Taphonomic and spatial analyses from the Early Pleistocene site of Venta Micena 4 (Orce, Guadix‑Baza Basin, southern Spain)

1 Vol.:(0123456789) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports Taphonomic and spatial analyses from the Early Pleistocene site of Venta Micena 4 (Orce, Guadix‑Baza Basin, southern Spain) Carmen Luzón1, Jose Yravedra 2,3*, Lloyd A. Courtenay 4, Juha Saarinen5, Hugues‑Alexandre Blain6,7, Daniel DeMiguel8,9, Suvi Viranta 10, Beatriz Azanza 11, Juan José Rodríguez‑Alba2, Darío Herranz‑Rodrigo2, Alexia Serrano‑Ramos 1, Jose A. Solano 12, Oriol Oms13, Jordi Agustí14,6,7, Mikael Fortelius 5,15,16 & Juan Manuel Jiménez‑Arenas 1,17* Venta Micena is an area containing several palaeontological sites marking the beginning of the Calabrian stage (Early Pleistocene). The richness of the fossil accumulation including species of Asian, African and European origin, makes Venta Micena a key site for the the palaeoecological and palaeoenvironmental study of southern Europe during the Early Pleistocene. Thus, research has been focused on Venta Micena 3, which was originally interpreted as a single palaeosurface associated with a marshy context, in which most of the fauna was accumulated by Pachycrocuta brevirostris. Recent excavations have unearthed a new site, Venta Micena 4, located in the same stratigraphic unit (Unit C) and in close proximity to Venta Micena 3. Here we show the first analyses regarding the taphonomic and spatial nature of this new site, defining two stratigraphic boundaries corresponding to two different depositional events. Furthermore, the taphonomic analyses of fossil remains seem to indicate a different accumulative agent than Pachycrocuta, thus adding more complexity to the palaeobiological interpretation of the Venta Micena area. These results contribute to the discussion of traditional interpretations made from Venta Micena 3. The south of the Iberian Peninsula is one of the regions in Europe with the largest number of Early Pleistocene sites. These sites are crucial to the study of the first Early Pleistocene human migrations into Europe. In the Guadix-Baza Basin, the localities of Barranco León (BL) and Fuente Nueva 3 (FN3) have provided clear evidence of human activity dated over ca. 1 million years ago, with the presence of lithic industries1, bones presenting cut OPEN 1Departamento de Prehistoria y Arqueología, Universidad de Granada, Granada, Spain. 2Departamento de Prehistoria, Universidad Complutense, Madrid, Spain. 3C.A.I. Arqueometría, Universidad Complutense, Madrid, Spain. 4Departamento de Ingeniería Cartográfica y del Terreno, Escuela Politécnica Superior de Ávila, Universidad de Salamanca, Ávila, Spain. 5Department of Geosciences and Geology, University of Helsinki, Helsinki, Finland. 6Institut Català de Paleoecologia Humana i Evolució Social (IPHES-CERCA), Tarragona, Spain. 7Departament d’Història i Història de l’Art, Universitat Rovira i Virgili, Tarragona, Spain. 8ARAID/Departmento de Ciencias de la Tierra (Paleontología), Universidad de Zaragoza, Zaragoza, Spain. 9Institut Català Paleontologia M. Crusafont ICP, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain. 10Department of Anatomy, University of Helsinki, Helsinki, Finland. 11Departamento de Ciencias de la Tierra (Paleontología), Universidad de Zaragoza/Instituto Universitario de Investigación en Ciencias Ambientales de Aragón (IUCA), Zaragoza, Spain. 12Departamento de Prehistoria y Arqueología, Universidad de Sevilla, Sevilla, Spain. 13Departament de Geologia, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain. 14ICREA, Barcelona, Spain. 15Finnish Museum of Natural History, Helsinki, Finland. 16Visiting Scholars Program, University of Granada, Granada, Spain. 17Instituto Universitario de la Paz y los Conflictos, Universidad de Granada, Granada, Spain. *email: [email protected]; [email protected] 2 Vol:.(1234567890) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ and percussion marks2, as well as human remains3. Together with these sites, Venta Micena 3 (VM3) (Fig.1) is also a well-known, palaeontological locality with no presence of human activity or anthropogenic evidence. Nevertheless, VM3 is strongly characterised by palaeoecological and palaeontological data in which animals of Asian, African and European origin converge4–9. The great faunal diversity of Venta Micena (VM) sites, and the quality of their remains, have led to the definition of some Pleistocene species for the first time10,11. Nevertheless, the VM sites are mainly notable because of their importance in characterizing the palaeolandscape and palaeoclimate of Southern Europe ca. 1.5 million years ago7. According to several studies carried out in VM3, the fossil accumulation is currently thought to have been produced by the giant hyaena Pachycrocuta brevirostris12–16, which makes VM3 one of the primary sites for the study of this super-scavenger’s behavioural attributes16. The presence of P. brevirostris has been frequently documented throughout Eurasian sites from the Early-Middle Pleistocene, including key sites such as Dmanisi (Georgia)17, Vallparadis (Spain)18 and Zhoukoudian (China)19. In contrast, however, VM3 is devoid of human presence and activity20,21, despite the controversial findings reported from the 1980s22–27. The present study describes Venta Micena 4 (VM4), a deposit located in proximity to VM3, and with very similar geological and sedimentological characteristics28. Evidence described in VM4, however, has presented contradictory data, namely regarding the stratigraphic and taphonomic data as originally described in VM3, as will be analyzed. Figure1. Geographic and topological location of Venta Micena 4. 3 Vol.:(0123456789) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ Geological and chronological context VM sites are located in the Guadix-Baza Basin, in the southeast of the Iberian Peninsula, forming part of the Baetic Chain (Fig.1). Located on the northeastern boundary of the Guadix-Baza basin, the Orce region was subjected to the lowstand and highstand dynamics of a large saline lake that dominated the basin. In this sense, it is relevant to point out that while the lake’s shoreline recedes, numerous fresh groundwater ponds would have emerged (Ref.28, Supplementary Notes 1, Supplementary Fig.S1). VM has been chronologically framed within the Matuyama magnetic Chron, between Jaramillo and Olduvai. The VM biozone has additionally been classed as MmQ-2, characterized by the occurrence of Allophaiomys ruffoi (originally, A. pliocaenicus), as well as the absence of suids. These finds have provided an approximate biochronological age of 1.6–1.5 Ma29. VM is located in a white micritic limestone sedimentary environment, having a lateral continuity of more than one kilometer (Supplementary Notes 1). This limestone is part of a sedimentary succession28,30 that reveals a landscape made by relatively freshwater shallow lacustrine and palustrine environments (pools and wetlands respectively), not connected with the large saline lake that occupied most of the region [Ref.28, Supplementary Notes 1, Supplementary Fig.S1]. The levels where palaeontological remains have been found are located in a unit between 80 to 120cm thick (Supplementary Notes 1, Supplementary Fig.S2). In general, vertebrate-rich levels are invertebrate-poor, finding the scarce ostracods and gastropods strongly recrystallized due to early carbonate dissolution. VM4 is one of the most prolific sites of the Orce region, currently consisting in a 39 m2 excavated window, where abundant fossil remains have been discovered in excavations starting in 2005. The VM sites seem to be slightly older than other archaeological sites in the Orce region, such as BL (1.4Ma) and FN3 (1.2Ma)1,2. As an example, this is supported by the occurrence of bovid Soergelia minor, which is absent in both BL and FN331. On the other hand, the big stenoid horse (Equus süssenbornensis), as well as Ammotragus europeus, are present in BL and FN331 and are absent. This is consistent with VM presenting an older biochronological age. Finally, VM4 is situated 300m southwest of VM3 in the same stratigraphical unit (Unit C, Supplementary Fig.S1), having a synchronic in age. Results Palaeontology. Although the majority of small vertebrates from VM3 and 4 are still under investigation, other VM localities such as VM1 and VM2 (Supplementary Tables1, Supplementary TablesS1a and S1b), as well as other sites of a similar age in the Guadix-Baza Basin (Cañada de Murcia 1, Fuente Nueva 2, Orce 7), are characterized by a sharp decrease in the diversity of small vertebrate associations when compared with other biozones. Rodent communities are dominated by the vole, Allophaiomys ruffoi (89%), a southern and more archaic variant of the Central-European A. pliocaenicus. This species is accompanied by representatives of the genus Apodemus (Apodemus sylvaticus in VM1 and, A. mystacinus in VM232,33), Castillomys rivas32–35 and Eliomys intermedius (VM1 and VM2)32, Oryctolagus cf. lacosti (VM1 and VM2) and Prolagus calpensis (VM1)32,36,37. Furthermore, Hystrix refosa (= H. major) is also associated with this species in VM232 and Galemys pyrenaicus in VM1 and VM238. In VM1, Galemys is found associated with Asoriculus gibberodon34, both of which are indicative of aquatic environments. A. gibberodon probably also indicates the presence of patchy landscapes adjacent to water bodies bushes and open lands39. The herpetofauna of VM1 and VM2 is represented by Discoglossus sp., Pelophylax cf. perezi, Testudines indet., Timon sp., Lacertidae indet. (cf. Podarcis sp.), and Ophidia indet36,40,41. With the exception of lacertids (Timon and Podarcis), the remaining taxa documented are indicative of sunny aquatic environments, while Timon and Podarcis seem to be suggestive of patchy landscapes, open areas and woody zones34. The large mammal associations identified so far in the palaeontological levels of VM4 are comprised of 21 large mammal species, including those belonging to the Felidae, Hyaenidae, Canidae, Ursidae, Elephantidae, Rhinocerotidae, Equidae, Bovidae and Cervidae families (see Supplementary Notes 2). Many of the species recorded at VM3 are also present in VM4 (Table1, Supplementary Notes 2, Supplementary TableS2), concurring that both sites contain species of African, Asian and European origin4–9. Both sites indicate a similar palaeoclimatic setting, dominated by warm and drier conditions than those suggested by the fauna present at BL and FN334,42. The bulk of the fauna is represented by equids alongside Mammuthus meridionalis, Stephanorhinus etruscus and Bison sp. Species from more wooded environments are also frequent, such as two cervids, as well as others from environments close to water sources, such as Hippopotamus antiquus (Supplementary Notes 2). Similarly to VM3, VM4 provides a large number of carnivore remains, including; hyenids (P. brevirostris), felids (Homotherium latidens, Megantereon cultridens, Panthera cf. gombaszoegensis, Lynx cf. pardinus) and canids (Xenocyon (= Lycaon) lycaonoides, Canis mosbachensis, Vulpes alopecoides) (Supplementary Notes 2). Taphonomy. The bulk of fauna from VM4 is represented by herbivores, comprising over 90% of the fossil record. E. altidens is the most abundant taxon, both in frequency of remains and number of individuals, contributing to 24.4% of identified herbivores and 18.5% of all individuals at the site (Table1). Cervids are also an important component of the assemblage, followed by bison, caprines and megaherbivores, such as elephants, rhinoceroses and hippopotamuses (Table1). Pachycrocuta are the most representative carnivore in the assemblage, followed by large felids, canids (wild dog-like canids, foxes and wolves), bears, and finally smaller felids such as lynx (Table1). Mortality patterns reveal that individuals of all ages are recorded in VM4, with a relatively higher amount of non-adult individuals. Among the large-sized species present, the number of non-adults among elephants and rhinoceroses is similar to, or higher than, that of adults (Table1). Among medium-sized species, E. altidens, Hemibos aff. gracilis and Praemegaceros cf. verticornis are represented by similar amounts of adults and nonadults. The only exception to this is in the case of bison, where non-adults only represent 33% of the individuals (Table1). Finally, among smaller species such as caprids and small cervids, adults outnumber new-born and 4 Vol:.(1234567890) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ juvenile individuals (Table1). Nevertheless, the total number of individuals in each species is too low to draw reliable conclusions on the resulting patterns. From this perspective, a prime-dominant, L or U shaped mortality profile cannot be clearly discerned. Regarding skeletal profiles, teeth are by far the most abundant anatomical elements, comprising of 36.9% of the faunal remains (Supplementary TablesS4, S4a). Species of size classes 1 (25–50kg) and 5 (> 1000kg), such as elephants, rhinoceroses and hippopotamuses, are not representative due to the scare number of remains (Table1 and Supplementary TablesS4, S4a). This is also the case for carnivores. Species of size class 2 (50–125kg), such as M. rhenanus, C. alba or S. minor, show biased skeletal profiles, with a predominance of teeth (55% of the sample), as well as anterior limbs (scapulae, humerii, radii, carpal bones and metacarpals). This is almost twice the number of posterior limbs (pelves, femora, tibiae, patellae, tarsal bones and metatarsals) (Supplementary TablesS4, S4a). Sizes 3, 3a and 3b species (125–500kg), on the other hand, are well represented by all anatomical elements. Cranial elements, together with teeth and mandibles, predominate, accounting for 30% of determinable bones, while axial and appendicular elements are also well represented with frequencies both higher than 20% (Supplementary TablesS4, S4b). Nevertheless, although all skeletal regions are represented, in some cases a certain Table 1. List of the Venta Micena 4 remains represented by Number of Identifiable Specimens (NISP) and Minimum Number of Individuals (MNI). Abbreviation for the mortality profiles; S: senile, A: adult, J: juvenile, I: infant. %Total represents all macrovertebrates (both carnivores and herbivores). aSee in Supplementary TableS3 for caption of species and relation to animal size. See in Supplementary TableS4. NISP % Herbivores % Total MNI % Herbivores % Total S/A/J/I % not adults Mammuthus meridionalis 4 0.3 0.2 2 4.9 3.6 1–0–1–0 50 Stephanorhinus etruscus 14 0.9 0.9 5 12.2 8.9 0–1–1–3 80.0 Equus altidens 119 8.0 7.4 10 24.4 17.9 1–4–3–2 50.0 Equus sp. 5 0.3 0.3 2 4.9 3.6 0–0–1–1 100.0 Hippopotamus antiquus 17 1.1 1.1 1 2.4 1.8 0–1–0–0 0.0 Bison sp. 43 2.9 2.7 3 7.3 5.4 0–2–1–0 33.3 Hemibos aff. gracilis 4 0.3 0.2 1 2.4 1.8 0–1–0–0 0.0 Sorgelia minor 13 0.9 0.8 1 2.4 1.8 0–1–0–0 0.0 Capra alba 19 1.3 1.2 3 7.3 5.4 1–1–1–0 33.3 Praemegaceros cf. verticornis 61 4.1 3.8 7 17.1 12.5 2–2–2–1 42.9 Metacervocerus rhenanus 35 2.3 2.2 6 14.6 10.7 1–3–1–1 33.3 Cervidae indet 1 0.1 0.1 Herbivore indet. size 0* 3 0.2 0.2 Herbivore indet. size 1* 1 0.1 0.1 Herbivore indet. size 2* 71 4.8 4.4 Herbivore indet. size 3* 96 6.4 6.0 Herbivore indet. size 3a* 55 3.7 3.4 Herbivore indet. size 3b* 198 13.3 12.3 Herbivore indet 733 49.1 45.6 Total Herbivore 1492 100.0 92.7 41 100.0 73.2 % Carnivore % Carnivore Canis mosbachensis 15 17.4 0.9 1 9.1 1.8 0–1–0–0 Canidae 18 20.9 1.1 Xenocyon (= Lycaon) lycaonoides 8 9.3 0.5 218.2 3.6 0–1–0–0 Vulpes alopecoides 1 1.2 0.1 1 9.1 1.8 0–1–0–0 Pachycrocuta brevirostris 15 17.4 0.9 2 18.2 3.6 1–1–0–0 Felidae 1 1.2 0.1 Lynx sp.3 3.5 0.2 1 9.1 1.8 0–1–0–0 Homotherium latidens 1 1.2 0.1 1 9.1 1.8 0–1–0–0 Megantereon cultridens 1 1.2 0.1 1 9.1 1.8 0–1–0–0 Panthera cf. gombaszoegensis 1 1.2 0.1 1 9.1 1.8 0–1–0–0 Ursus etruscus 11 12.8 0.7 1 9.1 1.8 0–1–0–0 Carnivora 11 12.8 0.9 Total carnivore 86 100.0 5.3 11 100.0 19.6 Lagomorpha 8 0.5 2 3.6 0–2–0–0 Testudines 14 0.9 1 1.8 0–1–0–0 Aves 9 0.6 1 1.8 0–1–0–0 Total 1609 100.0 56 100.0 5 Vol.:(0123456789) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ bias is observed. An example of this can be found in the disproportionate amount of posterior limb remains, as well as a modest number of anterior limb specimens (Supplementary TablesS4, S4b), which contrasts with the more balanced representation of these elements observed in VM37. The VM4 fossil remains show a moderate fragmentation. Only 36% of remains measure less than 3cm (Table2), with more than 55% of long bones presenting green fractures (Table2). In addition, some bones have been documented in anatomical connection. Examples include the attached humerus-radius of S. etruscus (NE area of the site), as well as a femur-tibiae, fibula and talus of this same species (SW area of the site). Also, a set of eight cervical vertebrae of M. meridionalis was retrieved on the western edge of the site. The almost complete forelimb of a X. lycaonoides individual was also found in the centre of the site, as well as two complete hindlimbs of the same species (NW corner of the site). Finally, two hemipelves belonging to an E. altidens individual was found towards the west, all of which present a good representation of bones found in anatomical connection at VM4. Bone surfaces are also well preserved, with only 31% of the remains presenting poor preservation (Table2). Such a high preservation rate can be the result of several factors, such as the low occurrence of weathering. From this perspective, only 9% of faunal remains have been observed to reach weathering stages 1–243, indicating short or/and low subaerial bone exposure. Diagenetic alterations are also rare, and are often limited to manganese oxide stains and calcite concretions (Table2). Evidence of hydric alterations are limited to abrasion (which affects a 40% of specimens), without the presence of rounded bones. Nevertheless, only 33% of these specimens show an intense degree of abrasion, implying hydric alterations to be notably low. On the other hand, 25% of the remains show alterations of biological origin, including biochemical corrosion as well as root-marks. Nevertheless, in most cases the impact of these alterations is low to moderate. Finally, carnivore alterations were only observed on 4.5% of the well-preserved bones (Table2). Only 3 bones show 3–6 tooth marks, while the remainder of tooth marks bones present no more than 2. Furthermore, salivary and gastric alterations are absent, with a similar lack of coprolites. Regarding tooth mark typologies, pits predominate over scores, accounting for 88% of the documented tooth marks (Table2). Most of the tooth marks are distributed on long bone diaphyses and axial elements (Table3). Extensive bone deletion is not frequent, but it has also been observed, with only 4.5% of the tooth-marked bones presenting evidence of furrowing (Table2). Likewise, several taphotypes (1, 3, 4, 5, 6, 11 and 15), proposed by Ref.44, have been documented on long bones with crushed epiphyses (Table4). Numerous long bones additionally present evidence of green fractures (Table2), with the additional presence of different notches types (Table5). Finally, complete long bones are rare, while bones that preserve > 50% of the total shaft circumference are also scarce (Table6). Table 2. Taphonomical characteristics of Venta Micena 4. Taphonomic characteristics NISP % %Representation Total amount of specimens 1609 Total amount of specimens excluding teeth 1374 Fragmentation Bones < 3cm 585 36.4 Bones 3.1–5cm 382 23.7 Bones > 5.1–9.9cm 382 23.7 Bones > 10cm 260 16.2 Long bone with green fracture 186 55.1 Long bone with dry fracture 152 44.9 Bone surfaces Badly preserved specimens 427 31.1 % respect to total amount of specimens excluding teeth Weathering Weathering stage 1–2 148 9.2 % respect to total amount of specimens Water alteration Abrasion 642 39.9 % respect to total amount of specimens Light stage abrasion 189 29.4 Intermediate stage abrasion 242 37.7 Intense stage abrasion 211 32.9 Calcitic concretions 155 9.6 Oxide staining 307 19.1 % respect to total amount of specimens Biochemical alterations 402 25.0 % respect to total amount of specimens Carnivore activity Bones with tooth marks 43 4.5 % Excluding bones with badly preserved bones and teeth Bones with tooth Marks with pits only 38 88.4 Bones with tooth marks with scores only 7 16.3 Bones with tooth marks with both pits and scores 4 9.3 Bones with punctures only 1 2.3 Bones with furrowing 17 4.5 Long bones with furrowing 9 5.8 % with respect to long bones Rodent tooth marks 1 6 Vol:.(1234567890) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ Table 3. Bones with tooth marks. Epd: Distal Epiphyses. Shaft: Diaphyses. Bones with tooth marks Species size class 2 3 3a 3b 5 Indet Vertebrae 1 Scapulae 1 Humerii 1 shaft 5 shaft 1 shaft Radii 1 epd 1 Metacarpals 2 shaft 1 epd Pelves 1 Femora 2 shaft Tibiae 6 shaft 1 epd Metatarsals 2 shaft Calcanei 1 Long bone shafts 1 1 8 1 Indeterminate 1 3 1 NR total tooth marks 5 1 3 31 1 2 Table 4. Taphotypes observed on different long bones according to (44). Taphotype No. Skeletal element NR total % 0Tibia 2 10.5 1 Humerus 1 5.3 3Tibia 5 26.3 4 Humerus 1 5.3 5Tibia 1 5.3 6 1 Radius, 1 Tibia 2 10.5 11 Radii 1 5.3 15 1 Femur/3 Humerii/1 Radius/1 Tibia 6 31.6 Table 5. Evidences of notches observed in VM4. Notches NISP % Single 8 14.3 Opposing 3 5.4 Incomplete type A 11 19.6 Incomplete type B 2 3.6 Incomplete type C 3 5.4 Double 3 5.4 Double opposing 4 7.1 Pseudonotch 1 1.8 Micronotch 15 26.8 Multiple 6 10.7 Total 56 100 Table 6. Degree of total circumference and fragment length vs complete bone length. Only green fractured bones are included. Degree of circumference NISP % % Length respect to the total length of the bone % < 25% 126 67.75 136 73.1 25–50% 33 17.75 40 21.5 > 50% 27 14.5 10 5.4 Total 186 100.0 186 100.0 7 Vol.:(0123456789) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ Spatial analyses. Artificially intelligent systems for the identification of discrete fossiliferous levels revealed 2 distinct and independent bone concentrations levels that could be clearly identified across the entire 39 m2 extension of the VM4 site. These levels have been subsequently named Level I (VM4-I) and Level II (VM4-II). VM4-II is located directly above VM4-I, approximately 200–230cm below the surface with a relatively homogeneous horizontal spread and slight NE-SW dip. VM4-I, on the other hand, is located approximately 250–280cm below the surface, and is observed to be a much denser horizontal plane. Each bone accumulation has a local thickness of generally < 30cm. VM4-I and VM4-II are vertically scattered 50 and 30cm (i.e., stratigraphic heights between 0 and 50cm and between 60 and 90cm, respectively). The separation between VM4-I and VM4-II is defined by a 10cm interval. When evaluating the quality of these defined levels, Random Forest (RF) algorithms proved to be the most confident models when associating each of the finds to their corresponding fossiliferous levels. RF, on average, presents a confidence of 100 [+ 0.0, −0.0]% probability when making new predictions. Support Vector Machines (SVM), on the other hand, saw a slight drop in confidence, assigning most finds a class probability of 99.9 [+ 0.001, −0.003]%. When considering the performance of both models in a system, both algorithms were successful in assigning 4219 fossils to a particular level; 3482 fossils were assigned to Level I, 737 to Level II, and 76 remains were considered indeterminable with < 80% confidence when assigning finds to any particular level. Among the classified remains, both SVM and RF agreed on the allocation of 97.5% of these remains. When disagreement did occur, RF appeared to be the most decisive algorithm at least 87.1% of the time. Detailed evaluation of agreement-disagreement rates additionally reveals an inter-rater reliability of 0.85, with near perfect agreement according to Cohen’s κ. Under this premise, while RF is in general a more confident classifier, the use of both algorithms in combination provides a more robust overall classification of the entire site (Fig.2). Detailed statistical analyses of each of these levels reveal VM4-I to present a strong concentration of faunal remains (Fig.3a), with the highest accumulated density of fossil remains per m2. Similarly, VM4-II shows higher concentrations of remains, which are oriented towards the NE (Fig.3b), while VM4-I is slightly more spread out. Overall, Monte Carlo tests for Complete Spatial Randomness (CSR) reveal inhomogeneous distribution patterns for both VM4-I (χ2 = 3297, p = 2e−04) and VM4-II (χ2 = 1582, p = 2e−04), while few quadrats comply with CSR (Fig.4c,d). Upon analysing spatial correlations with theoretical K(r) functions of an inhomogeneous Poisson process, both VM4-I and VM4-II can be seen to present general tendencies for more cluster-like patterns, as confirmed by the centered L(r) function (Fig.4). While VM4-II shows slight tendencies towards a regular point process, this is likely due to the smaller sample size and lower concentrations across the overall surface area. It is worth noting that Hopkins–Skellam tests are able to confirm that both levels present strong tendencies towards clustering across the overall spread of the spatial window (VM4-I: A = 0.01, p < 2.2e−16; VM4-II: A = 0.06, p < 2.2e−16). Upon quantifying the location of clusters through density based pattern recognition algorithms, 17 clusters were detected in VM4-I and 12 clusters in VM4-II (Fig.5). When considering orientation patterns for fossil remains, only 1396 fossils were documented with orientation values (VM4-I: n = 1125; VM4-II: = 271). Nevertheless, in both levels preferential orientations towards the NE (VM4-I = 34.65°; VM4-II = 45.33°) have been documented and calculated to be of notable importance (see Supplementary Methods). When analysing orientation patterns across the site (Fig.6), general trends reveal most clusters to share a similar central tendency, with most clusters being oriented between the NNE and the ENE. Two clusters in VM4-II show exceptions to this rule with a slight tendency towards the NW and NNW, however this only represents 8% of the total sample for this level. Similarly, with the exception of 3 clusters (2 in VM4-I: Figure2. Scatter plot showing the spatial distribution of levels VM4-I (dark green) and VM4-II (brown), as identified using artificially intelligent systems. Black points indicate indeterminable points with < 80% confidence when being assigned to a level. 8 Vol:.(1234567890) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ 34% of the sample; 1 in VM4-II: 36%), preferential orientations are strong across the entire site (See orientation uniformity data from Supplementary Methods, Supplementary TableS5). With regard to the general slope of fossil finds, only 4% present extreme azimuth values over 45°, while 79% of finds have been recorded relatively flat along the surface (Fig.6). When combining information, VM4 in general therefore presents a strong tendency for relatively flat slopes and NE–SW orientations, likely conditioned by the natural topography of each palaeosurface (Fig.7). Needless to say, when considering the natural topography of each level, gravity is the likely cause for the observed patterns at VM4 (see Supplementary Methods, Supplementary TableS5). Discussion The present study of VM4 reports evidence suggesting this site to be a palimpsest of various events, distributed over several palaeosurfaces28. This is corroborated by the vertical and spatial distribution of faunal remains throughout the stratigraphic sequence (Figs.2, 3, 4, 5). Taphonomic information has additionally provided an insight into the rate at which fossils were buried, revealing VM4 to be in the context of a series of short-time events, followed by rapid sedimentation. This is supported by the low degree of weathering (Table2, in accordance with Ref.43), low tooth mark frequencies, and also by the presence of some remains found in anatomical connection. From a different perspective, the absence of rounded bone surfaces helps confirm fluvial currents to not be responsible for the accumulations present at VM4 (Table2). Nevertheless, while hydraulic activities did not move the remains, sedimentary abrasion has been observed to have affect bone surfaces, product of circulating water moving mobile sediments over the bones. While these currents were not strong enough to remove osteological remains, sedimentary abrasion has had an impact on bone surface preservations, resulting in the poor preservation rates observed across 31.1% of specimens. As seen by the presence of tooth marks (Fig.8), furrowing (Table2), and notches (Table5), the influence carnivores had in the formation of VM4 is undeniable. Although insufficient information is currently available to discern the precise carnivore agencies present at this site, some important conclusions can be drawn about the activity of carnivores and their role in the formation of the fossil assemblage. In general, while the impact carnivores had on VM4 is notable, carnivore activity in general can be considered of a low intensity. This can be seen by low tooth mark frequencies, the reduced number of bite marked bones (4.5%, Table2), the low number of tooth marks per bone (< 2 marks per bone, Table2), the absence of digested bones and salivary alterations, the absence of coprolites, and the moderate percentages of bones with furrowing (4.5%, Table2). These observations contrast with the taphonomic data present at other P. brevirostris bone assemblages, such as Vallparadís45, Zhoukoudian19, VM312,13 and Fonelas P-146. The same can be said of other extant Figure3. Density (A, B) and heat (C, D) maps for Pearson residual counts (for each of the level VM4-I (A, C) and VM4-II (B, D). 9 Vol.:(0123456789) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ carnivore accumulations in South African sites47, among others48–51. In light of this, it can be concluded that VM4 should not be considered a den site, nor product of Pachycrocuta activities, as described in the case of VM37,12–15. The taphonomic evidence described in VM4 also differs greatly from bone accumulations typically associated with carnivore dens or open air rendezvous sites. Bone accumulations from hyaena dens, for example, tend to be characterized by high frequencies of tooth marked bones, as well as a high frequencies of tooth marks per specimen. Similarly, other diagnostic criteria for detecting intensive hyaena activities include; gastric alterations, high amounts of furrowing, absence of epiphyses, abundance of bone cylinders, as well as remains of infant carnivores47,49,52,53. This profile does not fit the case of VM4, considering the low tooth mark frequencies, the low number of tooth marks per specimen, as well as the absence of coprolites, digested bone, and general salivary alterations (Table2). When compared with the accumulations produced by felids, both leopards50,54 and lions51 usually leave complete skeletal profiles, while elements of the axial skeleton are often well represented. Felids are also known to leave carcasses in anatomical connection, leaving most bones complete47,50,51,54, presenting low tooth mark frequencies51, similar to the frequencies described here in VM4. Nevertheless, VM4 differs from felid assemblages as seen with skeletal profiles not dominated by axial bones, while most faunal remains are not found in anatomical connection. Similarly, the VM4 assemblage presents numerous specimens with notches (Table5), a feature uncharacteristic of felid activities, while taphotype number 15 (bone cylinders) is also present. Finally, pits clearly predominate over scores, another feature uncharacteristic of felids (Table2). Beyond these comparisons, the VM4 assemblage is also characterized by the presence of all skeletal elements, with a slight predominance of teeth (Supplementary TablesS4, S4a, S4b). Similarly, bone modifications characteristic of carnivores are abundant, including tooth marks (Fig.8), notches (Fig.8), taphotypes and green fracture planes. All the aforementioned evidence can be found in kill sites, where some carnivores hunt their prey, others scavenge, and others disperse the remains. To this effect, VM4 would have been a place used recurrently by different carnivores to hunt, presenting primary access to their prey. This scenario would, throughout time, produce a palimpsest comprised of periodic hunting events in the same space. In addition although coprolites are absent from VM4, this may be due to the fact that feces are no always present in recent kill sites55,56. Figure4. Empirical (black) and theoretical (red) spatial correlation functions for Venta Micena 4 levels I and II using Ripley’s K and Besag’s L. Empirical functions account for border correction estimates while both for K and L the inhomogeneous variants of these tests were performed. 16 Vol:.(1234567890) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ 39. Popov, V. V. Late Pliocene Soricidae (Insectivora, Mammalia) from Varshets (North Bulgaria). Acta Zool. Cracov. 46, 43–72 (2003). 40. Bailon, S. Amphibiens et reptiles du Pliocène et du Quaternaire de France et d’Espagne: mise en place et évolution des faunes. (Thèse 3ème cycle, Université de Paris VII, 1999). 41. Felix, J. & Montori, A. Los anfibios y los reptiles del Plio-Pleistoceno de la Depresión de Guadix-Baza (Granada). Paleontologia i Evolució 1, 283–286 (1987) (Memoria Especial). 42. Blain, H. A., Bailon, S., Agustí, J., Martínez-Navarro, B. & Toro, I. Paleoenvironmental and paleoclimatic proxies to the Early Pleistocene hominids of Barranco León D and Fuente Nueva 3 (Granada, Spain) by means of their amphibian and reptile assemblages. Quat. Int. 243, 44–53 (2011). 43. Behrensmeyer, A. K. Taphonomic and ecologic information from bone weathering. Paleobiology 4, 150–162 (1978). 44. Domínguez-Rodrigo, M. et al. A new methodological approach to the taphonomic study of paleontological and archaeological faunal assemblages: A preliminary case study from Olduvai Gorge (Tanzania). J. Archaeol. Sci. 59, 35–53 (2015). 45. Madurell-Malapeira, J., Alba, D. M., Minwer-Barakat, R., Aurell-Garrido, J. & Moyà-Solà, S. Early human dispersals into the Iberian Peninsula: A comment on Martínez etal. (2010) and Garcia etal. (2011). J. Hum. Evol. 62, 169–173 (2012). 46. Arribas, A. et al. Un nuevo yacimiento de grandes mamíferos villafranquienses en la Cuenca de Guadix-Baza (Granada): Fonelas P-1, primer registro de una fauna próxima al límite Plio-Pleistoceno en la Península Ibérica. Bol. Geol. Min. 112, 3–34 (2001). 47. Brain, C. K. The Hunters or the Hunted? An Introduction to African Cave Taphonomy (Chicago University Press, 1981). 48. Pobiner, B. New actualistic data on the ecology and energetics of hominin scavenging opportunities. J. Hum. Evol. 80, 1–16 (2015). 49. Cruz-Uribe, K. Distinguishing hyaena from hominid bone accumulations. J. Field Archaeol. 18, 467–486 (1991). 50. Arriaza, M. C., Aramendi, J., Maté-González, M. A., Yravedra, J. & Stratford, D. Characterising leopard as taphonomic agent through the use of micro-photogrammetric reconstruction of tooth marks and pit to score ratio. Hist. Biol. 33, 176–185 (2019). 51. Arriaza, M. C., Domínguez-Rodrigo, M., Yravedra, J. & Baquedano, E. Lions as bone accumulators? Paleontological and ecological implications of a modern bone assemblage from Olduvai Gorge. PLoSOne 11(5), e0153797. https:// doi. org/ 10. 1371/ journ al. pone. 01537 97 (2016). 52. Marean, C. W. & Kim, S. Mousterian large mammals from Kobech Cave. Curr. Anthropol. 39, 79–113 (1998). 53. Brugal, J. P. & Fosse, P. Carnivores et Hommes au Quaternaire en Europe de l’Ouest. Rev. Paléobiol. 23, 575–595 (2004). 54. Ruiter, D. & Berger, L. Leopards as taphonomic agents in dolomitic caves & implications for bone accumulations in the hominidbearing deposits of South Africa. J. Archaeol. Sci. 27, 665–684 (2000). 55. Yravedra, J., Lagos, L. & Bárcena, F. A Taphonomic study of wild wolf (Canis lupus) Modification of horse bones in Northwestern Spain. J. Taphonomy 9, 37–65 (2011). 56. Arriaza, M. et al. Striped hyenas as bone modifiers in dual human-to-carnivore experimental models. Anthropol. Archaeol. Sci. 11, 3187–3199 (2019). 57. Rodríguez Alba, J. J., Linares-Matás, G. & Yravedra, J. First assessments of the taphonomic behaviour of jaguar (Panthera onca). Quat. Int. 517, 88–96 (2019). 58. Yravedra, J., Andrés, M. & Domínguez Rodrigo, M. A taphonomic study of the African wild dog (Lycaon pictus). Archaeol. Anthropol. Sci. 6, 113–124 (2014). 59. Moclán, A., Domínguez Rodrigo, M. & Yravedra, J. Classifying agency in bone breakage: An experimental analysis of fracture planes to differentiate between hominin and carnivore dynamic and static loading using machine learning (ML) algorithms. Archaeol. Anthropol. Sci. 11, 4663–4680 (2019). 60. Courtenay, L. et al. Combining machine learning algorithms and geometric morphometrics: A study of carnivore tooth marks. Palaeogeogr. Palaeoclimatol. Palaeoecol. 522, 28–39 (2019). 61. Courtenay, L. A., Herranz-Rodrigo, D., González-Aguilera, D. & Yravedra, J. Developments in data science solutions for carnivore tooth pit classification. Sci. Rep. 11, 10209 (2021). 62. Pole, A., Gordon, I. J., Gorman, M. L. & Macaskill, M. Prey selection by African wild dogs (Lycaon pictus) in southern Zimbabwe. J. Zool. 262, 207–215 (2004). 63. Lagos-Abarzuza, L. Ecología del lobo (Canis lupus), del poni salvaje (Equus ferus atlanticus) y del ganado vacuno semiextensivo (Bos taurus) en Galicia (Universidad de Santiago de Compostela, 2013). 64. Mitchell, B. L., Shenton, J. B. & Uys, J. C. M. Predation on large mammals in the Kafue National Park, Zambia. Zool. Afric. 1, 297–318 (1965). 65. Kruuk, H. The Spotted Hyena: A Study of Predation and Social Behavior (Chicago University Press, 1972). 66. Stiner, M. C. The use of mortality patterns in archaeological studies of hominid predatory adaptations. J. Anthropol. Archaeol. 9, 305–351 (1990). 67. Schaller, G. B. The Serengeti Lion (University of Chicago Press, 1972). 68. Fitzgibbon, C. D. & Lazarus, J. Antipredator behavior of Serengeti ungulates. In Serengeti II: Dynamics, Conservation and Management of an Ecosystem (eds. Sinclair, A. R. E. & Arcese, P.) 274–296 (University of Chicago Press, 1995). 69. Rodriguez-Gomez, G., Palmqvist, P., Ros-Montoya, S., Espigares, M. P. & Martinez-Navarro, B. Resource availability and competition intensity in the carnivore guild of the Early Pleistocene site of Venta Micena (Orce, Baza Basin, SE Spain). Quat. Sci. Rev. 164, 154–167 (2017). 70. Espigares, M. P. Análisis y modelización del contexto sedimentario y los atributos tafonómicos de los yacimientos pleistocénicos del borde nororiental de la cuenca de Guadix-Baza (Universidad de Granada, 2010). 71. Domínguez-Rodrigo, M. et al. Autocthonous anisotropy of archaeological materials by the action of water: Experimental and archaeological reassessment of the orientation patterns at the Olduvai sites. J. Archaeol. Sci. 41, 44–68 (2014). 72. Domínguez-Rodrigo, M. et al. Fluvial spatial taphonomy: A new method for the study of post-depositional processes. Archaeol. Anthropol. Sci. 10, 1769–1789 (2018). 73. Domínguez-Rodrigo, M. et al. Autochtony and orientation patterns in Olduvai Bed I: A re-examination of the status of postdepositional biasing of archaeological assemblages from FLK North (FLKN). J. Archaeol. Sci. 39, 2116–2127 (2012). 74. Luzón, C. Análisis tafonómico preliminar del Corte IV del yacimiento de Venta Micena (Orce, Granada, España) (Universitat Rovira i Virgili, 2016). 75. Gibert J. & Caporicci, R. Tafonomía y paleoecología del yacimiento de Venta Micena. In Los restos humanos de Orce y Cueva Victoria (eds. Gibert J., Campillo D. & García-Olivares E.) 241–268 (Diputació de Barcelona, 1989). 76. Garrido, G. Paleontología sistemática de grandes mamíferos del yacimiento del Villafranquiense superior de Fonelas P-1 (Cuenca de Guadix, Granada) (Universidad Complutense de Madrid, 2006). 77. Pales, L. & Lambert, C. Atlas ostéologique des mammiféres. I. Membres (2 fascicules: Herbivores et Carnivores) (CNRS, 1971). 78. Pales, L. & García, M. A. Atlas ostéologique des mammiféres. II. (2 fascicules: Carnivores et Homme, Herbivores) (CNRS, 1981). 79. Brain, C. K. The probable role of leopards as predators of the Swartkrans Australopithecines. S. Afr. Archaeol. Bull. 24, 170–171 (1969). 80. Yravedra, J. & Domínguez-Rodrigo, M. The shaft-based methodological approach to the quantification of long limb bones and its relevance to understanding hominin subsistence in the Pleistocene: Application to four Paleolithic sites. J. Quat. Sci. 24, 85–96 (2009). 81. Villa, P. & Mahieu, E. Breakage patterns of human long bones. J. Hum. Evol. 21, 27–48 (1991). 82. Thompson, C. E. L., Ball, S., Thompson, T. J. U. & Gowland, R. The abrasion of modern and archaeological bones by mobile sediments: The importance of transport modes. J. Archaeol. Sci. 38, 784–793 (2011). 17 Vol.:(0123456789) Scientific Reports | (2021) 11:13977 | https://doi.org/10.1038/s41598-021-93261-1 www.nature.com/scientificreports/ 83. Binford, L. R. Bones: Ancient Men and Modern Myths. (New York Academic Press, 1981). 84. Blumenschine, R. J. Percussion marks, tooth marks, and experimental determination of the timing of hominid and carnivore access to long bones at FLK Zinjanthropus, Olduvai Gorge, Tanzania. J. Hum. Evol. 29, 21–51 (1995). 85. Martín-Perea, D. M., Courtenay, L. A., Soledad-Domingo, M. & Morales, J. Application of artificially intelligent systems for the identification of discrete fossiliferous levels. PeerJ 8, e8767. https:// doi. org/ 10. 7717/ peerj. 8767 (2020). 86. Baddeley, A., Rubak, E. & Turner, R. Spatial Point Patterns: Methodology and Applications with R. (Taylor and Francis, 2016). 87. Ripley, B. D. Modelling spatial patterns (with discussion). J. R. Stat. Soc. B 39, 172–212 (1977). 88. Besag, J. Contributing to the discussion of the paper by Ripley (1977). J. R. Stat. Soc. B 39, 193–195 (1977). 89. Hopkins, B. A new method of determining the type of distribution of plant individuals. Ann. Bot. 18, 213–227 (1954). Acknowledgements This research has been possible thanks to the support and approval of the Consejería de Cultura de la Junta de Andalucía through the General Research Project “Primeras ocupaciones humanas y contexto paleoecológico a partir de los depósitos pliopleistocenos de la cuenca Guadix-Baza. Zona arqueológica de la cuenca de Orce (Granada, España)” (BC.03.032/17). We are grateful to the “Archaeometrical Studies. Inside the Artefacts & Ecofacts” Unit of Excellence (University of Granada) for its support. We also thank the TIDOP Group from the Department of Cartographic and Land Engineering of the Higher Polytechnics School of Avila, University of Salamanca. We would also like to recognize the technical support provided by C.A.I. Arqueometry and Archaeological Analysis from the Complutense University which has been very useful in carrying out the present paper. We would like to thank to the whole Orce Research Project teams including the volunteers who year after year join the field work.Last but not least, we would like to thank the reviewers of this paper, in particular Paul Mazza, for his fruitful comments that have greatly improved the first version of this paper, and the editor of the collection Quaternary Taphonomy, Ruth Blasco. Author contributions C.L.: Data curation, investigation in Taphonomy, methodology, writing, editing. L.A.C.: Data curation, investigation, methodology, software, validation, visualization, writing—original draft, review and editing. J.Y.: Conceptualization, investigation, methodology, supervision and writing—original draft, review and editing. J.S.: Investigation and coordination in Macrovertebrates Taxonomy. H.-A.B.: Investigation in Micromammals Palaeontology. D.M.: Investigation in Macrovertebrates Taxonomy. S.V.: Investigation in Macrovertebrates Taxonomy. B.A.: Investigation in Macrovertebrates Taxonomy. J.J.R.A.: Investigation in Taphonomy. D.H.R.: Investigation in Taphonomy. A.S.R.: Investigation. J.A.S.: Investigation in Taphonomy. O.O.: Investigation and Supervision in Stratigraphy. J.A.: Investigation and Supervision in Micromammals Palaeontology. M.F.: Investigation and supervision in Macrovertebrates Taxonomy. J.M.J.A.: Formal analysis, project administration, resources, supervision and writing—original draft, review and editing. Funding C.L. and A.S.-R. are funded by the Regional Government of Andalusia, with both contracts associated to General Research Project BC.03.032/17. C.L. was also funded by the Spanish Public Employment Service (Spanish Ministry of Labour and Social Economy) with an unemployment benefit. L.A.C. is funded by the Spanish Ministry of Science, Innovation and Universities, with an FPI Predoctoral Grant (Reg. PRE2019-089411) associated to project RTI2018-099850-B-IOO and the University of Salamanca. The Institut Català de Paleoecologia Humana i Evolució Social (IPHES-CERCA) has received financial support from the Spanish Ministry of Science and Innovation through the “María de Maeztu” program for Units of Excellence (CEX2019-000945-M). This paper is also part of projects 2017SGR-859 (Gov. of Catalonia, AGAUR), and CGL2016-80000-P (Spanish Ministry of Economy and Competitiveness). J.M.J.-A. takes part of the Research Group HUM-607 (Junta de Andalucía). Competing interests The authors declare no competing interests. Additional information Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1038/ s41598- 021- 93261-1. Correspondence and requests for materials should be addressed to J.Y.orJ.M.J.-A. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. © The Author(s) 2021