Archeologia e Calcolatori 34.1 - 2023

Introduction to ArcheoFOSS 2022. Proceedings of the 16th International Conference on Open Software, Hardware, Processes, Data and Formats in Archaeological Research (Rome, 22-23 September 2022)

This paper presents an update to an earlier proposal for a standardized open format for archaeological stratigraphy data, the Harris Matrix Data Package, and the accompanying software tool implementation. The update is two-fold: firstly, it aims at a clear separation between data format and the software tool, particularly by defining the data format in more detail and independently from the software used to create or analyze it; secondly, it introduces a new software feature that allows the creation of a new ‘data package’ from scratch. A third issue that was identified is the lack of tools for converting existing data to and from the Harris Matrix Data Package, but this issue is not dealt with in this paper.

The collection of stratigraphic data has been done for years with offline tools like the Microsoft Access software, which was considered a user-friendly tool with the ability to print standardised context sheets (like the Italian ministerial US sheets) directly and ‘without’ the need for technical expertise. It is well known to the scientific community the limitation inherent in this type of approach: the data is not shared online and is not easy to be collaboratively edited; the data is locked within a proprietary format with repercussions on medium- to long-term preservation; and it is not immediately possible to integrate the data with other projects due to a lack of precise reference standard. Despite these issues, the offline approach remained viable in a whole range of situations where it is necessary to have a fast and easy-to-manage database. This contribution starts from the hypothesis that an offline standardised and encapsulated tool in an open format (such as SQLite, LibreOffice Base or MySQL), editable with open source software, can offer additional solution because it is easy to use and disseminate in the form of a free, downloadable template. EMdb aims to collect and manage not only stratigraphic data but also reconstructive unit sheets to cover the need to analyse, interpret and validate scientific hypotheses in the field.

This paper illustrates the usage and potential of the pyArchInit plugin, an open source tool created in Python language for the management and overall analysis of archaeological data on a single georeferenced platform (QGIS). Some of the functionalities of the application are highlighted in relation to the archaeological survey conducted in 2021 within the area of the Roman theater and the former Bosone spinning mill, in the historic center of Fano (PU), in the Marche region. Specifically, at the same time as the stratigraphic archaeological excavation operations were carried out, we proceeded directly on site, with the detailed management of both the identified stratigraphic Units and the finds recovered during the excavation, with direct data entry. In this way, it was possible to deliver to the Superintendence all georeferenced information layers in Gauss Boaga Est (EPSG3004) reference system, manageable in different GIS platforms and easily usable as a scientific research instrument for protection and planning of cultural heritage.

In the framework of the project ‘Castelseprio, centre of power’, the authors began excavating the structure known as Casa Piccoli in 2021. The area, already investigated by Piccoli in the 1970s, presents itself as an interesting case study for the application of an open and integrated solution for the management of stratigraphic data, specifically pyArchInit. Being an academic excavation project and, therefore, characterized by both research and training issues, it was decided to progressively and incrementally include the use of pyArchInit within the documentation protocols on site and post-excavation, over the three years of the permit granted by the Ministry of Culture for the excavations. Master’s degree students who participated in the excavation, at the end of the planned period, will have the basic skills to use the plugin also in a professional environment. At the end of the first two years of implementation, a SWOT analysis will show the results obtained within the site for both training and research purposes.

During the fieldwork season in November 2021-March 2022, the ‘Missione Archeologica della Sapienza nella Penisola Arabica e nel Golfo’ (MASPAG), as part of the research activities supported and financed by the Great Excavations of Sapienza since 2019 and MAECI since 2022, planned and launched a new landscape archaeological project in the Sultanate of Oman. The first survey was carried out in an area of the Al Batinah South Governorate unknown to archaeology, combining remote-sensing and ground verification activities. This operation also saw the first result of the collaboration between the MASPAG research group and adArte srl, developer of pyArchInit open sources plugin for QGIS. The first season of the survey not only made it possible to estimate the archaeological potential of the study area, but also served as a workshop, opening a dialogue between universities and private companies, to discuss open source solutions in archaeology.

In the field of Cultural Heritage, the technological advances of recent years have enriched and optimised the possibility of documenting and studying ancient graffiti with a wide range of low-cost and non-invasive methodologies. The most popular are digital photogrammetry SfM (Structure from Motion) and RTI (Reflectance Transformation Imaging) methodologies. The RTI is a powerful tool that, through the use of open source software, enables the documentation of data that are difficult to visualise, facilitating the recognition of traces and marks on the surface of objects. On the other hand, the SfM 3D models are increasingly replacing documentation with traditional photographs. This ‘almost excessive’ production of three-dimensional models is not often accompanied by an adequate exploitation of all their potential uses. This research aims to investigate the possibility of using a high-resolution 3D model for the implementation of virtual RTI processing, a hybrid method that combines 3D, virtual manipulation and 2D technologies in a fast and intuitive workflow suitable for the documentation of a wide range of archaeological monuments. The process sees the 3D model from the SfM survey being illuminated and photographed in a virtual dome in the open source Blender environment; therefore, the images generated are processed with RTI Builder software.

Building Information Modeling is the most consolidated work method for engineering design of buildings and infrastructural works. It allows to create a comprehensive database starting from a 3D model of a building. Its use in archaeology permits to test and transform a working method, born for engineering design, in an operational support for archaeologists during and after the field phase. Our contribution focuses on the application of BIM to archaeological evidence. It presents the initial stages of a research project, whose aim is the definition of operational solutions for the creation of BIM models. To implement, on a scientific basis, the BIM methodology and make the modelling of archaeological structures easier, a model of semantic library, based on Pompeian archaeological evidence, has been created. The BIM contains archeological objects that can be reproduced and possibly modified for other projects. They represent a support to share on a large scale the representation in BIM. All archaeological library’s objects can be exported in the IFC format. This format can be opened and edited by all BIM software and worked on all OS; the export of archaeological objects in IFC format leads BIM closer to the FOSS world.

In 2018 the Parco Archeologico del Colosseo set out on a three-year basis project, the ‘Risk Map of Floors Surfaces’, with the aim of preserving and monitoring all the in situ floor coverings of the Roman Forum and the Palatine Hill areas (mosaics, sectile, cement floors, spicata). In order to systematically address this methodological approach a team of archaeologists, architects and restorers designed and created a comprehensive and functional information management system, the ‘Risk Map of the Mosaic and Marble Surfaces’, together with a web-based application with integrated webGIS tools. The platform is used daily to record historical-archaeological and archival data and it has become an essential tool in planning interventions in the field. This approach brings the Parco to move from emergency maintenance to a continuous cycle of systematic maintenance. At the end of the first three-year phase of the project, an interactive web map was published online in May 2022 to share selected data related to the ancient floors of the Parco with public users. At this stage, the web map (https://cdrweb.parcocolosseo.it) allows to obtain descriptive texts and a gallery of images of the ancient floors; there are plans in the next future to improve data sharing through API and web map services.

The German Archaeological Institute (Deutsches Archäologisches Institut, DAI) conducts a variety of different types of field research, each with its own unique documentation requirements: excavations, surveys and architectural surveys. The resulting differences are reflected in the workflows, the recording methods and the documentation. In addition, the DAI’s international work has to comply with the guidelines of the respective heritage agencies in the host countries. iDAI.field is the system for documenting archaeological fieldwork at the DAI. From the very beginning it was developed to meet these very different requirements. The development spanned several years, major versions and associated, far-reaching technology changes. The latest iteration of the application relies exclusively on open source technologies and is published on GitHub under the Apache License 2.0 in accordance with DAI-IT’s open source policy. In order to open up the application to other interested researchers and/or developers, the focus of the last year has been the implementation of an extended configuration interface and the removal of dependencies from the DAI infrastructure. This article outlines the development history, introduces the currently available functionalities, and briefly discusses the data model, followed by an overview of the technologies used. It also describes the development into a real open source product and gives a short outlook on the future plans.

The case of city of Ancona highlighted the lack of specific software for the management and digitization of the archaeological data stored in the archives of the Superintendence. The archives contain many heterogeneous data that can help to understand the history of the archaeological sites, from their discovery up to the information archived from the numerous research or rescue archaeology excavations that have taken place over time. The normalization of all the archival data within a single relational database associated with their specific geographical nature, thanks to an overall view and an in-depth review of the data, shed new light on both edited contexts and archaeological evidence that had not yet received an adequate study and that had not been entered into a system that considers nearby data. This software does not replace existing cataloging systems, such as SIGECweb, but it aims to support the cataloging activity using the same standard and at the same time allowing the Superintendence to use the data for the protection activity and for their study.

The Linked Open Ogham Data Project was set up in 2019 by the Research Squirrel Engineers Network and supported by the Wikimedia Germany Open Science Fellows Program in 2020/2021. In 2022 an Ogham survey was done in Ireland to record Ogham stones in the field and museums. The project aims at providing and integrating Ogham Data in community hubs such as Wikidata and Open Street Map (OSM). This paper shows a hybrid Ogham LOD workflow, based on the idea of Open Science, Open Software, Open Data and the FAIR principles to create re-usability and modular IT infrastructure with community standards and commonly-used interfaces. Furthermore, the paper shows examples of Ogham stones from the Dingle and Iveragh Peninsulas in OSM and gives a deeper insight into the inscriptions and mentioned Ogham-specific formula words and names.

This paper proposes an overview of practices ensuring the gradual transition of printed archaeological journals, already internationalised, to new models of online scientific publishing. It also examines the economic and organisational means that guarantee the sustainability of these models. Our two research units, the Pôle éditorial of the Maison des Sciences de l’Homme Mondes and the team Archéologie du monde grec et systèmes d’information of the ArScAn équipe, are both based in the campus of Nanterre, and collaborative discussions between the numerous professionals based there give us a precious glimpse at the evolutions of practices in terms of data management and publication, in a context of Open Science where scholarly publications and data tend to be open more widely and faster.

The number of scientific articles published each year is on the rise, but the current system, which is dominated by a few for-profit publishers, has become prohibitively expensive for many institutions. This model of publishing is increasingly being criticized for its serious flaws. The deposit of preprints in open archives is a solution for the rapid dissemination of research. However, the quality of these preprints must be ensured. This is where Peer Community In (PCI) comes in, by organizing communities of researchers to assess the quality of the work deposited in open archives. In 2020, a PCI dedicated to Archaeology was established, with over 100 archaeologists acting as recommenders. These recommenders handle the submitted preprints as associate editors would in traditional journals, but at the end of the process, they write a recommendation text, and the entire editorial process is published with it. So far, PCI Archaeology has received 45 submissions, mostly pertaining to Prehistoric periods, and from authors located in different regions of the world. This open process has been widely accepted by reviewers, but there is still a need to promote the use of preprints in the community of archaeologists.

Scientific literature understanding benefits from visual resources, which is even more evident in the case of material cultural heritage. In recent years, journals and publishing platforms have been increasingly offering extensive access to publications via the contextual provision of visual media, such as images and 3D models. The diamond open access journal ‘Archeologia e Calcolatori’, founded in 1990, started publishing its articles in 2005 on its website and has always paid attention to giving proper value and presentation to visual contents related to publications. Indeed, it maintained an online image gallery displaying colour plates from volumes until coloured images started being embedded in the articles’ PDFs (since 2009). Then, in 2021, the journal added images and 3D models as resources together with publications and displayed them both as standalone content and in relation to articles. However, this later work did not include the previous thirty-year-long history of the journal, since it required close cooperation with authors. Thus a new dedicated web application was specifically developed to present a structured and visually appealing archive of about 4000 images. The paper illustrates this application, entitled A&C_IADI (Interactive Atlas of Digital Images).

This paper focuses on collaborative methods and open source tools aimed to analyze and query 3D photogrammetric models of ancient architectures. The processing of virtual models led to the constitution of a training dataset of around 1300 wall facing stones from four archaeological sites in Crete. Through a purposely-conceived add-on of the open source software Blender, some algorithms expressed in Python are able to extract archaeologically significant features and to perform processes of Machine Learning and data mining. The resulting data are imported into a dedicated DB managed through a web application based on the open source framework Django. This workflow addresses some peculiar challenges of the application of Artificial Intelligence to archaeological heritage: the lack of training dataset, particularly related to architecture; the lack of best practices for geometry processing and analysis of 3D data; the use of poorly predictive data in semi-automatic processes; the sharing of data into the scientific community; the importance of the open source technology and open data.

In 2020, the Samian Research database began a process of integrating its data within Wikidata through the creation of a set of Samian Research Wikidata items, including Samian Ware Discovery Sites, Samian Ware kiln sites and kiln regions, comprising accurate or approximate geospatial information and a backlink to the Linked Open Data hub ‘archaeology.link’. This approach of creating designated Wikidata items is an efficient way to map the enormous geographic reach of our subject and to call attention to many European archaeological sites and excavations that hitherto lacked a Wikidata identifier. The site of Corinth illustrates an exemplary issue to be solved: ambiguity and different archaeological concepts and ideas. E.g., is it correct to merge Corinth as a Samian Ware Discovery Site with the archaeological site of ancient Corinth? To solve the issue, the broader Wikidata community must be enlisted. This paper describes the challenges in the use case of Corinth and offers solutions within Wikidata.

The aim of this contribution is to explain workflows, methodologies and impact of the ‘MedAniene Project’, that was realised through the collaborative use of Wiki platforms and coordinated by the APS ArcheoFOSS. The project involved creating or modifying Wikipedia pages for the museums of the Aniene Valley, uploading images to the Wikimedia Commons repository and structuring relational information in the Wikidata system. The goal of the work was to improve the communication and increase the visibility of those museums by providing the public with open access scientific explanations and implementing data relationships through several heuristic systems which are available to the digital audience. Building such an informative apparatus offers the opportunity to learn about shared digital cultural heritage, increasing awareness in external users and in communities. It is now possible to analyse data on views, accesses and clicks on contents and evaluate the impact of specific activities in terms of audience targeting and engagement. Indeed, this work proposes a methodology for the establishment of a digital, open and collaborative communication space, improving the connection between cultural entities and communities.

The Viabundus pre-modern street map attempts to show medieval and early modern traffic connections. However, mapping medieval and pre-modern land routes comes with methodological challenges which are reflected upon in this paper. The reconstruction is based on written and archaeological sources, historical maps, and establishments of traffic infrastructure. Correlating the data with the origin places and finding places of pilgrim badges shows the research potential of the endeavor, as the simple co-visualization of the data already provides interesting connecting points.

The area of Castronovo di Sicilia was analysed by integrating different methodologies. In terms of the road network, it was decided to compare information from traditional written sources, such as the Itinerarium Antonini and texts from the Arab geographer al-Idrisi, with the results of the Least-Cost Path Analysis (LCPA) conducted using the QGIS plugin ‘movecost’. The primary objective of this analysis was to evaluate how the centrality of the Castronovo area was determined by environmental factors that made it easily accessible along the main long-distance routes connecting the island. At the same time, the analysis aimed to highlight similarities and differences between the written sources and the LCPA results.

The purpose of the paper is to introduce a QGIS plugin named ArcheoloGIS. It is developed in PYQGIS and tested by the community of Una Quantum Inc. (Italy). It consists of a decorator algorithm named Tabula Peutingeriana, that outputs points at a regular distance, every one Roman mile, along a given path. The article shows its use, the construction of a possible dataset and its evolution, as well as a case study of its application.

Since 2019, the LAD team has been working on the digitisation of the Carte topographique de l’Égypte through a GIS platform. The data contained in this historical cartography, published in the early years 1800s, play a key role in research on the ancient Egyptian landscape, yet they show a still image from the late 1700s. Taking a step towards a Linked Open Data (LOD), this paper illustrates the work of updating the already published dataset of the hydrography of Napoleonic cartography by the LAD team, to which new information useful for the study of the Ancient Egyptian landscape will also be added.

Introduction to the "Modelling the Landscape" section

The Iranian Plateau is an important geographical unit located in a key potential region for the Pleistocene population dispersals across Eurasia. Despite its important location and a long history of archaeological investigations, the Epipaleolithic sites distribution pattern and connectivity remained less explored compared to the Middle and Upper Paleolithic periods. In this study we used ecological niche modelling (Generalized Linear Models, Generalized Additive Models, Generalized Boosting Models, Maximum Entropy Modelling and Random Forest), together with corridor mapping methods, to reconstruct the Epipaleolithic settlements and their connectivity in the Zagros Mountains. We showed that the central parts and the western slopes of the Zagros Mountains were the most suitable areas for Epipaleolithic settlement during the Last Glacial Maximum (LGM). Topographic complexity was the most important variable in shaping Epipaleolithic settlement distribution with a positive association. The niche model and corridors maps developed for the Epipaleolithic humans show areas potentially suitable for the presence of Epipaleolithic settlements but no site has been discovered in this area so far. Thus, these areas are having high priority for future field excavations.

Southern Sweden, and especially the area around Malmo in southwestern Scania, is perhaps one of the most archaeologically investigated areas in the world. Our knowledge of the local Prehistory has greatly increased in the past decades although it is also the product of centuries of agricultural practices, urban expansion and a relatively early (18th-19th c.) interest for prehistoric monuments (e.g. burial mounds and megaliths). However, despite the deluging amount of available information (over 50,000 ancient sites recorded in Scania), their distribution is not homogeneous and archaeologists are restlessly trying to explain this pattern and its underlying causes. In addition, post-depositional factors (infrastructure works, agricultural practices, etc.) heavily affect site distribution and preservation, blurring the global interpretation. The aim of this paper is to reduce the impact of post-depositional factors on our interpretations on site distribution. In addition, the results can be used as a starting point for further and more elaborate analyses (spatial statistics and simulations). All the models presented here were computed in a reproducible way, relying on FOSS and open data only, in order to allow anyone interested to replicate the model and adapt it to their own purposes and study regions.

During the Upper Palaeolithic, Southern Tuscany was strongly affected by geomorphological changes that significantly altered its coastal seaboard. In particular, during the Last Glacial Maximum, the sea reached a level below 100 meters. As a result of this, the prehistoric coastland included also the present Tuscan Archipelago, in particular the Islands of Elba and Pianosa, assuming a different layout during MIS3 and MIS2. In this context, the process of prehistoric occupation took place, according to different needs and criteria. The present work explores the possibility of investigating the dynamic relationship between the prehistoric landscape and waterscape by a predictive-postdictive approach. Alongside the simulation of coastal changes, the study makes use of legacy data, taking into account those derived from artefact surface scatters collected over the past decades by various research groups. The latter provide further evidence of the prehistoric occupation process. In this scenario it is crucial to highlight areas that potentially still retain some relict features of the Palaeolithic landscape. These are examined in order to better understand settlement strategies taking place during the Upper Palaeolithic and, at the same time, to investigate the relationship between inland and coastal sites in a diachronic perspective. Although still ongoing, preliminary results provide new elements for the planning of future field surveys.

The aims of TPAA Project (Traces Prehistoric in the Alpine Environment) are the research, promotion and protection of the archaeological heritage in the Lanzo Valleys in Graian Alps, Western Alps (Turin, Piedmont, Northwest Italy). This paper illustrates the GIS predictive model results for the identification of archaeological sites in Lanzo Valleys and the 2019-2020 field survey. The archaeological data stem from occasional findings or traces of rock art. The aim of the GIS predictive model is to identify Potential Archaeological areas for the presence of archaeological sites and to hypothesize any reconstruction of human frequentation dynamics in Western Alps. Predictive GIS model has been elaborated through the interpolation and interpretation of the different environmental and archaeological data available. In the GIS predictive model, criteria such as the geomorphology, distance to water resources, aspect, slope and the use of land were considered. Also, the methodology is an evolution of the one that has already been successfully employed in the Sessera Valley. The results of the GIS model are compared with archaeological data collected during field surveys in the Potential Archaeological areas.

The paper illustrates the creation and integration of the environment as a multilevel landscape in AMPBV Simulator, a spatial Agent-Based Model (ABM) developed in NetLogo programming language. The model was conceived with the aim of investigating, through a simulative approach, the events and the circumstances (both anthropic and environmental) that presumably led, between the end of the Late Bronze Age, in the 12th cent. BC, and the beginning of the Final Bronze Age, the protohistoric communities of the Southern Verona plain (known as the Northern Terramare polity) from a climatic phase of maximum development and articulation to an anti-climatic phase of sudden collapse. The study context is an interesting application for an investigation through ABM, both because of the complexity of the case scenario, in which several interrelated actors and factors must have played an important role, and because of the availability of a number of geographical and archaeological data providing both a term of comparison and an excellent information base. With the development of an artificial environment and by modeling processes potentially critical for the fate of the Terramare system, the aim is, on the one hand, to give such a complex study case a new tool for historical analysis and, on the other hand, to experiment Agent-Based Modeling and assess its potential as a methodology for archaeological investigation in the Po Plain.

Thanks to the reconstruction of agro-pastoral land use of a territory, it is possible to obtain much information, both of an ecological nature, and about the populations. By the reconstruction of these dimensions of a community it is possible to understand not only the aspects linked to the exploitation of a territory, the subsistence and demography of a given group, but also more generally the group’s social organization itself. With a series of GIS tools, capable of applying the FAO’s land evaluation techniques, it has been possible to generate a predictive raster model of the landscape with the degree of agro-pastoral suitability inside each cell. Thanks to this model, the agro-pastoral exploitation of a territory can be simulated, calculating the food production of each settlement, as well as the consequent demography maximum sustainability. Thanks to the identification of specialized productions sites and of settlements capable of producing a ‘surplus’, or vice versa ‘not-self-sufficient’, it will be possible to articulate socio-political models, hypothesizing exchange networks or relationships between the different sites. The text illustrates in detail the structure and functioning of the developed model, as well as its applications in the archaeological context of the ‘Ager Tarquiniorum’ during the Final Bronze Age.

For about ten years this research team has been experimenting with techniques of spatial and visual perceptual analysis using GIS on the Nuragic landscapes (Middle and Late Bronze Age) of different areas of Southern Sardinia. Over time, various field campaigns have been developed on the monumental complexes built by the Nuragic communities. Towers, fortresses, villages, tombs of giants and sacred wells suggest a marked attention to the display of a message of power, strength and monumentality by the human group studied. An aspect often emphasized by these researches is the relationship of monuments with mobility and with the routes and strategic points of the territory, often verified through matches of different Least Coast Path Analyses (LCPA) that cross the territory by connecting different patterns of points. However, the real efficiency of the paths predicted by the GIS has never been directly tested on the ground. The paper presents the result of field analyses conducted by our group of archaeologists on the paths resulting from the GIS analysis in the Marmilla territories: travel times, energy expenditure and the real possibility of a path to actually cross a given territory are provided. Working with the LCPAs is still to be explored, however it remains a valid tool for territorial research, if an analysis unrelated to preconceptions and with a holistic evidence framework is carried out.

The Roman colony of Savaria was the first founded town in Pannonia (Colonia Claudia Savaria). Some inscriptions attest the so-called ‘deductio’, the settlement of veterans. After the first reconstruction of the centuriatio of Savaria no substantial archaeological attempt has been made in the last 40 years to verify the theory. In the last decade, research into the existence of the Savarian centuriatio has been pursued by using GIS methods, thanks to which we have managed to build a predictive model-network for the centuriatio, which is completely different from the previous reconstructions. The model has been continuously refined and validated by archaeological fieldwork and geophysical survey. The new reconstruction has led to new possibilities for interpreting the sites excavated in recent decades and the previously known Roman roads and aqueducts. Another interesting relationship between the watercourses running through the former colonia and the Roman centuriatio was also detected: the impact of Roman agriculture on the landscape transformation that has survived to the present day. Our pilot project, launched this year, plans to verify these effects using multispectral UAV surveys and geophysical measurements to show whether there were former streams along the presumed Roman channels that could provide evidence to support this hypothesis.

The Roman Limes represents the border line of the Roman Empire at its greatest extent in the 2nd century AD. It stretched over 5,000 km from the Atlantic coast of northern Britain, through Europe to the Black Sea, and from there to the Red Sea and across North Africa to the Atlantic coast. The remains of the Limes today consist of vestiges of built walls, ditches, forts, fortresses, watchtowers and civilian settlements. The limes system is focused both on the presence of natural physical barriers, such as the Rhine and Danube rivers in Europe and the Sahara Desert in North Africa, either on the presence of fortified sections such as the Hadrian’s wall or the Germanic-Rhaetian limes. The latter two are the best preserved and studied section. However, the limes sections in which natural barriers were exploited to mark the boundary of the area under Roman control are less well known. Over the past two decades considerable progress has been made in the knowledge of limes areas such as the Rhine sector. In this area the river was exploited as a natural barrier, and control of the area was based on the presence of two larger legionary camps around which, along the southern course of the Rhine, small auxiliary camps gravitated. Only some of these encampments have been investigated and their position confirmed by archaeological excavations. The position of the other encampments is still speculated and awaiting verification. In this contribution, in order to verify the position of these hypothetical forts, through GIS systems a visibility analysis and path distance analysis was carried out based on the location of certain sites and taking into account the ancient road routes and the geomorphology of the soil.

The Roman Limes represents the border line of the Roman Empire at its greatest extent in the 2nd century AD. It stretched over 5,000 km from the Atlantic coast of northern Britain, through Europe to the Black Sea, and from there to the Red Sea and across North Africa to the Atlantic coast. The remains of the limes today consist of vestiges of built walls, ditches, forts, fortresses, watchtowers and civilian settlements. The limes system is focused both on the presence of natural physical barriers, such as the Rhine and Danube rivers in Europe and the Sahara Desert in North Africa, either on the presence of fortified sections such as the Hadrian’s wall or the Germanic-Rhaetian limes. The latter two are the best preserved and studied section. However, the limes sections in which natural barriers were exploited to mark the boundary of the area under Roman control are less well known. Over the past two decades considerable progress has been made in the knowledge of limes areas such as the Rhine sector. In this area the river was exploited as a natural barrier, and control of the area was based on the presence of two larger legionary camps around which, along the southern course of the Rhine, small auxiliary camps gravitated. Only some of these encampments have been investigated and their position confirmed by archaeological excavations. The position of the other encampments is still speculated and awaiting verification. In this contribution, in order to verify the position of these hypothetical forts, through GIS systems a visibility analysis and path distance analysis were carried out based on the location of certain sites and taking into account the ancient road routes and the geomorphology of the soil.

How can we study the making of a medieval juridical space? Which were the agents that affected more? We try to answer to these questions thanks to a postdictive approach. We applied the QGIS algorithms to model the potential political space. We used several agents to get different outcomes. We tried both environmental and human agents to avoid the more deterministic side of this approach. We focused on the plain of Lucca in Northern Tuscany to study the making and development of its bishopric. Then we turned to the southern side of this region to study the district of a castle already excavated and whose territory is known quite well: Selvena. In this last example, we applied a regressive procedure, starting from the 19th century communities boundaries and making hypotheses about the relationship with medieval districts. This procedure can be applied to any context where a certain amount of data is available.

This paper concerns spatial analysis applied to urban and landscape scale; main aims are the reconstruction of the evolution in a specific settlement and the detecting of potential location for archaeological sites. Spatial analysis takes advantage of a GIS dataset containing different systems of sources (excavations, historical maps, toponymy, medieval documents, geomorphological data). Case studies at urban scale concern Monselice (Veneto) and Salpi (Northern Apulia). A retrogressive analysis aims to reconstruct the medieval urban fabric starting from the late morphology of these cities, using the modern cadastre or a magnetic survey. The Kernel Density Estimation and the evaluation of parcels orientation have been applied for a comprehension of the urban structure. At territorial scale, case studies regard two sectors of Northern Apulia. First step of these ongoing researches concerns the detection of landform (by TPI-Topographical Index, Geomorphons) and Wetness Index (TWI). This work helps us detect potential areas for settlements which are not preserved (dependencies of the city of Montecorvino and of the monastery of S. Egidio) and validate the outcomes of TWI (S. Marco in Lamis); moreover, it provides new hints about the relationship between settlement pattern, geomorphological elements, territorial/hydrological arrangement related to centuration.

The aim of this research is to define a new methodology in order to reconstruct the historical human mobility network in mountain areas in a perspective of ‘longue durée’. Through the match of different types of data, in particular historical and archaeological sources, the analysis of environmental features and the application of a series of algorithms on a GIS platform, we produced a series of maps of possible mobility networks. The comparison between them and with the historical cartography emphasizes both continuities and breaks over time and outlines the reliability of the elaborations obtained. Our focus is a small region in the North-Eastern Italy, called Feltrino (BL), on a time frame from the Bronze Age to the modern times