nondestructive, and has good resolution for most obsidians (Glascock Reference Glascock2020). It also enables analysis both of natural and archaeological obsidians: it identified patterns of use of this raw material and distances of occurrence of artifacts using fall-off curves and other methodologies (e.g., Fernández and Leal Reference Fernández and Leal2017; Pallo and Borrero Reference Pallo and Borrero2015; Pérez et al. Reference Pérez, Giesso and Glascock2019).
The increased number of geochemical analyses performed on archaeological artifacts made visible a greater diversity and complexity in the characterization of the different geochemical groups at variable scales. The recurrent appearance of unknown geochemical types highlighted the increasingly urgent need to deepen our knowledge of the regional obsidian base. These “unknown” or “indeterminate” sources gradually began to be characterized and integrated into archaeological discussions. Yet at least three main problems persist. First, there is a mismatch between the smaller spatial scale of the archaeological work—focused on excavations of individual sites or regional prospections—and the larger scale of the questions about human mobility in the past. Added to this is the fact that archaeological practice in Patagonia is being conducted in two countries, which makes it difficult to develop cross-cutting projects in different sectors. Second, the generation of fragmentary information within archaeology and geology has led to the use of different nomenclatures to refer to the same site/deposit. For example, the currently known Portada Covunco source was named initially “La Bandera” (Giesso et al. Reference Giesso, Berón and Glascock2008) or Río Kilca (Stern and Pereda Reference Stern and Pereda2011; Stern et al. Reference Stern, Pereda and Aguerre2012) because of the use of different deposits’ locality names during the early research (see Palacios et al. [Reference Palacios, Vázquez, Hajduk, Bertolino, Cattáneo and Izeta2010] for another example). Finally, the implementation of different analytical methods, the use of different chemical elements, or both have hindered comparisons of obsidian types between sources or even of results of the different published investigations (see Palacios et al. [Reference Palacios, Vázquez, Hajduk, Bertolino, Cattáneo and Izeta2010] for an example of northern Patagonian sources).
As a result, a coherent and complete set of data about the geographical and geochemical diversity of obsidians on a Patagonian macroscale is still lacking. This is a great obstacle to understanding human mobility, exchange, and landscape use at various scales.
Materials and Methods
Following Hughes (Reference Hughes, Ramenofsky and Steffen1998), we use the term “geochemical group” to refer to the unit that emphasizes the geochemical uniformity/homogeneity of each obsidian type, regardless of its spatial distribution. An “obsidian deposit” refers to the natural locations of obsidian, whether human groups have used it or not.
Each geochemical group has one or more associated deposits that may have varying characteristics and may be at different distances from their primary manifestations, depending on the geological and geomorphological processes at work. The number of locations may be increased with advances in the geomorphological and geological knowledge of the events that gave rise to the different obsidian deposits.
We distinguish three types of deposits: primary, secondary (Church Reference Church1994), and undetermined. Primary deposits consist of lava flows or pyroclastic bombs around volcanic centers, which are often grouped on structural lava plains or plateaus. Secondary deposits result from the erosive activity of geomorphological agents such as glaciers, rivers, and gravity that transport the material to places from its primary occurrence. Undetermined deposits lack sufficient information to be classified as either primary or secondary.
Within each deposit type, significant natural variants exist. One important consideration is the extent of a deposit and how to accurately delimit it. In previous studies we proposed a 1 km boundary: any occurrence of obsidian beyond this limit was considered a new deposit (Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019). However, at the macroregional scale, such a low scale limit cannot be homogeneously applied across the entire territory because of a lack of baseline information. Therefore, we labeled all reported deposits, regardless of their precise location, clustering, or distance from other occurrences.
To compile all the known obsidian geochemical groups and deposits in the Argentine and Chilean Patagonia, we reviewed the following geological and archaeological studies:
• Geological reports of the Servicio Geológico Minero Argentino south of the Colorado River available from the Departmento de Ciencias Geológicas de la Facultad de Ciencias Exactas, Universidad de Buenos Aires (N = 27)
• All the proceedings of the Argentine Geological Congresses (N = 20)
• All the proceedings of the Congresses of Mineralogy and Metallogeny (N = 12)
• All issues of the Revista de la Asociación Geológica Argentina (N = 308)
• GAEA, the database of all Argentinian geological publications before 1998
• International works of Chilean geology available on the internet (N >10)
• All the proceedings of the Jornadas de Arqueología de la Patagonia (N = 11)
• Archaeological papers published to date regarding both Chilean and Argentinian Patagonia (N > 50)
The geographic characterization of the deposits was carried out following the recommendations of Glascock and others (Reference Glascock, Braswell, Cobean and Shackley1998) and the categories proposed by the International Association for Obsidian Studies for the preparation of its database. In addition, we selected the following variables that we considered pertinent to the objectives of this work: geochemical group ID, name, and analysis; deposit ID, name, and type; province/region and country; meters above sea level; macroscopic description (color, quality, and maximum nodule size); and first references (the first paper in which each geochemical group was mentioned).
Based on this information, we built a map of all obsidian geochemical groups and deposits using QGIS 3.10 (http://www.qgis.org). When we located more than one deposit of the same geochemical group, the same color was used to identify them. In cases where the geographical coordinates were not published, the location was made by approximation and comparison with the graphical data in available publications. Deposit elevations were obtained from the published literature. When elevation was not explicitly stated, it was derived from the deposit’s geographical location using Google Earth Pro. This elevation measurement served as a proxy for the seasonal versus annual accessibility of each deposit.
To characterize each geochemical group, we selected the following variables: the main geochemical group, the geochemical subgroup, the geochemical method of identification, the number of natural deposits identified, and the archaeological representation.
Spatial distribution patterns of obsidian deposits were analyzed using point pattern analysis (PPA) techniques within the R spatstat package (Baddeley et al. Reference Baddeley, Rubak and Turner2015; Wickham et al. Reference Wickham, François, Henry and Müller2023). Deposit locations, categorized by type (primary, secondary and undetermined), were represented as a marked point pattern process (PPP) object, with the study area defined as its observation window (owin). Kernel density estimation provided visual insights into spatial variations. Formal assessment of spatial trends involved fitting inhomogeneous Poisson process models (PPMs) using x and y coordinates as covariates. The significance of these spatial trends was evaluated using ANOVA to test whether the inclusion of geographic coordinates as predictors of intensity provided a significantly better fit than a null model of complete spatial randomness, which assumes a constant intensity across the study area. Finally, a Monte Carlo segregation test was conducted on the marked pattern to evaluate intertype spatial mixing or segregation.
Results
Table 1 presents all the Patagonian obsidian geochemical groups, their associated deposits, and their main variables from north to south in continental Patagonia and then in insular territories. Altogether, 30 geochemical groups were spatially distributed in 57 known deposits throughout the Patagonian territory (Figure 2; Table 1). The number of geochemical groups could increase because there are nine deposits for which geochemical information is not yet available.
Localization of all obsidian geochemical groups and their natural deposits along Patagonia. 1–30, geochemical groups; A–I, undetermined deposits (see Table 1). (Color online)
Patagonia Obsidian Geochemical Groups, Deposits, and Their Main Associated Variables.
* The elevation was taken from Google Earth Pro.
Of these obsidian geochemical groups, six are located on the western slope of the Andean range (within present-day Chile) and 24 on the eastern slope (in current Argentinian Patagonia). On the eastern slope, these deposits are distributed across provinces as follows: 28 in Neuquén, six in Río Negro, five in Chubut, and nine in Santa Cruz. In Chile, the six obsidian groups and their deposits are highly dispersed, spanning from the northernmost to the southernmost parts of Patagonia: one is in the Maule region, one in Araucanía, two in Los Lagos, one in Aysén, and one in Magallanes. Although all these deposits were already identified in the literature, the amount and quality of associated information about them differ.
It is noteworthy that seven additional obsidian deposit locations were identified from geological information. Two of these deposits (Sierra Pailemán and Pire Mahuida) were identified in geological papers (Hayese and Manera Reference Hayase and Manera1969; Salani and Page Reference Salani and Page1989, respectively). The remaining five deposits were registered in geological reports of the Servicio Geológico Minero Argentino: Pichileufú (Giacosa and Heredia Reference Giacosa and Heredia2001), Escondida and Magdalena (Genini and Franchi Reference Genini and Franchi2001), and Ventisqueros and Valenciana (Panza and Cobos Reference Panza and Cobos2001). The geographic locations of the natural deposits of geochemical groups ID#28 (Cordillera Baguales) and ID#29 (Seno de Otway) have not yet been determined (e.g., Morello et al. Reference Morello, Roman, Prieto, Civalero, Fernández and Guráieb2004, Reference Morello, Stern and Román2015; Stern Reference Stern2018; Stern and Franco Reference Stern and Franco2000). However, possible locations have been published based on the frequency of obsidian artifacts of these geochemical groups as shown in Figure 2.
The published data indicate a great variability of scenarios in which obsidians naturally occur (see Figure 3 for scenarios examples). Of the total locations (N = 66), more than half are secondary deposits (59.1%, n = 39). Primary deposits were identified at 21 locations (31.8%), whereas the remaining 9.1% (n = 6) have not yet been determined.
(a) Laguna del Maule deposit (ID#1.1; Laguna del Maule 1-Laguna Negra geochemical group); (b) Río Barrancas deposit (ID#4.2; Laguna del Maule 2-Río Barrancas geochemical group); (c) Upper Varvarco River deposit (ID#2.1; Varvarco geochemical group); (d) Cerro Huenul deposit (ID#5.1; Cerro Huenul geochemical group). (Color online)
Only 37 deposits (56.1%) have published associated altitude data. Among these, elevations range from 161 to 2,435 m asl. However, most deposits with altitude data are located between 1,000 and 2,000 m asl (n = 26, representing 39.1% of the total 66 deposits and 70.3% of those with altitude data). For deposits without published altitudes (n = 29), estimated values based on location also indicate elevations not exceeding 1,639 m asl, with an average of 588 m asl.
Most deposits exhibit excellent obsidian quality for knapping (40.9%, n = 27). Only 13.6% (n = 9) show good or very good (medium) quality, 4.5% (n = 3) are of poor to good quality, and 15.1% (n = 10) are of poor quality and hence not suitable for knapping. Two deposits (3%) exhibit all qualities. Information regarding obsidian quality is unavailable for the remaining 22.7% (n = 15) of deposits.
The predominant color of Patagonian obsidians is black (Figure 3). However, most deposits present multiple colors or variants, including translucent, gray, reddish, greenish, and brownish variations. Banded obsidians are very common, with spotted ones being less frequent (Table 1).
The natural size of obsidians varies considerably among deposits. Notably, the largest obsidians are found in Chile’s VII Maule region and Argentina’s Neuquén Province, which are associated with primary Andean range deposits: Laguna del Maule 1 (Figure 3a) and Portada Covunco, respectively. Recently, a large extra-Andean primary deposit (Tres Cerros) was also reported in Santa Cruz Province in Argentina (Franco et al. Reference Franco, Brook, Ambrústolo, Iglesias, Stern, Glascock and Vetrisano2023).
Our initial observation reveals that only three geochemical groups—Portada Covunco, Río Cisnes, and Pampa del Asador—exhibit a high diversity of geochemical subgroups (three or more; Table 2). The remaining registered geochemical groups are characterized by greater chemical homogeneity.
Obsidian Geochemical Groups, Subgroups, Method of Determination, Number of Deposits with Obsidian of Each Group and Archaeological Representation.
Another pattern evident from Table 2 is the variability in the geographic occurrence of a few geochemical types, which contrasts with the homogeneity or geographic specificity of most groups. This variability is found in these geochemical groups—Varvarco, Maule 2-Río Barrancas, Cerro Huenul, Portada Covunco, Nevados de Solipulli, Cerro las Planicies, Volcán Chaitén, and Pampa del Asador—that occur in three or more geographic locations or deposits. It is possible, however, that this pattern reflects differences in sampling intensity more than the natural occurrence of the deposits.
Table 2 also details the diversity of geochemical methods used for obsidian characterization in Patagonia. Analyses were conducted using three methods: X-ray fluorescence (XRF), instrumental neutron activation analysis (INAA), and inductively coupled plasma–mass spectrometry (ICP-MS). Although many groups and subgroups were characterized using at least two methods, at the regional level ICP-MS was the most widely used analytical technique, followed by XRF analysis. INAA was used to a lesser extent and was typically applied when XRF analysis proved insufficient to distinguish among specific obsidian groups (e.g., Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; Fernández et al. Reference Fernández, Barberena, Rughini, Giesso, Cortegoso, Durán and Villanueva2017, Reference Fernández, Leal, Negra, Klesner, MacDonald, Glascock and Barberena2019). The use of portable XRF, however, has increased recently.
Finally, of the 30 known geochemical groups for Patagonian obsidians, 25 show evidence of archaeological use (Table 2), varying in intensity and transport distance. Among these 25 groups, only 12 were recovered from archaeological sites located more than 40 km from their source (i.e., nonlocal; sensu Meltzer Reference Meltzer, Ellis and Lothrop1989; see Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; Stern Reference Stern2018).
As anticipated, the distribution of obsidian deposits across the Patagonian landscape is not homogeneous. Figure 2 shows at least two major concentrations of deposits in northern Patagonia: one in the northwest and another in the southwest of Neuquén Province, Argentina, where the highest density and concentration are observed. Regionally, mountain sources are predominant, although extra-Andean sources are relatively abundant in eastern Patagonia’s basaltic plateaus.
To more precisely characterize the spatial distribution of these geographic occurrences, we employed two simple probability models: (1) complete spatial randomness, which assumes a random distribution with an equal probability of finding a deposit at any point, and (2) a nonstationary Poisson process, which models a scenario in which deposit distribution depends on specific spatial or temporal parameters.
The results indicate that the distribution of obsidian deposits in Patagonia is not random but instead conforms to a specific spatial pattern. The intensity of deposit distribution or the probability of finding one is compatible with the expectations of a nonstationary Poisson process exhibiting a longitudinal and latitudinal gradient. This signifies that deposit density significantly decreases toward the east and south (Npar Df Deviance; Pr (> Chi) = 44.681; 2.32e-11). As depicted in Figure 4, the model plotted with the geographic occurrence data shows an excellent fit; it also accounts for a certain density of occurrences, albeit on a smaller scale, toward the southwest of Patagonia.
Spatial analysis of obsidian deposit distribution. (Top left) Kernel density map of all deposits; (top right) estimated intensity based on the inhomogeneous Poisson process model (∼x + y), with observed deposit locations overlaid; (bottom) kernel density maps for primary, secondary, and undetermined deposit types. (Color online)
We used a Monte Carlo test of spatial segregation to analyze the distribution of deposits in relation to their type—primary, secondary, or undetermined: it did not reveal significant differences (t = 0.057977, p = 0.65). This indicates that the various deposit types are spatially mixed rather than showing significant tendencies for self-segregation or repulsion from other types within the study area.
Discussion
Obsidians in Patagonia
This discussion is organized at two scales: (1) a macroregional scale, examining broad variations in deposit distribution across Patagonia, and (2) a medium or regional scale, focusing on refining deposit concentrations.
The macroscale approach uncovers several key insights. Knowledge of a significant portion of the region’s known obsidian deposits derives from archaeological work. However, at least seven new locations were identified from the geological literature: ID A-Pichileufú, ID C-Sierra Pailemán, ID D-Pire Mahuida, ID E-Escondida, ID F-Magdalena, ID G-Ventisqueros, and ID H-Valenciana). These data, of course, need to be updated in terms of archaeologically relevant variables, such as the deposits’ quality, type, and geochemistry.
Our exhaustive compilation identifies at least 30 geochemical groups and 66 deposits, a number expected to increase with further research across the region. Although these figures represent a significant increase in the number of previously known geochemical groups and deposits for Patagonia, the area’s overall obsidian deposit density and diversity remain relatively low compared to other regions globally with similar extensions (e.g., Turkey and the Caucasus; Frahm Reference Frahm2023) or even much more limited areas such as Japan (Kuzmin Reference Kuzmin2024).
In addition, our results indicate a highly heterogeneous distribution of obsidian deposits across the region. There is a clear concentration of deposits in northwestern Patagonia, with decreasing density both eastward (toward the Atlantic Ocean) and southward. This finding holds important implications for characterizing diverse land-use patterns and the circulation of raw materials throughout the history of human occupation in the region. It suggests that certain Patagonian areas can be considered “source-scapes” (sensu Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019), rather than isolated provisioning locations. This definition provides a new framework for analyzing archaeological distributions, enabling the modeling and testing of hypotheses about the intensity and diversity of obsidian source use against the archaeological record. Based on our analysis, we identified two main source-scapes: one in northern and one in southern and western Neuquén Province. Each of these regions contains more than five distinct geochemical groups and numerous associated secondary deposits.
The location and altitude of obsidian deposits also provide information essential for understanding their accessibility and facilitating discussions on the seasonal use of different provisioning landscapes. If we set 1,500 m asl as the point beyond which sources are only accessible during the summer, then only 20 deposits from 11 geochemical groups would be restricted to summer accessibility, leaving 46 deposits accessible year-round. Among these 11 geochemical groups, two are also found in secondary deposits at lower elevations than their primary manifestations, meaning some locations are also accessible throughout the year.
We recognize that the obsidian deposits identified and mapped in this study are products of existing published archaeological and geological research and thus do not necessarily represent the total distribution or full natural variability of obsidian occurrences. Our results clearly demonstrate a significant disparity in the geographic coverage of available information, which affects not only the final count of known geochemical groups or deposits but also our understanding of their distribution, presentation, quality, and volume, all of which influence the characterization of human use.
These data gaps should not, however, limit efforts to generate multiscalar approaches for understanding the use of this essential resource throughout Patagonia’s human occupational history. On the contrary, we contend that developing macroregional maps is vital. Such maps not only allow us to contrast the natural occurrences of obsidians with their use at archaeological sites but also they effectively highlight information gaps, indicating areas where future research efforts should be focused. At this point, our work reveals a notable disparity in deposit frequency between the western and eastern slopes of the Andean range. It remains unclear whether this reflects a true difference in natural availability, a need for increased sampling efforts on the Andean western slope, or, more likely, a combination of these factors.
On a finer scale (regional level), our results highlight the critical need to design specific research programs to address issues that are neither strictly archaeological nor geological but rather are geoarchaeological and thus need a truly integrated approach. Such an approach moves beyond merely consulting a geologist as a technician for geochemical data interpretation to conducting joint archaeological and geological work. At a minimum, such collaboration involves specific research designs, exhaustive study of geological antecedents, and joint fieldwork and landscape interpretation. In this context, chemical characterization becomes just one of many ways to understand each deposit’s origin, development, and final disposition within the landscape.
Adopting such a geoarchaeological approach can yield fundamental insights into the role of different deposits in human dynamics. A compelling example is Laguna del Maule obsidian. Initially, it was categorized as a single primary source for both Maule-1 and Maule-2 geochemical groups and was identified as natural nodules near the main source or as archaeological artifacts in nearby or distant sites (Giesso et al. Reference Giesso, Durán, Neme, Glascock, Cortegoso, Gil and Sanhueza2011; Seelenfreund et al. Reference Seelenfreund, Rees, Bird, Bailey, Bárcena and Durán1996). Over time, adopting a geoarchaeological approach facilitated a more detailed understanding of the differential spatial arrangement of these two geochemical groups (Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; Fernández et al. Reference Fernández, Barberena, Rughini, Giesso, Cortegoso, Durán and Villanueva2017): the Laguna del Maule 1-Laguna Negra group is present in the highlands of the Barrancas River, whereas the Laguna del Maule 2-Río Barrancas obsidian group occurs in the lowlands of the Barrancas and Colorado Rivers. This finding significantly affected previous interpretations of regional mobility and landscape use, necessitating their thorough reevaluation (e.g., Giesso et al. Reference Giesso, Durán, Neme, Glascock, Cortegoso, Gil and Sanhueza2011; Salgán et al. Reference Salgán, Gil and Neme2014).
The heterogeneity observed within the most diverse geochemical groups is another area requiring further study. In Patagonia, these include the Portada Covunco, Pampa del Asador, and Río Cisnes geochemical groups. We recorded between three (Pampa del Asador and Río Cisnes) and four (Portada Covunco) geochemical subgroups within these groups. This variability could stem from several factors. First, it might relate to the deposits’ formation history, implying that internal variability is inherent. Second, it could be explained by some methodological issues (De Francesco et al. Reference De Francesco, Barca, Bocci, Cortegoso, Barberena, Yebra and Duran2018). In cases in which the obsidian is not completely glassy and may contain numerous microliths, some methods such as laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) yield multiple results for each fragment. That is because they analyze a smaller volume than other techniques, such as XRF, which may represent the whole composition only in perfectly vitreous and homogeneous obsidians (De Francesco et al. Reference De Francesco, Barca, Bocci, Cortegoso, Barberena, Yebra and Duran2018). Finally, the number of samples analyzed in each deposit could potentially lead to more thoroughly analyzed deposits appearing more diverse. However, this seems less plausible, at least in Neuquén Province and Chile’s VII Region, given the greater geochemical homogeneity observed in the most-studied areas (e.g., northwest Neuquén).
This last point also prompts a critical discussion on the definition of subgroups within larger geochemical groups: What constitutes the accepted limit between subgroups? What role does or should geoarchaeological spatial analysis in the field play in this definition? These issues are crucial for obsidian studies in Patagonia, given their extensive history that now enables such definitions to be refined.
Archaeological Obsidian Importance
Studying the regional raw material base is critical for assessing past land use, and few materials are as valuable for this research as obsidian. However, our work clearly demonstrates that a deeper understanding of the entire obsidian supply is necessary before there can be an accurate characterization of source use. Our results indicate a greater complexity in the presence of geochemical groups and deposits across Patagonia than previously recognized. In this context, the concept of a source-scape will aid in the development of research programs that incorporate, from their outset, not only geological analytical methods but also a comprehensive geoarchaeological perspective.
Of the 30 geochemical groups identified, 25 have evidence of use in archaeological sites across Patagonia (Alberti et al. Reference Alberti, Cardillo, Stern and Dubois2016; Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; Bellelli et al. Reference Bellelli, Pereyra, Carballido, Maggetti and Messiga2006, Reference Bellelli, Carballido and Stern2018; Boschín and Massaferro Reference Boschín, Massaferro, de Lara and Rocchietti2014; Gómez Otero et al. Reference Gómez Otero and Stern2005; López, Pérez, et al. Reference López, Pérez, Batres and Stern2009; López, Pérez, and Stern Reference López, Pérez and Stern2009, Reference López, Silveira and Stern2010; Méndez et al. Reference Méndez, César, Reyes and Mena2012, Reference Méndez, César, Delaunay, Reyes, Gutiérrez and Mena2018; Méndez Melgar et al. Reference Méndez Melgar, Stern and Báez2008–2009; Morello et al. Reference Morello, Stern and Román2015; Pérez and López Reference Pérez and López2004, Reference Pérez and López2007, Reference Pérez and López2010; Pérez et al. Reference Pérez, López and Stern2012, Reference Pérez, Giesso and Glascock2019; Seelenfreund et al. Reference Seelenfreund, Rees, Bird, Bailey, Bárcena and Durán1996, Reference Seelenfreund, Miranda, Dinator and Morales2005; Sierralta Navarro Reference Sierralta Navarro, Rodríguez, Castillo and Urbina Araya2025; Stern Reference Stern, Juan, Pablo, Rafael, Ana and Mariana1999, Reference Stern, Belardi, Marina and Espinosa2000, Reference Stern, Civalero, Fernández and Guráieb2004, Reference Stern2018; Stern et al. Reference Stern, Gómez Otero and Belardi2000, Reference Stern, Bellelli, de Micou, Morello, Martinic, Prieto and Bahamonde2007, Reference Stern, Navarro, Pino and Vega2008, Reference Stern, Pereda and Aguerre2012) and nearby or distant areas (e.g., Giesso et al. Reference Giesso, Berón and Glascock2008; Stern and Aguerre Reference Stern and Aguerre2013; Suárez et al. Reference Suárez, Pistón, Julia Melián, Silva, Faccio, Davenport, Arbiza and Glascock2025; Yebra et al. Reference Yebra, Cortegoso, Castro and Lucero2024). Not all these groups were used equally intensively or across the same spatial scale. Notably, 12 geochemical groups—Maule 1, Maule 2, Cerro Huenul, Portada Covunco, Nevados de Solipulli, Cerro de Las Planicies/Lago Lolog, Sierra Negra/Arroyo Telsen, Sacanana, Chaitén, Pampa del Asador, Cordillera Baguales, and Seno de Otway—were found in geographic contexts extending beyond the immediate vicinity (more than 40 km) of currently known natural obsidian deposits (Barberena et al. Reference Barberena, Hajduk, Gil, Neme, Durán, Glascock and Giesso2011, Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019; Bellelli and Pereyra Reference Bellelli and Pereyra2002; Bellelli et al. Reference Bellelli, Pereyra, Carballido, Maggetti and Messiga2006; Durán et al. Reference Durán, Giesso, Glascock, Neme, Gil and Sanhueza2004; Emperaire and Laming Reference Emperaire and Laming1961; Espinosa and Goñi Reference Espinosa, Goñi, Belardi, Fernández, Goñi, Guráieb and Nigris1999; Fernández et al. Reference Fernández, Leal, Negra, Klesner, MacDonald, Glascock and Barberena2019; Giesso et al. Reference Giesso, Berón and Glascock2008; Gómez Otero et al. Reference Gómez Otero, Belardi, Súnico, Taylor, Juan Bautista, Pablo Marcelo, Rafael, Ana Gabriela and Mariana De1999; Gómez Otero and Stern Reference Gómez Otero and Stern2005; Morello et al. Reference Morello, Roman, Prieto, Civalero, Fernández and Guráieb2004, Reference Morello, Stern and Román2015; Navarro Reference Navarro and Ximena2012; Palacios et al. Reference Palacios, Vázquez, Hajduk, Bertolino, Cattáneo and Izeta2010; Pérez and López Reference Pérez and López2004; Seelenfreund et al. Reference Seelenfreund, Rees, Bird, Bailey, Bárcena and Durán1996; Sierralta Navarro et al. Reference Sierralta Navarro, Rodríguez, Castillo and Urbina Araya2025; Stern Reference Stern, Juan, Pablo, Rafael, Ana and Mariana1999, Reference Stern, Belardi, Marina and Espinosa2000, Reference Stern2018; Stern and Curry Reference Stern and Curry1995; Stern and Franco Reference Stern and Franco2000; Stern and Porter Reference Stern and Porter1991; Stern and Prieto Reference Stern and Prieto1991; Stern et al. Reference Stern, Gómez Otero and Belardi2000, Reference Stern, Navarro, Pino and Vega2008). Some of these groups have been recovered as artifacts in sites located at distances greater than 300 km or even more than 1,000 km from their currently known deposits (Stern Reference Stern2018; Suárez et al. Reference Suárez, Pistón, Julia Melián, Silva, Faccio, Davenport, Arbiza and Glascock2025). Further research is needed to investigate the acquisition mechanisms for this raw material and its relative importance compared to other rock types. Moreover, considering temporal variations in obsidian use, certain geochemical groups—Cerro Huenul, Portada Covunco, Cerro de Las Planicies/Lago Lolog, and Pampa del Asador—stand out for showing evidence of early use, even from the initial stages of human occupation in various areas of Patagonia and the Pampas (Barberena et al. Reference Barberena, Hajduk, Gil, Neme, Durán, Glascock and Giesso2011; Civalero and Franco, Reference Civalero and Franco2003; Franco and Borrero Reference Franco, Borrero, Miotti, Salemme and Flegenheimer2012; Paunero Reference Paunero, Miotti, Salemme and Flegenheimer2003; Paunero and Castro, Reference Paunero and Castro2001; Stern Reference Stern2018).
Conclusion
This work’s combination of geological, archaeological, and, especially, geoarchaeological data, which have been integrated into a macroscale regional approach spanning the whole of Patagonia, provides a solid basis for constructing a significant source-scape (sensu Barberena et al. Reference Barberena, Fernández, Rughini, Borrazzo, Garvey, Lucero and Negra2019). In addition to compiling an updated map of all obsidian occurrences—a major undertaking never before achieved in Patagonia—the presentation of geochemical groups identified in each natural deposit offers a clearer picture of the possible provenance sites for archaeological artifacts. By integrating results related to these issues, we enhance our knowledge of the source-scape in which people selected rocks and moved across the landscape. Furthermore, the complete source-scape resulting from this effort will enrich and likely affect previous interpretations of obsidian provisioning strategies and mobility circuits across Patagonia. Each obsidian deposit holds a different value, derived not only from its intrinsic features (quality, density, abundance, size, etc.) but also, significantly, from its relationship with other deposits.
Finally, even though comparing and understanding the diverse analyses conducted over more than 30 years presents challenges, the results are invaluable. A substantial amount of archaeological and geological work now exists that requires integration. Not all deposits possess the same level of information or data resolution, which defines a clear work schedule for future research. Nor are the temporal differences in the use of various geochemical groups and deposits fully understood. We also still need to determine how paleoenvironmental variations may have affected the availability and distribution of certain deposits and, consequently, their potential for human use. These efforts undoubtedly need to be part of a future interdisciplinary research agenda.
Acknowledgments
We thank Marcelo Cardillo for his statistical support, Alhué Bay Gavuzzo for her graphical assistance in Figure 1, Nora Franco for helping locate 17 M and Tres Cerros obsidian deposits, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), and Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación, Argentina.
Funding Statement
The writing of this article was supported by Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación (PICT 2020- SERIE A-00577) granted to Federico Luis Scartascini and by Consejo Nacional de Investigaciones Científicas y Técnicas, Argentina (PIP-CONICET 11220210100784CO) granted to Ramiro Barberena.
Data Availability Statement
Our original data are presented in this article.
Competing Interests
The authors declare none.
