The road not taken: Modelling approaches to transport on local and regional scales

Posted on August 10, 2016


Every Wednesday I post videos of papers given at archaeology conferences. This weeks session comes from the CAA conference.

Session Abstract: Research on transport systems thus far has largely focussed on the documented and partly surviving road systems, ranging from the Roman imperial road systems known from the itineraries and the Peutinger Table to the road systems documented by medieval cartographers, even when it is well known that secondary road systems were in use simultaneously. Empirical evidence for the existence, location and chronology, particularly for secondary roads, is however scarce in many cases. In addition, transport over water (both fluvial and coastal) is often not considered. In order to bridge the gap between our theoretical notions of short- to medium-distance transport and the surviving (archaeological and historical) evidence for transport systems, in this session we want to focus on the practical and theoretical implications of using spatial modelling and analysis techniques, such as GIS-based cost surface modelling and social network analysis, for better understanding transport at the local and regional scales. We want to explore in what way spatial modelling can provide more insight into the organisation of local and regional transport, as well as the implications it has for the interpretation of the position, function and potential for trade of settlements within the local and regional transport system. We specifically invite papers that deal with: 1) New approaches to modelling transport networks, including aspects of differential access to the system, different modes of (wheeled) transport and diverse cost considerations (energetic, economic etc.); 2) Studies that combine transport network modelling and quantitative analysis approaches such as social network analysis; 3) Studies that link transport networks to models of trade/exchange at the local and regional scale; 4) Applications of transport network modelling in different landscapes and environments.

Session Organisers: Mark Groenhuijzen, Philip Verhagen

Transport and trade: An energyscape model and transport network approach for trade in Roman times

Movement of goods is vital in all human societies. Complex societies are characterised by an increased importance of movement as the production of goods tends to be distributed geographically among all members/cities of the society. Roman society is particularly known for its well-developed trade systems where goods – such as food, wine, olive oil, pottery and stone – were moved in enormous quantities and over long distances.
We believe that the distribution of bulky and heavy goods is particularly related to the cost of their transport and that physical geography can be a major constraint to this transport. While Roman water transport has already received significant scholarly attention, land transport is studied only superficially. Many urban centres could, however, not be reached without land transport. Land transport was thus inevitable and formed an integral part of Roman trade. In this paper, we present different energyscape models for depicting the predicted energy costs of transporting heavy loads by land. The resulting energy networks are compared to identify differences in the structures and the focal points of the networks, such as changes in the location of distribution hubs or changes in the locations of critical intermediary sites that determine the direction, quantity and volume of flow and therefore critically contribute to the cost of moving and trading heavy loads. This approach allows for a better evaluation of how land transport could have acted as a structuring variable for the Roman landscape and the Roman economy. As a case study, the trade of marble in Central Adriatic Italy in Roman times is selected. Stone resources suitable for ornamental use did not occur in central Adriatic Italy and thus had to be imported. The marble assemblages of several Roman towns in the area are compared with the results of the energyscape models and transport networks.

Authors: Devi Taelman, Dimitrij Mlekuž

From sea to land: Reflections on freight traffic during the Iron Age of Northeastern Iberian

It is known that maritime and fluvial shipping were the most efficient means of transport for distributing freights during the Antiquity. However; the distribution of goods to certain inner cities or significant Iberian settlements raises many questions. In this regard the use of carts and pack animals was common, but the roads the Iberian people travelled are mostly unknown. The objective of this paper is studying overland travels during the Second Iron Age in the northeast of the Iberian Peninsula. In this regard, the application of the least cost paths calculation methods should provide the opportunity to define the basis of a hypothetical Iberian road network during this period in the area. The results will be compared with the historic roads, in order to detect the presence or the lack of coincidences among them. This subject will be approached through two study cases, concerning two different political, ethnic and geographic territories. In one hand we have the Iberian Cessetània, located in the coastal area, which grosso modo coincides with the province of Tarragona. On the other we have the western Ilergetia, a territory wich is nowadays split among the provinces of Lleida and Huesca. This area has been from ancient times a well known communication hub towards the inner lands of the Iberian Peninsula and the Atlantic coast, due to the significance of its fluvial networks. In conclusion with this work we pretend use the GIS least cost paths calculations in order to approach the Iberian terrestrial transport networks and their persistence over time and landscape.

Authors: Joan Canela Gràcia, Núria Otero Herraiz

The need of topographic restitution in local mobility analysis

When we conduct a landscape archaeology study in a GIS environment (especially the analysis of mobility) we should be aware that any territory undergoes a transformation, especially in the last 50 years.
The Digital Terrain Models on which the study is based depart from the current topography (whether models from contours or the latest from LiDAR point clouds), so the lack of a correction and topographic restitution it can lead to certain errors, as can be the layout of an optimal path on a current road, an industrial area blocking sightlines or the the impossibility of study a valley submerged in a dam. The probability of obtaining these errors increase in inverse proportion to the cell size of the digital model, and we need high resolution models if we want a higher accuracy in the calculations and results over the previous relief; the examples in this paragraph are evident when we work with a 5×5 m. DEM.
For all these reasons, it is essential to make a correction that brings us as closer as possible to a hypothetical paleotopography that allows us, among other things, know the evolution of a particular territory or calculate least cost paths avoiding “current interferences” such as the aforementioned actual roads, towns or, if we have the help of historical planimetry and orthophotos, opencast mines, dams and other significant alterations in the landscape.

Author: Diego Torres-Iglesias

Testing the validity of network analysis results in research on local transport networks

Computational archaeology provides valuable tools for the reconstruction and analysis of transport networks. One such approach is a combination of a network constructed using least-cost paths and network analysis (Verhagen et al 2014, Groenhuijzen and Verhagen 2015), which can potentially provide valuable information regarding settlement location choice, site hierarchy, the role of settlements in transport networks and so on. However, testing the validity of the network analysis results and the archaeological interpretation thereof has so far been largely neglected. One of the key questions is thus: how reliant are the results of network analysis and their interpretation on nuances and uncertainties in the methodology and the dataset? This paper aims to test the robustness of network analysis results by measuring and analysing the development of local network statistics in randomly emerging transport networks. It is applied on a case study involving the Dutch part of the Roman limes, an area which is particularly interesting for research on local transport networks in the light of social and economic relations between the local rural population and the Roman military population, and an area for which a large amount of archaeological and palaeogeographical data is available.

Authors: Mark Groenhuijzen, Philip Verhagen

Groenhuijzen, M. R. and P. Verhagen 2015 “Exploring the dynamics of transport in the Dutch limes.” eTopoi Journal for Ancient Studies Special Volume 4: 25-47.
Verhagen, P., T. Brughmans, L. Nuninger, and F. Bertoncello 2014. “The Long and Winding Road: Combining Least Cost Paths and Network Analysis Techniques for Settlement Location Analysis and Predictive Modelling.” In CAA2012. Proceedings of the 40th Conference in Computer Applications and Quantitative Methods in Archaeology, Southampton, United Kingdom, 26-30 March 2012, edited by G. Earl, T. Sly, A. Chrysanthi, P. Murrieta-Flores, C. Papadopoulos, I. Romanowska, and D. Wheatley, 357─366. Amsterdam: Pallas Publications.

Comparison of regional and local transport networks of stone-walled structures in Gauteng Province

Since 2008, as part of the South African National Research Foundation funded ‘Southern Gauteng Stone Walled Structures’ project a dozen postgraduate research assistants in Archaeology and Geography at Wits University, along with another dozen photo interpreters employed at SBL Geospatial Services in Cochin, India, have pored over freely available satellite images of an 8000 sq km study area between Johannesburg and the Vaal River using Google Earth software. Over 5000 pre-colonial stone-walled structures have been detected and classified. This presentation focuses on the Group I and Group II pre-colonial SWS, comparing their distribution and probable transport networks within the 8000 square km study area. Group I structures date to the 16th and 17th centuries AD and represent the earliest Iron Age settlements in the study area. Previous studies on a restricted sample of Group I SWS suggest they represent a relatively egalitarian, more pastorally oriented society with a relatively dispersed settlement pattern. Group II structures date to the late 18th and early 19th centuries AD and are generally regarded as representing fairly complex and densely nucleated societies of Tswana-speakers in the immediately pre-colonial era. Using spatial modelling and analysis techniques, such as GIS-based cost surface and social network analysis, an attempt is made to compare the probable transport networks of Group I and II SWS at the local and regional scales. This involves the application of transport network modelling in the same landscape and environment but on two separate chronological periods. Combined with quantitative analysis, the aim is to see how the proposed differences in social, economic and political organization of Group I and II SWS that were ascertained through standard settlement pattern studies, might be reflected in their respective transport networks.

Karim Sadr

A network model for the evolution of terrestrial connections in Central Italy (1175/1150─500 BC ca)

The period between the Final Bronze Age and the Archaic Age is a time of changes and developments in the Italian Peninsula which led to the creation of regional ethnic and political groups and to the formation of the first city-states in Western Europe.
We study the evolution of the interaction patterns among settlements by analysing terrestrial communication networks in Latium Vetus (LV) and Southern Etruria (SE) in several time snapshots. At a first overall inspection, the two regions appear to be very similar, except for some interesting differences. In order to understand to what extent the observed results are a consequence of either differences on the spatial distribution of settlements, or dissimilarities in the process that generated those networks (cultural and political factors), we design a simple network model. After locating the nodes at the positions we know them from the archaeological record, we start adding links where they are more needed, according to a geographical criterion. The total link length is the only data-derived constraint. The model reproduces with good accuracy the features of every real network, except for LV in the Early Iron Age, where it systematically underestimate the average shortest path length while overestimating the local efficiency. Our hypothesis is that this model, that implements an optimal resources (new roads) distribution, cannot reproduce structures shaped by the unbalanced tensions of a harsh competition such as the Latin region of those times. On the contrary, it works fine in the case of a heterarchical system, such SE, or when considering a centralized, hierarchical one, as LV in the Archaic Period. We measured the hierarchicity (Trusina et al 2004) of both real and artificial networks as a preliminary test
obtaining results compatible with this explanation.

Author: Luce Prignano, Francesca Fulminante, Sergi Lozano, Ignacio Morer



Posted in: Uncategorized