Epidemic models for disease control in crop mixtures

Background and challenge

Modern agriculture faces a major challenge: maintaining high productivity while reducing reliance on chemical inputs such as pesticides. Crop mixtures, which consist of growing multiple plant species within the same field, are a promising strategy to address this challenge by enhancing both productivity and resilience.

Crop mixtures can increase yields through overyielding and reduce disease spread via two key epidemiological mechanisms: the dilution effect, which lowers host density, and the barrier effect, which limits pathogen transmission between plants. However, the relative importance of these mechanisms remains poorly understood, especially when agronomic interactions such as facilitation and competition between species are taken into account.

This PhD project aims to fill this gap by developing novel mathematical models that integrate both epidemiological and agronomic processes. By improving our understanding of how plant diversity affects disease dynamics and crop productivity, this research will contribute to the design of more sustainable and resilient agricultural systems, supporting agroecological transition and food security.

Objectives

The main objective of this PhD project is to develop generic mathematical models to better understand and optimize the role of crop mixtures in plant disease control. More specifically, the project aims to:

• Develop mechanistic epidemiological models describing disease spread in crop mixtures, distinguishing between dilution and barrier effects.
• Incorporate agronomic interactions (facilitation and competition) into these models to assess their combined effects with epidemiological processes.
• Identify conditions leading to overyielding, determining optimal mixture configurations that maximize yield while minimizing disease.
• Extend the models across temporal and spatial scales, including: continuous-time models, semi-discrete models accounting for seasonality, and spatially explicit models.

Develop interactive decision-support tools (R-Shiny or Python-Streamlit) to allow users to explore different crop mixture scenarios and their impacts on disease and yield. 

This work will produce generalizable insights valuable for both fundamental research in plant disease epidemiology and practical applications in sustainable agriculture.

The PhD is part of an international collaboration between the Institute for Genetics, Environment and Plant Protection (UMR IGEPP: INRAE – Institut Agro Rennes-Angers – University of Rennes) and the University of Douala (Cameroon).