New insights for understanding spatial patterning and formation processes of the Neanderthal occupation in theAmalda I cave (Gipuzkoa, Spain)

previous item | next item

New insights for understanding spatial patterning and formation processes of the Neanderthal occupation in theAmalda I cave (Gipuzkoa, Spain)

Material Information

New insights for understanding spatial patterning and formation processes of the Neanderthal occupation in theAmalda I cave (Gipuzkoa, Spain)
Series Title:
Scientific Reports
Sánchez-Romero, Laura
Benito-Calvo, Alfonso
B. Marín-Arroyo, Ana
Agudo-Pérez, Lucía
Karampaglidis, Theodoros
Rios-Garaizar, Joseba
Publication Date:


Subjects / Keywords:
Amalda I Cave ( local )
Anthropic ( local )
Neanderthal ( local )
serial ( sobekcm )


The Level VII of Amalda I cave (Gipuzkoa, Spain) represents one of the latest Middle Palaeolithic occupations in the Cantabrian Region. It is characterized by the presence of Middle Palaeolithic lithic industry and animal remains, with clear evidences of anthropic and carnivore manipulation. At this site, the Neanderthal presence has been questioned in relation to the role of carnivores in the accumulation of large, medium-sized and small mammals. It has also been proposed that the Neanderthal occupation could have consisted of short-term occupations, where different activities took place in a structured space within the cave. However, all hypotheses lacked any integrative analysis of the site formation processes. With the aim of understanding these processes, a combination of spatial techniques, based on GIS and inferential statistics (density analysis, hotspots tools and palaeotopographic reconstruction), along with the taphonomic study of identifiable and non-identifiable macromammals remains, were employed. This study has revealed distinct use of the cave space by Neanderthals and carnivores. The major concentrations of lithics and medium-size mammal remains were clearly accumulated by humans at the cave entrance, while the small-size mammals were gathered by carnivores in an inner zone. The activities of the Neanderthals seem to be distinctly structured, suggesting a parallel exploitation of resources.
Original Version:
Scientific Reports, Vol. 10, no. 8733 (2020-05-26).

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
This item is licensed with the Creative Commons Attribution License. This license lets others distribute, remix, tweak, and build upon this work, even commercially, as long as they credit the author for the original creation.
Resource Identifier:
K26-05796 ( USFLDC: LOCAL DOI )
k26.5796 ( USFLDC: LOCAL Handle )

USFLDC Membership

University of South Florida
Karst Information Portal

Postcard Information



This item is only available as the following downloads:

Full Text


1SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ þNNew insights for understanding spatial patterning and formation processes of the þNNeanderthal occupation in the Amalda þII cave (Gipuzkoa, Spain)Lauraþ  Sánchez-Romeroþ n þ n wR , Alfonsoþ  Benito-þCCalvox, Anaþ  B.þ  Marín-Arroyoþ n þ n y, Lucíaþ  Agudo-þPPé rezy, þTTheodorosþ  Karampaglidisx & Josebaþ  Rios-Garaizarþ n þ n xþTThe Level VþIIII of Amalda þII cave (Gipuzkoa, Spain) represents one of the latest Middle þP Palaeolithic occupations in the þC Cantabrian Region. þIIt is characterized by the presence of Middle þP Palaeolithic lithic industry and animal remains, with clear evidences of anthropic and carnivore manipulation. At this site, the þNNeanderthal presence has been questioned in relation to the role of carnivores in the accumulation of large, medium-sized and small mammals. þI It has also been proposed that the þNNeanderthal occupation …‘—Ž†Šƒ˜‡…‘•‹•–‡†‘ˆ•Š‘”–æ–‡”‘……—’ƒ–‹‘•á™Š‡”‡†‹¡‡”‡–ƒ…–‹˜‹–‹‡•–‘‘’Žƒ…‡‹ƒ•–”—…–—”‡† space within the cave. However, all hypotheses lacked any integrative analysis of the site formation processes. With the aim of understanding these processes, a combination of spatial techniques, based on GþIIS and inferential statistics (density analysis, hotspots tools and palaeotopographic ”‡…‘•–”—…–‹‘f჎‘‰™‹–Š–Š‡–ƒ’Š‘‘‹…•–—†›‘ˆ‹†‡–‹¤ƒ„Ž‡ƒ†‘æ‹†‡–‹¤ƒ„Ž‡ƒ…”‘ƒƒŽ• remains, were employed. þT This study has revealed distinct use of the cave space by þNNeanderthals and carnivores. þT The major concentrations of lithics and medium-size mammal remains were clearly accumulated by humans at the cave entrance, while the small-size mammals were gathered by carnivores in an inner zone. þT The activities of the þNNeanderthals seem to be distinctly structured, suggesting a parallel exploitation of resources. e Neanderthal presence in the Cantabrian region has been documented from the end of the Middle Pleistocene to the MIS4 and MIS3, as we can see in sites such as Lezetxiki1 , 2, Arlanpe3, Axlor4 – 7, Aranbaltza8, El Castillo9 , 10, El Esquilleu11– 14, Cueva Morín15, Covalejos16-10), El Sidrón17 or La Viña18. e data recovered from all these sites indicate a great variability in lithic technology, subsistence strategies and landscape uses, suggesting that Neanderthals successfully adapted to dierent climatic and environmental contexts, experiencing also important cultural changes, evidencing the complexity and particular history of these groups. Amalda I is a cave with a long sequence, with a single Middle Palaeolithic level (Level VII) with undisputed evidence of human and carnivore activity, both as bone accumulators. is has given rise to a certain amount of debate about the nature of the occupation and the relevance of the carnivores in the taphocenosis of the deposit19– 21. Previous analysis of the lithic assemblage and its spatial distribution has interpreted this deposit as an occasional logistic occupation for the exploitation of local resources, or as a temporary refuge for Neanderthal groups moving between regions22, 23. is earlier study did not include a detailed analysis of the site formation processes and the role of the carnivores, not only as bone accumulators but also in the possible postdepositional alterations aer human occupation24. e Amalda I site has great potential for assessing the spatial organisation of the activities and the space use by Neanderthals, which has essential implications for understanding their social organisation, such as the existence of well-structured productive processes, the division of labour or the existence w—ƒb˜‘Ž—–‹‘‡•‡ƒ”…Š‡–‡”áywvwƒŽŽ‡›‹ˆ‡…‹‡…‡•—‹Ž†‹‰á‹˜‡”•‹–›‘ˆƒŽ‹ˆ‘”‹ƒá‡”‡Ž‡›ááz}xvá äx‡–”‘ƒ…‹‘ƒŽ†‡f˜‡•–‹‰ƒ…‹×•‘„”‡Žƒb˜‘Ž—…‹×—ƒƒá»‹‡””ƒ†‡–ƒ’—‡”…ƒáyávvvxᗔ‰‘•á’ƒ‹äybn”‘—’áf•–‹–—–‘f–‡”ƒ…‹‘ƒŽ†‡f˜‡•–‹‰ƒ…‹‘‡•”‡Š‹•–×”‹…ƒ•†‡ƒ–ƒ„”‹ƒá‹˜‡”•‹†ƒ††‡ƒ–ƒ„”‹ƒ fá˜ä‘•ƒ•–”‘•á{xა–ƒ†‡”á’ƒ‹ä R ‡æƒ‹Žã Žƒ—•ƒ”‘;„‡”‡Ž‡›ä‡†—þOPENOPENThere are amendments to this paper


2SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ of some kind of principles in the organisation of the habitat. e present work aims to identify the spatial organisation, occupation pattern and factors that could have acted in the formation of the assemblage, as well as the alteration processes aerwards. is enables us to observe that the human and carnivore activities can be spatially distinguished, and that the human activity seems to be structured, both by the location of their main accumulations and the type of activity carried out.þCContexte Amalda I cave (Zestoa, Gipuzkoa) is located on the western hillside of the Alzolaras valley (Fig. 1 ), 110þ t m from the current base level of the Alzolaras stream, in a closed subsidiary valley of the Urola River, 11þ t km away from the current coastline. e Alzolaras gorge is deeper and narrower upstream from the cave, becoming wider towards the position of the site19. e Alzolaras stream runs 4þ t km until its conuence with the Urola River, which in turn currently ows towards Zumaia and into the Cantabrian sea. is connexion with the Urola River could have served as a natural communication route with the Aizkorri mountain range that allows the access to the Ebro Figure 1.þ (A) Digital elevation model (DEM) showing the position of Amalda I site in the context of the Alzolaras valley (Gipuzkoa, Basque Country). (B) Photo of the cave entrance taken with drone.


3SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ basin and Llanada Alavesa. e Urola River could also have served as access route to the coast23. Amalda I opens perpendicularly onto the river in a steep front of Urgonian limestones, and its formation is a consequence of the dissolution of these limestones to leave a subvertical joint (fracture) (W-E direction)19– 23. e cave has a characteristic triangular entrance, which gives way to a wide gallery 50þ t m long that narrows further in. e description of the lithostratigraphic sequence of Amalda I is based on the observations and analysis conducted during the excavation eldwork developed between 1979 and 198419 and the reworking of one section by ourselves in 2017 (Supplementary, Fig.S1). Level VII is characterized by a silty-clay matrix, with a low percentage of sand, no cementation and some angular gravels of limestone at the top, with Middle Palaeolithic lithic industry and faunal remains (Supplementary, Fig.S1). e Level has been recently dated between ca. 44,500 and 42,600 uncal BP by 14C AMS of ultraltered collagen samples25, being one of the most recent presences of Middle Palaeolithic in the region. From the same Level, another dated sample gave a result of 28,640þ t þ t 310 (OxA-32425) (Table 1 ), which jars with the Middle Palaeolithic dating but is very close to the results obtained for the immediately superior Gravettian Level VI25. A limited admixture between Levels VI and VII, also detected in the lithic assemblage23, would explain this odd date.ResultsDating.þ ree carnivore-modied bones were selected for dating, two from Altuna’s collection and one yielded from the excavation in 2017 by our team. e samples from the former come from the central part of the excavated area (9C and 6D) and were selected from the top and bottom of Level VII. ese remains were identied as Rupicapra rupicapra and had clear gnawing marks (Table 2 ). e sample from the 2017 eldwork comes from a marginal area of the excavation (G11-12). ese remains were also identied as Rupicapra rupicapra and all of them had gnawing and digestion traces (Table 2 ). Two of the three dates (Beta-451923 and Beta-451419) are consistent with the outlier obtained in 2018 (OxA-32425) (Table 1)25, which might suggest that the carnivore activity was coeval to the Early Gravettian occupation of the site. e third date (Beta-451922) seems to indicate that carnivores roamed the cave when it was unoccupied by humans, between the Late Middle Palaeolithic and the Early Gravettian (Table 2 ). ese results also indicate that some admixture exists between Middle Palaeolithic and Gravettian materials.þTT aphonomic analysis.þ e results obtained refer to the non-identiable elements of the assemblage, a total number of 4,589 fragments, including antler, bones and teeth. In terms of taxonomy, despite the fragmentation (the average length is 0.6þ t cm), it was still possible to attribute 18% of the assemblage to mammal-size categories and some specic ungulates (82% was completely indeterminate). Only 2% were identied as ungulates, such as Bos/Bison sp., Equus sp., Cervus elaphus , Capra pyrenaica and Rupicapra rupicapra . Also, some carnivores were identied, such as Ursus sp., Canis lupus and Vulpes vulpes . Apart from those, some avifauna and mustelids were also identied. Due the preservation state, 16% of the assemblage was only categorised as large, medium, small and very small mammals that included both ungulates and carnivores (Table3 ). Regarding the taphonomy of the non-identiable faunal elements, 10% showed anthropogenic modications, including percussion marks on long-bone shas (0.2%), fresh breakage patterns (1.2%), akes (2.3%), cut marks (1.3%) and thermoalterations (7.4%) (Fig. 2 , Table 4 ). On the other hand, carnivore traces on bones were better represented, since 3% of them showed carnivores marks, including scores, furrowing and punctures, 24% of them digestion traces and only 0.4% show both marks and digestive traces (Table 4). is high number of tiny digested bones can only be the result of defecation onto the scats on-site by the carnivores responsible for consuming the animal carcasses.þPP alaeotopographic reconstruction of Level VþII II.þ e reconstruction of the palaeotopography of the Level VII was based on 595 points located at the base of the Level VII. e results obtained for this level show a continuous surface characterised by a mean slope of 9.54°, where it is possible to observe a higher part located to the NW, towards to the interior of the cave, and a lower zone located to the SE, at the entrance (Fig. 3). Sample Species Bone Modication Lab Code Date (BP) Error %C %N C:N d13C A.12D.124 Equu s sp. Metacarpal Impact notch OxA-32425 28640 310 42.4 4.1 3.2 20.8 A8G.204.13 Cervus elaphus Metatarsal Anthropogenic breakage OxA-32500 44500 2100 42.1 3.8 3.4 20.2 7þ t G.221.16.X62.Y82 Cervus elaphus Tibia Anthropogenic breakage OxA-34933 42600 1600 43.7 15.9 3.2 21.2Table 1.þ 14C AMS (ultraltered) results of the bone samples from the Level VII of Amalda I25. Sample Species Bone Modication Lab Code Date (BP) Error %C %N C:N d13C A.9C.171 Rupicapra rupicapra Tibia Gnawed BETA-451923 28720 140 37.26 13.43 3.2 20 A.6D.208 Rupicapra rupicapra Falanx Gnawed BETA-451922 35590 270 39.9 14.47 3.2 19.8 A.17.10020 Indetermined Fragment Digested BETA-481419 28240 150 41.74 14.9 3.3 20Table 2.þ 14C AMS (ultraltered) results of the dated carnivore modied bone samples from the Level VII of Amalda I.


4SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ NR % Ungulates 73 1.59 Carnivores 10 0.22 Avifauna 12 0.26 Mustelids 2 0.04 Large size-mammals 194 4.23 Medium size-mammals 204 4.45 Small size-mammals 298 6.49 Very small size-mammals 38 0.83 Indeterminate 3758 81.89 TOTAL 4589 100Table 3.þ Taxonomy of the non-identiable faunal assemblage in terms of Number of Remains (NR) and their percentage of representation. Figure 2.þ Anthropogenic modications identied within the non-identiable faunal assemblage of Level VII at Amalda I. (A) Metaphysis of large-size mammal with cut marks. (B) Metapodial of Cervus elaphus with anthropic fresh breakage and cut marks (C) Proximal radius of Equus sp. with oblique cut marks in the cranial side. Image: Lucía Agudo.


5SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ Distribution patterns.þ e application of kernel density analysis indicates that the zone with the maximum concentration of lithic remains does not coincide with the zone with maximum concentration of faunal remains. e two clusters are located facing each other and separated by a gap (Fig. 4), in which there is a small amount of material, delimited by the two main concentrations of lithic and faunal remains. e clustered nature of the data was veried through the application of several statistical tests according to the number of points (nþ t þ t 1972) per square (nþ t þ t 96). Proceeding in this way, the chi-squared (X2) and Kolmogorov-Smirnov (K-S) tests revealed that both lithic and faunal remains are grouped (or clustered), showing lower critical values than the signicance level (pþ t þ t 0.95) (Supplementary, Section 2). Additionally, other analyses were carried out to delve into the distribution of the remains, due to the apparently dierentiated patterns observed between faunal and lithic industry, as well as the clustered pattern resulting from the dierent statistical tests applied. us, we have analysed the Average Nearest Neighbour (ANN) of the projected points, as well as the Global Moran’s I, incremental spatial autocorrelation and the Ripley’s K function according to the distribution of points and squares (Supplementary, TableS1).‡¤‹–‹‘‘ˆ–Š‡ƒ‹…Ž—•–‡”•äþ e identication and denition of the main accumulations of materials were made possible through hotspots analysis, calculating the Getis-Ord Gi and the Anselin Local Moran’s I statistics according to the variable of maximum length. ese techniques provided the statistical signicance of the clusters that comprise high ( hot ) and/or low ( cold ) values as a function of the value of the maximum length and their spatial location (Supplementary, Figs.S2 and S3).þCCharacterization of the groups.þ e clusters related to the lithic industry have been named CL1 and CL2, while the clusters of faunal remains are CF1 and CF2. Regarding the lithic clusters, there are some dierences between them, such as the number of remains making up each. Although both are mainly composed of small remains, the minimal length being equal or less than 10þ t mm (Table 5 ). e most abundant raw material in Level VII is int, although there are many others such as argillite, quartzite, quartz, limonite and mudstone. us, in both clusters the most represented raw material is int. Regarding the support, fragments dominate the assemblage of both clusters, followed by akes and resharpening akes (Supplementary, TableS4). With regard to the types of tools, the non-retouched pieces clearly dominate the record, both in CL1 and CL2. Side-scrapers are the most abundant retouched tool in both clusters (Supplementary, TableS4). For the faunal remains, the clusters N % Anthropogenic modications Fresh breakage 57 1.2 Bone akes 101 2.2 Cut marks 8 0.2 ermoalterations 340 7.4 Both cut marks carnivore marks 12 0.3 Carnivore modications Carnivore marks (scores, punctures, furrowing) 116 2.5 Digestive traces 1122 24.4 Both carnivore marks digestion 18 0.4Table 4.þ Anthropogenic and carnivore modications identied within the non-identiable faunal assemblage in terms of Number of Remains (NR) and their percentage. Figure 3.þ Palaeotopographic reconstruction and slope map of the Level VII from Amalda I.


6SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ Figure 4.þ Kernel density analysis and Jenks classication method applied to faunal (A, A’) and lithic (B, B’) remains from the Level VII, Amalda I cave. n length (mm) min. max. mean CL1 409 8 101 21.25 CL2 76 10 77 26.04 CF1 181 10 130 25.74 CF2 191 10 160 23.52Table 5.þ Clusters of lithic (CL1, CL2) and faunal remains (CF1, CF2) identied in the Level VII of Amalda I.


7SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ have been named CF1 and CF2, both being very similar regarding the amounts of remains contained but not in relation to the mean length (Table 5 ). ese clusters also present minimal lengths very similar to those found in the lithic clusters, although with some dierences in the maximum and mean lengths (Table 5 ). Regarding the species, the assemblage is dominated by Rupicapra rupicapra , which is the most abundant taxon in both clusters (Supplementary, TableS5).Directional patterns.þ All the groups, except CF2, show quite high eccentricity values, mainly in the case the whole assemblage (eþ t þ t 0.912). e ellipse CF2 is the most circular, with the lowest e-value (eþ t þ t 0.589). In the case of the ellipse orientations, there is no similarity among the groups, although the orientation pattern between the ellipse of CL2 (113.83°) and the ellipse of the whole assemblage (98.89°) are similar (Fig. 5 ). e CF1 ellipse is practically perpendicular to the cavity (Fig. 5 ), while the CF2 one tends to the east although the shape is almost circular. e ellipse of the CL1 also tends to the east, in comparison with the ellipse of CL2 (Fig. 5). All these data are very signicant, because all the groups, except CL2, show a very dierent orientation pattern with respect to the axis of the cavity. According to the MBR( minimum bounding rectangle ) calculation, the orientation of the cave is 175.9°. It is also important to take into account that all the ellipses present very dierent mutual orientation patterns.þNNon-identifiable faunal remains.þ The analysis of the distribution patterns according to the taphonomic modication of the non-identiable faunal remains shows several dierences between them, although the main concentrations are generally located in the limits established by Kernel. e remains with gnawing marks are mainly concentrated in the square 11C (Supplementary, Fig.S7), while the digested bones appear in 13C (Supplementary, Fig.S8). In both cases, the application of Getis-Ord Gi and Anselin Local Moran’s has allowed clear delimitation of the statistically signicant accumulation zones, reducing the area of accumulation of remains. Regarding the remains with anthropic alterations, the highest concentration is located in the square 12C (Supplementary, Fig.S9), as in the case of burnt bones (Supplementary, Fig.S10). However, in the rst case the remains seem to be more dispersed than for the burnt remains, something that in both cases is reduced when Getis-Ord Gi and Anselin Local Moran’s I are applied. e statistically signicant accumulation of burnt bones is larger than in the remains with evidence of anthropic alteration. e remains with evidence of water dissolution are located in the square 9C, this accumulation being the most concentrated in a specic point of the excavated area (Supplementary, Fig.S11).Discussione palaeotopographic reconstruction of Level VII of the Amalda I site shows that the accumulation patterns seem to be independent of the palaeotopography of the level. e densest accumulation area of lithic and faunal remains is regular and with a gentle slope. However, in the lowest zone towards the southeast of the excavated area there are hardly any remains. is pattern is general over the whole site, since the denser areas are located in the central and western part of the site, while towards the east the density of materials is considerably lower. e materials are not accumulated following the slope, nor in the more depressed zone of the palaeotopography Figure 5.þ Directional distribution ellipses of the whole assemblage (A) and the clusters classied (B) in the Level VII, Amalda I.


8SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ (Fig. 6). If the materials had been transported by agents like water, they probably would have accumulated in the more depressed zone. e Getis-Ord Gi statistic identies the remains located in this zone as a hotspot, although they are larger than those in the main accumulation. is pattern is the same for faunal and lithic remains, although the sample size in the case of the faunal remains is not signicant enough to establish comparisons with the results obtained by Getis-Ord Gi . As was seen earlier, the lithic materials identied as hotspots are more dispersed and located in the margins of the depressed zone, not in the deeper part. In the case of the faunal remains identied as hotspots, there are fewer and they are found in the lower part of the depressed zone. It is important to remark that the taphonomic analysis does not indicate any kind of evidence of crushing or breaking produced by falls of limestone blocks. Furthermore, the palaeotopographic proles of Level VII indicate that the coldspot clusters, where the smaller remains are clustered, are not found in depressed zones or with low slope (Fig. 6 ). e clusters of faunal and lithic remains are in very close spatial positions, even lying opposite each other. If the materials had been transported by water ows, it is reasonable to think that the accumulation of smaller remains would be in the more depressed zone. On the other hand, in this case the smaller remains would have been winnowed out and the larger ones would have remained, so the concentration of smaller items in the nearest margin would not be explicable. Additionally, no evidence of gravels or sand that would indicate the existence of water currents that could transport and select these materials has been observed or documented in this zone of the site6 , 19. at being said, the faunal and lithic remains show a little evidence of alteration by water, as we can see in the use-wear analysis22,23 and taphonomy (see Supplementary). However, we cannot know if this alteration by water could have been due to ows (small and ephemeral) or to zones with more humidity, maybe associated to dripping points, something common enough in caves. In the descriptions given in19 there are no references that suggest the presence of water ows. erefore, we can propose that in Amalda I the water eect was not strong Figure 6.þ N-S direction of the palaeotopographic proles that show the Level VII palaeo-relief and the position of the coldspots clusters.


9SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ enough to winnow or rearrange the spatial distribution of the remains. In addition, it is important to emphasize that the lithic assemblage does not show any sorting or selection by shape or size. Regarding the spatial patterning, there are elements other than water ows that can alter the arrangement of the archaeological materials and disturb a possible spatial organisation of the occupation area. e presence of carnivores in Amalda I is evident19–21,33, a point we can verify not only by the presence of carnivore alterations in bones, taxonomic prey representation, their skeletal and age proles, but also from the dating results that we have obtained, a fact that could be related to the disturbance activities of carnivores aer humans le the cave. e trampling of these animals could have provoked the percolation of remains to the lower levels. However, taking into account the impact of carnivore activity in anthropized spaces24, the alteration of the spatial patterning of Amalda I can be described as discrete. e carnivore activity in Amalda I could have been due to attraction by the remains previously le by the humans19, 21, 33. e chance that these activities altered the original position of the remains is very high, and some elements that could oer clues to the existence of structures like hearths could even have disappeared, which would explain the abundance of thermoaltered remains in some points of the occupied areas. However, this possible postdepositional alteration derived from the presence of carnivores would not explain the existence of distinct accumulations of faunal and lithic remains located in spatially dierent positions, even lying facing each other. If the activity of carnivores had been intense24, there would be a high probability that the two well-dierentiated areas of higher material concentration would not have been preserved. e site of Abric Romaní has evidence of carnivore access to remains leover by humans in a zone identied as toss34, in a zone further from the hearth and where the larger bones were located. In Amalda I, the main accumulation of remains with evidence of carnivore activity is observed around the cluster CF1, which is located in an inner zone, further away from the main accumulations of fauna and lithic remains (Supplementary, Figs.S7 and S8). Concerning the gap between the main clusters of faunal and lithic remains, this cannot be explained by the excavation process or by the presence of large blocks, which are lacking in this area of the cave. Since the evidence does not suggest that any natural process or recording error could have generated this spatial disposition, other explications can be considered. e well-dierentiated spatial distribution and clear clustering of small remains, together with the presence of accumulations of larger remains placed aside from the main accumulation area, recalls the model described by Binford34. However, we do not nd a mixture of lithic and faunal materials forming a single cluster. A clearly separated accumulation of lithic and faunal remains raises the possibility that this patterning arose from the conduct of parallel and simultaneous activities, such as dierent resource processing work (butchery, wood working, etc.) and/or knapping. e presence of several individuals carrying out similar or dierent activities in parallel can generate distributional patterns dierent from those described by Binford34. In the case of Amalda I, the accumulations coincide with the coldspots clusters, with materials no more than 20þ t mm long. e rest of the identied groups present slightly greater lengths, apart from the fact that their spatial location is clearly dierent than in other clusters with smaller remains. is structuration of activities seems to be related with technological organisation in these Neanderthal groups. As previous work has established22, 23, lithic technology is organised to supply dierent kind of tools for diverse activities. is has made it possible to identify the use of large mudstone akes, cleavers and bifaces for heavy activities, such as the initial phases (extraction) of butchering and woodworking; medium-sized akes and retouched tools made of int for skin processing or woodworking; and small int microlithic tools for nal phases, such as lleting, sinew cutting, tool maintenance, etc. is structuration of technology, together with that of the occupational space and activities, suggest an organisation of the production that could have involved some kind of social structure. e fact that the two main clusters (CF2 and CL1) are separated by a gap with a considerably lower density of remains opens up the possibility of a hearth as a structuring spatial element35. Several ethnographic studies have focused on the use of re and its possible traces at archaeological sites36–44, while other authors have centred on the study of those elements that allow intentional res to be distinguished from res of natural origin42, 45– 51. While at Amalda I there is evidence of re use (185 burnt remains according to Altuna’s data19, and 340 in the “non-identiable” assemblage), some of them even calcined19, we cannot conrm the existence of hearths, since no ashes, reddened sediments or any kind of basin shape to allow inferring the presence of this type of structures has been identied. e presence of burnt remains is indicative of the processing and consumption of prey, as well as the controlled use of re52. In addition, the calcined bones seem to be an indicator of controlled combustion, where the temperatures reached were high enough to cause the calcination of the remains46. Direct exposure can also produce this kind of alteration, either by direct contact or from the use of bones as fuel14, 57– 63. At Amalda I, the larger concentration of burnt bones is located in the squares 12C (nþ t þ t 56) and 11D (nþ t þ t 50), according to the data obtained aer the analysis of the non-identiable faunal remains (Fig. 7). In the case of the absence of ashes, this could be due to water eects, such as dripping during the rainiest seasons, causing their disappearance. e burnt remains, dehydrated due to their exposure to the re, are lighter, so their transport would have been easier as they are less dense64. It is important to highlight that exposure to re can produce greater fragmentation and fragility in these elements59, generating accumulations of small remains as a consequence of the exposure. e lack of clear evidence of domesticated res, such as structured hearths, burnt sediments or reddened stones, could be the result of taphonomic processes rather than the absence of re itself52, 53. Not all sites satisfy the necessary and required conditions for the conservation of re evidence54. In these cases, studies focused on the spatial distribution of materials have yielded one of the most appropriate tools for identifying and pinpointing of possible hearths52,53,55,56. e data obtained at Amalda I show that there is a dierent spatial layout of the main accumulations, not only in relation to the type of material, but also according to the length of the remains. e archaeological record is full of sites with dierential spatial distribution patterns, such as Qesem Cave52, 65, Kebara66– 68, Abric Romaní69– 80, inter alia ), Bolomor81 , 82, El Salt83 , 84, Grotte Vaufrey85 , 86 o Pincevent87 , 88, to name just a few. All these sites have in common the presence of structured hearths, but the absence of this kind of element does not mean that they


10SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ never existed, since dierent taphonomic processes can cause their disappearance52. Leroi-Gourhan and Brézillon referred to this type of pattern when they coined the term structures latentes87, dening those archaeological char acteristics that can be inferred from the study of the observable spatial distribution patterns of the archaeopaleontological materials. In the case of Amalda I, although there is no direct proof of hearths, we have indirect evidence of re use, such as thermoaltered remains and their concentration in some points of the cave, mainly in the gap between CL1 and CF2. is could be interpreted as a structuration of the space around a hearth that would have existed in this gap. However, the available data are insucient to verify this possibility. e spatial distribution patterns of remains found in Amalda I seem to point to multifunctional areas, where the exploitation of fauna and lithic resources would have taken place in the same area, producing well-dened clusters. us, it is possible that these activities were performed at the same time, causing contemporaneous accumulations75 according to the activity35, 40, 89. e distribution of the lithic pieces with traces of use also point in this direction, since their locations match the main accumulations and where there are clusters of smaller remains, both fauna and lithic materials. ere are also some dierences among the clusters identied, because there is a majority of cutting and scraping pieces in the accumulation zone of the cluster CF1, while there is a predominance of pieces focused on cutting in the cluster CL1 (Supplementary, TableS4). e abundance of cutting tools, together with the pieces with direct evidence of butchering activities, indicates that the processing and exploitation for consumption of red deer, bovines, ibex and horse carcasses took place in the cave itself22,23. Lastly, the arrangement of the main accumulations of materials could indicate the existence of a horizontal palimpsest. e reiterated occupation of the cave by humans and carnivores, reected in the activity traces, would have been seen in several diachronic accumulations, instead of being superimposed one on top of another. us, the human occupation is evidenced by the accumulation of lithic materials, prey taxa identied with evidence of anthropogenic modications, butchering activities documented on the bones and the presence of thermoalterations; while the carnivore occupation is seen in the accumulation of faunal remains, which also could be the result of human activity, since the carnivores could have come to the cave attracted by the remains le by humans19, 21, 33. e alternation in the occupation of the caves by carnivores and humans is amply demonstrated90– 94, while it is possible to generate horizontal palimpsests that could be wrongly assumed to be the result of the same moment of occupation of the site. Taking this into account, it is possible that most of the archaeopalaeontological assemblage has remained and the main area of activity preserved, but it is important to highlight the fact that the information most susceptible to be transported, like ashes or the smallest remains, could have been lost. e remains with evidence of intervention by carnivores are located in the most distant position and outside the main accumulation zone of humans. In this sense, it could be considered that the carnivores may have accessed the remains le by humans, transporting them to inner and more protected zones, thus prompting another accumulation parallel to the human one.þCConclusionse spatial study of Level VII has allowed the unravelling of the event succession that resulted in the formation of the Level, considering the anthropic, carnivore and natural activity. us, we have set out the possibility of a space structuration by Neanderthal groups, with the main activity area developed in the external part of the cave, where we nd the main concentrations of materials and, thanks to the information provided by bone taphonomy Figure 7.þ Density maps of burnt remains (A) and bones with cutmarks (B), overlapped in the main clusters of lithic (CL1, CL2) and faunal remains (CF1, CF2).


11SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ and lithic use-wear analysis, it is possible to infer the type of activities performed. us, the clusters CF2 and CL1 seem to indicate the conduct of parallel activities related with resource processing, allowing us to infer a dier ential deposition of lithic and faunal remains in the same main area of accumulation resulting from the possible task organisation of Neanderthal groups. e alteration by postdepositional processes of this space of activity has hardly been detected, since two clear areas of accumulation of materials have been perfectly distinguished by the type of material accumulated and the spatially layout of each accumulation. Also, the size of the materials is small, which seems to point to an intense exploitation of resources and the preservation of most of this activity. Apart from this, an accumulation has been observed in an inner zone of the cave (CF1), where there is a predominance of larger remains. is concentration seems to t with an accumulation of remains generated by car nivores aer the presence of humans in the cave. e absence of successive anthropic and carnivore marks on the same bones, as well as the dates obtained from the bone remains with carnivore marks, dating back to ca. 35-28 uncal BP, reinforce the idea that the carnivore visits were not immediately aer the Neanderthal occupation, but when Neanderthals had probably disappeared from this region. On the other hand, it is important to highlight the degree of alteration by other natural agents, such as water. Although there is evidence of alteration by this agent both in faunal and lithic remains, we have shown that alter ation by water was not strong enough to alter the spatial patterning, since water currents have not been detected in the sedimentary record. In this study, we have combined new spatial analysis tools and palaeotopographic reconstruction of the Level, together with the data provided by bone taphonomy and lithic analysis. All of this information has allowed us to unravel and shed light on the succession of events and processes that acted in the formation of Level VII of the Amalda I cave.Material and methodse spatial study that we present combines the palaeotopographic reconstruction of Level VII with the analysis of the spatial patterns of the lithic and faunal remains. e latter was tackled combining density analysis and hotspots tools to dene the empty areas and the clustering. e identied clusters were analysed by archaeological composition, size, shape and direction.Data.þ Several databases have been used to achieve this spatial analysis, both published and unpublished. Regarding the published ones, we have used the existing faunal data obtained by the multidisciplinary team led by J. Altuna during their excavations between 1979-1984, compiled in a detailed monograph19. On the other hand, we have also worked with the unpublished databases generated by J. Altuna and K. Mariezkurrena for the faunal remains, and the data of raw material, technology, typology and use-wear for the lithic assemblage produced by J. Rios-Garaizar23. Apart from these, we have obtained and included new taphonomic information from the Level VII faunal assemblage, including the non-identiable fauna that was barely studied in previous analysis20, 21. In addition, bones modied by carnivores from the Altuna collection and new remains from the recent excavation (2017) were selected for dating to establish the chronology of the carnivore activity in the cave. Regarding to the databases used, there are two types of data: one, materials with XYZ information recorded during the excavation (30%); and two, remains with information about the square of provenance and excavation spit (70%), to which random XY coordinates were assigned following the testing procedure detailed in23.þPP alaeotopographic reconstruction.þ e palaeotopographic reconstruction of Level VII is based on the selection of the set of points with information about the depth (Z coordinate). e interpolation method used was ordinary kriging (8 sectors), in order to produce a high-resolution topography based on the spatial correlation of the data. us, we created a continuous and estimated surface to study the relationships between the archaeological spatial distribution pattern and the oor relief during the occupation of the cave.Density analysis.þ e application of Kernel density analysis allows calculating the zones with high or low concentration of elements per unit area, according to the location and spatial proximity among the elements under study26. In the case of Amalda I we tested several search radii, selecting a nal radius of 0.50þ t m. Density maps were classied using natural groupings inherent to the data27. is classication allowed us to obtain a high-resolution estimation of the zones with a higher concentration of materials, separated from less representative low or medium density zones28.Hotspots analysis.þ e spatial study was not only focused on analysing whether the materials were clustered, dispersed or randomly distributed. It also provides the statistical signicance of clusters, according to their spatial location and the analysis of the variables that dene the groups. In this work, we applied the Getis-Ord Gi and Anselin Local Moran’s I techniques29,30 to analyse the spatial patterns according to the maximum length of the remains and the number of elements in each square (in the case of non-identiable remains) through the quadrat analysis method95. To discern the inuence of the spatial relationships in the hotspots analysis, we tested the inverse distance, the inverse distance squared and the xed band, obtaining no signicant dierences in the results. Additionally, we also tested the FDR correction (False Discovery Rate).Directional distribution.þ Directional distribution and shape of clusters were estimated using the standard deviational ellipse31, which measures the trend of a set of points, calculating the standard deviation of the X and Y coordinates from the mean centre to dene the axes of the ellipse32. is ellipse determines whether the clusters show compact or elongated shapes and their directional trend. Using this, it was possible to analyse the directional


12SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ features of each group identied for the lithic and faunal remains according to the variable of maximum length. us, information was obtained about the standard deviation of the materials that compose each group, both X-axis and Y-axis, as well as the rotation and orientation of every ellipse that the materials of the groups comprise. e data obtained from calculating each ellipse’s geometry has allowed their degrees of eccentricity to be known.þTT aphonomy.þ For this study, a taphonomic reassessment of the whole faunal assemblage from Level VII was carried out to plot both the anthropogenic and carnivore alterations identied on the ungulate remains spatially. In order to understand the formation of the assemblage and the use of the space use by humans and carnivores during the Middle Palaeolithic, an ArcGIS grid was devised to combine all the taphonomic data associated to the mammal specimens and its spatial location. For the rst time, a detailed taphonomic study of the non-identiable bone fragments is presented, recording the anthropogenic modications such as cut marks, breakage patterns, thermoalterations, but also the alterations by carnivores, including gnawing and digestionmarks.Received: 30 December 2019; Accepted: 22 April 2020; Published online: 26 May 2020Referencesþt 1.þt Baldeon, A. El yacimiento de Lezetxii (Gipuzoa, Pais Vasco). Los niveles musterienses. Munibe 45, 3–97 (1993).þt 2.þt Álvarez-Alonso, D. & Arrizabalaga, A. La secuencia estratigra ca inferior de la cueva de Lezetxii (Arrasate, País Vasco). Una reexion necesaria. Zephyrus 69, 15–29 (2012).þt 3.þt ios-Garaizar, J. et al . Short-term Neanderthal occupations in the Late Middle Pleistocene of Arlanpe (Lemoa, Northern Iberia Peninsula). Comptes endus Paleovol 14(3), 233–244 (2015).þt 4.þt Gonza lez-Urquijo, J. E. et a l. Excavaciones recientes en Axlor. Movilidad y planicacio n de actividades en grupos de neandertales. In Actas de la reunio n cientíca: Neandertales Canta bricos. Estado de la Cuestio n (eds. Montes, . & Lasheras, J. A.), 527–539 (Monografías del Museo Nacional y Centro de Investigacion de Altamira 20, 2005).þt 5.þt ios-Garaizar, J. Características de la produccio n lítica al nal del Paleolítico Medio en el País Vasco. El caso del nivel B de Axlor (Dima, Bizaia). In Actas de la reunio n cientíca: Neandertales canta bricos. Estado de la Cuestio n (Montes, . & Lasheras, J.A.), 333–348 (Monografías del Museo Nacional y Centro de Investigacion de Altamira 20, 2005).þt 6.þt ios-Garaizar, J. A new chronological and technological synthesis for Late Middle Paleolithic of Eastern Cantabrian egion. Quat. Int 433, 50–63 (2017).þt 7.þt Jones, J. . et al . Changing environments during the Middle-Upper Palaeolithic transition in the eastern Cantabrian egion (Spain): direct evidence from stable isotope studies on ungulate bones. Sci ep 8, (2018).þt 8.þt ios-Garaizar, J. et al . A Middle Palaeolithic wooden digging stic from Aranbaltza III, Spain. PLoS One 13(3), e0195044, https:// (2018).þt 9.þt Sa nchez-Ferna ndez, G. & Bernaldo De Quiro s, F. El nal del Musteriense Canta brico: El nivel 20e de la Cueva de El Castillo (Cantabria). Fe rvedes 5, 117–126 (2008).þt 10.þt Jones, J. . et al . Multi-Isotope investigations of ungulate bones and teeth from El Castillo and Covalejos caves (Cantabria, Spain): Implications for paleoenvironment reconstructions across the Middle-Upper Palaeolithic transition. J. Archaeol. Sci. ep 23, 1029–1042 (2019).þt 11.þt Baena, J. et al . Paleoecología y comportamiento humano durante el Pleistoceno Superior en la comarca de Lie bana. In Actas de la reunión cientíca: Neandertales canta bricos. Estado de la cuestio n (eds. Lasheras, J. & Montes, .), 461–487 (Monografías del Museo Nacional y Centro de Investigación de Altamira 20, 2005).þt 12.þt Baena, J., Carrio n, E., Cuartero, F. & Fluc, H. A chronicle of crisis: e Late Mousterian in north Iberia (Cueva de El Esquilleu, Cantabria, Spain). Quat. Int 247, 199–211 (2012).þt 13.þt Uzquiano, P. et al . Human behavior and adaptations to MIS 3 environmental trends ( 53-30 yrs BP) at El Esquilleu cave (Cantabria, northern Spain). Quat. Int 252, 82–89 (2012).þt 14.þt Yravedra, J. & Uzquiano, P. Burnt bone assemblages from El Esquilleu cave (Cantabria, Northern Spain): Deliberate use for fuel or systematic disposal of organic waste? Quat. Sci. ev. 68, 175–190 (2013).þt 15.þt Ma í llo-Ferna ndez, J. M. Aproximacio n tecnolo gica del nal del Musteriense de Cueva Morín (Villanueva de Villaescusa, Cantabria, Espaþn ñ a). Munibe 58, 13–42 (2007).þt 16.þt Sanguino, J. & Montes, . Nuevos datos para el conocimiento del Paleolítico Medio en el centro de la egión cantábrica: la Cueva de Covalejos. In Neandertales cantábricos. Estado de la cuestión (eds. Montes, . & Lasheras, J. A.), 489–504 (Monografías del Museo Nacional y Centro de Investigación de Altamira 20, 2005).þt 17.þt Torres, T. et al . Dating of the hominid (Homo neanderthalensis) remains accumulation from El Sidrón Cave (Piloña, Asturias, North Spain): an example of a multimethodological approach to the dating of Upper Pleistocene sites. Archaeometry 52, 680–705 (2010).þt 18.þt Wood, . E. et al . e chronology of the earliest Upper Palaeolithic in northern Iberia: New insights from L’Arbreda, Labeo oba and La Viña. J. Hum. Evol. 69, 91–109 (2014).þt 19.þt Altuna, J. (Ed.). La cueva de Amalda (Zestoa, País Vasco). Ocupaciones paleolíticas y postpaleolíticas (Monografía. Coleccio n Barandiara n 4, 1990).þt 20.þt Yravedra, J. Acumulaciones biolo gicas en yacimientos arqueolo gicos: Amalda VII y Esquilleu III-IV. Trab Preh 63(2), 55–78 (2006).þt 21.þt Yravedra, J. Nuevas contribuciones en el comportamiento cinegetico de la Cueva de Amalda. Munibe 58, 43–88 (2007).þt 22.þt ios-Garaizar, J. Organizacio n econo mica de las sociedades neandertales: El caso del nivel VII de Amalda (Zestoa, Gipuzoa). Zephyrus LXV , 15–37 (2010).þt 23.þt ios-Garaizar, J. Industria lítica y sociedad del Paleolítico Medio al Superior en torno al Golfo de Bizaia. Tesis Doctoral. Universidad de Cantabria (2012).þt 24.þt Camaro s, E. et al . Large carnivores as taphonomic agents of space modication: an experimental approach with archaeological implications. J. Archaeol. Sci. 30, 1361–1368 (2013).þt 25.þt Marín-Arroyo, A. B. et al . Chronological reassessment of the Middle to Upper Paleolithic transition and Early Upper Paleolithic cultures in Cantabrian Spain. PLoS One 13(6), e0199954, (2018).þt 26.þt Silverman, B. W. Estimacion de densidad para las estadísticas y el analisis de datos. (New Yor: Chapman and Hall, 1986).þt 27.þt Jens, G. F. e Data Model Concept in Statistical Mapping. Intern . Yearboo Cartography 7, 186–190 (1967).þt 28.þt Sánchez-omero, L., Benito-Calvo, A., Pérez-González, A. & Santonja, M. Assessment of accumulation processes at the Middle Pleistocene site of Ambrona (Soria, Spain). Density and orientation patterns in spatial datasets derived from excavations conducted from the 1960s to the Present. PLoS One 11(12), e0167595, (2016).þt 29.þt Moran, P. A. P. Notes on continuous stochastic phenomena. Biometria 37, 17–23 (1950).þt 30.þt Getis, A. & Ord, J. . e Analysis of Spatial Association by Use of Distance Statistics. Geographical Analysis 24, 3 (1992).


13SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ þt 31.þt De la Torre, I., Benito-Calvo, A., Arroyo, A., Zupancich, A. & Prott, T. Experimental protocols for the study of battered stone anvils from Olduvai Gorge (Tanzania). J. Archaeol. Sci. 40(1), 313–332 (2013).þt 32.þt Mitchell, A. La guía de ESI para el analisis SIG, 2. ESI Press (2005).þt 33.þt Altuna, J. & Mariezurrena, . Tafocenosis en yacimentos del País Vasco con predominio de grandes carnívoros. Consideraciones sobre el yacimiento de Amalda. In Actas de la I reunio n de cientícos sobre cubiles de hienas (y otros grandes carnívoros) en los yacimientos arqueologicos de la Península Iberica (Baquedano, E. & osell, J.), 214–228 (Zona Arqueológica 13, 2010).þt 34.þt Saþn ñ udo, P., Vallverdu -Poch, J. & Canals, A. Spatial patterns in Level J. In High resolution. Archaeology and Neanderthal behaviour. Time and space in Level J of Abric omaní (Capellades, Spain) (eds. Carbonell, E.), 47–76 (Vertebrate Paleobiology and Paleoanthropology Series. Springer, 2012).þt 35.þt Binford, L. . In Pursuit of the Past. Decoding the Archaeological ecord (ames and Hudson, 1983).þt 36.þt Murray, P. Discard location: e ethnographic data. American Anthropol 45, 490–502 (1980).þt 37.þt Binford, L.. Bones: Ancient Men and Modern Myths (Academic Press, New Yor, 1981).þt 38.þt Brain, C.. e Hunters or the Hunted? An Introduction to African cave taphonomy (University of Chicago Press, Chicago, 1981).þt 39.þt Yellen, J. E. Archaeological Approaches to the Present (New Yor: Academic Press, 1977).þt 40.þt Yellen, J. E. Cultural patterning in faunal remains: Evidence from the !ung Bushmen. In Experimental Archaeology (eds. Ingersoll, D., Yellen, J.E. & MacDonald, W.), 271–331 (Columbia University Press, New Yor, 1977).þt 41.þt O’Connell, J. F. Alyawara site structure and its archaeological implications. American Anthropol 52, 74–108 (1987).þt 42.þt Brain, C. . & Sillen, A. Evidence from the Swartrans cave for the earliest use of re. Nature 336, 464–466 (1988).þt 43.þt Fisher, J. W. Jr. & Stricland, H. C. Dwellings and replaces: eys to Efe Pygmy campsite structure. In Ethnoarchaeological approaches to mobile campsites (eds. Gamble, C. S. & Boismier, W. A.), 215–236 (International Monographs in Prehistory, Ann Arbor MI, 1991).þt 44.þt Nicholson, A. & Cane, S. Desert camps: Analysis of Australian aboriginal proto-historic campsites. In Ethnoarchaeological approaches to mobile campsites (eds. Gamble, C. S. & Boismier, W. A.), 263–354 (International Monographs in Prehistory, Ann Arbor MI, 1991).þt 45.þt Stinson, . J. & Wright, H. A. Temperature headres in the southern missed prairie of Texas. J. ange Management 22, 169–174 (1969).þt 46.þt Shipman, P., Foster, G. F. & Schoeninger, M. Burnt bones and teeth: And experimental study of colour, morphology, crystal structure and shrinage. J. Archaeol. Sci. 11, 307–325 (1984).þt 47.þt Grayson, D. . Danger Cave, last Supper Cave, and Hanging roc shelter: The faunas (American Museum of Natural History Anthropological Papers, 66 (1), 1988).þt 48.þt David, B. How was this bone burnt? In: Solomon, S., Davidson, I. and Watson, D. (Eds.): Problem solving in taphonomy: Archaeological and paleontological studies from Europe, Africa and Oceania. Tempus, Anthropol. Museum 2, 65–79 (1990).þt 49.þt Lyman, . L. Vertebrate Taphonomy (Cambridge University Press, Cambridge, 1994).þt 50.þt Stiner, M. C., uhn, S. L., Weiner, S. & Bar-Yosef, O. Dierential burning, recrystallization, and fragmentation of archaeological bones. J. Archaeol. Sci. 22, 223–237 (1995).þt 51.þt Mentzer, S. M. Microarchaeological approaches to the identication and interpretation of combustion features in prehistoric archaeological sites. J. Archaeol. Method eory , 1–53 (2012).þt 52.þt Blasco, ., osell, J., Saþn ñ udo, P., Gopher, A. & Barai, . What happens around a re: Faunal processing sequences and spatial distribution at Qesem Cave (300 a), Israel. Quat. Int. 398, 190–209 (2016).þt 53.þt Alperson-Al, N. et al . Spatial organisation of hominin activities at Gesher Benot Ya’aqov, Israel. Science 326, 1677–1680 (2009).þt 54.þt Gowlett, J. A. J. e early settlement of northern Europe: re history in the context of climate change and the social brain. Comptes endus Paleoevol 5, 299–310 (2006).þt 55.þt Sergant, J., Crombe, P. & Perdaen, Y. e “invisible” hearths: A contribution to the discernment of Mesolithic non-structured surface hearths. J. Archaeol. Sci. 33, 999–1007 (2006).þt 56.þt Goren-Inbar, N. et al . Evidence of hominin control of re at Gesher Benot Ya’aqov, Israel. Science 304, 725–727 (2004).þt 57.þt e ry-Parisot, I. Fuel management (bone and wood) during the Lower Aurignacian in the Pataud oc Shelter (Lower Palaeolithic, Les Eyzies de Tayac, Dordogne, France). Contribution of experimentation. J. Archaeol. Sci 29, 1415–1421 (2002).þt 58.þt Villa, P., Bon, F. & Castel, J.-C. Fuel, re and replaces in the Palaeolithic of Western. Europe. eview Archaeol 23, 33–42 (2002).þt 59.þt Cain, C. . Using burned animal bones to loo at Middle Stone Age occupation and behavior. J. Archaeol. Sci. 32, 873–884 (2005).þt 60.þt Yravedra, J., Baena, J. & Arrizabalaga, A. & Iriarte, M. J. El empleo de material oseo como combustible durante el Paleolítico Medio y Superior en el Canta brico. In Actas de la reunión cientíca: Neandertales cantábricos. Estado de la cuestión (eds. Montes Barquín, . & Lasheras Corruchaga, J. A.), 369–383 (Monografías del Museo Nacional y Centro de Investigación de Altamira 20, 2005).þt 61.þt Yravedra, J. et al . New evidence of bones used as fuel in the Gravettian level at Coímbre cave, northern Iberian Peninsula. Archaeol. Anthropol. Sci 9, 1153–1168 (2017).þt 62.þt Costamagno, S., e ry-Parisot, I., Castel, J.-C. & Brugal, J. P. Combustible ou non? Analyse multifactorielle et modèles explicatifs sur des ossements brûle s pale olithiques. In Gestion des combustibles au Pale olithique et au Me solithique: Nouveaux outils, nouvelles interpretations. UISPP XV Congress (eds. ery-Parisot, I., Costamagno, S. and Henry, A.), 65–84 (British Archaeological eports International Series, 2009).þt 63.þt Morin, E. Taphonomic implications of the use of bone as fuel. e Taphonomy of Burned Organic esidues and Combustion Features in Archaeological Contexts (Proceedings of the ound Table, Valbonne, CEPAM, Valbonne, 2010).þt 64.þt Behrensmeyer, A. . e taphonomy and paleoecology of Plio-Pleistocene vertebrate assemblages East of Lae udolf, enya. Bulletin of the Museum of Comparative Zoology 146(10), 473–578 (1975).þt 65.þt Barai, ., osell, J., Blasco, . & Gopher, A. Fire for a reason: Barbecue at Middle Pleistocene Qesem Cave, Israel. Current Anthropol. 58(suppl. 16), 314–328 (2017).þt 66.þt Meignen, L., Goldberg, P. & Bar-Yosef, O. e hearths at ebara Cave and their role in site formation processes. In ebara Cave, Mt. Carmel, Israel: e Middle and Upper Paleolithic archaeology (eds. Bar-Yosef, O. & Meignen, L), 91–122 (Part I. American School of Prehistoric esearch Bulletin 49. Harvard University, Cambridge, 2007).þt 67.þt Albert, . M., Berna, F. & Goldberg, P. Insights on Neanderthal re use at ebara Cave (Israel) through high resolution study of prehistoric combustion features: Evidence from phytoliths and thin sections. Quat Inter 247, 278–293 (2012).þt 68.þt Speth, J. D., Meignen, L., Bar-Yosef, O. & Goldberg, P. Spatial organization of Middle Palaeolithic occupation X in ebara Cave (Israel): Concentrations of animal bones. Quat. Int 247, 85–102 (2012).þt 69.þt Martínez, . & ando, J. M. Organizacio n y funcionalidad de la produccio n lítica en un nivel del Paleolítico Medio del Abric omaní. Nivel Ja (Capellades, Barcelona). Trabajos Prehistoria 58(1), 51–70 (2010).þt 70.þt Vaquero, M. & Pasto , I. e denition of spatial units in Middle Palaeolithic sites: e hearth-related assemblages. J. Archaeol. Sci. 28, 1209–1220 (2001).þt 71.þt Vaquero, M., ando, J. M. & Chaco n, M. G. Neanderthal spatial behavior and social structure: Hearth-related assemblages from the Abric omaní Middle Palaeolithic site. In Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age 2 (eds. Conard, N. J.), 367–392 (erns Verlag, Tübingen, 2004).þt 72.þt Vaquero, M. et al . The lithic assemblage of Level J. In High esolution Archaeology and Neanderthal Behavior, Vertebrate Paleobiology and Paleoanthropology Series (ed. Carbonell, E.), 189–311 (Springer, 2012).þt 73.þt Vaquero, M. et al . Time and space in the formation of lithic assemblages: e example of Abric omaní Level J. Quat. Int. 247, 162–181 (2012).


14SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ þt 74.þt Vaquero, M., Bargallo , A., Chaco n, M. G., omagnoli, F. & Saþn ñ udo, P. Lithic recycling in a Middle Paleolithic expedient context: Evidence from the Abric omaní (Capellades, Spain). Quat. Int. 361, 212–228 (2015).þt 75.þt Vaquero, M. et al . Moving things: Comparing lithic and bone rets from a Middle Palaeolithic site. J. Anthropol. Archaeol. 48, 262–280 (2017).þt 76.þt Vallverdu, J. et al . Short human occupations in the Middle Palaeolithic Level I of the Abric omaní roc-shelter (Capellades, Barcelona, Spain). J. Hum. Evol. 48, 157–174 (2005).þt 77.þt Vallverdu, J. et al . Sleeping activity area within the site structure of archaic human groups: Evidence from Abric omaní Level N combustion activity areas. Current Anthropol. 51(1), 137–145 (2010).þt 78.þt Modolo, M. & osell, J. econstructing occupational models: Bone rets in Level I of Abric omaní. Quat. Int 435, 180–194 (2016).þt 79.þt Bargallo , A., Gabucio, M. J. & ivals, F. Puzzling out a palimpsest: Testing an interdisciplinary study in level O of Abric omaní. Quat. Int 417, 51–65 (2016).þt 80.þt Gabucio, M. J. et al . Unraveling a Neanderthal palimpsest from a zooarchaeological and taphonomic perspective. Archaeol. Anthropol. Sci. 10, 197–222 (2018).þt 81.þt Blasco, . et al . Learning by Heart: Cultural patterns in the fauna processing sequence during the Middle Pleistocene. PLoS One 8 ( 2 ) , (2013).þt 82.þt Saþn ñ udo, P., Blasco, . & Ferna ndez Peris, J. Site formation dynamics and human occupations at Bolomor Cave (Valencia, Spain): An archaeostratigraphic analysis of levels I to XII (100–200 a). Quat. Int. 417, 94–104 (2016).þt 83.þt Machado, J., Molina, F. J., Herna ndez, C. M., Tarriþnñ o, A. & Galva n, B. Using lithic assemblage formation to approach Middle Palaeolithic settlement dynamics: El Salt Stratigraphic Unit X (Alicante, Spain). Archaeol. Anthropol. Sci., s12520-016-0318-z (2016).þt 84.þt Vidal-Matutano, P. Firewood and hearths: Middle Palaeolithic woody taxa distribution from El Salt, stratigraphic unit Xb (Eastern Iberia). Quat. Int 457, 74–84 (2017).þt 85.þt Bourguignon, L. et al . L’habitat mouste rien de La Folie (Poitiers, Vienne): Synthèse des premiers resultats. Paléo 14, 29–48 (2002).þt 86.þt Bourguignon, L. et al . Comple ments d’informations chronologiques sur le campement mouste rien de tradition acheule enne du gisement de La Folie (Poitiers, Vienne). Paléo 18, 37–44 (2006).þt 87.þt Leroi-Gourhan, A. & Brezillon, M. Fouilles de Pincevent. Essai d’analyse ethnographique d’un habitat Magdalenien (la Section 36), 263–385 (Paris: Editions du CNS,1972).þt 88.þt Enloe, J. G., Francine, D. & Hare, T. S. Patterns of fauna processing at Section 27 of Pincevent: e use of spatial analysis and ethnoarchaeological data in the interpretation of archaeological site structure. J. Anthropol. Archaeol. 13, 105–124 (1994).þt 89.þt Binford, L. . Nunamiut ethnoarchaeology (Academic Press, 1978).þt 90.þt Blasco, F., Montes, L. & Utrilla, P. Deux modþèè les de strategie occupationelle dans le Mousterien Tardif de la vallée de l’Ebre: les grottes de Peña Miel et Gabasa. In e last Neanderthals, the rst anatomically Modern Humans (eds. Carbonell, E. and Vaquero, M.), 289-313 (1996).þt 91.þt Discamps, E., Delagnes, A., Lenoir, M. & Tournepiche, J.-F. Human and hyena cooccurrences in Pleistocene sites: insights from spatial, faunal and lithic analyses at Camiac and La Chauverie (SW France). J. Taphon 10(3-4), 291–316 (2012).þt 92.þt osell, J. et al . A resilient landscape at Teixoneres Cave (MIS 3; Moià, Barcelona, Spain): e Neanderthals as disrupting agent. Quat. Int 435, 195–210 (2017).þt 93.þt Stiner, M. C. Competition theory and the case for Pleistocene hominincarnivore co-evolution. J. Taphon 10(3-4), 129–145 (2012).þt 94.þt Utrilla, P., Montes, L., Blasco, F., Torres, T. & Ortiz, J. E. La cueva de Gabasa revisada 15 años después: un cubil para las hienas y un cazadero para los Neandertales. In Actas de la I reunio n de cientícos sobre cubiles de hienas (y otros grandes carnívoros) en los yacimientos arqueologicos de la Península Iberica (eds. Baquedano, E. & osell, J.), 376–389 (Zona Arqueológica 13, 2010).þt 95.þt Lee, J. & Wong, D. W. S. Statistical Analysis with ArcView GIS (Wiley, New Yor, 2000).AcknowledgementsWe would like to acknowledge the Basque Government and Gipuzkoako Ondare Bildumen Zentroa for allowing us access to the fossils excavated during the Altuna team excavations. is research has received nancial support from Diputación de Gipuzkoa (Orden Foral 0210/239/2016 and 0210/209/2017). e results presented here have been partially funded by the research projects of the Spanish Science Ministry HAR2017-84997-P and the ERCCoG project (SUBSILIENCE Ref 818299) led by Ana B. Marín-Arroyo.e authors are very grateful to J. Altuna, K. Mariezkurrena and J. Yravedra for giving us access to their research databases. We would also like to thank Ruth Blasco for bringing us the opportunity to participate in this special volume on “Quaternary Taphonomy” and to the anonymous reviewers for their comments and suggestions.Author contributionsL.S.-R., A.B.-C. and J.R.-G. designed the paper. L.S.-R. performed the analyses, wrote the initial draft and prepared the gures, tables and supplementary information. L.S.-R.,J.R.-Gand A.B.M. A. obtained the funding for the Amalda project. T.K., A.B.-C, J.R.-G. and L.S.-R. compiled the stratigraphic interpretation. A.B.M.A. and L.A. performed the taphonomic study of the fauna assemblage. All the authors have interpreted and discussed the data obtained and commented on the manuscript.þCCompeting interestse authors declare no competing interests.Additional informationSupplementary information is available for this paper at Correspondence and requests for materials should be addressed to L.S.-R. Reprints and permissions information is available at Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional aliations.


15SCIENTIFIC REPOþrRTþsS | (2020) 10:8733 | Š––’•ã†‘‹ä‘”‰wväwvy~•zw{~ævxvæ|{y|zæ~ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. e images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit © e Author(s) 2020


Download Options


  • info Info

    There are only PDFs associated with this resource.

  • link PDF(s)

Cite this item close


Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.


Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.


Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.


Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.