1 International Workshop ERC World Seastems

International Workshop ERC World Seastems 1

PROGRAMME, PRACTICAL INFORMATION & ABSTRACTS

Paris, 16-18 June 2014

TABLE OF CONTENTS ------4 AIMS OF THE WORKSHOP ------5 PRACTICAL INFORMATION ------7 WORKSHOP PROGRAMME ------15 ABSTRACTS ------23 WHO’S WHO ------69 PARTICIPANTS’ CONTACT EMAILS ------85 WORLD SEASTEMS PROJECT OUTLINE & KEY FACTS ------88

More than 80% of world trade volumes occur by sea nowadays. Despite this enormous importance, little is known about the precise distribution and configuration of maritime flows. This international workshop on maritime flows and networks wishes to gather worldwide scholars and experts from all scientific and disciplinary backgrounds to introduce and discuss their researches on various related issues. This is a multi-disciplinary event which will investigate how maritime flows and networks of any kind can be analysed in order to answer major industry and wider societal challenges.

Contributions will deal with themes such as:

 ancient and contemporary shipping routes and corridors;

 regionalisation and globalisation dynamics in world maritime trade;

 global production networks and logistics/supply/value/commodity chains;

 multimodal transportation, hinterlands and the land-sea interface;

 port systems, centrality, hubs, maritime ranges and forelands;

 port-city relationships, merchants and intermediaries, urban systems, global cities;

 network security, congestion, disruption, vulnerability, and optimisation; network design and routing, port choice and shipping line competition; technological change in shipping and port operations, time and cost factors; climate change, bio-invasions and environmental sustainability;

 mobility, migrations, cultural exchanges, knowledge diffusion, tourism, voyages and discoveries.

The originality of the workshop is to put a strong emphasis on the methodological aspects across a wide spectrum:

 theories, concepts, models applicable to transport networks;

 database building from historical and other documents (AIS, schedules, movements...);

 network analysis from graph theory to complex systems;

 visualization and modelling of flow matrices;

 spatial analysis and statistics, cartography, Geographical Information Systems (GIS);

 simulation, agent-based models, forecast.

It is intended that this workshop will lead to a book publication, based on a selection of best papers following the workshop itself. Young scholars as well as confirmed experts are all welcome, from both social and natural sciences (i.e. archaeology, history, geography, economics, engineering, computer science, physics, mathematics, geomatics...).

WORKSHOP VENUE :

ISC-PIF (Institut des Systèmes Complexes, Paris-Île de France) 113 rue Nationale, 75013 Paris / Metro : line 6 (Nationale) or 14 (Olympiades).

USEFULL PHONE NUMBERS :

In case of any problem, please contact organisers :

- César Ducruet: +0033(0)628.196.048 (mobile phone);

- Sébastien Haule: +0033(0)674.581.315 (mobile phone);

- Kamel Ait-Mohand: +0033(0)628.344.274 (mobile phone) ;

- Martine Laborde (UMR Géographie-cités secretary): +0033(0)140.464.000 (office phone).

ACCOMMODATION: 5TH ARRONDISSEMENT

 Hôtel Beauvoir 43 avenue Georges Bernanos, 75005 Paris Phone: +0033(0)143.255.710 RER: line B (Port-Royal)

 Hôtel Libertel Austerlitz Jardin des Plantes 12-14 boulevard de L'hôpital, 75005 Paris Phone: +0033(0)143.376.080 Metro: lines 5 or 10 (Gare d’Austerlitz) or line 5 (Saint-Marcel); RER: line C (Gare d’Austerlitz)

ACCOMMODATION: 6TH ARRONDISSEMENT

 Hôtel Perreyve 63 Rue Madame, 75006 Paris Phone: +0033(0)145.483.501 Metro: line 4 (Saint-Placide)

 Hotel Saint Andre Des Arts 66 Rue St Andre Des Arts, 75006 Paris Phone: +0033(0)143.269.616 Metro: lines 4 or 10 (Odéon) or line 4 (Saint- Michel) ; RER: line B or C (Saint-Michel-Notre- Dame)

ACCOMMODATION: 13TH ARRONDISSEMENT

 Hôtel Paris Massena Olympiades (Ex Bercy Rive Gauche) 82-84 rue Regnault, 75013 Paris Phone: + 33(0)145.857.070 Metro: line 7 (Porte d’Ivry)

 Timhotel Place d'Italie - butte aux cailles 22 rue Barrault, 75013 Paris Phone: +0033(0)145.806.767 Metro: line 6 (Corvisart)

 Hôtel Le Richemont 17 rue Jean Colly, 75013 Paris Phone: +0033(0)145. 828.484 Metro: line 14 (Olympiades)

 Grand Hôtel Des Gobelins 57 Boulevard St Marcel, 75013 Paris Phone: + 0033(0)143.317.989 Metro: line 7 (Les Gobelins)

GALA DINNER (SEINE RIVER CRUISE) / TUESDAY 17TH JUNE 2014

 Marina de Paris - Port Solferino 23 Quai Anatole France, 75007 Paris RER: line C (Musée d’Orsay)

DAY 1 - Monday 16th June 2014

KEYNOTE SPEECHES Time slot Title Welcoming address 1 9h00-9h10 David CHAVALARIAS - Institute of Complex Systems Paris Ile-de-France (ISC-PIF) Welcoming address 2 9h10-9h20 Arnaud BANOS - CNRS & UMR 8504 Géographie-cités, Paris, France Workshop introduction 9h20-9h30 César DUCRUET - CNRS, Paris, France Shipping line networks 9h30-10h00 Antoine FREMONT - Réseau Ferré de France (RFF), Paris, France Merchant networks in the 'Asian Mediterranean' 10h00-10h30 François GIPOULOUX - Ecole des Hautes Etudes en Sciences Sociales (EHESS), Paris, France

10h30-11h00 Coffee break Time slot Title City networks and transportation systems on the long-term period 11h00-11h30 Anne BRETAGNOLLE - University of Paris 1, France Spatial networks 11h30-12h00 Marc BARTHELEMY - Commissariat à l'énergie atomiques et aux énergies alternatives (CEA), Paris, France 12h00-12h30 Round table (discussion)

12h30-14h00 Lunch break

SESSION 1: THE ANALYSIS OF HISTORICAL MARITIME NETWORKS [1] Time slot Title How technology and geography influence network dynamics Ray RIVERS - Imperial College London, United Kingdom 14h00-14h30 Carl KNAPPETT - University of Toronto, Canada Tim EVANS - Imperial College London, United Kingdom ORBIS: First of many worlds? Elijah MEEKS - Stanford University Libraries, USA 14h30-15h00 Karl GROSSNER - Stanford University Libraries, USA Walter SCHEIDEL - Stanford University Departments of Classics and History, USA Modeling Venice's maritime network - End 13th to Mid. 15th centuries Mélanie FOURNIER - Dalhousie University, Canada 15h00-15h30 Yannick ROCHAT & Frédéric KAPLAN - Ecole Polytechnique Fédérale de Lausanne, Digital Humanities Laboratory, Lausanne, Switzerland

15h30-16h00 Coffee Break SESSION 2: THE ANALYSIS OF HISTORICAL MARITIME NETWORKS [2] Time slot Title Navigocorpus. A database for shipping and trade 16h00-16h30 Silvia MARZAGALLI - Université Nice Sophia Antipolis & Institut Universitaire de France Networks of oil and water: Pennsylvania’s global petroleum industry and its urban footprint in 16h30-17h00 Philadelphia (1860s-2014) Carola HEIN - Bryn Mawr College (Pennsylvania), Growth and Structure of Cities Department, USA ‘Knots’, port authorities and governance: comparing the governance of port connectivity in three major European ports 17h00-17h30 Luis Lobo-GUERRERO & Anna STOBBE - Chair Group on History and Theory of International Relations, University of Groningen, The Netherlands

DAY 2 - Tuesday 17th June 2014

SESSION 3: DEDICATED "WORLD SEASTEMS" SESSION [1] Time slot Title Introduction to the World Seastems project (2013-2018) 9h00-9h15 César DUCRUET - CNRS, Paris, France The complex corpus of Lloyd's List periodicals: from collection to extraction 9h15-10h00 Sébastien HAULE & Kamel AIT-MOHAND - CNRS, Paris, France A review of other maritime databases as sources for research 10h00-10h30 César Ducruet, Sébastien Haule & Nora MAREÏ - CNRS, Paris, France

10h30-11h00 Coffee break SESSION 4: DEDICATED "WORLD SEASTEMS" SESSION [2] Time slot Title Changing Mediterranean maritime networks 11h00-11h30 Nora MAREÏ - CNRS, Paris, France Polarization of oceanic areas and shape of the global network of shipping containerized traffic at the 11h30-12h00 dawn of the 1990s Olivier JOLY - University of Le Havre, France Network structure and regional distribution of global maritime flows (1890-2010) 12h00-12h30 César DUCRUET - CNRS, Paris, France Bruno MARNOT - University of La Rochelle, France

12h30-14h00 Lunch break

SESSION 5: MARITIME NETWORK DYNAMICS Time slot Title The multiplier attachment: A shipping network architecture 14h00-14h30 Francesca MEDDA & Simone CASCHILI - UCL QASER Lab, University College London, London, United Kingdom Modelling the emergence of world’s direct shipping lines: An uncertainty analysis 14h30-15h00 Ronald HALIM & Lóránt TAVASSZY - Delft University of Technology, Transport and Logistic Group, Delft, The Netherlands Random walks in dynamic maritime networks 15h00-15h30 Frédéric GUINAND & Yoann PIGNÉ - LITIS Laboratory, University of Le Havre, France

15h30-16h00 Coffee break SESSION 6: DIFFUSION AND VULNERABILITIES IN MARITIME NETWORKS Time slot Title The complex network of cargo ship movements and its importance in marine bioinvasion 16h00-16h30 Michael GASTNER - Institute of Technical Physics and Materials Science, Research Centre for Natural Sciences, Hungarian Academy of Sciences, Budapest, Hungary Maritime network vulnerability to cascading failures 16h30-17h00 Serge LHOMME - Ecole des Ingénieurs de la Ville de Paris (EIVP), Paris, France Maritime network evolution under crisis: The case of North Korea 17h00-17h30 Stanislas ROUSSIN - ASEM Institute, Seoul, Republic of Korea

18h00 Departure to the Gala Dinner (Seine River Cruise)

DAY 3 - Wednesday 18th June 2014

SESSION 7: MARITIME NETWORKS AND TRADING FLOWS Time slot Title Impacts of Northern Sea Route on natural resources logistics map in East Asia: South Korea's case 9h00-9h30 Sung-Woo LEE - Korea Maritime Institute (KMI), Seoul, Republic of Korea Explaining international trade flows with shipping-based distances 9h30-10h00 David GUERRERO - Institut Français des Sciences et Technologies des Transports, de l'Aménagement et des Réseaux (IFSTTAR), France Think inside the box 10h00-10h30 Yann ALIX - SEFACIL Foundation, Le Havre, France Jean-François PELLETIER - CPCS, Canada

10h30-11h00 Coffee break SESSION 8: PATTERNS IN MARITIME NETWORKS Time slot Title Visualizing regional structures in global maritime flows 11h00-11h30 Françoise BAHOKEN - Institut Français des Sciences et Technologies des Transports, de l'Aménagement et des Réseaux (IFSTTAR), France Stochastic blockmodeling applied to global maritime flows 11h30-12h00 Charles BOUVEYRON - Laboratoire MAP5, Université Paris Descartes, France Pierre LATOUCHE & Rawyia ZREIK - Laboratoire SAMM, Université Paris 1 (Panthéon-Sorbonne), France Maritime trajectory patterns 12h00-12h30 Laurent ETIENNE - Naval Academy Research Institute, Brest, France Thomas DEVOGELE - University of Tours, Blois, France

12h30-14h00 Lunch break

SESSION 9: INTERMODAL MARITIME NETWORKS Time slot Title The Weastflows Project: freight transport spatial interactions and territorial dynamics in North West Europe 14h00-14h30 Juliette DUSZINSKI, Jean-François MARY, Lucile AUDIEVRE & Mathilde MUS Agence d'Urbanisme de la Région Havraise (AURH), Le Havre, France The Eurasian railway corridor and Lower Seine Valley ports: competition or complementarity? 14h30-15h00 Jérôme VERNY & Milhan CHAZE - NEOMA Business School, Rouen, France Modeling a system of cities based on spatial networks 15h00-15h30 Igor LUGO - National Autonomous University of Mexico (UNAM), Regional Center for Multidisciplinary Research (CRIM), Mexico

15h30-16h00 Coffee break SESSION 10: CHINA'S MARITIME NETWORKS Time slot Title The British and Japanese shipping network in China in 1920s 16h00-16h30 Liehui WANG - Center for Modern Chinese City Studies, East China Normal University, Shanghai, China Lie YANG - College of History and Social Development, Shandong Normal University, Jinnan, China Canton-Hong Kong-Macau: A historical review on the intercity competitions and interactions of the three commercial centers of the Pearl River Delta before the 20th century 16h30-17h00 Ka-Chai TAM - Department of History, Hong Kong Baptist University, Hong Kong, China Adolf K.Y. NG - Department of Supply Chain Management, I.H. Asper School of Business, University of Manitoba, Winnipeg, Manitoba, Canada

SPECIAL GUEST: MARITIME PIRACY ANALYSIS Time slot Title A geospatial analysis of maritime piracy and its challenges 17h00-17h30 Philippe LEYMARIE, Journalist, Le Monde Diplomatique

WORKSHOP CONCLUSION

Time slot Title 17h00-18h00 Workshop conclusion & discussion on publication opportunities

Shipping line networks

Antoine FRÉMONT

Réseau Ferré de France (RFF), Paris, France

Merchant networks in the “Asian Mediterranean”, 16th-21th cent.

François GIPOULOUX

Ecole des Hautes Etudes en Sciences Sociales (EHESS), Paris, France

City networks and transportation systems on the long-term period (13th-21th cent.)

Anne BRETAGNOLLE

Paris I Sorbonne University, Paris, France

Taking into account a long-term perspective of eight centuries do matter when exploring the complex relationship between transportation systems and city networks. We focus here on the European and American regions of the world and give a special attention to theoretical issues. Three main periods are distinguished, each of them being related to a particular globalization phase: the European world-economy system (13th-18th cent.), the industrial revolution & second globalization phase (19thcent. -beginning of 20th cent.), the current globalization phase (since the Second World War). We aim to show that the nature of the links between maritime transportation and systems of cities have dramatically changed. In the first globalization phase, inter-urban links are weak but deeply based on maritime exchanges (Fernand Braudel). We explore the consequences of this prominence at international, national and regional levels. In the second globalization phase, the maritime transportation systems (strongly articulated with rivers and railroads) are the main roots of the world systems of cities, the world metropolises being coined as “world emporia” (i.e. ports) by Paul Vidal de la Blache. Like for railroads, a process of geographical selection (Halford Mackinder) can be analyzed for maritime transport, with a positive retroaction feedback between improving accessibility and increasing centrality degree. In the current globalization phase, we hypothesize a progressive “decoupling” or deconstruction of the retroaction feedback between maritime accessibility and centrality. The main factor invoked to explain this decoupling is the relative speed of maritime transport, too slow in comparison with other information networks (telecommunications) and transportation networks (planes, rapid trains) for allowing it to keep a structuring role in urban systems at the global level (Roderick MacKenzie).

Spatial networks

Marc BARTHELÉMY

Commissariat à l'énergie atomique et aux énergies alternatives (CEA), Paris, France

How technology and geography influence network dynamics

Ray RIVERS

Imperial College London, SW7 2AZ, UK

In creating dynamical models for historic maritime networks we have a large range from which to choose; intermediate opportunity models, maximum entropy models, cost-benefit models and more. Although possessing a common methodology, expressed through the modelling cycle, different models make different assumptions about the ‘agency’ underlying the network formation.

In this paper [in collaboration with Tim Evans (Imperial College London) and Carl Knappett (U. of Toronto)] I shall compare and contrast the maritime networks of the Early Bronze Age (EBA) Cyclades, the Middle Bronze Age (MBA Minoan) Aegean and the late Bronze Age (LBA) East Mediterranean. [See Figures of vessels above, reading from top to bottom.] I shall show how the different marine technologies of these periods - particularly widespread use of sail – and the different scales of the networks require different model types, and thereby different agency, if we are to explain the archaeological record. Geography and technology, through the distribution and separation of sites and the ability of vessels to traverse these distances, come together in determining the ease with which links arise. A key issue is how robust our conclusions are, given the ambiguities of model-making. Different models encode robustness and ‘contingency’ (non-deterministic behaviour) differently. Understanding this is an important part of our analysis. We have several papers addressing these issues, but hitherto restricted to the MBA and not the Bronze Age as a whole. A pertinent one accessible on the arXiv is Evans TS, Rivers RJ and Knappett C., 2012 “Interactions in Space for Archaeological Models”, Advances in Complex Systems 15 1150009 (17 pages), arXiv:1102.0251

Keywords: Maritime networks, Bronze Age, Aegean

ORBIS: First of many worlds?

Elijah MEEKSa, Karl GROSSNERa, Walter SCHEIDELb

a Stanford University Libraries ; b Stanford University Departments of Classics and History

The popularity of ORBIS: The Stanford Geospatial Network Model of the Roman World suggests that a more generalized web software framework for presenting complex geographic systems will be broadly useful. This framework should support the representation of such systems in spatial-temporal databases and in multiple types of interactive visualizations, including geographic maps and network diagrams, as well as more traditional graphs and charts.

The ORBIS web application models the transport of goods and people for a particular period and region in terms of distance, time, and cost in currency, but in principle, it could be used for any period and region. Indeed, there have been suggestions for several extensions, for example forward in time and eastward along the Silk Road.

ORBIS 2.0, (http://orbis.stanford.edu/v2) retains the original geographic and temporal scope, but extends the model and presentation significantly. ORBIS 2.0 now accounts for the duration of mode transfers at ports. Dynamic cartograms can now be generated speedily with any of the 650 plus sites as its center. The ability to generate cost-based clusters and regions has been added as well.

In this presentation, we will give a short overview of the project, detail the modeling of maritime routes and links, and also describe how we imagine the technology and interface of ORBIS 2.0 could be extended further, to create both a more powerful scholarly platform and a more generic online historical atlas. What are now fixed weights for routes and fixed financial costs could be made parameters. Sites and roads could be annotated to any level of detail with information from authoritative sources. To the extent that information is structured, correlations between the network characteristics of sites (e.g. centrality) and other dynamic properties, like population and position in administrative hierarchies, could be explored. Encyclopedic linked data could be integrated with maps to aid understanding with historical context.

Whereas print atlases and print scholarly analytical papers can present only a handful of maps and other visualizations in support of interpretation and argument, a digital atlas such as an embellished ORBIS-based application could provide a potentially unlimited number. Teaching tools and historical narratives for popular audiences could be developed atop the same data store. There is also the opportunity to allow annotations of an ORBIS atlas by authorized users, the general public, or both.

The digital historical atlas is an emerging genre of scholarly work with enormous potential only barely tapped so far. We are actively seeking to promote their further development by extending the ORBIS framework. We will report on work to date, and look forward to discussing possibilities and challenges with our fellow workshop participants.

Modeling Venice's maritime network - End 13th to Mid. 15th Centuries.

Fournier Mélaniea, Rochat Yannickb, Kaplan Frédéricb

a Dalhousie University, Canada ; b Digital Humanities Laboratory, Ecole Polytechnique Fédérale de Lausanne

The maritime Republic of Venice has been described as the first colonial empire [1]. During six centuries (from the end of the 12th century until the end of the 18th century) the Republic ruled an empire made of continental and maritime possessions, growing and shrinking following the expansion of other empires and mainly the Ottoman Empire. Its authority extended on two dominions: one overseas called by the Venetians Stato da Mar and the other, continental called Stato da Terraferma. Those two dominions were made by a succession of islands, cities, fortresses, ports and bases of strategic importance along the rivers (Po, Adige and Brenta) and along the shores since the lagoon to Syria, Egypt and to the Azov Sea.

If we localize these sites on a map we see that Venice's possessions were distributed along the trading routes to the Levant [2] but the Venetian maritime network is not reduced to it. It is necessary to take into account the numerous ports of call which led to the western Mediterranean Basin, to England, to Brabant and to Flanders. And even the cities of caravans departures/arrivals such as Bagdad, Dams or Aleppo. To study this network in general is a long process because it has to take into account acquisitions and losses of territories, treaties with countries, conicts etc. but also various types of navigation known for the Venetians [3] [4]: free, private, public ships with auction, direct management etc. This is the reason why we decided to study one type of navigation at a time.

In 1992 a historian presented a thesis dedicated to the Venetian navigation by auction: the Incanti [5]. She is not the first one describing the evolution of the Incanto system, Tenenti and Vivanti wrote a small study in the 60s' with a complete cartography presenting the phenomenon year by year [6], but what was new here was the exhaustiveness in the perusal of the primary sources (not only Venetians) and the attempt of systematization in the retranscription of ports with the duration of the stopovers, prices, names and routes.

After scanning and pre-processing each pages via several computer vision-based procedures (e.g. adjustment of rotations, removal of noise), we made explicit the structure of the tables by creating a grid. This grid in conjunction with an OCR software permitted to extract structured data. The next two steps were to clean the tables and to project the sites into a GIS. We aligned the names and the geolocalizations with existing databases and geonames from digital historical atlases such as Pleiades or Digital Atlas of Roman and Medieval Civilizations from Harvard. The sites with coordinates and the data once cleaned have been decomposed

JUNE 2014 - UMR 8504 GÉOGRAPHIE-CITÉS - ALL RIGHTS RESERVED International Workshop ERC World Seastems 33 using a R script into individual segments. By connecting these directed segments and the nodes we created a global network encoding 170 years of navigation. This network only concerns the auctionning system called Incanto.

Keywords: Venice; maritime routes; Incanto system; maritime networks; GIS; ABBY Fine Reader; R

References:

[1] Thiriet F. Histoire de Venise. Presses Universitaires de France, 1952.

[2] Dudan B. Il Dominio Veneziano Di Levante. Filippi Editore Venezia, 2006. First edition in 1938.

[3] Luzzatto G. Storia economica di Venezia dal' XI al XVI secolo. Centro internazionale delle Arti e del Costume, 1961.

[4] Lane F.C. Venise une république maritime. Flammarion, 1985.

[5] Stoeckly. Le système de l'Incanto des galères du marché à Venise (Fin XIIIe-Milieu XVe siècle). PhD thesis, 1992.

[6] Tenenti A. and Vivanti C. Le film d'un grand système de navigation: les galères marchandes vénitiennes (xve-xvie siècles). Annales. Economies, Sociétés, Civilisations, 16(1):83-86, 1961.

The Incanto system visualized as a network

Navigocorpus. A database for shipping and trade

Silvia MARZAGALLI

Professor of early modern history, Université Nice Sophia Antipolis, senior member at the Institut Universitaire de France

In 2007, the French Agence Nationale de la Recherche financed a research program aiming at creating a database structure to collect data concerning shipping and maritime trade issued from sources of different nature and providing information on the presence of ships or their movements in time and space. The aim of Navigocorpus (http://navigocorpus.hypotheses.org/) was both to make it possible to stock data without losing any information provided by the sources, and to make the query as powerful as possible for researchers without predetermining the kind of historical questions they might wish to explore.

The result is a powerful, but complex tool. The fact that the database is an open one forced us to build a database centered on the points touched by a ship which are subsequently reordered chronologically as the database expands and new data on the same ship are added. A point is defined by its geo-localization, a date and an event which occurs in it. One of the major (and most time- consuming) features of Navigocorpus, which is not provided to my knowledge by any other existing major online databases, is the identification of ships and captains. Once they are identified, it is possible among other to reconstruct life-courses. In order to illustrate the potentialities of the database, we have inserted all available data on shipping in French ports in 1787. Although the ANR-program and funding is finished, we are still expanding the database.

Navigcorpus database is accessible online for all those who have a FileMaker 12 license. A simplified version is also available online (http://navigocorpus.org/).

Keywords: shipping; maritime trade; database building from historical documents; Navigocorpus

Networks of Oil and Water: Pennsylvania’s Global Petroleum Industry and its urban footprint in Philadelphia (1860s-2014)

Carola HEIN

Bryn Mawr College (Pennsylvania) in the Growth and Structure of Cities Department

Oil and ships have long been intimately connected, in the early years for river transportation and later for ocean transport. In the 1860s, at the beginning of petroleum extraction in Western Pennsylvania, wooden bulk boats carried oil down Oil Creek and the Allegheny River to Pittsburgh refineries and consumers. While these early barges were quickly replaced by rail transportation, water transport remained a cheap option and special deep-water barges worked on the Great Lakes; others carried oil products through the canal and river systems of Germany and the Netherlands Sea-going ships carried petroleum in barrels along the coasts and across the oceans. Standard Oil, for example, sold kerosene for lamps in China as early as the 1890s. The company owned its own fleet, including the sailing ships that carried oil from Philadelphia to cities in Asia, such as Shanghai and other treaty ports. Refineries needed rivers for industrial processes, for cooling and washing, and for dumping unwanted products - and also as drinking water for the employees who settled in their vicinity. Proximity to the port appears to have been one reason for the location of oil refineries, for example in Philadelphia and in Markus Hook on the Delaware waterfront, starting with the Bear Creek Refining Company in 1892. As oil became a global consumer good, even before the mass production of cars—its production and transportation was intimately entwined with urban growth and development. This paper explores the interface of land and water as witnessed through through the lens of Philadelphia’s urban history, focusing on the Schuylkill River refineries that were built in after 1865 and have shaped the city’s urban development until today.

‘Knots’, port authorities and governance: comparing the governance of port connectivity in three major European ports

Prof. dr. Luis LOBO-GUERRERO and Anna STOBBE Chair Group on History and Theory of International Relations, University of Groningen

Ports are not unitary entities. They are sites where multiple processes, agencies, and interests coalesce into a role that connects transport domains with goods in transit between markets. The connectivity that results from such role is not the simple sum of the parts but the effect of their interaction. The interaction is complex and heterogeneous, a port is under continuous transformation, adaptation and change. Material elements (e.g. containers), transportation systems, logistical and managerial principles and practices, market imperatives, and regulatory regimes, react to changing market and political conditions around the globe. How is such connectivity governed is the question that guides this paper.

Although the governance of ports is not restricted to the role of port authorities, these authorities play a central function in governing the connectivity effects of their ports. Detailing how port authorities publicly present their role and how other entities involved in port business understand such role provides an empirical ground from which to interrogate how port connectivity is governed. This paper focuses on the cases of the ports of Rotterdam, Hamburg and Antwerp.

How to conceptualise the governance of ports and the role of port authorities is a second challenge addressed in this paper. To that end, the figure of the ‘knot’ is explored as a metaphorical device with which to make sense of the governance of connectivity effects. Knots have historically been used in various cultures as systems for numbering and for keeping accounts, as ways of connecting ropes and fibres into useful instruments, as decorative objects and heraldic elements, as ways of measuring speed, and as representations of entanglement. Conceptually, they are employed in this paper as elements that perform and depict connectivity between otherwise disconnected elements and domains. They refer to the capacity and possibility to connect but also, and always, to ‘dis-connect’. As such, knots allow for the thinking of governance not in terms of permanent structures but as assemblages that can always be different, and especially, creatively different.

The paper employs a comparative approach to examine the role of the port authorities of Rotterdam, Hamburg and Antwerp as knots that create connectivity effects. In that way, the paper is written as an experimental piece with the clear objective of exploring alternative ways to conceptualise the role of contemporary major ports by focusing on the way they are governed. Keywords: connectivity; ports governance; Hamburg; Antwerp; Rotterdam

The complex corpus of Lloyd's List periodicals: from collection to extraction

Sébastien HAULE & Kamel AIT-MOHAND CNRS, Paris, France

“Lloyd’s list” was the oldest newspaper published in England – and probably in the world- till December 20, 2013, when the last paper issue was printed for the benefit of digital format. The “List”, published for the first time in 1726, was descended from “Lloyd’s news”, a three time a week newspaper created on Edward Lloyd’s initiative in 1696 and given shipping and commercial news thanks to an important network of home and foreign correspondents in the ports. Main periodical publication of Lloyd’s corporation, “Lloyd’s list” was also the most famous in the world of shipping and merchant navy – even evoked in Jules Verne’s adventure novel Around the World in Eighty Days. However, Lloyd’s corporation spread shipping and maritime data since the 1880’s years by means of several periodical publications, each one dedicated to a specific type of information. The first aim of this paper is to present this complex corpus from an historical standpoint, with titles specificity, changes and evolutions. Then, the paper focus on Lloyd’s Shipping Index and Lloyd’s Voyage Record as raw data sources for studying calls, ports and merchant maritime traffic. We will also deal with the set of problems risen by a representative sample of Lloyd’s periodical publications. Their digitization will be approached with the view to extract data with a specific optical character recognition system.

To effectively exploit the “Lloyd’s list” corpus, we must convert its textual content into machine-readable text (for use with Excel, for example). This digitization of the paper documents can be either:

 Manual (a human operator transcribes manually the contents of the documents) ;  Automatic: paper documents are scanned, then the obtained images are processed by a computer in order to « read » their textual content. This automated reading process is called OCR (Optical Character Recognition).

OCR is a complex technology that includes image processing techniques as well as pattern recognition algorithms. Existing OCR software are designed to perform well on a wide variety of documents. However, their performance can be quite low on some documents. Thus the tests we conducted with a commercial OCR software on some samples of the Lloyd's corpus did not give conclusive results. The text transcribed by OCR contains many errors and requires a tedious step of manual correction.

For best results, we decided to design an OCR software adapted to Lloyd's documents, in order to obtain the most efficient recognition (e.g. to minimize the number of errors).

As a first step, we have implemented a set of preprocessing operations designed so as to maximize recognition results on the Lloyd's document images by enhancing image quality and correcting image distortions.

The next step in our work is to create a "recognition engine" that is adapted to our documents. We used an artificial neural network (ANN) to model the shapes of the characters. ANN requires prior learning on a database of image samples: a set of text line images associated with their "ground truth" text. These samples closely reproduce the content of the Lloyd's documents: same font and same image degradation. Once his training is complete, the artificial neural network is able to provide, for each line image, the (most likely) sequence of characters that it contains.

The last part of our work concerns the language modeling, which consists first to identify the different fields in each line (ports of departure and arrival, dates, name of vessel, type of vessel, tonnage …). Then to identify the words likely to appear in each field. Word lists thus created will be used to constrain the recognition in each field: the recognition engine will only output words that appear on the list related to the current field.

Keywords: Lloyd’s corporation; Lloyd’s periodical publications; shipping and maritime data; OCR; image processing; neural networks

Illustration of preprocessing steps: (a) Original grayscale image. (b) Binarization. (c) Header and footer removal. (d) Segmentation in columns (left column). (e) Part of the left column. (f) Lines curvature estimation. (g) Correction of line curvature (dewarping). (h) Detection of text lines borders.

A review of other maritime databases and their potential for research

César DUCRUET, Sébastien HAULE & Nora MAREÏ CNRS, Paris, France

This paper provides a brief overview of a number of sources on port and maritime traffics that are either already been used for research or not yet exploited. Sources are categorized as follows without pretending to be exhaustive, from the least to the most precise in terms of the traceability of flows: archaeological records, overseas flows, origin-destination flows, shipping schedules, shipping routes, ship logs, and AIS data. We also discuss the usefulness of such sources and their possible connection with Lloyd’s List archives. It appears that a wide range of approaches are mobilized, such as climatology, linguistics and genealogy, environmental and topological studies, and globalization-urbanization studies. Finally, the review is extended to port traffic databases having a continental or global coverage.

Changing Mediterranean maritime networks

Nora MAREÏ CNRS, Paris, France

Since the late nineteenth century, with the opening of the Suez Canal, the Mediterranean sea became a crossroad for shipping trade between North and South and between East and West. Port infrastructures have rapidly grown in the basin which today hosts the main global carriers. Few studies have documented this evolution of port and maritime linkages across the basin. By analyzing vessel movements and port statistics evolution, we map and analyse shipping flows accross the whole Mediterranean basin. Confronting global and local dynamics in the formation of port networks and the maritime region this research allows to discuss the relation between global and regional integration.

Keywords: maritime networks; globalization; regionalization; Mediterranean Sea; international passages

Polarization of oceanic areas and shape of the global network of shipping containerized traffic at the dawn of the 1990s

Olivier JOLY University of Le Havre, France

Network structure and regional distribution of global maritime flows (1890-2010)

César DUCRUET & Bruno MARNOT CNRS, Paris, France; University of La Rochelle, France

This research is principally based on the analysis of the worldwide vessel movements reported by Lloyd’s List in its Shipping Index over the period 1890-2008 at different scales such as ports and continents. The first part of the presentation provides a snapshot of the global maritime network’s size and topological structure at selected years (1890, 1925, 1946, 1951, 1960, 1965, 1970, 1975, 1980, 1985, 1990, 1995, 2000, and 2008). The goal of this analysis is to verify major trends affecting maritime flows in relation with the wider economic and geopolitical context in which they take place, using graph theory and complex networks to produce various macro-level indices. Secondly, we map and analyze the geographic distribution of maritime flows among ports and continents that are visualized in a dynamic way, as a means highlighting important regional shifts across the globe, mainly from the Atlantic to the Pacific. A typology (clustering) of port traffic trajectories is also proposed to summarize the individual growth patterns of the connected ports. Lastly, we propose a zoom on the initial period (1890-1925) to get more insights on the link between maritime trade and economic development during the “first globalization”, with more detailed information on the distribution of steamers versus sailing vessels, and the concentration of flows at certain maritime ranges and gateways of the world.

Keywords: complex networks; globalization; graph theory; Lloyd’s Shipping Index; vessel movements

The multiplier attachment: a shipping network architecture

Francesca R. MEDDA & Simone CASCHILI UCL QASER Lab, University College London, London, UK

Countries with disadvantageous remote geographic positions from continents are described as sea-locked because they suffer similar problems as land-locked countries. In fact, small insular countries in the South Pacific, Caribbean and North Atlantic Ocean zones have difficulty connecting to international markets due to their remote locations which is worsened by poor national economies and institutional frameworks. Furthermore, from a logistics vantage point, these countries also have to reckon with rudimentary freight structures that can only handle low levels of out-bound and in-bound freight volumes. Consequently, these sea-locked countries are not yet able to optimize the use of their fleet and guarantee predictability of service. The maritime services that are in operation are irregular; they give rise to very long lead times and in general produce poor quality track-and-trace consignments as well as high costs for inventory and storage services. Port infrastructures are poorly equipped and operate at low levels of productivity.

Against this background, our objective in this study is to verify how specific shipping network structures can stimulate the consolidation of cargo and decrease transport cost by achieving economies of scale. We propose the multiplier attachment, a network architecture based on the modification of the Barabasi-Albert model (Barabasi and Albert, 1999) for spatial networks. Through the multiplier attachment we will analyze the effect of hub allocation and volume consolidation in order to minimize transport cost across the network. Whereas the main assumption of the spatial preferential attachment is that networks exhibit a spatial preferential attachment in which the likelihood of connecting to a node not only depends on the distance but also on a connectivity gain, in the multiplier attachment we assume that the gain is given by the potential benefits that a node may receive from connecting to another particular node. This may include economic benefits, cultural ties, or market opportunities.

We test the multiplier attachment on cargo shipping networks in which nodes can be identified as ports situated in sea-locked countries. Links between nodes represent shipping routes between and among ports.

Results show that the multiplier attachment allows for the emergence of the hub-and-spoke network for the logistics chain. This structure proves to be more effective in achieving economies of scale and fostering agglomeration economy processes. We suggest that governments in sea-locked regions initiate the development of a comprehensive shipping hub-and-spoke structure through the support of regional maritime programs among neighboring countries. Keywords: economies of scale; maritime trade; shipping network

Modelling the emergence of world’s direct shipping lines: an uncertainty analysis

Ronald Apriliyanto HALIM & Lóránt A. TAVASSZY Delft University of Technology, Transport and Logistic group

The world’s trade of the past couple of decades has proven to be very dynamics with the growth in the economies of the developing countries and the stagnation or crisis in the economies of the developed countries. It’s observable over time that the world’s maritime transport network adapts itself to these dynamic changes. The foreseen increase of trade between BRIC countries and the rest of the world, gives an example on how this trend will stimulate and promote the emergence of new shipping lines in the world’s maritime service network. This response from the service network is predicted to occur to provide adequate sea transport services for the growing demand between these developing countries and the rest of the world. Looking at this phenomenon, an insight on how these trends will influence or shape the future service network is highly valuable to deal with future uncertainties. Specifically, we’re investigating on:

 How the service network of the carriers will be restructured by the future demand of freight transport  How the emergence of new service/shipping lines will influence the performance of Ports in a certain economic block (such as EU, US,etc)

In order to answer these questions, we apply a methodology which can give a preliminary broader picture made of all the responses from the shipping line companies to the foreseen trends. Firstly, we examine how direct lines between two ports can emerge as a response to the change in the demand for transport. Secondly, as a study case, we investigate the influence of the emergence of these direct lines to the key performance indicators of the European ports such as throughput and transhipment values. Finally it’s worth noticing that the nature of the model that’s developed in this research is aimed to be used for exploratory research. This means, that the range of investigations that can be done is not limited to growth-related scenarios but also for the opposite scenarios.

Methodology

Step 1: Identifying port to port flow thresholds for availability of a direct connection

Direct connection is a transport service to ship freight from one port to another port that’s conducted without any transshipment or change in shipping service provider. It’s apparent from empirical observations that direct lines have a significant impact on the throughput and transshipment values of a port. As the first step to model the change in the service network, we investigate the threshold value by which shipping companies would provide a direct connection service or a new shipping line between two ports. We obtain this threshold value by observing both the amount of flow and transshipment made in more than 160.000 port-to-port flows (ptpFlows) of which the transport is served by the available service lines. A global strategic network choice model called the World Container Model (WCM) is used to perform the assignment of the trade flows to the worldwide maritime service network from which the data for all ptpFlows in the world can be obtained. By analyzing all the ptpFlows, we can get a flow value above which the transshipment number becomes zero. This is the threshold value.

Step 2: Applying growth scenario to worldwide O-D flows

Next, we use a future scenario where there is a significant increase on the transport demand which takes into account significant economic growth in the developing countries such as those included in BRICS. The increase is varied between 30-140% depending on the forecasted economic trade between the corresponding countries. In this scenario, most of the countries in the world have growth in their export and import activities leading to higher transport demand between countries.

Step 3: Identifying port to port relations with flows above the threshold found

The high growth scenario is used as an input for WCM to perform the assignment of the increased flows onto the current service network. Based on the threshold value found in step one, we implement a method to automatically identify all ptpFlows which values exceeds the threshold value.

Step 4: Creating the direct lines between the ports in the service line network

Based on the identified ptpflows, new service lines are automatically created and added to the current network. The new service line provides a direct connection between two ports which flow is above the threshold value. We assume that the capacity and frequency of the new service line will suffice to accommodate the increased demand.

Step 5: Re-assigning Port to Port flows based on the new service network

Finally WCM is used once more to perform the assignment of the flows from the high growth scenario onto the new service network in which the new service lines have been incorporated. We assume that the old service lines will remain available in the network due to the nature of the growth.

Result

We determine the threshold value by looking at the highest flow value above which the transshipment number for the flow becomes zero. From our study the cut-off value is found to be at 197.903 TEU/year. Based on the threshold value found and the high growth scenario, we identify 528 port-to-port flows which values exceed the threshold value. The emergence of new direct service lines in the worldwide maritime network, under high growth scenario, is found to have a negative impact to Europe’s busiest ports (such as that in Bremen-leHavre range) relative to the scenario where there are no new direct service lines. It is Interesting to observe that, in terms of throughput, Rotterdam isn’t severely impacted from by the emergence of the new shipping lines. While the rest of the ports in Europe suffer negative impact in their throughput, Rotterdam has a positive throughput change. However, in terms of transshipment, Rotterdam’s performance is the most negatively impacted. This is followed by port of Le Havre and port of Bremen. The negative change of transshipment values suffered by port of Rotterdam indicates that in the scenario where there’s a high growth in trade between countries, many of the emerging shipping lines would not have transshipment in Rotterdam. There might emerge new transshipment hubs for these new lines driven by the growth in world’s trade.

Keywords: global shipping lines; maritime service network; route choice; uncertainty

Random walks in dynamic maritime networks

Frédéric GUINAND & Yoann PIGNÉ LITIS Laboratory, University of Le Havre, France

Let G=(V,L) a graph and let S_G an ordered sequence of its subgraphs S_G = G_1,G_2,...,G_T. The System G_E=(G,S_G) is called an evolving graph. An evolving graph can also be represented as a graph F=(V,L,T) where V is a set of vertices, L a set of links and T a set of dates. Each edge (resp. vertex) is valuated by a set of dates indicating the presence of the edge (resp. vertex) in the corresponding subgraphs. In this work we study maritime traffic, based on real datasets, using evolving graph models. In particular we perform various analyses in relation with connectivity and with biaised random walks.

The complex network of cargo ship movements and its importance in marine bioinvasion

Michael T. GASTNER Institute of Technical Physics and Materials Science, Research Centre for Natural Sciences,

Hungarian Academy of Sciences, P.O. Box 49, H-1525 Budapest, Hungary

Figure 1: The routes of all cargo ships bigger than a gross tonnage of 10 000 during 2007. Brighter colours indicate shipping lanes with more tra_c. Ships are assumed to take the shortest (i.e. geodesic) routes. Reproduced from [1].

Cargo ships do not only move merchandise around the world, but can also transport unwanted stowaways: potentially harmful invasive species hitching a ride in the ballast tanks or clinging to the ship hull. Upon arrival in a new port, the alien species can then drive native species to extinction, modify entire ecosystems and impact human economy. The knock-on effects to fishermen, farmers, tourism and industry create billions of dollars in damage every year [2]. The spread of invasive species takes place on the global network of cargo ship movements where the nodes are ports and the links are nonstop ship journeys. The prediction of the invasion risk thus consists of two parts. (1) We represent records of ship movements as a complex network (Fig. 1). (2) We use this network as input to models of invasive spread. In this talk, we will highlight the main properties of the network and introduce a general model for invasion probabilities. We identify high-risk invasion routes, hot spots of bioinvasion, and major source regions from which bioinvasion is likely to occur. Comparing our prediction with observations in the field, we argue that network-based invasion models may serve as a basis for the development of effective, targeted bioinvasion management strategies [3].

Keywords: complex network; cargo shipping; marine bioinvasion

References:

[1] P. Kaluza, A. Kolzsch, M. T. Gastner and B. Blasius, J. R. Soc. Interface 7, 1093 (2010).

[2] D. Pimentel, L. Lach, R. Zuniga and D. Morrison, BioScience 50, 53 (2000).

[3] H. Seebens, M. T. Gastner and B. Blasius, Ecol. Lett. 16, 782 (2013).

Maritime network vulnerability to cascading failures

Serge LHOMME Ecole des Ingénieurs de la Ville de Paris (EIVP), Paris, France

The main objective of this presentation is to assess global vulnerability of the world maritime network and to identify the most critical harbors. In the same time, we hope to demonstrate that vulnerability measures can be commonly used to determine the importance of vertices or edges in a graph. Three types of vulnerability measures has been computed: basic measures which assess world maritime network vulnerability to a single node breakdown (in a static way); dynamic measures which assess network vulnerability to cascading failures (starting from a single node breakdown) requiring flow redistribution rules after nodes breakdown; attack and error measures which investigate network vulnerability to the removal of a group of nodes. We show that world maritime network is relatively resilient. Nevertheless, world maritime network is quite vulnerable to the breakdown of the most important harbors (particularly some Asian harbors). Finally this research shows that some European harbors are more critical than expected.

Maritime network evolution under crisis: The case of North Korea

Stanislas ROUSSIN ASEM Institute, Seoul, Republic of Korea

Although maritime transport plays a rather secondary role in North Korea’s development, all its largest cities are settled along its coasts and therefore use a port to connect the outside world. This study particularly underlines the changing spatial distribution of maritime connectivity in the context of a rapidly and profoundly declining economy, based on a database on worldwide vessel movements available between 1985 and 2010. Firstly, we observed a continuous shipping activity over the period, despite North Korea’s reputation of a fully isolated country. In addition, the share of maritime transport in North Korea’s international trade, albeit modest (about 15%), has remained stable all over the period. Secondly, infrastructure dereliction, trade embargo, economic decline, and political isolation have had profound changes in the country’s internal and external organization. Vessel traffics have drastically shifted from East coast to West coast ports, thereby concentrating upon Nampo, Pyongyang’s gateway. This trend continues nowadays, reflecting upon the prolonged industrial decline and lowering land-sea accessibility of the eastern provinces, but also the reinforced economic and demographic polarization of the capital region. Externally, a parallel shift has occurred, in accordance with the overall geographic shrinkage of North Korea’s overseas connections since the fall of the USSR and the Socialist Block in the late 1980s and early 1990s. In contrast, North Korean vessels have increasingly sailed through other neighboring hubs, such as Dalian (China), Vladivostok (Far-East Russia), and several Japanese ports (e.g. Niigata). Foreign vessels (and notably containers) have gradually concentrated their movements between North and South Korea, the second becoming the hub of the first, based on spatial proximity, improved inter-Korean relationships, and a parallel reduction of the importance of Hong Kong, Singapore, Suez Canal, and Panama Canal in ensuring North Korea’s long-distance relationships. Interestingly, and as a reflection of North Korea’s diplomatic pragmatism, economic factors have superseded political factors in the recent transformation of its maritime linkages.

Keywords: Democratic People’s Republick of Korea (DPRK); sea transport; port system

Impacts of Northern Sea Route on natural resources logistics map in East Asia: South Korea’s case

Sung-Woo LEE Korea Maritime Institute (KMI), Seoul, Republic of Korea

Climate change and global warming makes Northern Sea Route open up for the Arctic transportation. It is expected that the commercialization of the Northern Sea Route is likely to have stronger impact on natural resources transportation in Siberia and Russian Arctic region rather than the container logistics connecting northern Europe with Asia. If the natural resources transportation using the Northern Sea Route is tangible, a huge change in the existing maritime logistics networks, on which “troika of Northeast Asian economy”—China, Japan, and South Korea—are heavily dependent in their imports, is unquestionable. In Korea’s case, the opening up of the Northern Sea Route will accelerate the change of natural resources logistics network rapidly, especially in terms of natural resources imports such as iron ores—dependent on Brazil and Indonesia, coal—dependent on Indonesia and Australia, and oil—dependent on the Middle East.

In this respect, this study forecasts the impacts of the Northern Sea Route on the change of South Korea’s maritime logistics network related with natural resources, and comparing the competitiveness of the Northern Sea Route with the existing maritime logistics networks used mainly by South Korea. In addition, the possibility of the Northern Sea Route’s future commercialization is analyzed based on the profitability analysis of the logistics costs when the existing natural resources logistics networks are replaced by the Northern Sea Route focusing on Russia’s natural resources.

Keywords: Northern Sea Route; South Korea; natural resources logistics; logistics cost analysis

Explaining international trade flows with shipping-based distances

David GUERRERO Institut Français des Sciences et Technologies des Transports, de l'Aménagement et des Réseaux (IFSTTAR), France

This paper analyzes how the use of shipping-based distances improves the explanatory power of a gravity model of trade between countries. Its aim is to understand how the shipping network is linked with international trade. The gravity model is the dominant approach in empirical studies of international trade, since it delivers a good explanation of the variation in bilateral trade flows across a wide variety of countries and periods. Flows between two countries are considered to increase with their size (ex. GDP’s) and to decline with trade costs (ex. Distance between countries). The latter are represented most of the time the distance as the crow flies between countries. Few research works, include more complex measures as for example topological measures (distinguishing trade flows between countries that share a common border) and none of them integrate transport services. The purpose of this paper is to contribute to fill this void, by testing measures of distance that are based on the shipping services between countries as a proxy of trade costs. Our main hypothesis is that the functional distance delivers a better explanation of trade links than the spatial distance. We use homogeneous data on world trade obtained from Chelem database. Shipping-based distances have been calculated using the Lloyd’s Register database about the movements of ships. The latter are based on the number of weekly opportunities to send cargo between two countries in the same ship. To compare the explanatory power of the different measures of distance, a doubly constrained spatial interaction model has been used. Our results show that shipping based distance improves (considerably?) the explanatory power of the gravity model of international trade. This seems to indicate that the existence of frequent transport services between countries is determinant to explain their trade flows.

Keywords: international trade; gravity model; shipping networks; distance

Think inside the box

Jean-François PELLETIERa & Yann ALIXb aCPCS Canada; bSEFACIL Foundation

In this fast moving and changing era where originality and novelty are often considered to be some of the drivers for success, it is usually suggested to “think outside the box”. This metaphor also applies to the analysis of containerized networks and flows but paradoxically, such an approach requires to “think inside the box”. Or rather, to think about what is inside the box.

In 2004, Pelletier and Alix suggested that adopting such an approach in the analysis of containerized flows could provide a wealth of information which is currently hidden by simple analysis of box flows. Building on this work and on recent academic literature on containerization flows, the authors will develop this approach further.

This paper proposes to begin by a detailed description of the methodology. An appraisal of its applicability to global containerized flows will then follow. To illustrate the potential of the proposed methodology, the authors will then analyze Canadian container flows from 1983 to 2011 though a content perspective. This will provide insight and demonstrate how products contained in the box, and not the box itself, are major drivers of the underlying factors that determine containerized flows, trade routes and the relative position of container ports on specific trades.

Pelletier, J.-F. et Alix, Y. 2004. Vers une nouvelle approche méthodologique pour analyser le phénomène de la conteneurisation des marchandises au Canada. Dans : GRTC, Actes de la 39e Conférence annuelle. Vol. 1 pp. 326-341.

Keywords: container; product; flows; methodological approach; Canada

Visualizing regional structures in global maritime flows

Françoise BAHOKEN Institut Français des Sciences et Technologies des Transports, de l'Aménagement et des Réseaux (IFSTTAR), France

Stochastic blockmodeling applied to global maritime flows

Charles BOUVEYRON, Pierre LATOUCHE & Rawyia ZREIK Laboratoire MAP5, Université Paris Descartes, France

Laboratoire SAMM, Université Paris 1 (Panthéon-Sorbonne), France

We consider two statistical methods for network analysis: SBM and RSM. The stochastic block model (SBM) is one of the most popular approaches to model and cluster the nodes of a network. Its flexibility allows it to model a wide variety of networks (communities, stars …) and summarizes the network with a few interpretable model parameters. The random subgraph model (RSM) is a more recent approach which extends SBM to networks with categorial edges and for which a partition into subgraphs is known (a geographical partition for instance). Thus, RSM allows to find clusters which correspond to real latent groups and not to obvious (geographical, for instance) communities. We finally present the application of both SBM and RSM to a maritime flow network.

Maritime trajectory patterns

Laurent ETIENNE & Thomas DEVOGELE Naval Academy Research Institute, Brest, France; University of Tours, Blois, France

Nowadays, an abundance of sensors are used to collect very large data sets of moving objects. To analyse and understand the path of these objects, cluster of trajectories must be de_ned and pattern of each cluster could be discovered. In this article, we introduce different techniques to generate a pattern that summarize a set of trajectories. This one is based on statistical analysis of point clusters. This pattern is useful to analyse sets of trajectory, understand the spatio-temporal density of trajectory and detect outliers. A large real world data sets illustrate this article. A visual analysis highlight how the pattern performs when describing trajectory clusters density change over time.

1.1 Problem statement

Given a set of trajectories, the goal of the Trajectory pattern (TP) is to summarize the spatial and temporal distribution of the trajectory set. The first issue consist in defining a representative trajectory depicting the usual movement performed by the trajectory set (central tendency). The second problem tackled in this article focus on the spatio-temporal dispersion of the trajectory set around the central tendency. Finally, an effective way to visualize the trajectory pattern need to be defined. This visualisation must give a feedback to the user about the central tendency, the trajectories spatial and temporal dispersion and symmetry without cognitive overload.

1.2 Motivation

Ships are mobile objects that moves in a low-constrained open space. The movements of these mobile objects can be captured using sensors and tracking systems such as radars or the Automatic Identification System (AIS) [4]. Considering that a trajectory describe the movement behaviour of an object, trajectory clusters can be used to define the usual behaviour of object that moves in a similar way (following a same path, moving together...). An analysis of the spatial and temporal distribution of the trajectories give important feedback about the global behaviour of the mobile objects evolving in one area. Such patterns can then be used to specify usual behaviour of mobile object in an open area. An efficient visualisation of these patterns can also ease the detection of outliers.

1.3 Related work

Spatio-temporal clustering can be defined as a process of grouping objects based on their spatial and temporal similarity. Considering trajectory clustering, many di_erent works focus on the definition of similarity functions between trajectories to partition them into clusters based on distance or density [5]. To model the spatio-temporal behaviour of a trajectory set, some methods targets on deriving a global model capable of describing the whole data-set. These methods usually rely on a definition of a multivariate density distribution and try to fit parameters to the model. Gaffney and Smyth [3] proposed a clustering method based on a mixture model for continuous trajectories which were represented as functions. The trajectory cluster is considered as being generated from a core trajectory mixed with a gaussian noise. Chudova et al [2] consider the spatial and temporal shift of trajectories within each cluster. Alon et al [1] proposed an approach based on a model-based technique where the representative of a cluster is expressed using a Markov model. This Markov model estimates the transition between successive positions. The parameter estimation of the model is often performed using an EM algorithm.

1.4 Challenges

There are different research challenges related to the definition of trajectory patterns. To cluster similar trajectory together, the notion of similarity (distances) between trajectories must be defined. A first challenge consist in comparing trajectories together taking both space and time into account. Then, the central tendency of a trajectory set has to be modelled, this central tendency can be either, a real trajectory of the set having minimizing the dissimilarity with other trajectories of the set, or a virtual trajectory generated from positions clusters. Once the central tendency defined, the spatial and temporal spreading around this central tendency must be modelled. Finally the statistical spatio-temporal density of the model have to be efficiently visualized in order to ease traffic monitoring operator understanding of the model.

1.5 Contributions

This paper makes the following contributions claims: (1) we adapt trajectory partitioning techniques to generate ordered set of position clouds (2) we introduce a new concept to summarize the spatial distribution of a point cloud (3) we extend this concept to 3D space-time cube for position clouds (4) we define a Trajectory pattern that visually synthesize the spatio-temporal behaviour of a trajectory set. (5) experimental evaluation of the proposed algorithms on a real world maritime dataset.

Keywords: trajectory; data mining; patterns; AIS; maritime; spatio-temporal

References:

[1] Jonathan Alon, Stan Sclaroff, George Kollios, and Vladimir Pavlovic. Discovering clusters in motion time-series data. In Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition, pages 375{381. IEEE Computer Society, 2003.

[2] Darya Chudova, Scott Gaffney, Eric Mjolsness, and Padhraic Smyth. Translation-invariant mixture models for curve clustering. In Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining, pages 79{88. ACM, 2003.

[3] Scott Gaffney and Padhraic Smyth. Trajectory clustering with mixtures of regression models. In KDD '99: Proceedings of the Fifth ACM SIGKDD international conference on Knowledge discovery and data mining, pages 63-72, New York, NY, USA, 1999. ACM.

[4] IALA. The automatic identification system (ais), volume 1, part i, operational issues. Technical report, International Association of Marine Aids to Navigation and Lighthouse Authorities, 2004.

[5] J.G. Lee, J. Han, and K. Whang. Trajectory clustering: A partition-and-group framework. In SIGMOD '07: Proceedings of the 2007 ACM SIGMOD international conference on Management of data, pages 593-604, New York, NY, USA, 2007. ACM.

The Weastflows Project: freight transport spatial interactions and territorial dynamics in North West Europe

Juliette DUSZINSKI, Jean-François MARY, Lucile AUDIEVRE & Mathilde MUS Agence d'Urbanisme de la Région Havraise (AURH), Le Havre, France AURH as urban planning agency of Le Havre and the Seine Estuary area is a french non-profit organization located in Le Havre (Normandy). Town planning and geography are its core business to assist cities and local authorities in their decisions making. AURH works with different territory extends, at different stages and different timelines, for the future of the Seine Estuary and the Seine Valley. Since 2011, AURH has been involved in an Interreg IVB European project called “Weastflows” (contraction of West, East and freight flows). Weastflows focuses on freight flows and infrastructures network for the North-West Europe area, all goods (containers, bulk liquids, and grains) and all modes, roads, railways, waterways, airports, and maritime links within this scope. The aim of this project is to improve freight transport and logistics in the North-West Europe by developing west-east corridors. AURH is in charge of one of the twelve actions of the project: the Action 1 untitled “Assessment of sustainable transport capacity”. One of the works of AURH in the Weastflows project is to analyze the sustainability of the existing transport network and the potential connections/synergies between the partners regions considering their economic features and logistics flows. This second topic aims to provide a spatial analysis of freight flows in the North-West Europe area by using a matrix of all goods interactions between seventy six geographical zones in the world. This matrix was produced by SEStran (South East Scotland Transport Partnership) another Weastflows partner and performed for every mode: road, rail, inland waterways and maritime waterways, with the “tons” unit and the “ton-kilometer” unit. AURH proposed an added value by integrating the data produced in a Geographic Information System to provide a spatial analysis of the freight interactions within the North-West Europe area and outside. It gives some precious knowledge at regional level on issued and received flows according the different transport modes. It mainly allows us to identify the significant interactions between North-West Europe and the other areas and between regions inside North-West Europe, especially the maritime interactions. For further information, please do not hesitate to visit: The official website of the AURH: www.aurh.fr The official website of Weastflows project: www.weastflows.eu The website of an innovative map tool: geo.weastflows.eu Our blog: www.aurhinweastflows.com

The Eurasian railway corridor and Lower Seine Valley ports: competition or complementarity?

Jérôme VERNY & Milhan CHAZE MOBIS Research Institute, NEOMA Business School, Rouen, France

As part of a general lowering of customs barriers and the gradual opening up to international trade in Central Asian countries, the increasing trade between Europe and East Asia invites to be interested in new international trade routes. If the sea is, for several centuries, the main route of goods transportation from Asia to Europe, and vice versa, the rail corridor connecting the two continents could arise as a new contemporary "Silk Road". The infrastructure development and rail services between Asia and Europe could contribute to the expansion of port hinterlands, including the Seine Valley. Thus the question raises concerning to the competition in opposition of the complementarity between in the one hand the Eurasian railway corridor and in the other hand the traditional good entrance and exit doors, represented by the Seine Valley seaports. Can the establishment of rail solutions linking Western Europe, especially Seine Valley ports (Haropa) to East Asia, be an opportunity to develop the catchment area of the three ports Paris -Rouen -Le Havre (Haropa) to Eastern Europe in addition to Central and Eastern Asia? What will be the impacts of these new transcontinental land transport services offers on maritime trade routes?

It stems from this problem more specific questions: is the Eurasian rail corridor actually accompanied by a transcontinental door to door service offer or rather an intra-regional axes succession, both on infrastructural and servicial dimensions? Between Asia and Europe, could the introduction of transportation and specific logistics participate in a maritime and land trade routes specialization according to product families? Could the complementarity between the Eurasian rail corridor and shipping between East Asia and Seine Valley ports be possible?

To carry out this study in the work context of the new economic geography, we will start from the assumption that the rail corridor between Asia and Europe is not able to supplant the sea, but however, it can assert itself as an alternative international transportation way for some products families. In particular, it will analyze the setting up possibility of a delayed differentiation along the rail corridor, which would illustrate the emergence of the new value chain geography.

Modeling a system of cities based on spatial networks

Igor LUGO National Autonomous University of Mexico (UNAM), Regional Center for Multidisciplinary Research (CRIM), Mexico

A system of cities is a common concept used in different disciplines to explain a network configuration among urban areas. However, its geometric representation in theoretical and applied studies is not trivial. For example, one of the main problems in the field of economic geography is the symmetry in shapes, where cities are nodes interrelated by the distance in a circular configuration. On the other hand, transport networks models complex spatial structures, but they are isolated to city attributes. To date there has been little attention between disciplines to encompass a common framework for modeling real urban systems. Therefore, the purpose of this paper is to present a process for creating a 2D geometric representation of a system of cities connected by transportation modes based on geospatial data. The results of this study indicate that spatial networks could represent any type of geographic objects and their interrelations in space-time, and they suggest an interdisciplinary approach to cross-fertilize perspectives.

Keywords: system of cities; spatial networks; economic geography; transport networks; geospatial data

The British and Japanese shipping network in China in 1920s

Liehui WANGa & Lei YANGb aCenter for Modern Chinese City Studies, East China Normal University ,Shanghai; bCollege of History and Social Development, Shandong Normal University, Jinnan

After the first Opium War in 1840, China began to open and trade with foreigners. In 1920s, British and Japanese maritime occupied the top two in China. This paper uses social network analysis software GEPHI and GIS software ARCGIS to analyze on the characters and mechanism of the British and Japanese maritime network in China in 1920s. Firstly, based on some analytical tools of social network analysis such as degree, betweenness centrality, community, the paper studies the coverage, structure and connection of the two countries’ maritime network. For the coverage of the routes, Britain connected the ports in the South China more closely and Japan linked the ports in the North China more closely. Shanghai and Hong Kong are both important hubs in British network. Based on Shanghai, single- nuclei is the outstanding character in the Japanese maritime network. Secondly, the structure and character of the two countries’ maritime networks depend on three factors: geography, history and policy.

Keywords: network; structure; mechanism; Britain; Japan

Canton-Hong Kong-Macau: a historical review on the intercity competitions and interactions of the three commercial centers of the Pearl River Delta before the 20th century

Ka-chai TAMa and Adolf K.Y. NGb aDepartment of History, Hong Kong Baptist University, Hong Kong, China; bDepartment of Supply Chain Management, I.H. Asper School of Business, University of Manitoba, Winnipeg, Manitoba, Canada

This study examines the changing roles and relative positions of the three most prominent cities of the Pearl River Delta (PRD) in South China, namely, Canton, Hong Kong and Macau in the commercial networks and transportation system before the commencement of the 20th century. The triad of Canton-Hong Kong-Macau (Sheng-Gang-Ao) is a concept introduced only in the early 20th century which emphasized the cooperation among these Cantonese cities. In the course of their development, however, the division of labours had by no means been well defined. There were keen competitions amongst them, and the success of one usually meant the failure of the others. Understanding such, our first objective is to identify the relative roles of the three cities in different stages from late 16th century down to the end of the 19th century, with special focus on the effects of the ascendance of different Western Powers to the domestic city networks of the Guangdong Province of China. Following our understanding of the roles of the triad in different times, our second objective is to rebuild the changing transportation networks surrounding them over time by employing the geographical information system (GIS) techniques. Our final objective, therefore, is to analyze how institutional factors, i.e., political crisis, civil wars, colonial rule and imperialist invasions, contributed to the rise and fall of some nearby cities of similar conditions within a relatively small region. Through exploring the historical development of the tri-city of Canton, Hong Kong and Macau, the study aims to provide a firm background to evaluate their interactions in the 20th century and at giving good evidences to forecast the future competitions and cooperation amongst different port cities of the PRD.

A geospatial analysis of maritime piracy and its challenges

Philippe LEYMARIE

Journalist, Le Monde Diplomatique

Maritime piracy, despite its relative decline in the eastern and western parts of the Indian Ocean, is increasing around the Gulf of Guinea. In a report made by request from the UNITAR, a United Nations division, four French journalists and cartographers have proposed to refine statistical data related with maritime piracy, by delineating its sub-regions, introducing a “severity index” for incidents, and a distance criterion (distance to coast) to better highlight recent evolutions.

Keywords: maritime piracy; UNITAR; severity index; distance to coast

Kamel AIT-MOHAND Post-doctoral Research Fellow (CNRS, UMR 8504 Géographie-cités), Kamel Ait-Mohand received his Ph.D. degree in computer science in 2011 from the LITIS laboratory (Rouen, France), where he has worked on optical word recognition on ancient documents. He is in charge of the development of a document recognition system that will be used to extract the textual content from Lloyd's publications.

Yann ALIX Since 2011, Dr Yann Alix is the General Delegate of the Foundation SEFACIL which aims to set up, lead and animate a unique international think-tank of excellence around strategic & prospective analysis on themes related to the future of Maritime Transport, Ports& Logistics. He is Associate Researcher for Logistel in Portugal and collaborates actively in different training & engineering projects in Africa for the Port Authority of Le Havre. After receiving his PhD from Concordia University and his Doctorate in Transport Geography from Caen University, he started a career as consultant for Maritime Innovation in Rimouski, leading applied researches and studies for the Canadian & International port authorities, shipping companies and logistics entities. His main research issues focus on Corporate Management & Strategies in the Port & Maritime industries as well as port-city relationships and freight inland corridors/transportation chains, especially on the African, Chinese & South-American continents.

Lucile AUDIEVRE Geographer and urban planner, Lucile Audièvre, joined the AURH team since 2010. She is Project manager and developed an expertise on mapping exercise, Geographic Information System and projects of territorial development.

Françoise BAHOKEN Geographer and research engineer at the French Institute of Transport Sciences, Technologies and Networks (IFSTTAR), her research activities include network analysis and the analysis and visualization of statistical and spatial data and information.

Marc BARTHELEMY Marc Barthelemy is a former student of the Ecole Normale Superieure of Paris. In 1992, he graduated at the University of Paris VI with a thesis in theoretical physics titled "Random walks in random media". After his thesis, Marc Barthelemy focused on disordered systems and their properties. In 1999, he visited Prof. Stanley at Boston University and started to work on the properties of complex networks. Since 1992, he has held a position at the CEA (Paris) where his interests moved towards applications of statistical physics to complex systems. In particular, he worked on complex networks, theoretical epidemiology, and other problems in urban systems.

Charles BOUVEYRON Charles Bouveyron is Professor of Statistics and head of Department of Statistics at Université Paris Descartes, Paris, France. He received in 2006 the Ph.D. degree from Université Grenoble 1 (France) for his work on high-dimensional classification. In 2006–2007, he was a postdoctoral researcher in the Department of Mathematics and Statistics of Acadia University in Canada where he worked on the statistical analysis of networks. Then, he was Assistant Professor (2007-2012) and Associate Professor (2012-2013) at Université Paris 1 Panthéon-Sorbonne. His research interests include classification of high-dimensional data, classification under uncertainty and weak supervision, adaptive and online learning as well as network analysis.

Anne BRETAGNOLLE Anne Bretagnolle is Professor of Geography at University Paris 1-Panthéon-Sorbonne. Its main research topics concern cities and towns, which are studied through a historical and international approach. Emphasis is given to the emergence and evolution of systems of cities and to the effects of transportation networks on cities and towns dynamics (France, Europe and United States, from 16th to 20th century). Along with these theoretical investigations, she explored modelling aspects, in particular the dynamics of cities by using multi-agent systems (Europe/United States 17th-20th centuries) or the co-evolution between urban systems and transportation networks by using graph theory (France, 17th-19th cent., South Africa 19th-20th cent.). Emphasis is given to the research of comparability in city definitions in the long term period, by constructing harmonized perimeters of urban agglomerations and urban functional areas.

Simone CASCHILI Simone Caschili, enviromental engineer, received his Ph.D. in Ingegneria del Territorio (Land Engineering and Urban Planning) at University of Cagliari (IT) in February 2010. His thesis (Complex Networks and Spatial Analysis: an Integrated Approach for Community Detection) explores the linkages between Complex Network Theory and Spatial Planning proposing a contribution to the issue of “regionalization”. His research interests are the modelling of urban and regional systems as complex networks, the inclusion of time-space features in network modelling and policy evaluation for planning in both urban and environmental governance. He developed the GIS tool ComplexNetGIS, an ArcGIS® integrated tool. It provides several network functions and indices taken from Complex Network Theory and transport analysis. Since 2007 he has been a teaching assistant for the Spatial Planning course at the Faculty of Engineering in Cagliari. From August 2006 to September 2010 he worked as a consultant at the Urban Planning Office of the Autonomous Administration of Sardinia and supported the maintenance of the regional landscape plan. Since October 2010 he has been appointed at Centre for Advanced Spatial Analysis– University College London – as Research Associate for the ENFOLD- ing Project. He is in charge of the trade and economic development work stream where he directs his research interests (under the supervision of Francesca Medda and Alan Wilson) on the application of Complexity theory and Complex Adaptive Systems to the

study of international trade and maritime shipping. A substantial part of his research interest regards real time data acquisition, data manipulation and visualisation. Since November 2010 he is a senior research fellow at the UCL QASER Lab.

Milhan CHAZE Dr. Milhan Chaze is currently a research engineer at MOBIS-NEOMA Business School (Rouen). He hold a Phd in Geography, Land Planning and Urban Planning from the University of Clermont-Ferrand II-Blaise Pascal (CERAMAC). Milhan is specialized in economic geography, especially trade issues. After working on various programs of basic research (VPM / AttracVil) and applied Observatory (Highway A89), he is currently working on several research projects in logistics and transportation, including RISC and FASE international projects, with the aim to understand the territorial organization evolution of freight transport corridors in the context of trade globalization. Commission member of the Trade Geographic of the National French Geography Council (CNFG), Milhan is the author of a dozen of publications and communications in national and international conferences.

Thomas DEVOGELE Full Professor and member of the Computer Science Laboratory (LI) at Tours University (EA 6300), in the team BDTLN (Base de données et traitement des langues naturelles). He was associate professor at the French Naval School (Ecole Navale) between 1998 and 2010 after receiving his Ph.D. from the National Geographic Institute (IGN-COGIT) in 1997.

César DUCRUET César is geographer and research fellow at the French National Centre for Scientific Research (CNRS) since 2009, after having receiving his PhD from Le Havre University (2004). From 2005 to 2008, he was post-doctoral research fellow in South Korea (Korea Research Institute for Human Settlements) and the Netherlands (Erasmus University Rotterdam, Marie Curie). His research focuses on port and maritime geography as well as network analysis.

Juliette DUSZYNSKI Geographer and urban planner, Juliette Duszynski, joined the AURH team since 2002. She is Project leader and in charge of action 1 in Weastflows project. She developed an expertise on economics, logistics and port issues including in the framework of European Projects.

Laurent ETIENNE Dr. Laurent Etienne did a postdoctoral fellowship in geomatics at the University of Dalhousie in the Industrial Engineering department. He worked with Pr. Ronald Pelot in the MARIN research group. He tought two undergraduate courses to Industrial Engineering students (Algorithms and Computational Methods, Geographic Information Systems). Dr. Laurent Etienne was a previously a research assistant at the French Naval Academy Research Institute from 2008 to 2012. His main research interests are spatial data mining and knowledge discovery from moving objects databases, spatio-temporal patterns of mobile objects, outlier detection and maritime Geographic Information Systems (GIS). Dr. Etienne is involved in different projects related to knowledge extraction and analysis of movement such as the LoCoss project, which is part of the French Intelligent Transport Systems research cluster and aims to geo- localise, analyse and optimise movements of firemen in duty. He is also implicated in the Reconsurve European project (Reconfigurable Surveillance System with Smart Sensors and Communication). This project addresses the need to control the rapidly increasing number and complexity of maritime-surveillance issues such as illegal immigration, interoperability between heterogeneous systems and automated cost-effective and efficient decision support to improve maritime security. Dr. Etienne recently obtained his PhD in geomatics from the University of Brest, France. His PhD research were achieved in the Geographic Information Systems group of the French Naval Academy Research Institute (IRENav). The topic studied during this PhD was about spatio-temporal data mining and classification of ships’ trajectories. Previously, he created his own company and worked as a research engineer in geomatics. He also holds a research master’s degree (M.Sc) in computer science of the University of Rennes 1, France. Dr. Etienne was also an assistant lecturer in computer science at the French Naval Academy. His teachings topics are computer architecture, operating systems, algorithms and programming, computer graphics and virtual reality, computer-human interaction, geographic information systems, databases.

Tim EVANS I am a Senior Lecturer (equivalent to an associate professor) in the Theoretical Physics group and part of the Complexity and Networks programme at Imperial. I did my first degree in Natural Sciences (Physics) at Cambridge followed by a PhD here at Imperial, supervised by Ray Rivers. I then spent time as a researcher at the University of Alberta in Edmonton Canada, followed by research positions back here at Imperial, in part as a Royal Society University Research Fellow.

My main interest is in the behaviour of many body systems both in and out of equilibrium. I act as the Imperial node of the EPSRC NetworkPlus collaboration on Emergence and Non-Equilibrium Physics whose title is a pretty good summary of my interests. Currently I am interested in ideas falling under the broad area of complex systems. In particular the properties of Complex Networks, such as the "six degrees of separation", intrigue me. From a theoretical perspective, I am interested in community detection in networks and networks constrained by time or space. In terms of applications to practical problems I look at many different areas but I am particularly

interested in social systems such as bibliometrics (citation patterns and related problems), cultural transmission and archaeology. For instance I have an ongoing project which emerged out of meeting in Venice in November 2002 of an intriguing EU funded network, ISCOM, involving economists, geographers, physicists and mathematicians. I am part of the Connected Past team organsing events to develop the use of ideas from complexity and networks in the context of archaeological and historical problems.

A slightly unusual direction has been my work to develop new visualisations of complex systems using 3D printing. This led to my recent paper Scuplexity: Sculptures of Complexity using 3D printing, see my blog entry about 3D printing and Complexity for more information and links about this work.

I'm have also worked on Quantum Field Theory. My main focus was on many-body problems, a topic known as Thermal Field Theory or Finite Temperature Field Theory. I also studied generic features such as the nature of multiplicative anomalies and their implications for zeta-function renormalisation.

Mélanie FOURNIER Doctor in Geography, Mélanie Fournier after many years working for the French Ministry of Defense, the EU, the Civil Aviation in New Caledonia and the UN as GIS expert and imagery analyst, started in January 2013 as a postdoc within the Digital Humanities Lab at the EPFL under the direction of Frédéric Kaplan to model, among others topics, the old Venetian maritime routes.

Antoine FREMONT Antoine Frémont is geographer, in charge of spatial planning at the French Railway Network Organization (RFF). He has been research director at the French Institute of Transport Sciences, Technologies and Networks (IFSTTAR). His research activities focus on the maritime transport industry in relation with globalization.

Michael T. GASTNER After my PhD in physics from the University of Michigan, I held postdoctoral positions at the Santa Fe Institute, the Universitat Oldenburg and Imperial College London. After a lectureship in the University of Bristol’s Department of Engineering Mathematics, I am currently a Marie Curie Fellow in the Complex Systems Group of the Hungarian Academy of Sciences in Budapest.

François GIPOULOUX François Gipouloux is Research Director at the French National Centre for Scientific Research (CNRS) and the Research Center on Modern and Contemporary China (CECMC), and Director of the laboratory China-Korea-Japan (UMR 8173 Chine, Corée, Japon).

He acts as coordinator and scientific director of URBACHINA: Sustainable Urbanisation in China, Historical and Comparative Perspectives, Mega-trends towards 2050 (7th PCRD). His research topics include the formation of an Asian Mediterranean area and the fragmentation of economic areas in China and Japan.

Karl GROSSNER Karl Grossner is a geographer (Ph.D. 2010) working at Stanford University Libraries as a digital humanities research developer, consulting and collaborating with humanities and social science faculty seeking to integrate digital mapping and modeling methods in varied scholarly projects. His own research focuses on modeling historical knowledge, computational models of place, and the emerging genre of digital historical atlases. (http://kgeographer.org)

David GUERRERO David Guerrero is researcher at the French institute of science and technology for transport, development and networks (IFSTTAR). He holds a Ph. D. in economic geography from the Paris - 7 Paris Diderot University. His research in the Team « Production Systems, Freight, Logistics and Work » sets out to understand the complex interdependencies between productive systems and transportation, with particular focus on spatial issues. This leads him to develop two parallel axes respectively focusing on the individual transport choices of shippers (ECHO Quantitative Surveys) and on qualitative research field on the spatial organisation of freight flows, at French and world level, following the PhD works on Port Hinterlands and Forelands. He teaches at Ecole Supérieure des Transports in Paris.

Frédéric GUINAND Dr. Frédéric Guinand received his Master degree in Computer Science in 1991, and the PhD degree in Computer Science in 1995 both from the University Joseph Fourier (Grenoble – France). After being employed as an engineer in a startup in 1996, he then pursued researches at the Swiss Federal Institute of Technology of Lausanne (EPFL – Switzerland) thanks to an INRIA postdoctoral grant. In 1997 Dr Frédéric Guinand joined Le Havre University as Assistant Professor and he was appointed to a Full Professor of Computer Science at Le Havre Institute of Technology in 2005. His current research interests are in dynamic graph theory, distributed and mobile computing, and bio-inspired computing for optimization under uncertainties. He has published over 80 papers in peer- reviewed international journals and conferences. He has also co-edited the book “Artificial Ants” (Wiley, 2010). He is currently the Head of a research team within the NormaSTIC federation (CNRS FR 3638).

Ronald Apriliyanto HALIM Ronald Apriliyanto Halim is a PhD candidate in the Transport, Logistic, Organisation section at Delft University of Technology, faculty Technology, Policy and Management. His research is focused on the development of a global freight logistics model which will be used to support decision/policy makers to deal with uncertainty in the future. The model will be used to explore and analyse the impacts of different scenarios in the future global logistics system by means of simulation. In addition to that, optimization techniques will be used to derive a framework to perform adaptive policy making as a response against the uncertainty. Ronald graduated as a bachelor of science in industrial engineering from Pelita Harapan University (Indonesia) in 2007 (magna cum laude). After working as a junior business consultant for almost one year he continued his study in Systems Engineering, Policy Analysis, and Management (SEPAM) in 2008 in Delft University of Technology. He accomplished his master study with a scholarship from Shell (Shell Centenary Scholarship). During his master study, Ronald was specialized in the transport, infrastructure and logistic domain and also in the modelling, simulation and optimization field. His master thesis concerned the development of the Simulation-based Multi-objective Evolutionary Optimization (SIMEON) framework which was done in an internship with an IT consultancy company. In the beginning of 2011, Ronald started his PhD research.

Sébastien HAULE Research engineer (CNRS, UMR 8504 Géographie-cités), Sébastien Haule received his Master degree in Contemporary history in 2001, and was involved in several CNRS publications as cartographer. In 2009, he created a database recording interface as part of Eurobroadmap project. Between 2011 and 2013, he worked for Agglomeration community of Le Havre (Communauté de l'agglomération havraise) in major risks department where he was in charge of subterranean risk questions. Since october 2014, he is ERC World Seastems project data analyst.

Carola HEIN Professor at Bryn Mawr College (Pennsylvania) in the Growth and Structure of Cities Department. Her current research interests include transmission of architectural and urban ideas along international networks, focusing specifically on port cities and the global architecture of oil. With an Alexander von Humboldt fellowship she investigated large-scale urban transformation in Hamburg in international context between 1842 and 2008. Her books include: The Capital of Europe (2004), European Brussels. Whose capital? Whose city? (2006), Brussels: Perspectives on a European Capital (2007), Rebuilding Urban Japan after 1945 (2003), and Cities, Autonomy and Decentralisation in Japan. (2006).

Olivier JOLY

Olivier P. Joly is is currently an Associate Professor in Transport & Spatial Planning and Member of the CIRTAI IDEES Le Havre, University of Le Havre (ULH). Former Director of the Master « International Transports », he is the holder of a PhD in Transport Geography & Spatial Planning from the University of Le Havre and of a Master’s Degree in Applied Mathematics & Social Sciences from the University of Paris 7. His previous works focus on: the worldwide structuring of Maritime Container Traffic Networks ; setting indicators related to transport activities, and urban management that are appropriate to retain to compare port-cities in Europe ; setting indicators related Logistics Performance ; the Global Logistic Chain Security : Economic Impacts of the US 100% Scanning Law. His present research activities are focused on the Territorial Impacts of the New Forms of Port and Harbour Organization, the 'Reticular' Strategies of Transport & Logistics Operators, the Socio-Technical Acceptance of Capture, Transport & Storage of Carbone Dioxyde (CTSC) technologies and the improvement of the Motorways of the Sea in Europe.

Frédéric KAPLAN Prof Frederic Kaplan holds the Digital Humanities Chair at Ecole Polytechnique Federale de Lausanne (EPFL) and directs the EPFL Digital Humanities Lab. He conducts research projects combining archive digitisation, information modelling and museographic design. He is currently working on the "Venice Time Machine", an international project in collaboration with the Ca'Foscari University in Venice, aiming to model the evolution and history of Venice over a 1000 year period. Frederic Kaplan graduated as an engineer of the Ecole Nationale Superieure des Telecommunications in Paris and received a PhD degree in Artificial Intelligence from the University Paris VI. Before coming to Switzerland, he worked ten years as a researcher at Sony Computer Science Laboratory contributing in particular to the AIBO robot. Then he worked six years at CRAFT, the EPFL pedagogical research laboratory. He published more than a hundred scientific papers, 6 books and about 10 patents. His inventions and devices have been exhibited in several museums including the Centre Pompidou in Paris and the Museum of Modern Art in New York. He is also the founder and president of OZWE, a company that designs and produces innovative interfaces and consumer electronic products and of bookapp.com, a joint venture focusing on digital publications.

Carl KNAPPETT Carl Knappett is the Walter Graham/Homer Thompson Chair in Aegean Prehistory at the University of Toronto, and holds his degrees from the University of Cambridge (Ph.D. in Archaeology). His areas of specialization are Knossos on Crete, Minoan ceramics, and material culture and theory. He has received various awards for his work, and is a Fellow of the Society of Antiquaries of London. He has several articles and books forthcoming, including An Archaeology of Interaction: Network Perspectives on Material Culture (Oxford University Press).

Pierre LATOUCHE Pierre Latouche is Assistant Professor at Université Paris 1 Panthéon-Sorbonne. He received in 2010 the Ph.D. degree from Université Evry Val d’Essonne (France) for his work on statistical network analysis. His research interests include statistical network analysis, high-dimensional statistics and their applications to biology and digital humanities. He is also the maintainer of the Mixer and Rambo packages for the R software.

Sung-Woo LEE Dr. Sung-Woo Lee is a director at the International Logistics Research Department, Korea Maritime Institute, a public think-tank in Korea specializing in the area of maritime transport and international logistics. He has been involved in various governmental projects, many of which are associated with shipping management, dryport planning and port logistics park development. He was previously affiliated with the Centre of Urban Planning and Environmental Management at the University of Hong Kong as a visiting researcher. His interests include managerial and strategic aspects of international logistics and maritime transport, possibility of Arctic logistics system, port-city interaction and globalization. He is a member of International Association of Maritime Economists and is a guest editor of International Development Planning Review.

Philippe LEYMARIE

Serviceman working abroad on an aid project, and then lecturer in Madagascar, Philippe Leymarie has collaborated in France to numerous publications, before he joined Radio France International where he realized most of his career. He currently animates the blog “Défense en ligne” on the website of Le Monde Diplomatique (http://blog.mondediplo.net/-Defense-en-ligne-).

Serge LHOMME Serge Lhomme is lecturer researcher at the University of Paris Est Créteil and Lab’URBA (France). He teaches environmental geography and spatial analysis. His areas of expertise concern spatial interactions between urban system and natural hazard. Specific topics of research have included urban resilience analysis, network vulnerability assessment using graph theory, cascading effect optimizat, critical infrastructures recovery optimization, GIS.

Luis LOBO-GUERRERO Professor of History and Theory of International Relations, Member of the advisory board of the Research Institute for the Study of Culture (ICOG), of the editorial boards of the journals Security Dialogue; Journal of Intervention and State Building; and Resilience:

International Policies, Practices and Discourses. Member of the international advisory board of Revista Pleyade (Chile). Specialist of International Relations and expert on connectivity, its history and politics, international relations theory, emphasis on post-structuralist thought, global governance of uncertainty, insurance and liberal governance. I am currently exploring the meaning and role of connectivity in global politics. I focus on sites such as maritime and inland ports in the European Union and their role in creating European governance, identity, and value. I am also working on ideas such as knots as a way to understand connectivity within contemporary geopolitics. I continue to explore different (historical) modes of reasoning about order, power and governance through continental philosophy.

Igor LUGO Igor Lugo is Research Professor at the Centro Regional de Investigaciones Multidisciplinarias-UNAM in Mexico since 2012 where he focuses on systems of cities and transportation networks.

Nora MAREÏ Nora Mareï, Post-doctoral Research Fellow (CNRS, UMR 8504 Géographie-cités), has a PhD in geography from the University of Nantes. She carried out her thesis under the direction of Pr. Jacque Guillaume in the Laboratory Géolittomer (UMR 6554 LETG). Her research focuses on the relation between global transports and territories.

Bruno MARNO Bruno Marnot is Professor in Contemporary Economic History at the University of La Rochelle. He studies the history and politics of large public works, and more particularly the port economy of the 19th-20th centuries.

Jean-François MARY Geographer, Jean-François Mary has worked with the AURH team since 2008 to manage the geographic data base and the business datasets with a practical systemic approach. This includes achievement of process dedicated to the studies or projects, ranging from input data to ready-to-use information.

Silvia MARZAGALLI Full-time professor of Early Modern history at the University of Nice, senior member of the Institut Universitaire de France, and Director of the Centre de la Méditerranée Moderne et Contemporaine in Nice. Her research focuses on maritime trade, merchant networks, shipping, warfare reorganization of trade flows, and the interactions between political decisions, State institutions and economic actors

in 18th-century conflicts and during Napoleon’s continental blockade. The Atlantic and the Mediterranean are the main areas of her interests. She has coordinated an international program in order to create a shipping database called Navigocorpus, which is now available online. She is currently working on the United States shipping and trade in the Mediterranean during the French Wars. Among her recent books: Révolution Consulat et Empire (with M. Biard and P. Bourdin), Paris 2009; Rough Waters. The United States involvement in the Mediterranean, 18th-19t c., (with John McCusker and Jim Sofka, eds.) St. John’s 2010 ; Atlas de la Révolution française. Circulations des hommes et des idées, 1770-1804 (with Pierre-Yves Beaurepaire) Paris 2010, and Bordeaux et les États- Unis, 1776 – 1815 : politique et stratégies négociantes dans la genèse d’un réseau commercial (Geneva, 2014).

Francesca Romana MEDDA Francesca Romana Medda is a Professor in Applied Economics and Finance at the University College London (UCL). From 2010 she is the Director of the UCL QASER (Quantitative and Applied Spatial Economics Research) Laboratory. Since 2012 she serves as economic adviser to the UK Ministry of Environment and Agriculture (Defra) and in 2014 at the Ministry of Finance (Her Majesty’s Treasury). Her research focuses on project finance, financial engineering and risk evaluation in different infrastructure sectors such as: the maritime industry, energy innovation and new technologies, urban investments (smart cities), supply chain provision and 80ptimization, airport efficiency. Her work is published in leading academic and practitioner journals. She has worked and works actively with the private and public sector such as for The European Investment Bank, The World Bank, UNESCO, UN-Habitat, WILLIS Re, HALCROW, and UITP. At present she holds several grants, two of which pertain to the application of complexity analysis in the real world. She is Co-Investigator in the £6.2m EPSRC Programme Grant “Liveable Cities” and the £5.8m grant “New Business for Infrastructure Investments”. From 2007 she served on the Executive Board of Directors of a major public transport company in Italy.

Elijah MEEKS Elijah Meeks is the digital humanities specialist at Stanford University Libraries. He supports humanities scholars in their integration of computational methods in their research. He was the technical lead on ORBIS, Kindred Britain, and the Digital Gazetteer of the Song Dynasty, and has contributed to several other digital humanities projects.

Mathilde MUS Geographer, Mathilde Mus joined the AURH team in 2012 as mapping coordinator to work on Weastflows project. Since 2013, its fields of intervention were extended on studies or projects who require geographic information system, or who need to develop geographical analysis or cartographic tools.

Adolf K.Y. NG Adolf K.Y. Ng is an Associate Professor in Transport, Logistics and Supply Chain Management from the I.H. Asper School of Business, University of Manitoba, Canada. He obtained a doctoral degree (Dphil) from University of Oxford, UK, in transport geography, and excels in the research and teaching of port management and governance, transport geography and regional development, transport and the environment, maritime logistics and global supply chains, with more than 100 scholarly publications, including two scholarly books entitled Port-Focal Logistics and Global Supply Chains (Palgrave Macmillan) and Port Competition: The Case of North Europe (VDM Verlag). A recipient of the Fulbright Scholar Program, the Endeavour Awards and the Université Parisienne Fellowship, funded by the USA, Australian and French governments, he was a visiting scholar at Stanford University, Australian Maritime College and the French Institute of Science and Technology for Transport, Spatial Planning, Development and Networks (IFSTTAR), respectively, and had provided strategic advices on transport and logistics issues to inter-governmental and governmental agencies around the world. He currently serves as a Council Member for the International Association of Maritime Economists (IAME), the Co-Editor of Journal of Transport Literature, and a member of the Editorial Board of Maritime Policy and Management and the Asian Journal of Shipping and Logistics.

Jean-François PELLETIER With over 15 years of experience in applied research, consulting and training in Canada and abroad, Mr. Pelletier has extensive knowledge of the maritime industry. Before joining CPCS, he was project manager with Innovation Maritime from 1997 to 2007 and from 2010 to 2011. He holds a PhD in transport geography from Paris-Est University and an MA in marine resource management (Transport Option) from the University of Quebec in Rimouski.

Yoann PIGNE Yoann Pigné received a Ph.D. in Computer Science (Le Havre, 2008). He then joined the university of Luxembourg as an associate researcher (postdoc) from 2009 to 2011. Since October 2011 he is a lecturer at the university of Le Havre. He specialized in dynamic graph algorithmic in the field of mobile ad hoc networks and especially meta-heuristic methods for finding and maintaining structures in dynamic graphs. He also focused on vehicular ad hoc networks problems and obtained significant results on realistic vehicular traffic and wireless communication simulation. He then broaden his application fields where graphs can be used as a model to solve problems such as epidemic problems, transportation networks, or life cycle assessment.

Ray RIVERS Emeritus Professor and distinguished research fellow at Imperial College London.

Yannick ROCHAT

Yannick Rochat is a mathematician and a doctoral student at the Digital Humanities Laboratory (DHLAB) of the École Polytechnique Fédérale de Lausanne (EPFL). His research topics are network analysis, literary analysis and data visualisation. In his doctoral thesis, he explores the modelling of novel plots by networks of characters based on co-occurrences in the text, and applies families of centrality measures on these networks in order to describe the characters' roles along these narratives.

Stanislas ROUSSIN Mr. Stanislas Roussin, major in Economic Geography (Master degree – University of Nantes), and in Land and Transport planning (Post Master Degree and PhD School – University of Le Havre), moved to Korea in 2002 further to a researcher exchange between INHA University (South Korea) and Le Havre University. Since 2004, he joined SERIC, an International Business Consulting firm supporting foreign companies to develop their activities in Eastern Asia (Japan, South Korea, North Korea, Mongolia and China). At SERIC, he holds the position of Director. Mr. Roussin is also the representative of HAROPA (Le Havre-Rouen-Paris ports) in Korea since 2012. On an academic point of view, since its graduation Mr. Roussin wrote several articles and speeches on International business practices, North Korean logistic organization, spatial development of North Korea, and on Korean peninsula ports system and port-cities, especially using his knowledge of the North Korean territory. He joined as Senior Researcher, the SEOUL-ASEM Institute in 2011.

Walter SCHEIDEL Walter Scheidel is Dickason Professor in the Humanities and Professor of Classics and History at Stanford University. His research focuses on ancient social and economic history, premodern demography, and the comparative study of labor regimes and state formation. He is the author or (co-)editor of a dozen books and has published 200 papers. In 2012 he launched the interactive geospatial network model “Orbis” jointly with Elijah Meeks.

Anna STOBBE Chair Group on History and Theory of International Relations, University of Groningen, The Netherlands.

Ka-Chai TAM Assistant Professor, Department of History at Hong Kong Baptist University (DPhil University of Oxford, MPhil BA (Hons) Chinese University of Hong Kong). His research foci include Chinese legal history, maritime history of East Asia, economic development and social order of Ming-Qing China, history of the Chinese Book, and historical Geographic Information System.

Lóránt TAVASSZY Since 1 June, Lóri Tavasszy (Hasselt, 1967) has been working as endowed professor of Freight Transport and Logistics at TPM for two days a week. He is also a senior consultant in Mobility and Logistics at TNO in Delft, focusing on innovative solutions for freight transport and issues such as accessibility and urban quality of life. He lectures on the subject of freight transport modelling and leads the research programme on sustainable accessibility for freight transport within the Randstad, funded by the NWO. Tavasszy is also one of the key nominees who will represent TU Delft in the new Dutch Institute for Advanced Logistics (DINAG) in Breda.

Jérôme VERNY Dr. Jerome Verny is founder and director of the MOBIS research institute specialized in international transport. Transportation engineer graduated of the Ecole Nationale des Ponts et Chaussées and the Higher School of Transport (Ecole Supérieure des Transports, Paris), he obtained a PhD in transport geography. Jerome has worked internationally since the last fifteen years with firms and public institutions in order to implement logistics innovations and performance optimization. In 2009, Jerome received the International Young Researcher Award in transport by the OECD-ITF (International Transport Forum) in the presence of 52 transport’s ministers, met on this occasion in Leipzig. He is the author of many national and international publications in academic journals (International Journal of Production Economics, European Journal of Transport and Infrastructure Research, etc..), in collective books, in business news (The Independent, Les Echos, La Tribune, the New Economist, etc..) and in news specializing in transport and supply chain management. The work developed by Jerome fall within the framework of the new economic geography. Projects currently underway with his team, deal with the logistics performance of companies, the transport policies and the land-use planning.

Liehui WANG Associate professor of center for modern Chinese city studies, East China Normal University, Shanghai, China. He got the PHD degree of historical geography from Fudan University in 2008. He is interested in port geography. Now he is a visiting scholar in ITMMA of University of Antwerp.

Lei YANG College of History and Social Development , Shandong Normal University, Jinnan, China.

Rawya ZREIK Rawya Zreik is a first-year Ph.D. student in Statistics at the universities Paris 1 and Paris 5, under the supervision of C. Bouveyron and P. Latouche. She is working on the development of methods for the statistical analysis of temporal networks with applications in digital humanities. She received the Master degree in Statitics in 2013 from University Lille 2 (France).

Name Country Email

AIT-MOHAND Kamel France [email protected]

ALIX Yann France [email protected]

AUDIEVRE Lucile France [email protected]

BAHOKEN Françoise France [email protected]

BARTHELEMY Marc France [email protected] BOUVEYRON Charles France [email protected] BRETAGNOLLE Anne France [email protected] CASCHILI Simone UK [email protected] CHAZE Milhan France N/A DEVOGELE Thomas France [email protected] DUCRUET César France [email protected] DUSZYNSKI Juliette France [email protected] ETIENNE Laurent France [email protected] EVANS Tim UK [email protected] FOURNIER Mélanie Canada [email protected] FREMONT Antoine France [email protected] GASTNER Michael UK [email protected] GIPOULOUX François France [email protected] GROSSNER Karl USA [email protected] GUERRERO David France [email protected] GUINAND Frédéric France [email protected] HALIM Ronald The Netherlands [email protected] HAULE Sébastien France [email protected] HEIN Carola USA [email protected] JOLY Olivier France [email protected] KAPLAN Frédéric Switzerland [email protected] KNAPPETT Carl Canada [email protected] LATOUCHE Pierre France [email protected]

Name Country Email LEE Sung-Woo South Korea [email protected] LEYMARIE Philippe France [email protected] LHOMME Serge France [email protected] LOBO-GUERRERO Luis The Netherlands [email protected] LUGO Igor Mexico [email protected] MAREÏ Nora France [email protected] MARNOT Bruno France [email protected] MARY Jean-François France [email protected] MARZAGALLI Silvia France [email protected] MEDDA Francesca UK [email protected] MEEKS Elijah USA [email protected] MUS Mathilde France [email protected] NG Adolf K.Y. Canada [email protected] PELLETIER Jean-François Canada [email protected] PIGNE Yohann France [email protected] RIVERS Ray J. UK [email protected] ROCHAT Yannick Switzerland [email protected] ROUSSIN Stanislas South Korea [email protected] SCHEIDEL Walter USA [email protected] STOBBE Anna The Netherlands [email protected] TAM Ka-Chai Hong Kong [email protected] TAVASSZY Lorant The Netherlands [email protected] VERNY Jérôme France [email protected] WANG Liehui China [email protected] YANG Lei China N/A ZREIK Rawyia France [email protected]

GLOBALIZATION, REGIONALIZATION, URBANIZATION: AN ANALYSIS OF THE WORLDWIDE MARITIME NETWORK SINCE THE EARLY 18TH CENTURY

PROJECT ABSTRACT

The World Seastems project aims to map and to analyse the changing spatial pattern of the world economy across 300 years from a maritime perspective. It will exploit untapped vessel movement data on a world scale since 1734, date of the first publication of Lloyd’s List. Such data offer disaggregated information on weekly inter-port flows with detailed descriptions of vessels as well as their dates of departure and arrival at world’s ports. Despite the vital importance of maritime transport for economic development and international trade, no research has been done on the long-term evolution of the global maritime network.

There are three main goals of the project:

 First, it will map for the first time the spatial distribution of almost 300 years of maritime flows in a dynamic and interactive manner. A geomatics visualisation platform will also integrate advanced analytical tools to simplify the pattern of shipping routes and corridors, and to extract meaningful information from the original data, with both scientific and pedagogical outcomes.

 Second, the project will look at the topological and spatial structure of the global network of inter-port links with reference to graph theory, social network analysis, and complex networks. The global properties of the network can be compared with general models of networks, while the evolution of macroscopic measures will be explored in relation with wider structural and conjectural changes in the world system e.g. conflicts, revolutions, crises, territorial reconfigurations) in terms of network expansion, shrinkage, concentration and polarization. Internally, the search for tightly connected substructures (i.e. clusters, communities of ports, économies-mondes) will focus on the emergence of world regions and regional integration processes.

 Finally, we will examine the co-evolution of maritime flows and urban/regional development and compare the growth trajectories of port and non-port cities based on their situation in the combined sea-land network.

In a multidisciplinary fashion, the project questions both the contribution and the resilience of port activities and shipping routes to the transformations of the world system and economy from the local level to the global level. It will provide novel results about world systems theory, network theory, and location theory.

PROJECT OUTLINE

Being one of the oldest forms of human interaction, maritime flows are good indicators of economic circulation and a useful tool to "take the pulse of world trade and movement" (Ullman, 1949). We present this sector-specific approach to globalization and urban development at two complementary levels of analysis, global and local.

Dynamics of macro-structures and world regionalization

To date, the few existing empirical studies of global maritime flows remain rather static and focussed on specialized issues such as biological invasions (Kaluza et al., 2010), climatology (Herrera et al., 2003), container network structure, cost efficiency, and individual companies. While the ability of maritime flows to reveal wider economic and territorial structures has been argued by both geographers (Rodrigue et al., 1997) and historians (Lewis and Wigen, 1999), no study could have validated the materiality of such proposals. More likely in this perspective are studies of other global networks such as those shaped by airlines, multinational firms, trade, migration, knowledge, and communication flows (Van Hamme and Patris, 2011). Yet, the specificity of maritime flows is their early existence as long-distance transport and communication vectors. This sole fact justifies their high relevance to revisit the changing spatial pattern of global human interactions.

Early theories about the evolution of the so-called "world system" depict successive phases since the medieval times by which some regions of the world become dominant at the expense of others (Wallerstein, 1979; Braudel, 1985), the first being the "core" and the others the "peripheries". Although subsequent works have debated such ideas notably by studying longer time periods or focusing on new forms of global organizations, such as global commodity chains and production networks, a systematic analysis remains lacking. Notably, research on global production networks remain highly qualitative and "falls short of delivering a rigorous analysis that can give ‘the big picture’ of GPNs on a global scale" (Hess and Yeung, 2006: 1201). Studies of commodity chains and world city networks are more quantitative but often neglect the materiality of flows among locations by looking at advanced producer services (Leslie and Reimer, 1999; Hall and Hesse, 2012). The longstanding existence of maritime flows can thus offer strong evidence about the spatial patterns of trade dominance across the world (Vigarié, 1968; Vance, 1970) as well as on the impacts of cost and time fluctuations on such patterns (e.g. declining friction of distance).

The World Seastems project thus aims at providing a systematic analysis of the path-dependency of hierarchical structures affecting global interactions, by measuring and mapping the unequal integration, concentration, and connectivity of maritime flows across space and time as well as their changing geographic coverage. Main efforts will be put on trying to untangle the respective influence of territorial and network factors. Territorial factors are those outside the maritime and port industry such as market location and trade relations, which are in turn affected by major political and economic evolutions (e.g. rise and fall of empires and nations,

JUNE 2014 - UMR 8504 GÉOGRAPHIE-CITÉS - ALL RIGHTS RESERVED International Workshop ERC World Seastems 91 industrial revolutions, wars and crises, production shifts). Network factors are more internal and better relate with technological progress (i.e. wind, steam, combustion, containerization, mega-carriers, intermodalism), freight costs and navigation constraints. The combination of those factors confer maritime flows a different role in shaping the world economy and the links among its components according to the context. Their distribution will reflect but also transgress or contradict major spatial structures and dynamics in terms of core-periphery and polycentric configurations. The project will also examine the potential to discuss future evolutions based on the improved knowledge of past and current patterns and the identification of "symptoms" by which maritime flows incorporate but also anticipate upcoming events.

Co-evolution of urban development and maritime flows

On a local level, main efforts will be put on further understanding the co-evolution of maritime transport and urban development, thereby providing novel evidence about the evolution of cities in general and port cities in particular. Earlier contributions about the evolution of cities and regions from a maritime perspective remain theoretical (Murphey, 1989; Hoyle, 1989; Fujita and Mori, 1996) as a large-scale validation is lacking. Successive phases of synergy and separation between port and urban spaces and functions are often attributed to the impact of several factors such as technological change fostering port competition and traffic concentration, urban diversification towards more advanced functions, with respective planning decisions and imperatives to sustain or relocate port activities within the urban space. Port cities thus provide a perfect example of how certain locations evolve by absorbing innovations in specific sectors (Pumain et al., 2009) that in turn reinforce or modify their rank and specialization in the hierarchy of urban places. The World Seastems project thus wishes to reveal the path-dependency of urban development in relation to maritime flows and the uneven adaptability of cities to technological and economic changes. It will revisit earlier models of transport network evolution (Taaffe et al., 1963) by shedding more light on the long-term dimension of port dynamics, in terms of concentration and competition.

Because cities and systems of cities are often studied from a continental perspective (see Bretagnolle and Pumain, 2010), the combined analysis of maritime networks and land-based transport networks will challenge existing models of urban systems and their evolution (e.g. Christaller's central place theory) as well as the specificity of "gateway cities" (Bird, 1977). The project will thus provide a more complete picture of the centrality of cities while elucidating the difference between port cities and non-port cities. Although some recent efforts have been made to calculate the multimodal accessibility of cities on a world or European level, the results remain highly static (European Union, 2010). The local impacts of lowering spatial friction over hinterlands and lowering maritime transport costs are not yet sufficiently understood. As stated by Slack (1993), ports belong to a global transport system where gateways focus on hinterland accessibility and intermediate hubs target maritime transhipment (Rodrigue et al., 2009). This analysis shall focus dominantly on the post-1890 period when industrial and transport revolutions really started to influence continental communication systems such as roads, railways, canals, warehousing, and related industrial developments and also due to the difficulty accessing detailed land network information.

Mapping and modeling maritime flows

The first research direction is the elaboration of a geomatics platform with strong emphasis on Geographic Information Systems (GIS) methods of data representation and analysis. Flows will be represented in diverse forms such as 2D and 3D representations on a world map and sphere using various cartographic projections. This is in itself an innovation of the project that has many pedagogical outcomes. Each port of call is a location (node) and each voyage is a link between two or more ports of call at a given period. Each vessel circulation creates both a "chain" (successive stops) and a "complete graph" (all stops connected directly and indirectly). The global maritime database is thus built from successive port-to-port matrices to be analyzed as a non-planar, weighted, and directed graph on various levels of node aggregation (e.g. port, port city, country, macro-region, continent) and where nodes and links have several attributes based on vessel characteristics (i.e. flag nationality, vessel size and type) and movements (i.e. duration, length, frequency).

At the global level, the project will measure the size of the network (i.e. number of nodes and links, total tonnage) as well as the changing structure of flows through various statistical and network analytical tools. Classic concentration measures (e.g. Gini coefficient, HHI) will be completed by network-specific ones such as those provided by graph theory, social network analysis, and complex networks. The changing size and properties of the network will be one first evidence of the globalization process and its ongoing trend of transformation. The capacity of flows to reflect and/or anticipate major global events will be examined in terms of the network's robustness and vulnerability, with reference to major works on scale-free / small-world networks and their cumulative dynamics. Preferential attachment can be tested to explore whether newly added ports generally connect to already established and larger ports. Innovation diffusion in the maritime and ports sector has been highly selective at different stages; many ports were dropped from the network, but new ports were also created or reactivated. Other global indicators can be produced such as the average duration and distance of links so as to verify the concept of time-space shrink in world exchanges. Simulation experiments will be tested with the help of other institutions in order to forecast future changes and confront past dynamics with some models of network evolution, but this is not a core objective and competency of the principal investigator. Another important dimension of the analysis will be the search for coherent substructures in the network, with reference to the buoyant research field on clusters and communities. Various methods such as single linkage analysis, hierarchical clustering, modularity, trajectory analysis, blockmodeling, and structural equivalence (see for instance Snijder and Kick, 1979) will reveal in different ways the emergence and resilience of subsystems in the pattern of flows. The idea is to map the cores and the peripheries as well as to examine the influence of various proximities in their internal and external links (e.g. spatial, commercial, political). The role of physical distance in the evolution of connectivity can also be tested with reference to small-worlds where only a few nodes act as connectors between different and/or distant communities within which most links are local. This will contribute to the fast growing research field on spatial networks where territorial embedding plays an important role (see Barthélémy, 2010 for a review).

At the local level, comparing the evolution of maritime flows and urban population data (i.e. the most widely, if not the only, accessible indicator of urban importance on a world level and over time) will allow for comparing and categorizing individual trajectories of port cities. The interdependence between port growth and urban growth will be analyzed by measuring the changing statistical correlation between city size and traffic size as well as through applying other methods such as Granger causality tests. Traffic flows per location will be normalized for the comparison of local dynamics (e.g. London vs. New York 1734-2010). The specialization of cities' traffic will also be measured in terms of commodities and vessel types (e.g. raw materials vs. manufactured goods, steamers vs. sailing vessels) as well as geographic coverage (e.g. European vs. Asian forelands, long-range vs. short-range connections) and centrality (i.e. accessibility measures, degree, closeness, betweenness, etc.). An example of cartography based on one Lloyd's register in published in 1890 (next figure) is provided for illustrating possible treatments. Specific GIS methods can be used to map the results such as cartographic distortions (e.g. dynamic anamorphic visualization of traffic distribution), traffic isochrones to/from certain centres (tributary areas), edge aggregation based on Kernel density, etc. This will help to determine the dominance of certain cities not only in terms of traffic volume but also in terms of the geographic reach of their maritime connections. A complementary approach will be to measure the centrality of cities on the level of a combined sea and land network, where port cities act as connectors between the two spaces (foreland and hinterland). Based on the modelling of land-based transport networks, such an approach will better assess the role of hinterlands on the emergence of some gateways as opposed to maritime hubs as well as regional particularities in this interdependence (e.g. North European gateways having wide hinterlands, Asian port cities being more maritime- oriented). This will complement and challenge current works on the structure and dynamics of coupled and interdependent infrastructure networks (Vespignani, 2010).

WORLD SEASTEMS: AN ERC FUNDED PROJECT

Project duration

From 2013-03-01 to 2018-02-28.

Project details

 Project reference: 31384.  Status: Execution.  Total cost: EUR 1 500 000.  EU contribution: EUR 1 500 000.  Programme acronym: FP7-IDEAS-ERC.  Subprogramme area: ERC-SG-SH3.  Contract type: ERC Starting Grant

Principal Investigator

Dr. César DUCRUET, P.A.R.I.S, CNRS, UMR 8504 Géographie-cités.

Host Institution & beneficiaries

Centre National de la Recherche Scientifique, France.

To know more

 ERC Website: http://erc.europa.eu/te  Cordis Website: http://cordis.europa.eu/projects/rcn/107041_en.html  World Seastems Website: http://www.worldseastems.cnrs.fr