I am an agricultural engineer and hold a PhD in population ecology. I use statistical and mechanistic models to assess the status of natural populations, their responses to environmental changes, and thus predict their dynamics over time and space. I have worked on numerous biological models, ranging from large predators (Canis lupus, Puma concolor, Crocuta crocuta) to locust populations (Shistocerca gregaria, Oedaleus senegalensis). I also use decision-support tools to propose management measures for these species, often within a socio-economic context characterised by conflict.
I am currently developing models to predict the risk of locust swarms in West and North Africa. In collaboration with members of the Locust Team at CBGP We are seeking to gain a better understanding of the ecological factors that influence locust population dynamics (migration, reproduction) and phase polymorphism – a form of density-dependent phenotypic plasticity that alters individual behaviour (solitary/gregarious phases) – in order to develop tools to support preventive management.
In collaboration with CLCPRO/FAO, as part of a project funded by AFD, I use data from satellite imagery (vegetation indices, soil moisture, etc.), weather models (air temperature, wind, etc.), and field data collection (locust observations) to produce distribution maps of the Desert Locust. Using Bayesian hierarchical models and artificial intelligence tools, we will be able to extrapolate the probability of events (presence, egg-laying, swarming, etc.) based on new satellite imagery, in order to guide the survey teams of the locust control units in the countries concerned, with the aim of improving the effectiveness of preventive management.
In partnership with Arianne Cease (University of Arizona), as part of a project funded by USAID, I am developing an agent-based model in Netlogo that combines the life cycle of the Senegalese locust (Oedaleus senegalensis) with its feeding and behavioural ecology, in order to better understand the effect of soil amendment on population dynamics at the landscape scale.