Potential and limitations of LiDAR altimetry in archaeologicalsurvey. Copper Age and Bronze Age settlements in southernIberia

RESEARCH ARTICLE Potential and limitations of LiDAR altimetry in archaeological survey. Copper Age and Bronze Age settlements in southern Iberia Francisco Sánchez Díaz 1 | Leonardo García Sanjuán 2 | Timoteo Rivera Jiménez 3 1 Research Group ATLAS (HUM-694), Department of Prehistory and Archaeology, University of Sevilla, María de Padilla, Sevilla, 41004, Spain 2 Department of Prehistory and Archaeology, University of Sevilla, Sevilla, 41004, Spain 3 Research Group HUM-838, University of Huelva, Huelva, Spain Correspondence Leonardo García Sanjuán, Department of Prehistory and Archaeology, University of Sevilla, María de Padilla, 41004, Sevilla, Spain. Email: [email protected] 1|INTRODUCTION The archaeological use of images and data obtained through devices carried on mobile platforms (such as airplanes and satellites) is already one century old. Today, aerial photography and remote sensing are routinely used to capture, process and analyse archaeological evidence present on the surface of the earth, which is reflected in a large body of literature—see Bewley (1999), Corsi et al. (2013), Palmer and Cox (1993), Piccarreta and Ceraudo (2000), Riley (1987) and Wilson (1982) for the former and Campana and Forte (2001), Lasaponara and Masini (2012), Limp (1989), Lyons (1981), Wiseman and El-Baz (2007) and Wheatley and Gillings (2013), for the latter. In the last two decades, there has been a steady increase in the usage of altimetric analysis based on high-resolution techniques aimed at the detection of architectural elements both above ground and underground which are difficult to detect through conventional air photography and remote sensing methods. Prominent among those techniques is airborne laser scanning (ALS), which, like terrestrial laser scanning (TLS), allows for the detection and measurement of microtopographies with a level of precision not attainable with conventional techniques of surveying and photogrammetric restitution (Challis et al., 2008; Chase et al., 2010; Doneus & Briese, 2006; Doneus & Kühteiber, 2013; Fernandez-Diaz et al., 2014; Fontana, 2022; Gallagher & Josephs, 2008; Harmon et al., 2006; Opitz, 2013; Opitz & Cowley, 2013; Risbøl, 2010; Risbøl & Gustavsen, 2018). As is well-known, this technology uses active LiDAR (light detection and ranging) sensors which emit a beam of polarized infrared light which is discretized in pulses in order to measure the distance between the sensor and the scanned object by the time difference between the pulse emission and the reception of its reflection (time of flight, TOF). This offers a value of the relative position of the object with regards the sensor, which in turn must be converted in absolute terrestrial coordinates within a geodesic system through an accurate measurement of the position, altitude, orientation and sensor speed by means of a global navigation satellite system (GNSS) with differential correction and an inertial measurement unit (IMU). When LiDAR sensors are fixed on airplanes, decimetric levels of accuracy are achieved, which may turn centimetric on helicopters or drones. The final result is a three-dimensional scatter of points which may be treated through digital 3D-modelling applications to create precise altimetric models, using both the first returns to produce a digital surface model (DSM) or the ground returns (filtered) to produce a digital terrain model (DTM) (Opitz, 2013). The application of ALS technology to extensive archaeological reconnaissance is fairly recent. Over the last decade, LiDAR has proven extremely useful, particularly in densely forested regions of northern Europe, the American continent and Southeast Asia, although its usage in Mediterranean environments is still limited. After an initial phase of testing and calibration, highly innovative and even ground-breaking results have been achieved—see, for example, Barnes (2003), Doneus and Briese (2006), Doneus (2013), Harmon et al. (2006), Challis et al. (2008), Chase et al. (2010), Risbøl (2010), Crutchley (2013), Evans (2016), Canuto et al. (2018), Historic England (2018), Guyot et al. (2021), and Prümers et al. (2022). In Spain, public, freely accessible and updated altimetric data are issued periodically since 2014, which has fostered a variety of uses in a number of disciplines, including Archaeology. The first archaeological applications of Spanish LiDAR survey data have appeared over the last 5 years, including case studies centred on Neolithic monuments, Iron Age cities and Roman camps—see Cerrillo-Cuenca and L opez L opez (2020) for a synthesis. The first of them examined an area of the Portuguese Alentejo region and Spanish Extremadura with known Received: 9 September 2021 Revised: 19 May 2022 Accepted: 2 June 2022 DOI: 10.1002/arp.1869 Archaeological Prospection. 2022;1–20. wileyonlinelibrary.com/journal/arp © 2022 John Wiley & Sons Ltd. 1 fortified sites and ditched enclosures, using 1 m resolution DTM from the LiDAR datasets obtained through the facilities of the Spanish National Geographic Institute (IGN in its Spanish acronym) (Cerrillo- Cuenca & Bueno Ramírez, 2019). The same IGN data were used to map the topography of Iron Age, Ancient and Medieval Cordoba (Monterroso-Checa et al., 2021), the amphitheatre of the Roman city of Torreparedones, as well as to suggest a new location for the Phoenician temple of Melkart (Hercules) in San Fernando, Cádiz, combining the laser altimetry with sonar bathymetry produced by the Spanish Oceanography Institute (Monterroso-Checa, 2017,2019,2021). Other very recent examples also include the reconnaisance of 135 Iron Age ‘castros’(hillforts) in Galicia, including 25 previously unknown ones, with buried features, ditches, pathways, field boundaries and levelled defensive elements (Parcero-Oubiña, 2021), also a fresh cartography of the pre-Roman ‘castro’at Irueña, Salamanca, combining surface surveys with LiDAR and GIS technology (Berrocal-Rangel et al., 2017) as well as a study of the Roman military presence in the northern fringe of the Duero basin, where 66 new archaeological sites were discovered thanks to the combined use of different remote sensing techniques and open access geospatial datasets, mainly aerial photography, satellite imagery and airborne LiDAR (Menéndez Blanco et al., 2020). Recent papers have looked into the possibilities of aerial LiDAR for the detection of megalithic mounds in Galicia (NW Spain) (Carrero-Pazos et al., 2014; Carrero-Pazos & Vilas Estévez, 2016). A very recent line of work looks at the development of algorithms for the automatic detection of archaeologically relevant microtopographies through data mining and artificial intelligence, which has led to the successful location of a great number of megalithic mounds (Berganzo-Besga et al., 2021; Cerrillo-Cuenca, 2016). Thus, while Spanish LiDAR survey data have been used to explore a relatively wide range of archaeological sites and features, including megalithic mounds, Phoenician temples, pre-Roman towns and Roman military camps, to this date, no attempt has been made to examine one of the most powerful ‘segments’of Iberian Late Prehistory, namely, Copper Age (c. 3200–2200 BCE) and Bronze Age (c. 2200– 850 BCE) settlements. The general aim of this paper is to test to what extent the resolution of the LiDAR data (in this case, the public data available for Spain) and the accuracy of the existing algorithms allow for identifications and characterizations of specific features and architectural elements found at settlements dating to the Copper Age (CA) and Bronze Age (BA). As will be explained below, not only there is a wealth of such sites awaiting examination but also the results obtained by us suggest that much can be gained from the use of LiDAR data in order to build a scientific understanding of them. 2|CASE STUDY Late prehistoric settlements with major stone-walled features are a pervasive phenomenon across most of the southern half of Iberia. In its early stages, starting c. 3000–2900 BCE, this was linked to the gradual sedentarisation of Late Neolithic communities coupled with a reduction of residential mobility, both of which were the result of demographic growth and economic intensification, allowing a greater accumulation of staple surplus. At various southern Iberian locations, ‘fortified’or ‘walled’sites have been interpreted as resulting from either defensive needs or monumentalisation practices (García Sanjuán & Murillo-Barroso, 2013; Gonçalves et al., 2013: 35; Jorge, 2003; Mataloto & Boaventura, 2009: 59). At an advanced stage, beginning c. 2200 BCE (Early BA), this phenomenon resulted in nucleated and well-defended settlements located on inaccessible hilltops, which also demanded the construction of large terraces capable of providing horizontal space to live on. Thus, CA and BA settlements presenting major stone-walled features have been identified across much of southern Iberia (see distribution in the map shown in Figure 1). In the Spanish Southwest, they are common in the provinces of Huelva, where sites such as Cabezo de los Vientos, Cabezo Juré, El Trastej on and La Papúa have been excavated (García Sanjuán et al., 2011; Hurtado Pérez, García Sanjuán, & Hunt Ortiz, 2011; Nocete Calvo et al., 2004; Pérez Macías et al., 2019; Sánchez Díaz & Hunt Ortiz, 2021), as well as in Badajoz, where San Blas, La Pijotilla, Palacio Quemado, Los Cortinales and Castillo de Alange have been explored (Enríquez Navascues, 1990; Hurtado Pérez & Hunt Ortiz, 1999; Pav on Soldevila & Duque Espino, 2014). On the Portuguese side of the border, settlements located along the basins of the rivers Guadiana and Tagus, such as Santa Justa, Vila Nova de S˜ao Pedro, Leceia and Zambujal have been excavated over several decades (Cardoso, 1997). In the Spanish Southeast, numerous CA and BA settlements showing major stone architecture have been excavated since the first explorations by Luis and Henry Siret in the late 19th century (Siret & Siret, 1890). Of course, they would be too numerous to list here. Although some settlements with major stone architecture flourished already in the CA, notably Los Millares, they became much more frequent and more impressive within the so-called ‘Argaric’culture of the Early BA (c. 2200–1550 BCE)—see Chapman, 1990; Aranda Jiménez et al., 2015 for good syntheses in English. In the provinces of Jaén and Granada, excavated Argaric sites (not always well-published) include Peñalosa, Castell on Alto, Terrera del Reloj, Cerro de la Virgen and Cuesta del Negro (García Huerta & Morales Hervás, 2004). In the province of Almería, several major BA settlements located on hilltops and provided with very substantial stone-walled civil and defensive architecture were excavated, including the eponymous site of El Argar and others such as Fuente ´ Alamo, El Oficio, Fuente Vermeja, Lugarico Viejo, El Picacho and Gatas (Gilman & Thornes, 1985). At these sites, the steep slopes were levelled by means of terraces while hilltops were often surrounded by an acropolis, sometimes including cisterns and other major infrastructures (Molina González & Cámara Serrano, 2004). Numerous Argaric BA settlements have also been excavated in the neighbouring Murcia province, including Placica de Caravaca, Covaticas, Cerro del Cuchillo, La Morra del Moro and Ifre (Eiroa, 2004), as well as, more recently, La Bastida and La Almoloya. With more than 4 ha and several levels of stepped terraces on which houses of areas between 10 and 70 m 2 were built, La Bastida is one of the largest BA settlements in the Spanish Southeast. On its perimeter, which is protected on three sides by steep cliffs, stands out a 2S´ ANCHEZ DÍAZ ET AL. defensive wall that could have stretched over 300 m, negotiating steep slopes of up to 40%. The excavated sector of this wall is 3 m wide, with bastions that project up to 3.50 m from its external face and are separated between 2.80 and 4.70 m. Considering the volume of collapsed materials, the original height of this wall would have been 5 m. At the lowest end of this wall, there is a ‘covering gate’flanked by two 3 m wide forts, which make up a narrow 1.5 m wide corridor (Lull et al., 2014). Further north, in La Mancha (the southern sector of the Spanish Central Plateau), dozens of BA settlements with major stone architecture have been described (Ruiz Taboada, 1997; Fernández-Posse et al., 2007; etc.) grouped in various types referred to in the literature as ‘motillas’(such as Las Cañas, Los Romeros, El Retamar, El Acequi on and El Azuer), ‘morras’and ‘castillejos’(El Quintanar, La Encantada, El Acebuchal, Los Dornajos, El Recuenco, Las Alberquillas and El Romeral). The same applies to the so-called ‘Valencian BA’, in the central sector of the Spanish Mediterranean coast (Tarradell, 1961), as can be seen in San Ant on de Orihuela, Callosa de Segura, Lloma de Betxí, Tabaiá, Muntanya Assolada and Terlinques (Pedro Mich o& Martí Oliver, 2004). Altogether, although these settlements share underlying locational preferences seen across southern Iberia, they are considerably smaller than those found in the southwest and the southeast, while stone-made civil or defensive architecture appears much less frequently and at a smaller scale, which sets them apart from the patterns seen in the Argaric area or the Sierra Morena region. Given that, throughout southern Iberia, settlements dating to third and second millennia often show well-defined topographic FIGURE 1 Main Copper Age and Bronze Age settlements known in Iberia. Base map: © Instituto de Estadística y Cartografía de Andalucía [Colour figure can be viewed at wileyonlinelibrary.com] S´ ANCHEZ DÍAZ ET AL.3 features, caused by the large-scale stone architecture that was intrinsic to them, this paper departs from the assumption that LiDAR altimetry can contribute to a more precise planimetry of their morphology and features. The application of high-resolution LIDAR altimetry is aimed at examining the potential and limitations of the technology for future archaeological research and management. Strictly, the aim is not to assess the ability of high-resolution LIDAR altimetry to identify or locate ‘new’archaeological sites, as this ability has been sufficiently demonstrated over the last 20 years (as briefly discussed above). Instead, the aim is to test to what extent it is possible to identify and characterize specific features and architectural elements with a matching or superior quality to that already achieved by other means, such as, principally, fieldwalking or air photography (obviously, LiDAR cannot match the level of precision attained by archaeological excavation through means such as hand drawing or photogrammetry). With this aim in mind, a case study has been selected for which knowledge based on previous fieldwork, including fieldwalking and excavation, is available. This involves a series of CA and BA settlements located in the Sierra Morena highlands of southwestern Spain (Figure 2). Between the late 1980s and early 1990s, the University of Sevilla developed a research project across this region (northern provinces of Huelva and Sevilla). As part of said project, several survey seasons were undertaken which led to the discovery and characterization of multiple new settlements, two of which, namely, La Papúa II and El Trastej on, were also excavated in 1988, 1990 and 1994 (García Sanjuán, 1999; Hurtado Pérez, García Sanjuán, & Hunt Ortiz, 2011; Hurtado Pérez, García Sanjuán, Mondéjar Fernández de Quincoces, & Romero Bomba, 2011; Hurtado Pérez, Romero Bomba, & Rivera Jiménez, 2011; García Sanjuán et al., 2011). The fieldwork that led to the original data capture was executed by the same team (University of Sevilla) using the same criteria and methods, and therefore, there was a substantial unity in the quality and intensity of the field surveys that led to the discovery of the sites. Up to 36 of these settlements, for which previous documentation exists in the form of plans, photographs and written descriptions, have been selected in order to test to what extent can LiDAR altimetry improve and expand previously existing records. Specifically, the potential and limitations of LiDAR to establish parameters such a site size, perimeter and major associated stonebuilt architectural features, including defensive walls with gates, bastions and towers, terraces, dwellings and even pathways, streets and roads, will be examined. The underlying assumption is that if high-resolution LiDAR can match or even surpass what was previously mapped on the basis of direct observation by fieldwalking, field topography and aerial photography, then the method can be safely used to detect and characterize new settlements in regions with analogous geographic settings for which no previous research is available. In principle, this has great potential both for research and management purposes, as it could lead to the creation of inventories of sites which may be legally protected on the basis of LiDAR-based surveys alone with (or even without) ulterior ground truthing. FIGURE 2 Distribution map of the sites included in this study. Base map: © LiDAR-PNOA 2014 CC-BY 4.0 scne.es. Coordinates: EPSG 25829 in metres [Colour figure can be viewed at wileyonlinelibrary.com] 4S´ ANCHEZ DÍAZ ET AL. 3|METHODS The application of ALS technologies to archaeological survey has led to the consolidation of a new methodology with well-defined and well-tested phases, processes and parameters (Adamopoulos & Rinaudo, 2020; Lozi c&Štular, 2021). The usual workflow includes the capture, processing, analysis, interpretation and representation of data. In turn, each of these phases incorporates specific processes, some obligatory and some optional (Figure 3). For this study, we used the LiDAR data captured by the Spanish IGN in 2014 with a density of 1 p/m 2 , reduced to a 0.32 p/m 2 once ground-classified, rasterized into DTMs with a pixel size of 1 m using the open-source programme LASTools and visualized with relief visualization toolbox (RVT) and the 3D module of QGIS named Qgis2threejs to interpret archaeological structures. The capture of altimetric information is normally made in the winter, in order to minimize the ‘masking’effect caused by vegetation. Flight altitudes oscillate between 200 m for drone-based projects (UAVs), 650 m for helicopters and 3000 m for airplanes, with pulse emission frequencies between 45 and 500 kHz. With these altitude and frequency settings, data point densities normally ranging between 0.5 and 21 points per square metre are achieved, although there are major differences between official reconnaissance projects run by state agencies with multipurpose coverage (between 0.5 p/m 2 in Spain and 8 in the Netherlands, for example) and those tailor-made for archaeological purposes, which normally achieve 16 p/m 2 FIGURE 3 Usual workflow in an archaeological project with airborne laser scanning (ALS) [Colour figure can be viewed at wileyonlinelibrary.com] S´ ANCHEZ DÍAZ ET AL.5 (Table 1). On the other hand, altimetric precision is barely influenced by flight altitude, as with 650 m the root-mean-square error (RMSE) Z reaches 10 cm, whereas at 3000 m it reaches 20 cm, on account of the error induced by GNSS and IMU systems, which is bigger than that of the altimeter and independent from altitude. The field of view (FOV) is usually limited to a maximum of 50, given that with higher angles, the probability of the beam hitting the ground is too low. Once the data point scatter is captured, it is subjected to a process that includes quality control, geometric correction, transformation of altimetric datum, georeferencing within a coordinate system, colour attribution on the basis of orthophotos, differentiation and classification of returns, organization in squares and compression into LAS or LAZ formats (Lorite Martínez et al., 2017). In our case, a study area of 100 40 km was been defined for which LiDAR data were obtained from the Spanish National Centre of Geographic Information (CNIG in its Spanish acronym), 1 which supplies the LiDAR point coverages produced within the framework of the National Plan of Aerial Orthophotography (PNOA), with a 6-year periodicity and coverage for the whole country. CNIG supplies LAZ- format files with point scatters arranged by 2 2 km cells, automatically classified by FWF and coloured by RGB and infrared on the basis of 25 cm orthophotos. As well as being public and free, the CNIG licence of use incorporates a copyright cession which allows the reutilization of the data for any lawful purpose, with the only condition of recognizing and acknowledging the source of the data, as well as citing authorship. This is a product funded by the Spanish National Cartographic System, and recognition of copyright must be expressed as © LiDAR-PNOA 2014 CC-BY 4.0 scne.es. The technical specifications of PNOA's ALS coverage set the parameters concerning flight, sensors, processes and final data which this new cartographic product must satisfy (IGN, 2014). According to these specifications, the point scatter must include orthometric altitudes over the EGM08 reference geoid, UTM projections in various zones, 0.5 p/m 2 point density—or 1.5 in the new series—and a minimum 40 cm RMSE Z altimetric precision, with 20 cm average. Other relevant specifications for archaeological use include 3000 m maximum flight altitude, 50maximum angle, 45 kHz minimum scan frequency, ≤1.41 m point spacing, up to four returns per pulse with discrimination in vertical distance of at least 4 m, 8-bit radiometric resolution and global horizontal position lower than 30 cm RMSE X, Y (Lorite Martínez et al., 2017). These initial specifications, regarded as minimum standards, are basically satisfied within different regions of Spain. For western Sierra Morena, the 2014 coverage provides an even better quality. The usual resolution is in fact higher than 1 p/m 2 and in some cases above 2 p/m 2 , which allows to derive raster LiDAR- based DTMs with 1 m resolution. In addition, the average distance between points is reduced to 0.99 m and sometimes even below 0.7 m. Table 2shows the quality parameters of the LiDAR data processed as part of this study. In order to achieve the phase of classification of returns, fullwaveform (FWF) analysis of the reflected signal allows the decomposition of the various echoes that form it, assigning an altitude to each of the captured objects while at the same time differentiating first, TABLE 1 Specifications of ALS flights used in archaeological projects Zone Spain England Norway Netherlands Hebrides Brittany Guatemala Mexico Cambodia Source IGN, (2014) Crutchley (2013) https://hoydedata.no/ LaserInnsyn/ https://www. pdok.nl/ https://remotesensingdata. gov.scot/ Guyot et al. (2021) Canuto et al. (2018) Ringle et al. (2021) Evans et al. (2013) Pulses/ m 2 0.5–1.5 4 2 8 16 14 15 21 16 DTM m 5–1 1 1 0,5 0,25 0,5 1 0,5 0,25 FOV 50 26 28 45 RMSE Z m 0.20 0.15 0.10 0.05 0.15 0.10 0.10 0,10 0,15 Abbreviations: DTM, digital terrain model; FOV, field of view. 6S´ ANCHEZ DÍAZ ET AL. intermediate and last returns. Of these processes, the most archaeologically relevant is the classification of data points according to object type (Doneus et al., 2020), for which both return gap and intensity of the reflected signal are used. In the case of Spain, in order to differentiate between buildings and high vegetation, the vegetation coverage index (NDVI), calculated on the basis of infrared imagery, is used in addition. This classification process is automatized by means of algorithms connecting the position of each point with the nearest ones, although this involves a subsequent correction of errors that has to be done manually (Lorite Martínez et al., 2017). Filtering of returns to obtain the altitude of the bare ground once buildings and vegetation have been removed can be an issue. Scarcity of soil returns caused by vegetation density and imprecision of classifications can become major limitations of LiDAR-based remote sensing. In the case of PNOA's LiDAR, which offers already classified returns, only 6.3 million out of 20.7 million points processed for this study corresponded to Class 2 (ground), while 4.7 million were assigned to vegetation. The rest, almost half, were not been classified and were assigned to Class 12 (overlay), Class 7 (noise) or Class 1 (unclassified), which causes half the potential resolution to be lost. In practice, this reduction to 30% of the points available to derive a DTM causes usable returns to have an average density of only 0.32 p/m 2 , with peaks between 0.12 and 0.77 p/m 2 . It is important to note that when the Class 2 (ground) point density fell below 0.25 p/m 2 ,it was not possible to identify archaeological features. On the other hand, when the point density was above 0.5 p/m 2 , it was possible to identify terraces or walls in all cases. Once processed, LiDAR data are available for downloading from national cartographic agencies, divided into squares which in Spain measure 2 2 km (by comparison, they measure 5 5kmin England) and converted into a DTM for their analysis, typically using ground-classified points in order to eliminate vegetation and/or buildings. In principle, it is sufficient to use type 2 (ground), but depending on the quality of the classification, some points labelled as 6 (building) may contain archaeological features, or some points identified as 3 (low vegetation) may be ground (Costa-García et al., 2017; Costa- García & Fonte, 2017). Depending on the density of returns at ground level, more or less precise DTMs can be achieved (Opitz & Cowley, 2013). In the case of PNOA's LiDAR, the usable returns had an average density of only 0.32 p/m 2 , with peaks between 0.12 and 0.77 p/m 2 , which allowed to derive LiDAR-based DTMs with 1 m resolution through interpolation if, instead of rasterizing altitude values through K-Nearest Neighbours Algorithm (KNN) algorithm, the DTM is derived from an intermediate tridimensional model in triangulated irregular network (TIN) (Štular et al., 2021). In our case, the open-source programme LASTools, integrated in package QGIS, was used to calculate the DTMs in the GeoTIFF format. Starting from the DTM, the entire process of raster analysis can be achieved through map algebra. The detection of archaeological features is based on the analysis of relief variation details, for which algorithms automatizing form recognition are being developed but which still demand a visual interpretation (Verschoof-van der Vaart et al., 2020). Microtopography-enhancing methods range from simple hillshading (HS), to more complex techniques based on calculations of slope and aspect. Among the latter are algorithms normally used for archaeological survey such as sky view factor (SVF), openness, local relief model (LRM), principal components analysis (PCA), local dominance (LD), cumulative visibility (CV), multiscale integral invariants (MSII) or Laplacian-of-Gaussian (LoG) (Bennett et al., 2012). Especially, the red relief image maps (RRIM) visualization technique enhances the visibility of subtle features combining Slope, Hillshade and Differential Openness (Daxer, 2020). Thanks to work undertaken in the last decade, the parameters needed for each method are now more precise, including azimuth, elevation, filtering radius, number of directions or search distance (Kokalj & Hesse, 2017). In any case, for the identification of archaeological features, a combination of several of these techniques may be necessary, depending on factors such as feature size, slope of the terrain, land use or surface alterations (Costa-García & Fonte, 2017). In the case of the CA and BA settlements examined here, with a terraced relief, the most useful techniques were slope, SVF, PCA and RRIM, by means of software RVT and lidar visualization toolbox (LiVT) (Kokalj & Hesse, 2017). Table 3 shows the techniques, parameters and values used in this study for the visualization of settlements. The interpretation of images, whether visual or automatic, is aimed at the identification, inventory and cartographic representation of archaeological features. In the case of terraces, a very useful algorithm for their automatic identification is the GRASS function r.param. scale, which measures the convexity in the direction of maximum slope (Arnau-Rosalén et al., 2018). Normally, a distinction is made between polygonal features such as enclosures, terraces, dwellings or burials; linear features, such as roads, canals or walls; and point TABLE 2 Quality parameters of LiDAR data used in the 36 surveyed sites Area surveyed Total points Last returns Ground points Ground/ total Ground/ vegetation Angle max Spacing Points density Ground density Unit m 2 nnn%n  m p/m 2 p/m 2 Max 1 493 500 1363 237 1 105 854 535 114 54.25 3.78 43 1.65 2.15 0.77 Min 109 744 71 617 71 617 15 977 12.78 0.40 12 0.68 0.37 0.12 Mean 538 145 575 401 483 206 174 568 30.33 1.32 25 0.99 1.07 0.32 Sum 19 373 216 20 714 422 17 395 433 6 284 433 Abbreviation: LiDAR, light detection and ranging. S´ ANCHEZ DÍAZ ET AL.7 features (Mlekuž,2013). In the interpretation of LiDAR data and aerial photographs, Historic England uses a classification according to relief morphology, distinguishing four types: structure, ditch, bank and slope. This process involves the conversion of the shapes seen on the images from their raster format into new vectorial entities, which may be stored in spatial databases and used in geographic information systems (GIS) (Gillings et al., 2020). The last phase of ALS projects includes the representation of the identified features based on the TABLE 3 Techniques, parameters and values used for the visualization of terraced settlements Techniques Parameters Values Hillshade Azimuth 315to 45 Elevation 30 Stretch Linear, 2% cut-off Slope Unit Degrees Stretch Linear, 0to 60 Sky view factor Search radius (pixels) 10 Directions 16 Stretch Linear, 0.64 to 1 PCA of Hillshade Components 3 (RGB) Elevation 25 Directions 16 Stretch Linear, 2% cut-off Local relief model Search radius (pixels) 25 Stretch Linear, 1to1m Positive openness Search radius (pixels) 10 Directions 16 Stretch Linear, 0.6 to 1 Visualization for archaeological topography Sky View Factor Stretch Linear, 0.7 to 1 Blend Multiply Opacity 25 Positive Openness Stretch Linear, 68 to 93 Blend Overlay Opacity 50 Slope Stretch Linear, 0to 50 Blend Luminosity Opacity 50 Analytical Hillshading Stretch Linear, 0 to 1 Blend Normal Opacity 100 Red relief image maps Slope Stretch Linear, 0to 50 Blend Multiply (red) Opacity 80 Differential Openness Stretch Linear, 1% cut-off Blend Normal Opacity 100 Figure 7: Hypsographic maps Elevation Stretch Linear, mean +1s Blend Multiply (green to red) Opacity 50 Slope Stretch Linear, 0to 50 Blend Multiply Opacity 100 8S´ ANCHEZ DÍAZ ET AL. many possibilities offered by cartographic semiology. Normally, intuitive representations akin to natural vision are used, based on the shading of relief through low illumination, between 10and 35and 315azimuth (Kokalj & Hesse, 2017). Other useful cartographic representations can be achieved through altimetric colours combined with slopes (Kokalj & Somrak, 2019). Given the potential of LiDAR data to create 3D virtual environments, one of the most recurrent ways to display and disseminate results is through perspective views, which may even be interactive when published in web environments (Popovic et al., 2017). In this study, both the interpretation phase and the mapping and 3D render phase were carried out using the opensource software QGIS 3.2.2 with SAGA 2.3.2 and GRASS 7. There is no single computing package capable of handling the execution of all the processes described above in a fully integrated environment, although recently, the Open LiDAR Toolbox integrates an archaeology-specific LiDAR workflow into a unified interface, using GRASS, QGIS, LASTools and RVT libraries (Štular et al., 2021). Therefore, specific pieces of software must be used for each of those phases. The capture and treatment of raw data are linked to sensors and are usually achieved through licenced software such as Trimble MX or Leica LSS, a software tool for point-cloud generation and cleaning of raw LiDAR data. For all other processes, there are free open-source programmes, whether as plug-ins integrated in GIS or as independent applications. The classification of returns through FWF can be achieved with open-source software packages such as MCC- Lidar, Fusion or LASTools, although there are also commercial programmes, such as TerraSolid, VR Mesh or MARS. For point visualization, filtering, cutting, union, transformation or exportation, the most used programme is LASTools, which can be integrated in packages such as QGIS or ArcGIS, although there are other open-source programmes such as MeshLab, Geomagic XOS and FugroViewer. To convert an LAS file into a DTM raster file, the commercial software Surfer or the open-source Whitebox Geospatial Analysis Tools and LASTools are available, with exportation into different formats such as TIF, BIL, IMG, CSV or ASCII. For DTM analysis and exportation into image format, there is a number of raster GIS packages, such as SAGA and GRASS, but specific and valuable archaeology-oriented tools have been developed, as is the case with RVT and LiVT. The interpretation stage is the least automated of all, although it can be supported by generic GIS packages. Visualization through perspective-viewing and 3D visual environments, including web publication, can be achieved through both commercial (Geoweb3d) or open-source (Qgis2threejs) applications. 4|RESULTS The approach described above leads to a number of significant results concerning the usability of Spanish public LiDAR data to map CA and BA sites. Provided some conditions are met, such data can be used TABLE 4 Comparison of architectural features detected by field work and by LiDAR Fieldwalking LiDAR Name Municipality Chronology Area (ha) Features Area (m 2 ) Perimeter (m) Features La Papúa II Arroyomolinos de Le on BA 10 W/B/G 110,908 1747 W/B/G La Papúa I Arroyomolinos de Le on BA 5 W/G 55,402 994 W La Lapa Encinasola BA 0,1 T/W 42,836 984 T/W El Trastej on Zufre BA 3 T/W/B/G 32,404 883 T/W/B/G Atalaya del Trastej on Zufre BA 0,52 W 5,259 458 W Castrej on de Capote Higuera La Real BA/IA 3 W/G 29,450 851 W/G Las Pedreras Cañaveral de Le on BA 0,8 W 22,435 676 T/W/B/G La Bujarda La Nava BA 1,4 T/W/B 18,175 703 T/W/B/G Cerro del Cojo Cortegana CA 0,37 W 15,894 508 T/W/B Los Rehoyos Cala CA/BA T 13,308 484 T Santa Catalina Zufre BA 1,1 T 12,168 471 W Castillo Aroche BA 1 W 11,609 433 T El Torrej on Aroche CA/IA 0,8 W 8,224 485 T/W/B/G Los Castillejos Santa Olalla del Cala CA/IA 0,05 W 7,101 488 T/W/B Cerro Librero Corteconcepci on CA/BA 0,2 T 4,737 286 T El Castañuelo Aracena BA/IA 1,1 W 4,300 293 T La Muela Aroche CA/IA W 3,290 251 T Santa Marta III Santa Olalla del Cala BA T 3,152 212 T Santa Marta II Santa Olalla del Cala BA 0,7 T/W/B 3,089 224 T/W Cerro del Almendro Santa Olalla del Cala CA 1 T/W 1,158 157 T Abbreviations: CA, Copper Age; B, Bastions; BA, Bronze Age; G, Gates; IA, Iron Age; LiDAR, light detection and ranging; T, terraces; W, walls. S´ ANCHEZ DÍAZ ET AL.9 features are few, small and/or nondescript. A good example of this, which is currently under study, is mining features commonly associated to CA and BA settlements in western Sierra Morena, such as trenches, shafts and dump sites. On a LiDAR cartography, the image of such features can be very similar for sites dating to a wide chronology, from the CA to the Middle Ages. It is also worth mentioning that LiDAR-derived microreliefs can be envisaged as a basic constituent of prehistoric landscapes, incorporating a history of construction, use and reuse like a palimpsest (Johnson & Ouimet, 2018), as well as a series of long-term geomorphological transformations and post-depositional alterations. The interpretation of microreliefs as a succession of changes over time is useful to identify modern land uses as a determining element in order to assess the usefulness of ALS techniques. Comparing the 20 settlements of the study area for which features previously identified through fieldwork have been successfully mapped through LiDAR with the 16 for which this has not been possible, it becomes clear that certain contemporary land uses prevent the successful use of LiDAR for CA and BA settlements. This includes industrial forestry, irrigation agriculture, large-scale mining and major infrastructures. It goes without saying that the same is true for those CA and BA settlements on which major occupancy, including castles, occurred in the Middle Age (as is the case of Cala, Aracena, Cortegana and Cumbres Mayores). In a way, all the settlements successfully mapped in this study have in common that they failed to sustain a long-term occupation: they were abandoned at the end of the BA or during the IA and were never reoccupied in the Middle Ages. The results obtained in this study also suggest some of the caveats and problems to be considered when using LiDAR in other Mediterranean regions. One interesting point concerns vegetation coverage. Unlike deciduous forests of the northern reaches, where a leaf-free season allows an optimal laser penetration, Mediterranean xerophilous tree and bush vegetation coverage is constant throughout the year. Dehesa-like Mediterranean oak-tree forests (with quercus suber and quercus ilex) allow for good resolutions once returns are classified, as there is sufficient distance between trees and the canopies are not so dense as to prevent ground-type returns. In turn, while pine forests offer less resolution, areas covered with dense Mediterranean bush vegetation such as rockrose (cistus ladanifer), gorse (genista scorpius), mastic (pistacia lentiscus) and arbutus (arbutus unedo) are impenetrable with the available density of pulses. At any rate, these limitations do also affect field walking. While dehesas are normally relatively amenable to direct survey, dense bush areas are often impenetrable. In summary, the application of LiDAR-based surveys would greatly help to enhance and homogenize the data pertaining CA and BA settlements and other periods. In the sample of 36 settlements studied in this paper, for which previous documentation existed in the form of plans, photographs and written descriptions, although limitations in raw data accuracy and changes in land use made it impossible to obtain high-quality plans for about half of those sites, ALS reached satisfactory results for the other half, improving previously existing records. LiDAR data allow for the identifications and characterizations of specific features and architectural elements of matching or superior quality to that already achieved by other means, such as fieldwalking and/or air photography. Paraphrasing Lozi c and Štular (2021), the case study presented here confirms that LiDAR data can fundamentally redefine archaeological survey work and spatial analysis by splitting the activities into desk-based interpretative mapping and field-based ground assessment, by enabling unsurpassed metrical accuracy, by providing the widest possible landscape context and by enabling a precisely targeted and planned fieldwork that can be carried out very efficiently. LiDAR can provide massive advantages to the cartographic representation of CA and BA settlements: While in surface surveys only isolated sections of the walls, mostly a few dozen metres in length (at best, after strenuous vegetation clearance, several hundreds of metres), could be recorded, LiDAR has been able to fully draw the layout of whole enclosures measuring several hundred metres in perimeter. This is of great importance, as, often, these hilltop sites are hard to get to and appear densely covered by a thick vegetation, which makes direct surveying highly energy and time-consuming and slow, if at all possible. Thus, the new LiDAR-based plans make possible the delimitation, planimetry and mapping of the settlement size with metric precision. Furthermore, LiDAR data can help to substantially improve conventional interpretations in terms of settlement patterns (population density, hierarchical ranking and locational preferences). The pattern resulting from the case study presented here show that both in the CA and BA, settlements were located on hill tops, although some substantial differences emerge in terms of locational strategies. In the BA, settlements were built on very steep, easily-defensible hilltops and were provided with larger and more substantial stone-walled infrastructures, some of them of clearly defensive nature, a trend observed across the whole of southern Iberia. This defensive concern was much more marked in the BA than in the CA—see full discussion in García Sanjuán et al., 2011. Indeed, the application of LiDAR-based surveys would greatly help to enhance and homogenize the data pertaining to CA and BA settlements, and other periods in which similar features were built, such as the Iron Age or the Middle Ages. In turn, in Iberia, this would contribute to challenge long-held notions regarding the ‘uniqueness’of the Argaric BA culture, thus providing a more encompassing and reliable perspective of prehistoric settlement patterns. Despite some of the problems mentioned above regarding data accuracy, the application of LiDAR altimetry is nonexpensive and relatively uncomplicated, given the availability of freely-accessible public data, which are improved every 6 years in Spain. In any case, public data need to be increased in resolution and classification accuracy for more effective archaeological use. In their present configuration (more so in the future, as precision increases), LiDAR data can significantly contribute to refine, enhance and homogenize the description of CA and BA settlements, both in terms of locational preferences and architectural configuration. Many of the features normally present at these sites can be accurately mapped by LiDAR altimetry, as shown by about half the cases examined in this paper. The evidence thus represented accurately reflects features such as terraces, defensive walls 16 S´ ANCHEZ DÍAZ ET AL. with bastions and gates, common in later prehistoric sites and particularly in BA. LiDAR-based technology will no doubt help predict the location of such kinds of settlements wherever, unlike the Sierra Morena region studied here, no earlier work has been carried out and no previous documentation exists. ALS projects have great potential to support and expand regional and national inventories of prehistoric sites, thus contributing to a better protection of the archaeological heritage. ACKNOWLEDGEMENTS The authors would like to thank the Spanish National Centre of Geographic Information (CNIG) for the CC-BY 4.0 licence. We also thank the two anonymous reviewers for their thorough and very useful comments and suggestions. CONFLICT OF INTEREST There are no conflict of interest. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available in the Spanish National Geographic Institute (http://www.ign.es/web/ign/ portal). These data were derived from the following resources available in the public domain: Spanish National Geographic Institute (http://centrodedescargas.cnig.es/CentroDescargas/catalogo.do? Serie=LIDAR). ORCID Francisco Sánchez Díaz https://orcid.org/0000-0001-6277-8799 Leonardo García Sanjuán https://orcid.org/0000-0002-8404-9252 Timoteo Rivera Jiménez https://orcid.org/0000-0001-9064-0682 ENDNOTES 1 http://centrodedescargas.cnig.es/CentroDescargas/catalogo.do?Serie= LIDAR 2 The dehesa is a multifunctional agro-sylvo-pastoral system and cultural landscape found in central and south-western Iberia, based on humanmanaged natural forests of oaks, usually holm oak (Quercus ilex) and cork oak (Quercus suber). As well as forestry exploitation, the dehesa is used primarily for grazing of animals, very especially Iberian pigs, which are allowed to roam freely, feeding on acorn. REFERENCES Adamopoulos, E., & Rinaudo, F. (2020). UAS-Based archaeological remote sensing: review, meta-analysis and state-of-the-art. Drones,4,1–28. https://doi.org/10.3390/drones4030046 Aranda Jiménez, G., Mont on-Subías, S., & Sánchez Romero, M. (2015). The Archaeology of Bronze Age Iberia. Argaric Societies. Routledge. Arnau-Rosalén, E., Calsamiglia, A., Calvo, A., Estrany, J., & Symeonakis, E. (2018). Cartografía semi-automática de terrazas de cultivo a partir de datos LiDAR. In P. C. González, J. S. Carrasco, M. J. L. García, & J. M. A. Puebla (Eds.), Tecnologías de la informaci on geográfica: perspectivas multidisciplinares en la sociedad del conocimiento (pp. 135– 144). Universitat de València. Barnes, I. (2003). Aerial remote-sensing techniques used in the management of archaeological monuments on the British Armys Salisbury Plain Training Area, Wiltshire, UK. Archaeological Prospection,10, 83–91. https://doi.org/10.1002/arp.197 Bennett, R., Welham, K., Hill, R. A., & Ford, A. (2012). A comparison of visualization techniques for models created from airborne laser scanned data. Archaeological Prospection,19,41–48. https://doi.org/ 10.1002/arp.1414 Berganzo-Besga, I., Orengo, H. A., Lumbreras, F., Carrero-Pazos, M., Fonte, J., & Vilas-Estévez, B. (2021). Hybrid MSRM-Based Deep Learning and Multitemporal Sentinel 2-Based Machine Learning Algorithm Detects Near 10k Archaeological Tumuli in North-Western Iberia. Remote Sensing,13(20), 1–18. https://doi.org/10.3390/ rs13204181 Berrocal-Rangel, L., Paniego Díaz, P., Ruano, L., & Manglano Valcárcel, G. (2017). Aplicaciones LiDAR a la topografía arqueol ogica: El Castro de Irueña (Fuenteguinaldo, Salamanca). Cuadernos de Prehistoria y Arqueología,43, 195–215. Bewley, R. (Ed.) (1999). Archiving Aerial Photography and Remote Sensing Data: A Guide to Good Practice. Oxbow Books. Campana, S., & Forte, M. (Eds.) (2001). Remote Sensing in Archaeology. XI Ciclo di Lezioni Sulla Ricerca Applicata in Archeologia. Universitá di Siena. Canuto, M., Estrada-Belli, F., Garrison, T., & Shrestha, R. (2018). Ancient lowland Maya complexity as revealed by airborne laser scanning of northern Guatemala. Science,361(6409), 201–219. Cardoso, J. L. (1997). Génese, apogeu e declínio das fortificaçoes calcolíticas da Extremadura. Zephyrus,50, 249–261. Carrero-Pazos, M., Estévez, B., Fariña, E., & Casal, A. (2014). La necr opolis del Monte de Santa Mariña revisitada: aportaciones del LIDAR aéreo para la cartografía megalítica de Galicia. Gallaecia: revista de arqueoloxía e antigüidade,33,39–57. Carrero-Pazos, M., & Vilas Estévez, B. (2016). The possibilities of the aerial LiDAR for the detection of Galician megalithic mounds (NW of the Iberian Peninsula). The case of Monte de Santa Mariña, Lugo. In Proceedings of the 43rd Annual Conference on Computer Applications and Quantitative Methods in Archaeology (pp. 901–908). Archaeopress. Cerrillo-Cuenca, E. (2016). An approach to the automatic surveying of prehistoric barrows through LiDAR. Quaternary International,435, 135–145. https://doi.org/10.1016/j.quaint.2015.12.099 Cerrillo-Cuenca, E., & Bueno Ramírez, P. (2019). Counting with the invisible record? The role of LiDAR in the interpretation of megalithic landscapes in south-western Iberia (Extremadura, Alentejo and Beira Baixa). Archaeological Prospection,26,1–14. https://doi.org/10.1002/ arp.1738 Cerrillo-Cuenca, E., & L opez L opez, A. (2020). Evaluaci on y perspectivas del uso del LiDAR en la arqueología española. Boletín del Museo Arqueol ogico Nacional,39, 221–238. Challis, K., Kokalj, Z., Kincey, M., Moscrop, D., & Howard, A. J. (2008). Airborne LiDAR and historic environment records. Antiquity,82(318), 1055–1064. https://doi.org/10.1017/S0003598X00097775 Chapman, R. (1990). Emerging Complexity. The Later Prehistory of South- East Spain, Iberia and the West Mediterranean. Cambridge University Press. https://doi.org/10.1017/CBO9780511735486 Chase, A. F., Chase, D. Z., Weishampel, J. F., Drake, J. B., Shrestha, R. L., Slatton, K. C., Awe, J. J., & Carter, W. E. (2010). Airborne LiDAR, archaeology, and the ancient Maya landscape at Caracol, Belize. Journal of Archaeological Science,38, 387–398. https://doi.org/10.1016/j. jas.2010.09.018 Corsi, C., Slapsak, B., & Vermeulen, F. (Eds.) (2013). Good Practice in Archaeological Diagnostics. Non-Invasive Survey of Complex Archaeological Sites. Springer. https://doi.org/10.1007/978-3-319- 01784-6 Costa-García, J. M., & Fonte, J. (2017). Scope and limitations of airborne LiDAR technology for the detection and analysis of Roman military sites in Northwest Iberia. In V. M. Herrera, C. Parcero-Oubiña, & P. Fábrega- ´ Alvarez (Eds.), Archaeology and Geomatics. Harvesting the S´ ANCHEZ DÍAZ ET AL.17 Benefits of 10 Years of Training in the Iberian Peninsula (2006–2015) (pp. 57–73). Sidestone Press. Costa-García, J. M., Fonte, J., Gago, M., Menéndez Blanco, A., & ´ Alvarez Martínez, V. (2017). Hallazgos arqueol ogicos recientes para el estudio de la presencia militar romana en el oriente gallego. Gallaecia: revista de arqueoloxía e antigüidade,35,39–70. https://doi.org/10.15304/gall. 35.4086 Crutchley, S. (2013). Using lidar data - drawing on 10 year-experience at English Heritage. In R. Opitz & D. Cowley (Eds.), Interpreting Archaeological Topography: Airborne Laser Scanning, 3D Data and Ground Observation (pp. 136–145). Oxbow Books. https://doi.org/10.2307/j. ctvh1dqdz.16 Daxer, C. (2020). Topographic Openness Maps and Red Relief Image Maps in QGIS. Institute of Geology. Doneus, M. (2013). Openness as visualization technique for interpretative mapping of Airborne Lidar Derived Digital Terrain Models. Remote Sensing,5, 6427–6442. https://doi.org/10.3390/rs5126427 Doneus, M., & Briese, C. (2006). A digital terrain modelling for archaeological interpretation within forested areas using full-waveform laser scanning. In M. Ioannides, D. Arnold, F. Niccolucci, & K. Mania (Eds.), The 7th International Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST, Aire-La-Ville 2006 (pp. 155–162). Doneus, M., & Kühteiber, T. (2013). Airborne laser scanning and archaeological interpretation - bringing back the people. In R. Opitz & D. Cowley (Eds.), Interpreting Archaeological Topography: Airborne Laser Scanning, 3D Data and Ground Observation (pp. 32–50). Oxbow Books. https://doi.org/10.2307/j.ctvh1dqdz.8 Doneus, M., Mandlburger, G., & Doneus, N. (2020). Archaeological ground point filtering of airborne laser scan derived point-clouds in a difficult Mediterranean environment. Journal of Computer Applications in Archaeology,3(1), 92–108. https://doi.org/10.5334/jcaa.44 Eiroa, J. J. (2004). La Edad del Bronce en Murcia. Real Academia Alfonso X El Sabio. Enríquez Navascues, J. J. (1990). El Calcolitico o Edad del Cobre de la cuenca extremeña del Guadiana: los poblados. Museo Arqueol ogico Provincial. Evans, D. (2016). Airborne laser scanning as a method for exploring longterm socio-ecological dynamics in Cambodia. Journal of Archaeological Science,74, 164–175. https://doi.org/10.1016/j.jas.2016.05.009 Evans, D., Fletcher, R., Pottier, C., Chevance, J. B., & Boornazian, G. (2013). Uncovering archaeological landscapes at Angkor using lidar. Proceedings of the National Academy of Sciences of the United States of America, 110(31), 1–6. https://doi.org/10.1073/pnas.1306539110 Fernandez-Diaz, J., Carter, W., Shrestha, R., & Glennie, C. (2014). Now You See It …now you dont: understanding airborne mapping LiDAR collection and data product generation for archaeological research in Mesoamerica. Remote Sensing,6, 9951–10001. https://doi.org/10. 3390/rs6109951 Fernández-Posse, M. D., Gilman, A., Martín, C., & Brodsky, M. (2007). Las Comunidades Agrarias de la Edad del Bronce en La Mancha Oriental (Albacete). Biblioteca Praehistorica Hispana 25. CSIC. Fontana, G. (2022). Italy's hidden hillforts: a large-scale Lidar-based mapping of Samnium. Journal of Field Archaeology,47, 245–261. https:// doi.org/10.1080/00934690.2022.2031465 Gallagher, J. M., & Josephs, R. L. (2008). Using LiDAR to detect cultural resources in a forested environment: An example from Isle Royale National Park, Michigan, USA. Archaeological Prospection,15, 187–206. https://doi.org/10.1002/arp.333 García Huerta, R., & Morales Hervás, F. J. (Eds.) (2004). La Península Ibérica en el II milenio A. C.: Poblados y Fortificaciones. Universidad de Castilla- La Mancha. García Sanjuán, L. (1999). Los Orígenes de la Estratificaci on Social. Patrones de Desigualdad en la Edad del Bronce del Suroeste de la Península Ibérica (Sierra Morena Occidental c. 1700–1100 ane/2100–1300 ANE). British Archaeological Reports International Series S823. Archaeopress. García Sanjuán, L., Hurtado Pérez, V., & Márquez Pérez, J. (2011). El marco territorial de El Trastej on en el contexto de las estrategias de asentamiento en la Edad del Bronce de Sierra Morena occidental. In V. H. Pérez, L. G. Sanjuán, & M. H. Ortiz (Eds.), El Asentamiento de El Trastej on (Huelva). Investigaciones en el Marco de los Procesos Sociales y Culturales de la Edad del Bronce en el Suroeste de la Península Ibérica (pp. 279–316). Junta de Andalucía. García Sanjuán, L., & Murillo-Barroso, L. (2013). Social Complexity in Copper Age Southern Iberia (c. 3200–2200 cal BC): reviewing the ‘state’ hypothesis at Valencina de la Concepci on (Seville, Spain). In M. C. Berrocal, L. G. Sanjuán, & A. Gilman (Eds.), The Prehistory of Iberia: Debating Early Social Stratification and the State (pp. 119–140). Routledge. Gillings, M., Hacigüzeller, P., & Lock, G. (2020). Archaeological Spatial Analysis: A Methodological Guide. Routledge. https://doi.org/10.4324/ 9781351243858 Gilman, A., & Thornes, J. B. (1985). Land-Use and Prehistory in South-east Spain. George Allen & Unwin. Gonçalves, V., Sousa, A. C., & Costeira, C. (2013). Walls, gates and towers. Fortified settlements in the south and centre of Portugal: some notes about violence and walls in the 3rd millennium BCE. Cuadernos de prehistoria y arqueología de la Universidad de Granada,23,35–97. Guyot, A., Lennon, M., Lorho, T., & Hubert-Moy, L. (2021). Combined Detection and Segmentation of Archeological Structures from LiDAR Data Using a Deep Learning Approach. Journal of Computer Applications in Archaeology,4(1), 1–19. https://doi.org/10.5334/jcaa.64 Harmon, J. M., Leone, M. P., Prince, S. D., & Snyder, M. (2006). LiDAR for archaeological landscape analysis: a case study of two eighteenthcentury Maryland plantation sites. American Antiquity,71(4), 649–670. https://doi.org/10.2307/40035883 Historic England. (2018). Using Airborne LiDAR in Archaeological Survey: The Light Fantastic. Historic England. Hurtado Pérez, V. (2004). El asentamiento fortificado de San Blas (Cheles, Badajoz). III milenio a.C. Trabajos de Prehistoria,61(1), 141–155. Hurtado Pérez, V., García Sanjuán, L., & Hunt Ortiz, M. (Eds.) (2011). El Asentamiento de El Trastej on (Huelva). Investigaciones en el Marco de los Procesos Sociales y Culturales de la Edad del Bronce en el Suroeste de la Península Ibérica. Junta de Andalucía. Hurtado Pérez, V., García Sanjuán, L., Mondéjar Fernández de Quincoces, P., & Romero Bomba, E. (2011). El asentamiento de La Papúa II. In V. H. Pérez, L. G. Sanjuán, & M. H. Ortiz (Eds.), El Asentamiento de El Trastej on (Huelva). Investigaciones en el Marco de los Procesos Sociales y Culturales de la Edad del Bronce en el Suroeste de la Península Ibérica (pp. 113–137). Junta de Andalucía. Hurtado Pérez, V., & Hunt Ortiz, M. (1999). Extremadura. In G. Delibes & I. Montero Ruiz (Eds.), Las Primeras Culturas Metalúrgicas de la Península Ibérica (pp. 241–274). Instituto Universitario Ortega y Gasset. Hurtado Pérez, V., Romero Bomba, E., & Rivera Jiménez, T. (2011). La Atalaya del Trastej on. In V. Hurtado Pérez, L. García Sanjuán, & M. Hunt Ortiz (Eds.), El Asentamiento de El Trastej on (Huelva). Investigaciones en el Marco de los Procesos Sociales y Culturales de la Edad del Bronce en el Suroeste de la Península Ibérica (pp. 105–112). Junta de Andalucía. Instituto Geográfico Nacional. (2014). Especificaciones Técnicas para la realizaci on del vuelo lidar que permita la obtenci on de informaci on altimétrica en el ámbito de Andalucía. Instituto Geográfico Nacional. Johnson, K. M., & Ouimet, W. B. (2018). An observational and theoretical framework for interpreting the landscape palimpsest through airborne LiDAR. Applied Geography,91,32–44. https://doi.org/10.1016/j. apgeog.2017.12.018 Jorge, S. O. (2003). Revisiting some earlier papers on the late prehistoric walled enclosures of the Iberian Peninsula. Journal of Iberian Archaeology,5,89–136. 18 S´ ANCHEZ DÍAZ ET AL. Jover Maestre, F. J., L opez Mira, J. A., & L opez Padilla, J. A. (1995). El Poblamiento Durante el II milenio a. C. en Villena (Alicante). Ayuntamiento de Villena. Kokalj, Z., & Hesse, R. (2017). Airborne Laser Scanning Raster Data Visualization: A Guide to Good Practice. Založba ZRC. https://doi.org/10.3986/ 9789612549848 Kokalj, Z., & Somrak, M. (2019). Why Not a Single Image? Combining Visualizations to Facilitate Fieldwork and On-Screen Mapping. Remote Sensing,11, 747–772. https://doi.org/10.3390/rs11070747 Lasaponara, R., & Masini, N. (Eds.) (2012). Satellite Remote Sensing; A New Tool for Archaeology. Springer. https://doi.org/10.1007/978-90-481- 8801-7 Legarra Herrero, B. (2014). Estructura territorial y estado en la cultura argárica. Menga,4, 149–171. Limp, W. F. (1989). The Use of Multispectral Digital Imagery in Archaeological Investigations. Research Series 24. Arkansas Archaeological Survey. Lorite Martínez, S., Ojeda Manrique, J. C., Rodríguez-Cuenca, B., González Crist obal, E., & Muñoz, P. (2017). Procesado y distribuci on de nubes de puntos en el proyecto PNOA-LiDAR. XVII Congreso de la Asociaci on Española de Teledetecci on (Murcia, 2017), 329–332. Lozi c, E., & Štular, B. (2021). Documentation of Archaeology-Specific Workflow for Airborne LiDAR Data Processing. Geosciences,11, 26. https://doi.org/10.3390/geosciences11010026 Lull, V., Mic o, R., Rihuete-Herrada, C., & Risch, R. (2014). The La Bastida fortification: New light and new questions on Early Bronze Age societies in the western Mediterranean. Antiquity,88(340), 395–410. https://doi.org/10.1017/S0003598X00101073 Lyons, T. R. (Ed.) (1981). Remote Sensing: Multispectral Analyses of Cultural Resources, Chaco Canyon and Bandelier National Monument. National Park Service. Mataloto, R., & Boaventura, R. (2009). Entre vivos e mortos nos IV e III milénios a.n.e. do Sul de Portugal. Um balanço relativo do povoamento com base em datações pelo radiocarbono. Revista Portuguesa de Arqueologia,12(2), 31–77. Menéndez Blanco, A., García Sánchez, J., Costa-García, J. M., Fonte, J., González- ´ Alvarez, D., & Vicente García, V. (2020). Following the Roman Army between the Southern Foothills of the Cantabrian Mountains and the Northern Plains of Castile and Le on (North of Spain): Archaeological applications of remote sensing and geospatial tools. Geosciences,10, 485–530. https://doi.org/10.3390/ geosciences10120485 Mlekuž, D. (2013). Skin deep: LiDAR and good practice of landscape archaeology. In C. Corsi, B. Slapšak, & F. Vermeulen (Eds.), Good Practice in Archaeological Diagnostics (pp. 113–129). Springer. https://doi. org/10.1007/978-3-319-01784-6_6 Molina González, F., & Cámara Serrano, J. A. (2004). Urbanismo y fortificaciones en la cultura de El Argar: homogeneidad y patrones regionales. In R. García Huerta & F. J. Morales Hervás (Eds.), La Península Ibérica en el II milenio A. C.: Poblados y Fortificaciones (pp. 9–56). Universidad de Castilla-La Mancha. Montanero Vico, D., & Asensio i Vilar o, D. (2009). Puertas fortificadas del Mediterráneo: Orígenes y evoluci on. Revista dArqueologia de Ponent, 19, 177–204. Monterroso-Checa, A. (2017). Remote sensing and archaeology from Spanish LiDAR-PNOA: Identifying the amphitheatre of the Roman city of Torreparedones (C ordoba-Andalucía-Spain). Mediterranean Archaeology and Archaeometry,17,15–22. Monterroso-Checa, A. (2019). Geoarchaeological characterisation of sites of Iberian and Roman Cordoba using LiDAR data acquisitions. Geosciences,9, 205–221. https://doi.org/10.3390/ geosciences9050205 Monterroso-Checa, A. (2021). La ubicaci on del santuario de Melqart en Gadir: aportaci on de los datos PNOA-LiDAR. SPAL-Revista de Prehistoria y Arqueología,1(30), 137–164. https://doi.org/10.12795/ spal.2021.i30.05 Monterroso-Checa, A., Moreno-Escribano, J. C., Gasparini, M., Conejo- Moreno, J. A., & Domínguez-Jiménez, J. L. (2021). Revealing archaeological sites under mediterranean forest canopy using LiDAR: El Viandar Castle (husum) in El Hoyo (Belmez-C ordoba, Spain). Drones,5, 72–89. https://doi.org/10.3390/drones5030072 Nocete Calvo, F., Lizcano Prestel, R., Nieto, J. M., ´ Alex Tur, E., Inácio, N. M., Rodríguez Bayona, M., Delgado Huertas, A., Orihuela Parrales, A., & Linares Catela, J. A. (2004). La ordenaci on espaciotemporal del registro arqueol ogico de Cabezo Juré. In F. Nocete Calvo (Ed.), Odiel. Proyecto de Investigaci on Arqueol ogica para el Análisis del Origen de la Desigualdad Social en el Suroeste de la Península Ibérica (pp. 129–232). Junta de Andalucía. Opitz, R. (2013). An overview of airborne and terrestrial laser scanning in archaeology. In R. Opitz & D. Cowley (Eds.), Interpreting Archaeological Topography: Airborne Laser Scanning, 3D Data and Ground Observation (pp. 13–31). Oxbow Books. https://doi.org/10.2307/j.ctvh1dqdz.7 Opitz, R., & Cowley, D. (2013). Interpreting Archaeological Topography: Airborne Laser Scanning, 3D Data and Ground Observation. Oxbow Books. https://doi.org/10.2307/j.ctvh1dqdz Palmer, R., & Cox, C. (1993). Uses of Aerial Photography in Archaeological Evaluations. Institute of Field Archaeologists. Parcero-Oubiña, C. (2021). Habelos, hainos. Detecci on remota de indicios arqueol oxicos mediante fotografía aérea e LiDAR en castros de Galicia. Cuadernos de Estudios Gallegos,134,15–45. https://doi.org/10.3989/ ceg.2021.134.01 Pav on Soldevila, I., & Duque Espino, D. (2014). El Cerro del Castillo de Alange (Extremadura, España): Un paisaje de la Edad del Bronce. In R. Vilaça & M. Serra (Eds.), Idade do Bronze do Sudoeste. Novas Perspectivas sobre una Velha Problemática (pp. 51–74). Universidade de Coimbra. Pedro Mich o, M. J., & Martí Oliver, B. (2004). Los poblados de la cultura del Bronce Valenciano. In R. García Huerta & F. J. Morales Hervás (Eds.), La Península Ibérica en el II milenio A. C.: Poblados y Fortificaciones (pp. 299–333). Universidad de Castilla-La Mancha. Pérez Macías, J. A., Macías Forte, R., & Rabadán Vázquez, M. (2019). Notas sobre la Edad del Bronce en el Andévalo (Huelva, España). Setúbal Arqueol ogica,18, 105–120. Piccarreta, F., & Ceraudo, G. (2000). Manuale di Aerofotografia Archeologica. Metodologia, Techniche e Applicazioni. Edipuglia. Popovic, D., Govedarica, M., Jovanovic, D., Radulovic, A., & Simeunovic, V. (2017). 3D Visualization of Urban Area Using Lidar Technology and CityGML. IOP Conference Series: Earth and Environmental Science, 95(4), 1–10. https://doi.org/10.1088/1755-1315/95/4/042006 Prümers, H., Betancourt, C. J., Iriarte, J., Robinson, M., & Schaich, M. (2022). Lidar reveals pre-Hispanic low-density urbanism in the Bolivian Amazon. Nature,606(7913), 325–328. https://doi.org/10.1038/ s41586-022-04780-4 Riley, D. N. (1987). Air Photography and Archaeology. Duckworth. Ringle, W. M., Gallareta, T., May Ciau, R., Seligson, K. E., Fernández- Diaz, J. C., & Orteg on Zapata, D. (2021). Lidar survey of ancient Maya settlement in the Puuc region of Yucatan, Mexico. PLoS ONE,16(4), e0249314. https://doi.org/10.1371/journal.pone.0249314 Risbøl, O. (2010). Towards an improved archaeological record through the use of airborne laser scanning. Third International Conference on Remote Sensing in Archaeology, BAR International Series 2118, 105–112. Risbøl, O., & Gustavsen, L. (2018). LiDAR from drones employed for mapping archaeology: potential, benefits and challenges. Archaeological Prospection,25, 329–338. https://doi.org/10.1002/arp.1712 Romero Bomba, E., & Rivera Jiménez, T. (2008). El asentamiento fortificado del Cerro del Almendro (Santa Olalla del Cala, Huelva). Vipasca,2, 128–138. Rouco Collazo, J., Benavides, J. A., & Civantos, J. M. (2020). Falling from the sky. Aerial photogrammetry and LiDAR applied to the archaeology of architecture and landscape: two fortifications from the Alpujarra S´ ANCHEZ DÍAZ ET AL.19 (Granada, Spain). In M. Ramírez Galán & R. Sandiferd Bard (Eds.), Studies in Archaeometry. Proceedings of the Archaeometry Symposium at NORM 2019, Portland, Oregon (pp. 87–174). Archaeopress. https:// doi.org/10.2307/j.ctv1zckxdm.7 Ruiz Taboada, A. (1997). Asentamiento y subsistencia en la Mancha durante la Edad del Bronce. Complutum,8,57–71. Sánchez Díaz, F., & Hunt Ortiz, M. (2021). Hipsometría mediante LiDAR del territorio calcolítico de La Zarcita. XI Encontro de Arqueologia do Sudoeste Peninsular, Loulé,2021,1–16. Siret, E., & Siret, L. (1890). Las Primeras Edades del Metal en el Sudeste de España. Henrich y Cía. Štular, B., Eichert, S., & Lozi c, E. (2021). Airborne LiDAR Point Cloud Processing for Archaeology. Pipeline and QGIS Toolbox. Remote Sensing,13(16), 3225. https://doi.org/10.3390/rs13163225 Tarradell, M. (1961). Sobre la identificaci on de los poblados eneolíticos valencianos. In VI Congreso Nacional de Arqueología (Oviedo 1959) (pp. 86–91). Universidad de Zaragoza. Verschoof-van der Vaart, W. B., Karsten Lambers, K., Wojtek Kowalczyk, W., & Bourgeois, Q. (2020). Combining deep learning and location-based ranking for large-scale archaeological prospection of LiDAR data from The Netherlands. International Journal of Geo-Infor- mation,9(5), 293–314. https://doi.org/10.3390/ijgi9050293 Wheatley, D. W., & Gillings, M. (2013). Spatial Technology and Archaeology. The Archaeological Application of GIS. Taylor & Francis. https://doi.org/ 10.1201/b12806 Wilson, D. R. (Ed.) (1982). Air Photo Interpretation for Archaeologists. Batsford. Wiseman, J., & El-Baz, F. (2007). Remote Sensing in Archaeology. Springer. https://doi.org/10.1007/0-387-44455-6 How to cite this article: Sánchez Díaz, F., García Sanjuán, L., & Rivera Jiménez, T. (2022). Potential and limitations of LiDAR altimetry in archaeological survey. Copper Age and Bronze Age settlements in southern Iberia. Archaeological Prospection, 1–20. https://doi.org/10.1002/arp.1869 20 S´ ANCHEZ DÍAZ ET AL.