https://orcid.org/0000-0002-0836-6862
Figures
Abstract
The European Final Palaeolithic witnessed marked changes in almost all societal domains. Despite a rich body of evidence, our knowledge of human palaeodemographic processes and regional population dynamics still needs to be improved. In this study, we present regionally differentiated population estimates for the Greenland Interstadial 1d-a (GI-1d-a; 14-12.7 ka cal BP) and the Greenland Stadial 1 (GS-1; 12.7-11.6 ka cal BP) for Southern, Western, Northern and Central Europe. The data were obtained by applying the Cologne Protocol, a geostatistical approach for estimating prehistoric population size and density, to a newly compiled dataset of Final Palaeolithic sites. On a large spatio-temporal scale and compared to preceding Upper Palaeolithic phases, areas north of the Alps become the dominant demographic growth area for the first time since the dispersal of anatomically modern humans into Europe. At smaller scales, we observe divergent regional trends, with a conspicuous lack of archaeological evidence appearing in previously occupied areas of central France and Germany. Our study also shows that during the Final Palaeolithic, the climatic cooling of GS-1 coincides with a pronounced population decline in most parts of the study area. An apparent increase in population density occurs only in north-eastern Central Europe and north-eastern Italy. Our estimates suggest that the total population was reduced by half. Similar results, with a relationship between decreasing temperatures and decreasing populations, have already been observed for the late phase of the Gravettian, when populations were reduced to only one third of those estimated for the early phase. Yet, in contrast to the collapse of local populations during the late Gravettian, the increase in population densities in Central Europe during GS-1 indicates population movements eastwards, possibly in response to deteriorating climatic conditions, particularly in western regions during the Younger Dryas.
Citation: Schmidt I, Gehlen B, Winkler K, Arrizabalaga A, Arts N, Bicho N, et al. (2025) Large scale and regional demographic responses to climatic changes in Europe during the Final Palaeolithic. PLoS ONE 20(4): e0310942. https://doi.org/10.1371/journal.pone.0310942
Editor: Przemysław Mroczek, Maria Curie-Sklodowska University: Uniwersytet Marii Curie-Sklodowskiej, Poland
Received: September 9, 2024; Accepted: February 19, 2025; Published: April 2, 2025
Copyright: © 2025 Schmidt et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are stored at doi:10.5281/ZENODO.13132357 or within the manuscript and its Supporting Information file.
Funding: AZi and IS acknowledge funding from the German Research Foundation, Project-ID 57444011 – SFB 806 ‘Our Way to Europe’; AM and IS acknowledge funding from the Ministry of Culture and Science of the State of North Rhine-Westphalia, 'HESCOR', PB22-081 ‘Profilbildung 2022’; BVE, MJW, SBG acknowledge funding from the German Research Foundation, Project-ID 290391021-SFB 1266, ‘Scales of Transformation’; WS acknowledges funding from the German Research Foundation, Project-IDs 386654307 and 57444011; and FR acknowledges funding through the ERC Consolidator Grant project CLIOARCH from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (No. 817564). AZa received funding from a British Academy Postdoctoral Fellowship at the University of York. None of the funders played any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The Late Glacial, comprising Greenland Interstadial 1 (GI-1) and Stadial 1 (GS-1) [1], was a period when the impact of rapidly changing environmental conditions drove high-magnitude and high-amplitude ecosystem shifts. The extensive dispersal of shrub and tree species into northern Western and Central Europe alongside the associated fauna during GI-1 created a European habitat with particularly pronounced regional differences. Although the so-called ‘mammoth steppe’ [2] of preceding periods was also never a uniform habitat [3], this emerging, regionally diverse ecotonal patchwork presumably had consequences for most fundamental properties of hunter-gatherer societies, such as subsistence strategies, social networks, and – prominently – the structure and distribution of the underlying meta-population.
In Europe, especially in the areas studied here, these changes are culturally associated with the transition from the Late Upper to the Final Palaeolithic period (e.g., [4–7]). Subsistence strategies change from hunting migratory game to a preference for more stationary, thermophilous game. Settlement and land use patterns are marked by sites with functionally differentiated areas, such as Le Closeau or Reichwalde [8,9] rather than large agglomerative sites, such as Gönnersdorf [10–12]. Changes in lithic technology, particularly in core reduction strategies, signalled a less normative use of lithic raw material of varying quality, alongside a regionalisation in lithic projectile morphology [4,13–19]. Visual representations are no longer found in caves but begin to appear frequently on pebbles. In addition, abstract representations become much more dominant than figurative ones.
From a demographic perspective, these profound changes went hand in hand with expanding hunter-gatherer groups into the northern European lowlands. From the extensive dispersal of anatomically modern humans into Europe during the Aurignacian up to the repopulation of Central Europe after the Last Glacial Maximum, the meta-population in Europe had undergone substantial fluctuations regarding absolute numbers and densities of people as well as social interconnectedness of sub-populations and spatial distributions of genetic haplotypes [20–24]. In its broadest outline, however, the large-scale distribution pattern of hunter-gatherers remained remarkably stable, with one demographic centre in the Franco-Cantabrian region and another in eastern Central Europe, covering Lower Austria, Moravia, southern Poland, and northern Hungary. In the Final Palaeolithic, this pattern changed drastically.
The change occurred between 15 and 14 ka BP, but there is no generally accepted date or event marking the end of the Late Upper and the beginning of the Final Palaeolithic in Western and Central Europe. Particularly problematic for a chronological delimitation is a plateau in the calibration curve during this period that causes wide calibration age ranges [25,26] and, thus, reduces chronological resolution. Further complicating matters are regionally differentiated environmental responses to the climatic shift at the transition from the Late Pleniglacial to GI-1e, which appears rather abrupt in the Greenland ice core records and may have been related to a substantial meltwater discharge into the North Atlantic [27] and tipping points in sea-ice cover [28]. Atmospheric warming over Europe would have started well before, and the associated ecosystem change likely was more gradual [29,30]. There is growing evidence that such an early start and comparatively slow change of ecosystems did not only take place south of the Alps [31,32], but also in Central Europe [33,34]. Variable response periods of the vegetation in diverse parts of Europe (e.g., [34–37]) further indicate regionally distinct ecosystem reactions differentiated by both leads and lags. Such differences are not limited to warming events but also occurred during the cooling of GS-1 ( ≈ Younger Dryas) and subsequent warming when de- and increases in seasonal temperature and precipitation varied markedly between different areas throughout Europe [38–42]. An earlier and more gradual ecosystem change prior to GI-1e also has consequences for our understanding of the timing of human population dispersal into Northern Europe. It is in accord with an apparently earlier northward expansion of hunter-gatherers of the Creswellian and Hamburgian [43–45].
From a palaeogenetic point of view, the limited data of this period indicates a major turnover between 15 and 14 ka BP, during which the so-called ‘Oberkassel cluster’ became the dominant haplogroup in Central Europe [24]. In terms of socio-cultural developments, it is widely accepted that archaeological phenomena after 14 ka BP (post-GI-1e) are attributed to the Final Palaeolithic [4,19,46,47]. Therefore, we chose this date as the older boundary of our study, while the younger boundary coincides with the end of GS-1 and the end of the Pleistocene at 11.6 ka BP.
Here, we present regionally differentiated estimates of absolute numbers and densities of people for two phases of the Final Palaeolithic according to the Cologne Protocol [48]. A previous study considered the Final Palaeolithic as a whole [22], but limited data granularity has hitherto precluded analyses at a finer resolution. Through the involvement of regional experts from different countries, we compiled an extensive database [49], allowing for a chronological subdivision of the assemblages into two phases. The older phase, dates between 14 and 12.7 ka BP and thus corresponds to GI-1d-a or the Older Dryas and the Allerød sensu Iversen [50], while the younger phase dates to between 12.7 and 11.6 ka BP and thus corresponds to GS-1 or the Younger Dryas (Fig 1). Based on our estimates, we address how the human meta-population responded demographically to environmental change associated with the climatic cooling from GI-1d-a to GS-1 at a subcontinental and regional spatial scale. We compare these results with the demographic response during the Gravettian [52], that was similarly linked to major cooling, and discuss similarities and differences, as well as implications for social networks and behavioural adjustments.
The time-bins defined in the upper part of the graph indicate the different typo-chronological resolutions of the assemblages that were taken into account for the database and analysis (NGRIP δ18O data from [51]).
Materials and methods
The database underlying this study comprises archaeological sites from Southern, Western, Northern and Central Europe. It has been compiled with the regional expertise of the participating authors and is available online as an open-access resource [49]. In order to compare this detailed record of evidence across the study area, unpublished (and therefore unverifiable) material and individual finds have been excluded. Some differences, however, turned out to be inevitable due to data accessibility (publication, database management of heritage authorities) or regionally varying properties of the record (e.g., chronological resolution). In sum, the database considers material from 1788 sites (Fig 2, excavations and surface collections). 1548 assemblages can either be directly dated and/or typologically or stratigraphically assigned to one of the two phases. GI-1d-a (14-12.7 ka BP) comprises 988 assemblages and GS-1 (12.7-11.6 ka BP), comprises 560 assemblages. Material from assemblages that cover larger time-bins (see Fig 1) contain 237 assemblages assigned to the Final Palaeolithic in general, 131 assemblages assigned to the Final Palaeolithic and the Preboreal in general, and 75 assemblages are recorded separately because their chronological placement relates to the final GS-1 and the succeeding Preboreal phase. Frequent generic attribution is most problematic for evidence from Central Europe.
Green dots: GI-1d-a; blue dots: GS-1. Filled circles: dataset B (chronologically subdivided sites); filled and open circles: dataset A (all sites; see text and Fig 1). Base map made with Natural Earth (see also supporting information).
Submerged coastal areas and so-called Doggerland provided large, formerly inhabitable landmasses. Evidence of human occupation from these areas is extremely scarce and, therefore, virtually absent in the dataset. Examples of long-distance raw material transport, e.g., of Helgoland flint [53–55], speak to human presence across these now submerged areas. If diagnostic archaeological artefacts are recovered from below current sea level, they often lack geospatial information. Our calculations of the estimates, regional comparisons and subsequent interpretations of the results therefore rest upon the available evidence from the current landmass, as mapped in Fig 2, and leaves estimates about the population of Doggerland to further investigations.
On the North European Plain and adjacent areas, GI-1d-a mainly comprises sites attributed to the Federmesser-Gruppen or Federmesser culture (e.g., [56–61]). The beginning of this phase overlaps slightly with the Havelte phase of the Hamburgian [43,47,58,62,63] and potentially with the latest Creswellian sites [56,64,65]. Given that most sites attributed to these industries predate GI-1d-a (see also [66]), Havelte and Creswellian assemblages were consistently excluded from the dataset. The disappearance of these ‘cultures’ may in itself be indicative of a demographic break at this important tipping-point transition [45,58,67] as indicated by results from the British Isles [68]. During GS-1, assemblages assigned to the Ahrensburgian and Swiderian occur in the Central and Northern lowlands [60,69–71], at time persisting into the subsequent Preboreal period (e.g., [72]). The position of the Bromme culture in this framework, situated at the transition from Allerød to Younger Dryas, is subject to discussion [58,73,74]. A high number of surface collections characterises the archaeological evidence on the northern Plains in general.
In the Benelux, Ahrensburgian and Epi-Ahrensburgian assemblages [75–79] replace the Federmesser-Gruppen during GS-1, with most dates placing the Ahrenbsurgian notably late into GS-1 or even into the Holocene [80]. Long Blade industries [81,82] and Belloisian/Long Blade assemblages [82–86] emerged in the UK and the Paris Basin, respectively. Assemblages with so-called ‘Riesenklingen’ (giant blades) also occur in the Netherlands [78], north-western Germany (e.g., [87]) and Denmark [88], where they are assigned to the younger half of GS-1 and occasionally seem to diffuse into the Preboreal [89].
In France, the term ‘Azilian’ is used in competition with ‘Federmesser-Gruppen’. The latter is more commonly used in northern France, particularly in the Somme valley. This distinction has more to do with taxonomy and the history of research [90] than with any archaeological reality. However, differences cannot be overlooked typologically and technologically, especially when comparing certain sites (e.g., Le Closeau vs Saleux), although it is not yet possible to explain the exact meaning of this variability. This taxonomic question also applies to the ‘Belloisian’, ‘Laborian’ and ‘Epi-Ahrensbourgian’ entities, which are sometimes supposed to be part of the same technical trend (‘FBBT’; [91–93]). The situation is, of course, different for assemblages related to the different phases of the Epigravettian period, and particularly to the most recent ones. They reflect the probable spread of these traditions beyond southeastern France - at least along the Rhone corridor and certainly as far as the Jura range [94–97] - during the GS-1 period. In the actual state of knowledge, we can highlight offsets between their chronometric boundaries and typological changes [85,92,98–100].
Along the Atlantic façade of the Iberian Peninsula, regional chronologies differ in the assignment of taxonomic units to certain periods [101–104]. In Portugal, the Magdalenian is seen as ending before the Younger Dryas. The following phase is frequently called Epipalaeolithic [103,105] or Early Mesolithic [104]. Either way, the lithic assemblages appear very similar in the two phases in both technology and typology. In Cantabrian Spain and Pyrenees, Azilian is the most used term.
In Italy, complexes dated to the Late Glacial interstadial and the Younger Dryas are framed within the late and terminal Epigravettian, although differences exist between regions in chronology and terminology [106,107]. Indeed, from the end of the Last Glacial Maximum up to the Pleistocene/Holocene boundary, the Epigravettian modified its technological features, the typology of tools and of weapons, as related to the use of bow and arrow, particularly in the latest phases [108]. The shortening of certain tools, such as frontal end-scrapers [109], and the spread of backed knives occur alongside with the appearance of geometric microliths and the increasing use of the microburin blow technique. These modifications seem to be shared at larger scale in Eastern and Mediterranean Europe [13].
The greatest difficulties for a chronological subdivision were encountered along the northern range of the Alps, particularly in southern Germany and the Czech Republic. In contrast to many other areas, there is no discernible shift in typology between GI-1d-a and GS-1 in south-eastern Germany, and assemblages of both phases are attributed to the Federmesser-Gruppen [110]. The absence of chronometric data further complicates matters since most assemblages are surface collections and organic preservation is poor. In the Czech Republic, Federmesser-Gruppen are only present in the north-western part of Bohemia towards the end of GS-1 [111,112]. In contrast, assemblages of GI-1d-a and the early GS-1 and in other parts of the country are at times assigned to the Epimagdalenian or Final Palaeolithic. However, many sites are surface collections and can therefore not be securely associated with either of the two phases [113,114]. In the sandstone regions, stratified and radiometrically dated sites are known [34,115,116], which may help refine the chronology of the Final Palaeolithic in this region. Better dated sites and assemblages occur in southernmost Germany and in Switzerland [117–122].
Alongside the database on Final Palaeolithic sites and assemblages, we collected data on lithic raw material transport from the literature. Where possible, the information was divided into the two chronological phases (Fig 1), too. If a site was known to contain assemblages from both periods but raw material data are presented in the literature generically as ‘Final Palaeolithic’, these data were assigned to both periods equally. If a site was known to be only generically assigned to the Final Palaeolithic, the raw material data was recorded but not considered for the phase-specific calculation of the demographic estimates. This accounts especially for sites from the area of SE Germany and the Czech Republic (e.g., [110,111]). We also excluded raw materials documented by single pieces and only considered raw materials representing at least 1% of an assemblage (see also [48]). In a second step, we determined the Raw Material Catchment Areas (RMCA) for each assemblage by adding a 5 km radius buffer around the archaeological site and circumscribing the area of the site-buffer and the related raw material sources by constructing a convex hull around them. We excluded RMCAs which only comprised a single source-to-site distance, as well as RMCAs which were smaller than 500 km2 since they turned out to be exceptional outliers among the overall RMCA sizes (these assemblages are from Grotte du Bichon and Blot for GI-1d-a and Gahlen-Schermbeck, Bad Breisig and Babí Pec for GS-1). Eventually, for GI-1d-a, 53 out of 108 RMCAs and for GS-1, 39 out of 97 RMCAs were used to calculate demographic estimates (S1 and S2 Tables). The scarcity or even lack of available raw material data in some Core Areas (e.g., Great Britain, Denmark, and northern Spain) required transferring RMCA-sizes from adjacent areas.
Population estimates are calculated according to the Cologne Protocol, which is presented in [48] and explained in detail for hunter-gatherer societies in the supplementary information provided there. In brief, it statistically divides a certain map section, called Total Area of Calculation, into areas of high hunter-gatherer activities, called Core Areas, and areas of little or no activities. The Core Areas are delimited by calculating areas with equal site densities, encircled by the so-called optimally describing isoline. To account for differences in site density within Europe, the optimally describing isolines have been modelled individually for three regions (for details, see S1 and S2 Figs). The Total Area of Calculation for the current study covers 2.6 million km2 of Southern, Western, Northern, and Central Europe, following the current coastlines. To estimate the number of hunter-gatherer groups per Core Area, the size of each optimally describing isoline (in km2) is divided by the quartiles (Q1, Q2, Q3) of the related RMCA. The raw material catchments are considered indicative of the spatial extent of a single socio-economic unit, with smaller RMCAs resulting in higher, and larger RMCAs in lower population size and density estimates. While the observed variability of RMCA sizes is reflected in our estimates by the Q1 and Q3 maxima and minima, the Q2 is used for further discussions and comparisons. The results are then multiplied with an average group size (42.5) derived from ethno-historic cases selected based on socio-economic considerations (see [48,123]), providing the estimate of the absolute number of people. Population densities are then calculated for each Core Area by dividing the respective numbers of people by the size of the Core Area, and for the Total Area of Calculation by dividing the number of all people by the size of the selected map section.
The Cologne Protocol does not require to adjust data for differences in time length or site numbers when comparing two different periods. This has been discussed in detail elsewhere [48]. However, to account for the fact that a chronological subdivision of the dataset was more feasible in some regions and less so in others, we decided to calculate for each phase two demographic estimates based on different datasets: the so-called ‘dataset A’ includes all sites, subdivided assemblages as well as sites that are only generically assigned to the Final Palaeolithic period or the time-bins indicated in Fig 1. Information about human presence recorded from sites with low chronological resolution is thus retained in the modelling, which may then, however, lead to overestimations for a given phase at regional scales. The second demographic estimate, based on’dataset B‘, in contrast, considers only assemblages with a clear chronological subdivision assigned to them, attributable either to GI-1d-a or to GS-1, or sites that can be clearly assigned to both phases. Due to the overall smaller sample size of ‘dataset B’, it is expected to produce lower estimates – likely underestimating the number of people – than the ones derived from dataset A, but it provides the best available high chronological control. All estimates are given as a mean estimate and, based on our calculation protocol, with a minima and maxima estimates (related to Q1 and Q3 of the RMCA).
Results
In the following, we will discuss the results of the estimates from dataset A and dataset B for each chronological phase, first considering dataset A with all sites, and then the smaller but chronologically highly resolved dataset B.
For the estimates from dataset A, all Core Areas of GI-1d-a (Fig 3) enclose an area of approximately 635,000 km2 and contain 90% of the sites (1220 out of 1356), while those of GS-1 (Fig 4), in contrast, enclose only about 481,000 km2 and contain 82% of the sites (825 out of 1003; for further details, see S1 Fig). The mean estimate for the total population for GI-1d-a is approximately 8100 people at any given time, with minima and maxima ranging from 4800 to 14,300 (Table 1). For the GS-1, the mean estimate for the total population drops to 4250 people living at the same time, with minima and maxima ranging from 2200 to 8800 people (Table 2). This decrease in population size is accompanied by a decrease in population density within the Core Areas from a mean value of 1.3 people per 100 km² (p/100km²) in GI-1d-a (min. 0.8, max. 2.2) to 0.9 p/100km² in GS-1 (min. 0.5, max. 1.8). In the regional results, the Czech Republic and SE Germany are combined by the Core Area and have, with around 2500 people, the largest and with 2.2 p/100 km² also the densest population in GI-1d-a. However, they also show the strongest decline to only 460 people and 0.9 p/100 km² in GS-1. The second highest estimates for GI-1d-a are found in northern Spain and the French Pyrenees with 1200 people, closely followed by Poland and NE Germany with 1100 people. Despite these similar numbers, the density varies considerably between 1.7 p/100 km² in the former and 0.8 p/100 km² in the latter areas. The remaining Core Areas, with mean estimates below 500 people during GI-1d-a, show sometimes moderate declines compared to the data from GS-1, such as N and SE France, while other also experience reductions by half, such as Great Britain and Portugal. Especially for S Scandinavia, the estimates indicate a dramatic decline. For the Total Area of Calculation (‘study area’ in Fig 2), the estimated population density drops from 0.003 in GI-1d-a to 0.002 p/km2 in GS-1.
The data from dataset A (thick lines, empty and filled dots) and dataset B (shaded areas, filled dots only) are shown. For details on areal data and population estimates see Table 1 (data A) and S3 Table (data B). Reconstructions of ice sheets (white), river channels, and coastline (>45°N) correspond to GI-1c-a conditions. Dashed line: study area (base map: see details in supporting information; for ice sheets and drainage system see [4] and references therein).
The data from dataset A (thick lines, empty and filled dots) and dataset B (shaded areas, filled dots only) are shown. For details on areal data and population estimates see Table 2 (data A) and S4 Table (data B). Reconstructions of ice sheets (white), river channels, and coastline (>45°N) correspond to GS-1 conditions. Dashed line: study area (base map: see details in supporting information; for ice sheets and drainage system see [4] and references therein).
For the estimates from dataset B (see S3 and S4 Tables), the size of all Core Areas of GI-1d-a (Fig 3: shaded areas) is reduced to 489,000 km2 (85% of the sites; 842 out of 988), and those of GS-1 to 392,000km2 (Fig 4: shaded areas; 74% of the sites; 415 out of 560). The mean estimate for the total population for GI-1d-a sums up to only roughly 5700 people (min. 3100, max. 9900) and is reduced to 3500 people (min. 1900, max. 6200) for GS-1. The mean density of GI-1d-a, 1.2 p/100 km² (min. 0.6, max. 2.0), in contrast, is only slightly lower than for dataset A, and remains at 0.9 p/100 km² for GS-1 (min. 0.5, max. 1.6). The largest regional population for dataset B estimates during GI-1d-a is found in northern Spain and the French Pyrenees with 1100 people, followed by Benelux and NW Germany with 820 people. With 770 people, NE Germany and Poland display estimates similar to those of the Czech Republic and SE Germany. For the Total Area of Calculation, the estimated population density drops in dataset B from 0.002 in GI-1d-a to 0.001 p/km2 in GS-1.
Differences of the estimates and diachronic tendencies between dataset A and B might indicate difficulties with the underlying dataset (chronological resolution, typological attribution, etc.). Therefore, to better understand the data presented above, we also highlight differences in the diachronic tendencies. Interestingly, while general tendencies from GI-1d-a to GS-1 of the regional estimates remain relatively stable across several northern and southern European regions, the results show a clear difference between dataset A and B for N France, with a much more pronounced demographic decline, and for the area of Poland and NE Germany where dataset B indicates a more pronounced increase in population size. For Switzerland and SW Germany, the tendency remains stable in dataset B (Fig 5, for details, see S5 Fig). For the Czech Republic and SE Germany, although the estimates are generally much lower (dataset A estimates around 460 people and dataset B only around 80 people), the estimated increase is higher within dataset B (S6 Fig), although the density for this area remains at 2.2 p/100 km² for GI-1d-a and 0.9 p/100 km² for GS-1, fully in line with the dataset A estimates.
Depicted are the mean estimates from dataset A (bright fill) and dataset B (dark fill).
A consistent signal in our palaeodemographic reconstructions is a strong decline in the meta-population from GI-1d-a to GS-1 (Fig 5). In Portugal, N and SE France, and S Scandinavia, likewise Czech Republic and SE Germany, the regional population may have dropped below a hundred people. Notable exceptions from this trend are Poland & NE Germany, and Italy (especially the northeastern area, compare Figs 4 and 5). Moreover, in all estimates, the former region hosts the largest regional population of around 1200 during GS-1, thus being almost twice as large as the second highest regional population (dataset A = 740; B = 590) located in (northern) Spain and the French Pyrenees. In Switzerland and SW Germany, the estimates from dataset B result in a relatively stable population (dataset A = 230, B = 260).
Discussion
Large-scale distribution patterns of Palaeolithic sites are influenced by several confounding factors that lie outside the decision-making processes of past hunter-gatherers. Many other parameters can influence the likelihood of site discovery, such as the geomorphological development of a landscape as shown, for instance, in the Great Adriatic Po-Plain [124] and other coastal areas [125]. Modern land use (e.g., forestry, vineyards, ploughed fields, quarries) or specific taphonomic settings (presence of caves and rock shelters or wetlands; loess or coversand deposition) can also play an important and sometimes even dominant role [126]. Other factors are the research histories of the different areas, in some cases linked to the proximity of universities and other institutes (research-driven versus development-led archaeological excavations) with strong prehistoric research traditions, the accessibility of collected data (such as regional or national databases, or grey literature) and their revision by experts (cf. [69,110,127,128]). Finally, differences in artefact size or the recognisability of iconic artefacts, such as the tanged points of the Ahrensburgian and Swiderian, can skew the observed patterning, either by making them harder to find (cf. [75,129]), or by skewing site numbers towards those characterised by particularly large artefacts (cf.[130]). These biases need to be discussed and necessitate caution, but they should not prevent research from addressing large-scale questions such as population dynamics in general [16].
The distribution of Final Palaeolithic sites in the study area (Fig 2) is notably uneven but comparable between GI-1d-a and GS-1 (Figs 3 and 4). Generally – and in sharp contrast to all periods of the preceding Upper Palaeolithic – the main area of the human population is not located in the Franco-Cantabrian region but in regions north of the Alps. Given that modern land use will not affect Upper and Final Palaeolithic sites differently and given further that the Late Glacial warming made higher-latitude areas newly accessible for Final Palaeolithic hunter-gatherers, it seems highly unlikely that this large-scale spatial pattern is exclusively the result of research biases [126]. At smaller scales, however, the effects of research biases might be more pronounced. For our dataset, this is particularly relevant for a conspicuous void that stretches in the south-west to north-east direction from southwestern France up to the Upper Rhine Valley, separating the investigated area into a south-western and north-eastern part. Final Palaeolithic sites are scarce in this void, but it is unclear whether this reflects prehistoric reality or results from multiple biases listed above. The latter view is supported by the absence of strong research traditions and academic structures in this region compares with, for instance, the southwest or the northeast of France and a relative scarcity of contexts favouring preservation and recognition of Upper and Final Palaeolithic sites. Difficulties in locating raw material sourcing sites and a general scarcity of typological markers (e.g., at Le Closeau) might additionally hamper the attribution of sites to the Final Palaeolithic in some areas. Also, modern population densities and industrial activity are either low or very high in some of the western areas of Germany – some sites are surely waiting to be discovered while others have long since been destroyed. The comparably long and strong archaeological research tradition here, however, speaks in favour of the existence of a void, at least in Germany (i.e., Hesse and western Thuringia). A comparison of the extent of the void with Core Areas calculated for the immediately preceding period of the Final Magdalenian (Fig 6, [123]), shows that the void widens considerably in the Final Palaeolithic towards the southwest and northeast. Core Areas are modelled in the void, at least for the French area, and also during earlier periods of the Upper Palaeolithic record, the early Gravettian and Solutrean technocomplexes [52,131]. It is difficult to see why potential biases would particularly affect Final Palaeolithic sites, leaving those of the Upper Palaeolithic in place. Also, differences in material culture on either side of the void (e.g., occurrences of painted cobbles), a lack of evidence for objects that were transported across it, and hints from palaeogenetic data warrant a closer examination of this observation.
16-14 ka BP, orange areas, after [123]) to Core Areas of the GI-1d-a (green areas). For details on background map, see Fig 3; dashed line: study area.
In order to assess biases by typological visibility, we looked at the number of radiocarbon-dated sites in both phases based on the entries in the Radiocarbon Palaeolithic Europe Database v.27 ([132]; we are aware of a later version but used v.27 when conducting this study). The analysis shows similar trends to our demographic estimates. For the entire map section and in absolute numbers, there is a decline of more than 50% in radiocarbon dated sites from GI-1d-a to GS-1 (S5 Table). Regionally differentiated and in relative terms (S7 Fig), we see a decline in all areas, but it is specifically pronounced in Central Europe. A fundamental difference, however, is our increasing estimates for NE Italy, Poland, and NE Germany (Fig 5). It is thus possible that our estimates for these regions are positively skewed due to greater typological recognisability of assemblages produced during GS-1. The estimated demographic increase, however, can also be the result of behavioural responses to contemporaneous climate change, such as population movements towards the east.
A comparison to other palaeodemographic studies might also reveal potential biases. Yet, in contrast to periods of the Upper Palaeolithic, such a comparison is difficult for the Final Palaeolithic estimates because of divergent chronological frameworks. For instance, the youngest period in the study by Bocquet-Appel et al. [133], is termed ‘Late Glacial’ and comprises the interval roughly between 19.1 and 13.3 ka BP. Tallavaara et al. [21] provide estimates in 1 ka steps, but the last two steps are centred at 14 and 13 ka BP, respectively, therefore essentially missing the Younger Dryas. While the former period is too large for a meaningful comparison, the latter provides two estimates at the fringes for our first phase and none for the second phase of GS-1. The chronologically matching study by French and Collins for southwestern France [20] also sees a decline in demographic signatures from GI-1d-a to GS-1. A revision of the Final Magdalenian assemblages in this area resulted in a reattribution to the Laborian, chronologically placing the sites to the GS-1 [134]. These findings and very recent research developments on the Laborian could change the demographic picture here. Rather more synoptically on a European scale, Ordonez and Riede [135] reconstruct forager population densities with reference to ethnographic data and climatic limiting factors as derived from high-resolution climate data. Their study similarly documents a population downturn during GS-1.
A large-scale shift to the east as a potential reaction to the GS-1 cooling
At a large spatial scale, a ~ 25% decrease in the size of the Core Areas from GI-1d-a to GS-1 is observable in most parts of the investigated area and across both datasets A (Fig 7) and B (S8 Fig). At the same time, Core Areas in western Britain, Brittany, and Denmark disappeared, whereas new Core Areas formed in eastern Poland, particularly in Masuria. Core Areas in Iberia and southwestern France shrank but remained rather stable in their geographic location [138]. Notably, in the rest of the investigated area, Core Areas showed a pronounced geographic shift towards the east during GS-1. Given that it seems unlikely that this large-scale shift is exclusively caused by biases (see discussion above), the question arises as to whether it can be seen as a reaction to changing climatic conditions. While other factors, such as technological change, varying subsistence strategies, or conflicts likely had effects on the population dynamics, we focus in the following on climate change as a long-term and large-scale factor, continuously affecting the meta-population in all regions of the investigated area.
For a comparison of dataset B Core Areas, see S8 Fig. Purple diamonds (after [136]) indicate documented volcanic ash (=tephra, microtephra or heavy minerals) from the Laacher See eruption, dated to 13 ky BP [137]. For details on background map, see Fig 3; dashed line: study area.
Here, a comparison to the developments during the Gravettian is instructive. Demographic studies, which compared the internal shift from an early phase (33-29 ka BP) to a late phase (29-25 ka BP) of the Gravettian is the only other instance since the arrival of anatomically modern humans in Europe where we see a pronounced demographic decline coinciding with climatic cooling [52]. During the late phase of the Gravettian, the estimates indicate a reduction across all regions, suggesting that no or only little population movement occurred over larger distances. This may indicate that a behavioural response to environmental change in terms of mobility did not happen or that population movement was not a sufficiently effective risk mitigation strategy, given that the climatic changes affected all of Europe [139]. Instead, it seems that the regional populations north of 50° latitude collapsed entirely, leaving western Central Europe largely depopulated until the Late Magdalenian [52,140]. This is in accord with palaeogenetic findings, indicating a continuity in western and southwestern Europe (Fournol cluster) and a disappearance of traits associated with the Věstonice cluster in Central and Southern Europe [24]. Interestingly – and in stark contrast to the Gravettian – the estimates for the Final Palaeolithic show a regionally more differentiated demographic pattern. In the following, we discuss the environmental change of the Younger Dryas in more detail, focusing on regional differences that might have contributed to the observed pattern.
The Younger Dryas (GS-1) is a phenomenon driven by processes in the North Atlantic and with partly varying effects on the ecosystems in different parts of Europe, often with a west-east-trending gradient (e.g., [35–38,42,110,141,142]). In northern Germany, western Belgium [143] and Poland [144], for instance, the climate of GS-1 drove the return of open landscapes, while the vegetation in Switzerland and eastern France seems to have been affected only weakly [145–147]. In a general comparison to the GI-1d-a, there is a sharp decline in winter temperatures with much colder temperatures between September and May, with strongly increased snow depth in north-western Europe during May, particularly in western Britain, Brittany, and Central France and a lesser extent in northern Central Europe [42].
As a consequence of this marked cooling, the thawing of the permafrost, ice-covered lakes, and other bodies of water was delayed. The period with stable warm summer temperatures was considerably shorter, restricted to about three months (June to August) in eastern Europe and to only two months (July to August) in north-western Europe (e.g., [41,148,149]). This led to a considerable decline in the length of the growing season, again with a pronounced west-east trend. While the regions along the Atlantic coast of France and Britain experienced a shortening of more than a month, Eastern Britain, Central France, Iberia, and Italy saw a reduction of 10-20 days, while northern Central Europe and the Balkans saw a decrease of between 2 and 10 days [42]. Shifts in the phenological patterns of the growing season can have considerable influence on the direction of migrations of herbivores [150] and might, over the course of centuries also influence the large-scale distribution of hunter-gatherer populations [23,151]. Interestingly, isotope analyses on reindeer teeth from the Hamburgian and Ahrensburgian layer at Stellmoor (northern Germany) suggest that reindeer herds generally migrated in an east-west direction on the North European Plain during the Younger Dryas [152], although these patterns could have been more complex in detail [153], cf. [39]. Another highly relevant phenomenon is a general shift towards dryer conditions during the Younger Dryas, which probably led to the lowering of the groundwater table and, thus, reduced water availability in some areas [154,155]. This phenomenon is particularly well documented in the Scheldt basin in north-western Belgium. Here, a lowering of the groundwater table probably led to a disappearance of numerous shallow lakes and ponds in the interior of the Lower Scheldt basin, while the neighbouring Meuse basin seems to have been less affected [142]. In addition, the lowlands of the North European Plain were affected by intensified winds forming cover sand dunes [156,157]. Lower groundwater tables, sand cover in which water seeps away easily and longer periods with ice-cover of the remaining water bodies might have posed challenges to animals and humans alike. Yet, the cover increased the in situ preservation of sites but also reduced their archaeological visibility [158].
In sum, the Younger Dryas has precipitated a swath of largely unfavourable environmental changes for humans. These changes were more pronounced closer to the Atlantic seaboard and less so towards the east. This is in line with our observation of an eastward shift of the Core Areas in north-western and Central Europe. While populations declined in most areas, there is a clear increase in NE Germany and Poland as well as stable estimates for Switzerland and SW Germany in the scenario based on dataset A. At the same time, the estimates for SE Germany and the Czech Republic show the sharpest decline of all regions in both scenarios, dataset A and B. In S Scandinavia, i.e., Denmark, the population dropped almost to zero, with only a few camps appearing in the East. Considering the overall decline of regional populations, independent population growth in NE Italy, NE Germany, and Poland seems unlikely. Therefore, these regional population dynamics may be seen as first evidence of large-scale population movement as a response to climatic cooling. If the population shift represents a prehistoric reality, then the source of the movement into NE Germany and Poland would have mainly been SE Germany and the Czech Republic, with some people moving towards Switzerland but the majority moving towards Poland. The sink areas of Poland, NE and NW Germany also could have had an influx from Denmark. NE Italy might have received an influx from SE France or western Italy, although there is a lack of archaeological evidence for contact in this direction [159]. Possible source areas are also regions at the margins or outside the investigated area, such as central-southern Italy, the western Balkans, or now submerged coastal areas.
Finally, environmental events beyond climate trends may also have played a role in shaping human responses at this time. The spatial congruence between the void area across Central Europe and the most massive ash fallout of the cataclysmic Laacher See eruption that occurred at 13,006 ± 9 cal BP [137] in the Eifel volcanic zone is notable, especially to the south-east and north-east of the eruptive centre (Fig 7). The eruption occurred at the tail end of the cool GI-1b which likely stressed contemporaneous forager populations. In Belgium and Central Germany, several sites evidence stratigraphic relations where Federmesser-Gruppen occupation is capped by Laacher See fallout, terminating these occupations [160,161]. The short- and medium-term environmental impacts of the eruption were varied and, depending on local fallout thickness, likely considerable [162–166]. These may have depressed population in the affected regions already prior to the onset of GS-1 and may have resulted in sustained legacies of landscape avoidance that lasted into GS-1.
Demographic dynamics, minimal viable population, and social networks
Depending on whether one favours the mean estimate of the total population during GI-1d-a from dataset A (8100, also including generically assigned sites) or the one from dataset B (5700, using only chronologically subdivided sites) as more reliable, there is either an increase (dataset A) or decrease (dataset B) after the Final Magdalenian, which produced a mean estimate of 7300 (min. 4500, max. 10200) people [123]. However, some northern regions were frequently occupied for the first time after the Last Glacial Maximum (Fig 6), suggesting a stable metapopulation if not an increase. Less ambiguous is the subsequent decline in GS-1 to 4300 and 3500 people for the estimates based on datasets A and B, respectively. In both scenarios, the population is reduced by roughly half. This marked decline raises questions about the viability of the remaining population and the consequences for regional as well as long-distance social networks.
Defining what constitutes minimum viable populations in humans is difficult. A lower threshold of ~ 1500 individuals, comprising around 770 individuals in the reproductive age cohort, has been suggested [167–169]. While the effects of drift-related inbreeding can be avoided at even lower numbers of 100 to 500 people [168,170], such minuscule populations become highly sensitive to negative stochastic effects, especially in high-risk environments [171]. Despite being relatively low, our estimates are in the range of viable populations, including, notably, the estimates for GS-1, which are still high compared to those of most other periods of the Upper Palaeolithic [131,172]. This difference is particularly striking regarding the Gravettian, where mean estimated population sizes dropped from 2800 people during the early phase to only 1000 during the late phase [52].
The estimated population decline likely also affected the size, structure, and connectivity of social networks. On the one hand, a reduction in the number of people who can maintain social contacts is likely to lead to distortions in the networks. Here, networks at regional scales are usually maintained by more people than networks at larger scales, which are thus more prone to disturbances. On the other hand, actively increasing the extent and connectivity of networks can be a way to mitigate environmental risk and population decline [173] – as long as the increasing physical distances can be bridged (cf. [174–176]). Social networks exist at different spatial scales, and besides palaeogenetic studies [24], they usually become archaeologically visible through different classes of objects circulating within them that have characteristic transport distances and patterns [177]. For networks at a regional scale, lithic raw materials are probably the most important source of information [178]. The materials usually transported in economically significant numbers and over distances well below 300 km are here considered likely to be acquired by embedded procurement [179,180] and thus by members of the same regional group that also discarded them (but see [181]). In the available raw material data, there is indeed an increase in the size of the catchments during the Final Palaeolithic, very pronounced in all areas of the study region except for the NW European area (see RMCA median values in Tables 1 and 2; S1 and S2 Tables).
For large-scale networks (connecting two or more regional networks) to become visible, transport patterns of objects with a presumably different acquisition mode are better suited. Ideally, these objects are not procured directly by the group that discards them but acquired via exchange between groups. Here, sub-recent and fossil mollusc shells seem to be highly informative. In contrast to lithic raw materials, they are regularly transported over more than 600 km and during the Late Upper Palaeolithic, they trace a network that spans between the Atlantic and Mediterranean coast, the Paris Basin up the Rhine-Meuse Area and across to the Swabian Jura [182–185]. They thus also span across the area of the central void in the Final Palaeolithic dataset (Fig 7). For the Final Palaeolithic itself, however, information on transported mollusc shells is extremely rare. There are singular instances of long-distance transports of lithic raw materials (e.g., obsidian found in Poland), yet there is no available evidence that systematically reflects social networks at that scale. The absence of evidence for regular long-distance transports of objects in contrast to the Late Upper Palaeolithic may signal reduced contact at that scale. By this token, the extensive void in the centre of the investigated area might also be an expression of reduced social connectivity between hunter-gatherers in southwestern France and Iberia on the one hand and in Northern and Central Europe on the other. The latter area can possibly be further divided by mid-scale networks through the distribution of raw materials such as red Helgoland flint [54] or the use of amber [186,187] or by differentiated symbolic use [98,187]. If there had been decreasing contact between the south-western and north-eastern parts of the investigated area during GI-1d-a, the marked population decrease in GS-1 probably would have additionally distorted the networks.
Recent palaeogenetic evidence suggests a differential development in both regions during the Final Palaeolithic. There appears to be a probably south-north directed movement of people during GI-1d-a, passing west of the Alps from Italy into Central Europe and the British Isles, whereas the population in Iberia shows more pronounced genetic continuity with several admixture events [24,68]. Whether this suggested south-north dispersal is consilient with the proposed northward dispersal of backed points from south-eastern France along the Rhône and Saône axis and alongside the disappearance of reindeer, as often emphasised by Thévenin [46,188,189], is beyond the scope of this paper. However, these proposals have mainly been based on typological analyses carried out on assemblages whose archaeostratigraphic reliability is difficult to prove. Furthermore, given the available data, the Azilian traditions appear after the Magdalenian, and some can be considered as a final episode of these traditions [4,92,190]. Other, more recent traditions of the Azilian (GI-1) are part of their extension and of a common evolution of technical traditions on a European scale. The situation is perhaps more complex at the very end of the Palaeolithic (GS-1) with the existence of a certain duality of technical traditions in France. The presence of assemblages with characteristics close to those known from the Laborian and Epigravettian periods has been proposed [85,94,97,191]. However, the chronological framework of these phenomena remains to be defined and the techno-economic data needs to be improved. At the moment, identifying this large empty area with very little to no evidence being assignable to the Final Palaeolithic calls for future research.
Conclusion
Our study of the Final Palaeolithic in Southern, Western, Central and Northern Europe provides new insights into the overall as well as regionally differentiated population dynamics of hunter-gatherers groups. Based on the population estimates derived using the Cologne Protocol we can state that:
From the Final Magdalenian to GI-1d-a:
- The estimates of the total population size increase only slightly across the study area, while the Core Areas expanded significantly;
- distances of raw material transport became longer, as reflected in larger raw material catchment areas.
During the GI-1d-a:
- An unprecedented regional dominance of human presence is evident for the western and central European lowlands;
- a void appears in central France, extending to the Upper Rhine.
From GI-1d-a to GS-1:
- The total population decreases by half, but regional dynamics do vary considerably;
- the size of Core Areas is reduced, while the average size of raw material catchment areas continues to increase;
- There is a continuation and regionally even slight increase in the extent of the void;
- regionally different responses to the climatic cooling become evident in the pattern of Core Areas, with an eastward shift being the dominant response of populations in Northern Europe and parts of Central Europe.
During the GI-1d-a, the higher-latitude areas of Europe – parts of which were previously unoccupied or only ephemerally visited by anatomically modern humans – became more regularly and frequently visited by contemporaneous hunter-gatherer groups. At this time, the western and central European lowlands became a hotspot of population dynamics for the first time. The subsequent climatic cooling during GS-1 (Younger Dryas) led to a marked population decline but also different regional responses. Populations in the Iberian Peninsula and southwestern France shrank while staying in the same area, a response comparable to the one observed during the Gravettian. Populations in Northern and Central Europe, in contrast, reacted with a pronounced shift towards the east. While populations declined in the western part, they increased in the eastern part. We argue this was not because of population growth but because of population movement from west to east. The same phenomenon can be observed for the area of south-eastern France up to north-eastern Italy, where the former region also showed a population decline, while the latter saw a population increase, again possibly due to population influx. The shift of populations towards the east was most probably a reaction to deteriorating environmental conditions brought about by the Younger Dryas, including longer winters, higher snow cover and a strongly reduced growing season. These changes affected the western parts of Europe particularly hard, especially the Atlantic coast, and became more and more buffered towards the east.
The question thus arises as to why the two sub-populations in the southwestern part and the rest of the investigated area reacted so differently. Answers to this question are not readily evident. Maybe the two sub-populations had already become largely decoupled during the Allerød, as might be indicated by the central void area appearing in the archaeological record after the Magdalenian. Maybe hunter-gatherers in the topographically rather uniform lowlands had a different attitude towards and aptitude for long-distance movements or a more permeable territorial structure. The possibility that some behavioural or technological innovation – using poison for efficiently killing prey, snowshoes, skis, travois or the like – transformed their ability to move longer distances at lower costs should not be discounted. Future studies are needed to illuminate these aspects. To further refine our palaeodemographic estimates, a higher chronological resolution of the dataset will likely reveal currently invisible patterns, which in combination with high-resolution climate models and palaeoenvironmental studies will allow for a more differentiated view of the biocultural processes that occurred at the end of the Pleistocene. In-depth studies of material culture change and a better understanding of transport patterns of raw materials and objects of far-distance transports could provide data on the size and connectivity of social networks, adding detail to the coarse picture and possibly identifying non-environmental explanations for observed patterns of population dynamics.
Supporting information
S1 Table. Sites with data on Raw Material Catchment Areas (RMCA) assigned to GI-1d-a.
The table comprises only data that were considered for calculation in this study (n = 53). Below each region, the corresponding quartiles (Q1, Q2, Q3) used to calculate the palaeodemographic estimates are shown (see table 1).
https://doi.org/10.1371/journal.pone.0310942.s002
(DOCX)
S2 Table. Sites with data on Raw Material Catchment Areas (RMCA) assigned to GS-1.
The table comprises only data that were considered for calculation in this study (n = 39). Below each region, the corresponding quartiles (Q1, Q2, Q3) used to calculate the palaeodemographic estimates are shown (see table 2).
https://doi.org/10.1371/journal.pone.0310942.s003
(DOCX)
S3 Table. Regionally distinguished demographic estimates for GI-1d-a based on dataset B (i.e., only directly dated or specifically attributable to GI-1d-a).
Thin horizontal lines indicate regions with transfer of RMCA data (cf. S3 Fig). ODI = optimally describing Isoline in km²; Q = Quartile [1 = lower, 2 = median, 3 = upper]; RMCA (km²) = raw material catchment area in km²; Nraw = number of RMCAs; Ngroups = number of groups; Npersons = number of persons; Dpopulation = population density within Core Areas as p/km2, and the Total Area of Calculation (bottom rows).
https://doi.org/10.1371/journal.pone.0310942.s004
(DOCX)
S4 Table. Regionally distinguished demographic estimates for GS-1 based on dataset B (i.e., only directly dated or specifically attributable to GS-1).
Thin horizontal lines indicate regions with transfer of RMCA data (cf. S4 Fig). For explanation on abbreviations see S2 Table.
https://doi.org/10.1371/journal.pone.0310942.s005
(DOCX)
S5 Table. Number of radiocarbon-dated sites divided by GI-1d-a and GS-1.
The numbers are given for the entire database (“all radiocarbon dates”, upper), for the Total Area of Calculation (middle) and for each region (lower). Dates were taken from the Radiocarbon Palaeolithic Europe Database v.27 [49], see details in main publication. Except for the first dataset (n = 1758), radiocarbon dates per site and time-bin were counted as one occurrence each (n = 731). To account for the slightly shorter phase of the GS-1 (1.2 ky) compared to GI-1d-a (1.3 ky), the GS-1 radiocarbon counts were normalised for the duration. Note that the regions slightly differ from the ones defined in Fig 1 (that is for “Benelux and W Germany” and Switzerland).
https://doi.org/10.1371/journal.pone.0310942.s006
(DOCX)
S1 Fig. Maps (left) and diagrams (right) from the definition of the Core Areas based on the sites dated to Greenland Interstadial (GI) 1d-a. The maps of dataset A (upper, sites n = 1356) and dataset B (lower, n = 988) include dashed lines which divide areas for which Core Areas (green areas) – areas of equal site densities – were calculated separately (see [ 48 ]), the dotted line delimits the Total Area of Calculation.
Regional areal increase of isolines (right) was interpolated by using distance values of Largest Empty Circles for GI-1d-a (right). X-axis of diagrams display radius of Largest Empty Circle, y-axis areal increase (km2) encircled by the Isoline. To define the Optimal Describing Isoline (ODI, arrows in diagrams), the first peak or plateau, encircling around 70% or more of all sites, is selected. Core Areas of dataset A comprises 90% of the sites, Core Areas of dataset B 85%.
https://doi.org/10.1371/journal.pone.0310942.s007
(TIF)
S2 Fig. Maps (left) and diagrams (right) from the definition of the Core Areas based on the sites dated to Greenland Stadial (GS) 1. The maps of dataset A (upper, sites n = 1003) and dataset B (lower, n = 560) include dashed lines which divide areas for which Core Areas (blue areas) – areas of equal site densities – were calculated separately (see [48]).
Dataset A comprises 82% of the sites, Core Areas of dataset B 74%. For explanation on details see S1 Fig.
https://doi.org/10.1371/journal.pone.0310942.s008
(TIF)
S3 Fig. Map of all Raw Material Catchment Areas (RMCA) from sites dated to GI-1d-a that were collected from the literature (n = 108). Several RMCAs were excluded from further analysis (see explanation in main article and remaining sites [n = 53] in table S1).
Thick dashed lines delimit regions within which the same RMCA data was used to calculate the demographic estimates. Background: Core Areas and sites from dataset A.
https://doi.org/10.1371/journal.pone.0310942.s009
(TIF)
S4 Fig. Map of all Raw Material Catchment Areas (RMCA) from sites dated to GS-1 that were collected from the literature (n = 97). Several RMCAs were excluded from further analysis (see explanation in main article and remaining sites [n = 39] in S2 Table).
Thick dashed lines delimit regions within which the same RMCA data was used to calculate the demographic estimates. Background: Core Areas and sites from dataset A.
https://doi.org/10.1371/journal.pone.0310942.s010
(TIF)
S5 Fig. GI-1d-a: Comparison of estimates of people for dataset A (left bar) and B (right bar) per region. The largest discrepancies occur in areas of NE, SE and SW Germany and adjacent regions.
https://doi.org/10.1371/journal.pone.0310942.s011
(TIF)
S6 Fig. GS-1: Comparison of estimates of people for dataset A (left bar) and B (right bar) per region. Discrepancies occur in areas the Czech Republic and SE Germany.
https://doi.org/10.1371/journal.pone.0310942.s012
(TIF)
S7 Fig. Diachronic tendencies in the frequency of regionally differentiated radiocarbon dates per region. The percentage of dates for GI-1d-a (green dot, left side of the arrow) and GS-1 (blue dot, right) are displayed for each region.
Dates were taken from the Radiocarbon Palaeolithic Europe Database v.27 [132]. We counted each time-bin covered by radiocarbon date(s) from an archaeological site as one occurrence (see S3 Table). To account for the slightly shorter phase of the GS-1 compared to GI-1d-a, the GS-1 radiocarbon counts were normalised for the duration of the phase.
https://doi.org/10.1371/journal.pone.0310942.s013
(TIF)
S8 Fig. Comparison of Final Palaeolithic Core Areas (data B) from GI-1d-a (green areas) to GS-1 (blue areas). For a comparison of data A Core Areas, see Fig 7. Grey diamonds indicate documented (micro)tephra from the Laacher See eruption, dated to 13 ky BP.
https://doi.org/10.1371/journal.pone.0310942.s014
(TIF)
Acknowledgments
The authors would like to thank Andrea Darida, Martin Müller and Nina Avci for their help in compiling the database and Dennis Batz for editing the references. We also like to thank the editor, Przemysław Mroczek, and Małgorzata Kot and three other reviewers for their helpful and insightful comments, which improved the quality of the paper. All authors revised and added new data on sites and assemblages for regions of their expertise. Felix Riede, Katja Winkler, Florian Sauer and Birgit Gehlen provided regional databases at an early stage of the project, that were incorporated after adjustment to the specific requirements. The authors are solely responsible for the content of this publication.
References
- 1. Rasmussen SO, Bigler M, Blockley SP, Blunier T, Buchardt SL, Clausen HB, et al. A stratigraphic framework for abrupt climatic changes during the Last Glacial period based on three synchronized Greenland ice-core records: refining and extending the INTIMATE event stratigraphy. Quat Sci Rev. 2014;106:14–28.
- 2. Guthrie RD. Origin and causes of the mammoth steppe: a story of cloud cover, woolly mammal tooth pits, buckles, and inside-out Beringia. Quat Sci Rev. 2001;20(1–3):549–74.
- 3. Drucker DG. The Isotopic Ecology of the Mammoth Steppe. Annu Rev Earth Planet Sci. 2022;50(1):395–418.
- 4.
Grimm SB. Resilience and reorganisation of social systems during the Weichselian lateglacial in North-West Europe: an evaluation of the archaeological, climatic, and environmental record. Mainz, Regensburg: Verlag des Römisch-Germanischen Zentralmuseums, Schnell & Steiner; 2019.
- 5. Jacquier J, Mevel L, Bodu P. Less simple than expected. New technological and functional insights on the Late Azilian site of Le Closeau (France). Quartär. 2021;68(1):181–214.
- 6.
Rozoy J-G. Les Derniers Chasseurs: L’Épipaléolithique en France et en Belgique: Essai de Synthèse. Charleville: Champenoise; 1978.
- 7.
Terberger T, Eriksen BV, editors. Hunters in a changing world: environment and archaeology of the Pleistocene - Holocene Transition (ca. 11000-9000 B.C.) in Northern Central Europe; workshop of the U.I.S.P.P.-Commission XXXII at Greifswald in September 2002. Rahden/Westfalen: Verlag Marie Leidorf; 2004.
- 8.
Vollbrecht J. Spätpaläolithische Besiedlungsspuren aus Reichwalde. 2005.
- 9. Bodu P, Debout G, Bignon O. Variabilité des habitudes tardiglaciaires dans le Bassin parisien : l’organisation spatiale et sociale de l’Azilien ancien du Closeau. Bull Soc Préhist Fr. 2006;103(4):711–28.
- 10.
Bolus M. Die Siedlungsbefunde des späteiszeitlichen Fundplatzes Niederbieber (Stadt Neuwied): Ausgrabungen 1981–1988. Bonn: Habelt; 1992.
- 11.
Bosinski G. Upper and Final Palaeolithic settlement patterns in the Rhineland, West Germany. In: Dibble HL, Montet-White A, editors. Upper Pleistocene Prehistory of Western Eurasia. Philadelphia: The University Museum, University of Pennsylvania; 1988. pp. 375–86.
- 12.
Terberger T. Die Siedlungsbefunde des Magdalénien-Fundplatzes Gönnersdorf: Konzentrationen III und IV. Stuttgart: Steiner; 1997.
- 13. Fasser N, Visentin D, Duches R, Peresani M, Fontana F. Lithic projectile technology in the western Late Epigravettian: The case study of north-eastern Italy. Quat Int. 2024;694:70–90.
- 14.
Fat Cheung C. L’Azilien pyrénéen parmi les sociétés du Tardiglaciaire ouest-européen: apport de l’étude des industries lithiques. Archéologie et Préhistoire. Ph.D., Université Toulouse le Mirail - Toulouse II. 2015.
- 15.
Fischer A. Pioneers in deglaciated landscapes: the expansion and adaptation of late Palaeolithic societies in Southern Scandinavia. In: Barton N, Roberts AJ, Roe DA, editors. The late Glacial in north-west Europe: human adaptation and environmental change at the end of the Pleistocene. London: Council for British Archaeology; 1991. pp. 100–121.
- 16. Riede F, Matzig DN, Biard M, Crombé P, de Pablo JF-L, Fontana F, et al. A quantitative analysis of Final Palaeolithic/earliest Mesolithic cultural taxonomy and evolution in Europe. PLoS One. 2024;19(3):e0299512. pmid:38466685
- 17.
Valentin B. Les groupes humains et leurs traditions au Tardiglaciaire dans le Bassin parisien. Apports de la technologie lithique comparée. Ph.D., Université Panthéon-Sorbonne. 1995.
- 18.
Valentin B. L’usage de la pierre tendre pour la débitage des lames. Circonstances de quelques innovations au cours du Tardiglaciaire dans le Bassin parisien. In: Pion G, editor. Le paléolithique supérieur récent: nouvelles données sur le peuplement et l’environnement Actes de la Table ronde de Chambéry. Paris: Société Préhistorique Française; 2000. pp. 253–260.
- 19. Valentin B, Fosse G, Billard C. Aspects et rythmes de l’azilianisation dans le Bassin parisien: caractérisation de l’industrie lithique recueillie au Cornet (locus 33) à Ambenay, Eure. Gallia Préhist. 2004;46:171–209.
- 20. French JC, Collins C. Upper Palaeolithic population histories of Southwestern France: a comparison of the demographic signatures of 14C date distributions and archaeological site counts. J Archaeol Sci. 2015;55:122–34.
- 21. Tallavaara M, Luoto M, Korhonen N, Järvinen H, Seppä H. Human population dynamics in Europe over the Last Glacial Maximum. Proc Natl Acad Sci U S A. 2015;112(27):8232–7. pmid:26100880
- 22.
Schmidt I, Gehlen B, Zimmermann A. Population estimates for the Final Palaeolithic (14,000 to 11,600 years cal. BP) of Europe - challenging evidence and methodological limitations. In: Mevel L, Weber M-J, Maier A, editors. En Mouvement/ On the Move/ In Bewegung. Paris: Société Préhistorique Française; 2021. pp. 221–237.
- 23. Maier A, Ludwig P, Zimmermann A, Schmidt I. The sunny side of the ice age: Solar insolation as a potential long-term pacemaker for demographic developments in Europe between 43 and 15 ka ago. PaleoAnthropology. 2022;1(1):35–51.
- 24. Posth C, Yu H, Ghalichi A, Rougier H, Crevecoeur I, Huang Y, et al. Palaeogenomics of Upper Palaeolithic to Neolithic European hunter-gatherers. Nature. 2023;615(7950):117–26. pmid:36859578
- 25. Adolphi F, Muscheler R, Friedrich M, Güttler D, Wacker L, Talamo S, et al. Radiocarbon calibration uncertainties during the last deglaciation: Insights from new floating tree-ring chronologies. Quat Sci Rev. 2017;170:98–108.
- 26. Reimer PJ, Austin WEN, Bard E, Bayliss A, Blackwell PG, Bronk Ramsey C, et al. The IntCal20 Northern Hemisphere Radiocarbon Age Calibration Curve (0–55 cal kBP). Radiocarbon. 2020;62:725–57.
- 27. Deschamps P, Durand N, Bard E, Hamelin B, Camoin G, Thomas AL, et al. Ice-sheet collapse and sea-level rise at the Bølling warming 14,600 years ago. Nature. 2012;483(7391):559–64. pmid:22460900
- 28. Rasmussen TL, Thomsen E, Moros M. North Atlantic warming during Dansgaard-Oeschger events synchronous with Antarctic warming and out-of-phase with Greenland climate. Sci Rep. 2016;6:20535. pmid:26847384
- 29. Naughton F, Costas S, Gomes SD, Desprat S, Rodrigues T, Goñi M, et al. Coupled ocean and atmospheric changes during Greenland stadial 1 in southwestern Europe. Quat Sci Rev. 2019;212:108–20.
- 30. Landais A, Capron E, Masson-Delmotte V, Toucanne S, Rhodes R, Popp T, et al. Ice core evidence for decoupling between midlatitude atmospheric water cycle and Greenland temperature during the last deglaciation. Clim Past. 2018;14:1405–15.
- 31. Vescovi E, Ravazzi C, Arpenti E, Finsinger W, Pini R, Valsecchi V, et al. Interactions between climate and vegetation during the Lateglacial period as recorded by lake and mire sediment archives in Northern Italy and Southern Switzerland. Quat Sci Rev. 2007;26:1650–69.
- 32. Samartin S, Heiri O, Lotter AF, Tinner W. Climate warming and vegetation response after Heinrich event 1 (16 700–16 000 cal yr BP) in Europe south of the Alps. Climate of the Past. 2012;8:1913–27.
- 33. Lemanik A, Baca M, Wertz K, Socha P, Popović D, Tomek T, et al. The impact of major warming at 14.7 ka on environmental changes and activity of Final Palaeolithic hunters at a local scale (Orawa-Nowy Targ Basin, Western Carpathians, Poland). Archaeol Anthropol Sci. 2020;12:66.
- 34. Reade H, Grimm SB, Tripp JA, Neruda P, Nerudová Z, Roblíčková M, et al. Magdalenian and Epimagdalenian chronology and palaeoenvironments at Kůlna Cave, Moravia, Czech Republic. Archaeol Anthropol Sci. 2021;13(1):4. pmid:33365102
- 35. Hoek W. Vegetation response to the ∼14.7 and ∼11.5 ka cal. BP climate transitions: is vegetation lagging climate?. Glob Planet Change. 2001;30:103–15.
- 36. Ammann B, Van Leeuwen JFN, Van Der Knaap WO, Lischke H, Heiri O, Tinner W. Vegetation responses to rapid warming and to minor climatic fluctuations during the Late-Glacial Interstadial (GI-1) at Gerzensee (Switzerland). Palaeogeogr Palaeoclimatol Palaeoecol. 2013;391:40–59.
- 37. Krüger S, Mortensen MF, Dörfler W. Sequence completed – palynological investigations on Lateglacial/Early Holocene environmental changes recorded in sequentially laminated lacustrine sediments of the Nahe palaeolake in Schleswig-Holstein, Germany. Rev Palaeobot Palyn. 2020;280:104271.
- 38. Rea BR, Pellitero R, Spagnolo M, Hughes P, Ivy-Ochs S, Renssen H, et al. Atmospheric circulation over Europe during the Younger Dryas. Sci Adv. 2020;6(50):eaba4844. pmid:33310841
- 39. Krüger S. Of birches, smoke and reindeer dung – tracing human-environmental interactions palynologically in sediments from the Nahe palaeolake. J Archaeol Sci Rep. 2020;32:102370.
- 40. Lane CS, Brauer A, Blockley SPE, Dulski P. Volcanic ash reveals time-transgressive abrupt climate change during the Younger Dryas. Geology. 2013;41:1251–4.
- 41. Schenk F, Väliranta M, Muschitiello F, Tarasov L, Heikkilä M, Björck S, et al. Warm summers during the Younger Dryas cold reversal. Nat Commun. 2018;9(1):1634. pmid:29691388
- 42. Schenk F, Wohlfarth B. The imprint of hemispheric-scale climate transitions on the European climate during the last deglaciation (15.5 ka to 9 ka BP). 2019. Report No.: AB 2019-01. Available: https://www.skb.com/publication/2492488/
- 43. Grimm SB, Weber M-J. The chronological framework of the Hamburgian in the light of old and new 14C dates. Quartär. 2008;55(2):17–40.
- 44. Jacobi RM, Higham TFG. The early Lateglacial re-colonization of Britain: new radiocarbon evidence from Gough’s Cave, southwest England. Quat Sci Rev. 2009;28(19–20):1895–913.
- 45. Pedersen JB, Assmann JJ, Normand S, Karger DN, Riede F. Climate Niche Modeling Reveals the Fate of Pioneering Late Pleistocene Populations in Northern Europe. Current Anthropology. 2023;64(5):599–608.
- 46. Thévenin A. L’“Azilien” et les cultures à pointes à dos courbe : esquisse géographique et chronologique. bspf. 1997;94(3):393–411.
- 47. Holzkämper J, Maier A, Richter J. “Dark Ages” illuminated – Rietberg and related assemblages possibly reducing the hiatus between the Upper and Late Palaeolithic in Westphalia. Quartär. 2013:115–36.
- 48. Schmidt I, Hilpert J, Kretschmer I, Peters R, Broich M, Schiesberg S, et al. Approaching prehistoric demography: proxies, scales and scope of the Cologne Protocol in European contexts. Phil Trans R Soc B 2021;376(1816):20190714. pmid:33250025
- 49. Schmidt I, Gehlen B, Winkler K, Arrizabalaga A, Arts N, Bicho N, et al. Database for the Final Palaeolithic archaeological record in western, southern and central Europe. Zenodo 2024.
- 50. Iversen J. The late-glacial flora of Denmark and its relation to climate and soil. Danmarks Geologiske Undersøgelse II Række. 1954;80(80):87–119.
- 51. North Greenland Ice Core Project members. High-resolution record of Northern Hemisphere climate extending into the last interglacial period. Nature. 2004;431(7005):147–51. pmid:15356621
- 52. Maier A, Zimmermann A. Populations headed south? The Gravettian from a palaeodemographic point of view. Antiquity. 2017;91:573–88.
- 53. Beuker J, Niekus JLT. Het gebruik van rode Helgoland-vuursteen in het laat-paleolithicum. Nieuwe Drentse Volksalmanak. 2003;120:112–20.
- 54. Terberger T, Schneider J, Fries JE. The site of Edewecht 82-East, district of Ammerland, and the role of Red Heligoland flint in the Late Glacial. Praehist Zeitschr. 2020;95(1):1–16.
- 55. Baales M, Heuschen W, Kehl M, Manz A, Nolde N, Riemenschneider D, et al. Western visitors at the Blätterhöhle (city of Hagen, southern Westphalia) during the Younger Dryas? A new final palaeolithic assemblage type in western Germany. PLoS One. 2023;18(5):e0284479. pmid:37134047
- 56.
De Bie M, Caspar J-P. Rekem: A Federmesser camp on the Meuse River bank: Archeologie in Vlaanderen 3. Leuven: Leuven University Press; 2000.
- 57.
Grimm SB, Weber M-J, Mevel L, Sobkowiak-Tabaka I, editors. From the Atlantic to Beyond the Bug River: Finding and Defining the Federmesser-Gruppen/ Azilian: Proceedings of Session A5b (Commission »The Final Palaeolithic of Northern Eurasia«) of the XVIIth UISPP Congress: 2014 Sep 1-7; Burgos: Spain. Heidelberg: Propylaeum; 2020. https://doi.org/10.11588/PROPYLAEUM.575
- 58.
Grimm SB, Eriksen BV, Krüger S, Reuter T, Wild M, Weber M-J. Late Glacial occupation of northern Germany and adjacent areas: Revisiting the archives. In: Gaudzinski-Windheuser S, Jöris O, editors. The Beef behind all possible pasts The tandem-Festschrift in honour of Elaine Turner and Martin Street. Heidelberg: Propylaeum; 2021. pp. 433–457. Available: https://books.ub.uni-heidelberg.de/propylaeum/catalog/book/950/c12572
- 59.
Sobkowiak-Tabaka I, Diachenko A. Approaching Late Paleolithic Seasonal Mobility on the North European Plain. In: Sobkowiak-Tabaka I, Diachenko A, Wiśniewski A, editors. Quantifying Stone Age Mobility: Scales and Parameters. Cham: Springer International Publishing; 2022. pp. 97–128. https://doi.org/10.1007/978-3-030-94368-4
- 60.
Sobkowiak-Tabaka I. Społeczności późnego paleolitu w dorzeczu Odry. Poznań: Instytut Archeologii i Etnologii Polskiej Akademii Nauk; 2011.
- 61.
Sobkowiak-Tabaka I. Living on the edge, or how resilient people settled the North. In: Djindjian F, editor. Les sociétés humaines face aux changements climatiques. Oxford: Archaeopress Archaeology; 2022. pp. 297–319.
- 62. Weber M-J, Clausen I, Housley RA, Miller CE, Riede F, Usinger H. New information on the Havelte Group site Ahrenshöft LA 58 D (Nordfriesland, Germany) - Preliminary results of the 2008 fieldwork. Quartär. 2010;57(1):7–24.
- 63.
Holzkämper J, Maier A. Neue Ergebnisse zur jungpaläolithischen Besiedlung Nordrhein-Westfalens. 25 Jahre Archäologie im Rheinland 1987 - 2011. Stuttgart: Theiss; 2012. pp. 58–60.
- 64. Barton RNE. Regional and chronological patterns in lithic raw material behaviour during the late glacial and some implications for the British Later Upper Palaeolithic. Lithics. 2010;31:33–43.
- 65.
Jacobi R, Higham T. The Later Upper Palaeolithic Recolonisation of Britain: New Results from AMS Radiocarbon Dating. The Ancient Human Occupation of Britain. Amsterdam: Elsevier; 2011. pp. 223–247. https://doi.org/10.1016/B978-0-444-53597-9.00012-1
- 66. Pedersen JB, Maier A, Riede F. A punctuated model for the colonisation of the Late Glacial margins of northern Europe by Hamburgian hunter-gatherers. Quartär. 2018;65(4):85–104.
- 67. Brovkin V, Brook E, Williams JW, Bathiany S, Lenton TM, Barton M, et al. Past abrupt changes, tipping points and cascading impacts in the Earth system. Nat Geosci. 2021;14(7):550–8.
- 68. Charlton S, Brace S, Hajdinjak M, Kearney R, Booth T, Reade H, et al. Dual ancestries and ecologies of the Late Glacial Palaeolithic in Britain. Nat Ecol Evol. 2022;6(11):1658–68. pmid:36280785
- 69.
Winkler K. Ahrensburgien und Swiderien im mittleren Oderraum: Technologische und typologische Untersuchungen an Silexartefakten der Jüngeren Dryaszeit. Kiel: Wachholtz Verlag; 2019.
- 70. Stefański D, Wilczyński J. Extralocal raw materials in the Swiderian Culture: Case study of Kraków-Bieżanów sites. Anthropologie. 2012;50:427–42.
- 71. Weber M-J, Grimm SB, Baales M. Between warm and cold: Impact of the Younger Dryas on human behavior in Central Europe. Quat Int. 2011;242:277–301.
- 72.
Grimm SB, Groß D, Gerken K, Weber M-J. On the onset of the Early Mesolithic on the North German Plain. In: Zander A, Gehlen B, Gramsch B, editors. From the Early Preboreal to the Subboreal period – Current Mesolithic research in Europe: studies in honour of Bernhard Gramsch. Kerpen-Loogh: Welt und Erde Verlag; 2020. pp. 15–37.
- 73.
Eriksen BV. Late Palaeolithic settlement in Denmark – how do we read the record? In: Kobusiewicz M, Kozlowski JK, editors. Post-pleniglacial re-colonisation of the great European Lowland: Papers presented at the Conference organised by the International Union of Prehistoric and Protohistoric Sciences, Commission 8, held at the Jagellonian University, Kraków, in June 1998. Kraków: Polska Akademia Umiejętności; 1999. pp. 157–73.
- 74. Riede F. The Laacher See-eruption (12,920 BP) and material culture change at the end of the Allerød in Northern Europe. J Archaeol Sci. 2008;35(5):591–9.
- 75.
Crombé P, Deeben J, Van Strydonck M. Hunting in a changing environment: The transition from the Younger Dryas to the (Pre)boreal in Belgium and the Southern Netherlands. In: Jaubert J, Fourment N, Depaepe P, editors. Transitions, ruptures et continuité en préhistoire: actes du XXVII Congrès préhistorique de France: Bordeaux-Les Eyzies: 31 mai-5 juin 2010 Volume 2: Paléolithique et Mésolithique. Paris: Société Préhistorique Française; 2014. pp. 583–604.
- 76. Crombé P, Sergant J, Verbrugge A, De Graeve A, Cherretté B, Mikkelsen J, et al. A sealed flint knapping site from the Younger Dryas in the Scheldt valley (Belgium): Bridging the gap in human occupation at the Pleistocene–Holocene transition in W Europe. J Archaeol Sci. 2014;50:420–39.
- 77.
Vermeersch PM. La transition Ahrensbourgien-Mésolithique ancien en Campine belge et dans le sud sableux des Pays-Bas. In: Fagnart J-P, Thévenin A, Ducrocq T, Souffi B, Coudrec P, editors. Le début du Mésolithique en Europe du Nord-Ouest, Mémorie XLV de la Societé préhistorique française. Paris: Societé Préhistorique Française; 2008. pp. 11–29.
- 78.
Niekus MJ, Boekschoten GR, Deeben JHC. A Late Preboreal site from Zwolle, province of Overijssel, and some remarks on the Ahrensburgian in the Netherlands. In: Eriksen BV, Rensink E, Harris S, editors. The Final Palaeolithic of Northern Eurasia: proceedings of the Amersfoort, Schleswig and Burgos UISPP commission meetings. Kiel: Ludwig; 2019. pp. 51–79.
- 79.
Vandendriessche H. Flintknapping from the Late Glacial to the Early Holocene: the Belgian Scheldt Valley sites of Ruien and Kerkhove. Leiden: Sidestone Press; 2022.
- 80. Crombé P, Pironneau C, Robert P, van der Sloot P, Boudin M, De Groote I, et al. Human response to the Younger Dryas along the southern North Sea basin, Northwest Europe. Sci Rep. 2024;14(1):18074. pmid:39103459
- 81.
Barton RNE. Long Blade Technology in Southern Britain. In: Bonsall C, editor. The Mesolithic in Europe: papers presented at the third international symposium, Edinburgh, 1985. Edinburgh: John Donald; 1989. pp. 264–271.
- 82.
Barton RNE, Roberts AJ. The transition from the Younger Dryas to the Pre-Boreal in southern Britain: some new perspectives on the spatial patterning and chronology of long blade sites. In: Montoya C, Fagnart J-P, Locht J-L, editors. Préhistoire de l’Europe du nord-ouest: mobilités, climats et identités culturelles Volume 2: Paléolithique supérieur ancien, Paléolithique final – Mésolithique, Congrès préhistorique de France: Compte Rendu de la 28ème session. Paris: Société Préhistorique Française; 2019. pp. 381–389.
- 83.
Fagnart J-P. La fin des temps glaciaires dans le nord de la France: approches archéologiques et environnementales des occupations humaines du Tardiglaciaire. Paris: Société Préhistorique Française; 1997.
- 84.
Valentin B. Jalons pour une Paléohistoire des derniers chasseurs (XIVe-VIe millénaire avant J.-C.). Paris: Publications de la Sorbonne; 2008.
- 85. Naudinot N, Fagnart J-P, Langlais M, Mevel L, Valentin B. Les dernières sociétés du Tardiglaciaire et des tout débuts de l’Holocène en France. Gallia Préhist. 2019;(59):5–45.
- 86.
Weber M-J, Valentin B, Bodu P. It’s all about volume - New observations on specific aspects of core reduction at the Belloisian site Donnemarie-Dontilly (Seine-et-Marne, France). In: Eriksen BV, Rensink E, Harris S, editors. The Final Palaeolithic of Northern Eurasia: proceedings of the Amersfoort, Schleswig and Burgos UISPP commission meetings. Kiel: Ludwig; 2019. pp. 91–104.
- 87.
Street M, Baales M, Gehlen B, Heinen M, Heuschen W, Orschiedt J, et al. Archaeology across the Pleistocene-Holocene boundary in western Germany: Human responses to rapid environmental change. In: Montoya C, Fagnart J-P, Locht J-L, editors. Préhistoire de l’Europe du nord-ouest: mobilités, climats et identités culturelles Volume 2: Paléolithique supérieur ancien, Paléolithique final – Mésolithique, Congrès préhistorique de France: Compte Rendu de la 28ème session. Paris: Société Préhistorique Française; 2019. pp. 491–510.
- 88.
Mevel L, Weber M-J, Berg-Hansen IM. Towards a reconstruction of the techno-economic variability of the Ahrensburgian in Schleswig-Holstein. In: Montoya C, Fagnart J-P, Locht J-L, editors. Préhistoire de l’Europe du nord-ouest: mobilités, climats et identités culturelles Volume 2: Paléolithique supérieur ancien, Paléolithique final – Mésolithique, Congrès préhistorique de France: Compte Rendu de la 28ème session. Paris: Société Préhistorique Française; 2019. pp. 433–453.
- 89.
Sørensen M, Lübke H, Groß D. The Early Mesolithic in Southern Scandinavia and Northern Germany. In: Milner N, Conneller C, Taylor B, editors. Star Carr: a persistent place in a changing world. Heslington, York: White Rose University Press; 2018. pp. 305–329.
- 90. Bodu P, Valentin B. Groupes à Federmesser ou Aziliens dans le sud et l’ouest du Bassin parisien. Propositions pour un nouveau modèle d’évolution. Bull Soc Préhist Fr. 1997;94(3):341–8.
- 91. Naudinot N, Jacquier J. Socio-economic organization of Final Paleolithic societies: New perspectives from an aggregation site in Western France. J Anthropol Archaeol. 2014;35:177–89.
- 92. Langlais M, Chevallier A, Fat Cheung C, Jacquier J, Marquebielle B, Naudinot N. The Pleistocene-Holocene transition in Southwestern France: A focus on the Laborian. Quat Int. 2020;564:37–47.
- 93. Gibaud A, Delvigne V, Langlais M, Bracco J-P. Frontière de préhistorique·s ou frontière de préhistorien·ne·s: quid des litho-espaces durant la transition Pléistocène-Holocène en France centrale?. Gallia Préhist. 2023;2023:75–102.
- 94.
Fornage-Bontemps S. Le niveau A4 de Rochedane, l’est de la France et la question des influences épigravettiennes à la fin du tardiglaciaire. Besançon: Presses universitaires de Franche-Comté; 2017.
- 95.
Fornage-Bontemps S. Quand les idées franchissent les montagnes. L’Est de la France et la question de la diffusion des influences épigravettiennes au nord des Alpes entre l’Allerød et la fin du Dryas recent. In: Naudinot N, Meignen L, Binder D, Querre G, editors. Les systèmes de mobilité de la préhistoire au Moyen Age: XXXVe rencontres internationales d’archéologie et d’histoire d’Antibes: actes des rencontres, 14-16 octobre 2014. Nice: APDCA; 2016. pp. 337–352.
- 96.
Mevel L, Fornage-Bontemps C, Bereiziat G. Au carrefour des influences culturelles? Les industries lithiques de la fin du Tardiglaciaire entre Alpes du nord et Jura, 11 500-9 500 CalBC. In: Langlais M, Naudinot N, Peresani M, editors. Les groupes culturels de la transition Pléistocène - Holocène entre Atlantique et Adriatique: actes de la séance de la Société préhistorique française Bordeaux 24-25 mai 2012. Paris: Société Préhistorique Française; 2014. pp. 45–81.
- 97.
Mevel L, Pion G, Fornage-Bontemps S. Changements techniques et géographie culturelle à l’extrême fin du Paléolithique dans les Alpes du nord françaises. Les stratigraphies de l’abri de La Fru (Savoie) revisitées. In: Jaubert J, Fourment N, Depaepe P, editors. Transitions, ruptures et continuité en préhistoire: actes du XXVII Congrès préhistorique de France: Bordeaux-Les Eyzies: 31 mai-5 juin 2010 Volume 1. Paris: Société Préhistorique Française; 2014. pp. 527–546.
- 98. Naudinot N, Tomasso A, Messager E, Finsinger W, Ruffaldi P, Langlais M. Between Atlantic and Mediterranean: Changes in technology during the Late Glacial in Western Europe and the climate hypothesis. Quat Int. 2017;428:33–49.
- 99. Tomasso A, Cheung C, Fornage-Bontemps S, Langlais M, Naudinot N. Winter is coming: What happened in western European mountains between 12.9 and 12.6 ka cal. BP (beginning of the GS1). Quat Int. 2018;465:210–21.
- 100. Langlais M, Fat Cheung C, Laroulandie V, Lefebvre A, Marquebielle B, Pétillon J-M. Dix mille ans de peuplement humain sur le versant nord des Pyrénées (21000-11000 cal BP) entre unification et régionalisation culturelles. Munibe, Antropologia-Arkeologia. 2020;71:59–71.
- 101. Aura JE, Jordá JF, Montes L, Utrilla P. Human responses to Younger Dryas in the Ebro valley and Mediterranean watershed (Eastern Spain). Quat Int. 2011;242:348–59.
- 102. Bicho N, Haws J. The Magdalenian in central and southern Portugal: Human ecology at the end of the Pleistocene. Quat Int. 2012;272(273):6–16.
- 103. Bicho N, Haws J, Almeida F. Hunter–gatherer adaptations and the Younger Dryas in central and southern Portugal. Quat Int. 2011;242:336–47.
- 104. Gameiro C, Aubry T, Almeida F, Dimuccio L, Gabriel S, Gaspar R. Archaeology of the Pleistocene-Holocene transition in Portugal: Synthesis and prospects. Quat Int. 2020;564:113–37.
- 105. Bicho N, Cascalheira J, Marreiros J, Gonçalves C, Pereira T, Dias R. Chronology of the Mesolithic occupation of the Muge valley, central Portugal: The case of Cabeço da Amoreira. Quat Int. 2013;308(309):130–9.
- 106. Naudinot N, Tomasso A, Tozzi C, Peresani M. Changes in mobility patterns as a factor of 14C date density variation in the Late Epigravettian of Northern Italy and Southeastern France. J Archaeol Sci. 2014;52:578–90.
- 107. Ruiz-Redondo A, Vukosavljević N, Tomasso A, Peresani M, Davies W, Vander Linden M. Mid and late upper Palaeolithic in the Adriatic Basin: Chronology, transitions and human adaptations to a changing landscape. Quat Sci Rev. 2022;276:107319.
- 108. Duches R, Peresani M, Pasetti P. Success of a flexible behavior: Considerations on the manufacture of Late Epigravettian lithic projectile implements according to experimental tests. Archaeol Anthropol Sci. 2018;10:1617–43.
- 109.
Peresani M, Tomio C, Dalmeri G. Les grattoirs épigravettiens et leur « raccourcissement » durant le Tardiglaciaire en Italie. Reflets d’un changement dans l’économie du débitage. In: Langlais M, Naudinot N, Peresani M, editors. Les groupes culturels de la transition Pléistocène - Holocène entre Atlantique et Adriatique: actes de la séance de la Société préhistorique française Bordeaux 24-25 mai 2012. Paris: Société Préhistorique Française; 2014. pp. 205–220.
- 110.
Sauer F. Late Palaeolithic Land Use Patterns in Bavaria. Ph.D., Friedrich-Alexander-Universität. 2018. Available: urn:nbn:de:bvb:29-opus4-92875
- 111.
Vencl S. Prehistory of Bohemia 1: The Palaeolithic and Mesolithic. Praha: Archeologický ústav AV CR; 2013.
- 112. Mlejnek O, Záhorák V. Předběžná zpráva o výzkumu mezolitické a pozdně paleolitické locality u Uherska ve východních Čechách. Excavation of the Mesolithic and Late Palaeolithic site near Uhersko in Eastern Bohemia: preliminary report. Ročenka. 2018; 50-66.
- 113.
Eigner J, Prekop F. Tašovice 1 a 2: dvě pozdně paleolitické a mezolitické lokality na Karlovarsku = Tašovice 1 and 2: two late palaeolithic and mesolithic sites in the Karlovy Vary region. Praha: Národní muzeum; 2021.
- 114. Kapustka K, Eigner J, Parkman M, Řezáč M, Přichystal A, Pokorný P, et al. Pozdně paleolitické a mezolitické osídlení Šumavy:možnosti výzkumu, datování a interpretaceLate Palaeolithic and Mesolithic Settlement of Šumava: The Possibilities of Research, Dating and Interpretation. Pam Arch. 2020; 5–59.
- 115. Svoboda J, Pokorný P, Horáček I, Sázelová S, Abraham V, Divišová M, et al. Late Glacial and Holocene sequences in rockshelters and adjacent wetlands of Northern Bohemia, Czech Republic: Correlation of environmental and archaeological records. Quat Int. 2018;465: 234–250.
- 116.
Šída P, Pokorný P. Mesolithic of the Northern Bohemia III: evolution of prehistoric landscape in Bohemian Paradise: vegetation, fauna, humans. Brno: Archeologický ústav AV ČR; 2020.
- 117. Jochim MA, Kind C-J, Kleinmann A, Merkt J, Stephan E. Eine spätpaläolithische Fundstelle am Ufer des Federsees: Bad Buchau-Kappel, Flurstück Gemeindebeunden. Fundber Baden-Württemberg. 2015;35:37–134.
- 118.
Hornauer-Jahnke TK, Noack ES. Der Übergang vom Jung- zum Spätpaläolithikum in Südwestdeutschland: Der Zigeunerfels. In: Baales M, Pasda C, editors. “All der holden Hügel ist keiner mir fremd.”: Festschrift zum 65 Geburtstag von Claus-Joachim Kind. Bonn: Habelt; 2019. pp. 315–336.
- 119.
Schön W, Affolter J, Gehlen B, Guggenmos A, Schröppel J, Guggenmos S. Siedlungsrelikte eiszeitlicher Menschen im Allgäu. In: Batzel K, editor. Natur und Mensch während der Eiszeit im Allgäu. Aichach-Friedberg: Likias-Verlag; 2024.
- 120.
Schön W, Affolter J, Gehlen B, Gulisano G, Guggenmos A, Guggenmos S, et al. Mesolithische Besiedlung und Rohstoffversorgung im Allgäu. In: Uthmeier T, Maier A, editors. Stone Age: studying technologies of non-analogous environments and glacial ecosystems: papers in honor of Jürgen Richter. Bonn: Habelt; 2024. pp. 399–422.
- 121.
Nielsen EH. Paläolithikum und Mesolithikum in der Zentralschweiz: Mensch und Umwelt zwischen 17 000 und 5 500 v. Chr. Luzern: Kantonaler Lehrmittelverlag; 2009.
- 122.
Sedlmeier J. Die letzten Wildbeuter der Eiszeit: neue Forschungen zum Spätpaläolithikum im Kanton Basel-Landschaft. Basel: Schwabe Verlag; 2015.
- 123.
Kretschmer I. Demographische Untersuchungen zu Bevölkerungsdichten, Mobilität und Landnutzungsmustern im späten Jungpaläolithikum. Rahden/Westf: Verlag Marie Leidorf; 2015.
- 124. Peresani M, Monegato G, Ravazzi C, Bertola S, Margaritora D, Breda M, et al. Hunter-gatherers across the great Adriatic-Po region during the Last Glacial Maximum: Environmental and cultural dynamics. Quat Int. 2021;581(582):128–63.
- 125.
Niekus MJ, Cohen KM, Johansen L, Stapert D. Opgezogen, opgespoten en opgeraapt: vuurstenen en hardstenen artefacten van de Zandmotor en hun sedimentaire context. Amersfoort: Rijksdienst voor het Cultureel Erfgoed; 2023.
- 126. Boemke BJ, Maier A, Schmidt I, Römer W, Lehmkuhl F. Testing the representativity of Palaeolithic site distribution: The role of sampling bias in the European Upper and Final Palaeolithic record. Quat Sci Rev. 2023;316:108220.
- 127.
Baales M. Umwelt und Jagdökonomie der Ahrensburger Rentierjäger im Mittelgebirge. Bonn: Dr. Rudolf Habelt GmbH; 1996.
- 128. Sauer F, Riede F. A Critical Reassessment of Cultural Taxonomies in the Central European Late Palaeolithic. J Archaeol Method Theory. 2019;26(1):155–84.
- 129. Vaquero M, Romagnoli F. Searching for lazy people: the significance of expedient behavior in the interpretation of Paleolithic assemblages. J Archaeol Method Theory. 2018;25:334–67.
- 130.
Riede F. The ‘Bromme Problem’ – Notes on Understanding the Federmessergruppen and Bromme Culture Occupation in Southern Scandinavia during the Allerød and Early Younger Dryas Chronozones. In: Sørensen M, Buck Pedersen K, editors. Problems in Palaeolithic and Mesolithic research. Copenhagen: University of Copenhagen & Museum of Southeast Denmark; 2017. pp. 61–85.
- 131. Maier A, Lehmkuhl F, Ludwig P, Melles M, Schmidt I, Shao Y. Demographic estimates of hunter–gatherers during the Last Glacial Maximum in Europe against the background of palaeoenvironmental data. Quat Int. 2016;425:49–61.
- 132. Vermeersch PM. Radiocarbon Palaeolithic Europe database: A regularly updated dataset of the radiometric data regarding the Palaeolithic of Europe, Siberia included. Data Brief. 2020;31:105793. pmid:32577447
- 133. Bocquet-Appel J-P, Demars P-Y, Noiret L, Dobrowsky D. Estimates of Upper Palaeolithic meta-population size in Europe from archaeological data. J Archaeol Sci. 2005;32:1656–68.
- 134. Langlais M, Fat Cheung C. Le site laborien d’Auberoche (Le Change, Dordogne), collections Daleau et Daniel: Anciennes fouilles, nouvelle attribution. Bull Soc Préhist Fr. 2019;116:155–8.
- 135. Ordonez A, Riede F. Changes in limiting factors for forager population dynamics in Europe across the last glacial-interglacial transition. Nat Commun. 2022;13(1):5140. pmid:36068206
- 136. Riede F. LSTbase - a collation of Laacher See tephra occurrences. Zenodo; 2024.
- 137. Reinig F, Wacker L, Jöris O, Oppenheimer C, Guidobaldi G, Nievergelt D, et al. Precise date for the Laacher See eruption synchronizes the Younger Dryas. Nature. 2021;595(7865):66–9. pmid:34194020
- 138.. Straus LG. Were there human responses to Younger Dryas in Cantabrian Spain? Quat Int. 2011;242(2):328–35.
- 139.
Halstead P, O’Shea J, editors. Bad Year Economics: Cultural Responses to Risk and Uncertainty. Cambridge: Cambridge University Press; 1989. https://doi.org/10.1017/CBO9780511521218
- 140. Miller R. Mapping the expansion of the Northwest Magdalenian. Quat Int. 2012;272–273:209–30.
- 141. Theuerkauf M, Joosten H. Younger Dryas cold stage vegetation patterns of central Europe - climate, soil and relief controls. Boreas. 2012;41:391–407.
- 142.
Crombé P. The impact of environmental changes on the human occupation of the Scheldt basin during the Younger Dryas. In: Montoya C, Fagnart J-P, Locht J-L, editors. Préhistoire de l’Europe du nord-ouest: mobilités, climats et identités culturelles Volume 2: Paléolithique supérieur ancien, Paléolithique final – Mésolithique, Congrès préhistorique de France: Compte Rendu de la 28ème session. Paris: Société Préhistorique Française; 2019. pp. 407–416.
https://doi.org/
10.1111/j.1502-3885.2011.00240.x
- 143. Bos JAA, Gelorini V, Van Der Meeren T, Peleman J, Court-Picon M, Demiddele H, et al. The Younger Dryas and Preboreal landscape in the Moervaart area (northwestern Belgium) and the apparent decrease in human occupation. Veget Hist Archaeobot. 2018;27:697–715.
- 144.
Latałowa M. Late Glacial 14,000 –10,000 14C yr BP (ca. 15,500 (16,000)-11,500 cal yr BP). In: Ralska-Jasiewiczowa M, Latałowa M, Wasylikowa K, Tobolski K, Madeyska E, Wright jr HE, et al., editors. Late glacial and holocene history of vegetation in Poland based on isopollen maps. Kraków: W. Szafer Institute of Botany; 2004. pp. 385–391.
https://doi.org/
10.1007/s00334-017-0666-3
- 145. Ammann B, Birks HJB, Brooks SJ, Eicher U, Von Grafenstein U, Hofmann W, et al. Quantification of biotic responses to rapid climatic changes around the Younger Dryas — a synthesis. Palaeogeography, Palaeoclimatology, Palaeoecology. 2000;159:313–47.
- 146. Magny M, Aalbersberg G, Bégeot C, Benoitruffaldi P, Bossuet G, Disnar J, et al. Environmental and climatic changes in the Jura mountains (eastern France) during the Lateglacial–Holocene transition: a multi-proxy record from Lake Lautrey. Quat Sci Rev. 2006;25:414–45.
- 147. Cupillard C, Magny M, Bocherens H, Bridault A, Bégeot C, Bichet V, et al. Changes in ecosystems, climate and societies in the Jura Mountains between 40 and 8 ka cal BP. Quat Int. 2015;378:40–72.
- 148. Isarin RFB, Bohncke SJP. Mean July temperatures during the Younger Dryas in Northwestern and Central Europe as inferred from climate indicator plant species. Quat Res. 1999;51:158–73.
- 149. Van Asch N, Kloos ME, Heiri O, De Klerk P, Hoek WZ. The younger dryas cooling in northeast Germany: summer temperature and environmental changes in the Friedländer Groβe Wiese region. J Quat Sci. 2012;27:531–43.
- 150. Merkle JA, Monteith KL, Aikens EO, Hayes MM, Hersey KR, Middleton AD, et al. Large herbivores surf waves of green-up during spring. Proc R Soc B. 2016;283(1833):20160456. pmid:27335416
- 151. Maier A, Tharandt L, Linsel F, Krakov V, Ludwig P. Where the grass is greener — large-scale phenological patterns and their explanatory potential for the distribution of Paleolithic hunter-gatherers in Europe. J Archaeol Method Theory. 2024;31(1):918–45.
- 152. Price TD, Meiggs D, Weber M-J M, Pike-Tay A. The migration of Late Pleistocene reindeer: isotopic evidence from northern Europe. Archaeological and Anthropological Sciences. 2017;9:371–94.
- 153. Rivals F, Drucker DG, Weber M-J, Audouze F, Enloe JG. Dietary traits and habitats of the reindeer (Rangifer tarandus) during the Late Glacial of Northern Europe. Archaeol Anthropol Sci. 2020;12:98.
- 154. Usinger H. Ein weit verbreiteter Hiatus in spätglazialen Seesedimenten: mögliche Ursache für Fehlinterpretationen von Pollendiagrammen und Hinweis auf klimatisch verursachte Seespiegelbewegungen. Eiszeitalter und Gegenwart. 1981;31:91–107. https://doi.org/ 10.1007/s12520-020-01052-y
- 155. Renssen H, Goosse H, Roche DM, Seppä H. The global hydroclimate response during the Younger Dryas event. Quat Sci Rev. 2018;193:84–97.
- 156. Kolstrup E. Lateglacial older and younger coversand in northwest Europe: chronology and relation to climate and vegetation. Boreas. 2007;36(1):65–75.
- 157. Brauer A, Haug GH, Dulski P, Sigman DM, Negendank JFW. An abrupt wind shift in western Europe at the onset of the Younger Dryas cold period. Nat Geosci. 2008;1:520–3.
- 158. Clausen I, Hartz S. Fundplätze des Spätglazials am Sorgetal bei Alt-Duvenstedt, Kreis Rendsburg-Eckernförde. Offa. 1988;45: 17–35. https://doi.org/ 10.1038/ngeo263
- 159. Mussi M, Peresani M. Human settlement of Italy during the Younger Dryas. Quat Int. 2011;242:360–70.
- 160. Riede F. Changes in mid- and far-field human landscape use following the Laacher See eruption (c. 13,000 BP). Quat Int. 2016;394:37–50.
- 161.
Riede F. Splendid isolation: the eruption of the Laacher See Volcano and Southern Scandinavian Late Glacial hunter-gatherers. Aarhus: Aarhus University Press; 2017.
https://doi.org/
10.1016/j.quaint.2014.07.008
- 162. Riede F, Bazely O. Testing the ‘Laacher See hypothesis’: a health hazard perspective. Journal of Archaeological Science. 2009;36:675–83.
- 163. Riede F, Kierdorf U. The eruption of the Laacher See volcano (~13,000 years BP) and possible fluoride poisoning amongst contemporaneous wildlife and human foragers—Outline of a hypothesis and the way to test it. Intl J Osteoarchaeology. 2020;30:855–71.
- 164. Riede F, Wheeler JM. Testing the ‘Laacher See hypothesis’: tephra as dental abrasive. Journal of Archaeological Science. 2009;36:2384–91.
- 165. Niemeier U, Riede F, Timmreck C. Simulation of ash clouds after a Laacher See-type eruption. Clim Past. 2021;17:633–52.
- 166. Andreasen LS, Cornér J, Abbott PM, Sinclair VA, Riede F, Timmreck C. Changes in Northern Hemisphere extra-tropical cyclone frequency following volcanic eruptions. Environ Res: Climate. 2024;3(1):025002.
- 167. Hey J. On the number of New World founders: a population genetic portrait of the peopling of the Americas. PLoS Biol. 2005;3(6):e193. pmid:15898833
- 168. Smith CM. Estimation of a genetically viable population for multigenerational interstellar voyaging: Review and data for project Hyperion. Acta Astron. 2014;97:16–29.
- 169. Bradshaw CJA, Ulm S, Williams AN, Bird MI, Roberts RG, Jacobs Z, et al. Minimum founding populations for the first peopling of Sahul. Nat Ecol Evol. 2019;3(7):1057–63. pmid:31209287
- 170. Traill LW, Brook BW, Frankham RR, Bradshaw CJA. Pragmatic population viability targets in a rapidly changing world. Biol Conserv. 2010;14328–34.
- 171. Eller E, Hawks J, Relethford JH. Local extinction and recolonization, species effective population size, and modern human origins. Hum Biol. 2009;81(5–6):805–24. pmid:20504198
- 172. Schmidt I, Zimmermann A. Population dynamics and socio-spatial organization of the Aurignacian: Scalable quantitative demographic data for western and central Europe. PLoS One. 2019;14(2):e0211562. pmid:30759115
- 173. Whallon R. Social networks and information: Non-“utilitarian” mobility among hunter-gatherers. J Anthropol Archaeol. 2006;25(1):259–70.
- 174. Mandryk CAS. Hunter-gatherer social costs and the nonviability of submarginal environments. J Anthropol Res. 1993;49(1):39–71.
- 175.
Riede F. Climate Change, Demography and Social Relations: An Alternative View of the Late Palaeolithic Pioneer Colonization of Southern Scandinavia. In: McCartan S, Schulting R, Warren G, Woodman P, editors. Mesolithic Horizons Papers Presented at the Seventh International Conference on the Mesolithic in Europe. Oxford: Oxbow; 2009. pp. 3–10.
https://doi.org/
10.1086/jar.49.1.3630629
- 176.
Riede F. Success and Failure during the Late Glacial Pioneer Human Re-Colonisation of Southern Scandinavia. In: Riede F, Tallavaara M, editors. Lateglacial and postglacial pioneers in Northern Europe. Oxford: Archaeopress; 2014. pp. 33–52.
- 177. Maier A, Sauer F, Bergsvik KA. Transport patterns as heuristic testing variables for the social coherence of taxonomic units at different spatial scales. J Paleo Arch. 2022;5:8.
- 178.
Delvigne V, Lafarge A, Fernandes P, Pesesse D, Angevin R, Bindon P, et al. Quels territoires en préhistoire? Une analyse par réseaux de lieux pour penser l’espace au Paléolithique supérieur. In: Mevel L, Weber M-J, Maier A, editors. En Mouvement/ On the Move/ In Bewegung. Paris: Société Préhistorique Française; 2021. pp. 27–69.
https://doi.org/
10.1007/s41982-022-00120-0
- 179.
Floss H. Rohmaterialversorgung im Paläolithikum des Mittelrheingebietes. Bonn: Habelt; 1994.
- 180. Binford LR. Organization and formation processes: Looking at curated technologies. J Anthropol Res. 1979;35(3):255–73.
- 181. Tomasso A, Porraz G. Hunter-gatherer mobility and embedded raw-material procurement strategies in the mediterranean upper paleolithic. Evol Anthropol. 2016;25(3):164–74. pmid:27312188
- 182.
Eriksen BV. Fossil mollusks and exotic raw materials in late glacial and early postglacial find contexts – a complement to lithic studies. In: Fisher LE, Eriksen BV, editors. Lithic raw material economies in late Glacial and early Postglacial Europe. Oxford: Archaeopress; 2002. pp. 27–52.
https://doi.org/
10.1002/evan.21488
- 183. Álvarez-Fernández E. Magdalenian personal ornaments on the move: a review of the current evidence in Central Europe. Zephyrus. 2009;63:45–59.
- 184. Schwendler RH. Diversity in social organization across Magdalenian Western Europe ca. 17–12,000 BP. Quat Int. 2012;272(273):333–53.
- 185.
Maier A. The Central European Magdalenian: Regional Diversity and Internal Variability. Dordrecht: Springer; 2015. https://doi.org/10.1007/978-94-017-7206-8
- 186.
Burdukiewicz JM. Further research on amber in Palaeolithic archaeology. In: Burdukiewicz JM, Cyrek K, Dyczek P, Szymczak K, editors. Understanding the past: papers offered to Stefan K Kozłowski. Warsaw: Center for Research on the Antiquity of Southeastern Europe, University of Warsaw; 2009. pp. 69–74.
https://doi.org/
10.1007/978-94-017-7206-8
- 187. Petersen PV. Zigzag lines and other protective patterns in Palaeolithic and Mesolithic art. Quat Int. 2021;573:66–74.
- 188.
Thévenin A. Géographie et cultures au Tardiglaciaire- L’impact de l’axe Rhône Saône. In: Pion G, editor. Le paléolithique supérieur récent: nouvelles données sur le peuplement et l’environnement Actes de la Table ronde de Chambéry. Paris: Société préhistorique française; 2000. pp. 67–79.
https://doi.org/
10.1016/j.quaint.2019.09.029
- 189.
Thévenin A. L’espace culturel «Méditerranéen» ses expansions de la fin du Tardiglaciaire au début du Postglaciaire. In: Barbaza M, Jaubert J, editors. Territoires, déplacements, mobilité, échanges pendant la Préhistoire: terres et hommes du Sud Actes du 126e Congrès national des sociétés historiques et scientifiques, Toulouse, 2001. Paris: Comité des travaux historiques et scientifiques; 2005. pp. 429–451.
- 190. Mevel L. Les premières sociétés aziliennes: nouvelle lecture de la genèse du phénomène d’azilianisation dans les Alpes du Nord à partir des deux niveaux d’occupation de l’abri de La Fru (Saint-Christophe-la-Grotte, Savoie). Bull Soc Préhist Fr. 2013;110:657–89.
- 191. Jacquier J, Langlais M, Naudinot N. Late Laborian trapezoids: Function and origin of the first transverse projectile tips of Western Europe prehistory. Quat Int. 2020;564:48–60.