My work combines biophysical accounting, network science, historical analysis and foresight to understand what system structure makes possible, efficient or vulnerable.
Using detailed nitrogen-flow networks, I study how crop mixes and herd structure are associated with agro-environmental efficiency and distinct forms of metabolic vulnerability across 33 French territories from 1852 to 2014.
SSRN preprint
Unsupervised classification identifies recurring biophysical structures and reconstructs a metabolic history of French regional agriculture. Over- and under-represented transitions are consistent with forms of path dependence.
Historical datasetI translate stakeholder scenarios into E-GRAFS to compare assumptions and trade-offs involving livestock, diets, inputs, trade and competing uses of biomass.
Scenario data
With Jérémie Unterberger, we reformulate agro-food nitrogen metabolism as a continuous-time flow network inspired by chemical reaction networks, allowing us to study metabolic phases and transitions between dominant recycling cycles.
arXiv preprintIn September and October 2026, I am developing a multisectoral model of energy dependence across the 27 EU Member States, extending my work on material dependencies beyond agriculture.
None of the articles below has yet been published in a peer-reviewed journal.