Papers seminar ‘Detection and Modelling of Ancient Pathways’

From this page you can download the pdfs of the presentations held at the seminar ‘Detection and Modelling of Ancient Pathways’, held at VU University Amsterdam, 27June 2016

Rachel Opitz (CVAST, University of South Florida)
Trends and state-of-the-art in the use of LiDAR for archaeology
Archaeologists have been using airborne laserscanning (ALS) for over a decade in projects ranging from heritage management schemes for post­industrial uplands in the UK or state­managed forests in Germany to research on cities now obscured by tropical jungle canopy in Central Mexico. The basic methods for the analysis and interpretation of this data have matured considerably and data is increasing available. Building on this increasing accessibility and an established basic methodology, archaeologists are addressing a growing variety of ground conditions and research and heritage management objectives through this technology. With this diversification comes the need to adapt the basic methods used to new landscapes and types of archaeological remains, and to integrate the practice of working with ALS with diverse fieldwork and research practices. In this paper I provide an overview of the history of lidar in archaeology and present key recent trends.

Philip Verhagen, Žiga Kokalj & Laure Nuninger (VU University Amsterdam / ZRC-SAZU, Ljubljana / CNRS Chrono-Environnement UMR 6249, Besançon)
Techniques to enhance the detection of paths with LiDAR

Over the past few years, significant progress has been made in the development of techniques to improve the recognition of archaeological remains in Lidar-based images. The accompanying tools are available for free, and have led to an increased awareness and use of Lidar in archaeological survey projects. Detection, however, is still predominantly manual labour heavily depending on expert judgment, since progress in (semi-)automated detection has been slow and haphazard. This means that the detection process is difficult to standardize for larger areas, and this in turn makes comparison between regions difficult. In this paper, we will sketch the state-of-the-art in current detection techniques, and discuss how useful these are in the context of detecting ancient pathways. We suggest that we need to develop an ontology of pathways in order to become more successful at detecting, interpreting and comparing the features of interest.

Cathérine Fruchart (University of Bourgogne-Franche Comté, Besançon)
Ground-truthing of detected paths

With LiDAR, we can detect many unmapped road elements, either unmapped current paths or remains of ancient tracks that are now abandoned. Once mapped, these features can reveal unknown communication networks at local or regional scale. To understand the local dynamics of these networks and to document their period of use, the dating of ancient paths provides fundamental information. While some of them can be dated by their identification on old maps, most of the time field walking provides the most significant results by collecting dated artefacts, which are key elements to reconstruct ancient routes and define their evolution over time. Based on some studies from north-eastern France, the presentation shows how field walking, LiDAR and historical information have been used to document the evolution of communication routes at the local scale since the Roman period.

Dimitrij Mlekuž (University of Ljubljana)
LiDAR and the messy nature of landscapes

My argument is that landscapes are largely messy. How can we with archaeological landscapes as a mess? Landscapes change. They are always in the process of being built and they are never complete. Landscapes are messy because they were constantly patched together from the things that were at hand, through the reshaping and alteration of things from different times, through the assembling of things in new, different ways. Never finished. Instead of focusing on static things, we might focus on the significance of networks, connections, flows and mobilities in the ongoing making of landscape and the bodies that inhabit it. The role of landscape archaeology is to understand how people, landscape, animals, things, nature society are entangled and dependent, and how these complex entanglements emerged and changed over time. Messy landscapes teach us that we need to become more accepting of these blurred, messy things rather than demanding sharp clear ones. Our job is not to make them clean. Trying to describe complex, diffuse and messy things, landscapes, in simple terms would only make more of a mess out of them.

Zoran Čučković (Université de Bourgogne-Franche Comté, Besançon)
Conceptualising movement: modelling topographic feature networks

Paths do not equal movement, paths are blueprints for potential movement. In any landscape there is much movement out of trodden paths and, crucially, prior to their materialisation on the ground. Foraging, hunting, pastoralism and many more activities are intimately related to efficient and well-planned movement across the terrain, yet they are not determined by the layout (or existence) of materialised paths – which are more likely to appear as a by-product of repeated movement, rather than its incentive. However, it is precisely this phenomenon of crystallisation of path networks from the incessant negotiation of terrain that gives an impression that some kind of blueprint could exist, that movement does get organised without being materialised on the ground. Indeed, the existence of purely conceptual path networks for pastoral purposes or for hunting and gathering is well attested ethnographically.

The problem which thus arises is understanding and modelling “conceptual” path networks and, more generally, the cognitive aspect of movement across terrain. By conceptual I do not mean virtual or imaginary: these paths are composed of real-world objects, terrain features such as ridges, valleys and rivers or more punctual landmarks such as individual trees, rock outcrops and ancient monuments. In order to model such path networks we cannot rely solely on traditional energy/time consumption calculus – an additional problem of “cognitive economy” has to be taken into account as well. The proposed communication will introduce a new approach to these problems through topographic networks. These are a representation of terrain in terms of networks, much like watershed models, which can be used for modelling shape and degree of singularity of terrain features.

Frédérique Bertoncello ( CNRS CEPAM UMR7264 / University of Côte d’Azur, Nice)
Long-term settlement development and prospects for path modelling

More than 30 years of research on ancient settlement systems have evidenced the variety of settlement patterns and dynamics in different areas of France from the Iron Age to the late Roman period. One core question is to identify the factors – both social and natural – involved in these various density, structure and evolution of the settlement. Related to the relative position of places, which is central in geography, accessibility and connectivity of the settlements are often considered as key elements to explain settlement pattern and evolution. Going back to the meaning of these concepts in geography, we will examine how they have been used in archaeology and what are the underlying hypotheses. Using the results of several research programs (Archaeomedes, ArchaeDyn and IHAPMA), we will present some issues and prospects of path modelling to understand Ancient settlement and territorial dynamics.

Élise Fovet (University of Bourgogne-Franche Comté, Dijon)
Path modelling and the development of network and settlement patterns

Possibilities and constraints on movement of people and goods is a key factor that determines practices and structures space. Therefore, understanding communication networks is very important when analysing changes in settlement patterns. Ancient communication network modelling is a useful approach for studying the development of this pattern, considered as a changing network rather than a succession of creations and abandonments of settlements.

In regional or supra-regional approach, road systems are usually considered as a coherent, planned network connecting strategic places (notably for Roman and Modern times) while auto-organisation principles are rarely taken into account to study the historical development of road networks. This presentation will discuss some of these principles through the case of the historical development of the network of regional centres in Languedoc (south-eastern France) from pre-Roman to late Roman periods.

Mark Groenhuijzen (VU University Amsterdam)
Path modelling and network analysis in the Dutch Roman limes

In our research on the Dutch part of the Roman limes we are interested in transport networks that link the rural population with the Roman military population. Yet we know very little about local transport in the Roman period, which is enhanced by the fact that most connections are likely mere routes rather than archaeologically visible roads. In this paper I will discuss how I have applied path modelling to reconstruct local routes in order to get a better understanding of transport networks in the Dutch Roman limes, as well as how formal network analysis contributed to our research.

Laure Nuninger & Philip Verhagen (CNRS Chrono-Environnement UMR 6249, Besançon / VU University Amsterdam)
The integration of path detection and modelling: state-of-the-art and future prospects

In this paper, we want to summarize and highlight the main issues discussed during the seminar, and open the debate about future research directions. Pathway detection and modelling have operated from different perspectives that seem to have little in common apart from using the same base data, the DTMs themselves. A crucial factor in both domains is the scale of analysis and modelling, both in spatial (2D and 3D) and temporal dimensions, but also in non-spatial terms of hierarchies of networks. We suggest that the gap can be bridged by considering the historical trajectory of pathways, from their construction and associated practices up to their development during their use and after their abandonment. An evolutionary, multi-scalar perspective of pathways is crucial to enhance our ability to detect and interpret structures, and to link the detected elements to their modelled counterparts.