lable at ScienceDirect

Quaternary Science Reviews 121 (2015) 75e88

Contents lists avai

Quaternary Science Reviews

journal homepage: www.elsevier .com/locate/quascirev

Dancing to the rhythms of the Pleistocene? Early Middle Paleolithicpopulation dynamics in NW Iberia (Duero Basin and CantabrianRegion)

Policarpo S�anchez Yustos*, Fernando Diez MartínUniversity of Valladolid, Pza. del Campus s/n, 47011 Valladolid, Spain

a r t i c l e i n f o

Article history:Received 27 January 2015Received in revised form5 May 2015Accepted 7 May 2015Available online

Keywords:Middle PaleolithicMousterianIberiaDuero BasinCantabrian RegionPopulation dynamics

Abbreviations: EMP, Early Middle Paleolithic; SSMAdaptive Cycle Mode; CM, Cascade Mode; DB, Duero BEM, Early Mousterian; LA, Later Acheulean.* Corresponding author. Tel.: þ34 670301253.

E-mail address: policarpo.sy@gmail.com (P. S�anch

http://dx.doi.org/10.1016/j.quascirev.2015.05.0050277-3791/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

The Northwest of Iberia has yielded one of the most complete European Middle Paleolithic records.Despite this wealth of information, very little is known about population dynamics during this period.For that reason, the main concern of this paper is to provide socio-environmental models that may helpexplain Early Middle Paleolithic (EMP) population dynamics in NW Iberia, assessing to what extent theywere shaped by climate forces. The archaeological record is analyzed on the basis of the heuristics ofecological models, already employed in the European Pleistocene record but never at a regional scale, inorder to detect long-term changes in the composition of EMP populations, and the environmental,biological and sociocultural process influencing those changes. According to the models proposed, wehave detected a long-term population dynamic between MIS 11 and MIS 6, characterized by low envi-ronmental stress, high biological productivity, interaction among populations and socioculturalcomplexity. Eventually, this population dynamic was broken due to an extreme climate phase in late MIS6 that had a profound impact on populations and sociocultural structures. As a result, the Upper Pleis-tocene population of NW Iberia was concentrated in the Cantabrian region. This area became an isolatedNeanderthal glacial refugium that hosted a population with different origins and fragile long-term de-mographic stability.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The impact of climate and environmental changes on humandemographic, migrational and cultural patterns during the Pleis-tocene is a topic of great current interest in Quaternary studies.Although there are interesting proposals that place less emphasison the climate background to demographic and cultural change inthe Paleolithic (e.g. Tzedakis et al., 2007; Banks et al., 2008;Roebroeks, 2008; Rodríguez et al., 2011; Moncel, 2012), the main-stream view among researchers is that climate had a tremendousimpact both on population and cultural dynamics (e.g. Gamble,1993; Housley et al., 1997; Stringer et al., 2003; Finlayson andCarri�on, 2007; Shea, 2009; Hublin and Roebroeks, 2009; Dennell

, Source-Sink Model; ACM,asin; CR, Cantabrian Region;

ez Yustos).

et al., 2011). In fact, the increasing number of high-resolutionclimate records for Europe during the Last Glacial cycle (i.e.Ganopolski and Rahmstorf, 2001; d’Errico and S�anchez-Go~ni, 2003)are currently fueling this approach (i.e. Banks et al., 2008; Mülleret al., 2012; Schmidt et al., 2012).

As has been amply demonstrated over the last decades, theMiddle Pleistocene climatic oscillations continually moved both thehuman and geographic frontiers in Europe and, consequently, thepattern of hominin occupation responded to repeated expansionsand contractions: populations expanded northwards in favorablecircumstances, and then retreated southwards into refugia whenconditions deteriorated (i.e. Gamble, 1993; Housley et al., 1997;Bocquet-Appel and Demars, 2000; Gamble et al., 2004; Bocquet-Appel et al., 2005; Green et al., 2010). The current emphasis thatsome researchers have put on regional extinction is seen in sce-narios of population fragmentation, recombination, extinction andexpansion, and thus Pleistocene glacial refugia have been renamedas bottlenecks (i.e Hewitt, 1999; J€oris et al., 2003; Hublin andRoebroeks, 2009; Verpoorte, 2009; Dennell et al., 2011;Bradtm€oller et al., 2012; Widlok et al., 2012). According to this,

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8876

Southern Europe would have worked as a glacial source for popu-lating the northerly areas when climate conditions allowed, but incertain periods it may have also become depopulated, andrecolonized from a source outside Europe (Dennell et al., 2011;MacDonald et al., 2012; Bermúdez and Martin�on, 2013). Regardlessof whether successful colonizing movements or population col-lapses are emphasized, the Pleistocene settlement of Europe wascharacterized by frequent spatial and geographic discontinuities(Bermúdez et al., 2013).

In these fluctuating conditions in Europe, the Lower to MiddlePaleolithic transition occurs during the second part of the MiddlePleistocene (Chazan, 2009; Villa, 2009; Moncel et al., 2012). Thisphenomenon runs parallel to the biological process of speciationfromHomo heidelbergensis toHomo neanderthalensis (Hublin, 2009;Arsuaga et al., 2014). Since the beginning of the twentieth century ithas been thought that the beginning of the European MiddlePaleolithic is marked by the appearance of Levallois technology(S�anchez Yustos, 2012). Nowadays, empirical evidence suggeststhat the Levallois method is part of a complex technological reor-ganization process that also implies a progressive abandonment oflarge-sized tools, increasing variability and formal standardizationof both production and retouching of small flakes, and intensefragmentation of lithic reduction and the consequent importance ofthe traveling component (i.e. Geneste, 1989; Chase, 1990; Wyneret al., 1993; Gamble and Roebroeks, 1999; White and Ashton,2003; Bourguignon et al., 2004; Barsky et al., 2005; Santonja andP�erez, 2006; Fern�andez Peris et al., 2008; White et al., 2011;Moncel et al., 2012; Oll�e et al., 2013; Picin et al., 2013; Turq et al.,2013; Santonja et al., 2014). Furthermore, this technological reor-ganization process is part of profound behavioral changes thatincluded the habitual use of fire (Roebroeks and Villa, 2011), newmobility patterns (Chazan, 2009), elaborate hunting strategies(Blasco et al., 2010), the mastery of hafting (Rots, 2013) or themanipulation of pigments (Roebroeks, 2012).

Early Middle Paleolithic (EMP) technology is traditionallydistinguished fromMousterian technologymainly on chronologicalgrounds (EMP S MIS 5) and the generalization of the above-mentioned techno-economical changes, particularly Levalloistechnology. The appearance of such changes during the EMP isgeographically and temporally discontinuous and coexisted for along time with the Acheulean technocomplex. This fact has fueledthe still open debate about the relationships between both tech-nocomplexes in the second part of theMiddle Pleistocene in Europe(i.e. Foley and Lahr, 1997; Tuffreau et al., 1997; Roebroeks andTuffreau, 1999; Wynn and Coolidge, 2004; Bourguignon et al.,2004, 2008; Moncel, 2006; Monnier, 2006; Peris, 2007; Brenetet al., 2008; Picin et al., 2013; Adler et al., 2014; Santonja et al.,2014). Despite these disagreements, many archaeologists agreethat Levallois technology resulted from the gradual synthesis of theshaping method characteristic of bifaces, and Acheulian bifacialtechnology and Levallois technology reflect an ancestor-descendant relationship (i.e. Bordes, 1971; Pigeot, 1991; Rolland,1995; Tuffreau, 1995; De Bono and Goren-Inbar, 2001; White andAshton, 2003; Bar-Yosef and Dibble, 2005; White et al., 2011; Adleret al., 2014).

The NW of Iberia, particularly the Cantabrian Region (CR) andthe nearby Duero Basin (DB), has provided one of the most com-plete European Middle Paleolithic records, with numerousarchaeostratigraphic sequences well contextualized geochrono-logically and paleo-ecologically. The CR is one of the areas ofgreatest interest for the study of the final Middle Paleolithic atEuropean scale (i.e. Maíllo-Fern�andez et al., 2004; Bernardo deQuir�os et al., 2008; Baena et al., 2012). Its different ecologicalunits (coast, inland valleys and mountains) are especially attractivefor analyzing this period. The Middle Paleolithic settlement in this

region is quite recent compared to the nearby DB, the largestCenozoic basin in Iberia, where a long-term EMP settlement hasbeen registered. The abundance of the archaeological record in allthe ecological units of this basin (plains, plateaus and mountainousborders) has provided one of the most complete pictures of hom-inin settlement and techno-economic behavior during the secondpart of the European Middle Pleistocene. Notwithstanding thiswealth of information, very little is known about the kind of rela-tionship that existed between the populations in these neighboringregions. For this reason, the interest of this paper is to providemodels that may help explain the Pleistocene population scenarioin NW Iberia. Recalling Gamble's words, part of in the interest ofthis paper is also to indicate to what extent these populations were“dancing to the rhythms of the Pleistocene” (Gamble, 1999: 125); orin other words, how far population dynamics were shaped byclimate forces.

2. Materials and methods

The chronological, economic, technological, paleo-ecologicaland paleo-anthropological data presented here derive from theEMP sites (SMIS 5) in the DB and CR. An overview of the lateAcheulean (LA) open-air sites in both regions is also presented dueto the chronological and techno-typological parallels that keepwith the early Middle Paleolithic sites. The level of resolution of thedata is the distribution of archaeological sites within the landscapeplotted against chronology. The data are analyzed on the basis onthe heuristics of the following ecological models, already employedin Pleistocene Europe but never at regional scale, in order to detectlong-term changes in the composition of EMP populations, and theenvironmental, biological and sociological process influencingthose changes. Population is used here as a synonym of “deme”,defined as ‘‘the aggregate of local populations of a species inhab-iting a geographic subdivision of the range of the species’’ (Howell,1999: 8e9).

2.1. Source-Sink Model

Pleistocene human population dynamics have been modeled as“sources and sinks” (Eller et al., 2004; Hawks, 2009; Dennell et al.,2011; Bermúdez and Martin�on, 2013). “Source-sink dynamics'’ is atheoretical ecological model, originally developed by Pulliam(1988), to analyze the impact of habitat-specific demographicrates on population growth and regulation. According to the“Source-Sink Model” (SSM), which describes how variation inhabitat quality may affect the growth or decline of a population, theparticular species assemblage occupying any region may consist ofa mixture of source and sink populations and may be as much ormore influenced by type or proximity of another habitat as by theresources and other conditions in the region.

A sink habitat is a regionwhere the average rate of reproductionis below replacement levels. Despite this, it may support largepopulations that would eventually disappear without continuedimmigration from an adjacent, more reproductive area that isnamed a source habitat. Therefore, the population regulation be-tween source and sink depends on active dispersal from sourcehabitats. Individuals choose to leave the source whenever theirexpected reproductive success is higher in the sink. There is noreason why the source-sink dynamic needs to be constant.

2.2. Adaptive Cycle Model

The “Adaptive Cycle Model” (ACM) was generated from ob-servations in ecosystems (Holling, 2001; Gunderson and Holling,2002) and successfully adapted to the analysis of current socio-

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e88 77

environmental systems (Abel et al., 2006; Folke, 2006) andPleistocene systems (Bradtm€oller et al., 2012; Widlok et al., 2012).Over time, the structures and functions of systems change as aresult of internal dynamics and external influences, resulting infour characteristic phases described by Holling (1986, 2001)(Fig. 1):

- Growth phase (r): is a moment of growth with high availabilityof resources, the establishment of internal structures and con-nections, and of high resilience. As structure and connectionsamong system components increase, more resources and en-ergy are required to maintain them.

- Conservation phase (k): in this phase the system growth slowsdown, internal structures are determined, and becomeincreasingly interconnected, less flexible, and more vulnerableto external disturbances. Growth and conservation phasescorrespond to ecological succession in ecosystems and consti-tute a development mode in organizations and societies.

- Release phase (U): disturbances lead to the next phase, thecrises, inwhich the accumulated structure collapses. The systemis most vulnerable to change

- Reorganization phase (a): the system is reorganized, new di-versity is established, innovative strategies of resource exploi-tation are developed, technology and tool production areadapted, and networks are reorganized. All of which createsnew structures and new contents for exchange between sub-systems. Multiple modes of reorganization are possible, and anynewmodes will be most apparent during the phases that followthe crises. The new r phase may be very similar to the previous rphase, or it may be quite different.

The ACM represents an ambitious attempt to construct aframework aimed at describing and comparing the internal dy-namics of social-ecological systems, because it not only sheds lighton situations of crises but also on the subsequent phases of reor-ganization. The ACM is one of the five heuristics used to understandsocio-ecological system behaviour, the other four heuristics are:resilience, panarchy, transformability, and adaptability (Walker,2006). Indeed, the concept of adaptive cycle is related to the pan-archy (hierarchical) structure of the components within the adap-tive cycle as well as between adaptive cycles on different scales.Gunderson and Holling use the term of panarchy to “capture theadaptive and evolutionary nature of adaptive cycles that are nestedone within the other across space and time scales” (Gunderson and

Fig. 1. Adaptive Cycle Model (modified from Holling, 2001).

Holling, 2002: 73). The four phases flow on different time scales:whereas Phases r and k may last for long time periods, crises andreorganization represent extremely rapid processes (Redman andKinzig, 2003). Depending on the particular configuration of thesystem, it can then begin a new adaptive cycle or alternatively itmay transform into a new configuration, shown as an exit arrow inFig. 1.

Changes in adaptive cycle are determined by two dimensions,connectedness and potential (Holling, 2001). The former dimensionis the visual depiction of a cycle and stands for the ability tointernally control its own destiny. In words of Gunderson andHolling, it “reflects the strength of internal connections thatmediate and regulate the influences between inside processes andthe outside world” (Gunderson and Holling, 2002: 73). The po-tential dimension is represented by the vertical axis (Fig. 1), andstands for the inherent potential of a system that is available forchange. Social or cultural potential can be characterised by theaccumulated networks of relationships. Along the four phases thelevels of both dimensions differ, showing distinct combinations ofhigh or low potential and connectedness (Fig. 1).

2.3. The Cascade Model

As explained above, Pleistocene climate and environmentalchanges have been put forward on many occasions to explainmultiple human population dynamics, as well as to explain majorcultural changes. To further understand the relation betweenPleistocene human populations and climate variations, a number ofmodels have been developed (i.e. Gamble, 1993; Bradtm€oller et al.,2012). Bradtm€olle�rs “Cascade Model” (CM), successfully applied toIberian Late Pleistocene populations, is focused on the beginning ofthe release phase (U) of the ACM (Bradtm€oller et al., 2012). Climaticinstability and the restricted tolerance of the system to stress areused as the basic release factors, while minor disturbances can beunderstood as components of the normal system development. Asa reaction of the system towards increasingly extreme climates,Bradtm€oller et al. (2012) propose four modes of societalreorganization:

- Resistance: it is the primary mode of reorganization as a reac-tion to climate change. The subsistence patterns can bedescribed as a cultural adjustment within the same territory.

- Retreat: climatic instability leads to retreat into the followingreorganization mode. This corresponds to the observation thathuman groups may occasionally only abandon certain parts oftheir settlement area.

- Micro-extinction: a first reaction to accelerated climatic forcingwould be micro-extinction of peripheral groups.

- Macro-extinction: corresponds to the complete collapse of thecultural system followed by population breakdown.

3. Early Middle Paleolithic record in the Duero Basin

3.1. Earliest evidence

Some of the earliest Middle Paleolithic evidence in Europe isregistered in the DB (Fig. 2). The sequence at Gran Dolina TD10(Atapuerca complex, northeast DB) has revealed a progressivereplacement of macro tools by highly standardized small tools, anincrease in hierarchized and predetermined flaking methods(Levallois-like methods) and specialization in the management ofthe different lithic resources (Oll�e et al., 2013). Other indicators inthis sequence, such as the presence of bone tools and hammers,evidence of hafting and even of big cat hunting, reveal theappearance of complex behaviors (M�arquez et al., 2001; Rodríguez,

Fig. 2. First Population Dynamic Model: adaptive phases and dates of the main Early Middle Paleolithic sites in NW Iberia.

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8878

2004; Blasco et al., 2010; Rosell et al., 2011). Available geochrono-logical studies of the upper Units of TD10 provide a series of ESR/UTh dates that include: two dates of 418 ± 63 and 337 ± 51 ka forTD10.2, a date of 379 ± 57 ka for the bottom of TD10.1, and a meandate of 337 ± 29 for its top (Falgu�eres et al., 1999). However, TLdating on polymineral fine-grain fractions in the lower section atTD 10.2 has produced a younger age of 244 ± 26 ka (Berger et al.,2008.). Quite similar technological traits have been documentedat the open-air site of Ambrona (Santonja and P�erez, 2006), locatedin the eastern junction of the DB with the Ebro and Tajo basins,dated to ~350 ka by combined ESR/U-series of fossil teeth(Falgueres et al., 2001).

3.2. Populations on the plateaus and plains

An abundant open-air mid-late Middle Pleistocene lithic recordwith important technoeconomical implications has been preservedon the DB plateaus, located in the central region of the eastern halfof the Basin. In the studied area (1000 km2), intensive surveying hasbeen undertaken in 555 sample units located randomly across theplateaus, and these have providedmore than 25,000 artifacts (Diez-Martín, 2000; Diez-Martín et al., 2008a; S�anchez Yustos, 2009;S�anchez Yustos et al., 2010) (Fig. 3). A total of 44 high-densitypatches have been recorded in a continuous background of low-density evidence scattered over these tablelands (Fig. 3). Plowingis the main post-depositional disturbance force that has affectedthe Paleolithic record documented, destroying the original strati-graphic sequence, as plateau soils tend to be no thicker than 40 cm(Diez-Martín, 2010). Only at rare locations a deeper stratigraphicsequence is preserved. The site of Valdecampa~na 4 is one of thesesites where the lower stratigraphic horizons are not altered byplowing, as documented by the excavation conducted in the centraland lowermost zone of this sinkhole (Diez-Martín et al., 2008b). TLanalysis on four burnt quartzite samples retrieved from differentsites, one of which is the Valdecampa~na 4 excavation, brackets

human occupation on the DB plateaus between 265 and 132 ka(Diez-Martín et al., 2008a) (Fig. 2).

The wealth of information recovered through the surveys carriedout on the DB plateaus reveals a discard, settlement and techno-logical patterning, providing valuable insights into hominin settle-ment systems in this ecological unit. There is a recurrent discardpattern in the archaeological continuum: high-density areas ofstone tools are exclusively linked to the exo-karst formations pre-sent on the plateaus (sinkholes and karst valleys), while low-densityevidence is randomly distributed (Fig. 3). The continuous low-frequency of stone tools across this ecological unit shows thatPaleolithic groups were repeatedly accessing large areas, whereashigh levels of activity would always have been limited to theabundant sinkholes and karst valleys scattered over these table-lands. Patches with high-density of material could operate as focalpoints, understood in the context of short-term activities carried outrepeatedly in discrete spots over time, and consistently their lithicassemblages should be the result of multiple discard episodes.

The accumulation of stone tools in sink-holes suggests thathominins systematically occupied areas with small lakes and pools.The network of shallowwater bodies scattered across the highlandsplayed a key role in settlement patterns. The sink-holes closer tothe escarpments, which facilitated the control of extensive areas ofthe valleys below, are in most cases the preferred zones (revisitedover time) for stone tool accumulation. Therefore, settlement se-lection was also driven by visual prominence over the valleys.Human groups that inhabited the plateaus exhibited a close eco-nomic relationship with the main adjacent river valleys. A perma-nent flux is clearly observed through the transfer of raw materialsfrom river terraces to plateaus, since these highlands lack rawmaterial sources.

The distance to raw material sources has important conse-quences in both settlement patterns and reduction sequences. Thelonger the distance to raw material procurement localities, thelower the frequency of stone tools (20 km is the longest distance

Fig. 3. Location of the Duero Basin Plateaus, distribution of archaeological sample units and types of density in the Duero-Jaramiel-Esgueva Plateaus.

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e88 79

documented). The longer the distance to these localities is, thebigger the spatial fragmentation of reduction sequences. No sig-nificant differences in technological behaviors exist according tothe density of objects and the whole lithic record documented onthe DB plateaus corresponds to a single techno-typological horizon.However, the reduction sequences documented display significantdifferences in terms of spatial fragmentation and transport ofartifacts.

Levallois cores and handaxes constituted the most mobile ele-ments within the plateaus and their production sequences arecharacterized by large-scale fragmentation. These traveling com-ponents are absent from most collections and, when found, theirrepresentation is quite limited. The same is true for choppers,chopping tools, trihedral picks and small cleavers. Among macrotools, choppers are the most common and cleavers are the leastrepresented. Bifaces were knapped into standardized, symmetricaland non-varied shapes (mainly lanceolates, amygdaloids andovates), showing regular flaking patterns. Trimming occurs regu-larly and soft-hammer use is observed in 41% of the bifacesanalyzed. The bulk of the reduction sequences are characterized bydiversity and formal standardization both in small/medium-sizedflake production systems and flake retouching. In some cases,these reduction sequences are fragmented spatially and some ob-jects (either cores or flakes) are transported across the plateaus.

Changes in settlement and land use patterns could trigger thecomplex technological behavior documented on the DB plateausduring MIS 7e6, characterized by: the still abundance of cobbletools and recurrent presence of Large Cutting Tools (mostly bifaces),the multiplication and formal standardization of production andconfiguration processes, the presence of a well-developed Levalloismethod, the large-scale fragmentation of the most elaboratedreduction sequences and the consequent emergence of a travelingtool-kit, Levallois cores and bifaces being the most remarkable el-ements of such a tool-kit. In sum, the recurrent although scarce

presence of the Levallois method, the large-scale fragmentation ofsome reduction sequences and the recurrent transport of the mostoperative tools justify the EMP ascription of these assemblages.However, the recurrent presence of cobble tools and Large CuttingTools, that resembles the Acheulean assemblages recorded in thevalleys (S�anchez Yustos, 2009; S�anchez Yustos and Diez-Martín,2010), prevent the Mousterian ascription of these assemblages(Table 1).

At the other extreme should be placed the abundant lithiccollections, usually retrieved from secondary contexts, collectedon the intermediate terraces located in the middle and lowercourse of the main rivers of the DB plains. These assemblages, inwhich cobble tools and bifaces are abundant and the Levalloismethod is absent or rare, are characterized by: simplicity oftechnical production processes; high uniformity, which results in apoor morphological and technical variability; low functionalspecialization; and remarkable immediacy in the procurement,processing, use and discard of objects (Martín Benito, 2000;S�antonja and P�erez, 2010; S�anchez Yustos, 2009). The few fluvialsequences excavated in the DB have not provided precisegeochronological information (Santonja and P�erez, 1984, 2005;Arnaiz, 1995). In the nearby Tajo Basin, intermediate terracesfrom the Henares, Jarama and Manzanares rivers have been datedbetween MIS 11 and MIS 6 (Benito et al., 1998; Baena et al., 2010;Panera et al., 2011).

3.3. Population on the borders

The intense occupation registered during the second half theMiddle Pleistocene in the Duero depression (plains and plateaus) isfollowed by the conspicuous absence of an Upper Pleistocene re-cord, neither Acheulean, Mousterian or Upper Paleolithic (Santonjaand P�erez, 2002, 2010; Delibes and Diez-Martín, 2006; S�anchezYustos, 2009; S�anchez Yustos et al., 2011), with the exception of

Table 1Technological features of the 44 large lithic concentrations on the Duero Basin Plateaus.

Sites Artefacts Cobble tools Large cutting tools Cores/Levallois

Llano de la Encina III 804 16 6 142/3Monte de Olmos 685 6 2 104/3La Hoyada del Pleito 669 4 2 97/2Valdecampa~na IV-Survey 607 e e 76/3Los Altillos V 584 2 e 72/3Pe~nalba 570 12 e 90/1Valdecampa~na I 531 3 2 76/1C�aquera 507 12 4 159/1Las Callejas 490 2 e 101/2La Cerca de la Serrana I 489 2 e 70/-Redondillo 486 14 4 148/2Valdegallaras II 466 2 1 84/-Cuesta Alta 442 14 3 71/1Mesamediana 429 1 2 94/3Valdegallaras I 427 e e 8/-La Casa del Cura 422 4 3 81/1Las Hontanillas 416 11 e 74/-La Hoyada 398 6 3 137/2La Cerca de la Serrana II 390 4 4 156/2P�aramo I 384 21 6 109/-P�aramo III 382 14 4 138/1Los Altillos I 365 e e 59/5Valdecampa~na II-Survey 365 e e 68/2P�aramo VI 357 16 4 114/1Las Canteras 357 17 6 165/4Los Altillos III 352 e e 46/2Fuente de los Frailes 333 20 6 123/2Llano de la Encina I 323 12 4 119/1P�aramo IV 322 18 4 142/-Llano de la Encina II 321 7 3 136/1Chorro de Luquillas 317 4 8 93/2Fuente de la Bodeguilla 300 24 6 118/3El Herrador 290 4 4 75/2Los Altillos II 279 2 e 49/1P�aramo II 263 18 4 113/1Fuente de Arriba 263 16 3 121/5Valdeperros II 251 1 1 28/3P�aramo V 244 26 e 91/3Cuesta del Pico 243 1 e 55/-Vadecampa~na III 239 e e 37/2Valdecampa~na IV-Excavation 204 e e 16/-Valdecampa~na II-Excavation 198 e 1 9/-Torozos V 191 1 e 36/1Los Altillos IV 147 1 1 16/-Torozos IX 122 e e 13/-

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8880

the scarce and controversial open-air Mousterian record docu-mented in the central area of the basin (S�anchez Yustos and Diez-Martín, 2006-7). To date, the Mousterian Upper Pleistocene re-cord is exclusively located on the borders of the basin and ispractically restricted to a cave habitat. However there are a fewopen-air sites located in the northeast border of the basin, largepart of them are very close to the Atapuerca karst system (Arnaizand Mediavilla, 1986; Mosquera et al., 2007; Navazo andCarbonell, 2014).

Mousterian sites in the DB have provided very fragmentary data(Moure and García Soto, 1983a: 26; Delibes et al., 1988; Neira et al.,2006: 116), although there are a small number of sites with datedsequences and abundant archaeological record. With the exceptionof a few sites dated between MIS 4 and early MIS 3 (Navazo andCarbonell, 2014; �Alvarez Alonso et al., 2014), the other Mouste-rian dated sequences are placed in MIS 5 (Quam et al., 2001; Díezet al., 2008; Diez-Martín et al., 2011; new dates of La Ermita cavepresented here).

Cueva Coraz�on (Mave, Palencia) is located on the southernborder of the natural corridor connecting the DB with the Canta-brian central area, in a canyon cut by the Pisuerga River (Fig. 4). TLdates for the Mousterian levels excavated, on thermo-alteredquartzite, are 96.5 ± 7.8 ka and 95.7 ± 7.4 ka (Díez-Martín et al.,

2011). The lithic assemblage, consisting of 149 objects, is charac-terized by the absence of macro tools and a relative abundance ofprepared core technologies made on non-local flints, while the highfragmentation of the reduction sequences suggest high territorialmobility (S�anchez Yustos et al., 2011). The analysis of the faunalremains suggests an anthropogenic accumulation of horses, deer,goats and other herbivores. Cut marks have been identified inalmost all anatomical parts, indicating a variety of activities relatedto skinning, disarticulation and fleshing, as well as marrowextraction (Yravedra et al., 2013).

Valdegoba Cave (Hu�ermeces, Burgos) is located on the southernborder of the Cantabrian Cordillera, in a small canyon cut by theUrbel River which provides access to the DB (Fig. 4). TheMousterianlevels underlie a laminar calcite concretion dated between 95 kaand 73 ka (Quam et al., 2001). Remains of five Neanderthals werefound together with abundant lithic and faunal remains (Díez et al.,88-9, Díez, 1991; Quam et al., 2001). The technological strategies,employed on local materials, aimed at small-sized flake productionthrough standardized flaking methods (discoidal is the mostfrequent), and the Levallois method is only occasionally used. Twoflint bifaces were recovered. The reduction sequences display nullor reduced fragmentation. The faunal remains exhibit significantdifferences in the various levels, suggesting that the Neanderthal

Fig. 4. Second Population Dynamic Model: adaptive phases, possible population dynamics (hypothesis) and dates of the main Early Mousterian sites in NW Iberia (Cova Eir�os:Lazu�en et al., 2011; Cueva del Camino: Arsuaga et al., 2012).

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e88 81

occupants changed their subsistence strategies from the exploita-tion of grassland species to an increasing reliance on mountainspecies (Díez, 1991).

The upper course of the Arlanza River, in the western foothills ofthe Iberian Range, has revealed an intense Neanderthal occupation(Moure and García Soto, 2000; Díez et al., 2008) (Fig. 4). The ex-cavations carried out first in La Ermita Cave and then in theneighboring Mill�an Cave have unearthed Mousterian levels in bothsites (Moure, 1978; Moure and García Soto, 1983a). At the time ofthe excavations, these levels were dated by 14C (Moure and GarcíaSoto, 1983b; see Table 2). Due to the anomalous results obtained inLa Ermita Cave, new samples were dated by the 14C-method (Moureet al., 1997) (Table 2). The new date was more coherent with theMIS 3 dates obtained inMill�an Cave, however it was still consideredtoo young (Moure et al., 1997: 80). For this reason, two samples forU-series dating were taken from the laminar calcite concretion thatcaps the Pleistocene sedimentary sequence in La Ermita. Afterrecalculating the original dates based on more recent decay con-stant values, the results, presented here for the first time, are101.8 ± 4.0 ka and 95.1 ± 5.7 ka. Another group of researchers has

Table 2Dates from the Mousterian levels in La Ermita Cave and Mill�an Cave.

Level Dates (ka)

La Ermita Cave5b 13.05 ± 195a 11.45 ± 165a 31.1 ± 555a 128.83 ± 39.195a 114.33 ± 41.92Calcite flowstone 101.80 ± 4.0Calcite flowstone 95.10 ± 5.70Mill�an Cave1b 37.45 ± 651a 37.60 ± 70

recently conducted Paleolithic research in this area and dated sixteeth of adult horses unearthed in the old excavations in La Ermitaby the acid racemization method. The dates obtained are128.8 ± 39.1 ka and 114.3 ± 41.9 ka (Díez et al., 2008). Althoughthese results were considered to be too old by these researchers,they are consistent with the U/Th dates now available. According tothe new datings (Table 2), the Mousterian levels in La Ermita areolder than previously thought and may be placed in MIS 5.

The abundant lithic collections from both caves in the uppercourse of the Arlanza River display strong techno-typological sim-ilarities (Moure and García Soto, 1983a). They are characterized bylocal raw material procurement, null or reduce fragmentation ofthe operative sequences, which are aimed at producing small-sizedflakes, low representation of the Levallois method, high represen-tation of Quina scrapers and total absence of macro tools (Moure,1978; Moure and García Soto, 1983a). The faunal remainsretrieved in La Ermita reveal differences in each level that suggest achange in subsistence patterns, from a recurrent exploitation ofgrassland species to an increasing consumption of woodland spe-cies, while the intense exploitation of mountain species is

Dating method Reference

14C Moure and García, 1983a14C Moure and García, 1983a14C Moure et al., 1997Acid racemization Díez et al., 2008Acid racemization Díez et al., 2008U/Th This paperU/Th This paper

14C Moure and García, 198214C Moure and García, 1982

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8882

maintained throughout the sequence (Delibes de Castro, 1972; Díezet al., 2008). In accordancewith the paleo-environmental data fromLa Ermita (Delibes de Castro, 1972; Moure et al., 1997), the uppercourse of the Arlanza River was occupied by Neanderthals groupsduring a warm climate phase.

4. Early Middle Paleolithic record in the Cantabrian Region

The Acheulean record, the earliest Paleolithic evidence in thisregion, is limited in size, biased and composed of few elements andbadly chrono-stratigraphically and techno-typologically defined(�Alvarez-Alonso, 2012). The few stratified deposits with Acheuleanartifacts have not provided numerical chronologies (Rodríguez-Asensio, 1993, 2001; Arrizabalaga and Ríos, 2012). However,based on techno-typological, geological and paleo-environmentaldata, the Acheulean horizon has been tentatively ascribed to thelast third of the Middle Pleistocene and beginning of the UpperPleistocene (Montes, 2003; Rodríguez-Asensio and Arrizabalaga,2004; Arrizabalaga and Ríos, 2012; �Alvarez-Alonso, 2012).

The first evidence of technological change towards the Mous-terian is identified at the beginning of the Upper Pleistocene (MIS5). The best sequences to contextualize the early Mousteriantechnology in this region are El Castillo (Puente Viesgo, Cantabria)and Lezetxiki (Arrasate, Basque Country) (Fig. 4). Level 23 at ElCastillo, dated to 92.8 ka and 89 þ 11/-10 ka, provides a minimumchronology for the Mousterian. The lower levels (24e26) with aminimum age of MIS 5c were first ascribed to the Acheulean(Cabrera et al., 2000). However, later studies have ascribed Levels24e25 to theMousterianwith the presence of macro tools (Montes,2003; �Alvarez Alonso, 2012). The lower levels of Lezetxiki (VI, VIIand VIII) are placed between the end of MIS 6 (Level VIII, archae-ologically sterile) and the beginning of MIS 5 (Level VI). Levels VIand VII are ascribed to the early Mousterian, and macro tools arepractically absent. The earliest Mousterian level registered inCovalejos (Pi�elagos, Cantabria) is placed between 101 ka and 94 ka.This level has not yielded more information because the excavatedarea is very reduced. There are other sites that have provided lithicassemblages classified as early Mousterian with macro tools andhave been tentatively assigned to MIS 5, such as La Garma A exte-rior (Santander, Cantabria), the lower levels in Arlanpe (Lemoa,Basque Country) and Ba~nugues (Goz�on, Asturias) (Rodríguez-Asensio, 1993; Tapia, 2010; Ríos et al., 2011; �Alvarez Alonso,2012). The scarcity of the fossil record retrieved in these earlyMousterian levels does not allow a characterization of Neanderthalsubsistence strategies.

According to some archaeologists, the chronological andtechno-typological parallels that the early Mousterian (EM) showswith the later Acheulean (LA), would suggest that the Acheulean-Mousterian gradient took place without any rupture or fissureduring MIS 6e5. This techno-cultural continuum is called “Canta-brian Ancient Paleolithic” (Rodríguez-Asensio, 2000; �AlvarezAlonso, 2012). Following this line of inference, it is suggested thatthe Ancient Paleolithic settlement pattern followed the coastalcorridor from east to west, and Neanderthals populations fromAquitaine would penetrate in the CR (Rodríguez-Asensio andArrizabalaga, 2004: 59; Djema, 2008). Anyway, the EM is likely tobe the substrate from which the subsequent classical Mousterianeventually originated between MIS 4 and MIS 3 (�Alvarez-Alonso,2012).

5. Modeling population dynamics

With the outline of basic EMP techno-economic behavior inmind, long-term population dynamics during the second half of theMiddle Pleistocene and the beginning of the Upper Pleistocene in

the study area can be defined and, following the heuristicsexplained in previous sections, the socio-environmental model ormodels that best fit each population dynamic may be proposed.

5.1. Population dynamic before MIS 5

Differences in location, physiographic conditions and occu-pation intensity would suggest that the DB was a source habitat,while the CR was a sink habitat. Due to the peripheral and iso-lated character of the CR within the Iberian Peninsula, flanked onthe north by coast and a long mountain range on the south andwith the “basque crossroads” (the eastern edge of the Cantabrianmountain Range) as its most accessible entrance (Arrizabalagaand Ríos, 2012), it is quite reasonable that the continuity ofthe CR population would depend on adjacent more reproductiveareas as its rate of reproduction would be insufficient to balancelocal mortality. The Lower and Middle Paleolithic settlement inthis region is very recent compared to nearby regions (DB orAquitania) which seems to support its correct attribution as asink habitat. Conversely, the DB is placed in a more advanta-geous position within the Iberia Peninsula in terms of proximityof another regions (Tajo Basin, Ebro Basin, Portuguese-DouroRegion and CR). The DB could support a larger and reproduc-tive population due to its location and extension (the largestCenozoic basin in Iberia). The abundance of archaeological re-cords during the second part of the Middle Pleistocene is inaccordance with its attribution as source habitat during this spanof time.

Growth and conservation are the adaptive phases that bestdescribe the development cycle of slow growth that representsthe intermittent presence of the behavioral changes that charac-terized the EMP. The chronological range of this developmentcycle goes from an early phase of MIS 11 to a mid-late phase of MIS6 (Fig. 5). The maintenance of this cycle requires an ecologicalsuccession in ecosystems and increasing demand for resources.One of the adaptive responses that the source population maycarry out in this sense would be the expansion of its operatingterritories through the occupation of new ecological units andemigration to neighboring regions. It is possible to envisage thatthe source population of the DB could keep in recurrent contactwith other neighboring populations or could seasonally occupyneighboring regions (Tajo and Ebro Basins). It is interesting to notein this regard that Atapuerca and Ambrona are placed on strategicnatural corridors (Fig. 2). Following this line of inference, the CRwould probably be colonized by groups from the DB; although, wecannot exclude that this region was colonized before by pop-ulations from other regions (Ebro Basin and/or Aquitaine). In anycase, the population of the CR during the second half of the MiddlePleistocene could be the result of several migration waves fromdifferent source populations.

Throughout this adaptive phase two different techno-complexes (LA and EMP) are documented in the DB, while thelimited number of well-documented and chronostratigraphicallycontextualized sequences in the CR prevents an appropriate char-acterization of the techno-economic behavior of the sink popula-tion during this period. This lithic variability could be interpreted asresult of either different cultural traditions or eco-functional dif-ferences within the same tradition. However, with the level ofresolution provided by the current record and methods appliedhere, the most parsimonious hypothesis would be that bothtechno-complexes were made by the same population, althoughwithin this population there might be two cultural traditions(about the controversial surrounding the concepts of “culturaltradition” and “cultural evolution” in Paleolithic Archaeology seeS�anchez Yustos, 2012).

Fig. 5. Complete adaptive cycle and main Middle Paleolithic sites in NW Iberia.

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e88 83

5.2. Population dynamic during MIS 5

The Middle-Upper Pleistocene transition entails a drasticchange of population dynamic in NW Iberia. The intense occupa-tion registered during the second half of the Middle Pleistocene onthe plains and plateaus of the DB is followed by the conspicuousabsence of an Upper Pleistocene record, which could respond to ademographic vacuum. The fewMousterian sites are now located onthe borders of the basin, and are concentrated mostly during MIS 5.Conversely, the CR appears to show more solid Neanderthal set-tlement from MIS 6e5 up to MIS 3. Two different socio-environmental models (Fig. 4) may explain this change of popula-tion dynamic as the reaction to an increasingly extreme climatethat triggered the beginning of a release phase and the subsequentreorganization phase (Fig. 5).

5.2.1. Retreat modelClimate deterioration in late MIS 6 could have had a tremendous

impact on the source population in the DB. Unfortunately there isno available paleo-environmental or paleo-ecological informationabout MIS 6 in the DB. Although population breakdown couldhappen on a small scale, the population retreated to the border ofthe Basin, more precisely to the upper course of the rivers whereecological diversity is greater and there are corridors to other re-gions (Fig. 4). The groups that retreated to the North met thepopulation in the CR. At that moment, the Middle Pleistocenesource-sink population dynamic established between the DB andCR was broken, and the latter area became an isolated Neanderthalglacial refugium on the northern edge of Iberia. As a refugium, itcould also host populations from other areas such as Aquitaine thatcould has been played as source region during the second part ofthe Middle Pleistocene. In any case, the CR remained a sink habitatin the sense that it needed recruitment from source populations tomaintain a demographically stable population. After a short releasephase at the end of MIS 6, the interstadial MIS 5e represent thereorganization phase in which innovative strategies of resource

exploitation are developed and networks were reorganized and, inaccordance with this, technology and tool production was alsoreorganized (Fig. 5). In this context, Mousterian technology wouldemerge in NW Iberia.

5.2.2. Micro-extinction modelAnother possible scenario could be that accelerated climatic

deterioration in late MIS 6 had a dramatic demographic impact onthe source population of the DB, pushingmuch of this population toextinction. The source-sink dynamic between the DB and CR wasbroken. However, this release phase was less dramatic in the CRpopulation. This region functioned as a glacial refugium and,therefore, could receive populations from other regions andmaintain a demographically stable population, at least during MIS5. The Mousterian emergence could have been the result of areorganization phase within the CR population or introduced byimmigrant populations. Be that as it may, with the onset of warmconditions (MIS 5), the population expanded to its maximumextent, occupying the northern border of the DB and even reachingthe southeastern border (Fig. 4). In this expansive context, Mous-terian technology would spread throughout NW Iberia.

6. Discussion

The non-lineal scenarios that we have proposed are in fullaccordance with the lack of linearity in the Neanderthalizationprocess, currently proposed (Hublin and Roebroeks, 2009; Greenet al., 2010; Dennell et al., 2011). The complex population dy-namics that we have suggested, in which population isolation,fragmentation, recombination and extinction would have workedas key factors, may help explain the morphological variability ofEuropean Middle Pleistocene hominins. One interesting possibilityin such complex and dynamic scenarios is that populations thatwere ancestral to another could have interbred with their owndescendants (Dennell et al., 2011; Bermúdez and Martin�on, 2013).Indeed, the scenario of possible interbreeding between

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8884

Neanderthals and their own ancestors has been already suggested(Pa€abo, 2009).

The population landscape outlined here resembles the ‘‘tidalislands’’model proposed by Dennell et al. (2011). According to this,populations were sometimes isolated and confined to refugia inglacial phases, and at other times, in interglacials or interstadials,when populations expanded to their maximum extent, able tointerbreed with other populations. The genetic tide thus ebbed andflowed to the climatic rhythms of the Pleistocene, namely: coldperiods of increased environmental stress and lessened biologicalproductivity, and warm periods of lessened environmental stressand increased biological productivity.

The same researchers have also suggested that Middle Pleisto-cene lithic variability may have resulted from non-lineal populationscenarios under conditions of repeated climate and environmentalinstability (Dennell et al., 2011). This is not the case of the lithicvariability registered in the DB during the second part of theMiddlePleistocene. According to the first model suggested here, the DBsource population transited between growth and conservationadaptive phases, and no demographic discontinuities have beendetected (Fig. 5). The maintenance of this cycle required anecological succession in ecosystems. The paleo-ecological evidenceretrieved in Atapuerca has evidenced the absence of extremelyharsh conditions during the mid-Middle Pleistocene in the DB (MIS11e7) (Rodríguez, 2006; Rodríguez et al., 2011). Accordingly, in thestudy area EMP type behavior would not be interpreted as a directresponse to the climate rhythms of the Pleistocene. The reason forthe EMP technological changes should be better sought in thementioned development cycle; in particular, in the increasingcomplexity of the sociocultural systems, which were likely to havebeen fueled by progressive demographic expansion. As a result,territorial mobility increased, and new settlement and land usepatterns emerged, while both predation pressure and culturalinteraction escalated. The occupation of the DB plateaus representsthe consolidation of EMP type behavior among the sourcepopulation.

It is interesting to note the importance that many researchershave granted to mobility patterns as a central key to restructuringMiddle Paleolithic technology (i.e. Geneste, 1985; F�eblot-Augustins,1993; Floss, 1994). Furthermore, it has been suggested that theLevallois technique is the result of changes in mobility strategies(Geneste, 1989; Gamble and Roebroeks, 1999; Auguste et al., 2005;Bocquet-Appel and Tuffreau, 2009; Villa, 2009; Scott and Ashton,2011; contra Moncel et al., 2012: 662). On the other hand, behav-ioral overlaps and a mosaic of changes in subsistence and technicalbehavior during the second half of the Middle Pleistocene wouldsuggest that Levallois technology in Eurasia was developed in situby different populations (Moncel et al., 2012; Adler et al., 2014). Insum, EMP type behavior may be interpreted as a multifacetedprocess in which sociocultural structures become more complex,while the population landscape favors both lithic and geneticvariability (i.e. Hublin and P€a€abo, 2005; Fabre et al., 2009; Villa,2009; Green et al., 2010; Dennell et al., 2011; Moncel et al., 2012;Arsuaga et al., 2014). There is no way at present to evaluatewhether lithic variability during MIS 11e6 in the study area is theresult of different cultural traditions or not.

The second population dynamic proposed here is largely sup-ported under the assumption that climatic deterioration in late MIS6 would have a profound impact on the source population of theDB. There is no paleo-environmental or paleo-ecological informa-tion about MIS 6 in the DB, but paleoclimate reconstruction fromGreenland ice cores and deep-sea cores suggest that during MIS 6the Earth experienced combinations of climatic boundary condi-tions (McManus et al., 1999; Kandiano and Bauch, 2003), while inmainland Europe ice sheet extent was greater than at any time

during the Last Glacial (Ehlers and Gibbard, 2004, 2007; Calleoniet al., 2009). Regardless of those recent geologically-based re-constructions of the Eurasian ice sheet, relatively little attention hasfocused on characterizing the details of climatic and environmentalresponses during MIS 6, compared with the increasing number ofhigh-resolution climate records for Eurasia during the Last Glacialcycle, which has allowed researchers to link population turnoversand cultural shifts with the North Atlantic Heinrich Events (i.e.Stringer et al., 2003; Finlayson and Carri�on, 2007; Banks et al.,2008; Shea, 2009; Müller et al., 2012; Schmidt et al., 2012). Thehighest-resolution environmental record for MIS 6 produced todate came from north-west Greece (Rocoux et al., 2011). Accordingto the authors of the study, the record from this refugial area re-sembles the pattern of changes during the Last Glacial: an earlyperiod (185e155 ka) is similar to MIS 3 and a later period (155e135ka) is similar to MIS 2 (Roucoux et al., 2011).

In order to evaluate the impact of the late climatic phase of MIS6 on the DB source population, we have focused our attention onthe distribution of archaeological sites on the landscape plottedagainst chronology. The youngest dates obtained in the DB plateausreveal that in late MIS 6 there were groups in the central area of thebasin that resisted this extreme climate phase (Fig. 3). As a conse-quence, the occupation of this ecological unit could be extended toearly phases of the Upper Pleistocene. In either case, these latteroccupants of the plateaus with a still pre-Mousterian technologylikely represent the end of the source population that inhabited theDB during the second half of theMiddle Pleistocene, as according tothe archaeological evidence during the Upper Pleistocene thecentral area of the basin seems to have been deserted, perhaps withthe exception of sporadic visits in warm phases of the Last Glacialcycle. By contrast, a large number of caves with Mousterian levelsare known in the mountainous borders of the basin or in adjacentterritories during MIS 5 (Fig. 4).

According to the heuristics of the ecological models proposedand in light of the current evidence, the change of population dy-namic registered at the Middle-Upper Pleistocene transition in NWIberia (Fig. 4) involved a short-term release phase at late MIS 6,followed by a reorganization phase in which socio-economic net-works were reorganized at early MIS 5. However two differentpopulation scenarios (retreat hypothesis or micro-extinction hy-pothesis) could explain such a population dynamic. However,neither hypothesis can be conclusively validated with the level ofresolution that provides the current archeological evidence.

The micro-extinction hypothesis would imply that the impact ofthe late climatic phase of MIS 6 on the DB source population wasextreme to the point of its extinction, and when the climatic con-ditions improved (MIS 5e) the sink population from the CRincreased and expanded its territories southward. However, a sinkpopulation would disappear without continued immigration fromsource areas. The hypothetical demographic growth of the CR sinkpopulation would had thus required a regular immigration fromother source areas that would had not been dramatically affectedby the deterioration of the climatic conditions at late MIS 6 asseems to be the case of the DB. However, the scarce archaeologicalevidence documented in the CR at MIS 6/5 can not support thatpremise. Another weakness point of this hypothesis is that in lateMIS 6 there were groups in the central area of the DB that mayresist this extreme climate phase, therefore population break-downs could have just been on small scale. Likewise, according tothe micro-extinction hypothesis the presence of EM sites in thewestern (Cova Eir�os), eastern (Cueva de la Ermita) and southern(Cueva del Camino) mountain borders that surround the DB (Fig. 4)have to been explained due to a notable population movement(from North to South) that should have been based on an extraor-dinary demographic growth. However, precisely such demographic

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e88 85

growth is the most controversial point of the micro-extinctionhypothesis.

The distribution of EM sites in NW Iberia is thus more consistentwith the hypothesis of a centrifugal retreat of the source populationto the mountains and adjacent territories rather that a scenario ofmicro-extinction and posterior Mousterian colonization (Fig. 4).According to the retreat hypothesis, late MIS 6 was an extremeclimate phasewith a profound impact on the DB population but notto the point of its total extinction. Although all Neanderthal groupsthat composed the DB population did not respond in the same waynecessarily, the tendency was to abandon their settlement terri-tories and move to new ones. The DB population is fragmented andretreated to the different mountainous borders of the DB. Suchfragmentation of the DB source population and its meeting withother populations would mark the beginning of a new populationdynamic in NW Iberia.

Summarizing, the most reliable interpretative tool available upto date is the distribution of archaeological sites on the landscapeplotted against chronology which is more consistent with theretreat hypothesis. However, further research is required to verifyor falsify either hypothesis or assess to what extent differentpreservations conditions or spatial distribution of research pro-jects may influence in the demographic discontinuities hereidentified.

Finally, as we have attempted to show, the CR acted as Nean-derthal glacial refugium during the Upper Pleistocene, and it ismost likely that it would host populations from different origins.The technological parallels suggested between Aquitaine and CRcould point in this direction (Djema, 2008; �Alvarez-Alonso, 2012).This fact confirms the idea that “lithic variability may haveresulted from demographic discontinuities and regroupings un-der conditions of instability” (Dennell et al., 2011: 1522).Furthermore, the lithic record in the CR between MIS 6 and MIS 5is characterized by high lithic variability with significant techno-typological differences among assemblages, consequently they“have always fallen into one or other category (LA or EM)depending on the emphasis placed on their most relevant fea-tures” (�Alvarez-Alonso, 2012: 302). Without any apparent de-mographic discontinuity, the classical Mousterian record in the CRwas made by the Neanderthal population that inhabited thisnorthern glacial refugium in Iberia during MIS 4 and MIS 3. Thedemographic stability of this population would have been veryfragile without the reproductive surpluses from the productivesource habitat. The reduced genetic diversity identified amonglater Neanderthals in western and central Europe is consistentwith this scenario (Dalen et al., 2012).

7. Conclusions

The EMP archaeological record in NW Iberia fits the ACM verywell, registering a complete adaptive cycle betweenMIS 11 andMIS5 (Fig. 5). The beginning of the cycle is marked by the first evidenceof Middle Paleolithic assemblages recovered in the upper units ofTD10 Gran Dolina site. However, the lower Units of TD 10, as well asother sites in the Atapuerca archaeological complex such as Galeriaand Sima de los Huesos, have yielded an older Middle Pleistocenerecord (Oll�e et al., 2013; Arsuaga et al., 2014; García Medrano et al.,2014). Although techno-cultural differences between them havebeen suggested (Lower Paleolithic vs Middle Paleolithic) (Oll�e et al.,2013), it seems quite reasonable to establish a population contin-uum between the Middle Pleistocene populations, pre- and post-MIS 11.

The early phases of the adaptive cycle (growth and conserva-tion) took place between MIS 11 and MIS 6, and are characterizedby progressive demographic expansion and sociocultural

complexity (Fig. 5). The behavioral changes resulting from thislong-term process were not rooted in environmental stress, sinceduring this period there was an absence of extremely harsh con-ditions in NW Iberia, which favored high biological productivity. Anabundant archaeological record has been documented in the DBduring this period, while in the CR the evidence is very limited andtentatively ascribed to later phases of this period. Such disparity ofevidence, which could be interpreted as different degrees ofoccupation intensity, as well as the locational and physiographicconditions in both regions would suggest that the DB was a sourcehabitat, while the CR was a peripheral or sink habitat. The sourcepopulation in the DB might maintain recurrent contact with otherneighboring populations or may seasonally occupy other regions(CR, Tajo and Ebro basins), while the CR population would be theresult of several migration waves from different sources pop-ulations (DB, Ebro Basin and Aquitaine). Furthermore, during thisperiod the NW of Iberia may have acted as a glacial refugium fornorthern European populations. This complex population land-scape would favor lithic variability, while it may help explain themorphological variability that characterizes the European MiddlePleistocene hominins.

This population dynamic is broken in late MIS 6, when anextreme climate phase had a profound impact over the DB pop-ulation, whose reaction was to retreat to the mountain margins ofthe Basin, where the ecological diversity is greater (ecotones) andthere are also corridors that lead to more favorable regions. Ashort-term release phase (late MIS 6) was followed by a short-term reorganization phase in which socio-economic networkswere reorganized (MIS 5) (Fig. 5). In a context of high environ-mental stress and low biological productivity, Mousterian tech-nology emerged in NW Iberia. In the case of the CR, it became aNeanderthal glacial refugium during the Upper Pleistocene, and isquite likely to have hosted populations from different origins. Thisfact could explain the lithic variability documented in the regionbefore the classical Mousterian, while the reduced genetic di-versity identified among later Neanderthals in western Europecould be explained by the absence of immigration from sourcepopulations.

Finally, the model proposed here has provided a theoreticaland methodological frame to interpret EMP population dynamicsin NW Iberia (Fig. 5). Although future discoveries and subsequentstudies can confirm or falsify the models proposed, it can bestated that the populations studied “were not simply dancing tothe rhythms of the Pleistocene” (Gamble, 1999: 125) since, as wehave tried to explain, during a long time their socioculturalsystem was largely shaped by internal forces. However, Paleo-lithic population dynamics and sociocultural systems cannot beunderstood outside the climate and environmental rhythms ofthe Pleistocene.

Acknowledgements

We wish to thank Profs. J.A. Moure, G. Delibes, and the late Prof.M. Hoyos for allowing us to use their unpublished U/Th dates of LaErmita Cave (Duero Basin, Spain).We also thank Prof. Hillaire-Marcel(Univerit�e du Qu�ebec �a Montreal), who undertook the dating pro-gram at La Ermita and kindly helped us recalculating and updatingthe results for this publication. We thank reviewers for providingcomments that significantly strengthened the manuscript.

References

Abel, N., Cumming, D.H.M., Anderies, J.M., 2006. Collapse and reorganization insocial-ecological systems: questions, some ideas, and policy implications. Ecol.Soc. 11 (1), 17. http://www.ecologyandsociety.org/vol11/iss1/art17/.

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8886

Adler, D.S., Wilkinson, K.N., Blockley, S., Mark, D.F., Pinhasi, R., Schmidt-Magee, B.A.,Nahapetyan, S., Mallol, C., Berna, F., Glauberman, P.J., Raczynski-Henk, Y.,Wales, N., Frahm, E., J€oris, O., MacLeod, A., Smith, V., Cullen, V., Gasparian, B.,2014. Early Levallois technology and the Lower to Middle Paleolithic transitionin the Southern Caucasus. Science 345, 1609e1613.

�Albarez-Alonso, D., 2012. First Neanderthal settlements in northern Iberia: theAcheulean and the emergence of Mousterian technology in the Cantabrianregion. Quat. Int. 326e327, 288e306.

�Alvarez-Alonso, D., de Andr�es-Herrero, M., Díez-Herrero, A., Rojo, J., Medialdea, A.,Benito, G., Vegas, J., 2014. Geoarqueología del yacimiento Musteriense delAbrigo del Molino (Segovia). In: Schnabel, S., G�omez Guti�errez, A. (Eds.),Avances de la Geomorfología en Espa~na, XIII Reuni�on Nacional de Geo-morfología, C�aceres, pp. 446e449.

Arnaiz, A., 1995. El Paleolítico Inferior en el tramo medioalto del río Pisuerga:situaci�on actual de la investigaci�on. In: Actas del III Congreso de Historiade Palencia. Palencia: Excma. Diputaci�on de Palencia. Departamento deCultura.

Arn�aiz, M., Mediavilla, O., 1986. Villarmero: un yacimiento “premusteriense” al airelibre en la zona oriental de la submeseta norte. Numantia II, pp. 7e31.

Arrizabalaga, A., Ríos, J., 2012. The first human occupation of the basque crossroads.J. World Prehistory 25, 157e181.

Arsuaga, J.L., Baquedano, E., P�erez, A., Sala, N., Quam, R., Rodríguez, L., García, R.,García, N., �Alvarez-Lao, D., Laplana, C., Huguet, R., Sevilla, P., Maldonado, E.,Blain, H., Blanca, M., Sala, P., Gil-García, M.J., Uzquiano, P., Pantoja, A.,M�arquez, B., 2012. Understanding the ancient habitats of the last-interglacial(late MIS 5) Neanderthals of central Iberia: Paleoenvironmental and tapho-nomic evidence from the Cueva del Camino (Spain) site. Quat. Int. 275,55e75.

Arsuaga, J.L., Martínez, I., Arnold, L.J., Aranburu, A., Gracia-T�ellez, N., Sharp, W.,Quam, R., Falgu�eres, C., Pantoja, A., Bischoff, J., Poza-Rey, E., Par�es, J.M.,Carretero, J.M., Demuro, M., Lorenzo, C., Sala, N., Martin�on, M., García, N.,Alc�azar, A., Cuenca, G., G�omez, A., Moreno, D., Pablos, A., Shen, C., Rodríguez, L.,Ortega, A., García, R., Bonmatí, A., Bermúdez, J.M., Carbonell, E., 2014. Nean-dertal roots: cranial and chronological evidence from Sima de los Huesos. Sci-ence 344, 1358e1362.

Auguste, P., Lamotte, A., Locht, J.-L., Tuffreau, A., 2005. Le traitement de la mati�erepremi�ere lithique et osseuse au Pal�eolithique inf�erieur et moyen dans le nordde la France: �etat des recherches r�ecentes. In: Molines, N., Moncel, M.-H.,Monnier, J.-L. (Eds.), Donn�ees r�ecentes sur les modalit�es de peuplement et sur lecadre chronostratigraphique, g�eologique et pal�eoanthropologique des indus-tries du Pal�eolithique inf�erieur et moyen en Europe, BAR International Series,vol. S1364, pp. 419e431.

Baena, J., Baquedano, I., Ben�eitez, P., Ca~naveras, J.C., Carri�on, E., Lario, J., Martín-Escorza, C., Mazo, A.V., Mill�an, A., S�anchez, S., Sanz, E., Ses�e, C., Silva, P.G.,Yravedra, J., 2010. Síntesis: interpretaci�on general del yacimiento de Tafesa(Madrid). In: Baena, J.Y., Baquedano, I. (Eds.), Las huellas de nuestro pasado,Estudio del yacimiento del Pleistoceno madrile~no de Tafesa (antigua Transfesa),vol. 14. Zona Arqueol�ogica, pp. 191e202.

Baena, J., Carri�on, E., Cuartero, F., Fluck, H., 2012. A chronicle of crisis: the LateMousterian in north Iberia (Cueva del Esquilleu, Cantabria, Spain). Quat. Int. 24,199e211.

Banks, W.E., d’Errico, F., Townsend Peterson, A., Kageyama, M., Sima, A., S�anchez-Go~ni, M.-F., 2008. Neanderthal extinction by competitive exclusion. PLoS ONE 3(12), e3972. http://dx.doi.org/10.1371/journal.pone.0003972.

Bar-Yosef, O., Dibble, H.L., 2005. The Definition and Interpretation of LevalloisVariability. Prehistory Press, Madison.

Barsky, D., Gregoire, S., Moigne, A.-M., 2005. Variabilit�e des types d’occupation etd’exploitation des territoires m�editerran�eens entre 600,000 et 300,000 ans. In:Molines, N., Moncel, M.-H., Monnier, J.-L. (Eds.), Donn�ees r�ecentes sur lesmodalit�es de peuplement et sur le cadre chronostratigraphique, g�eologique etpal�eoanthropologique des industries du Pal�eolithique inf�erieur et moyen enEurope, BAR International Series, vol. S1364, p. 565577.

Benito, A., P�erez-Gonz�alez, A., Santonja, M., 1998. Terrazas rocosas, aluviales ytravertínicas del valle alto del río Henares (Guadalajara, Espa~na). Geogaceta 24,55e58.

Berger, G.W., P�erez-Gonz�alez, A., Carbonell, E., Arsuaga, J.L., Bermúdez deCastro, J.M., Ku, T.-L., 2008. Luminescence chronology of cave sediments at theAtapuerca paleoanthropological site, Spain. J. Hum. Evol. 55, 300e311.

Bermúdez, J.M., Martin�on, M., 2013. A new model for the evolution of the humanPleistocene populations of Europe. Quat. Int. 295, 102e112.

Bernaldo de Quir�os, F., Arrizabalaga, A., Maíllo-Fern�andez, J., Iriarte, Mª J., 2008. Latransici�on paleolítica medio-superior en la regi�on centro-oriental de la cornisacant�abrica. Espac. Tiempo Forma, Ser. 1, 33e46 (Nueva �Epoca).

Blasco, R., Rosell, J., Arsuaga, J.L., Bermúdez de Castro, J.M., Carbonell, E., 2010. Thehunted hunter: the capture of a lion (Panthera leo fossilis) at the Gran Dolinasite, Sierra de Atapuerca, Spain. J. Archaeol. Sci. 37, 2051e2060.

Bocquet-Appel, J.-P., Tuffreau, A., 2009. Technological responses of Neander-thals to macroclimatic variations (240,000-40,000 BP). Hum. Biol. 81,287e307.

Bocquet-Appel, J.-P., Demars, P.-Y., Noiret, L., Dobrowsky, D., 2005. Estimates ofUpper Palaeolithic meta-population size in Europe from archaeological data.J. Archaeol. Sci. 32, 1656e1668.

Boquet-Appel, J.-P., Demars, P.-Y., 2000. Population kinetics in the Upper Palae-olithic in Western Europe. J. Archaeol. Sci. 27, 551e570.

Bordes, F., 1971. Physical evolution and technological evolution in man: a paral-lelism. World Archaeol. 3, 1e5.

Bourguignon, L., Faivre, J.-P., Turq, A., 2004. Ramification des chaînes op�eratoires:une sp�ecificit�e du Moust�erien? Pal�eo 16, 37e48.

Bourguignon, L., Djema, H., Bertran, P., Lahaye, C., Guibert, P., 2008. Le gisementSaalien de Petit-Bost (Neuvic, Dordogne). In: Jaubert, J., Bordes, J.-G., Ortega, I.(Eds.), Les Soci�et�es du Pal�eolithique dans un Grand Sud-Ouest de la France:nouveaux gisements, nouveaux r�esultats, nouvelles m�ethodes. M�emoiresSocie�et�e Pr�ehistorique Française, pp. 41e57.

Bradtm€oller, M., Pastoors, A., Weninger, B., Weniger, G.-C., 2012. The repeatedreplacement model e rapid climate change and population dynamics in LatePleistocene Europe. Quat. Int. 247, 38e49.

Brenet, M., Folgado, M., Lenoble, A., Bertran, P., Viellevigne, E., Guibert, P., 2008.Interpr�etation de la variabilit�e de deux industries du Pal�eolithique moyenancien du Bergeracois: Cantalouette I et Combe Brune 3 (Creysse, Dor-dogne). In: Jaubert, J., Bordes, J.-G., Ortega, I. (Eds.), Les Soci�et�es duPal�eolithique dans un Grand Sud-Ouest de la France: nouveaux gisements,nouveaux r�esultats, nouvelles m�ethodes. M�emoires Socie�et�e Pr�ehistoriqueFrançaise, pp. 57e83.

Cabrera, V., Bernaldo de Quir�os, F., 2000. Excavaciones arqueol�ogicas en la cueva deEl Castillo (Puente Viesgo). Actuaciones Arqueol. Cantab. 1984e1999, 23e32(Santander).

Chase, P.G., 1990. Tool-making and Middle Palaeolithic behaviour. Curr. Anthropol.31, 443e447.

Chazan, M., 2009. Assessing the Lower to Middle Paleolithic Transition. In:Camps, M., Chauhan, P. (Eds.), Sourcebook of Paleolithic Transitions, Methods,Theories, and Interpretations. Springer, New York, pp. 237e243.

Colleoni, F., Krinner, G., Jakobsson, M., Peyaud, V., Ritz, C., 2009. Influence ofregional parameters on the surface mass balance of the Eurasian ice sheetduring the peak Saalian (140 kya). Glob. Planet. Chang. 68, 132e148.

Dalen, L., Orlando, L., Shapiro, B., Durling, M., Quam, R., Gilbert, M., Diez, C.,Willerslev, E., Arsuaga, J., Gotherstrom, A., 2012. Partial genetic turn-over inneanderthals: continuity in the east and population replacement in the west.Mol. Biol. Evol. http://dx.doi.org/10.1093/molbev/mss074.

De Bono, H., Goren-Inbar, N., 2001. Note on a link between Acheulean Handaxes andthe Levallois method. J. Israel Prehist. Soc. 31, 9e23.

Delibes, G., Diez-Martín, F., 2006. El Paleolítico superior en la Meseta norte espa-~nola. In: Studia Archaeologica, vol. 94. Univ. de Valladolid y Fundaci�on Duquesde Soria, Valladolid, p. 213.

Delibes de Castro, M., 1972. Informe Paleontol�ogico de la fauna de la cueva de LaErmita. Not. Arqueol. Hisp�anico Prehist. I, 41e46.

Delibes, G., Esparza, A., García, E., L�opez, J.R., Marin�e, M., 1988. La colecci�onarqueol�ogica del Padre Saturio Gonz�alez en Santo Domingo de Silos. Diputaci�onProvincial de Burgos, Burgos.

Dennell, R.W., Martin�on, M., Bermúdez, J.M., 2011. Hominin variability climaticinstability and population demography in Middle Pleistocene Europe. Quat. Sci.Rev. 30, 1511e1524.

Díez, C., 1991. La grotte de Valdegoba (Hu�ermeces, Burgos, Espagne). Un gisementdu Pal�eolithique Moyen avec des restes humaines. L’Anthropologie 95,329e330.

Díez, J.C., García, M.A., Gil, E., Jord�a, J.F., Ortega, A.I., S�anchez, A., S�anchez, B., 1988-1989. La cueva de Valdegoba (Burgos). Primera campa~na de excavaciones.Zephyrus 41e42, 55e74.

Díez, J.C., Alonso, R., Bengoechea, A., Colina, A., Jord�a, J.F., Navazo, M., Ortíz, J.E.,P�erez, S., Torres, T., 2008. El Paleolítico medio en el valle del Arlanza (Burgos).Los sitios de La Ermita, Mill�an y La Mina. Cuaternario Geo-morfología 22,135e157.

Diez-Martín, F., 2000. El poblamiento paleolítico en los p�aramos del Duero. Serviciode Publicaciones de la Universidad de Valladolid, Valladolid.

Diez-Martín, F., 2010. Evaluating the effect of plowing on the archaeological record:the early middle palaeolithic in the river Duero basin plateaus (north-centralSpain). Quat. Int. 214, 30e43.

Diez-Martín, F., S�anchez, P., G�omez, J.A., G�omez de la Rúa, D., 2008a. EarlierPalaeolithic settlement patterns. Landscape archaeology on the riverDuero basin plateaus (Castilla y Le�on, Spain). J. World Prehistory 21,103e137.

Diez-Martín, F., S�anchez, P., G�omez, J.A., G�omez de la Rúa, D., 2008b. La ocupaci�onpaleolítica en los p�aramos del Duero. Nuevos datos procedentes deValdecampa~na (Olivares de Duero, Valladolid). Zephyrus 60, 19e39.

Diez-Martín, F., S�anchez Yustos, P., G�omez, J.A., G�omez de la Rúa, D., Yravedra, J.,Díaz, Mª I., 2011. La ocupaci�on neandertal en el Ca~n�on de la Horadada (Mave,Palencia). Nuevas perspectivas arqueol�ogicas en Cueva Coraz�on. Munibe 62,65e85.

Djema, H., 2008. Le Pal�eolithique moyen ancien de la corniche Cantabrique etdu bassin Aquitain ou le ph�enom�ene culturel des premiers n�eandertaliens.Anal- yses techno-�economiques comparatives de gisements cantabriques etp�eri- gourdins (Ph.D. thesis (unpublished)). Universit�e Panth�eon-Sorbonne,París.

d’Errico, F., Sanchez-Go~ni, M.F., 2003. Neanderthal extinction and the millenniascale climatic variability of OIS 3. Quat. Sci. Rev. 22, 769e788.

Ehlers, J., Gibbard, P.L., 2004. Extent and chronology of glacials. Quat. Sci. Rev. 22,1561e1568.

Ehlers, J., Gibbard, P.L., 2007. The extent and chronology of Cenozoic global glaci-ation. Quat. Int. 164e165, 6e20.

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e88 87

Eller, E., Hawks, J., Relethford, R., 2004. Local extinction size and recolonization,species effective population size, and modern human origins. Hum. Biol. 76 (5),689e709.

Fabre, V., Condemi, S., Degioanni, A., 2009. Genetic evidence of geographical groupsamong Neanderthals. Plos One 4, 5151.

Falgu�eres, C., Bahain, J.J., Yokoyama, Y., Arsuaga, J.L., Bermúdez de Castro, J.M.,Carbonell, E., Bischoff, J.L., Dolo, J.-M., 1999. Earliest humans in Europe: the ageof TD6 Gran Dolina, Atapuerca, Spain. J. Hum. Evol. 37 (3/4), 343e352.

Falgu�eres, C., Bahain, J.J., Yokoyama, Y., Bischoff, J.L., Arsuaga, J.L., Bermúdez deCastro, J.M., Carbonell, E., Dolo, J.M., 2001. Datation par RPE et U-Th des sitespl�eistoc�enes d’Atapuerca: sima de los Huesos, Trinchera Dolina et TrincheraGalería. Bilan g�eochronologique. L’Anthropologie 105 (1), 71e81.

F�eblot-Augustins, J., 1993. Mobility strategies in the late Middle Palaeolithic ofcentral Europe and western Europe: elements of stability and variability.J. Anthropol. Archaeol. 12, 211e265.

Fernandez Peris, J., Barciela Gonzalez, V., Blasco, R., Cuartero, F., Sa~nudo, P., 2008. ElPaleolítico Medio en el territorio valenciano y la variabilidad tecno-economicade la Cova del Bolomor. Treballs d'Arqueologia 14, 141e169.

Finlayson, C., Carri�on, J.S., 2007. Rapid ecological turnovers and its impact onNeanderthal and other human populations. Trends Ecol. Evol. 22 (4), 213e222.

Floss, H., 1994. Rohmaterialversorgung im Pal€aolithikum des Mittelrheingebietes.In: Monographien des R€omisch-Germanischen Zentralmuseums Mainz 21.Habelt, Bonn.

Foley, R.A., Lahr, M.M., 1997. Mode 3 technologies and the evolution of modernhumans. Camb. Archaeol. J. 7, 3e36.

Folke, C., 2006. Resilience: the emergence of a perspective for social-ecologicalsystems analyses. Glob. Environ. Change 16, 253e267.

Gamble, C., 1993. People on the move: interpretations of regional variation inPalaeolithic Europe. In: Chapman, J.C., Dolukhanov, P. (Eds.), Cultural Trans-formations and Interactions in Eastern Europe. Aveburg, Aldershot, pp. 37e55.

Gamble, C., 1999. Palaeolithic Societies of Europe. Cambridge University Press,Cambridge.

Gamble, C., Roebroeks, W., 1999. The Middle Palaeolithic: a point of inflexion. In:Roebroeks, W., Gamble, C. (Eds.), The Middle Palaeolithic Occupation of Europe.University of Leiden, Leiden, pp. 3e21.

Gamble, C.S., Davies, W., Pettitt, P.B., Richards, M., 2004. Climate change andevolving human diversity in Europe during the last glacial. Philosophical Trans.R. Soc. 359, 243e254.

Ganopolski, A., Rahmstorf, S., 2001. Rapid changes of glacial climate simulated in acoupled climate model. Nature 409, 153e158.

García-Medrano, P., Oll�e, A., Mosquera, M., C�aceres, I., Díez, C., Carbonell, E., 2014.The earliest Acheulean technology at Atapuerca (Burgos, Spain): Oldest levels ofthe Galería site (GII Unit). Quat. Int. 333 (5), 170e194.

Geneste, J.M., 1985. Analyse lithique d

industries mousterennes du P�erigord: uneapproche technologique du comportament des groupes humains auPal�eolithique moyen (Tesis doctoral). Universit�e de Bordeaux I, in�edita.

Geneste, J.M., 1989. Syst�emes d’approvisionnement en mati�eres premi�eres auPal�eolithique moyen et au Pal�eolithique sup�erieur en Aquitaine. In: Otte, M.(Ed.), L'Homme de N�eandertal, La Mutation, vol. 8. Universit�e de Li�ege, ERAUL,pp. 61e70.

Green, R.E., Stenzel, U., Kircher, M., Patterson, N., Li, H., Zhai, W., Hsi-Yang Fritz, M.,Hansen, N.F., Durand, E.Y., Malaspinas, A.-S., Jensen, J.D., Marques-Bonet, T.,Alkan, C., Prüfer, K., Meyer, M., Burbano, H.A., Good, J.M., Schultz, R., Aximu-Petri, A., Butthof, A., H€ober, B., H€offner, B., Siegemund, M., Weihmann, A.,Nusbaum, C., Lander, E.S., Russ, C., Novod, N., Affourtit, J., Egholm, M., Verna, C.,Rudan, P., Brajkovic, D., Kucan, Z., Gusic, I., Doronichev, V.B., Golovanova, L.V.,Lalueza-Fox, C., de la Rasilla, M., Fortea, J., Rosas, A., Schmitz, R.W.,Johnson, P.L.F., Eichler, E.E., Falush, D., Birney, E., Mullikin, J.C., Slatkin, M.,Nielsen, R., Kelso, J., Lachmann, M., Reich, D., P€a€abo, S., 2010. A Draft sequence ofthe Neandertal genome. Science 328, 710e722.

Hawks, J., 2009. Local Adaptation in Population Sinks. http://johnhawks.net/taxonomy/term/260.

Hewitt, G.M., 1999. Post-glacial re-colonization of European biota. Mol. Genet.Anim. Ecol. 68, 87e112.

Holling, C.S., 1986. Resilience of ecosystems; local surprise and global change. In:Clark, W.C., Munn, R.E. (Eds.), Sustainable Development of the Biosphere.Cambridge University Press, Cambridge (UK), pp. 292e317. Reprinted in:Cleveland C, Costanza R, Perrings C (Eds.). 1997. The Development of EcologicalEconomics. Brookfields: Edward Elgar Publishing.

Holling, C., 2001. Understanding the complexity of economic, ecological, and socialsystems. Ecosystems 4, 390e405.

Holling, C., Gunderson, L.H., 2002. Panarchy: Understanding Transformations inHuman and Natural Systems. D.C, Washington.

Housley, R., Gamble, C., Street, M., Petitt, P., 1997. Radiocarbon evidence for the Lateglacial human recolonisation of Northern Europe. Proc. Prehist. Soc. 63, 25e54.

Howell, F.C., 1999. Paleo-demes, species clades, and extinctions in the Pleistocenehominin record. J. Anthropol. Res. 55, 191e243.

Hublin, J.-J., 2009. The origin of Neanderthals. Proc. Natl. Acad. Sci. 106,16022e16027.

Hublin, J.J., Pa€abo, S., 2005. Neanderthals. Curr. Biol. 16, 113e114.Hulbin, J.J., Roebroeks, W., 2009. Ebb and flow or regional extinctions? on the

character of Neandertal occupation of northern environments. Comptes RendusPalevol 8, 503e509.

J€oris, O., Alvarez, E., Weninger, B., 2003. Radiocarbon evidence of the Middle toUpper Paleolithic transition in Southwestern Europe. Trab. Prehist. 60, 15e38.

Kandiano, E.S., Bauch, H.A., 2003. Surface ocean temperatures in the north-eastAtlantic during the last 500.000 years: evidence from foraminiferal censusdata. Terra Nova 15, 265e271.

Lazu�en, T., F�abregas, R., Lombera, A., Rodríguez, X.P., 2011. La gesti�on del utillaje depiedra tallada en el Paleolítico Medio de Galicia. El nivel 3 de Cova Eir�os (Tri-acastela, Lugo). Trab. Prehist. 68 (2), 237e258.

Maíllo, J.M., Cabrera, V., Bernaldo de Quir�os, F., 2004. Le d�ebitage lamellaire dans leMoust�erien final de Cantabrie (Espagne): le cas de El Castillo et Cueva Morin.L’anthropologie 108, 367e393.

M�arquez, B., Oll�e, A., Sala, R., Verg�es, J.M., 2001. Perspectives m�ethodologiquesde l’analyse fonctionnelle des ensembles lithiques du Pl�eistoc�ene inf�erieuret moyen d’Atapuerca (Burgos, Espagne). L’Anthropologie 105 (2),281e299.

Martín Benito, J.I., 2000. El Achelense en la cuenca media occidental del Duero.Instituto de Estudios Zamoranos Florían de Ocampo (CECEL-CSIC), Zamora.

MacDonald, K., Martinon-Torres, M., Dennell, R.W., Bermúdez de Castro, J.M., 2012.Discontinuity in the record for hominin occupation in south-western Europe:implications for occupation of the middle latitudes of Europe. Quat. Int. 271,84e97.

McManus, J.F., Oppo, D.W., Cullen, J.L., 1999. A 0.5-million-year record of millennial-scale climate variability in the North Atlantic. Science 283, 971e975.

Moncel, M.-H., 2006. Answer to paper of G. Monnier, an evaluation of the Lower/Middle Palaeolithic Periodization in Western Europe. Curr. Anthropol. 47,709e745.

Moncel, M.H., Moigne, A.M., Combier, J., 2012. Towards the Middle Palaeolithic inWestern Europe: the case of Orgnac 3 (Southeastern France). J. Hum. Evol. 63(5), 653e666.

Monnier, G., 2006. An evaluation of the Lower/Middle Palaeolithic periodization inWestern Europe. Currrent Anthropol. 47, 709e745.

Montes, R., 2003. El primer poblamiento de la Regi�on cant�abrica. El Paleolíticoinferior cant�abrico. Museo Nacional y Centro de Investigaci�on de Altamira, 18.

Mosquera, M., Oll�e, A., P�erez Gonz�alez, A., Rodríguez, X.P., Vaquero, M., Verg�es, J.M.,Carbonell, E., 2007. Valle de las Orquídeas: un yacimiento al aire libre delPleistoceno Superior en la Sierra de Atapuerca (Burgos). Trab. Prehist. 64 (2),143e155.

Moure, A., 1978. Mousterian occupation of the Cueva de La Ermita (Burgos, Spain).Curr. Anthropol. 19, 456e457.

Moure, A., García Soto, E., 1982. Dataci�on radiocarb�onica del Musteriense de CuevaMill�an (Hortigüela, Burgos). Boletín del Semin. Estud. Arte Arqueol. 48, 71e72.

Moure, A., García Soto, E., 1983a. Cueva Mill�an y La Ermita: dos yacimientos mus-terienses en el valle medio del Arlanza. Boletín del Semin. Estud. Arte Arqueol.49, 5e30.

Moure, A., García Soto, E., 1983b. Radiocarbon datting of the Mousterian in CuevaMill�an (Hortigüela, Burgos, Spain). Curr. Anthropol. 19, 155e157.

Moure, A., García Soto, E., 2000. �Economie et utilisation du territoire pendant leMoust�erien de la vall�ee moyenne de l’Arlanza (Burgos, Espagne). Anthropol.Pr�ehistoire 111, 186e189.

Moure, A., Delibes, G., Castanedo, I., Hoyos, M., Ca~naveras, J.C., Housley, R.A.,Iriarte, M.J., 1997. Revisi�on y nuevos datos sobre el musteriense de la cueva deLa Ermita (Hortigüela, Burgos). In: de Balbín, R., Bueno, P. (Eds.), Actas del IICongreso de Arqueología Peninsular, Paleolítico y Epipaleolítico. Fundaci�on ReiAlfonso Henriques, Zamora, pp. 67e83.

Müller, U.C., Pross, J., Tzedakis, P., Gamble, C., Kotthoff, U., Schmiedl, G., Wulf, S.,Christanis, K., 2012. The role of climate in the spread of modern humans intoEurope. Quat. Sci. Rev. 30, 273e279.

Navazo, M., Carbonell, E., 2014. Neanderthal settlement patterns during MIS 4e3 inSierra de Atapuerca (Burgos, Spain). Quat. Int. 331, 267e277.

Neira, A., Fuentes, N., Fern�andez, C., Bernaldo de Quir�os, F., 2006. Paleolítico su-perior y epipaleolítico en la provincia de Le�on. In: Delibes, G., Diez-Martín(Eds.), El Paleolítico superior en la Meseta norte espa~nola. Studia Archaeologica.Universidad de Valladolid, Valladolid, pp. 113e148.

Olle, A., Mosquera, M., Rodríguez, X.P., de Lombera-Hermida, A., García-Anton, M.D.,García-Medrano, P., Luna, P., Menendez, L., Navazo, M., Terradillos, M.,Bargallo, A., Marquez, B., Sala, R., Carbonell, E., 2013. The Early and MiddlePleistocene technological record from Sierra de Atapuerca (Burgos, Spain). Quat.Int. 295, 138e167.

Pa€abo, S. (2009). http://apps.nc3.biz/kunden/mpg/2009/02/12/index.php?status1/4;4prechoosenplayer1/4;4wmtcountme1/4;4true.

Panera, J., Torres, T., Perez, A., Ortiz, J.E., Rubio, S., Uribelarrea, D., 2011. Geo-cronología de la Terraza Compleja de Arganda en el valle del río Jarama(Madrid, Espa~na). Estud. Geol. 67 (2), 495e504.

Peris, J.F., 2007. La Cova del Bolomor (Tavernes de la Valldigna, Valencia). Lasindustrias líticas del Pleistoceno medio en el �ambito del mediterr�aneo penin-sular. In: Serie de Trabajos Varios, vol. 108. Servicio de investigaci�on pre-hist�orica, Diputaci�on provincial de Valencia.

Picin, A., Peresani, M., Falgu�eres, C., Gruppioni, G., Bahain, J.J., 2013. San Bernardinocave (Italy) and the appearance of levallois technology in Europe: results of aradiometric and technological reassessment. Plos One 8 (10).

Pigeot, N., 1991. Reflexions sur �lhistoire technique de �lhomme: de evolutioncognitive a �levolution culturelle. Pal�eo 3, 167e200.

Pulliam, H.R., 1988. Sources, sinks, and population regulation. Am. Nat. 132,652e661.

Quam, R., Arsuaga, J.L., Bermúdez de Castro, J.M., Díez, C., Lorenzo, C., Carretero, J.M.,García, N., Ortega, A.I., 2001. Human remains from Valdegoba Cave (Hu�ermeces,Burgos, Spain). J. Hum. Evol. 41, 385e435.

P. S�anchez Yustos, F. Diez Martín / Quaternary Science Reviews 121 (2015) 75e8888

Redman, C.L., Kinzig, A.P., 2003. Resilience of past landscapes: resilience theory,society, and the longue dur�ee. Conserv. Ecol. Jg 7. http://www.consecol.org/vol7/iss1/art14.

Ríos, J., Garate, D., Gomez Olivencia, A., Iriarte Aviles, E., Aramburu, A., Arcedillo, D.,Garcia, A., Iriarte Chiapusso, M.J., Moreno, J., Murelaga, X., Ortiz, J.E., Torres, T.,San Pedro, Z., Zapata, L., 2011. The Lower to Middle Palaeolithic transition inNorthern Iberia: new data from Arlanpe cave. Antiquity 85.

Rodríguez, X.P., 2004. Atapuerca y el inicio del Paleolítico medio en Europa. In:Baquedano, E., Rubio Jara, S. (Eds.), Miscel�anea en homenaje a Emiliano Aguirre.Arqueología, vol. IV. Museo Arqueol�ogico Regional, Alcal�a de Henares,pp. 416e431 (Zona Arqueol�ogica, 4).

Rodríguez, J., 2006. Structural continuity and multiple alternative stable States inMiddle Pleistocene European mammalian communities. Palaeogeogr. Palae-oclimatol. Palaeoecol. 239, 355e373.

Rodríguez-Asensio, J.A., Arrizabalaga, A., 2004. El poblamiento m�as antiguo de laregi�on: las ocupaciones previas al IS4. Desde el inicio del poblamiento a circa80.000 AP. Kobie 8, 51e90.

Rodríguez, J., Burjachs, F., Cuenca-Besc�os, G., García, N., Van der Made, J., P�erezGonz�alez, A., Blain, H.-A., Exp�osito, I., L�opez-García, J.M., García Ant�on, M.,Allu�e, E., C�aceres, I., Huguet, R., Mosquera, M., Oll�e, A., Rosell, J., Par�es, J.M.,Rodríguez, X.P., Díez, C., Rofes, J., Sala, R., Saladi�e, P., Vallverdú, J., Ben-nasar, M.L., Blasco, R., Bermúdez de Castro, J.M., Carbonell, E., 2011. One millionyears of cultural evolution in a stable environment at Atapuerca (Burgos, Spain).Quat. Sci. Rev. 30 (11e12), 1396e1412.

Rodríguez-Asensio, J.A., 1993. La presencia humana m�as antigua en Asturias.Estudios de Arqueología Asturiana N� 2, Oviedo.

Rodríguez-Asensio, J.A., 2000. El Paleolítico antiguo en Asturias. SPAL 9, 109e123.Rodríguez-Asensio, J.A., 2001. Yacimiento de Cabo Busto. Los orígenes prehist�oricos

de Asturias. GEA.Roebroeks, W., 2008. Time for the Middle to Upper Paleolithic transition in Europe.

J. Hum. Evol. 55, 918e926.Roebroeks, W., Tuffreau, A., 1999. Paleoenvironment and settlement patterns of the

Northwest European Middle Palaeolithic. In: Roebroeks, W., Gamble, C. (Eds.),The Middle Palaeolithic Occupation of Europe. Leiden University Press, Leiden,pp. 121e138.

Roebroeks, W., Villa, P., 2011. On the earliest evidence for habitual use of fire inEurope. Proc. Natl. Acad. Sci. 108, 5209e5214.

Roebroeks, W., Sier, M.J., Nielsen, T.K., De Loecker, D., Par�es, J.M., et al., 2012. Use ofred ochre by early Neandertals. Proc. Natl. Acad. Sci. 109, 1889e1894.

Rolland, N., 1995. Levallois technique emergence: single or multiple? A review ofthe Euro-African record. In: Dibble, H., Bar-Yosef, O. (Eds.), The Definition andInterpretation of Levallois Technology, Monographs in World Archaeology, vol.23. Prehistory Press, pp. 333e359.

Rosell, J., Blasco, R., Campeny, G., Díez, J.C., Alonso, R., Men�endez, L., Arsuaga, J.L.,Bermúdez, J.M., Carbonell, E., 2011. Bone as a technological raw material at theGran Dolina site (Sierra de Atapuerca, Burgos, Spain). J. Hum. Evol. 61 (1),125e131.

Rots, V., 2013. Insights into early Middle Palaeolithic tool use and hafting inWestern Europe. The functional analysis of level IIa of the early Middle Palae-olithic site of Biache-Saint-Vaast (France). J. Archaeol. Sci. 40, 497e506.

Roucoux, H., Tzedakis, P.C., Lawson, I.T., Margarin, V., 2011. Vegetation history of thepenultimate glacial period (Marine Isotope Stage 6) at Ioannina, north-westGreece. J. Quat. Sci. 26 (6), 616e626.

S�anchez Yustos, P., 2009. El Paleolítico Antiguo en la Cuenca del Duero. Instru-mentos Te�oricos para el Desarrollo de un Modelo Interpretativo de ArqueologíaEcon�omica (Unpublished PhD dissertation). Universidad de Valladolid,Valladolid.

S�anchez Yustos, P., 2012. Crítica a la cultura f�osil. La estructura econ�omica comounidad de an�alisis del cambio cultural paleolítico. Complutum 23, 27e40.

S�anchez Yustos, P., Diez-Martín, F., 2006-07. Historia de las investigaciones paleo-líticas en la provincia de Valladolid. El caso Mucientes. Boletín del Semin. Estud.Arqueol. LXII-LXXIII, 7e38.

S�anchez Yustos, P., Diez-Martín, F., 2010. El Paleolítico Antiguo en la Cuenca delDuero: la Unidad Ecol�ogica de P�aramos. Zephyrus LXVI, 39e56.

S�anchez Yustos, P., Diez-Martín, F., Díaz, I.M.ª, G�omez, D., G�omez, J.A., 2011. Estra-tegias de talla en Cueva Coraz�on (Mave, Palencia). Un yacimiento del Muster-iense antiguo en las estribaciones meridionales de la Cordillera Cant�abrica.Trab. Prehist. 68 (1), 51e63.

Santonja, M., P�erez-Gonz�alez, A., 1984. Las industrias paleolíticas de La Maya I en su�ambito regional. Ministerio de Cultura: Direcci�on General de Bellas Artes yArchivos. Subdirecci�on General de Arqueología y Etnografía, Valladolid.

Santonja, M., P�erez-Gonz�alez, A., 2002. El Paleolítico inferior en el interior de laPenínsula Ib�erica. Un punto de vista desde la Geoarqueología. Zephyrus 53/54,27e77.

Santonja, M., P�erez-Gonz�alez, A., 2005. Geoarqueología del yacimiento Achelensede El Basalito (Castraz de Yeltes, Salamanca). Discusi�on acerca de su naturalezay significado. In: Baquedano, E., Rubio, S. (Eds.), Miscelanea en Homenaje aEmiliano Aguirre. Miscelanea en Homenaje a Emiliano Aguirre. ZonaArqueol�ogica, Madrid, pp. 472e483.

Santonja, M., P�erez-Gonz�alez, A., 2006. La industria lítica del miembro estratigr�aficomedio de Ambrona (Soria, Espa~na) en el contexto del Paleolítico antiguo de laPenínsula Ib�erica. Zephyrus 59, 7e20.

Santonja, M., Perez-Gonzalez, A., 2010. Mid-pleistocene Acheulian industrial com-plex in the Iberian Peninsula. Quat. Int. 223/224, 154e161.

Santonja, M., Perez, A., Domínguez, M., Panera, J., Rubio, S., Sese, C., Soto, E.,James, J., Duval, M., Demuro, M., Ortiz, J., de Torres, T., Mercier, N., Barba, R.,Yravedra, J., 2014. The Middle Paleolithic site of Cuesta de la Bajada (Teruel,Spain): a perspective on the Acheulean and Middle Paleolithic tech-nocomplexes in Europe. J. Archaeol. Sci. 49, 556e571.

Schmidt, I., Bradtm€oller, M., Kehl, M., Pastoors, A., Tafelmaier, Y., Weninger, B.,Weniger, G.-C., 2012. Rapid climate change and variability of settlement pat-terns in Iberia during the Late Pleistocene. Quat. Int. 274, 179e204.

Scott, B., Ashton, N., 2011. The Early Middle Palaeolithic: the European context. In:Ashton, N., Lewis, S.G., Stringer, C. (Eds.), The Ancient Hominin Occupation ofBritain. Developments in Quaternary Science, pp. 91e112.

Shea, J., 2009. Transitions or turnovers? Climatically-forced extinctions of Homosapiens and Neanderthals in the East Mediterranean Levant. Quat. Sci. Rev. 27,2253e2270.

Stringer, C., P€alike, H., van Andel, T.H., Huntley, B., Valdes, P., Allen, J.R.M., 2003.Climatic stress and the extinction of the Neanderthals. In: van Andel, T.H.,Davies, W. (Eds.), Neanderthals and Modern Humans in the Euro-pean Land-scape during the Last Glaciations. Cambridge University Press, Cambridge,pp. 233e240.

Tapia, J., 2010. El dep�osito paleolítico de La Garma A: estratigrafía y aproximaci�ontafon�omica. Trabajo de Investigaci�on de 3er ciclo. Departamento de CienciasHist�oricas, Universidad de Cantabria (unpublished).

Tuffreau, A., 1997. The variability of Levallois technology in northern France andneighbouring areas. In: Dibble, H., Bar Yosef, H.L. (Eds.), The Definition andInterpretation of Levallois Technology. Prehistory Press, Madison, pp. 413e427.

Turq, A., Roebroeks, W., Bourguignon, L., Faivre, J.-P., 2013. The fragmented char-acter of Middle Palaeolithic stone tool technology. J. Human Evol. 65, 641e655.

Tzedakis, P.C., Hughen, K.A., Cacho, I., Harvati, K., 2007. Placing late Neanderthals ina climatic context. Nature 449, 206e208.

Verpoorte, A., 2009. The first modern humans in Europe? A closer look at the datingevidence from the Swabian Jura (Germany). Antiquity 79, 269e279.

Villa, P., 2009. Discussion 3: the Lower to Middle Paleolithic transition. In:Camps, M., Chauhan, P.R. (Eds.), Sourcebook of Paleolithic Transitions. Methods,Theories, and Interpretations. Springer, New York, pp. 265e270.

Walker, B.H., Gunderson, L.H., Kinzig, A.P., Folke, C., Carpenter, S.R., Schultz, L., 2006.A handful of heuristics and some propositions for understanding resilience insocial-ecological systems. Ecol. Soc. 11 (1), 13. http://www.ecologyandsociety.org/vol11/iss1/art13/.

White, M., Ashton, N., 2003. Lower Palaeolithic core technology and the origins ofthe Levallois method in North-Western Europe. Curr. Anthropol. 44, 598e609.

White, M., Ashton, N., Scott, B., 2011. The emergence, diversity and significance ofmode 3 (prepared core) technologies. In: Ashton, N., Lewis, S.G., Stringer, C.(Eds.), The Ancient Hominin Occupation of Britain. Developments in QuaternaryScience, pp. 53e67.

Widlok, T., Aufgebauer, A., Bradtm€oller, M., Dikau, R., Hoffmann, T., Kretschmer, I.,Panagiotopoulos, K., Pastoors, P., Peters, R., Sch€abitz, F., Schlummer, M.,Solich, M., Wagner, B., Weniger, G., Zimmermann, A., 2012. Towards a theo-retical framework for analyzing integrated socio-environmental systems. Quat.Int. 274, 259e272.

Wymer, J.J., Gladelter, B.G., Singer, R., 1993. The industries at Hoxne and the LowerPaleolithic of Britain. In: Singer, R., Gladfelter, B.G., Wymer, J.J. (Eds.), The LowerPalaeolithic Site at Hoxne, England. University of Chicago Press, Chicago,pp. 218e224.

Wynn, T., Coolidge, F.L., 2004. The expert Neanderthal mind. J. Hum. Evol. 46,467e487.

Yravedra, J., Diez-Martín, F., S�anchez Yustos, P., G�omez, D., Díaz, I.Mª, G�omez, J.A.,2013. Estrategias de subsistencia durante el Paleolítico Medio en la SubmesetaNorte. La Cueva Coraz�on (Mave, Palencia). Zephyrus LXXI, 45e60.