A single changing hypernetwork to represent (social-)ecological dynamics

Cédric Gaucherel (1) , Maximilien Cosme (1) , Camille Noûs (2) , Franck Pommereau (3)

1 UMR AMAP - Botanique et Modélisation de l'Architecture des Plantes et des Végétations
2 Laboratoire Cogitamus = Cogitamus Laboratory
3 IBISC - Informatique, BioInformatique, Systèmes Complexes

To understand and manage (social-)ecological systems, we need an intuitive and rigorous way to represent them. Recent ecological studies propose to represent interaction networks into modular graphs, multiplexes and higher-order interactions. Along these lines, we argue here that non-dyadic (non-pairwise) interactions are common in ecology and environmental sciences, necessitating fresh concepts and tools for handling them. In addition, such interaction networks often change sharply, due to appearing and disappearing species and components. We illustrate in a simple example that any ecosystem can be represented by a single hypergraph, here called the ecosystem hypernetwork. Moreover, we highlight that any ecosystem hypernetwork exhibits a changing topology summarizing its long term dynamics (e.g., species extinction/invasion, pollutant or human arrival/migration). Qualitative and discrete-event models developed in computer science appear suitable for modeling hypergraph (topological) dynamics. Hypernetworks thus also provide a conceptual foundation for theoretical as well as more applied studies in ecology (at large), as they form the qualitative backbone of ever-changing ecosystems.

Mots clés: Ecosystems; Interaction networks; Hypergraphs; Petri nets; Discrete-event models; Ecosystems interaction networks hypergraphs Petri nets discrete-event models; interaction networks; hypergraphs; discrete-event models

Domaines: Sciences de l'environnement; Modélisation et simulation

Cédric Gaucherel, Maximilien Cosme, Camille Noûs, Franck Pommereau. A single changing hypernetwork to represent (social-)ecological dynamics. 2024. ⟨hal-04506250⟩