Solving a ‘puzzle’. The global 4.2 ka Bond Event at El Mirador cave (Sierra de Atapuerca,Burgos,Spain) and the importance of small mammal taphonomy to the interpretation of past environments and their climatic controls

Abstract

The regional climatic context in which Chalcolithic (MIR5) and Bronze Age (MIR4) levels from El Mirador cave (Sierra de Atapuerca, Burgos, Spain) are framed is affected by the 4.2 ka cal. BP event, a global event defined as a cooling and aridification phase. Previous works based on palaeoenvironmental inferences indicate conflicting results regarding the possible impact of the event on vegetation and small mammals from MIR5. Pollen record illustrates a possible aridification episode that could match with the 4.2 ka cal. BP event, while the signal of this event is not clearly recorded in the small mammal assemblage, which indicates more humid environmental conditions than pollen record. Taphonomic analyses confirmed that the small mammal assemblages from MIR4 and MIR5 are the result of predation, supporting the involvement of European eagle owls (Bubo bubo) in its formation. This avian raptor shows a marked preference for hunting animals living in the more open and wetter parts of their hunting range. Likewise, spontaneous specialisation on abundant prey species could also be observed under certain environmental conditions. This characteristic behaviour of eagle owls may have provided the contradictory results observed between the small mammal assemblage and palynological evidence. Nonetheless, taphonomic analyses also provided information about climatic conditions and fluctuations along time. The low incidence of manganese coatings and carbonate crusts deposits in small mammal bone remains from MIR5 support the presence of arid conditions during the formation of this level, which agreed with the aridification phase probably related to the 4.2 ka Bond Event inferred by palynological data from MIR5. These results provided a more robust conclusion about the paleoenvironmental contexts during the formation of Chalcolithic and Bronze Age levels at El Mirador cave.

Keywords

anthropic influence 4.2 Global Bond Event Holocene palaeoclimate palaeoecology palaeoenvironment

Introduction

Small vertebrates are considered one of the principal sources of palaeoecological information in archaeological and palaeontological contexts (e.g. Andrews and O’Brien, 2000, 2010; Blain et al., 2021; Cuenca-Bescós et al., 2009, 2010; Fernández-García et al., 2016; Sánchez-Bandera et al., 2020). Nonetheless, the main factor accumulating microvertebrate assemblages in both, modern and fossil sites is predation (e.g. Andrews, 1990; Bisbal-Chinesta et al., 2020; Fernández and Pardiñas, 2018; García-Morato et al., 2019a; Lloveras et al., 2014; Marin-Monfort et al., 2019, 2021). Different factors such as the size ratio of the predator/prey relationship, diurnal or nocturnal hunting behaviour, hunting strategy and territory size may vary from one predator to another, and this will be reflected in the species represented in the small vertebrate assemblages (Andrews, 1990). Predation together with the presence of other biotic and abiotic factors involved during the accumulation and preservation processes may compromise the palaeoecological value of the microfauna species recorded in a fossil site. The combination of all these factors may result in altered species representation, which could be reflected in the reduction or increasing of the species richness/abundance when certain groups are selected against others or when species from more than one habitat are mixed together (Andrews, 2006; Fernández-Jalvo et al., 2011). Therefore, to use microvertebrates for biochronologic, biostratigraphic, palaeoenvironmental, palaeoclimatic and palaeoecological studies, taphonomic interpretations need to be preferentially addressed before any of these analyses are undertaken (Fernández-Jalvo et al., 2011).

Taphonomic studies may provide valuable information to recognise the origin of the small vertebrate assemblage but also about the climatic conditions and fluctuations along time (Behrensmeyer et al., 2000). A long-term monitoring experiment of 40 years in southern Kenya demonstrated the high fidelity of the ecological information preserved in bone assemblages (Western and Behrensmeyer, 2009). Thus, post-depositional processes such as weathering, plant activity, soil corrosion or mineral deposits could be a reliable source for interpreting climatic conditions during the formation of the fossil site (Fernández-Jalvo and Andrews, 2016) improving environmental interpretations.

The application of taphonomic analyses to the small mammal assemblages from Chalcolithic (MIR5) and Bronze Age (MIR4) levels from El Mirador cave (Sierra de Atapuerca, Burgos, Spain) could help to further consolidate palaeoecological inferences of this upper part of the sequence. The regional climatic context in which both levels are framed is affected by the 4.2 ka cal. BP event, a global event defined as a cooling and aridification phase which led to the simultaneous collapse of several complex civilisations (e.g. Bini et al., 2019; Bond et al., 1997, 2001; DeMenocal, 2001; Liu and Feng, 2012; Magny, 2004; Magny et al., 2009, 2013; Roland et al., 2014; Staubwasser and Weiss, 2006; Weiss et al., 1993). Although its climatic impact remains poorly understood and very variable at local and regional scale, Walker et al. (2012) identified the 4.2 ka cal. BP event as a suitable marker to subdivide the Holocene period into ‘middle’ and ‘late’ chronozones. Conflicting results regarding the possible impact of the event on vegetation and small mammals have emerged in palaeoenvironmental evidence from MIR5. The palynological data illustrates a possible aridification episode that could match with the 4.2 ka cal. BP event. Nonetheless, the signal of this event is not clearly recorded in the small mammal assemblage, which indicates more humid environmental conditions than pollen records (Bañuls-Cardona et al., 2017a). The information provided by taphonomic analyses of the small mammal assemblages would allow us to interpret palaeoecological inferences at a high level of resolution. At this point, recognition of the site formation processes will lead to a better understanding of the relationship between living and fossil environments, and a more precise palaeoenvironmental interpretation of this Global Climate Change. The main objective of this work is to provide new evidence about past environmental changes in the Chalcolithic and Bronze Age levels from El Mirador cave through the taphonomic study of the small mammal assemblages. For this purpose, taphonomic analyses were applied in order to: (1) dilucidate the origin of the small mammal assemblages; (2) discard potentially distortive processes (i.e. predator preferences, resedimentation and reworking processes) in the palaeoecological inferences and (3) assess the environmental information provided by the small mammal remains through post-depositional surface modifications.

El Mirador cave

El Mirador cave (42° 20′ 58″ N and 03° 30′ 33″ W) is located in the southern slope of the Sierra de Atapuerca, in the municipality of Ibeas de Juarros (Burgos, Spain) at 1033 m.a.s.l (Figure 1). Presently, the site has a large entrance that provides a shelter-like form, enlarged by a collapse of part of the roof which occurred during the late-Pleistocene. The mouth of this cavity is approximately 23 m wide and 4 m high, penetrating some 15 m inwards (Vergès et al., 2016). The cave has been systematically excavated since 1999. During the first 10 years, the archaeological work in the cave was focused on the excavation of 6 m² area test pit in the central area of the western half of the cave to determine the archaeological potential of the site. This test pit constitutes a 20 m deep profile with 14 m of Pleistocene deposits and 6 m of Holocene sediment (Vergès et al., 2002, 2016). In 2009, fieldwork began in two new sectors named 100 and 200 at the NW and NE ends of the cave, respectively. Both sectors are in contact with the current wall and are being dug in steps, following the line of the cave roof with the purpose of documenting the inward retreat of the cave and the stratigraphic variations between the different areas (Vergès et al., 2016).

Figure 1.

Location of El Mirador Cave and main associated river basins (upper panel) together with the stratigraphic profile from the Late Neolithic-Bronze Age south section of the test pit (Modified from Vergès et al., 2002) (lower panel). The dotted line indicates the levels analysed in this work. 1: limestone; 2: burrows; 3: ashes; 4: ashes and organic matter; 5: thermoaltered sediments; 6: charcoal accumulations; 7: lithic remains; 8: bone remains; 9: preserved pellets.

The Chalcolithic (MIR5 – 4530–4417 cal. yr BP) and Bronze Age (MIR4 – 3730–3070 cal. yr BP) levels analysed in this work belong to the 6 m² test pit located in the central area. The large Holocene sequence which contains these levels bears 24 differentiated archaeological layers displaying high lateral and vertical variability due to the sedimentary characteristics of the cave. A total of 18 radiocarbon dates were obtained from the sequence, revealing that the cave was occupied between the Early Neolithic and the Bronze Age. Holocene levels are mainly the result of the use of the cave as a livestock pen (Vergès et al., 2016). The activities related to animal husbandry left sedimentary layers, dung, which was piled together and burned at regular intervals in order to reduce the volume and to eliminate parasites (Angelucci et al., 2009). The alternation of burned and unburned layers of manure and nodules of ash from burned dung results in the formation of ‘fumiers’, an artefact record related to domestic occupations. Therefore, the labelling and excavation of the site was decided to differentiate contexts, distinguishing subsequent facies which are characteristic of this kind of anthropized units (Vergès et al., 2002).

The two levels (MIR5 and MIR4) here analysed reflect two different moments of the use of the cave and both yielded human remains. MIR5 is a thin level with scarce anthropogenic contributions, rich in small vertebrate remains (Vergès et al., 2002). This level is considered contemporaneous with a Chalcolithic collective burial in which at least 23 human individuals have been identified. The burial was discovered in the test pit 200 (MIR203) at the NE sector of the cave, indicating that the site was used with burial purposes at the time of the deposition of MIR5. Radiocarbon dates obtained from two of the individuals recovered yield dates between 4550 and 4390 cal. yr BP and 4880 and 4480 cal. yr BP (Ceperuelo et al., 2014, 2015; Lozano et al., 2017; Vergès et al., 2016). A new dating obtained on rodent bones from a pellet sample recovered from MIR5 level in the central test pit area indicates dates of 4530–4417 cal. yr BP (Bisbal-Chinesta, 2020). The sedimentation in MIR5 is attributed to natural processes exclusively and the low sedimentation rates observed in this level have been associated with the interruption of anthropic contributions during MIR5 formation (Vergès et al., 2002). This fact is supported by the preservation of pellets in this level, that would have been rapidly disintegrated by the human activities whilst occupying the cave, and by the evidence of plants and non-Pollen Palynomorphs registered, which indicates an opposite scenario to those associated with anthropic contexts. A possible aridification phase that match with the 4.2 ka cal. BP event has been inferred during this period of low anthropic activity in El Mirador cave (Expósito et al., 2017).

MIR4 level yields dates which correspond to the Bronze Age based on charcoal remains (3730–3070 cal. yr BP). Archaeological record from MIR4 suggests an intensification of human activity with respect to MIR5 (Vergès et al., 2016). This is supported by the presence of breeding of mixed goat (Capra hircus) and sheep flocks (Ovis aries) (Martín et al., 2016). Cattle breeding also increased in parallel with an increase in the representation of wild consumed animals (Cervus elaphus, Capreolus capreolus, Meles meles, Felis silvestris, Vulpes vulpes, lagomorphs and birds). This fact has been linked to the diversification of faunal source exploitation and to the intensification of human activity in the site (Vergès et al., 2016).

The level is composed by archaeological sediment in situ and the presence of animal husbandry activities resulted in the formation of the burnt manure levels or ‘fumiers’. Pollen record indicates that the base of MIR4 is also characterised by arid conditions and by the lack of evidence of anthropogenic impact on vegetation (Expósito et al., 2017). This palynological evidence is contemporaneous to the presence of cannibalised human bones of six individuals from the Early Bronze Age which provided ages from 4430 to 3880 cal. yr BP (Cáceres et al., 2007). Nonetheless, an increment of human activities after this episode is evidenced by the presence of possible cultivation areas in the vicinity of the cave (Expósito et al., 2017; Rodríguez et al., 2016).

Palaeoenvironments and the 4.2 ka Bond event in El Mirador cave

Palynological analyses from El Mirador cave showed the existence of an open landscape formed by mixed forest of pines (Pinus sp.) and oaks (Quercus sp.) from the early Neolithic to the Bronze Age. These forest formations alternated with open lands, crop fields (probably located close to the cave) and pasturelands. (Expósito et al., 2017; Rodríguez et al., 2016). This landscape has suffered several transformations along the Holocene succession, mainly characterised by a decreasing trend in tree cover (Bañuls-Cardona et al., 2017a; Expósito et al., 2017; Rodríguez et al., 2016). The transformation of the landscape documented along the sequence is related to the intensification of human activities, although climatic variations related to arid increase episodes are also present (Expósito et al., 2017; Rodríguez et al., 2016).

Amongst these arid episodes, the 4.2 ka cal. BP Bond event (Magny, 2004; Magny et al., 2009) is evidenced at MIR5 and at the lowermost part of the base of MIR4 level through palynological analyses (Expósito et al., 2017). This event is considered of global extent but variable in its climatic expression at the local and regional level (Bini et al., 2019; Magny et al., 2009). In the Mediterranean Basin, between 4.3 and 3.8 ka, the so-called ‘4.2 ka event’ has been observed in several pollen records (Di Rita and Magri, 2009, 2012, 2019). Although it is considered as an interval of cooling and drying (e.g. Cullen et al., 2000; Dixit et al., 2014; Drysdale et al., 2006), some researchers have suggested that this event is best described as a complex succession of dry/wet events, rather than a single long, dry event (Magny et al., 2009; Railsback et al., 2018). Nevertheless, the Mediterranean region shows some of the most consistent evidence of the 4.2 ka BP event (Bini et al., 2019). Likewise, in the Iberian Peninsula the expression of this event is variable. The northwest region is characterised by a relatively moderate signature of the event, while in the northern and southern meseta its impact is very apparent. In the southwest, aridification dynamics during the 4.2 ka BP event were also observed (Blanco-González et al., 2018).

Material and methods

Study samples

The small mammal (orders Rodentia and Eulipotyphla) material analysed in this study come from levels MIR4 (Bronze Age) and MIR5 (Chalcolithic) from El Mirador cave. Most of the small mammals analysed come from unburned facies, as scarce materials were recovered from burned ones (Bañuls-Cardona et al., 2017a; SGM under study). Small-vertebrate remains were collected through a system of water screening with sieves of decreasing mesh size (1, 0.5, and 0.05 cm). Microfossils were separated from the dry sediment distinguishing between cranial and post-cranial elements. Small mammal cranial remains and some post-cranial diagnostic elements were selected by Bañuls-Cardona et al. (2017a, 2017b) with the main purpose of obtaining taxonomic and palaeoecological results. Taxonomic analysis from MIR4 and MIR5 revealed a total of seven different small mammal species (Table 1): Crocidura russula, Sorex gr. coronatus-araneus, Microtus arvalis, Microtus agrestis, Microtus (Terricola) duodecimcostatus, Apodemus sylvaticus and Eliomys quercinus, which is only present in MIR4. Both small mammal assemblages indicate that woodlands are the best represented environment in both levels, followed by open humid habitats in MIR5 and open dry habitats in MIR4. Moreover, taxonomic composition is indicating that synanthropic species (Crocidura russula, Microtus arvalis, Microtus (Terricola) duodecimcostatus and Eliomys quercinus) are more frequent in MIR4 than in MIR5 (Table 1).

Table 1.

Small mammal species identified (MNI and %) in the levels analysed in this work from El Mirador cave. Synanthropic species are indicated in bold. Data from Bañuls-Cardona et al. (2017a, 2017b).

TAXA	MIR4 (Bronze Age)	%	MIR5 (Chalcolithic)	%
Crocidura suaveolens	19	19.79	6	6.74
Sorex gr. araneus-coronatus	6	6.25	4	4.49
Microtus (Terricola) duodecimcostatus	17	17.71	14	15.73
Microtus agrestis	14	14.58	37	41.57
Microtus arvalis	21	21.88	20	22.47
Apodemus gr. sylvaticus-flavicollis	18	18.75	8	8.99
Eliomys quercinus	1	1.04	0	0.00
Total	96		89

Taphonomic analyses in this work have been carried out using the samples previously sorted by Bañuls-Cardona et al. (2017a, 2017b) considering both, cranial and post-cranial small mammal elements. All the fossil materials were studied at the Micropalaeontology laboratory of the Catalan Institute of Human Palaeoecology and Social Evolution (IPHES), where they are housed. Samples were analysed under a light microscope (Leica MZ75 10×–80×) and photographs of some specimens were taken under a binocular light microscope motorised in Z (Leica M205A).

Taphonomic analysis

Taphonomic study on small mammal fauna fossil specimens is based on the methodology proposed by Andrews (1990), Fernandez-Jalvo and Andrews (1992), Fernández-Jalvo et al. (2016) and Marin-Monfort et al. (2019). Three main taphonomic variables were analysed: anatomical representation, breakage, and digestion. Amongst these three variables, digestion is the most direct evidence of predation and the most useful to identify the predator. Anatomical representation and breakage are variables that are susceptible of being affected by post-depositional processes (e.g. transport and/or trampling) and recovery procedures. Thus, the original values formerly produced by the predator involved in the accumulation of the assemblage may be distorted (Andrews, 1990). The study of this variables allows us to evaluate how intensively post-depositional taphonomic processes and recovery procedures affected the small mammal assemblages analysed.

Anatomical representation to evaluate the loss of skeletal elements was calculated using the formula proposed by Dodson and Wexlar (1979):

R i % = [M N E i / (M N I * E i)] * 100

Where Ri is the relative abundance of the element i; MNE, is the minimum number of element i observed in the sample, MNI is the minimum number of individuals and Ei is the number of the element i in a complete skeleton.

Skeletal element proportions of post-cranial in relation to cranial remains are based on the following indices proposed by Andrews (1990):

a) Pc/C compares five post-cranial elements (Pc) (femora, humeri, radii, ulnae and tibiae) with mandibles, maxillae and isolated molars (C). A correction factor (×8/×5) was applied.

b) F + H/Mx + Md: the MNE of femora and humeri (F + H) is compared with the MNE of mandibles and maxillae (Mx + Md).

c) T + R/F + H: this index represents the loss of distal elements comparing the sum of tibiae and radii (T + R) with femora and humeri (F + H).

Proportion of isolated teeth and tooth loss were also calculated. Isolated teeth index compares the number of molars and incisors isolated in the sample with the number of empty alveoli. Tooth loss compares the number of empty alveoli with the number of total empty alveoli calculated for the sample analysed.

Breakage pattern was identified for cranial and post-cranial remains following the scheme proposed by Andrews (1990):

a) Maxillae: complete, preserving the zygomatic arch, without the zygomatic arch and hemipalates.

b) Mandibles: complete, with ascending ramus broken and without ascending ramus plus inferior border broken.

c) Post-cranial elements: complete, proximal fragments, distal fragments and shafts.

Damage caused by digestion process is analysed in both, isolated and in situ molars and incisors. The digestion pattern is identified following the descriptions by Andrews (1990), Fernandez-Jalvo and Andrews (1992) and Fernández-Jalvo et al. (2016) for arvicolids, murids and soricids. Glirids were also identified following the classification scheme proposed in Marin-Monfort et al. (2019).

Apart from modifications produced by predatory activity, other taphonomic traits were analysed, such as: the presence of burnt bones (from burning dung), root-marking, manganese staining, humidity cracking or rounding. Burnt bones were classified in five different stages following descriptions provided by Cáceres (2002). Stage 1 is recognised by the presence of brown colour in limited areas of the bone; bones which show a brownish colour and are not cracked are classified in stage 2. Stage 3 burning shows a dull black colour; stage 4, the bone is dark grey with extensive cracking of the surface. Finally, stage 5 burning is characterised by a bright white colour of the bone surface.

Root-marking on the bone surface has been classified according to the percentage of coverage following the classification proposed by Martínez et al. (2019). A total of five categories are distinguished: 0 (absent), 1 (25%), 2 (50%), 3 (75%) and 4 (100%). Same classification was applied for manganese staining and carbonate crusts.

Transport and dispersion by water streams is evaluated by considering the abundance of those skeletal elements with better hydrodynamic shape (following the classification proposed by Korth, 1979) together with the presence of rounding, polishing and surface abrasion in teeth and post-cranial elements (Fernández-Jalvo et al., 2014; García-Morato et al., 2019b).

Results

Skeletal element representation

A total of 1703 small mammal minimum number of elements (MNE) are present in the sample analysed considering both, MIR4 and MIR5 levels. Amongst them, 996 correspond to cranial and 707 to post-cranial skeleton. All the skeletal elements are represented in MIR4 while in MIR5 astragali are absent (Table 2). A visual inspection of the skeletal element frequency (Ri) (Table 2 and Figure 2) reveals some differences between the levels analysed. Relative abundance of isolated molars is higher in MIR5 and indices of skeletal proportions support this observation indicating that cranial elements are also especially abundant in this level (Table 3). The relationship between distal and proximal elements shows a better representation of distal elements in MIR4 while it is the proximal elements in MIR5 (Table 3).

Table 2.

Skeletal element frequencies calculated for MIR4 and MIR5 levels from El Mirador cave.

		MIR4		MIR5
Skeletal elements	Expected	MNE	Ri%	MNE	Ri%
Hemimaxillae	2	48	53.93	29	47.03
Hemimandibles	2	89	100.00	55	89.19
Incisors	4	87	48.88	88	71.35
Molars	12	230	43.07	370	100.00
Vertebrae	30	106	7.94	124	13.41
Ribs	24	9	0.84	1	0.14
Scapulae	2	15	16.85	17	27.57
Humeri	2	49	55.06	37	60.00
Radii	2	5	5.62	10	16.22
Ulnae	2	35	39.33	34	55.14
Pelves	2	51	57.30	24	38.92
Femora	2	46	51.69	30	48.65
Tibiae	2	31	34.83	25	40.54
Metapodials	20	23	2.58	8	1.30
Calcanei	2	12	13.48	2	3.24
Astragali	2	3	3.37	0	0.00
Phalanges	56	7	0.28	3	0.17
Total	846			857
Average Ri	31.47%			36.05%
MNI	45			31

Table 3.

Indices of skeletal element proportions for the levels analysed.

	MIR4 (%)	MIR5 (%)
Pc/C	72	48
F + H/Mx + Md	69	80
T + R/F + H	38	52

Figure 2.

Representation of the skeletal element frequencies for MIR4 and MIR5 levels.

Breakage

Cranial and post-cranial breakage percentages are summarised in Figure 3. Results for cranial remains are classified distinguishing between arvicolids, murids and soricids. Overall, assemblages from MIR4 and MIR5 show a high degree of breakage in cranial and post-cranial elements. However, breakage amongst the different taxa considered show differences, especially in soricids, where most of the mandibles are complete in MIR4 and all of them are complete in MIR5 (Figure 3). The higher completeness observed in soricids mandibles is probably the result of the robustness of these mandibles in comparison with the ones from arvicolids and murids. Molar and incisor loss indices are relatively low for soricids, in contrast to murids and arvicolids, which show the highest breakage percentages and tooth loss (Figure 3). In fact, percentages for isolated teeth are especially high for arvicolids, mainly due to the unrooted teeth that characterises this group. Total percentages for both levels indicate that the detachment of molars and incisors from jaws is higher in MIR5. Percentages of isolated teeth also indicate that the destruction of jaws is higher in MIR5 while in MIR4 some of the molars and incisors are lost from the sample probably during its recovering. The difference observed between both levels in loss and isolated teeth indices is due to the fact that arvicolid remains were more frequently recovered in MIR5 than in MIR4.

Figure 3.

Breakage percentages of the different portions of small mammal cranial and long bones. For cranial remains a distinction between arvicolids, murids and soricids has been done due to differences in robustness and teeth detachment from skulls and mandibles. Total values are referred to the sum of arvicolids, murids and soricids. Note that grey colour is referred to MIR4 and black colour to MIR5; number of identified fossils (NISP) are displayed in brackets. *A: Arvicolids; M: Murids; S: Soricids.

A high degree of breakage is also observed in long bones for both levels. Amongst the fragmented skeletal limbs, overall, the best-preserved portions in MIR4 are the epiphyses of radii, ulnae and tibiae. For humeri and femora, values from proximal and distal parts are similar. In MIR 5, distal portions of humeri are the most abundant while for femora and tibiae are the proximal ones. In the case of ulnae, most of them are complete. Shafts are generally scarce in both levels, and they are only found in MIR4 for femora and tibiae showing percentages below 6% (Figure 3). Total breakage of long bones was relatively higher in MIR4 (68.51%) than in MIR5 (56.86%).

Digestion

Corrosive effects of digestion on bones and teeth have been observed in both levels analysed (Figure 4a–c). Table 4 lists the number and proportion of molars, incisors and postcranial elements which exhibit digestive etching. Teeth showing no signs of digestion traits on the surface are the most abundant, although some of them exhibit signs of digestive corrosion in both levels, especially concentrated in light and moderate grades. Regarding molars, the 17.94% are digested in MIR4 while in MIR5 this percentage reaches the 20.05%. As for incisors, the 41.32% and the 40.59% of them are digested in MIR4 and MIR5, respectively. Although light degree of digestion is the most abundant followed by the moderate one, heavy and extreme degrees are also observed in both levels but at low percentages (Table 4).

Table 4.

Digestive corrosion percentages obtained for cranial and postcranial elements analysed in MIR4 and MIR5.

	Cranial digestion	Non-digested		Light		Moderate		Heavy		Extreme		TOTAL DIGESTED (%)
		NISP	%	NISP	%	NISP	%	NISP	%	NISP	%
MIR4	In situ molars	123	87.86	8	5.71	8	5.71	1	0.71	0	0.00	12.13
	Isolated molars	179	78.51	42	18.42	4	1.75	3	1.32	0	0.00	21.49
	TOTAL	302	82.06	50	13.59	12	3.26	4	1.09	0	0.00	17.94
	In situ incisors	28	75.86	3	8.11	6	16.22	0	0.00	0	0.00	24.33
	Isolated incisors	43	51.19	31	36.90	7	8.33	2	2.38	1	1.19	48.8
	TOTAL	71	58.68	34	28.10	13	10.74	2	1.65	1	0.83	41.32
	Post-cranial digestion	NISP	%	NISP	%	NISP	%	NISP	%	NISP	%
	Femora	20	54.05	11	29.73	3	8.10	2	5.41	1	2.70	45.94
	Humeri	31	79.49	7	17.95	1	2.56	0	0.00	0	0.00	20.51
	TOTAL	51	67.11	18	23.68	4	5.26	2	2.63	1	1.32	32.89
	Cranial digestion	Non-digested		Light		Moderate		Heavy		Extreme		TOTAL DIGESTED (%)
		NISP	%	NISP	%	NISP	%	NISP	%	NISP	%
MIR5	In situ molars	35	79.55	4	9.09	4	9.09	1	2.27	0	0.00	20.45
	Isolated molars	296	80.00	58	15.68	10	2.70	5	1.35	1	0.27	20.00
	TOTAL	331	79.95	62	14.98	14	3.38	6	1.45	1	0.24	20.05
	In situ incisors	9	69.23	1	7.69	2	15.38	1	7.69	0	0.00	30.76
	Isolated incisors	51	57.95	30	34.09	5	5.68	2	2.27	0	0.00	42.24
	TOTAL	60	59.41	31	30.69	7	6.93	3	2.97	0	0.00	40.59
	Post-cranial digestion	NISP	%	NISP	%	NISP	%	NISP	%	NISP	%
	Femora	24	82.76	4	13.79	1	3.45	0	0.00	0	0.00	17.24
	Humeri	22	64.71	10	29.41	2	5.88	0	0.00	0	0.00	35.29
	TOTAL	46	73.02	14	22.22	3	4.76	0	0.00	0	0.00	26.98

Values referred total digestion percentages are in bold.

Figure 4.

(a) Arvicoline molars showing light digestion, (b) incisors showing light digestion mainly concentrated on the tip, (c) femur with light digestive corrosion concentrated on the proximal end, (d) examples of concretions on a distal and proximal end of a femur in MIR4, (e) humerus with manganese stains from MIR4, (f) burnt humerus at grade two-thirds according to Cáceres (2002) from MIR4.

Long bones are also affected by digestive corrosion (Figure 4c). Approximately 30% of the proximal ends of the femora and the distal portions of the humeri are modified by digestive corrosion in the two levels analysed at El Mirador cave. These elements show digestion traces mainly classified in light and moderate degrees, which only affect the epiphyses without penetrating to the diaphysis of the bone. Only in MIR4 heavy and extreme degrees of corrosion are found for femora but at low percentages as it is observed for cranial elements (Table 4).

Post-depositional modifications

Both levels are characterised by a low incidence of abiotic post-depositional modifications. Alterations such as weathering, cracking produced by changes in humidity or abrasion caused by transport processes are not found. Additionally, biotic alterations such as root marks caused by plant coverage are also absent in the small mammal assemblages from the levels analysed. Both levels show a good skeletal representation without any specific abundance of any of the Korth groups (Figure 5) indicating lack of hydrodynamic selection. Only a certain deficit of ribs was observed, although this element in small mammal fossil assemblages is rarely recovered. MIR4 also indicates an absence of isolated molars in Korth Group II/III. This group was calculated using only those molars that exceed the empty alveoli, but in the case of MIR4 isolated molars were in deficit, which indicates that some of them have been detached from the jaws. No evidence of polishing and rounding linked to abrasion by wind or water transport processes has been observed in the fossil assemblages here analysed. Consequently, the absence of weathered and/or abraded bones, together with the simultaneous presence of all Korth groups (equivalent to the Voorhies, 1969 classification for large mammals, see Figure 6) suggests a natural accumulation performed by predators slightly modified by post-depositional events.

Figure 5.

Korth groups calculated for each of the levels analysed. MNE: Minimum Number of Elements; Group I: ribs; Group I/II: vertebrae, radius, ulna, pelvis; Group II: scapula, maxilla, calcaneum, astragalus, humerus, femur; Group II/III: molars; Group III: mandible, tibia.

Figure 6.

Summary of the most frequent alterations found in the levels analysed from El Mirador cave.

The most frequent post-depositional evidence found on the fossils are the precipitation of manganese and the presence of carbonate crusts mainly on long bones (femur, tibia, humerus, ulna and radius), and more abundant in MIR4. In this level, from a total NISP of 235 long bones the 22.97% of them show certain level of carbonate crusts (Figure 4d) mainly covering the 25% of the surface (Figure 6). Likewise, the 16.17% of the long bones show manganese mineralisation (Figure 4e) covering the 25% of the bone surface or less than the 25% of the bone surface (grade 1) (Figure 6). In MIR5, from a total NISP of 153, only the 7.19% presents carbonate crusts covering less than the 25% of the surface and only one element (0.01%) shows manganese covering less than the 50% (grade 2) of the bone surface (Figure 6).

Burnt bones are also found, some of them reaching a stage 3 of cremation (black) (Figure 4f), although most of the bones are classified as a stage 2 (brown). In MIR4, the 2.19% of the long bones show this alteration while in MIR5 the 0.01% is affected by fire (Figure 6).

Predator classification

Previous results described suggest a natural predator accumulation of the small mammal assemblages from Chalcolithic and Bronze Age levels at El Mirador cave. Although anatomical representation and breakage patterns originally produced by the predator may have been slightly modified by post-depositional processes, digestive corrosion on teeth and post-cranial elements are clearly recognised. A summary of the results obtained compared with the predator classification done by Andrews (1990) is shown in Table 5. Predator categories obtained are relatively low for most of the taphonomic variables assessed. Nonetheless, breakage in cranial elements, especially in mandibles, is probably influenced by the high completeness found in soricid mandibles.

Table 5.

Summary of the predator category classification done by Andrews (1990) compared with the data in the present study (modified from Marin-Monfort et al., 2022).

Taphonomic variables	MIR4	Category	MIR5	Category	Category 1	Category 2	Category 3	Category 4	Category 5
Average of Ri	31.47%	3−4	36.05%	3−4	80%–70%	70%–60%	60%–30%	40%–20%	<20%
					Barn owl, long eared owl, short eared owl, Verreaux owl and great grey owl	Spotted eagle owl and European eagle owl	Tawny owl and little owl	Kestrel and hen harrier	Mammalian carnivores
PC/C	72%	2	48%	2	>200%	100%–150%	50%–70%	<30%	-
F + H/Mx + Md	69%	2	80%	2	100%–150%	90%–70%	50%–30%	<60%	-
					Barn owl, long eared owl, short eared owl, European eagle owl, great grey owl, bat-eared fox and pine marten	Tawny owl, Verreaux owl, spotted eagle owl, kestrel and genet	Hen harrier and artic fox	Snowy owl, little owl, mongoose, coyote and red fox	-
T + R/F+H	38%	5	52%	4	≈100%	85%–70%	70%–60%	60%–40%	40%–20%
					Barn owl, long eared owl, Verreaux owl and tawny owl	Short eared owl, European eagle owl, great grey owl, artic fox and coyote	Little owl, kestrel	Spotted eagle owl, hen harrier and red fox	Pine marten, mongoose, genet, bat-eared fox
Complete long bones	31.49%	4−5	40.49%	4	90%–100%	90%–80%	70%–40%	40%	<30%
					Barn owl, long eared owl, great grey owl, Verreaux owl	Snowy owl and European eagle owl	Tawny owl and spotted eagle owl	Little owl, kestrel, hen harrier, mongoose, genet, bat-eared fox	Coyote, red fox, artic fox and pine marten
Complete skulls	4.17%	5	0.00%	5	70%–100%	50%–70%	30%–50%	-	0
					Barn owl, long eared owl, great grey owl	Short eared owl, spotted eagle owl, tawny owl and European eagle owl	Little owl, kestrel and hen harrier	-	Mammalian carnivores
Complete mandibles	28.09%	2	25.45%	2	70%–90%	15%–40%	2%–10%	0%	-
					Barn owl, long eared owl, Verreaux owl and great grey owl	Snowy owl, short eared owl, European eagle owl and tawny owl	Spotted eagle owl, kestrel and hen harrier	Little owl, mammalian carnivores	-
Incisor loss	72.26%	2−3	84.52%	3	20%–40%	50%–60%	80%–90%	90%–100%	100%
					Barn owl, snowy owl, long eared owl, short eared owl, Verreaux owl, spotted eagle owl and European eagle owl	Great grey owl, mongoose and coyote	Tawny owl, little owl and bat-eared fox	Kestrel, genet, red fox and pine marten	Hen harrier and artic fox
Molars loss	65.94%	2−3	82.54%	3	20%–40%	50%–60%	80%–90%	90%–100%	100%
					Barn owl, snowy owl, long eared owl, short eared owl, Verreaux owl, spotted eagle owl and European eagle owl	Great grey owl, mongoose and coyote	Tawny owl, little owl and bat-eared fox	Kestrel, genet, red fox and pine marten	Hen harrier and artic fox
Molars digested	17.93%	3	20.05%	3	<3%	4%–6%	11%–22%	50%–70%	50%–100%
					Barn owl, long eared owl, short eared owl and Verreaux owl	Snowy owl, spotted eagle owl, great grey owl	European eagle owl, tawny owl, bat-eared fox, mongoose and genet	Little owl, kestrel, pine marten	Hen harrier, coyote, red fox and artic fox
Incisors digested	43.65%	2−3	41.75%	2−3	5%–13%	10%–30%	50%–70%	60%–80%	100%
					Barn owl, short eared owl and snowy owl	Long eared owl, Verreaux owl, great grey owl and bat-eared fox	European eagle owl, spotted eagle owl, tawny owl, little owl, pine marten and mongoose	Kestrel	Hen harrier, coyote, red fox and artic fox
Postcranial digested	32.89%	2	26.98%	2	6%–20%	25%–50%	60%–100%	-	100%
					Barn owl, short eared owl, long eared owl, Verreaux owl and great grey owl	European eagle owl, spotted eagle owl and tawny owl	Little owl, kestrel and hen harrier	-	Mammalian carnivores

Values in bold are referred to the small mammal assemblages analysed in this work.

Percentages and grades of digestion obtained from the analysis of these fossil associations agree with Category 3 predators (11%–22% for molars, 50%–70% for incisors and 25%–50% for post-cranial elements, see Table 5) according to Andrews (1990) classification. This category includes the European eagle owl (Bubo bubo), the spotted eagle owl (Bubo africanus) and the tawny owl (Strix aluco). Amongst them, both the European eagle owl and the tawny owl inhabits the Iberian Peninsula. The main difference between these predators is the more abundant and intense digestion produced by tawny owl on skeletal elements in comparison with the European eagle owl (sensu Andrews, 1990). Results obtained in this study fit better with the involvement of Bubo bubo. The presence of fossil remains linked to this avian raptor are not known yet for the two levels analysed in this work.

Discussion

Palaeoenvironmental analyses from El Mirador cave have been carried out through a multiproxy approach (zooarchaeological, microvertebrate and pollen studies), providing a comprehensive view of the evolution of the landscapes surrounding the cave (e.g. Bisbal-Chinesta, 2020; Bisbal-Chinesta et al., 2020; Expósito et al., 2017; Martín et al., 2016; Vergès et al., 2016). Both, small mammals and palynological data have provided evidence of the intensification of anthropization in the area. However, a phase of increased aridity in MIR5 and the lowermost part at the base of MIR4 in contact to MIR5 (4530–3730 cal. yr BP) is observed through pollen analyses (Expósito et al., 2017) in contrast to more humid conditions indicated by taxonomic identifications of the small mammal assemblages (Bañuls-Cardona et al., 2017a). Small mammal sampling does not record the lowermost part of MIR4 as pollen studies do, so the study is only referred to MIR5 and MIR4 and the contrast between these two entire levels.

Palynological studies of MIR5 and the lowermost part at the base of MIR4 show aridity indicated by a decrease in riparian taxa, mesophilous plants and deciduous Quercus species. The taxonomic diversity of trees and shrubs decreases, including those taxa linked to anthropization (Poaceae and Cerealia) and livestock pressure (nitrophilous vegetation) (Expósito et al., 2017). This arid phase shown by palynology is supported by taphonomic post-depositional traits related to the low presence of wet or damp conditions such as manganese coatings and carbonate crusts (Fernández-Jalvo and Andrews, 2016; Marin-Monfort et al., 2021). These taphonomic modifications show relative higher percentages in MIR4 than in MIR5, indicating slightly more humid conditions in the former than in the latter level. The palaeoenvironmental evidence provided by the small mammal taxonomy from MIR5 indicates humid conditions in this level, supported by an increase in open dry landscapes together with a decreased in precipitation values from MIR5 to MIR4 (Bañuls-Cardona et al., 2017a, 2017b).

The participation of eagle owls in the site as the main agent responsible for the small mammal production have several implications in palaeoecological results that need to be carefully considered. The European Eagle owl (Bubo bubo) is a large nocturnal strigiform, whose hunting habits are influenced by the abundance or accessibility of mammal prey. In the Iberian Peninsula, the Eagle owl is considered a rabbit (Oryctolagus cuniculus) specialist predator in central and southern areas due to its relative high abundance and profitability (Fernandez-de-Simon et al., 2014; Penteriani et al., 2008; Tobajas et al., 2016). Nevertheless, when lagomorphs are at low densities, the eagle owl is considered a complete opportunist predator due to its varied diet (Andrews, 1990; Mikkola, 1983). According to these authors, in the absence of more profitable prey items, the diet of the eagle owls is mainly composed by microtines, which usually make up 30%–80% of prey numbers (Mikkola, 1983). A detailed inspection of the small mammal species recorded at MIR4 and MIR5 indicates that microtines represent the 54.17% and 79.78% of the prey items, respectively (Bañuls-Cardona et al., 2017a, 2017b; Table 1). Although it is claimed that eagle owls show a generalist feeding behaviour, it is necessary to carefully considered two main biases in its prey assemblages: nocturnal animals are taken more commonly due to its time of hunting activity and animals living in the more open and wetter parts of the owl’s hunting range may be over-represented (Andrews, 1990; Mikkola, 1983; Olsson, 1979). Likewise, spontaneous specialisation on abundant prey species probably favours successful hunting of this predator under certain conditions (Mikkola, 1983). This characteristic behaviour of the eagle owls could explain the apparent contradictory palaeoenvironmental signal observed in MIR5 between small mammals, palynological evidence and post-depositional taphonomic traits.

Eagle owls still being present in MIR4 where palaeoenvironmental analyses indicate an increase in open dry landscapes to the detriment of open humid ones (Bañuls-Cardona et al., 2017a). This environmental change is also linked to an increase in synanthropic taxa (Crocidura russula, Microtus arvalis, Microtus (Terricola) duodecimcostatus and Eliomys quercinus) (Table 1) in MIR4 with respect to MIR5 (Bañuls-Cardona et al., 2017b) also supported by pollen remains, which indicate a significant increase in crops around the cave in MIR4 level while in MIR5 the anthropogenic impact on vegetation is scarce (Expósito et al., 2017). The intensification of human activities during the Bronze Age period (MIR4) may have caused a homogenisation of the landscape favouring the dominance of the more generalist and synanthropic species, which became the more accessible prey items surrounding the cave. Considering that eagle owls can easily specialised on abundant prey species, a change in its diet composition as it has been observed in other studies could be expected (Obuch and Bangjord, 2016).

In turn, it cannot be discarded that the population dynamics of the small mammals living in the surrounding environment of the cave may have responded to the presence of human activity in the area. The more humid conditions inferred in MIR5 through small mammals are mainly the consequence of the higher representation percentages of Microtus agrestis (41.57%) with respect to MIR4 (14.58%) (Bañuls-Cardona et al., 2017a, 2017b; Table 1). This species inhabits wet areas in meadows, riverside habitats, or forests with dense herbaceous cover and with low or null presence of livestock (Corbet and Southern, 1977; Gosalbez and Luque-Larena, 2002; Hansson, 1977; Wheeler, 2005). Although humans may still be in the area, the higher abundance of M. agrestis in MIR5 may have been favoured by the low livestock pressure which characterises this Chalcolithic level when the site was used as a burial cave (Vergès et al., 2016). This period is also characterised by a low presence of synanthropic small mammal taxa (Bañuls-Cardona et al., 2017b). A similar situation in which the percentage of small mammal synanthropic taxa decreases during a burial phase is observed in the Early Chalcolithic level (5290–4850 cal. yr BP) from El Portalón cave (Sierra de Atapuerca, Burgos, less than a kilometre from El Mirador) according to Rofes et al. (2021). Despite a possible delay in the response of the microfauna in MIR4 to the arid phase of MIR5 could be considered, the intensification of the human impact in the surroundings of El Mirador cave during the Bronze Age prevents from supporting this assumption (Bañuls-Cardona et al., 2017a, 2017b; Cabanes et al., 2009; Expósito et al., 2017; Rodríguez et al., 2016).

In summary, the incidence of more humid conditions indicated in MIR5 by small mammal taxonomy contrast with taphonomic post-depositional processes and palynological evidence (Expósito et al., 2017). Although pollen deposits may respond to changes at macro/regional level, the low incidence of manganese coatings and carbonate crusts deposits in bone remains from MIR5 agreed with the general scenario inferred by palynological data which indicates the presence of an aridification phase in MIR5 probably related to the 4.2 ka Bond Event.

Conclusions

This work presents the first taphonomic analysis realised for the small mammal fossil assemblage of levels MIR4 and MIR5 of El Mirador cave. These levels are of special interest as the regional climatic context in which they are framed is affected by the 4.2 ka cal. BP Bond event. The taphonomic study provided evidences to dilucidate the origin of the small mammal assemblages of MIR4 and MIR5 levels and also further environmental information through the identification of post-depositional surface modifications.

Results obtained from the taphonomic analyses pointed out to a European eagle owl (Bubo bubo) as the main accumulation agent of the small mammal fossil assemblage in both levels. Eagle owls are considered generalist predators but with a marked preference for hunting animals living in the more open and wetter parts of their hunting range. The predator hunting preferences may have provided the contradictory results observed between the small mammal assemblage and palynological evidence. Therefore, the impact of the 4.2 ka cal. BP Bond event in the small mammal species recovered in the site is very likely affected by the taphonomic source of the fossil assemblage (i.e. predation).

The absence of abrasion or weathered small mammal fossils together with the simultaneous presence of all Korth groups suggest that the remains were not transported from close proximity to the cave. Root-marks were also scarce in both level indicating that the fossil assemblage has been located far from the cave entrance. Post-depositional taphonomic modifications, such as manganese coatings and carbonate crusts are more abundant in level MIR4 than in MIR5. The low incidence of these post-depositional process in MIR5 is in accordance with the arid phase indicated by palynological results. Post-depositional taphonomic surface modifications reinforced the hypothesis of a more arid environment during the formation of MIR5 level and the possibility of an altered species representation induced by predator hunting preferences.

Footnotes

Acknowledgements

The authors wish to express their gratitude to Fernando J. Fernández, Hugues-Alexandre Blain and an anonymous reviewer whose comments and suggestions greatly improved this paper.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work has been supported by the Ministerio de Ciencia, Innovación y Universidades (MICINN-FEDER) [Project PID2021-122355NB-C32], the Ministerio de Ciencia, Innovación y Universidades through the María de Maeztu excellence accreditation to the IPHES [project CEX2019-000945-M], the Agència de Gestió d’Ajutas Universitaris i de Recerca de la Generalitat de Catalunya [project 2017-SGR1040], Aragosaurus (Gobierno de Aragón): Recursos Geológicos y Paleoambientes, the Ministerio de Ciencia, Innovación y Universidades AEI cofinanced by the Fondo Europeo de Desarrollo Regional (MICINN/AEI/FEDER, UE) [Project PGC2018-093925-B-C33] and Escuela de Doctorado de la Universidad Complutense de Madrid through a funded research stay at IPHES (Tarragona, Spain) gave to SGM [grant number EDUCM2021]. SGM has a predoctoral grant funded by the Universidad Complutense de Madrid and Banco Santander [grant number CT42/18-CT43/18]. DMM has a CONICET postdoctoral grant [RESOL-2020-134-APN-DIR#CONICET].

ORCID iDs

Sara García-Morato

Gloria Cuenca-Bescós

References

Andrews

(1990) Owls, Caves and Fossils. London: University of Chicago Press.

Andrews

(2006) Taphonomic effects of faunal impoverishment and faunal mixing. Palaeogeography Palaeoclimatology Palaeoecology 241(3–4): 572–589.

Andrews

O’Brien

(2000) Climate, vegetation, and predictable gradients in mammal species richness in southern Africa. Journal of Zoology 251: 205–231.

Andrews

O’Brien

(2010) Mammal species richness in Africa. In: Werdelin

Sanders

(eds) Cenozoic Mammals of Africa. New York, NY: Columbia University Press, pp.929–947.

Angelucci

Boschian

Fontanals

, et al. (2009) Shepherds and karst: The use of caves and rock-shelters in the Mediterranean region during the Neolithic. World Archaeology 41(2): 191–214.

Bañuls-Cardona

López-García

Morales Hidalgo

, et al. (2017a) Lateglacial to Late Holocene palaeoclimatic and palaeoenvironmental reconstruction of El Mirador cave (Sierra de Atapuerca, Burgos, Spain) using the small-mammal assemblages. Palaeogeography Palaeoclimatology Palaeoecology 471: 71–81.

Bañuls-Cardona

Martín Rodríguez

López-García

, et al. (2017b) Human impact on small-mammal diversity during the middle- to late-Holocene in Iberia: The case of El Mirador cave (Sierra de Atapuerca, Burgos, Spain). The Holocene 27(8): 1067–1077.

Behrensmeyer

Kidwell

Gastaldo

(2000) Taphonomy and paleobiology. Paleobiology 26(S4): 103–147.

Bini

Zanchetta

Perşoiu

, et al. (2019) The 4.2 ka BP Event in the Mediterranean region: an overview. Climate of the Past 15(2): 555–577.

10.

Bisbal-Chinesta

(2020) Biogeografía e impacto humano en las comunidades de anfibios y reptiles del Cuaternario final en la Península Ibérica: la Cueva del Mirador (Atapuerca, Burgos) y la población de Chalcides ocellatus (Scincidae) de la Serra del Molar (Alicante). Doctoral dissertation, Universitat Rovira i Virgili, Tarragona, Spain.

11.

Bisbal-Chinesta

Bañuls-Cardona

Fernández-García

, et al. (2020) Elucidating anuran accumulations: massive taphocenosis of tree frog Hyla from the Chalcolithic of El Mirador cave (Sierra de Atapuerca, Spain). Journal of Archaeological Science Reports 30: 102277.

12.

Blain

Fagoaga

Ruiz-Sánchez

, et al. (2021) Coping with arid environments: A critical threshold for human expansion in Europe at the Marine Isotope Stage 12/11 transition? The case of the Iberian Peninsula. Journal of Human Evolution 153: 102950.

13.

Blanco-González

Lillios

López-Sáez

, et al. (2018) Cultural, demographic and environmental dynamics of the copper and Early Bronze Age in Iberia (3300–1500 BC): Towards an interregional multiproxy comparison at the time of the 4.2 ky BP Event. Journal of World Prehistory 31(1): 1–79.

14.

Bond

Kromer

Beer

, et al. (2001) Persistent solar influence on North Atlantic climate during the holocene. Science 294: 2130–2136.

15.

Bond

Showers

Cheseby

, et al. (1997) A pervasive millennial-scale cycle in North Atlantic holocene and glacial climates. Science 278: 1257–1266.

16.

Cabanes

Burjachs

Expósito

, et al. (2009) Formation processes through archaeobotanical remains: The case of the Bronze Age levels in El Mirador cave, Sierra de Atapuerca, Spain. Quaternary International 193(1–2): 160–173.

17.

Cáceres

(2002) Tafonomía de yacimientos antrópicos en karst. Complejo Galería (Sierra de Atapuerca, Burgos), Vanguard Cave (Gibraltar) y Abric Romaní (Capellades, Barcelona). Doctoral dissertation, Universitat Rovira i Virgili, Tarragona, Spain.

18.

Cáceres

Lozano

Saladié

(2007) Evidence for bronze age cannibalism in El Mirador cave (Sierra de Atapuerca, Burgos, Spain). American Journal of Physical Anthropology 133(3): 899–917.

19.

Ceperuelo

Lozano

Duran-Sindreu

, et al. (2014) Root canal morphology of Chalcolithic and Early Bronze Age human populations of El Mirador Cave (Sierra de Atapuerca, Spain). Anatomical Record 297: 2342–2348.

20.

Ceperuelo

Lozano

Duran-Sindreu

, et al. (2015) Supernumerary fourth molar and dental pathologies in a Chalcolithic individual from the El Mirador Cave site (Sierra de Atapuerca, Burgos, Spain). Homo Journal of Comparative Human Biology 66: 15–26.

21.

Corbet

Southern

(1977) The Handbook of British Mammals. Oxford: Blackwell.

22.

Cuenca-Bescós

Rofes

López-García

, et al. (2010) Biochronology of Spanish quaternary small vertebrate faunas. Quaternary International 212(2): 109–119.

23.

Cuenca-Bescós

Straus

González Morales

, et al. (2009) The reconstruction of past environments through small mammals: From the Mousterian to the Bronze Age in El Mirón Cave (Cantabria, Spain). Journal of Archaeological Science 36(4): 947–955.

24.

Cullen

DeMenocal

Hemming

, et al. (2000) Climate change and the collapse of the Akkadian empire: Evidence from the deep sea. Geology 28(4): 379–382.

25.

DeMenocal

(2001) Cultural responses to climate change during the late holocene. Science 292(5517): 667–673.

26.

Di Rita

Magri

(2009) Holocene drought, deforestation and evergreen vegetation development in the central Mediterranean: A 5500 year record from Lago Alimini Piccolo, Apulia, southeast Italy. The Holocene 19(2): 295–306.

27.

Di Rita

Magri

(2012) An overview of the holocene vegetation history from the central Mediterranean coasts. Journal of Mediterranean Earth Sciences 4: 35–52.

28.

Di Rita

Magri

(2019) The 4.2 ka event in the vegetation record of the central Mediterranean. Climate of the Past 15(1): 237–251.

29.

Dixit

Hodell

Petrie

(2014) Abrupt weakening of the summer monsoon in northwest India 4100 yr ago. Geology 42: 339–342.

30.

Dodson

Wexlar

(1979) Taphonomic investigations of owl pellets. Paleobiology 5: 275–284.

31.

Drysdale

Zanchetta

Hellstrom

, et al. (2006) Late Holocene drought responsible for the collapse of Old World civilizations is recorded in an Italian cave flowstone. Geology 34(2): 101–104.

32.

Expósito

Burjachs

Vergès

(2017) Human trace on the landscape during the holocene at El Mirador Cave (Sierra de Atapuerca, Spain): The palynological evidence. The Holocene 27(8): 1201–1213.

33.

Fernandez-de-Simon

Díaz-Ruiz

Cirilli

, et al. (2014) Role of prey and intraspecific density dependence on the population growth of an avian top predator. Acta Oecologica 60: 1–6.

34.

Fernández

Pardiñas

UFJ

(2018) Small mammals taphonomy and environmental evolution during late pleistocene-Holocene in Monte Desert: The evidence of Gruta del Indio (central west Argentina). Journal of South American Earth Sciences 84: 266–275.

35.

Fernández-García

López-García

Lorenzo

(2016) Palaeoecological implications of rodents as proxies for the late Pleistocene–Holocene environmental and climatic changes in northeastern Iberia. Comptes Rendus Palevol 15(6): 707–719.

36.

Fernandez-Jalvo

Andrews

(1992) Small mammal taphonomy of Gran Dolina, Atapuerca (Burgos), Spain. Journal of Archaeological Science 19(4): 407–428.

37.

Fernández-Jalvo

Andrews

(2016) Atlas of Taphonomic Identifications. 1001+ Images of Fossil and Recent Mammal Bone Modification. Dordrecht: Springer.

38.

Fernández-Jalvo

Andrews

Denys

, et al. (2016) Taphonomy for taxonomists: Implications of predation in small mammal studies. Quaternary Science Reviews 139: 138–157.

39.

Fernández-Jalvo

Andrews

Sevilla

, et al. (2014) Digestion versus abrasion feature in rodent bones. Lethaia 47: 323–336.

40.

Fernández-Jalvo

Scott

Andrews

(2011) Taphonomy in palaeoecological interpretations. Quaternary Science Reviews 30(11–12): 1296–1302.

41.

García-Morato

Sevilla

Panera

, et al. (2019a) Rodents, rabbits and pellets in a fluvial terrace (PRERESA site, Madrid, Spain). Quaternary International 520: 84–98.

42.

García-Morato

Marin-Monfort

Fernández-Jalvo

(2019b) Rolling bones: A preliminary study of micromammal abrasion on different initial taphonomic stages. Palaeontologia Electronica 22: 1–16.

43.

Gosalbez

Luque-Larena

(2002) Microtus agrestis (Linnaeus,1761). Topillo agreste. In: Palomo

Gisbert

(eds) Atlas de Los Mmamíferos Terrestres de España. Madrid: Dirección General de Conservación de la Naturaleza, pp.390–393.

44.

Hansson

(1977) Spatial dynamics of field voles Microtus agrestis in heterogeneous landscapes. Oikos 29: 539–544.

45.

Korth

(1979) Taphonomy of microvertebrate fossil assemblages. Annals of Carnegie Museum of Natural History 48: 235–285.

46.

Liu

Feng

(2012) A dramatic climatic transition at ~4000 cal. Yr BP and its cultural responses in Chinese cultural domains. The Holocene 22(10): 1181–1197.

47.

Lloveras

Moreno-García

Nadal

, et al. (2014) Blind test evaluation of accuracy in the identification and quantification of digestion corrosion damage on leporid bones. Quaternary International 330: 150–155.

48.

Lozano

Bermúdez de Castro

Arsuaga

, et al. (2017) Diachronic analysis of cultural dental wear at the Atapuerca sites (Spain). Quaternary International 433: 243–250.

49.

Magny

(2004) Holocene climate variability as reflected by mid-European lake-level fluctuations and its probable impact on prehistoric human settlements. Quaternary International 113(1): 65–79.

50.

Magny

Combourieu-Nebout

de Beaulieu

, et al. (2013) North-south palaeohydrological contrasts in the central Mediterranean during the Holocene: Tentative synthesis and working hypotheses. Climate of the Past 9: 2043–2071.

51.

Magny

Vannière

Zanchetta

, et al. (2009) Possible complexity of the climatic event around 4300—3800 cal. BP in the central and western Mediterranean. The Holocene 19(6): 823–833.

52.

Marin-Monfort

Fagoaga

García-Morato

, et al. (2021) Contribution of small mammal taphonomy to the last Neanderthal occupations at the El Salt site (Alcoi, southeastern Spain). Quaternary Research 103: 208–224.

53.

Marin-Monfort

García-Morato

Andrews

, et al. (2022) The owl that never left! Taphonomy of earlier Stone Age small mammal assemblages from Wonderwerk Cave (South Africa). Quaternary International 614: 111–125.

54.

Marin-Monfort

García-Morato

Olucha

, et al. (2019) Wildcat scats: Taphonomy of the predator and its micromamal prey. Quaternary Science Reviews 225: 106024.

55.

Martínez

Alcaráz

, et al. (2019) Geoarchaeology and taphonomy: Deciphering site formation processes for late holocene archaeological settings in the eastern Pampa-Patagonian transition, Argentina. Quaternary International 511: 94–106.

56.

Martín

García-González

Nadal

, et al. (2016) Perinatal ovicaprine remains and evidence of shepherding activities in early Holocene enclosure caves: El Mirador (Sierra De Atapuerca, Spain). Quaternary International 414: 316–329.

57.

Mikkola

(1983) The Owls of Europe. Calton: T. and A.D. Poyser Ltd.

58.

Obuch

Bangjord

(2016) The Eurasian eagle-owl (Bubo bubo) diet in the Trøndelag region (Central Norway). Slovak Raptor Journal 10(1): 51–64.

59.

Olsson

(1979) Studies on a population of Eagle Owls, Bubo bubo (L.), in southeast Sweden. Swedish Wildlife Research 11(1): 1–99.

60.

Penteriani

Del Mar Delgado

Bartolommei

, et al. (2008) Owls and rabbits: Predation against substandard individuals of an easy prey. Journal of Avian Biology 39(2): 215–221.

61.

Railsback

Liang

Brook

, et al. (2018) The timing, two-pulsed nature, and variable climatic expression of the 4.2 ka event: A review and new high-resolution stalagmite data from Namibia. Quaternary Science Reviews 186: 78–90.

62.

Rodríguez

Allué

Buxó

(2016) Agriculture and livestock economy among prehistoric herders based on plant macro-remains from El Mirador (Atapuerca, Burgos). Quaternary International 414: 272–284.

63.

Rofes

Ordiales

Iriarte

, et al. (2021) Human activities, biostratigraphy and past environment revealed by small-mammal associations at the Chalcolithic levels of El Portalón de Cueva mayor (Atapuerca, Spain). Quaternary 4(2): 16.

64.

Roland

Caseldine

Charman

, et al. (2014) Was there a ‘4.2 ka event’ in Great Britain and Ireland? Evidence from the peatland record. Quaternary Science Reviews 83: 11–27.

65.

Sánchez-Bandera

Oms

Blain

, et al. (2020) New stratigraphically constrained palaeoenvironmental reconstructions for the first human settlement in Western Europe: The Early pleistocene herpetofaunal assemblages from Barranco León and Fuente Nueva 3 (Granada, SE Spain). Quaternary Science Reviews 243: 106466.

66.

Staubwasser

Weiss

(2006) Holocene climate and cultural evolution in late prehistoric–early historic West Asia. Quaternary Research 66(3): 372–387.

67.

Tobajas

Fernandez-de-Simon

Díaz-Ruiz

, et al. (2016) Functional responses to changes in rabbit abundance: Is the eagle owl a generalist or a specialist predator? European Journal of Wildlife Research 62(1): 85–92.

68.

Vergès

Allué

Angelucci

, et al. (2002) La Sierra de Atapuerca durante el Holoceno: Datos preliminares sobre las ocupaciones de la Edad del bronce en la Cueva de el Mirador (Ibeas de Juarros, Burgos). Trabajos de Prehistoria 59(1): 107–126.

69.

Vergès

Allué

Fontanals

, et al. (2016) El Mirador cave (Sierra de Atapuerca, Burgos, Spain): A whole perspective. Quaternary International 414: 236–243.

70.

Voorhies

(1969) Taphonomy and Population Dynamics of an Early Pliocene Vertebrate Fauna. Laramie, WY: University of Wyoming.

71.

Walker

MJC

Berkelhammer

Björck

, et al. (2012) Formal subdivision of the holocene Series/Epoch: A discussion paper by a Working Group of INTIMATE (Integration of ice-core, marine and terrestrial records) and the Subcommission on Quaternary Stratigraphy (International Commission on Stratigraphy). Journal of Quaternary Science 27(7): 649–659.

72.

Weiss

Courty

Wetterstrom

, et al. (1993) The genesis and collapse of third millennium north mesopotamian civilization. Science 261: 995–1004.

73.

Western

Behrensmeyer

(2009) Bones track community structure over four decades of ecological change. Science 324: 1061–1064.

74.

Wheeler

(2005) The diet of field voles Microtus agrestis at low population density in upland britain. Acta Theriologica 50: 483–492.