Defense thesis / Margaux DARNIS

The jury members appreciated her clear, dynamic, and particularly educational presentation. She was able to explain complex concepts in simple terms and demonstrated a solid mastery of various fields ranging from modeling to population genetics. After deliberation, the jury awarded her the title of Doctor of Biology and congratulated her on the quality of her work.

President : Christine Meynard
Examiners : Cindy Morris, Karine Chalvet-monfray, Jean Peccoud, Nicolas Sauvion, Alexandra Schoeny, Manuel Plantegenest
Rapporteurs  : Karine Chalvet-monfray, Jean Peccoud

Abstract : French agriculture is currently facing a dual challenge: managing ever-changing plant health issues that threaten crop yields and quality, and the need to reduce pesticide use for environmental and public health reasons. Epidemics caused by plant pathogens are particularly difficult to manage when lacking curative methods. For phytopathogens transmitted by insect vectors, one of the major levers for action is vector control, whose effectiveness depends on the right balance between the methods used and the ecology of the vectors. In particular, understanding and predicting the dispersal patterns of these insect vectors is the cornerstone of this control, enabling the optimization of surveillance, risk assessment, and the implementation of appropriate prophylactic measures. This thesis thus addresses two key questions: (1) What are the main drivers (local or distant, wind transport, etc.) influencing vector dispersal, and what is their relative importance? and (2) How can this knowledge be integrated into modelling tools for better plant health surveillance? To answer these questions, we studied two contrasting biological models: the psyllids Cacopsylla pruni, vector of the phytoplasma responsible for European stone fruit yellows (ESFY) in Prunus species, and several species of aphids vectoring viruses (CABYV and WMV) in melons. Our methodology is based on an integrative approach combining population and landscape genetics, statistical analyses, modelling of atmospheric trajectories and connectivity (HYSPLIT-Tropolink model), and vector population dynamics. Using this framework, we were able to assess the relative influence of various drivers — local climate, atmospheric connectivity and wind dispersal, landscape, host distribution, and geography — on vector dispersal and identify probable sources and dispersal patterns of arrival in fields and orchards. Our results reveal that vector dispersal cannot be explained by a single factor, but emerges from the interaction between local abiotic factors (e.g., winter temperature, contemporary climate), atmospheric connectivity (i.e., wind dispersal), landscape (e.g., host availability, physical barriers), geographic distances (limiting possible flights), and the biology and ecology of the studied insect. In addition, modelling flight pathways has made it possible to identify the likely source regions for each vector species studied and to improve predictive models of their abundance at the start of the cropping season. Our approach thus offers a transferable methodological framework for studying the dispersal of insect vectors in other systems (e.g., culicoides, mosquitoes), provided that the method is adapted to the specific characteristics of the model. This thesis thus provides new fundamental information on the movements of insect vectors and the role of atmospheric connectivity in these movements. It also provides a methodological framework, adaptable to other pathosystems, for studying these dispersions and developing more precocious and optimized surveillance and warning systems for better management of vector-borne diseases.

Scientific productions indexed in the open archive HAL INRAE