[Thesis defense] 11/12/23: Andrea Radici: "A modelling framework for the development of eco-efficient control strategies against aerial plant pathogens: the cases of black rust in wheat and moniliosis in peaches" (INRAE BioSP)

Title of the thesis

"A modelling framework for the development of eco-efficient control strategies against aerial plant pathogens: the cases of black rust in wheat and moniliosis in peaches".

Date and place

11 December at 14h00
Salle Alvéole,
INRAE PACA,
228 route de l'Aérodrome,
CS 40509 Domaine Saint-Paul - Site Agroparc,
84914 Avignon Cedex 9

Discipline

Agricultural Sciences

Laboratory

INRAE BioSP

Framing

• Daniele BEVACQUA (Director)
• Davide MARTINETTI (co-director)

Composition of the jury

• Virginie RAVIGNÉ
• Thibaud PORPHYRE
• Cindy MORRIS
• Suzanne TOUZEAU
• Daniele BEVACQUA
• Davide MARTINETTI

Summary of the thesis

The emergence of plant pathogens is accelerating worldwide, threatening food safety. There is an urgent need to design innovative plant protection tools, combining surveillance strategies with early disease control strategies, in order to guarantee food safety while ensuring the environmental sustainability of agricultural practices. In this context, airborne pathogens represent a major challenge, as they can spread over long distances.

In this thesis, I propose a modelling framework for designing strategies based on complex networks for monitoring and controlling plant pathogens moving through air masses. My subjects of study are the fungal pathogen Puccinia graministhe causal agent of black rust in wheat, and Monilinia fructicolathe causal agent of peach brown rot.

First, I present the modelling of host-pathogen interactions as complex networks and trace how scientists have used the properties of networks to develop disease protection strategies. Next, I reconstruct the global epidemic network of wheat stem rust, where wheat-growing regions are linked by potential pathogen transport connections. I estimate these connections using an aerobiological model based on Lagrangian trajectory simulations. I show how an algorithm based on complex networks can help identify the best sentinels, i.e. places where an epidemic can be detected quickly. Thirdly, I integrate the aerobiological model into a metapopulation epidemiological framework in order to simulate the spatial spread of a plant disease epidemic. In particular, I couple a climate-dependent model describing host susceptibility and epidemiology within orchards with Lagrangian trajectory simulations determining pathogen transport between orchards. I use brown rot of peaches in France as a case study, for which I produce epidemiological risk maps to facilitate the development of protection strategies. Finally, I am assessing the overall loss of surveillance efficiency due to the lack of communication and coordination between countries in the case of cross-border diseases. I use the global wheat rust epidemic network as a case study. I evaluate the surveillance effort (i.e. the number of sentinels) deployed by each country to achieve a given surveillance objective in a cooperative scenario (i.e. without taking borders into account) and then in a scenario where each country is independent.

Given the high density of the global epidemic network of PucciniaIt is possible to find a small set of sentinels (1% in the network) that indirectly monitor half of the wheat-growing regions (50% in the network). I demonstrate that connectivity based on area masses helps the reconstruction of observations of the incidence of brown rot in France, and identifies the places most at risk in the Rhône valley. The advantages of a cooperative strategy, which correctly interprets the disease's dispersal scale, are obvious for cross-border diseases, but are distributed heterogeneously between countries: compensation mechanisms should be implemented to obtain unanimous support for an international cooperative surveillance system.