Settlement patterns during the Magdalenian in the south-eastern Pyrenees, Iberian Peninsula. A territorial study based on GIS

11 pages
19 views
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
ABSTRACT: The territorial characteristics and environmental factors involved in the selection of a specific site for establishing a settlement are key features in the analysis of hunter-gatherers' knowledge of (and dominance over) their
Transcript
  Contents lists available at ScienceDirect Journal of Archaeological Science: Reports  journal homepage: www.elsevier.com/locate/jasrep Settlement patterns during the Magdalenian in the south-eastern Pyrenees,Iberian Peninsula. A territorial study based on GIS Bàrbara Mas a,b, ⁎ , Ethel Allué a,b , Marta Sánchez de la Torre c,d , Óscar Parque a,b,d ,José Miguel Tejero d,e , Xavier Mangado d , Josep Maria Fullola d a  Àrea de Prehistòria, Universitat Rovira i Virgili (URV), Avinguda de Catalunya 35, 43002 Tarragona, Spain b  IPHES, Institut Català de Paleoecologia Humana i Evolució Social, Zona Educacional 4, Campus Sescelades URV (Edi  fi ci W3), 43007 Tarragona, Spain c  PPVE, Universidad de Zaragoza, 12 Pedro Cerbuna, 50009 Zaragoza, Spain d  SERP (Seminari d'Estudis i Recerques Prehistòriques), Universitat de Barcelona, Montalegre 6-8, E-08001 Barcelona, Spain e CNRS, UMR 7041, ArScAn Équipe Ethnologie Préhistorique, 92023 Nanterre, France A R T I C L E I N F O  Keywords: GISSettlement patternsSite locationMacrospatial analysisLate PalaeolithicSouth-eastern Pyrenees A B S T R A C T The territorial characteristics and environmental factors involved in the selection of a speci fi c site for estab-lishing a settlement are key features in the analysis of hunter-gatherers' knowledge of (and dominance over)their surroundings and in the attempts to understand the survival strategies that they deployed. This paperpresents a macrospatial analysis using GIS tools, which provides an objective comparison of territorial variablesat several Magdalenian archaeological sites located in the south-eastern Pyrenees (NE Iberia). To establish thesettlement patterns, we analyse territorial values: orientation, elevation, slope, sites aspect (caves, rock sheltersor open air) and distance from rivers. With the data obtained, we create solar radiation models, construct groupsof sites according to visibility, and calculate the displacement costs of mobility. The results suggest a series of di ff  erent settlement patterns during the Magdalenian. The visibility of rivers from the archaeological sites andpotential sunlight are characteristic features throughout the period, but the distance between rivers and set-tlements decreases diachronically. Comparison of the climate models indicates that settlements in the vicinity of the river were more frequent at times with evidence of low rainfall. Likewise, the costs of displacement from thesurrounding territory to the archaeological sites increase; access to the Lower Magdalenian sites is easier, andaccess to the Upper Magdalenian sites much more di ffi cult. 1. Introduction The Magdalenian is one of the most documented and well-knownPalaeolithic chrono-cultural phases in western Europe, characterized bythe diversity of symbolic art, fauna and lithic remains in the archae-ological record (Fullola et al., 2012; Vega et al., 2013). These records have allowed the study of changes in strategies for procuring biotic andabiotic resources and in social and symbolic behaviour, and re fl ect thevariability and development of technical systems of lithic knapping andproduction of bone tools (Benito-Calvo et al., 2009; Calvo et al., 2009, 2008; Esteve, 2009; Fullola, 2001; Fullola et al., 2006; García-Diez and Vaquero, 2015; Langlais, 2010; Maluquer de Motes, 1983 – 1984;Mangado et al., 2014, 2013, 2010, 2009; Martínez-Moreno et al., 2007; Mithen, 1988; Montes, 2005; Mora et al., 2011; Morales and Verges, 2014; Peña and Cruz, 2014; Roman, 2016; Sánchez de la Torre, 2015; Sánchez de la Torre and Mangado, 2013; Tejero, 2009; Tejero et al., 2013; Utrilla et al., 2013, 2010; Utrilla and Mazo, 2014; Utrilla and Montes, 2009). These studies have used a variety of approaches, butspatial analyses have not been applied to date in the Magdalenian sitesof north-eastern Iberia. The spatial analysis of prehistoric sites providesinteresting insights into the study of settlement dynamics and mobilitystrategies. Moreover, the conservation and excavation of the Magdale-nian sites in north-eastern Iberia has allowed paleoenvironmental re-constructions of the southern Pyrenees at the end of the Last Glacialperiod, coinciding with the deglaciation of the Pyrenees around 20kyrBP (Alcolea, 2017; Allué, 2009; Allué et al., 2012a, 2012b, 2018; Aura et al., 2011; Bergadà, 1991; Burjachs, 2009; Fullola et al., 2012; Fumanal and Ferrer, 2014; Soler et al., 2009). During the Magdalenian, the climatic sequence underwent a transition between the colder epi-sode GS-2a and the GI-1 interstadial. The interstadial GI-1 began with a https://doi.org/10.1016/j.jasrep.2018.10.002Received 11 June 2018; Received in revised form 1 October 2018; Accepted 2 October 2018 ⁎ Corresponding author at: Àrea de Prehistòria, Universitat Rovira i Virgili (URV), Avinguda de Catalunya 35, 43002 Tarragona, Spain.  E-mail addresses:  barbara.mas86@gmail.com (B. Mas), eallue@iphes.cat (E. Allué), sanchezdelatorre@unizar.es (M. Sánchez de la Torre),  jose-miguel.tejero@mae.cnrs.fr (J.M. Tejero), mangado@ub.edu (X. Mangado), fullola@ub.edu (J.M. Fullola). Journal of Archaeological Science: Reports 22 (2018) 237–2472352-409X/ © 2018 Published by Elsevier Ltd.    rapid increase in temperatures, similar to current temperatures. Fa-vourable temperatures were interrupted by two colder oscillations, thesub-stadials GI-1d and GI-1b (sensu Walker et al., 1999). The warmpulsation after GI-1a oscillation gradually decreased until the beginningof stadial GS-1, with very cold temperatures in a large part of Europeuntil the beginning of the Holocene.Models of the di ff  erences in the forms of territorial occupation andexploitation can help to de fi ne the adaptation and survival strategiesused by palaeolithic human groups (Binford, 1980, 1982; Brück and Goodman, 1999; Cleland, 1966; Gamble, 1986; Kelly, 1983). Inter- preting the possible interactions between these groups and their im-mediate environment is essential to an explanation of the in fl uence of the di ff  erent strategies used for land management and natural resourcecatchment (Allen et al., 1990; Butzer, 1989; Chatters, 1987; Clark, 1972; Clarke, 1977; Eriksen, 1997). The choice of the site for a settle- ment depends on the season in which it is to be occupied, the pattern of territorial mobility and the procedures involved in the decisions madeby human groups (Binford, 1980; Clark, 1972; Kelly, 1983; Saaty, 1972: 1061). A variety of factors of the landscape must be taken into con-sideration when deciding on its exploitation, for example its topo-graphy (location, riverbeds, vegetation), resource availability (proxi-mity and accessibility), and conditions of liveability (dimensions,visibility and thermal comfort).Quantitative and qualitative information on the territorial and en-vironmental characteristics of archaeological sites allows the descrip-tion of settlement patterns and mobility strategies of hunter-gatherergroups (Grove, 2009; Jochim, 1976). Using Geographical Information Systems (GIS) it is possible to obtain, quantify, and analyse the terri-torial factors that are observed qualitatively in the immediate en-vironment of an archaeological site (Allen et al., 1990; Conolly and Lake, 2006; Eriksen, 1997; Hodder and Orton, 1990). Site location preferences play a role in social organization. There area range of landscape factors, such as proximity to speci fi cs resources(like rivers or forests), terrain conditions (like slope or accessibility) orhabitability conditions that could a ff  ect these preferences. Subsistencein the southern Pyrenees during the Magdalenian is not determinedexclusively by the climatic conditions, cultural behaviour or by tech-nological adaptations (Fullola et al., 2012). There are certain territorialvariables that de fi ne di ff  erent landscapes, like latitude, altitude andtopography. In many cases, these preferences have been intuitivelyrelated, but they have been seldom analysed in a systematic way. Theobjective of this study is to de fi ne the settlement patterns and territorialcharacteristics in the Magdalenian sites located in the southern Pyr-enees. Studies of the spatial characteristics around the Magdaleniansites have never been analysed jointly in the south-eastern Pyrenees.However, the settlement patterns of Magdalenian sites in the adjoiningarea of Cantabrian region have been carried out (see García, 2013;García-Moreno, 2013, 2015). In order to evaluate how the changes insettlement patterns could have been related the landscape conditions,the territorial characteristics are analysed. Using GIS, sites orientation,solar radiation, visibility and energy mobility cost from a set of Mag-dalenian sites are analysed and evolution in preferences are de fi ned.The spatial analysis performed should broaden our understanding of thechanges in the territorial management strategies and site location pre-ference by human groups, during the Lower Magdalenian to the UpperMagdalenian in south-eastern Pyrenees. 2. Material and methods The area of study comprises the modern-day regions of Aragon andCatalonia, north-eastern Iberian Peninsula, on the southern slope of thePre-Pyrenees. This slope is situated between the structure or apex of theSegre river, and between the Mesozoic and Cenozoic outcrops thatconstitute the eastern boundary from the Mediterranean Sea and, to thewest, the Cinca river valley and his tributaries (Vera, 2004) (Fig. 1). Table 1 presents the reoccupations and the presence/absence of humans at the sites analysed, according to the di ff  erent chrono-culturalphases of the Magdalenian (Fullola, 2001; Mangado et al., 2010, 2014; Montes, 2005; Mora et al., 2011; Tejero et al., 2013; Utrilla and Mazo, 2014; Utrilla et al., 2010, 2012). For the analysis of the settlement patterns, three chrono-cultural groups have been established accordingto the spatial characteristics of sites with evidence of occupation, alsoincluding archaeological sites with more than one occupation phase:the Lower Magdalenian group, comprising Montlleó, Cova Alonsé,Forcas I and Cova Gran; the Middle Magdalenian group, comprisingCova del Parco, Fuente del Trucho and Cova Gran; and the UpperMagdalenian group, including Cova del Parco, Cueva de Chaves, BoraGran d'en Carreras, Forcas I, Legunova, Peña 14 and Cova Gran.  2.1. Radiocarbon dating  Human presence or absence was established based on radiocarbondating reported in other studies, from di ff  erent Magdalenian archae-ological sites (Table 2). We perform the calibration because we believeit is necessary to join the set of Magdalenian sites dates, currentlyisolated, and put them in a territorial context.The calibration was carried out using the online tool OxCal v.4.3.2(Ramsey, 2017), with a deviation range lower than 101. We used In-tCal-13 curve (Reimer et al., 2013) and NGRIP ice core (Rasmussen et al., 2014).  2.2. GIS analysis GIS was used to generate a hypothetical reality model for a speci fi cterritory. This allows the interpretation of the relationship betweenhuman behaviour and the environment by selecting precise territorialvariables. The territorial variables are discussed below.  2.2.1. Vector and raster data: DEM and rivers layers Data were spatially analysed and referenced from a GIS projectthrough free software program QGIS, and its application GRASS version10.12.3. Digital cartographic base maps were generated from the to-pographic maps in raster (Digital Elevation Models), vector and meta-data layers (rivers) downloaded for the provinces of Lleida, Huesca,Zaragoza and Girona, obtained from the  Institut Cartogrà  fi c de Catalunya (ICC) and the  Instituto Geográ  fi co Nacional  (IGN). To undertake themacrospace analysis, we downloaded the Digital Elevation Models(DEM) (mdt25) and rivers layers of the municipalities where theMagdalenian sites are located …  The rivers vectors have been necessaryfor the distances measurements, between the exact location points of the archaeological sites up to the nearest river course.  2.2.2. Analysis of settlement patterns The term  “ spatial model ”  refers to the use of spatial data to simulatea process, to understand a complex relationship, to predict a result, orto analyse a problem (Conolly and Lake, 2006: 73). Spatial models areapplied to establish the relationship between the location of archae-ological sites and territorial variables such as slope and orientation,classi fi cation of archaeological sites according to site type (i.e., cave,open air and rock shelter), visible area and distance from settlements torivers.  2.2.2.1. The orientation of archaeological sites and site types: solar radiation analysis . Sunlight and sun heating are usually considereddesirable factors for good habitability conditions (Fano, 1998; García- Moreno, 2015). In our study, we include the Palaeolithic habitat in thecaves, since certain domestic activities were carried out outside thecave or near the entrance (Mangado et al., 2014). The QGIS-GRASS'ssolar radiation analysis computes direct, di ff  use and re fl ected solarradiation raster maps for a given day, the latitude, the surface and theatmospheric conditions. The latitude used in the calculation of thepotential insolation was 42° 30 ′ , since it constitutes the approximate  B. Mas et al.  Journal of Archaeological Science: Reports 22 (2018) 237–247  238  average latitude of the entire study area, and in any case a variation of several minutes of degree in the latitude used did not modify thecalculation of potential insolation (sensu Hesse, 2013). Also, theposition of the Earth with respect to the Sun is di ff  erent from thecurrent one with respect to the existing at the end of the Pleistocene.This is due to variations in the obliquity of the ecliptic, that is, the anglebetween the axis of Earth rotation with respect to the plane of theecliptic. This angle varies with a periodicity of about 41.000yearsbetween 21.5° and 24.5°, being currently 23.5° (Uriarte-Cantolla,2000). In order to carry out this calculation, it is not necessary totake into account the length of the point on which the insolation is to becalculated, since the solar incidence will be similar along the parallel.In our case, solar parameters are calculated in every cell in theraster map (e.g., sunrise, sunset times, declination, extraterrestrial ir-radiance, daylight length), and are saved in the map history  fi le. We Fig. 1.  Map of north-eastern Iberian Peninsula and location of Magdalenian sites analysed in this work. 1: Cova del Parco; 2: Montlleó; 3: Cova Gran; 4: Legunova; 5:Abrigo Forcas I; 6: Cova Alonsé; 7: Fuente del Trucho; 8: Cueva de Chaves; 9: Bora Gran d'en Carreras; 10: Peña 14. Table 1 Reoccupations and presence/absence of the sites analysed during theMagdalenian (LM: Lower Magdalenian; MM: Middle Magdalenian; UM: UpperMagdalenian). Site Site type LM MM UMCova del Parco Rock shelter  –  X XMontlleó Open air X  – – Cueva de Chaves Cave  – –  XBora Gran d'en Carreras Rock shelter  – –  XCova Alonsé Rock shelter X  – – Forcas I Rock shelter X  –  XLegunova Rock shelter  – –  XPeña 14 Rock shelter  – –  XFuente del Trucho Cave  –  X  – Cova Gran Rock shelter X X X Table 2 Compilation of published radiocarbon dating, ordered by chrono-cultural periods (LM: Lower Magdalenian; MM: Middle Magdalenian; UM: Upper Magdalenian). Site Level Period Ref. lab. Radiocarbon dating Deviation± Method Sample PublicationCova Gran EA3 LM Beta-233606 16,800 80 AMS Charcoal Mora et al., 2011Montlleó 1S·B LM OxA-9017 15,440 80 AMS Molar of equine Mangado et al., 2011: 29Cova Gran 6P LM Beta-265984 15,120 70 AMS Charcoal Mora et al., 2011Cueva Alonsé m LM GrA-21536 15,069 90 AMS Charcoal Montes, 2005Cueva Alonsé m LM GrA-21537 15,069 90 AMS Charcoal Montes, 2005Cova Gran 4P LM Beta-259273 14,760 70 AMS Charcoal Mora et al., 2011Forcas I 15 LM GrA-25979 14,440 70 AMS Bone Utrilla and Mazo, 2014Cova Gran SH4 MM Beta-187424 13,660 50 AMS Charcoal Mora et al., 2011Cova del Parco MM OxA-23650 13,475 50 AMS Charcoal Mangado et al., 2014Cova del Parco MM OxA-29336 13,255 50 AMS Charcoal Mangado et al., 2014Cova del Parco II UM OxA-17730 13,095 55 AMS Charcoal Mangado et al., 2010Bora Gran d'en Carreres ? UM OxBGA-2153 13,080 90 AMS Bone Fullola, 2001Cova del Parco II UM OxA-13596 13,025 50 AMS Charcoal Mangado et al., 2010Cueva de Chaves 2b UM GrN-15635 12,950 70 Conv Bone Utrilla, 1995Cova del Parco II UM OxA-13597 12,995 50 AMS Charcoal Mangado et al., 2010Bora Gran d'en Carreres ? UM PxBGA-2222 12,830 80 AMS Bone Fullola, 2001Cueva de Chaves 2b UM GrN-14561 12,660 70 Conv Bone Utrilla et al., 2010Forcas I 14 UM GrA-33986 12,600 60 AMS Bone Utrilla and Mazo, 2014Legunova q UM GrA-22089 12,500 90 AMS Charcoal Utrilla et al., 2010Cova del Parco II UM OxA-10797 12,460 60 AMS Charcoal Mangado et al., 2010Forcas I 13d UM GrA-32957 12,440 50 AMS Bone Utrilla and Mazo, 2014Legunova q UM GrA-24296 12,060 60 AMS  –  Tejero et al., 2013  B. Mas et al.  Journal of Archaeological Science: Reports 22 (2018) 237–247  239  followed the methodology proposed by García (2013). The calculationinvolved the preparation of twelve models for each archaeological site,equivalent to one day for twelve months per year. The hours of sunlightwere also calculated, using three mapping inputs: A) elevation, B) slopeand C) orientation (Fig. 2).The hours of sunlight received were extracted from the values of theraster map output cells, in the geographical coordinates where eacharchaeological site is located. The optimum mean hours were extractedfor an area of 3km 2 . Furthermore, the sunlight hours lost or not usedwere calculated by subtracting the optimal value from the obtainedvalue.  2.2.2.2. From mean slope: viewshed analysis and mobility energy cost  . Viewshed analysis is based on the idea of the visual basin thatan observer can perceive from a certain place. In south-easternPyrenees, visual characteristics of sites have merely been mentionedin ambiguous way, such as  “ great visibility ”  or  “ strategic location ” .Likewise, usually related to the visual control of the territory or speci fi cresources, like the control of routes of large herbivores (e.g. Mangadoet al., 2013). However, visual characteristics of sites have never beenquanti fi ed in systematic analysis. In this study we also considered thatthe topography of the land near the sites would condition the variablesfor the choice as well as the visibility, in order to control thesurrounding area from the settlement. The visibility analysis wascarried out with the free software QGIS-GRASS, establishing a seriesof parameters that have helped to create a greater proximity whenmapping the present-day model of visibility (García-Moreno, 2013;Wheatley and Gillings, 2002). The variables were established inrelation to the height of the observer (1.70m) and the maximumvisibility range (10km). This process calculated the hectares observedfrom the archaeological sites, as well as their mapping.For the slope analysis, a categorical classi fi cation according to thedegree of inclination of the slopes was assigned, delimiting a radius of 3km from the location of the Magdalenian sites. The following se-paration of the groups by degrees of inclination was proposed by García(2013): Flat: 0 – 5°; Moderate: 5 – 15°; Abrupt: 15 – 30°; Steep:> 30°.Accessibility to the archaeological sites (the mobility energy value)was analysed applying the principle of William W. Naismith(1856 – 1935), according to which an adult in good physical conditionwalks, on  fl at ground, 10km in 2h, which is 1km every 12min. Weemphasize that walking on foot is a physical activity, and so the in-dividual physiological limitations of each person should not be under-estimated. As the study area contains uneven reliefs and varied slopes, avariable regarding the e ff  ort depending on the degree of inclination of the slopes was added. Slope is a key factor in recognizing accessibilityto a territory, in addition to conditioning both the walker's e ff  ort andspeed. For the analysis, we have reclassi fi ed a cost surface based on aninput dataset and a cost function, where cost is measured in time. Thecumulative pixel values of the cells of the DEM maps containing in-formation on the degrees of the slopes were reclassi fi ed with thegeoanalytical tool  Reclass , by GIS, integrating a reclassi fi es layer withranges of colours from Naismith's time scale (12min), and theE ff  ort+1 (friction). Subsequently, cost of movement was classi fi edwith a range of colours in the cell map (Fig. 3).According to Optimal Foraging Theory, that presumes in certainarenas human decisions are made to maximize net rate of energy gain(Clark and Mangel, 1986; Charnov, 1976; MacArthur and Pianka, 1966), regarding the settlement patterns and mobility of hunter-gath-erer human groups (Binford, 1977, 1980, 1982), and considering theidea of Site Catchment Analysis method (Bailey and Davidson, 1983;Vita-Finzi and Higgs, 1970), we assume three di ff  erent types of catch-ment areas around settlements: 1) the immediate area, around 3km 2 , inwhich resources are quickly absorbed (wood, plants and small prey); 2)the area of harvesting and smaller hunting, between 3 and 7.5km 2 ; 3)the largest hunting area, related to the maximum distance that can becovered in solar hours per day (between 20 and 40km 2 ). The GIS cal-culation model bases their analysis on the calculation of cumulativevalues of the di ff  erent cells or pixels in raster layer. The cell values arereclassi fi ed from 0 to 9, representing the di ffi culty or cost of displace-ment in a certain territory and distinguishing the gradients of the slopesaccording to their steepness. Areas of about 3km 2 around the archae-ological sites were analysed and the values extracted for each rastercell. Fig. 2.  Example of raster maps used to calculatethe solar irradiation analysis at the sites CovaGran and Cova del Parco. A: srcinal ElevationMap (DEM) mdt25-0327-H31-LIDAR; B: map of slopes created in degrees from A; C: slope or-ientation map in degrees, created from A+B; D:map of solar irradiation emulating May 15th,created from A+B+C.  B. Mas et al.  Journal of Archaeological Science: Reports 22 (2018) 237–247  240  3. Results 3.1. Radiocarbon data In the  fi rst period in which there are radiocarbon records attribu-table to anthropic activity during the Lower Magdalenian, the datingranges between 20,000 and 18,900cal. BP (Fig. 4).A second period of non-occupation is observed during the transitionfrom the GS-2b stadial episode to GS-2a. The climatic phase of the GS-2a turned out to be colder than the previous one (GS-2b); however, itwas during this transition that the greatest drop in temperatures in bothperiods was recorded. This cold period coincided with the end of theLower Magdalenian in this area, at a time without radiocarbon datingpresence. The third absence of documented dating in the graph oc-curred during the Upper Magdalenian and coincided with the inter-stadial episode GI-1e and the onset of GI-1d, more temperate than theprevious episodes. On the calibration, this appears as a lack of an-thropic occupation at a time during which the temperatures are the bestof the Magdalenian period. 3.2. Macrospatial analysis The data extracted from the DEM analysis with GIS and site typeinformation according to other studies are included in Table 3. 3.2.1. Orographic characteristics The classi fi cation of the Magdalenian sites according to theirchronology shows large changes in the settlement sites according to thedata extracted from Table 3. Among these sites we observe a preferencefor rock shelters; only three are in caves and only one is an open-air site.However, it is important to point out that no cave settlements weredocumented during the Lower Magdalenian; in contrast, during theMiddle Magdalenian cave occupation was the preferred type of settle-ment (Fig. 5).Analysing the distance from the settlement to the nearest waterpoint, a trend towards shorter distances is observed over time. From ageneral perspective, Fig. 6 shows that during the Lower Magdaleniansettlements were located at a considerable distance from the rivers, butmuch closer in the subsequent chrono-cultural phases. 3.2.2. Solar radiation analysis Solar radiation in the sites is an important variable for the estab-lishment of settlements. The median of sunlight hours at the LowerMagdalenian sites was 11.02h/day, in the Middle Magdalenian11.22h/day and in the Upper Magdalenian is 11.33h/day (Fig. 7).However, seasonal values show variability in potential sunlight.Very high values are obtained between May and August, with 14 and15h/day, in the Bora Gran d'en Carreras, Legunova and Peña 14, andmuch lower values in the months of December and January at sitesForcas I and Fuente del Trucho, with 5 and 6h/day respectively. Therest of the sites remain more or less stable throughout the year.The GIS analysis calculating the optimal hours within a perimeter of 3km around the archaeological sites, displayed in Table 4. An analysisof K-means for two clusters can di ff  erentiate between those sites withlow insolation. Thus, Forcas I, Fuente del Trucho and Cueva de Chavessites are in locations with fewer hours of sunlight than the other sitesanalysed, losing 27, 24 and 16h/year respectively. Cova Alonsé, CovaGran, Peña 14 and Cova del Parco are more favourably located withrespect to hours of sunlight, while Legunova and Bora Gran d'en Car-reras have the best locations of all, with maximum hours of sunlight.Also, the sites that receive less sunlight are oriented in the West andSoutheast directions. Thus, the most favourable orientations are East,South and Northeast (Table 5). 3.2.3. Viewshed analysis The results for the visibility analysis, based on the number of hec-tares visible from the sites, showed three distinct groups. Group Acomprises sites with>1000ha of visibility, group B sites with visibilitybetween 1000 and 350ha, and  fi nally, group C with a visual basinof<350ha (Fig. 8).The analysis of the visibility superimposed on the DEM-LIDAR mapshows that nine of the ten sites have direct visual contact with the river.Cova Gran is the only one where the river is not visible.To establish these dynamics according to the chrono-culturalphases, sites occupied during the Lower Magdalenian might be ascribedto any of the three groups: A, B and C. Sites from the Middle Fig. 3.  To the left: graph of friction according to the percentage of the values of slopes; right: values and colour ranges to integrate into raster map cells with Slopevalues (output).  B. Mas et al.  Journal of Archaeological Science: Reports 22 (2018) 237–247  241
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks