Abstract
The Crimean Peninsula contains numerous important Palaeolithic sites. It has been considered a potential Neanderthal refugium in Eurasia prior to their disappearance or assimilation, and subsequent replacement by Homo sapiens. This understanding is primarily drawn from lithic technological analysis and radiocarbon dating. However, recent developments in the latter suggest that many of the previously obtained radiocarbon dates from Crimean sites may be too young. The chronological sequence from the site of Kabazi II is a case in point, where differences between dates of bulk collagen and those of single amino acids range up to 10,000 radiocarbon years. These discrepancies are argued to arise from sample contamination and its lack of removal in bulk collagen dating. In this paper, we undertake a critical re-evaluation of results from seven Palaeolithic archaeological sites in the Crimean Peninsula: Kabazi II, Siuren I, Buran-Kaya III, Zaskalnaya V and Zaskalnaya VI, Kabazi V and Starosel’e. Our analysis of radiocarbon results from various sites supports the view that radiocarbon dates obtained so far tend to be too young, which has been indicated in the stratigraphy and lithic assemblages. By harnessing robust radiocarbon protocols in future chronometric work, such as the extraction of hydroxyproline, ninhydrin derivatisation or the purification of amino acids using XAD-2 resin, there is potential for constructing accurate chronologies. Furthermore, improved dating accuracy holds the promise of significant additional insights into the prehistoric sequence of the Crimean Peninsula and its potential role as a refugium for Neanderthals prior to their disappearance.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
The Crimean Peninsula is a region rich in preserved stratified Middle and Upper Palaeolithic sites. Within the Middle Palaeolithic, two distinct lithic industries have been revealed. The Crimean Micoquian (CM) appears to be present across approximately 100,000 years (Chabai et al., 2004). This technocomplex is primarily characterised by soft hammer bifacial thinning and retouching to produce bifacial tools, amongst which Keilmessers are the most indicative tool forms; unifacial tools such as side scrapers are often made on bifacial thinning flakes (Chabai, 2004). The Western Crimean Mousterian (WCM) appears later and is thought to have been present over 50,000 years until the end of the CM industry. The WCM is characterised by an absence of bifacial tools and presence of tools made on flakes and from Levallois reduction processes and, in its final phase, volumetric core reduction (Marks & Chabai, 2006). The Upper Palaeolithic is represented by two lithic industries that appear, based on the current evidence, to have coexisted partially alongside CM and WCM (Chabai, 2005). The first of these is the Aurignacian, a technocomplex characterised by pronounced carinated technology, finely retouched twisted and non-twisted bladelets and blades as well as micro-blades and bone tools (Demidenko & Otte, 2001). The second Upper Palaeolithic technocomplex, the Eastern Szeletian, is notable for bifacial leaf points, end-scrapers on flakes, geometric trapezoidal microliths and bone tools (Laroulandie & d’Errico, 2004). This industry also exhibits radial core reduction practices, albeit with blades being a relatively infrequent component (Monigal, 2004). The CM has conventionally been attributed to Neanderthals due to the presence of Neanderthal remains in Crimean sites such as Kiik Koba (Demidenko & Uthmeier, 2013), Zaskalnaya V (Chabai & Uthmeier, 2017) and Zaskalnaya VI (Stepanchuk et al., 2017). Based on a human tooth from Siuren I, the Aurignacian in the Crimea can be tentatively linked with what we call here derived Homo sapiens,Footnote 1 as indeed it is across Europe (Benazzi et al., 2015; Slimak et al., 2022). The association between the WCM and one of these human species is unknown due to the lack of human remains.
Refuge
In general, the period spanning approximately 47,000–42,000 cal BP marks a pivotal juncture in the biocultural transition from the Middle to the Upper Palaeolithic in Eurasia (Davies et al., 2015; Higham et al., 2014; Hublin, 2015; Hublin et al., 2020). This temporal range aligns with the increasing presence of derived Homo sapiens and the waning of Neanderthals throughout Eurasia (Hublin, 2015). The Initial Upper Palaeolithic (IUP) assumes key importance within this transition period, ushering in significant changes in culture and technology increasingly linked with derived Homo sapiens (e.g. Mylopotamitaki et al., 2024). The IUP itself was then replaced by industries of the Upper Palaeolithic (Rybin et al., 2023). This can be seen in various sites across Europe, one notably being Bacho Kiro, Bulgaria. A lithic assemblage was found, which did not relate or contribute to later assemblages and populations (Hajdinjak et al., 2021).
The Crimea, like the southern Iberian Peninsula and Gibraltar, are regions with unusually recent dates for Middle Palaeolithic assemblages (Finlayson et al., 2008; Jennings et al., 2011). Soffer (1994) suggested that the Crimea was a technological and anthropological eastern European cul-de-sac or refugium (Soffer, 1994; Stepanchuk, 2006: Burke, 2006, see also Ramírez-Pedraza et al., 2020). A refugium describes a region/area which supports an isolated species or community during a period where unfavourable conditions prevail elsewhere (Birks, 2015; Jones, 2021). In classical biological terms, the area must also provide sufficient temperatures to live and reproduce, and shelter against strong winds, and should have no permafrost present (Birks & Willis, 2008). Presently, the term refugium (local environment to have escaped ecological change) or refuge (general state of safety) is used in various fields and can be defined by different functions, for example macro or micro refugia (Jones, 2021). Macro refugia refers to regional refugial scales, whilst micro refers to a microclimate.
The Crimea can be defined as a macro refugium because of its large size and the fact that it is a peninsula. The region was inhabited long before the Middle to Upper Palaeolithic transition. At Kabazi II, absolute dates and palaeoenvironmental evidence show that the lowermost archaeological horizons date to Marine Isotope Stage (MIS) 5. Similarly, at Starosel’e, CM artefacts are found in the lowermost occupation from level 4, dating to between 80,000 and 77,000 uncal BP (Hardy et al., 2001; McKinney, 1998; Rink et al., 1998). The earliest layer radiocarbon dated at Starosel’e has been dated to ~ 42,500 14C years BP (without ultrafiltration) (Marks et al., 1997). This date of course may be too young, given the common underestimates associated with non-ultrafiltered dates. The age of the earliest occupations of Kabazi II fall into a phase of rising sea levels which made the Crimea an island (Marks & Chabai, 2006). This suggests the presence of Neanderthals, pre-dating MIS 5, during the Saalian complex, when it was still possible to reach continental land from the Crimea. On a longer chronological perspective, the Crimea might have experienced refugial phases with human presence cut off from the mainland during interglacial periods on the one hand, and refugial phases with connectivity to the larger northern Black Sea region during glacial periods on the other. The transitional palaeoenvironment during the Middle to Upper Palaeolithic shift was characterised by colder and more arid conditions. Environmental studies based on small mammal fauna, malacofauna, pollen, pedology and large mammal fauna all indicate that the conditions during MIS 3 during both stadials and interstadials were moderate rather than severe (Markova, 2012; Ramírez-Pedraza et al., 2020).
While the idea of the Crimean Peninsula being a refugium for late surviving Neanderthals holds some merit, doubts have begun to surface regarding the chronological evidence used in its support. At the Crimean Palaeolithic site of Kabazi II, for example, AMS radiocarbon dates ranging from 40,000 to 35,000 cal BP for WCM contexts were taken as evidence for a Neanderthal presence later than the disappearance of Neanderthals elsewhere in Europe, as shown by Higham et al. (2014). Recent redating of the bones that provided these initial results suggest that they likely date to > 50,000 cal BP, much older than was first assumed (Spindler et al., 2021), casting doubt on the notion of an unusually late Neanderthal presence in the Crimea.
Other European regions, such as Iberia, Italy, Belgium and the Balkans, have also been posited as potential refugia due to their geography, which might at times have offered benefit to human groups faces with progressively unfavourable climatic conditions (Bicho & Carvalho, 2022). Southern Iberia has been suggested as an optimal location for Neanderthal populations seeking to evade the more severe glacial conditions further in the north, much like a potential Crimean scenario (Chabai, 2005). Pleistocene Iberia boasted a rich biodiversity and sustained a persistence of rainfall (Jennings et al., 2011). Among the notable sites postulated as Neanderthal refugia, Gorham’s Cave, Gibraltar, stands out (Finlayson et al., 2006). The cave features a Mousterian level, radiocarbon dated to between 33,000 and 23,000 cal BP (López-García et al., 2011), although the reliability of these dates has been criticised for potential contamination and uncertainties surrounding the dated charcoal remains within the stratigraphy (Zilhão & Pettitt, 2006; Higham et al., 2012). Southern Iberia is not favourable overall for the preservation of organic remains and biomolecules due to the high average temperatures. This has deleteriously affected the bone preservation at various sites. At the Palaeolithic sites of Jarama VI and Zafarraya, for example, only 27 out of 215 bones contained a nitrogen yield high enough to proceed with radiocarbon dating (Wood et al., 2013). The challenge of dating sites in southern Iberia makes it difficult to assess whether the region really was a refugium for late Neanderthals. Our view is that this remains to be demonstrated. By comparison, the Crimean Peninsula holds greater promise for reliable radiocarbon dating, due to the improved preservation of biomolecules, including collagen. At Buran Kaya III, for example, collagen yields as determined by %nitrogen was significantly better (Péan et al., 2013). However, it is important to note that collagen yield does vary between the Crimean archaeological sites. For this reason, the Crimea offers us the possibility of testing the hypothesis of whether the Neanderthal refugia hypothesis is robust or not.
Based on the interstratification of Middle and Upper Palaeolithic industries in the Buran-Kaya III sequence and the presence of Middle Palaeolithic toolkits in the two lowermost Aurignacian units of Siuren I, it seems likely there was a co-existence or periodic overlap between the two populations within the wider framework of the Middle to Upper Palaeolithic transition (Bataille, 2017; Chabai et al., 2004). This implies the co-existence of Middle and Upper Palaeolithic industries on the relatively small territory of the Second range of Crimean Mountains. The movement of people into the Crimean would, at this time, have been facilitated due to the decreased sea levels of the Black Sea during MIS 3 and in fact during the entire last Glacial cycle (Winguth et al., 2000). The Crimea was not a peninsula at this time but part of dry land, which stretched from the Lower Danube to the Northern Caucasus and thus enabled population movements from east-to-west (Chabai et al., 2004).
Radiocarbon Dating
In assessing the veracity of late Neanderthal refugia, chronology is crucial. Bone, like all other carbonaceous materials, is prone to contamination, which can affect the reliability of radiocarbon dates unless removed. Carbon contamination may derive from human/laboratory/conservation sources, but most commonly it comes from soil organic acids, such as humic and fulvic compounds which come from the decay and degradation of plant remains (Devièse et al., 2021). When the bone is of Palaeolithic age, particularly > 30,000 years ago, errors due to contamination can be significant because of the overwhelming influence of modern sources of carbon contaminants on dating samples which are necessarily low in remaining radiocarbon. Underestimated ages resulting in dating errors of thousands of years have been found from past dated sites across Eurasia (Higham, 2019). Up to 70% of previous results are thought to be affected and usually are too young (Higham, 2011).
It is important to review and describe the various protocols for radiocarbon dating that have been applied to bone in Crimean archaeological sites over the last few decades if we are to understand the potential for errors in the corpus of dates we have currently. There are two broad methodological approaches that have previously been applied to AMS dating of bone from the Crimea. The first comprises methods used to isolate bulk type 1 mammalian collagen; the second has been to date compound-specific single or purified hydrolysed amino acids, or carboxylic carbon. The vast majority of dates in the literature fall into the first category. This paper will discuss previous radiometric analysis, which has been undertaken on the Palaeolithic sites.
Collagen is the most prominent protein within bone, comprising ~ 20% by weight in modern material. Collagen molecules comprise three polypeptides of alpha chains, each with approximately 1000 amino acids per alpha chain (Higham et al., 2006). Collagen is bound within a mineral matrix (hydroxyapatite). Bulk collagen dating undergoes a multi-step extraction process, which involves sequential steps of decalcification, ABA (acid–base-acid) treatment, gelatinisation and freeze-drying (Brock & Higham, 2009). The ‘collagen’ extracted is known widely as ‘Longin’ collagen after the researcher who originally published the gelatinisation method (Longin, 1971). It is the most commonly applied method in dating bone and is essentially the precursor approach for all chemical pretreatment protocol discussed below. Data shows that, while this method can provide reliable radiocarbon dates, it is also prone to errors when bone has been contaminated with greater age, resulting in errors manifestly larger.
In the early 2000s, an ultrafiltration method was applied to purify bulk Longin collagen samples from Palaeolithic bone routinely dated at the Oxford Radiocarbon Accelerator Unit (Higham et al., 2006). This was based on work previously undertaken by Brown et al. (1988). Ultrafiltration is an additional step in bone pretreatment designed to separate low molecular weight (MW) components (often including organic contaminants and broken up degraded collagen) from the high MW fraction (> 30 kD), thereby isolating individual alpha chains of polypeptides which themselves weigh ~ 110 kD. This has been shown to improve the collagen quality as well as remove much of the contaminating substances that affect so many Palaeolithic samples (Higham et al., 2006). A limitation is that, in cases where the MW size of the contaminant exceeds the MW cut-off of the ultrafilter, the contamination will not be removed. This could be significant in specific cases where museum consolidants have been applied to bones or where there has been cross-linking of the collagen with organic ground acids derived for example via the Maillard reaction. This limitation aside, ultrafiltered collagen approaches are the most reliable methods for dating bulk collagen extracted from archaeological bones. Sometimes, however, more rigorous methods are required to obtain accurate dates. The most common involve hydrolysing the collagen to its constituent amino acids. Two main approaches are currently used. The first is using XAD-2 resin; the second, hydroxyproline.
The XAD-2 method revolves around the hydrolysis of collagen amino acids and their recovery using ion-exchange columns (Stafford et al., 1991). The hydrolysis step is important because it effectively breaks the strong bonds between the collagen and humates where they occur. The purification process occurs through an ion-exchange column comprising a styrene-divinyl-benzene resin, usually XAD-2. Contaminants such as humates and fulvic acids remain on the resin while the amino acids pass through, to be collected, deacidified and passed for AMS dating (Minami et al., 2016; Van der Sluis et al., 2023).
Single compound dating is a challenging but attractive approach. Rather than purifying the bulk collagen, the method focuses on extracting a single compound from the collagen amino acid pool. The most common target has been hydroxyproline (HYP) (Stafford et al., 1982; Gillespie et al., 1984). HYP serves as a valuable biomarker for mammalian collagen type 2, due to its rarity in nature, with its abundance constituting approximately ~ 10% of bone amino acids. The collagen is hydrolysed then the amino acids injected onto an HPLC where they are separated chromatographically and the HYP collected (Devièse et al., 2017). HYP provides contaminant-free measurements, ensuring reliable and accurate results, because only the HYP is being dated (Devièse et al., 2017; Marom et al., 2012). This has resulted in several high-profile cases where previously contaminated and very valuable archaeological specimens have been reliably dated for the first time (e.g. Becerra-Valdivia et al., 2018; Devièse et al., 2019, 2021; Marom et al., 2013). Some notable exceptions have been identified, for example when samples are contaminated with amino acids from modern bone glues (see Devièse et al., 2018; Prüfer et al., 2021). In these cases, the HYP method will not discriminate the ancient from the contaminating amino acids, and only the use of ancient DNA or palaeoproteomics will enable the identification of the source of this carbon. The low proportion of HYP also means that sample size requirements are generally much higher than for other methods, with extensive lab work. This, along with the added cost, means that this method is often used only to date severely contaminated bone, or bone which is from particularly important single specimens.
The use of ninhydrin derivatisation for isolating carboxylic carbon is an alternative method first described by Nelson (1991). This method involves a chemical reaction between carboxylic batches of proteinaceous molecules and the CO2 which is extracted from them (Tisnérat-Laborde et al., 2003). The protocol involves two treatments of ninhydrin, which removes amino acids which are weakly attached to the collagen. Acid hydrolysis and filtration is completed in between the two steps (Dumoulin et al., 2017). The method is attractive in that it is straightforward to use. It has been suggested, however, that it is not possible to exclude the inclusion of carbon derived from exogenous carboxylic groups (Hedges & Klinken, 1992).
Following these protocols, the process of dating follows the same general set of procedures, involving combustion, purification, graphitisation and usually measurement by accelerator mass spectrometry.
For each site, tables are included with previous radiometric dates, which were undertaken. The method for each date is included if this is known.
Location
The present-day Crimean Peninsula covers approximately 27,500 km2, bounded by the Black Sea to the west and south, and the Azov Sea to the east. The peninsula encompasses a predominantly steppe environment, interwoven with a mountainous terrain separated into three distinct ranges. The first comprises mountains over 1000 m in elevation. This topography offered not only elevated vantage points but also an abundance of water sources, rendering it probably a favourable and stable habitat prior to the onset of the last glacial maximum. The second range, situated more centrally within Crimea, hosts a wealth of rock shelters and a profusion of high-quality flint outcrops, distributed within an altitude range of 300–800 m above sea level (asl). The third consists of low hills with occasional rock shelters and flint outcrops (Ferring, 1998).
Materials—Archaeological Sites and Previous Radiometric Dates
This review will cover the key Middle and Early Upper Palaeolithic archaeological sites of the Crimean Peninsula, with a particular focus on the chronostratigraphy in the light of hypotheses regarding its roles as a possible refugia. Seven sites will be critically considered: Kabazi II, Siuren I, Buran-Kaya III, Zaskalnaya V, Zaskalnaya VI, Kabazi V and Starosel’e. These sites are all located within the middle mountain range of the Crimea (Fig. 1). In a broader context, the peninsula harbours an impressive array of more than 35 stratified Middle Palaeolithic sites and more than 10 Upper Palaeolithic sites (Chabai, 2005).
Map of the Crimean Peninsula, showing the position of Siuren I, Starosel’e, Kabazi II, Kabazi V, Buran-Kaya III, Zaskalnaya V and VI. The red to brown colours represent the first and highest mountain range, which reaches elevations over 1000 m, the lighter and less frequent red to orange colour indicates the second middle mountain range; due to low elevations, the lowermost third range in the north of the middle range is almost not visible in this map. Map made with Mapbox and Topographic map (2024)
Of the selected archaeological sites we review here, six are rock shelters, whilst Kabazi II is an open-air site (for radiometric date ranges, see Table 1). The rockshelters are situated within Eocene nummulite limestones, some of which exhibit evidence of rock fall as discerned in the stratigraphy. Kabazi V and Siuren I are good examples. These sites contain an exceptionally good preservation of the archaeological faunal remains (Bataille, 2012; Uthmeier & Chabai, 2018), favoured by high sedimentation rates (Chabai & Uthmeier, 2017). In contrast, open-air sites present unique challenges. These sites are susceptible to erosion or disturbances due to contemporary human activities, potentially leading to the destruction of valuable deposits. Furthermore, open-air sites can deleteriously affect the preservation of biomolecules. The downward leaching of soil organic acids can lead to problems with cross-contamination of bone and charcoal, which can make radiocarbon dating significantly more difficult (Spindler et al., 2021). Another challenge for radiocarbon dating is the often-low presence of intact collagen for dating, which may not be adequately preserved in the bone. Charcoal samples can serve as an efficient substitute if the collagen content is low. However, the efficacy of this approach hinges on the quality of charcoal preservation and the yield of viable samples, as well as the context of the material of course.
Currently, considering the absolute dates obtained so far, the sites under review here encompass a temporal range spanning from 28,000 cal BP to > 50,000 cal BP, and thus cover both late Middle and early to middle Upper Palaeolithic periods. Only Kabazi II has, in our view, undergone recent robust re-dating using the most up-to-date protocols: HYP dating. Because these dates were considerably older than expected, it is quite plausible that other sites might also contain dates which underestimate their ‘true’ age, because of unremoved contaminants, as there is a wide range of literature attesting to age offsets between bulk and compound-specific dating. Such offsets between dates with different sampling pretreatment have been reported from many other sites from this temporal range, some defined as ‘open-air’ contexts (Becerra-Valdivia et al., 2018; Bourrillon et al., 2018; Devièse et al., 2017, 2018, 2019; Dinnis et al., 2018; Kosintsev et al., 2019; Marom et al., 2013; McColl et al., 2018; Moreno-Mayar et al., 2018; Nalawade-Chavan et al., 2013; Reynolds et al., 2017; Sikora et al., 2017). It is critically important to employ refined protocols to ensure the accuracy of dating results across archaeological sites such as these.
In this review, we consider the individual sites and their chronologies based on our experience of AMS dating of bone, and we also build Bayesian age models to explore outliers and offsets. We built Bayesian models, where appropriate, from previously published radiocarbon determinations with OxCal 4.4.3 software (Bronk Ramsey, 2009a) and the INTCAL20 calibration curve (Reimer et al., 2020).
Kabazi II
Kabazi II, part of a larger group of sites including Kabazi V, is a multi-layered open-air archaeological site situated along the right bank of the Alma River, within a forest-steppe environment (Gerasimenko, 2005; Markova, 2005). It boasts one of the most extensive and important Palaeolithic sequences in Crimea, extending from the end of the Last Interglacial period, approximately 115,000 years ago (MIS 5d), to the Late Middle Palaeolithic period, less than 50,000 years ago. Kabazi II has been subject to comprehensive analysis and research, including recent radiocarbon dating, as mentioned above, contributing to a deeper understanding of its chronology. New radiocarbon dates confirm that the site dates to more than 50,000 years ago and was not, therefore, a site of very late Neanderthal presence, as originally suspected (Spindler et al., 2021).
The stratigraphy comprises seven stratigraphical units (Fig. 2, further subdivided into 55 in situ archaeological levels (Uthmeier & Chabai, 2018). In total, 20 levels at Kabazi II relate to the WCM and 35 to the CM. Unit II includes a WMC lithic assemblage, characterised by Levallois core reduction and absence of bifacial technology (Chabai, 2005).
Kabazi II Drawings of the western (a) and eastern (b) sections. Arabic numbers indicate lithological layers and Roman numerals indicate archaeological layers. From Uthmeier and Chabai (2018)
Faunal remains and lithics suggest that Kabazi II is an open-air kill and butchering site, most likely occupied by Neanderthals (Patou-Mathis, 1999, 2005, 2006). This inference is drawn from the analysis of the lithic assemblages and is further bolstered by Neanderthal burials at nearby sites with CM material such as Kiik-Koba at Zaskalnaya V and VI, and the absence of H. sapiens remains in the Crimea until the appearance of the Aurignacian.
Siuren I
Siuren I is located within the second range of the Crimean Mountains (Demidenko, 2014). It is a rock shelter at an elevation of 15 m above the Belbek River.
The site exhibits a stratigraphic sequence comprising four distinct undisturbed occupational archaeological levels of the Aurignacian (see Fig. 3), belonging to three broader archaeological units. The archaeological levels are separated by sterile rock fall sediments. In units F, G and H, Aurignacian assemblages have been uncovered with the presence of finely retouched non-geometric microlithic, e.g. Dufour bladelets, which are interpreted as being indicative of the Aurignacian and thereby probably linked with Homo sapiens. However, there is the intriguing possibility of an overlap with Neanderthals, as evidenced by a CM component in the Aurignacian assemblages of archaeological units G and H (Chabai, 2005; Demidenko & Otte, 2001).
Siuren 1. Stratigraphy of the site. Profiles II and IV. From Chabai (2005)
The Aurignacian units at Siuren I can be subdivided into two major phases: an earlier one, encompassing unit H and G, each with several archaeological levels, and a later one, consisting mainly of levels from unit G (Demidenko & Otte, 2001). The latter is characterised by retouched bladelets with a twisted length profile (Dufour bladelets, sub-type Roc du Combe), also found in Central Europe (Chabai, 2005). The previous calibrated radiocarbon date for unit F indicates that the Aurignacian was present around 31–35,000 cal BP, which would be very late within a European context, and might be seen perhaps as unlikely, since there is evidence for the Aurignacian being present around 43–35,000 years ago (Chu & Richter, 2020). The uncalibrated radiocarbon results are in Table 2. Units G and H can be assigned to an earlier phase of the European Aurignacian; like other Proto-Aurignacian assemblages, the main argument is the presence of retouched Dufour bladelets of the Dufour sub-type that are straight in length profile. In addition, and as already mentioned above, there are bifacial and unifacial thinning flakes, scrapers, and points typical of the CM technocomplex (Bataille, 2012; Demidenko, 2014). This Middle Palaeolithic component is often patinated, implying a longer exposure to weathering, contrasting with the Aurignacian artefacts, which are preserved with unpatinated, fresh surfaces. Around 10% of the assemblages observed in units G and H fit this pattern (Demidenko & Noiret, 2012). The presumed oldest grinding stone linked with Homo sapiens was also found within unit G, relating to the Aurignacian. This suggests food was being mechanically processed at the site (Lango et al., 2021).
Archaeological units G and H hold significance for understanding the humans that once inhabited Siuren I. The combination of Upper and Middle Palaeolithic artefacts, represented by Aurignacian and CM assemblages, within one and the same in situ levels raises the possibility of a co-existence of Upper and Middle Palaeolithic humans. This could be interpreted as the periodic presence of Neanderthal visits to the site during a period when derived Homo sapiens were already widely spread in Europe but not perhaps occupying the Crimean Peninsula as permanently (Demidenko & Chabai, 2012a, 2012b). Alternatively, it might indicate that there were more extensive interactions between Neanderthals and derived Homo sapiens groups (Bataille, 2012). This underlines the complexities inherent in archaeological interpretation and the need for clearer understanding of the potential mixing of lithic assemblages at the site itself. From a methodical point of view, this might be solved through refined dating techniques to determine whether ages are mixed or consistent, more extensive refitting studies, in-depth sediment micromorphology or sediment DNA analyses.
Buran-Kaya III
Buran-Kaya III sits within an overhanging rock shelter within a narrow valley in the second mountain range of the Crimean Peninsula (Uthmeier, 2024). The rock shelter is 5 m wide and 6 m deep, sitting 8 m above the current Burulcha river (Marks & Monigal, 2000). The rock shelter’s location encompasses diverse biomes, including steppe, forests and plateau, making it potentially an ideal habitat for hominins during the Palaeolithic period. This ecological diversity can be seen throughout the extensive stratigraphy of the site (Péan et al., 2013).
A total of 19 cultural layers have been identified at Buran-Kaya III. These layers were initially documented following the 2001 excavations, encompassing a time span ranging from the Medieval period to the Middle Palaeolithic. Subsequent excavations in 2009, as illustrated in Fig. 4, specifically focused on nine archaeological layers, which span from the Middle to the Upper Palaeolithic (Péan et al., 2013). Layers 5.2, 6.1 and 6.2 have been characterised as Gravettian, while layers 6.3, 6.4 and 6.5 are associated with the Aurignacian. Layer B has been identified as CM and Layer C as Eastern Szeletian (Péan et al., 2013; Prat et al., 2011). The available radiometric dates can be found in Table 3.
Stratigraphic sections of Buran-Kaya III and plan of the excavations. From Péan et al. (2013) with permission
One hundred and sixty-two Homo sapiens remains have been discovered within the Upper Palaeolithic layer 6.1, but no evidence for intentional burial has been identified. These human remains, as evaluated through dental wear analysis, have been determined to belong to juveniles or sub-adults (Prat et al., 2011). Isotopic analysis for palaeodiet using δ15N and δ13N has been undertaken on the human remains, which has revealed a likely high amount of mammoth and plant protein-based consumption, which indicates the environment was rich in game and nutritious plants (Drucker et al., 2017). Layer B/B1 (Demidenko, 2004; Uthmeier, 2004) shows evidence of bifacial tool production, with few cores but a rich assortment of bifacial and unifacial points attributed to the Kiik-Koba facies for the CM (Demidenko, 2004). Below this, in Layer C, an Eastern Szeletian assemblage, sometimes termed Streletskayan, has been identified, comprising bifacial foliate points, geometric trapezoidal microliths and endscrapers, as well as snapped bone tubes and handles (Monigal, 2004; Uthmeier, 2012). Residues found on these tools provide evidence of their hafted use in cutting animals and avian tissues, as well as plants (Chabai, 2005; Hardy et al., 2001). Level C is conventionally attributed to the Upper Palaeolithic (Monigal, 2004), so it is interesting that layer B above signals a markedly different industry. As in the case with Siuren I, this stratigraphic position raises intriguing questions about the possibility of a coexistence of Neanderthals and Upper Palaeolithic humans not only at the site but also on Crimean Peninsula as a whole. However, interpreting this coexistence is complex due to the nature of the Eastern Szeletian, which is described as a transitional industry (Adams, 2009; Uthmeier, 2012) or an Upper Palaeolithic technocomplex (Monigal, 2004) within the archaeological record. In any case, the lack of human remains does not allow for a secure identification of the makers of this industry.
Various worked faunal remains have been discovered, some of which were modified through bone tube manufacturing (Laroulandie & d’Errico, 2004). These modifications observed in Layer C involved the use of materials from a wolf, bird and hare. One of these remains exhibits signs of having been heated. Additionally, evidence of sawing and cutting on different bone specimens has been documented. Remains that show evidence of scraping marks evidence the use of hare (Laroulandie & d’Errico, 2004), which is highly unusual in the Crimean Middle Palaeolithic. Furthermore, it appears that humans employed a specific technology in the creation of lithic artefacts for working bone (Laroulandie & d’Errico, 2004). The archaeozoological analysis of the faunal material from Layer C suggests that not only humans, but also carnivorous animals, played a role in the formation of the faunal assemblage. This is again an exception in the Crimean Middle Palaeolithic and interpreted as indicating an ephemeral presence of humans at the site that was scavenged by carnivores afterwards. In the Aurignacian and Gravettian layers above level C, there are numerous artefacts crafted from bone, and ornaments were made from both marine and freshwater shells, mammal teeth and ivory. In the Gravettian layers, there is strong evidence of cultural ornamentation of objects, a trend that is evident not only at this site, but also across the Crimea and mainland Ukraine. This suggests the possibility of extensive social interactions and exchanges among Gravettian Homo sapiens (Yanevich, 2014).
Zaskalnaya V
The rock shelter site of Zaskalnaya V is situated within a limestone cliff at the lower level of the Krasnaya Balka. The rock shelter is 60 m above the Bijuk Karasu River and faces south towards the Krasnaya Balka Valley. The reconstructed size of Zaskalnaya V, based on rockfall excavated downslope the actual rock shelter, is similar in size to Siuren I: approximately 25 m in width, 8 m in depth, and 4 m in height. High-quality flint outcrops are near the site (Chabai & Uthmeier, 2017). Zaskalnaya V has one of the most intensively occupied sequences within the Crimean Middle Palaeolithic, comprising 23 lithological layers and 87 archaeological levels, which were uncovered during the most recent excavations conducted in 2012 and 2013 (Chabai & Uthmeier, 2017). The archaeological levels were sub-divided into different groups (termed A, B, C, D) based on taphonomy and archaeological features, and cross-correlated with excavation units of Y. Kolosov concluded in 1994 (Fig. 5).
Group A consists of archaeological levels with lithic artefacts, burnt material and faunal remains. Group B comprises levels with primarily burnt material, indicating the former presence of hearths, but significant erosion makes it challenging to draw confident interpretations about precise taphonomic processes. Group C levels are categorised by depressions within the sediment, which have been filled up with a combination of lithic artefacts, burnt material and faunal remnants. In some cases, depressions may have been used to construct a hearth or for the anthropogenic deposition of burnt remains from nearby fireplaces. However, most of the depressions and their filling are assumed to go back to natural site formation processes. Group D comprises water eroded sediments.
From top to bottom, the 87 archaeological levels of the most recent excavations were correlated to the eight ‘cultural layers’ defined by Y. Kolosov during his excavations, to enable comparison with older data, and labelled units I, Ia, II to VII. Across stratigraphical units I–V, a consistent pattern emerges regarding the dominant lithic tools, which are primarily scrapers. Notably, a high proportion of these scrapers are bifacially worked. Additionally, other tool types such as points, denticulates, burins and notches were present in all levels. The lithic artefacts all show characteristics of the CM, indicating a Neanderthal occupation. This is further supported by the discovery of a Neanderthal upper deciduous molar within cultural unit IV. The tooth is currently under further analysis to isolate aDNA (Chabai & Uthmeier, 2017). A fragment of a skull of a 5- to 6-year-old child was found within unit V and a metacarpal bone of a 25-year-old woman from units II–III (Danilova, 1979a, 1979b: Kolosov, 1983: Stepanchuk, 2022). Two facies of the CM have been recognised at the site, these being the Kiik Koba (units I and IV) and Ak Kaya facies (units II, IIA and III). Both can be described as a sub-type of lithic assemblages within the CM, demonstrating either small lithic artefacts with intense reworking in the Kiik Kova Facies, or comparably large, unreduced bifacial tools in the Ak Kaya facies (Chabai & Uthmeier, 2017; Majkić et al., 2018). The uncalibrated radiometric dates can be seen in Table 4.
Zaskalnaya VI
Zaskalnaya VI is a multi-layered rock shelter site, situated near to the Biyuk-Karasu river and facing south. The rock shelter is located 60 m above the river level, which is 10–12 m in width (Stepanchuk et al., 2017).
The stratigraphy unearthed by Y. Kolosov consists of at least seven archaeological cultural layers, with the lower levels being associated with MIS 5 and the upper layers (I–III) with MIS 3 (Stepanchuk et al., 2017). Also, in the light of the much higher stratigraphical resolution obtained from the re-excavation of Kolosov’s sequence at Zaskalnaya V described above, it is highly likely that each ‘cultural layer’ at Zaskalnaya VI also consists of several separate archaeological levels. The six cultural layers defined in Fig. 6. The radiometric dates can be seen in Table 5.
The presence of CM tools at the site provides strong evidence for Neanderthal occupation, which is strongly supported by Neanderthal remains discovered in cultural layer III and IIIa. In the latter, a triple dismembered child burial was found, indicating the possibility that Neanderthals may have practised complex burial rituals. The accumulation of features such as pits, hearths, bone fragments, charcoal and lithic artefacts across cultural layer III suggests an extended period of occupation at the site. This accumulation reflects various activities and behaviours of Neanderthals over time (Stepanchuk et al., 2017). There is also evidence that mammoth bones were brought to the camp, where they were knapped and possibly used as fuel (Fedorchenko, 2018). In total, 120 animal bones were found with splitting marks, caused by anthropogenic lithic abrasion. These abrasion marks include striking blows or negatives of previous chips (Fedorchenko, 2018).
Interestingly, the remains of a young dolphin were found, despite the coast being 50 km away from the site. The presence of marine mammal remains is taken as argument to suggest that Neanderthals were engaged in small-scale migrations and effectively exploited resources, including those from coastal areas (Stepanchuk et al., 2017). However, the lack of cut marks leaves open the possibility that scavenging birds brought the dolphin bones to the site. Traces of abraded ochre found on lithic artefacts may indicate the Neanderthals at the site were engaged in activities that may have had a symbolic or cultural significance (Stepanchuk, 2022).
Kabazi V
Kabazi V is a rock shelter situated within the second range of the Crimean Mountains and only a few hundred meters east of the open-air site of Kabazi II (Uthmeier et al., 2008). The location marks the transition into the higher mountainous terrain of the region. Within the Kabazi Mountain, it stands out as one of eight Middle Palaeolithic sites, and notably is the second deeply stratified site dating to the Upper Pleistocene period from this locale. Kabazi V is situated 120 m above the Alma River, facing south-east (Uthmeier & Chabai, 2018).
The geological genesis of the site is different from others, as the geological formation partly resulted from the effects of running water instead of being primarily caused by gelifraction of the nummulites limestone (Uthmeier & Chabai, 2018). Later, phases of rockfall and sedimentation processes relate to intense weathering of the rock shelter walls. In its final phase, the Kabazi V rock-shelter became an open-air site after the final collapse of the roof. These events, while potentially disruptive, have had a positive impact on the preservation of the site’s archaeology and stratigraphy (Housley et al., 2007).
The stratigraphic sequence is 8 m deep, containing 27 lithological layers and 40 archaeological levels (Uthmeier & Chabai, 2018).
Four major archaeological units can be seen in Fig. 7. Unit III/1B represents the phase during which the site became an open-air site, due to a potential rock fall. This would have changed the dynamics of the site entirely (Uthmeier & Chabai, 2018).
Regarding Pleistocene human use, the development of Kabazi V can be divided into three major stages. The first involves the major inflow of water, which hindered any human occupation. The second stage sees the first presence of human occupation, with rock fall, sedimentation and weathering of limestone. The third stage is characterised by sedimentation accompanied by different degrees of erosion within the rock shelter; in this phase, the site was used as a residential camp. Numerous hearths with lithic scatters were found, some of them being placed towards the back of the cave (Uthmeier et al., 2008; Uthmeier & Chabai, 2018; Uthmeier & Chabai, 2018).
In addition, artificial pits are present in greater number at Kabazi V dissociated with CM occupations. Such archaeological evidence points to extensive and diverse activities at the site and the surrounding area, including the exploitation of flint outcrops for tool production. The stratigraphy of the hearths suggest that the site was occupied multiple times during short-term occupations (Chabai & Veselsky, 2007). Radiocarbon dates from archaeological units III/1A and III/5-3B2 have suggested that the cave site was occupied at least 39,000 cal BP. Dates were obtained from a charcoal fragment and a charred bone sample (Housley et al., 2007). The charcoal sample was dated using the humic acid fraction, which is less reliable than dating the black carbon fraction. Charred bone is considered an unreliable sample type, but at Kabazi V there were few options available due to low collagen levels in bone.
Starosel’e
The site of Starosel’e is located within a steep canyon, slightly off the Churuksu river valley (Burke, 2000). At the time of its initial discovery, the site spanned a north–south distance of 40 m with a depth of approximately 1 m (Marks et al., 1997).
The initial excavation of the site was conducted in 1952 by Formosov (Hardy et al., 2001) and led to the discovery of a single occupation thought to correspond to the late Mousterian. Subsequently, between 1993 and 1995, the site was re-excavated by A. E. Marks and Y. E. Demidenko, revealing the presence of four distinct cultural archaeological levels (Fig. 8). These were assigned to the CM (levels 1–2 and level 4), and the Starosel’e level 3 industry. The combined AMS and ESR dates of the site range from around 100,000–40,000 uncal BP (Hardy et al., 2001; Marks et al., 1997, 1998). Level 1 has been dated to around 40,000 uncal BP (see Table 6), while level 2 has not been dated. Levels 3 and 4 are separated by a rock fall event, with level 4 representing a relatively brief occupation period, preserved by a fine clay sediment layer. This level is currently dated to around 100,000 uncal BP by radiometric dating methods (Burke, 2000; Marks et al., 1997, 1998).
East/West profile. From Marks et al. (1997)
Human remains, including those of a juvenile individual, were discovered at Starosel’e (Burke, 2000; Monigal et al., 1998). The juvenile remains, initially unearthed during the 1953 excavation, were considered to exhibit transitional features between Neanderthals and derived Homo sapiens in terms of morphology (Alexeyev, 1976), while others suggested that these characteristics were caused by hydrocephalus (Wolpoff, 1980; Spitery, 1980). The juvenile individual was estimated to be approximately 18–19 months old and was found in the southeastern quarter of the excavation. It was noted that the juvenile was discovered beneath level 1; however, it is important to mention that there are no available photographs or detailed field notes that can confirm its exact positioning of the juvenile (Marks et al., 1997). The re-excavations from 1993 to 1995 clearly showed that the so-called Starosel’e child was in fact of recent origin (18th century) and represents one out of three burial pits at the site that follow Muslim ritual practises (as demonstrated by the position of the bodies and face orientation (Monigal et al., 1998)).
The faunal assemblage at Starosel’e is dominated by Equus hydruntinus (horse), Cervus sp. (deer), various carnivores and several bird species (Burke, 1999; Hardy et al., 2001). The presence of anthropogenic cut marks on some of these remains provides strong evidence of butchering activities. The faunal remains of level 3 suggest that complete carcasses were processed at the site, rather than just specific parts. It is therefore assumed that the site served as a location for trapping animals, particularly horses, within the box canyon, possibly for mass killing. In level 2, to the contrary, faunal elements from the skulls and less meaty parts of the horse carcasses dominate. This may indicate scavenging, probably as the result of natural events like flash flooding, which may have created opportunities for scavenging (Burke, 2000). Such natural or human-induced accumulation of horse remains could potentially explain the scarcity of other taxa in level 4 (Burke, 2000).
There is evidence for a change in the technology of lithic artefacts, whereas levels 1, 2 and 4 belong to different facies of the CM, level 3 represents an industry otherwise unknown from the Crimea termed the Starosel'e level 3 industry. One fundamental difference is the presence of bifacial tools in the levels of the CM, and the lack thereof in level 3. Furthermore, the blank production of level 3 exhibits characteristics of the discoidal method. This raises intriguing questions about possible shifts in either the site function (within the CM) or in the groups that were using the site (Burke, 2000). Feather barbules from swans, geese and ducks found at the site were at least potentially used for enhancing the stability of projectile weapons (Hardy et al., 2001). With all caution, this is interpreted as a possible hint at innovative hunting practices connected to the utilisation of the site for hunting activities.
Methods
Bayesian Age Modelling
OxCal v. 4.4.3 (Bronk Ramsey, 2009a) and INTCAL20 calibration curve (Reimer et al., 2020) were used to calibrate previous radiocarbon results. INTCAL20 the limit of 55,000 cal BP, in contrast to previous curves, contains some important updates that influence the calibration and modelling of Palaeolithic sites such as these (Reimer et al., 2020). We used Bayesian chronometric modelling to analyse previous uncalibrated dates for the sites of Siuren I, Buran Kaya III, Zaskalnaya V, Zaskalnaya VI and Starosel’e. We built age models using established methods based on building a prior framework corresponding to the site stratigraphy, using Sequence and Phase commands. All stratigraphic models used the General outlier model which allows an assessment to be made of whether individual radiocarbon dates and combined dates were outliers within the constraints of the model (Bronk Ramsey, 2009b). We set the prior sample outlier probability to 5%. We also used Simple outlier models to assess the statistical reproducibility of paired dates using R_Combine. The CQL code used in the models is in the Supplementary Methods along with a more detailed description of the Bayesian modelling method.
We did not construct a Bayesian model for Kabazi II, as this has already been discussed extensively by Spindler et al. (2021), nor did we build a model for Kabazi V due to the lack of chronometric data available. We feel it will be more beneficial to create a model when more data is available. In the models, we use different colours to represent the different dated technological assemblages. Gravettian is green, Aurignacian is purple, Micoquian is blue and Szeletian is red. We use grey when this is not clear. Brackets under the distributions represent the 68.2 and 95.4% probability ranges.
Results and Discussion
The Bayesian models built from the various sites are shown in Figs. 9, 10, 11, 12 and 13.
A Bayesian model for the site of Siuren I. The model comprises dates primarily made at Oxford University and published in Demidenko and Otte (2001). The model comprises three archaeological units, purple representing the Aurignacian (unit F) and grey representing unit G and H, due to their mixed assemblages. Outliers can be seen within H and F. These are given in the form (O: posterior probability/prior probability) with the priors set at 0.05
A Bayesian model for the site of Buran Kaya III. The model comprises seven archaeological units, green representing Gravettian, purple Aurignacian blue Micoquian and red Szeletian. The General Outlier Model shows significant outliers in every unit except horizon 20 and layer 6.4
A Bayesian model for the site of Zaskalnaya V. The model comprises dates primarily made at Oxford University. The model comprises four archaeological units, blue representing CM. The ‘General Outlier Model’ shows significant outliers in units III and II
A Bayesian model for the site of Zaskalnaya VI. The model comprises five archaeological layers, blue representing Micoquian
A simple one phase Bayesian model for the site of Starosel’e. The model comprises dates measured at the University of Oxford and published in Marks et al. (1997). The model comprises one archaeological layer, representing the Micoquian
Buran Kaya III is the site with the most robust chronometric sequence in the Crimean Peninsula with several radiocarbon determinations obtained using the more reliable methodological approaches (Péan et al., 2013; Table 3 and Fig. 10). As mentioned above, previously it was suggested that Neanderthals of unit B might have inhabited the site as late as around 28.5 ka uncal BP (34.4 kacal BP). This was based on older radiocarbon pre-treatment protocols, which implies caution.
More recent dating work has shown that several dates produced in the 1990s were underestimates of the ‘true age’. Péan et al. (2013), for example, redated GrA-53942, a Saiga tatarica bone from layer 6.1. The original result of 29,640 ± 170 BP was not duplicated when the same bone was dated in Oxford using ultrafiltration. There, the result was 32,3200 ± 450 BP (OxA_25669). Another comparison can be made between OxA-6673 (28,840 ± 460 BP) and OxA-36921 (37,700 ± 900 BP), another sample which underwent ultrafiltration. The substantial age shifts between the results implies that the newer ultrafiltered results are to be preferred. They also cast some doubt on samples previously dated using a simple Longin collagen approach. This cannot be the pattern in all cases, however, because Péan et al. (2013) also showed that there was agreement between some Longin treated dates and others pretreated using Ninhydrin derivatisation methods, although the comparison was helped by very imprecise Longin dates sometimes. This is not surprising, since the reliability of bone determinations depend in specific instances on the level of contamination present. In cases where contamination is low to non-existent, less robust pretreatment methods will produce accurate results. The problem is that it is difficult, if not impossible, to determine which bulk collagen determinations are accurate without redating them.
We built two Bayesian models to test the sensitivity of the posterior results to including or not some of the determinations from the site that were obtained in the 1990s. Model 1 included some of the previous Oxford dates from the unit C Szeletian (OxA-6672) (Péan et al., 2013). We gave these a 40% prior outlier value due to the potential for underestimated ages (other dates from the same period have been shown to underestimate the true age). We left out OxA-13302, as it was originally found in a supposed Mesolithic layer (after Péan et al., 2013). OxA-4126 and OxA-4127 from layer 6.1 were not included, due to their dates being extremely young compared to the other radiocarbon results. The final model contained 8 outliers, roughly 30% of the total determinations (SOM). In model 2, we left these results out completely. The results for the main boundaries in both models were similar, which is not surprising given the high prior outlier probabilities we applied. We favour model 2, and this is shown in Fig. 10. The model has only 3 outliers, all downweighed by the outlier posterior percentage.
The model confirms the age of the Homo sapiens remains in the Gravettian layer 6.1 of Buran-Kaya III between 37 and 36.4 ka cal BP (Prat et al., 2011, 2018). For the Micoquian layer B, the model gives an age range from 42,810 to 40,870 cal BP (95.4% prob.), which compares closely to the ages obtained for terminal Neanderthal Mousterian archaeological sites in Europe (Higham et al., 2014). For the Eastern Szeletian layer C, Buran Kaya III, the date ranges between 43 and 39 cal BP (Bataille, 2017). These more recent dates suggest that Crimea may not have served as a refugium for Neanderthals much later than 41 cal BP. In sum, Buran Kaya III challenges the idea that Crimea was a late Neanderthal refuge. However, it is important to reiterate that these interpretations are contingent upon accurate dating, and it would still be important to re-date some of the samples of faunal remains previously analysed, with newer protocols, to establish if these dates are correct and provide more determinations for the Bayesian model. For the time being, we have the highest level of confidence in the Buran-Kaya III chronometric sequence. In Fig. 14, we summarise the date ranges for the key horizons in the model.
Date ranges for the Aurignacian and Micoquian levels at Buran-Kaya III
Suiren I presents a very different case. Layers F, G and H from the site have been previously dated by AMS (Table 2), but the results are extremely variable and young, compared with sequences like Buran-Kaya III. We suspect a combination of contamination, poor collagen preservation and post-depositional or taphonomic influences. Several dates were not included in our tentative Bayesian model (see Table 2), due to extremely young ages (e.g. Lv-2132). Previous results indicated that Neanderthals may have been present at the site as late as 35,000–30,000 cal BP, but there are rightfully considered much too young (Demidenko, 2014). Charcoal was also used for dating in some instances, but yielded dates that were much too young and therefore no additional material was dated (Demidenko & Noiret., 2012). The dating of the Aurignacian/CM assemblages in Siuren I present at around 35,000–12,000 cal BP is obviously erroneous, since the CM has only been evident until early MIS 3 in various sites in Europe (Andreeva et al., 2022; Kozłowski, 2014; Picin et al., 2020) and probably until around 40,000 cal BP in the Crimea (Spindler et al., 2021). However, caution is required due to some dates being based on charcoal, being only reliable with pre-treatment methods such as oxidation and stepped combustion (Brock & Higham, 2009; Wood et al., 2012). It can be estimated that the previous radiocarbon dates are around 5–10,000 years too young, based on the technological assemblages present in the archaeological units alone. Early iterations of the modelling with did show high outlier probabilities and the likelihood is that the dates are most likely incorrect or unreliable. Our tentative Bayesian model includes only AMS dates that we have a degree of confidence in. Interestingly, the results appear to produce a fairly consistent pattern within which all of the dates sit with a tight wind down centred on 34–33 ka cal BP. The assessment of these archaeological layers as being more aligned with the Proto-Aurignacian is obviously chronologically at odds with this picture, given the fact that this industry is much earlier in all other parts of western Eurasia. More work is needed. This should focus on a re-dating of samples from archaeological units F, G and H utilising the latest radiocarbon protocols. Given the complexities surrounding dating bone remains from these units, it is plausible that the dates shown in our model are potentially too recent. This assessment aligns with the observations made by Demidenko (2014), highlighting the necessity for a rigorous re-analysis to provide a more precise chronology and a deeper understanding of the site’s history.
Zaskalnaya V comprises archaeological units I, II, III and IV, which have previously been radiocarbon dated using bone and charcoal (see Table 4 for a complete list of ESR and uncalibrated dates). Unit I has two uncalibrated radiocarbon dates from a bulk collagen protocol, which range from 30,080 to 28,850 uncal BP. For unit II, an age range of 41,800–24,900 uncal BP (radiocarbon and ESR) was determined, and for unit III, a date range from 42,700 to 20,900 uncal BP was obtained (radiocarbon and ESR). For unit IV, an age of 47,000 to 46,000 uncal BP was calculated. Outliers can be seen within the Bayesian model for three charcoal samples within units III and II (Fig. 11). The charcoal dates should be assessed carefully. It is noteworthy that the lowermost levels of units V and VII, based on the pedogenesis of the sediments and pollen analysis, are proposed to date within MIS 5d to MIS5a (Chabai & Uthmeier, 2006; Chabai & Uthmeier, 2017). The ESR dates should also be assessed extremely carefully, due to the large differences compared with the radiocarbon dates. This is a significant time period within the Late Pleistocene and includes multiple climatic and environmental fluctuations, so refining the dating of these layers can help place the archaeological record within this broader context. However, to obtain dates from these levels, other protocols for radiocarbon dating must be applied.
The Bayesian model results for Zaskalnaya VI can be seen in Fig. 12, for units I–IV. The site of Zaskalnaya VI has been determined as Middle Palaeolithic, relating to Micoquian (Stepanchuk et al., 2017). This alone strongly indicates that the radiocarbon dates especially from the lowermost units are extremely young to be Micoquian, and thus unlikely the technological assemblages belonging to Neanderthals. Also, it is quite interesting that the dates within this unit spread across a wide range of around 6000 years. This strongly indicates that further work needs to be undertaken on the radiocarbon analysis for this site.
Bayesian modelling was not undertaken for Kabazi V, due to the possibility of the charcoal dates being affected by contamination due to their treatment method. Dating was also attempted on four bones from the site, but these samples proved to be challenging due to low collagen yields. All four bone samples yielded less than 1% collagen. Nevertheless, two of the samples produced dates of 38,780 to 30,980 uncal BP (Housley et al., 2007).
Finally, we turn to Starosel’e, which has only two radiocarbon dates (Fig. 13), both of which were treated using ultrafiltration. Starosel’e also has some ESR dates; however, we have decided to only concentrate on the analysis of radiocarbon dates within this paper. The results from the simple Bayesian model shows a date range of 49,000 to 43,000 cal BP. Further work on this important Palaeolithic is needed, due to its significance.
Bayesian modelling is of great importance in reliably interpreting the radiocarbon and other dating chronologies in archaeology. In the Crimea, the challenge is in deciphering which of the radiocarbon dates available from the Palaeolithic sites are reliable and can be faithfully incorporated into models. No Bayesian model can compensate for problematic dates, of which there are many in this instance, as this review can show. Only Buran Kaya III has a reliable chronometric sequence. Some other sites, such as Zaskalnaya VI, have had different protocols tested on bones and some that appear reliable, but more work is needed to compare different protocols and include them in statistical analyses and Bayesian modelling.
Great caution is therefore required in the interpretation of some of the corpus of results at face value. The application of improved methods of decontaminating bones, such as the hydroxyproline single amino acid dating (HYP), the use of XAD-2 resin approaches, or methods such as ninhydrin derivatisation dating of carboxylic carbon, offers a promising solution to obtain more accurate dates (Devièse et al., 2017; Higham, 2019). As this review has shown, many sites likely have dates that are not accurate and almost certainly too young. There are some limitations with HYP, especially in regard to large quantities of collagen needed and lab work. However, work is currently underway to test HYP results against XAD-2. This will help us understand further how the two methods compare and then, if results are encouraging, to apply the method more widely.
The ubiquitous problem of low collagen levels in bones from Crimean sites also has implications for reliability (as collagen yields decrease so the impact of constant low-level contamination in sites increases). In the future, the use of %nitrogen or NIR approaches as a means of screening bones for their collagen yield before proceeding with full AMS dating is a valuable approach (Jacob et al., 2018; Sponheimer et al., 2019) that can save considerable time and resources. Additionally, when collagen levels are insufficient for AMS dating, alternative materials like charcoal can be considered. Charcoal is a valuable resource for dating purposes and has been found at numerous Palaeolithic sites in Crimea, including Kabazi V and Zaskalnaya V. However, the application of oxidation and stepped combustion pre-treatment method should be undertaken to ensure a complete as possible contamination removal (Brock & Higham, 2009).
Questions of a seemingly intractable nature, concerning which human groups were present through the late Palaeolithic in the Crimea, may also be addressed through a wider application of zooarchaeology by mass spectrometry (ZooMS) for screening previously excavated faunal remains. This method enables the identification of unidentifiable bone fragments to species/genus level through collagen peptide fingerprinting. By comparing the spectra of unidentifiable bones to a comprehensive database of published peptide markers, it is possible to identify those that have the characteristic marker of the Hominidae (Richter et al., 2022). ZooMS can be used to screen thousands of bone fragments and has proved to be an immensely beneficial method for identifying the species of non-diagnostic fragmented bones. It proved to be pivotal in the discovery of the Denisovans F1 hybrid “Denny” at Denisova Cave, Russia, for example (Brown et al., 2021). As at Denisova, a high percentage of fragmented bones in the Crimea could not be assigned to species, so ZooMS could be useful. At Suiren I, for example, a staggering 88% of fragmented bones could not be assigned to a species (Massé & Patou-Mathis, 2009). This method opens exciting possibilities, as it may aid in the discovery of additional hominin bones, for Buran-Kaya III, or even potentially the first hominin bones, as would be the case at Suiren I. These advances in species identification techniques continue to enrich our understanding of the past and may lead to significant discoveries in the field of palaeoanthropology. Research on this subject is currently ongoing, and we hope to share the results later this year. Similarly, the wider application of sediment DNA methods (Gelabert et al., 2021) would also provide the possibility of identifying human presence in some of the key Crimean sites.
Conclusion
The Crimean Peninsula has emerged as a profoundly significant region for understanding the process of the biocultural shift associated with the Middle to Upper Palaeolithic transition. This area holds great importance, particularly over the question of when this transition occurred in the Crimea, the extent to which Neanderthals and derived Homo sapiens may have overlapped with one another, and whether Neanderthals may have used the Peninsula as a refugium or not. Recent research has demonstrated that there are challenges in obtaining accurate dates, such as the case with bulk collagen extracts from Kabazi II, which yielded dates that were deemed too young. This highlights the need for more advanced and precise dating methods to sharpen the chronological lens during this key period. New radiocarbon dates have the potential to significantly contribute to our understanding of the Middle and Upper Palaeolithic technocomplexes observed in the region and their chronological relationships. Only with a strong and reliable underpinning chronology will it be possible to obtain a deeper understanding of the relationships between various hominins through this period and whether there is weight in the refugium hypothesis of late Crimean Neanderthals.
Data Availability
No datasets were generated or analysed during the current study.
The CQ data in the supplementary material can be used to replicate the OxCal figures.
Notes
The precise terms used to refer to what were previously known as ‘anatomically modern humans’ or Homo sapiens are undergoing a great deal of debate currently. Stringer (pers. comm’d in Bermúdez de Castro and Martinón-Torres, 2022) suggested that ‘basal H. sapiens’ and ‘derived H. sapiens’ might be more preferred over the more traditional ‘archaic’ and ‘modern’. We follow this suggestion here for ‘modern’ humans.
References
Adams, B. (2009). The Bükk mountain Szeletian: Old and new views on “transitional” material from the eponymous site of the Szeletian. In M. Camps &, P. Chauhan (Eds.) Sourcebook of Palaeolithic transitions: Methods, theories and interpretations. Springer. https://doi.org/10.1007/978-0-387-76487-0_28
Alexeyev, V. P. (1976). Position of the Staroselye find in the hominid system. Journal of Human Evolution, 5, 413–421.
Andreeva, T. V., Manakhov, A. D., Gusev, F. E., Patrikeev, A. D., Golovanova, L. V., Doronichev, V. B., Shirobokov, I. G., & Rogaev, E. I. (2022). Genomic analysis of a novel Neanderthal from Mezmaiskaya Cave provides insights into the genetic relationships of Middle Palaeolithic populations. Science and Reports, 12, 13016. https://doi.org/10.1038/s41598-022-16164-9
Bataille, G. (2017). Neanderthals of Crimea – Creative generalists of the late Middle Paleolithic. Contextualizing the leaf point industry Buran-Kaya III Level C. Quaternary International, 435, 211–236. https://doi.org/10.1016/j.quaint.2015.12.071
Bataille, G. (2012). Flakes and blades. The role of flake production in the Aurignacian or Siuren I (Crimea, Ukraine). In A. Pastoors & M. Persani (Eds.), Flakes not blades: The role of flake production at the onset of the Upper Palaeolithic in Europe 5 (pp. 261–293). Wissenschaftliche Schriften des Neanderthal Museum.
Becerra-Valdivia, L., Waters, M. R., Stafford, T. W., Jr., Anzick, S. L., Comeskey, D., Devièse, T., & Higham, T. F. G. (2018). Reassessing the chronology of the archaeological site of Anzick. Proceedings of the National Academy of Sciences of the United States of America, 115(27), 7000–7003. https://doi.org/10.1073/pnas.1803624115
Benazzi, S., Slon, V., Talamo, S., Negrino, F., Peresani, M., Bailey, S. E., Sawyer, S., Panetta, D., Vicino, G., Starnini, E., Mannino, M. A., Salvadori, P. A., Meyer, M., Pääbo, S., & Hublin, J. J. (2015). The makers of the Protoaurignacian and implications for Neandertal extinction. Science, 348(6236), 793–796. https://doi.org/10.1126/science.aaa2773
Bicho, N., & Carvalho, M. (2022). Peninsular southern Europe refugia during the Middle Palaeolithic: An introduction. J Quaternary Sci, 37, 133–135. https://doi.org/10.1002/jqs.3410
Birks, H. J. B. (2015). Some reflections on the refugium concept and its terminology in historical biogeography, contemporary ecology and global-change biology. Biodiversity, 16(4), 196–212. https://doi.org/10.1080/14888386.2015.1117022
Birks, H. J. B., & Willis, K. J. (2008). Alpines, trees, and refugia in Europe. Plant Ecol and Divers, 1, 147–160.
Bourrillon, R., White, R., Tartar, E., Chiotti, L., Mensan, R., Clark, A., Castel, J. C., Cretin, C., Higham, T., Morala, A., Ranlett, S., Sisk, M., Devièse, T., & Comeskey, D. J. (2018). A new Aurignacian engraving from Abri Blanchard, France: Implications for understanding Aurignacian graphic expression in Western and Central Europe. Quart Int, 491, 46–64. https://doi.org/10.1016/j.quaint.2016.09.063
Brock, F., & Higham, T. F. G. (2009). AMS radiocarbon dating of Paleolithic-aged charcoal from Europe and the Mediterranean rim using ABOX-SC. Radiocarbon, 51, 839–846.
Bronk Ramsey, C. (2009a). Bayesian analysis of radiocarbon dates. Radiocarbon, 51(1), 337–360.
Bronk Ramsey, C. (2009b). Dealing with outliers and offsets in radiocarbon dating. Radiocarbon, 51(3), 1023–1045.
Brown, T. A., Nelson, D. E., Vogel, J. S., & Southon, J. R. (1988). Improved collagen extraction by modified Longin method. Radiocarbon, 30(2), 171–177.
Brown, S., Massilani, D., Kozlikin, M. B., Shunkov, M. V., Derevianko, A. P., Stoessel, A., Jope-Street, B., Meyer, M., Kelso, J., Pääbo, S., Higham, T., & Douka, K. (2021). The earliest Denisovans and their cultural adaptation. Nature Ecology & Evolution, 6, 28–35. https://doi.org/10.1038/s41559-021-01581-2
Burke, A. (2000). The view from Starosel’e: Faunal exploitation at a Middle Palaeolithic site in Western Crimea. International Journal of Osteoarchaeology, 10, 325–355.
Burke, A. (2006). Neandertal settlement patterns in Crimea: A landscape approach. Journal of Anthropological Archaeology, 25, 510–523.
Burke, A. (1999). Butchering and scavenging at the Middle Paleolithic site of Starosel’e. In V. P. Chabai & K. Monigal (Eds.), The Middle Paleolithic of western Crimea (Vol. 2, pp. 1–29). ERAUL 87.
Chabai, V. P., & Uthmeier, T. H. (2006). Settlement systems in the Crimean Middle Palaeolithic. In V. P. Chabai, J. Richter, & T. H. Uthmeier (Eds.), Palaeolithic sites of Crimea, Vol 2. Kabazi II: 70000 years since the Last Interglacial (pp. 297–359). Shlyakh, Simferopol & Cologne.
Chabai, V. P., & Veselsky, A. P. (2007). Kabazi V: Hearths & pits. In V. P. Chabai, J. Richter, & T. H. Uthmeier (Eds.), (pp. 27–49). Simferopol-Cologne.
Chabai, V. P., Marks, A. E., & Monigal, K. (2004). Crimea in the context of the eastern European Middle Paleolithic and early Upper Paleolithic. In V. Chabai, K. Monigal, & A. Marks (Eds.), The Middle Paleolithic and early Upper Paleolithic of Eastern Crimea (Vol. 3, pp. 419–460). ERAUL 104.
Chabai, V. P., & Uthmeier, T. H. (2017). New excavations at the Middle Paleolithic site Zaskalnaya V, Crimea: The 2012-2013 field seasons A Preliminary Report. Quartär, 64, 27–61.
Chabai, V. P. (2004) The Middle Palaeolithic of Crimea. Simferopol: Shlyakh. (In Russian).
Chabai, V. P. (2005). The late middle and early Upper Paleolithic in Crimea (Ukraine). In J. Zilhão, T. Aubry, & A. F. Carvalho (Eds.), Les premiers hommes modernes de la péninsule ibérique Vol 17. Actes du colloque de la commission VIII de l’UISPP, Vila nova de Foz Cóa (22–24 October 1998) (pp. 25–35). Trabalhos de Arqueologia, 17.
Chu, W., & Richter, J. (2020). Aurignacian cultural unit. In C. Smith (Ed.), Encyclopedia of global archaeology (pp. 1–10). Springer. https://doi.org/10.1007/978-3-030-30018-0_3441
Danilova, E., & I. (1979a). The occipital bone of Neanderthal from Zaskalnaya V trench, near Ak-Kaya. In Y. G. Kolosov (Ed.), Palaeolithic studies in Crimea (pp. 76–83). Naukova Dumka (In Russian.).
Danilova, E., & I. (1979b). The Neanderthal metacarpal fragment. In Y. G. Kolosov (Ed.), Palaeolithic studies in Crimea (pp. 84–85). Naukova Dumka (In Russian.).
Davies, W., White, D., Lewis, M., & Stringer, C. (2015). Evaluating the transitional mosaic: Frameworks of change from Neanderthals to Homo sapiens in eastern Europe. Quaternary Science Reviews, 118, 211–242. https://doi.org/10.1016/j.quascirev.2014.12.003
de Castro, J. B., & Martinón-Torres, M. (2022). The origin of the Homo sapiens lineage: When and where? Quaternary International, 634, 1–13.
Demidenko, Y. E., Chabai, V. P. (2012a). Unit H: Lithic Artifacts. In Y. E. Demidenko, M. Otte, P. Noiret (Eds.), Siuren I Rockshelter. From Late Middle and Early Upper Palaeolithic to Epi-paleolithic in Crimea. The Palaeolithic of Crimea, IV. ERAUL 129, Liège, (pp. 109–133).
Demidenko, Y. E., & Chabai, V. P. (2012b). Unit G: Lithic artifacts. In Y. E. Demidenko, M. Otte, & P. Noiret (Eds.), Siuren I Rockshelter: From Late Middle and Early Upper Paleolithic to Epi-paleolithic in Crimea, Volume IV (pp. 135–211). ERAUL 129, Université de Liège.
Demidenko, Y. E., & Noiret, P. (2012). The Siuren-I Aurignacian of Krems-Defour type industries in the context of the European Aurignacian. In Y. E. Demidenko, M. Otte, & P. Noiret (Eds.), Siuren I rock-shelter: From Late Middle paleolithic and early upper Paleolithic to Epi-paleolithic in Crimea, Volume IV (pp. 343–357). ERAUL 129.
Demidenko, Y. E., & Otte, D. (2001). Siuren-I (Crimea) In the context of a European Aurignacian. Préhistoire Européenne, 16–17, 133–146.
Demidenko, Y. E., & Uthmeier, T. H. (2013). kiik-Koba Grotto Crimea (Ukraine): Re-analysis of a key site of the Crimean Micoquian. Verlag marie leidorf GmbH.
Demidenko, Y. E. (2004). Buran-Kaya III layer B: The lithic assemblage. In V. P. Chabai, K. Monigal, & A. E. Marks (Eds.), The Palaeolithic of Crimea, The Middle Paleolithic and Early Upper Paleolithic of Eastern Crimea (Vol. 3, pp. 113–149). ERAUL 104.
Demidenko, Y. E. (2014). Siuren I Rockshelter: From the Late Middle Paleolithic and Early Upper Paleolithic to the Epipaleolithic in Crimea. In C. Smith. (Ed.), Encyclopaedia of Global Archaeology. Springer, United States, New York. https://doi.org/10.1007/978-1-4419-0465-2_1867
Devièse, T., Comeskey, D., McCullagh, J., Bronk Ramsey, C., & Higham, T. (2017). New protocol for compound-specific radiocarbon analysis of archaeological bones. Rapid Communications in Mass Spectrometry, 32, 373–379.
Devièse, T., Stafford, T. W., Waters, M. R., Wathen, C., Comeskey, D., Becerra-Valdivia, L., & Higham, T. F. G. (2018). Increasing accuracy for the radiocarbon dating of sites occupied by the first Americans. Quaternary Science Reviews, 198, 171–180.
Devièse, T., Massilani, D., Seonbok, Y. I., Comeskey, D., Nagel, S., Nickel, B., Ribechini, E., Lee, J., Tseveedorj, D., Gunchinsuren, B., Meyer, M., Pääbo, S., & Higham, T. F. G. (2019). Compound-specific radiocarbon dating and mitochondrial DNA analysis of the Pleistocene hominin from Salkhit Mongolia. Nature Communications, 10, 274.
Devièse, T., Abramsb, G., Hajdinjakd, M., Pirsone, S., Grootef, I. D., Modica, K. D., Toussainth, M., Fischeri, V., Comeskeya, D., Spindlera, L., Meyerd, M., Semalj, P., & Highama, T. (2021). Reevaluating the timing of Neanderthal disappearance in Northwest Europe. Proceedings of the National Academy of Sciences U S.A., 118, 12. https://doi.org/10.1073/pnas.2022466118
Dinnis, R., Bessudnov, A., Reynolds, N., Devièse, T., Pate, A., Sablin, M., Sinitsyn, A., & Higham, T. (2018). New data for the Early Upper Paleolithic of Kostenki (Russia). Journal of Human Evolution, 127, 21–40.
Drucker, D. G., Naito, Y. I., Péan, S., Prat, S., Crépin, L., Chikaraishi, Y., Ohkouchi, N., Puaud, S., Lázničková-Galetová, M., Patou-Mathis, M., Yanevich, A., & Bocherens, H. (2017). Isotopic analyses suggest mammoth and plant in the diet of the oldest anatomically modern humans from far southeast Europe. Scientific Reports, 7, 6833. https://doi.org/10.1038/s41598-017-07065-3
Dumoulin, J. P., Messager, C., Valladas, H., Beck, L., Caffy, I., Delqué-Količ, E., Moreau, C., & Lebon, M. (2017). Comparison of two bone-preparation methods for radiocarbon dating: Modified longin and ninhydrin. Radiocarbon, 59(6), 1835–1844. https://doi.org/10.1017/RDC.2017.132
Fedorchenko, O. (2018). Taphonomic ‘chronicles’: The case-study of Zaskalnaya VI site (layer III). Revista Arta, 14, 2.
Ferring, C. R. (1998). The geologic setting of Mousterian sites in Western Crimea. In A. E. Marks & V. P. Chabai (Eds.), The Middle Paleolithic of western Crimea (Vol. 1, No. 84, pp. 17–30). ERAUL.
Finlayson, C., Pacheco, F. G., Rodríguez-Vidal, J., Fa, D. A., López, J. M. G., Pérez, S., Finlayson, G., Allue, E., Preysler, J. B., Cáceres, I., Carrión, J. S., Jalvo, Y. F., Gleed-Owen, C. P., Espejo, F. J. J., López, P., Sáez, J. A. L., Antonia, J., Cantal, R., Marco, A. S., … Sakamoto, T. (2006). Late survival of Neanderthals at the Southernmost extreme of Europe. Nature, 443, 850–853. https://doi.org/10.1038/nature05195
Finlayson, C., Darren, A., Espejo, F. J., Carrión, J. S., Finlayson, G., Geraldine, F., Pacheco, F. G., Vidal, J. R., Stringer, C., & Ruiz, F. M. (2008). Gorham’s Cave, Gibraltar - the persistence of a Neanderthal population. Quaternary International, 181(1), 64–71.
Gelabert, P., Sawyer, S., Bergström, A., Margaryan, A., Collin, T. C., Meshveliani, T., Belfer-Cohen, A., Lordkipanidze, D., Jakeli, N., Matskevich, Z., Bar-Oz, G., Fernandes, D. M., Cheronet, O., Özdoğan, K. T., Oberreiter, V., Feeney, R. N. M., Stahlschmidt, M. C., Skoglund, P., & Pinhasi, R. (2021). Genome-scale sequencing and analysis of human, wolf, and bison DNA from 25,000-year-old sediment. Current Biology, 31(16), 3564-3574.e9. https://doi.org/10.1016/j.cub.2021.06.023
Gerasimenko, N. (2005). Vegetation evolution of the Kabazi II site. In V. Chabai, Richter. J, Uthmeier. T (Eds.), Kabazi II: Last Interglacial occupation, environments and subsistence, Chapter III. Simferopol-Cologne: Shlyakh, (pp. 25–49).
Gillespie, R., Hedges, R. E. M., & Wand, J. O. (1984). Radiocarbon dating of bone by accelerator mass spectrometry. Journal of Archaeological Science, 11(2), 165–170.
Hajdinjak, M., Mafessoni, F., Skov, L., Vernot, B., Hübner, A., Fu, Q., Essel, E., Nagel, S., Nickel, B., Richter, J., Moldovan, O. T., Constantin, S., Endarova, E., Zahariev, N., Spasov, R., Welker, F., Smith, G. M., Sinet-Mathiot, V., Paskulin, L., … Pääbo, S. (2021). Initial Upper Palaeolithic humans in Europe had recent Neanderthal ancestry. Nature, 592, 253–257. https://doi.org/10.1038/s41586-021-03335-3
Hardy, B. L., Kay, M., Marks, A. E., & Monigal, K. (2001). Stone tool function at the paleolithic sites of Starosel’e and Buran Kaya III, Crimea: Behavioural implications. PNAS, 98, 19. https://doi.org/10.1073/pnas.191384498
Hedges, R. E., & Van Klinken, G. J. (1992). A review of current approaches in the pretreatment of bone for radiocarbon dating by AMS. Radiocarbon, 34(3), 279–291.
Hedges, R. E. M., Housley, R. A., Pettitt, P. B., Bronk Ramsey, C., & Van Klinken, G. J. (1996). Radiocarbon dates from the Oxford AMS system. Archaeometry, 38(1), 181–207.
Higham, T. (2011). European Middle and Upper Palaeolithic radiocarbon dates are often older than they look: Problems with previous dates and some remedies. Antiquity, 85(327), 235–249. https://doi.org/10.1017/S0003598X00067570
Higham, T. (2019). Removing contaminants: A restatement of the value of isolating single compounds for AMS dating. Antiquity, 93(370), 1072–1075.
Higham, T., Jacobi, R. M., & Bronk Ramsey, C. (2006). AMS Radiocarbon dating of ancient bone using ultrafiltration. Radiocarbon, 48, 179–195.
Higham, T., Bronk Ramsey, C., Brock, F., Bake, D., & Ditchfield, P. (2007). Radiocarbon dates from the Oxford AMS system. Archaeometry, 49(1), 1–60.
Higham, T. F. G., Douka, K., Wood, R., Bronk Ramsey, C., Brock, F., Basell, L., Camps, M., Arrizabalaga, A., Baena, J., Barroso-Ruíz, C., Bergman, C., Boitard, C., Boscato, P., Caparrós, M., Conard, N. J., Draily, C., Froment, A., Galván, B., Gambassini, P., … Jacobi, R. (2014). The timing and spatio-temporal patterning of Neanderthal disappearance. Nature, 512, 306–309. https://doi.org/10.1038/nature13621
Higham, T. F. G., Bronk Ramsey, C., Cheney, H., Brock, F., & Douka, K. (2012). The radiocarbon chronology of Gorham’s cave. In R. B. E. Barton, C. B. Stringer, & J. C. Finlayson (Eds.), Neanderthals in Context: A report of the 1995-1998 excavations at Gorham’s and Vanguard Caves, Gibraltar, University School of Archaeology, Monograph 75 (pp. 62–76).
Housley, R. A., Sanderson, D. C. W., Burbidge, C. I., Richter, D., & Higham, T. F. G. (2007). In V. P. Chabai, J. Richter, & T. Uthmeier (Eds.), Luminescence analysis (OSL and TL) and AMS radiocarbon determinations from Kabazi V, Kabazi V: Interstratification of Micoquian and Levallois-Mousterian camp sites, Part 1. Palaeolithic sites of Crimea (Vol. 3, pp. 51–67). Simferopol-Cologne.
Hublin, J. J. (2015). The modern human colonization of western Eurasia: When and where? Quaternary Science Reviews, 118, 194–210. https://doi.org/10.1016/j.quascirev.2014.08.011
Hublin, J. J., Sirakov, N., Aldeias, V., Bailey, S., Bard, E., Delvigne, V., Endarova, E., Fagault, Y., Fewlass, H., Hajdinjak, M., Kromer, B., Krumov, I., Marreiros, J., Martisius, N. L., Paskulin, L., Sinet-Mathiot, V., Meyer, M., Pääbo, S., Popov, V., … Tsanova, T. (2020). Initial Upper Palaeolithic Homo sapiens from Bacho Kiro Cave, Bulgaria. Nature, 581, 299–302. https://doi.org/10.1038/s41586-020-2259-z
Jacob, E., Querci, D., Caparros, M., Ruiz, B., Higham, T. F. G., & Devièse, T. (2018). Nitrogen content variation in archaeological bone and its implications for stable isotope analysis and radiocarbon dating. J Archael Sci, 93, 68–73.
Jennings, R., Finlayson, C., Fa, D., & Finlayson, G. (2011). Southern Iberia as a refuge for the last Neanderthal populations. J Biogeogrgorha, 38, 1873–1885.
Jones, E. L. (2021). What is a refugium? Questions for the Middle-Upper Palaeolithic transition in peninsular southern Europe. Journal of Quaternary Science, 16(4), 196–212. https://doi.org/10.1080/14888386.2015.1117022
Kolosov, Yu G., & Stepanchuk, V. N. (2002). New radiocarbon dates of Crimean Paleolithic sites. In L. L. Zalizniak (Ed.), Stone age of Ukraine (pp. 18–29) (In Ukranian).
Kolosov, Y. G., Kuharchuk, Y. V., & Ryzhov, S. N. (1997). The investigations nearby the Ak Kaya Rock. In V. A. Kutaisov (Ed.), The archaeological investigations in Crimea, 1994 (pp. 143–150). Simferopol (in Russian).
Kolosov, Y. G. (1983). Mousterian sites of the Belogorsk area. Naukova Dumka (In Russian).
Kolosov, Y. G. (1986). Ak-Kaya Mousterian culture. Naukova Dumka (In Russian).
Kosintsev, P., Mitchell, K. J., Devièse, T., van der Plicht, J., Kuitems, M., Petrova, E., Tikhonov, A., Higham, T., Comeskey, D., Turney, C., Cooper, A., van Kolfschoten, T., Stuart, A. J., & Lister, A. M. (2019). Evolution and extinction of the giant rhinoceros Elasmotherium sibiricum sheds light on late Quaternary megafaunal extinctions. Nat, Ecol & Evol. https://doi.org/10.1038/s41559-018-0722-0
Kozłowski, J. K. (2014). Middle Palaeolithic variability in Central Europe: Mousterian vs Micoquian. Quaternary International, 326-327, 344–363. https://doi.org/10.1016/j.quaint.2013.08.020
Lango, L., Shakun, N. N., Pantyukhina, I. R., Terekhina, V. V., & Sorrentino, G. (2021). Aurignacian grinding stone from Surein I (Crimea): “tracing” the roots of starch-based diet. Journal of Archaeological Science: Reports, 38, 102999. https://doi.org/10.1016/j.jasrep.2021.102999
Laroulandie, V., & d’Errico, F. (2004). Worked bones from Buran-Kaya III Level C and their taphonomic context. In V. P. Chabai, K. Monigal, & A. E. Marks (Eds.), The Paleolithic of Crimea Volume 3, The Middle Paleolithic and Early Upper Paleolithic of Eastern Crimea (pp. 83–94). ERAUL 104.
Longin, R. (1971). New method of collagen extraction for radiocarbon dating. Nature, 230(5291), 241–242.
López-García, J. M., Cuenca-Bescós, G., Finlayson, C., Brown, K., & Pacheco, F. G. (2011). Pal-aeoenvironmental and palaeoclimatic proxies of the Gorham’s cave small mammal sequence, Gibraltar, southern Iberia. Quaternary International, 243, 137–142. https://doi.org/10.1016/j.quaint.2010.12.032
Majkić, A., d’Errico, F., & Stepanchuk, V. (2018). Assessing the significance of Palaeolithic engraved cortexes. A case study from the Mousterian site of Kiik-Koba Crimea. PLoS ONE, 13(5), e0195049. https://doi.org/10.1371/journal.pone.0195049
Markova, A. K. (2005). Small mammals from the Palaeolithic site of Kabazi II, Western Crimea. In V. Chabai, J. Richter, & T. Uthmeier (Eds.), Kabazi II: Last Interglacial occupation, environments and subsistence, Chapter III (pp. 59–73). Simferopol-Cologne.
Markova, A. K. (2012). Small mammals from Paleolithic site Siuren I. In Y. E. Demidenko, M. Otte, & P. Noiret (Eds.), Siuren I rock-shelter: From Late Middle Paleolithic and Early Upper Paleolithic to Epi-Paleolithic in Crimea (Vol. 129, pp. 65–71). ERAUL.
Marks, A., & Chabai, V. (2006). Stasis and change during the Crimean Middle Paleolithic. In E. Hovers & S. L. Kuhn (Eds.), Transitions before the transition: Evolution and stability in the Middle Paleolithic and middle stone age (pp. 121–135). Springer. https://doi.org/10.1007/0-387-24661-4_7
Marks, A. E., Monigal, K. (2000). The Middle to Upper Palaeolithic intetrface at Buran-Kaya III, Eastern Crimean. In J. Orschiedt, G. C. Weniger, (Eds.), Neanderthals and Modern Humans – Discussing the Transition: Central and Eastern Europe from 50.000–30.000 B.P. Wissenschaftliche Schriften 2, Mettmann, Neanderthal Museum, (pp. 212–226).
Marks, A. E., Monigal, K., Usik, V. I., Ferring, C. R., Burke, A., Rink, J., & McKinney, C. (1997). Starosel’e and the Starosel’e child: New excavations, new results. Current Anthropology, 38, 112–123. https://doi.org/10.1086/204591
Marks, A. E., Demidenko, Y., Monigal, K., & Usik, V. (1998). Starosele, the 1993–1995 excavations. In A. E. Marks & V. P. Chabai (Eds.), The Middle Paleolithic of Western Crimea (Vol. 1, pp. 67–100). ERAUL 84.
Marom, A., McCullagh, J. S. O., Higham, T. F. G., Sinitsyn, A. A., & Hedges, R. E. M. (2012). Single amino acid radiocarbon dating of Upper Palaeolithic modern humans. PNAS, 109(18), 6878–6881. https://doi.org/10.1073/pnas.1116328109
Marom, A., McCullagh, J. S. O., Higham, T. F. G., & Hedges, R. E. M. (2013). Hydroxyproline dating: Experiments on the 14C analysis of contaminated and low-collagen bones. Radiocarbon, 55(2), 698–708. https://doi.org/10.1017/S0033822200057854
Massé, J., Patou-Mathis, M. (2009). Siuren I: New zooarchaeological results. In V. P. Chabai, (Eds.), The top issues of the Eastern European prehistoric archaoelogy, Actes du Colloque Conference dedicated to the 100 anniversary of Sergei Nikolaevich Bibikov, Alushta (Crimée, Ukraine), 1–3 October 2008, Archaeological Almanac n° 20. Donbass, Donetsk, (pp. 149–158).
McColl, H., Racimo, F., Vinner, L., Demeter, F., Gakuhari, T., van Morenomayar, J., Víctordriem, G., Gram Wilken, U., Seguin-Orlando, A., de la Fuente Castro, C., Wasef, S., Shoocongdej, R., Souksavatdy, V., Sayavongkhamdy, T., Saidin, M. M., Allentoft, M. E., Sato, T., Malaspinas, A. S., Aghakhanian, F. A., … Willerslev, E. (2018). The prehistoric peopling of southeast Asia. Science, 361(6397), 88–92.
McKinney, C., & Rink, R. (1996). The absolute chronology of the Middle Palaeolithic of the Crimea. In The paper was presented on the 61st annual meeting of the society of American archaeology.
McKinney, C. (1998). U-series dating of enamel, dentine, and bone from Kabazi II, Starosel’e, Kabazi V and GABO. In A. E. Marks & V. P. Chabai (Eds.), Middle Paleolithic of Western Crimea (Vol. 1, pp. 341–354). ERAUL.
Minami, M., Sakata, K., Takigami, M., Nahaoka, T., & Nakamura, T. (2016). Ultrafiltration pretreatment for 14C dating of fossil bones from archaeological sites in Japan. Radiocarbon, 55(2), 481–490. https://doi.org/10.1017/S0033822200057611
Monigal, K., Marks, A. E., Usik, V. I. (1998). Human remains at Starosel’e. In: A. E. Marks, & V. P. Chabai (Eds.), The Middle Paleolithic of western Crimea, Vol. 1 (No. 84, pp. 101-116). ERAUL.
Monigal, K. (2004). The lithic assemblage from Buran-Kaya III level C. In: V. P. Chabai, K. Monigal, & A. E. Marks (Eds.), The Middle Paleolithic and Early Upper Paleolithic of Eastern Crimea. The Paleolithic of Crimea III. ERAUL 104. Etudes et Recherches Archéologiques de l’Université de Liège, Liège, (pp. 419–460).
Moreno-Mayar, J. V., Vinner, L., de Barros Damgaard, P., de la Fuente, C., Chan, J., Spence, J. P., Allentoft, M. E., Vimala, T., Racimo, F., Pinotti, T., Rasmussen, S., Margaryan, A., Iraeta Orbegozo, M., Mylopotamitaki, D., Wooller, M., Bataille, C., Becerra-Valdivia, L., Chivall, D., Comeskey, D., … Willerslev, E. (2018). Early human dispersals within the Americas. Science, 362, 6419. https://doi.org/10.1126/science.aav2621
Mylopotamitaki, D., Weiss, M., Helen Fewlass, H., Zavala, E. I., Rougier, H., Sümer, A. P., Hajdinjak, M., Smith, G. M., Karen Ruebens, K., Sinet-Mathiot, V., Pederzani, S., Essel, E., Harking, F. S., Xia, H., Hansen, J., Kirchner, A., Lauer, T., Stahlschmidt, M., Hein, M., … Hublin, J.-J. (2024). Homo sapiens reached the higher latitudes of Europe by 45,000 years ago. Nature, 626, 341–346.
Nalawade-Chavan, S., McCullagh, J., & Hedges, R. (2013). New hydroxyproline radiocarbon dates from Sungir, Russia, confirm early Mid Upper Palaeolithic burials in Eurasia. PLoS ONE, 9(1), e76896.
Nelson, D. E. (1991). A new method for carbon isotopic analysis of protein. Science, 251, 552–554.
Otte, M. (2021). Buran Kaya, Gravettian and Aurignacian in Crimea. L, Anthropologie, 125(4), 102918. https://doi.org/10.1016/j.anthro.2021.102918
Patou-Mathis, M. (1999). Archeozoological analysis of the Middle Paleolithic Fauna from selected levels of Kabazi II. In V. P. Chabai & K. Monigal (Eds.), The Middle Paleolithic of western Crimea (Vol. 2, pp. 41–74). ERAUL 87.
Patou-Mathis, M. (2005). Analyses archéozoologiques des unités V et VI de Kabazi II. In V. Chabai, J. Richter, & T. Uthmeier (Eds.), Kabazi II: Last interglacial occupation, environment & subsistence (Vol. 1, pp. 73–98). Simferopol-Cologne.
Patou-Mathis, M. (2006). Analyse Archéozoologique de l’Unite II, Niveaux II/7AB à IIA/4B. In V. Chabai, J. Richter, & T. Uthmeier (Eds.), Kabazi II: The 70000 years since the Last Interglacial. Palaeolithic site of Crimea (Vol. 2, pp. 37-62). Simferopol – Cologne.
Péan, S., Puaud, S., Crépin, L., Prat, S., Quiles, A., van der Plicht, J., Valladas, H., Stuart, A. J., Drucker, D. G., Patou-Mathis, M., Lanoë, F., & Yanevich, A. (2013). The Middle to Upper Paleolithic Sequence of Buran-Kaya III (Crimea, Ukraine): New stratigraphic, paleoenvironmental, and chronological results. Radiocarbon, 55, 3. https://doi.org/10.1017/s0033822200048384
Pettit, P. B. (1998). Middle and early Upper Paleolithic Crimea - The radiocarbon chronology. In M. Otte (Ed.), Préhistorie de’Anatolie, Genèse de deux mondes (Vol. 1, pp. 329–338). ERAUL 85.
Picin, A., Hajdinjak, M., Nowaczewska, W., Benazzi, S., Urbanowski, M., Marciszak, A., Fewlass, H., Bosch, M. D., Socha, P., Stefaniak, K., Żarski, M., Wiśniewski, A., Hublin, J. J., Nadachowski, A., & Talamo, S. (2020). New perspectives on Neanderthal dispersal and turnover from Stajnia Cave (Poland). Science and Reports, 10, 14778. https://doi.org/10.1038/s41598-020-71504-x
Prat, S., Péan, S. C., Crépin, L., Drucker, D. G., Puaud, S. J., Valladas, H., Lázničková-Galetová, M., van der Plicht, J., & Yanevich, A. (2011). The oldest anatomically modern humans from far southeast europe: Direct dating, culture and behavior. PLoS ONE, 6, 6. https://doi.org/10.1371/journal.pone.0020834
Prat, S., Péan, S., Crépin, L., Puaud, S., Drucker, D. G., Lázničková-Galetová, M., van der Plicht, J., Valladas, H., Verna, C., & Patour-Mathis, M. (2018). The first anatomically modern humans from south-eastern Europe. In Contributions from the Buran-Kaya III Site (Crimea). Bulletins et Mémoires de la Société d’anthropologie de Paris, Volume 30 N° 3–4, October 2018 (pp. 169–179). https://doi.org/10.3166/bmsap-2018-0032
Prüfer, K., Posth, C., Yu, H., Stoessel, A., Spyrou, M. A., Deviese, T., Mattonai, M., Ribechini, E., Higham, T., Velemínský, P., & Brůžek, J. (2021). A genome sequence from a modern human skull over 45,000 years old from Zlatý kůň in Czechia. Nature Ecology and Evolution, 5(6), 820–825.
Ramírez-Pedraza, I., Rivals, F., Uthmeier, T., & Chabai, V. (2020). Paleoenvironmental and seasonal context of the Late Middle and Early Upper Palaeolithic occupations in Crimea: an approach using dental wear patterns in ungulates. Archaeological and Anthropological Sciences, 12, 268. https://doi.org/10.1007/s12520-020-01217-9
Reimer, P., Austin, W., Bard, E., Bayliss, A., Blackwell, P., Bronk Ramsey, C., Butzin, M., Cheng, H., Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Hajadas, I., Heaton, T., Hogg, A., Hughen, K., Kromer, B., Manning, S., Muscheler, R., Palmer, J., Pearson, C., Van der Plicht, J., Reimer, R., Richards, D., Scott, E., Southon, J., Turney, C., Wacker, L., Adolphi, F., Büntgen, U., Capano, M., Fahrini, S., Fogtmann-Schultz, A., Friedrich, R., Köhler, P., Kudsk, S., Miyake, F., Olsen, J., Reinig, F., Sakamonto, M., Sookdeo, A., Talamo, S. (2020) The IntCal20 Northern manufacture? In A, Pastoors. & M, Peresani (Eds.), Flakes Not Blades: The Role of Flake Production at the Onset of the Upper Palaeolithic. Wissenschaftliche Schriften des Neanderthal Museums 5. Neanderthal Museum, pp 239–260.
Reynolds, N., Dinnis, R., Bessudnov, A. A., Devièse, T., & Higham, T. F. G. (2017). The Kostënki 18 child burial and the cultural and funerary landscape of Mid Upper Palaeolithic European Russia. Antiquity, 91(360), 1435–1450. https://doi.org/10.15184/aqy.2017.150
Richter, K. K., Codlin, M. C., Seabrook, M., & Warinner, C. (2022). A primer for ZooMS applications in archaeology. Proceedings of the National Academy of Sciences U.S.A., 119(20), e2109323119. https://doi.org/10.1073/pnas.2109323119
Rink, W., Lee, H., Rees-Jones, K., Goodker, K. (1998). Electron spin resonance (ESR) and mass spectrometric U-series (MSUS) dating of teeth in Crimean Paleolithic sites: Starosel’e, Kabazi II, and Kabazi V. In: A. Marks, & V. Chabai (Eds.), The Middle Paleolithic of Western Crimea, Vol. 1. ERAUL 84 Etudes et Recherches Archéologiques de l’Université de Liège, (pp. 323–340).
Rybin, E, P., Belousova, N, E., Derevianko, A, P., Douka, K., Higham, T. (2023) The Initial Upper Palaeolithic of the Altai: New radiocarbon determinations for the Kara-Bom site. J Hum Evol, 185. https://doi.org/10.1016/j.jhevol.2023.103453
Sikora, M., Seguin-Orlando, A., Sousa, V. C., Albrechtsen, A., Korneliussen, T., Ko, A., Rasmussen, S., Dupanloup, I., Nigst, P. R., Bosch, M. D., Renaud, G., Allentoft, M. E., Margaryan, A., Vasilyev, S. V., Veselovskaya, E. V., Borutskaya, S. B., Devièse, T., Comeskey, D., Higham, T., … Willerslev, E. (2017). Ancient genomes show social and reproductive behavior of early Upper Paleolithic foragers. Science, 358(6363), 659–662. https://doi.org/10.1126/science.aao1807
Slimak, L., Zanolli, C., Higham, T., Frouin, M., Schwenninger, J. L., Arnold, L. J., Demuro, M., Douka, K., Mercier, N., Guérin, G., Valladas, H., Yvorra, P., Giraud, Y., Seguin-Orlando, A., Orlando, L., Lewis, J. E., Muth, X., Camus, H., Vandevelde, S., … Metz, L. (2022). Modern human incursion into Neanderthal territories 54,000 years ago at Mandrin France. Science Advances, 8, 6. https://doi.org/10.1126/sciadv.abj9496
Soffer, O. (1994). Ancestral lifeways in Eurasia – The Middle and Upper Palaeolithic records. In M. H. Nitecki & D. V. Nitecki (Eds.), Origins of Anatomically Modern Humans (pp. 101–120). Plenum Press.
Spindler, L., Comeskey, D., Chabai, V., Uthmeier, T., Buckley, M., Devièse, T., & Higham, T. (2021). Dating the last Middle Palaeolithic of the Crimean Peninsula: New hydroxyproline AMS dated from the site of Kazi II. Journal of Human Evolution, 156, 102996.
Spitery, E. (1980). L’enfant de Staroceljé et sa pathologie. Bulletins et Mémoires de la Société d’Anthropologie de Paris 7 (XIII): 65–73.
Sponheimer, M., Ryder, C. M., Fewlass, H., Smith, E. K., Pestle, W. J., & Talamo, S. (2019). Saving Old Bones: A non-destructive method for bone collagen prescreening. Science and Reports, 9(1), 13928.
Stafford, T. W., Duhamel, R. C., Haynes, V. C., & Brendel, K. (1982). Isolation of proline and hydroxyproline from fossil bone. Life Sciences, 31, 931–938. https://doi.org/10.1016/0024-3205(82)90551-3
Stafford, T. W., Hare, P. E., Currie, L., Jull, A. J. T., & Donahue, D. J. (1991). Accelerator radiocarbon dating at the molecular level. Journal of Archaeological Science, 18(1), 35–72. https://doi.org/10.1016/0305-4403(91)90078-4
Stepanchuk, V. N., Kovalyukh, M. M., & van der Plicht, J. (2004). Radiocarbon age of Late Pleistocene Palaeotihic sites of the Crimea. Kamiana Doba Ukrainy, 5, 34–61. In Ukrainain.
Stepanchuk, V. N., Vasilyev, S. V., Khaldeeva, N. I., Kharlamova, N. V., & Borutskaya, S. B. (2017). The last Neanderthals of Eastern Europe: Micoquian layers IIIa and III of the sites of Zaskalnaya VI (Kolosovskaya), anthropological records and context. Quaternary International, 428, 132–150. https://doi.org/10.1016/j.quaint.2015.11.042
Stepanchuk, Vadim. (2006). Lower and Middle Paleolithic of Ukraine. Zelena Bukovyna, Chernivtsy (In Ukranian).
Stepanchuk, V, N. (2022). Pebbles with ochre residues from Neanderthal sites of Eastern Crimea. In O. Cîrstina, E. C. Niţu (Eds.), A life dedicated to the paleolithic: Studies in honorem Marin Cârciumaru. Târgovişte, Cetatea De Scaun.
Tisnérat-Laborde, N., Valladas, H., Kaltnecker, E., & Arnold, M. (2003). AMS radiocarbon dating of bones at LSCE. Radiocarbon, 45(3), 409–419.
Topographic-map (2024) de-at.topographic-map.com. Accessed: 20.11.2023
Uthmeier, T. H. (2024). Supplementary data to: Identification of workpieces, transformation analysis, and the transition from the Middle to the Upper Paleolithic in the Northern Black Sea Region: the case of Buran-Kaya III, Level C (Crimea, Ukraine). In T. Uthmeier & A. Maier (Eds.), STONE AGE. Studying technologies of non-analogous environments and glacial ecosystems. Papers in Honor of Jürgen Richter. Universitätsforschung zur prähistorischen Archäologie (UPA) Band 393. Verlag Dr. Rudolf Habel (pp. 231–256). https://doi.org/10.5281/zenodo.8414106
Uthmeier, T. H., & Chabai, V. (2018). Formation processes at sites with high-resolution sequences in the Crimean Middle Palaeolithic: The Kabazi V rock shelter and the open-air site of Kabazi V compared. Quaternary International, 485, 44–67. https://doi.org/10.1016/j.quaint.2018.01.017
Uthmeier, T. H., Ickler, S., & Kurbjuhn, M. (2008). Site catchment analysis in the late crimean middle palaeolithic: A Gis based approach. In V. P. Chabai, J. Richter, & T. H. Uthmeier (Eds.), Kabazi V: Interstratification of Micoquian & Levallois-Mousterian camp sites. Palaeolithic sites of Crimea, Volume 3, Part 2 (pp. 481–508). Simferopol-Cologne.
Uthmeier, T. (2004). Transformation analysis and the reconstruction of on-site and off-site activities: Methodological remarks. In V. P. Chabai, K. Monigal, & A. E. Marks (Eds.), The Middle Paleolithic and early upper Paleolithic of Eastern Crimea: The Paleolithic of Crimea III. Etudes et recherches archéologiques de l’Université de Liège 104 (pp. 175–191). Université de Liège.
Uthmeier, T. H. (2012). The Transition from Middle- to Upper Palaeolithic at Buran Kaya III, Crimea (Ukraine): A case of conceptual continuity of lithic artefact manufacture? In A. Pastoors & M. Peresani (Eds.), Flakes not blades: The role of flake production at the onset of the Upper Palaeolithic. Wissenschaftliche Schriften des Neanderthal Museums 5 (pp. 239–260). Neanderthal Museum.
Van der Sluis, L. G., Zazzo, A., Tombret, O., Thil, F., & Pétillon, J. M. (2023). Testing the use of XAD resin to remove synthetic contamination from archaeological bone prior to radiocarbon dating. Radiocarbon, 65(5), 1160–1175. https://doi.org/10.1017/RDC.2023.100
Winguth, C., Wong, H. K., Panin, N., Dinu, C., Georgescu, P., Ungureanu, G., Krugliakov, V. V., & Podshuveit, V. (2000). Upper Quaternary water level history and sedimentation in the northwestern Black Sea. Marine Geology, 167(1–2), 127–146.
Wolpoff, M., & H. (1980). Paleoanthropology. Alfred A. Knopf.
Wood, R. E., Douka, K., Boscato, P., Haesaerts, P., Sinitsyn, A., & Higham, T. F. G. (2012). Testing the ABOx-SC method: Dating known age charcoals associated with the nCampanian Ignimbrite. Quaternary Geochronology, 9, 1626.
Wood, R. E., Ruíz, B., Caparrós, M., Pardo, J. F. J., Santos, B. G., & Higham, T. F. G. (2013). Radiocarbon dating casts doubt on the late chronology of the Middle to Upper Palaeolithic transition in southern Iberia. Proceedings of the National Academy of Sciences USA, 110(8), 2781–6. https://doi.org/10.1073/pnas.1207656110
Yanevich, A. (2014). Les occupations gravettiennes de Buran-Kaya III (Crimée): contexte archéologique. L’Anthropologie, 118, 5. https://doi.org/10.1016/j.anthro.2014.10.006
Zilhão, J., & Pettitt, P. (2006). On the new dates for Gorham’s Cave and the late survival of Iberian Neanderthals Before Farming. Hunter Gatherer Research, 3, 1–9.
Acknowledgements
The excavations at Zaskalnaya V as well as the subsequent analysis of the excavated materials were funded by the grant ‘The dispersal of modern humans into an Eastern European refugial area of late Neanderthals: interdisciplinary studies of contemporaneous industries from the Middle to Upper Palaeolithic transition in the Crimea (Ukraine)’ of the Deutsche Forschungsgemeinschaft (DFG) given to T.U. in cooperation with V. C. (Project number 227159777). We would also like to thank the two reviewers who provided exceptional feedback and comments, dedicating their time and efforts, to help enhance the paper. Their insightful thoughts and suggestions for improvement have contributed to the papers overall quality.
Funding
Open access funding provided by University of Vienna.
Author information
Authors and Affiliations
Contributions
Pigott, E. M1,2., Uthmeier, T3., Chabai, V4. and Higham, T.F.G1,2. Pigott wrote the main manuscript, with the main reviewer being Higham Pigott prepared all figures and tables All authors reviewed manuscript.
Corresponding author
Ethics declarations
Conflict of Interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Pigott, E.M., Uthmeier, T., Chabai, V. et al. The Late Middle and Early Upper Palaeolithic in Crimea (Ukraine)—A Review of the Neanderthal Refugium Hypothesis. J Paleo Arch 7, 27 (2024). https://doi.org/10.1007/s41982-024-00194-y
Accepted:
Published:
DOI: https://doi.org/10.1007/s41982-024-00194-y