Arnau Garcia

This paper presents new developments on drone‐based automated survey for the detection of individual items or fragments of material culture visible on the ground surface. Since the publication of our original proof of concept, awarded with the Journal of Archaeological Science and Society for Archaeological Sciences Emerging Investigator Award 2019, additional funding has allowed us to implement a series of improvements to the method. These aim to improve detection capabilities and the extraction of items' shapes and increase flight autonomy, control, area covered per flight and the type of environments in which the method can be applied while reducing computing needs, processing time and expertise necessary for its application. This paper provides an account of the methods followed to achieve these objectives, their preliminary results and the current development for their implementation into a free and open‐source system that can be used by the archaeological community at large.

Historical maps present a unique depiction of past landscapes, providing evidence for a wide range of information such as settlement distribution, past land use, natural resources, transport networks, toponymy and other natural and cultural data within an explicitly spatial context. Maps produced before the expansion of large-scale mechanized agriculture reflect a landscape that is lost today. Of particular interest to us is the great quantity of archaeologically relevant information that these maps recorded, both deliberately and incidentally. Despite the importance of the information they contain, researchers have only recently begun to automatically digitize and extract data from such maps as coherent information, rather than manually examine a raster image. However, these new approaches have focused on specific types of information that cannot be used directly for archaeological or heritage purposes. This paper provides a proof of concept of the application of deep learning techniques to extract archaeological information from historical maps in an automated manner. Early twentieth century colonial map series have been chosen, as they provide enough time depth to avoid many recent large-scale landscape modifications and cover very large areas (comprising several countries). The use of common symbology and conventions enhance the applicability of the method. The results show deep learning to be an efficient tool for the recovery of georeferenced, archaeologically relevant information that is represented as conventional signs, line-drawings and text in historical maps. The method can provide excellent results when an adequate training dataset has been gathered and is therefore at its best when applied to the large map series that can supply such information. The deep learning approaches described here open up the possibility to map sites and features across entire map series much more quickly and coherently than other available methods, opening up the potential to reconstruct archaeological landscapes at continental scales.

This paper presents an innovative multisensor, multitemporal machine-learning approach using remote sensing big data for the detection of archaeological mounds in Cholistan (Pakistan). The Cholistan Desert presents one of the largest concentrations of Indus Civilization sites (from ca. 3300 to 1500 BC). Cholistan has figured prominently in theories about changes in water availability , the rise and decline of the Indus Civilization, and the transformation of fertile monsoonal alluvial plains into an extremely arid margin. This paper implements a multisensor, multitemporal machine-learning approach for the remote detection of archaeological mounds. A classifier algorithm that employs a large-scale collection of synthetic-aperture radar and multispectral images has been implemented in Google Earth Engine, resulting in an accurate probability map for mound-like signatures across an area that covers ca. 36,000 km 2. The results show that the area presents many more archaeological mounds than previously recorded, extending south and east into the desert, which has major implications for understanding the archaeological significance of the region. The detection of small (<5 ha) to large mounds (>30 ha) suggests that there were continuous shifts in settlement location. These shifts are likely to reflect responses to a dynamic and changing hydrological network and the influence of the progressive northward advance of the desert in a long-term process that culminated in the abandon-ment of much of the settled area during the Late Harappan period. multitemporal and multisensor satellite big data | machine learning | archaeology | Indus Civilization | virtual constellations A rtificial mounds are a characteristic feature of permanent and semipermanent settlement locations in past cultural landscapes, particularly on sedimentary plains, but also in arid and semiarid regions. These mounds can be readily visible due to their prominence and shape, and the fact that they are composed of accumulated debris such as mud bricks and pottery sherds, which creates specific soil with distinct color and surface texture. These characteristics make them detectable using different methods, and their number and distribution have seen them play an important role in addressing questions about the formation of early urbanism, states, and economic systems. The use of remote sensing (RS) to detect and map archaeological mounds has been attempted in many parts of the world (1-4). Much research has focused on arid and semiarid areas in the Levant and the Near East, where the geomorphological and sedimentary properties of mounds make them highly visible in digital elevation models and aerial and satellite imagery (5). Mounds can also leave specific multispectral soil signatures in highly anthropized landscapes with leveled or irrigated fields (6). When available, the use of declassified historical photographs such as CORONA imagery has been critical to the detection of mounds (7-9). Georeferenced historical map series have also been used solely or in combination with contemporary declas-sified data (10-14). In recent years, RS-based archaeological research has gradually incorporated machine-learning techniques and algorithms that facilitate the automated detection of sites and features. Most of those applications have focused on the detection of small-scale features using high-resolution data-sets such as lidar (15) or WorldView imagery (16-18). In the Near East, Menze and Ur (19) applied a random forest (RF) classifier over a multitemporal collection of ASTER imagery to identify probable anthrosols. Some other attempts have used multitemporal data to monitor archaeological sites and human impact such as urban sprawl and looting (20-22). The detection of anthropic signatures, such as those that characterize mounded sites, across a very large area, remains seldom attempted, presumably due to the large computational resources, coding expertise , and large amount of satellite data required. Significance This paper illustrates the potential of machine learning-based classification of multisensor, multitemporal satellite data for the remote detection and mapping of archaeological mounded settlements in arid environments. Our research integrates multitemporal synthetic-aperture radar and multispectral bands to produce a highly accurate probability field of mound signatures. The results largely expand the known concentration of Indus settlements in the Cholistan Desert in Pakistan (ca. 3300 to 1500 BC), with the detection of hundreds of new sites deeper in the desert than previously suspected including several large-sized (>30 ha) urban centers. These distribution patterns have major implications regarding the influence of climate change and desertification in the collapse of the largest of the Old-World Bronze Age civilizations.

This paper presents an innovative multisensor, multitemporal machine-learning approach using remote sensing big data for the detection of archaeological mounds in Cholistan (Pakistan). The Cholistan Desert presents one of the largest concentrations of Indus Civilization sites (from ca. 3300 to 1500 BC). Cholistan has figured prominently in theories about changes in water availability , the rise and decline of the Indus Civilization, and the transformation of fertile monsoonal alluvial plains into an extremely arid margin. This paper implements a multisensor, multitemporal machine-learning approach for the remote detection of archaeological mounds. A classifier algorithm that employs a large-scale collection of synthetic-aperture radar and multispectral images has been implemented in Google Earth Engine, resulting in an accurate probability map for mound-like signatures across an area that covers ca. 36,000 km 2. The results show that the area presents many more archaeological mounds than previously recorded, extending south and east into the desert, which has major implications for understanding the archaeological significance of the region. The detection of small (<5 ha) to large mounds (>30 ha) suggests that there were continuous shifts in settlement location. These shifts are likely to reflect responses to a dynamic and changing hydrological network and the influence of the progressive northward advance of the desert in a long-term process that culminated in the abandon-ment of much of the settled area during the Late Harappan period. multitemporal and multisensor satellite big data | machine learning | archaeology | Indus Civilization | virtual constellations A rtificial mounds are a characteristic feature of permanent and semipermanent settlement locations in past cultural landscapes, particularly on sedimentary plains, but also in arid and semiarid regions. These mounds can be readily visible due to their prominence and shape, and the fact that they are composed of accumulated debris such as mud bricks and pottery sherds, which creates specific soil with distinct color and surface texture. These characteristics make them detectable using different methods, and their number and distribution have seen them play an important role in addressing questions about the formation of early urbanism, states, and economic systems. The use of remote sensing (RS) to detect and map archaeological mounds has been attempted in many parts of the world (1-4). Much research has focused on arid and semiarid areas in the Levant and the Near East, where the geomorphological and sedimentary properties of mounds make them highly visible in digital elevation models and aerial and satellite imagery (5). Mounds can also leave specific multispectral soil signatures in highly anthropized landscapes with leveled or irrigated fields (6). When available, the use of declassified historical photographs such as CORONA imagery has been critical to the detection of mounds (7-9). Georeferenced historical map series have also been used solely or in combination with contemporary declas-sified data (10-14). In recent years, RS-based archaeological research has gradually incorporated machine-learning techniques and algorithms that facilitate the automated detection of sites and features. Most of those applications have focused on the detection of small-scale features using high-resolution data-sets such as lidar (15) or WorldView imagery (16-18). In the Near East, Menze and Ur (19) applied a random forest (RF) classifier over a multitemporal collection of ASTER imagery to identify probable anthrosols. Some other attempts have used multitemporal data to monitor archaeological sites and human impact such as urban sprawl and looting (20-22). The detection of anthropic signatures, such as those that characterize mounded sites, across a very large area, remains seldom attempted, presumably due to the large computational resources, coding expertise , and large amount of satellite data required. Significance This paper illustrates the potential of machine learning-based classification of multisensor, multitemporal satellite data for the remote detection and mapping of archaeological mounded settlements in arid environments. Our research integrates multitemporal synthetic-aperture radar and multispectral bands to produce a highly accurate probability field of mound signatures. The results largely expand the known concentration of Indus settlements in the Cholistan Desert in Pakistan (ca. 3300 to 1500 BC), with the detection of hundreds of new sites deeper in the desert than previously suspected including several large-sized (>30 ha) urban centers. These distribution patterns have major implications regarding the influence of climate change and desertification in the collapse of the largest of the Old-World Bronze Age civilizations.

Incomplete datasets curtail the ability of archaeologists to investigate ancient landscapes, and there are archaeological sites whose locations remain unknown in many parts of the world. To address this problem, we need additional sources of site location data. While remote sensing data can often be used to address this challenge, it is enhanced when integrated with the spatial data found in old and sometimes forgotten sources. The Survey of India 1" to 1-mile maps from the early twentieth century are one such dataset. These maps documented the location of many cultural heritage sites throughout South Asia, including the locations of numerous mound features. An initial study georeferenced a sample of these maps covering northwest India and extracted the location of many potential archaeological sites-historical map mound features. Although numerous historical map mound features were recorded, it was unknown whether these locations corresponded to extant archaeological sites. This article presents the results of archaeological surveys that visited the locations of a sample of these historical map mound features. These surveys revealed which features are associated with extant archaeological sites, which were other kinds of landscape features, and which may represent archaeological mounds that have been destroyed since the maps were completed nearly a century ago. Their results suggest that there remain many unreported cultural heritage sites on the plains of northwest India and the mound features recorded on these maps best correlate with older archaeological sites. They also highlight other possible changes in the large-scale and long-term distribution of settlements in the region. The article concludes that northwest India has witnessed profound changes in its ancient settlement landscapes, creating in a long-term sequence of landscapes that link the past to the present and create a foundation for future research and preservation initiatives.

This paper explores the historical inundation of the city of Dera Ghazi Kkan (Punjab, Pakistan) in 1909. The rich documentation about this episode available—including historic news reports, books and maps—is used to reconstruct the historical dynamics between an urban settlement and the river morphodynamics in the Indus alluvial plain. Map and document-based historical regressive analysis is complemented with the examination of images obtained through different Remote Sensing techniques, including the use of new algorithms specifically developed for the study of topography and seasonal water availability which make possible to assess long-term changes in the Indus River basin. This case of study provides an opportunity to examine: (1) how historical hydrological dynamics are reflected in RS produced images; (2) the implications of river morphodynamics in the interpretation of settlement patterning; and (3) the documented socio-political responses to such geomorphological change. The results of this analysis are used to consider the long-term dynamics that have influenced the archaeo/cultural landscapes of the Indus River basin. This assessment provides critical insights for: (1) understanding aspects of the formation, preservation of representation of the archaeological record; (2) identifying traces of morphodynamics and their possible impact over the cultural heritage; and (3) offering insights into the role that recent historical documents can have in the interpretation of RS materials. This paper should be read in conjunction with the paper by Cameron Petrie et al. in the same issue of Geosciences, which explores the Survey of India 1” to 1-mile map series and outlines methods for using these historical maps for research on historical landscapes and settlement distribution

For the first time, we offer the edition of a latin hexameter considered by the specialists as apocryphal of Eugenius of Toledo: first hexameter of Appendix's poem 26, In baculo. The text has been found in an inscription from Aiguafreda (Catalonia), which can be dated in the VII th century by its archaeological context. We connect the archaeological, paleographic and epigraphic data in order to propose a renewed point of view of a text whom nature has been only known through manuscripts written some centuries after the so called spurious au-tor's death.

The territorial organisation of rural areas in the context of the Roman Empire is a widely discussed issue in archaeomorphological research but it is also a topic of great interest for the study of long-term landscapes dynamics. Most of these studies deal with large territorial management works related to known ancient cities. In this paper we propose a discussion about a study case carried out on a much more reduced scale, focused in a small river valley, placed in an inland area near Barcelona in the NorthEastern sector of the Iberian Peninsula. In this research both the analysis of road network from aerial photographs and the study of settlement patterns from the Archaeological Heritage database are complemented by fieldwork, survey and the regressive analysis of historical maps. A geomorphological study of the area is actually in progress. A significant relationship has been highlighted between an ancient Roman road and an intensive settlement network along this historical road axis. Settlement distribution along the studied stretch seems to show a pattern related to well-known measures used in Roman land-surveys. According to the chronology of these sites, this spatial organisation may coincide with the consolidation processes of a Roman urban network at the Tarra-conensis province. This process took place during the first century BC, especially from the second half of that century. In this paper, the results will be discussed and analysed in order to understand how far the data available allow us to interpret landscape organisation in relation to geomorphological and environmental dynamics.

Free access link:
http://authors.elsevier.com/a/1RysK15SlTUT0H

This paper introduces a novel workflow for the reconstruction of nowadays disappeared cultural landscapes based on the extraction of morphological information from historic aerial photographs. This methodology has been applied for the first time for the detection, classification and characterisation of upstanding, flattened and buried archaeological sites and various off-site ancient landscape features in the plain of Karditsa, western Thessaly. Although Thessaly has been the focus of prehistoric, and especially Neolithic, research in Greece, since the beginning of the 20th century, western Thessaly has not received as much archaeological attention and its archaeological record remains rather scanty. Moreover, an extensive land reclamation project implemented in the western Thessalian plain during the early 1970s resulted in the flattening of habitation tells and funerary sites of all periods. Thus, recognition of archaeological sites and relict landscape features becomes extremely difficult, whereas standard landscape analysis and application of mainstream Remote Sensing (RS) techniques based on multispectral satellite images are problematic.

Digital photogrammetric reconstruction techniques and the subsequent GIS-based treatment of the results allowed overcoming these challenging limitations: the combined use of pre-1970s aerial photographs with later imagery provided a powerful means to reconstruct the landscape before the land reclamation process, using a workflow designed to highlight photogrammetry-derived topographic differences and multi-temporal imagery analysis.

Hundreds of previously unknown mounded archaeological sites, as well as other ancient landscape traits such as roads, city grids and field systems were detected. More importantly, invaluable insights into the type and character of these archaeological features were gained, which would have been impossible to obtain by conventional RS techniques.

Landscape Archaeology in the surrounding of Lauro: A micro-regional approach to Roman territorial shaping in a north-eastern sector of the Iberian Peninsula:

The link between the place-name Lauro and the ancient history of the eastern part of Vallès region (Barcelona) is a constant throughout historiography. In this paper, this ancient Lauro serves as a guide to present the results of a research conducted in a small area, which due to archaeological data complexity has been of great interest to study the dynamics of territorial structuration in one of the first rural areas conquered by Rome outside the Italian peninsula.
From Landscape Archaeology conceptual and methodological tools, extensive and intensive archaeological surveys have been developed. The collected materials and the documented structural remains have allowed us to define settlement and human occupation in a chronology ranging from the development of indigenous (Iberian) societies to the consolidation of the Roman Empire –IVth BC-IIth AD–.
The results of these works set out the elements to analyse the impact of Roman conquest in settlement evolution at a micro-regional scale. The studied area may represent a model to contrast in other regions and to discuss in future works.
Key words: Cultural landscapes; archaeological surveys; settlement dynamics; Late Iron Age period; Roman
period; Eastern Vallès.

This contribution will present the first results of the MSCA-IF funded project Water management strategies and climate changes in the Indus civilization (WaMstrIn). This project participates in a coordinated program carried out by researchers from the McDonald Institute for Archaeological Research with the aim to develop consistent methodologies for the use of big sets of multi-temporal satellite images in the study of prehistoric and historical landscapes. The specific object of study is the relationship between the settlement dynamics and the changing hydrographic network during the Indus Civilisation, first urban culture in South Asia (3300-1900BC).
The methodological approach is based on the analysis of large repositories of multi-temporal multi-spectral satellite images and other types of satellite-derived data, using purposefully created algorithms and parallel cloud computing applications for the detection of (1) paleo-channels and (2) signatures of disappeared human settlements. These data will be complemented with the creation of a geodatabase of historical sources, which will serve to validate the results obtained and support the implementation of machine-learning processes.
WaMStrIn project is applying this methodological approach in the Indus middle basin (historical region of Punjab, Eastern Pakistan and North-western India). It represents one of the world’s most productive agricultural areas, capable of sustaining large populations, including the Bronze Age Indus cities. The poster will discuss the potentialities and difficulties in using large multi-temporal datasets and cloud computing in highly anthropized landscapes affected by important transformations in recent periods.

Archaeological excavations at Aiguafreda de Dalt are the first stage of the research project Paisatges Culturals de l’Alt Congost (PAICCONGOST) on archaeology of cultural landscapes and knowledge sharing , to be developed in 2014-2017.
The ancient parish church of Sant Martí d’Aiguareda was built on a travertine deposit at the northeast of Aiguafreda (Barcelona province), at a height of 600 m. According to the latest archaeological findings, the church history may have begun at some time between the end of the Late Antiquity and the beginning of the Early Middle Ages.
The church was built on a massif formed by travertine deposits built up from the stagnation and sedimentation of mineral springs from the cliffs of “Les Queredes”. In the same area we can also find some small waterfalls which create travertine formations. 
Inside this travertine massif there is a cave that had been used as a crypt, and it was probably related to an ‘ecclesia’ built before 898. The new parish church dedicated to Sant Martí, placed on top of the massif and the crypt, was consecrated the same year. The consecration could have involved a reorganization of the burial space which would have finished in 1105, when the new Romanesque church with a sacred graveyard –the ‘sacraria’ (‘cimiterium triginta passuum’) – was consecrated. This new association between the worship area and the burial area was the main feature of the church until the mid nineteenth century.
Between the eighteenth and nineteenth centuries, the cave became a quarry for travertine. The extracted materials would be partly used in the church built on top of the cave. Because of this quarrying, most of the cave disappeared, leaving some spaces with rock-cut tombs out in the open and starting new travertine-formation processes which made an actual humanized landscape seem “natural”. Within the framework of the project a long-term use of the travertine quarry was confirmed through archaeological and documentary research.

This paper explores the historical inundation of the city of Dera Ghazi Kkan (Punjab, Pakistan) in 1909. The rich documentation about this episode available-including historic news reports, books and maps-is used to reconstruct the historical dynamics between an urban settlement and the river morphodynamics in the Indus alluvial plain. Map and document-based historical regressive analysis is complemented with the examination of images obtained through different Remote Sensing techniques, including the use of new algorithms specifically developed for the study of topography and seasonal water availability which make possible to assess long-term changes in the Indus River basin. This case of study provides an opportunity to examine: (1) how historical hydrological dynamics are reflected in RS produced images; (2) the implications of river morphodynamics in the interpretation of settlement patterning; and (3) the documented socio-political responses to such geomorphological change. The results of this analysis are used to consider the long-term dynamics that have influenced the archaeo/cultural landscapes of the Indus River basin. This assessment provides critical insights for: (1) understanding aspects of the formation, preservation of representation of the archaeological record; (2) identifying traces of morphodynamics and their possible impact over the cultural heritage; and (3) offering insights into the role that recent historical documents can have in the interpretation of RS materials. This paper should be read in conjunction with the paper by Cameron Petrie et al. in the same issue of Geosciences, which explores the Survey of India 1" to 1-mile map series and outlines methods for using these historical maps for research on historical landscapes and settlement distribution.

A range of data sources are now used to support the process of archaeological prospection, including remote sensed imagery, spy satellite photographs and aerial photographs. This paper advocates the value and importance of a hitherto under-utilised historical mapping resource-the Survey of India 1" to 1-mile map series, which was based on surveys started in the mid-late nineteenth century, and published progressively from the early twentieth century AD. These maps present a systematic documentation of the topography of the British dominions in the South Asian Subcontinent. Incidentally, they also documented the locations, the height and area of thousands of elevated mounds that were visible in the landscape at the time that the surveys were carried out, but have typically since been either damaged or destroyed by the expansion of irrigation agriculture and urbanism. Subsequent reanalysis has revealed that many of these mounds were actually the remains of ancient settlements. The digitisation and analysis of these historic maps thus creates a unique opportunity for gaining insight into the landscape archaeology of South Asia. This paper reviews the context within which these historical maps were created, presents a method for georeferencing them, and reviews the symbology that was used to represent elevated mound features that have the potential to be archaeological sites. This paper should be read in conjunction with the paper by Arnau Garcia et al. in the same issue of Geosciences, which implements a research programme combining historical maps and a range of remote sensing approaches to reconstruct historical landscape dynamics in the Indus River Basin.