Introduction
Although a consensus exists within the archaeological community regarding the relative chronology of the Chinchorro mummification practices along the Andean south-central coast (between Ilo, 18°28′, 70°19′S, and the Loa River, 17°37′, 71°20′S), during the Middle Archaic and Late Archaic periods (7000–3000 BP), there is ongoing debate regarding the absolute chronology of various mummification strategies employed by the Chinchorro populations and regional variations. Discrepancies persist in datasets and samples (Arriaza et al. Reference Arriaza, Doubrava, Standen and Haas2005, Reference Arriaza, Figueroa, Ogalde, Standen, Halcrow, Amarasiriwardena, Van Hoesen and Castelleti2023; Silva et al. Reference Silva-Pinto, Cáceres, Tonko-Huenucoy and Carpary2021) hindering a unified understanding.
Current research on Morro de Arica’s northeastern slope, a densely occupied Chinchorro area (Figure 1), yields new insights into domestic contexts associated with these classic funerary sites. This study provides a suite of radiocarbon dates, enabling evaluation of existing data and radiocarbon sample methodologies. Our analysis incorporates absolute datasets reported from Morro funerary area, specifically Morro 1, Morro 1/5 and Morro 1/6 (Arriaza and Standen Reference Arriaza and Standen2016; Focacci and Chacón Reference Focacci and Chacón1989; Guillén Reference Guillén1997; Standen Reference Standen1997, Reference Standen2003),
(a) Map of Chinchorro archaeological sites at the Arica coast; (b) test pits and funerary archaeological sites at Chinchorro reserves in the Morro de Arica.
The aforementioned research facilitates a dual evaluation: assessing pre-mummification domestic occupational at Morro de Arica (Castelleti et al. Reference Castelleti, Arriaza, Solís, Rodríguez Ceja, Silva, de Ugarte, Cisternas and Vega2024) and comparing classic radiocarbon data from local funerary (e.g., mummies) with newly obtained dates from recent Morro area field work. This updated information enables us direct comparison of paired terrestrial and marine samples by stratigraphic level, shedding light on disturbed local stratigraphy. Furthermore, our study aims to elucidate implications of the reservoir effect on the dating of malacological and charcoal samples (cf. Kennet et al. Reference Kennet, Ingram, Southon and Wise2002; Kim et al. Reference Kim, Wright, Hwang, Kim and Oh2019; Ortlieb et al. Reference Ortlieb, Vargas and Saliège2011).
Background
The Arica region (18°28′−70°18′S) in the Atacama Desert of Chile is defined by a tropical hyper arid desert bioclimate with little associated vegetation. The territory of Arica presents an exorheic water regime of reduced flow, with some streams draining to the sea, which together with the camanchaca or coastal fog has allowed the formation of vegetation such as Nolana adansonii and Nolana lycioides tropical desert scrubs in the coast, and Malesherbia auristipulata and Tarasa operculata to the interior, in association with arboreal species like Prosopis sp. (tamarugo), Schinus molle, and Vachellia macracantha (Crom Reference Crom1988−Reference Crom1989; Luebert and Pliscoff Reference Luebert and Pliscoff2006). This vegetation allowed for the cohabitation of a limited terrestrial fauna. In addition, the river mouth’s natural wetlands support diverse vegetation, including many marsh plants like club-rush (Scirpus sp.), bulrush (Typha sp.), foxtail (Polypogon interruptus), and horsetail (Equisetum giganteum), alongside shrubs such as brea or sorona (Tessaria absinthioides), and chilca (Pluchea chingoyo) with grassy area dominated by saltgrass (Distichlis spicata) (García and Belmonte Reference García, Belmonte, Arriaza and Standen2016).
The Arica region falls within “Andean Western Valleys” cultural zone, an important mobility corridor connecting the Andean Altiplano (highlands) to the coast (Crom Reference Crom1988−Reference Crom1989; Muñoz et al. Reference Muñoz, Agüero, Valenzuela, Falabella, Uribe, Sanhueza, Aldunate and Hidalgo2016; Osorio Reference Osorio2014). This unique ecological setting is enhanced by a diverse coastal and pelagic marine biota, shaped by the Pacific Ocean’s General Circulation System (Ortlieb et al. Reference Ortlieb, Vargas and Saliège2011).
The earliest human occupation in Arica is found located at the Acha ravine mouth, where a camp and human remains with natural mummification date back to 10,762–9480 calibrated BP (2σ) (Muñoz, Arriaza and Aufderheide Reference Muñoz, Arriaza and Aufderheide1993). The first crude evidence of artificial mummification by Chinchorro fishermen appears in Camarones-14 site, approximately 10 km south of Arica, dating 8038−7578 cal BP (2σ) (Schiappacasse and Niemeyer Reference Schiappacasse and Niemeyer1984). Morticians employed marsh plants (reed or Scirpus sp. and totora or Thypa sp.), wood, clay, and pigments (manganese and iron oxide) to prepare the bodies for the afterlife, developing increasingly elaborate mummification techniques over time (Arriaza and Standen Reference Arriaza and Standen2016; Arriaza et al. Reference Arriaza, Doubrava, Standen and Haas2005, Reference Arriaza, Figueroa, Ogalde, Standen, Halcrow, Amarasiriwardena, Van Hoesen and Castelleti2023). However, from the fourth to fifth millennium BP, sites along the Arica coast including Morro 1, Playa Chinchorro, Maderas Enco and Camarones 14, among others, show a proliferation of probably coexisting mummification styles, such as the Red Mummies and Bandaged Mummies (5582−5213 to 4830−4287 cal BP, 2σ) the Mummies covered with a layer of mud (5480−4959 cal BP, 2σ), natural mummification (4655−4246 to 3458−2991 cal BP, 2σ), and Black Mummies (8038–7578 to 6313−4837 cal BP, 2σ) (Arriaza Reference Arriaza2003; Arriaza and Standen Reference Arriaza and Standen2016; Silva-Pinto et al. Reference Silva-Pinto, Cáceres, Tonko-Huenucoy and Carpary2021) (Figure 1, Table 1).
AMS dating of burial sites and recent sampling in the northeast slope of the Morro de Arica
* ∆R discarded due to disturbance.
The Chinchorro population would also have articulated domestic spaces, which were subsequently used by the Arica culture. Thus, the domestic and funerary areas are superimposed, with ceramic evidence from post-Chinchorro periods, where the use of acacia and tamarugo for firewood becomes significant (Dauelsberg Reference Dauelsberg1985; Focacci and Chacón Reference Focacci and Chacón1989; García and Belmonte Reference García, Belmonte, Arriaza and Standen2016; Santoro et al. Reference Santoro, Rivadeneira, Latorre, Rotthammer and Standen2012). For example, the northwest slope of the Morro de Arica features domestic deposits associated with burials at the sites Playa Miller 8 (PLM 8) and Quiani 1 and 9. Notably, the latter two have produced dates ranging from 7505−6553 to 6753−6176 cal BP (2σ) and 8382−6178 to 7021−5030 cal BP (2σ), respectively (Bird Reference Bird1967; Dauelsberg Reference Dauelsberg1974; Muñoz and Chacama Reference Muñoz and Chacama1982).
The above highlights the necessity to refine the radiocarbon dates sequence for each occupational phase and sites (Arriaza and Standen Reference Arriaza and Standen2016; Castelleti et al. Reference Castelleti, Arriaza, Solís, Rodríguez Ceja, Silva, de Ugarte, Cisternas and Vega2024; Silva-Pinto et al. Reference Silva-Pinto, Cáceres, Tonko-Huenucoy and Carpary2021), particularly considering the highly disturbed cultural deposit at Morro de Arica due to recent human occupation (Castelleti and Arriaza Reference Castelleti and Arriaza2021) and the need to evaluate the environmental impact on archaeological site conservation.
Nevertheless, recent studies conducted across diverse planetary bioclimates have emphasized the necessity to refine human occupation chronologies based on radiocarbon dates of marine (considering the reservoir effect) and terrestrial origin. This refinement addresses issues related to dating long-lived woods (Carré et al. Reference Carré, Jackson, Maldonado, Chase and Sachs2016; Kennet et al. Reference Kennet, Ingram, Southon and Wise2002; Kim et al. Reference Kim, Wright, Hwang, Kim and Oh2019; among others). However, such studies are scarce for the Chilean arid coast (Latorre et al. Reference Latorre, de Pol-Holz, Carter and Santoro2017; Merino Reference Merino2017; Ortlieb et al. Reference Ortlieb, Vargas and Saliège2011).
Material and methods
In this study, we aim to provide new insights into the radiocarbon dating of the Chinchorro culture and subsequent periods. This research enables us to assess the existing database and stratigraphic features, considering of impact of the reservoir effect on marine samples and wood types in terrestrial samples. This work was conducted as part of cultural management efforts, specifically the enhancements to the perimeter fencing of the Chinchorro reserves “Estanques” and “Reserva 2” in the Morro de Arica´s northeastern slope, carried out by the Universidad de Tarapacá between 2019 and 2020.Footnote 1
Twenty-five pit tests were excavated using artificial levels of 10 cm thickness (Figure 1b), facilitating the creation of a detailed stratigraphic record of strata and interfaces. The excavated soil was sifted through 3–4 mm sieves to ensure through recovery of cultural remains. All the recovered cultural material was meticulously labeled and curated at the Centro de Gestión Chinchorro, Universidad de Tarapacá.
Charcoal samples and malacological remains, including food debris, were found in numerous excavated levels. We selected marine samples of bivalves and terrestrial samples of woody species from two test pits near the Morro 1 and Morro 1/6 funerary areas for radiocarbon dating. This selection was based on observed diameter categories of the samples, aiming to evaluate the master stratigraphic sequence.Footnote 2 To compare and assess significance differences, we utilized delta means reservoirs for terrestrial and marine samples per stratigraphic level and woody remains of varying diameters.Footnote 3
A Harris matrix Composer diagram was created to assess the depositional context of the sampling (Harris Reference Harris1991). Charcoal samples were sent to Beta Analytic, while shell samples were sent to the Accelerator Mass Spectrometry Laboratory at the UNAM Physics Institute in Mexico City for AMS dating. All published and newly obtained Chinchorro radiocarbon measurements from the northeastern slope of the Morro (this study) were calibrated and analyzed using multiple plots, employing the currently recommended calibration curves for terrestrial (charcoal) and marine (malacology) samples (Heaton et al. Reference Heaton, Kölher, Butzin, Bard, Reimer, Austin, Bronk Ramsey, Grootes, Hughen, Kromer, Reimer, Adkins, Burke, Cook, Olsen and Skinner2020; Hogg et al. Reference Hogg, Heaton, Hua, Palmer, Turney, Southon, Bayliss, Blackwell, Boswijk, Bronk Ramsey, Pearson, Petchey, Reimer, Reimer and Wacker2020). Marine-origin samples were calibrated using the Marine20 curve, while atmospheric-origin samples employed the SHCal20 curve, utilizing Calib 8.2 and Oxcal 4.4 (Heaton et al. Reference Heaton, Kölher, Butzin, Bard, Reimer, Austin, Bronk Ramsey, Grootes, Hughen, Kromer, Reimer, Adkins, Burke, Cook, Olsen and Skinner2020; Hogg et al. Reference Hogg, Heaton, Hua, Palmer, Turney, Southon, Bayliss, Blackwell, Boswijk, Bronk Ramsey, Pearson, Petchey, Reimer, Reimer and Wacker2020). Oxcal 4.4 facilitated a summation of dates to identify significant temporary groupings between funerary and domestic areas, and samples of different origins, assessing their correlation. To further test this correlation, a linear correlation analysis was conducted in Excel between charcoal and shell dates.
To compare the radiocarbon dates, we determined the marine reservoir effect index (R) and regional delta marine reservoir effect (∆R) for samples by level/excavation layer following Latorre et al. (Reference Latorre, de Pol-Holz, Carter and Santoro2017) and Merino (Reference Merino2017). We applied the conversion method outlined on the calib.org/deltar/ page (Reimer and Reimer Reference Reimer and Reimer2017), considering the overlap interval between marine and atmospheric sample density probabilities (Table 1).
Finally, we discuss absolute dating, reservoir effect, wood species identification, anomalous cases, and stratigraphy findings from Morro de Arica’s northeastern slope, contrasting them with information from other Peruvian and Chilean areas to illuminate chronological relationships (Carré et al. Reference Carré, Jackson, Maldonado, Chase and Sachs2016; Kennet et al. Reference Kennet, Ingram, Southon and Wise2002; Ortlieb et al. Reference Ortlieb, Vargas and Saliège2011).
Stratigraphy and absolute dating in the Morro de Arica northeastern slope
The stratigraphic record of the probed area reveals successive layers of sandy, semi-compacted, and scarcely organic sediments, associated with landslides and wind action on the Morro de Arica. The Harris matrix facilitates period grouping based on sedimentology and cultural materiality, dated through interfaces and chronological milestones. Notably, it highlights a recent component intruding into the basal layers (Figure 2).
(a) Stratigraphic sequence for Chinchorro site “Estanques” Faldeos del Morro (calibrated dates with SHCal 20 and Marine20 curves), and (b) Harris Matrix, recorded on the northeastern slope of Morro de Arica (arrows denote chrono-stratigraphic relationship).
The obtained stratigraphic sequence reveals a preceramic occupation span of 3715−7026 cal BP (2σ) comprising two successive domestic phases (Figure 2). Phase 1 dates to 6618−7026 cal BP while Phase 2 spans 5933−6971 cal BP, coinciding with mud-covered mummies. A third preceramic phase corresponds to the classic artificial mummification period, 3715−5933 cal BP (2σ), matching the original stratigraphic deposit (levels 3−6 in test pit C8 or layer A2), characterized by sandy, redeposited sediments, and abundant cultural remains.
The radiocarbon chronology correlated with the recorded stratigraphy suggests a significant alteration of the original sequence of the three preceramic phases, likely due to soil block slippage on the slope. Human intervention increases in these deposits from the Late Intermediate pottery period (ca. 750−500 BP) onward, especially in recent times. Pottery (one instance) and various contemporaneous materials (e.g., metals, bottle caps) appear up to level 7 in pit C8 and in disturbed burial contexts in pits C6 and C12, likely associated with Morro 1 and Morro 1/6 (Allison et al. Reference Allison, Arriaza, Standen, Focacci, Rivera and Löwenstein1984; Focacci and Chacón Reference Focacci and Chacón1989). These stratigraphic phases will be evidenced in the deposit as interfaces of extraction or alteration from the original order. Ancient and historical disturbance contributed to depositing new sediments around looted burial areas at various times, severely altering the original archaic shell deposits and burial grounds, likely exacerbated by trampling.
The absolute dating for the Morro de Arica
A comparison of the AMS dating results from our recent study with radiocarbon dates from burial sites on the northeast slope of the Morro de Arica is presented in Table 1.
The delta reservoir effect (∆R) was calculated from samples collected during recent excavations (Table 1, Figure 3), yielding an average of 84 years (range of <2 ± 117 to 176 ± 113 years). Stratigraphic disturbances are responsible for anomalous cases observed at surface levels.
(a) Multiple plot (Oxcal v4.4) with dates for domestic (Lema/Beta) and funerary sites (others laboratories) at Morro de Arica northeastern slope; (b) terrestrial samples (calibrated SHCal 20 curve; (c) marine samples (calibrated Marine20 curve).
Analysis of reservoir effect via comparison of terrestrial and marine origin samples across excavation level reveals an upward increase and downward stability, potentially resulting from recent fills and alterations (Table 1, Figure 4a). Charcoal size categories support this pattern, exhibiting increased the differentiated recording pattern of R and ∆R at surface levels, which is corroborated medium and large diameters at deeper levels.
(a) Radiocarbon dating summation comparison between charcoal and shell samples (Test pits, terrestrial samples calibrated with SHCal 20 curve and marine samples with Marine20); (b) linear correlation between shell versus charcoal sample dating, considering concordance by excavation level.
Figures 4a and 4b show a significant correlation between terrestrial and marine samples from Morro de Arica’s Chinchorro reserves, with differences explained by the malacological reservoir effect. Other indices show minimal variations. Radiocarbon dates and wood samples reveal negligible differences between terrestrial and marine samples (Table 2).
Discussion
The radiocarbon dates enable us to discuss two key aspects of the Chinchorro reserves’ archaeological sites on Morro de Arica’s northeastern slope: stratigraphy and comparative chronology of domestic deposits linked to funerary occupation. Furthermore, test pit excavations provided valuable insights into various factors shaping the area’s cultural deposits, spanning Chinchorro phases, ceramic periods, and historical to contemporary components.
Initial findings suggest the enclave on the Arica’s northeastern slope was heavily used for domestic purposes around ∼7090−6675 cal BP, likely by early coastal communities (Figure 3). However, the area will only begin to be used for funerary purpose from ca. 6200 cal BP, with a registered first phase of black mummies until ∼5800 cal BP, and then a second component between ∼5300−3700 cal BP in which the red mummies and bodies covered with a layer of mud/sand were preponderant. By the end of this period we see the presence of bodies with natural mummification (Arriaza et al. Reference Arriaza, Doubrava, Standen and Haas2005; Focacci and Chacón Reference Focacci and Chacón1989). There appears to be reduced occupation intensity at the site (domestic and funerary nucleus) towards ∼3926−3715 cal BP, despite previous studies identifying Chinchorro mummies within this date range (Arriaza et al. Reference Arriaza, Doubrava, Standen and Haas2005). Additionally, no dates were obtained for the 3000−1000 cal BP period, marking the disappearance of artificial mummification in Arica.
Nevertheless, a pre-Hispanic pottery occupation (ca. 800−600 cal BP) and recent historic and contemporary activities have altered the Archaic stratigraphy. Looting burials, soil extraction, and redepositing, combined with natural sand accumulation due to Morro de Arica’s steep slope, have significantly modified the archaeological context (Arriaza and Standen Reference Arriaza and Standen2016; Bird Reference Bird1967; Castelleti et al. Reference Castelleti, Arriaza, Solís, Rodríguez Ceja, Silva, de Ugarte, Cisternas and Vega2024). This shows that the northeastern slope of Morro was utilized for domestic purposes nearly a thousand years before the first burials at Morro 1 and 1700 years before Morro 1/6, as evidenced by shells and charcoal dating. Comparing samples of terrestrial and marine origin supporting the dating of funerary and domestic areas (Table 2, Figure 4), reveals two key findings: a higher date average of marine samples over terrestrial samples in the domestic area (suggesting the incidence of the reservoir effect), and a higher date average of terrestrial samples over marine samples among burial dates, which could be attributed to the use of old woodFootnote 4 or inherent sampling issues. Independently, there is a generally good correlation (R2= 0.77) between shell and charcoal samples dated and coming from similar strata.
However, differences in radiocarbon dating and ∆R between shell and charcoal at surface levels of the domestic deposit, and their stabilization towards lower levels, corroborate the occurrence of stratigraphic disturbances, and/or suggest the possibility of catastrophic events at the regional level (Latorre et al. Reference Latorre, de Pol-Holz, Carter and Santoro2017). Furthermore, the retained values exhibit similarity with those obtained in previous investigations comparing marine and terrestrial samples along the arid and semiarid Andean coastline (Carré et al. Reference Carré, Jackson, Maldonado, Chase and Sachs2016; Latorre et al. Reference Latorre, de Pol-Holz, Carter and Santoro2017; Merino Reference Merino2017; Merino et al. Reference Merino-Campos, de Pol-Holz, Southon, Latorre and Collado-Fabri2019; Ortlieb et al. Reference Ortlieb, Vargas and Saliège2011).
Conclusions
A level-by-level comparison of radiocarbon dates from the Morro de Arica excavation refined the Chinchorro culture’s occupational sequence. This study also made it possible to determine the Arica’s delta marine reservoir effect, averaging 84 years (range <2±117 to 176±113 years), confirmed by older radiocarbon dates from marine-origin samples in Morro domestic deposits. Additionally, terrestrial-origin samples dates from Chinchorro mummies suggest potential old wood contamination or ancient reuse.
Finally, comparing marine and terrestrial radiocarbon dates allowed evaluation of stratigraphic alterations via delta reservoir effect, assessing pottery fill’s and recent periods’ impact on El Morro’s absolute chronology and confirming the in-situ character of pre-mummification domestic occupation. Hopefully, these findings will pave the way for future research to explore the local reservoir effect’s implications for archaeological studies and its potential as a stratigraphic alterations indicator.
Acknowledgments
We are grateful to the projects UTA Mayor N° 3757-22 “Sondeo y caracterización del sitio arqueológico Villa Frontera, Arica, Región de Arica y Parinacota” and Fondecyt N°1210036. And the following individuals and organizations: Drs. Corina Solís, María Rodríguez Ceja, Quan Hua, and Maikel Díaz-Castro, LEMA Laboratory (UNAM Physics Institute), Laboratory of Bioanthropology (Instituto de Alta Investigación-UTA), UTA students and staff, the Morro de Arica neighborhood council, Consejo de Monumentos Nacionales and the anonymous reviewers of this paper.
