Carlos PAIS-MONTES, Universidade da Coruña, Instituto de Estudios Maritimos, Spain
Jean-Claude THILL, University of North Carolina at Charlotte, Department of Geography & Earth Sciences, United States
David GUERRERO, Université Gustave Eiffel, Ifsttar, Ame-Splott, France
In port geography a number of studies have tried to identify port regions based on common forelands. However these regions are unstable by essence because they depend on how maritime companies deploy their fleets on different trades and these decisions can rapidly change depending on market conditions and the availability of vessels. Previous works using network analysis techniques such community detection suffer from this instability, which make comparisons over time difficult. To limit this problem, this research proposes an alternative partition of the maritime network obtained through a combination of network analysis techniques (community detection) with empirical approaches. For each port region, it distinguishes a core (more stable) and a periphery (less stable). To assess the relevance of the method, we compare the structure of regions over different periods before and after the Covid-19 outbreak. We use a AIS dataset on vessel movements between ports between January 2018 and August 2021.
Keywords: regionalization|container transport|Covid-19|AIS data
A102673DG