Biostatistique et processus spatiaux

Domaines scientifiques déclarés au RNSR

Mots-clés des publications HAL repérées

Projets ANR

• EXtremes, apprentissage STatistique et Applications · 2023
• Associer des variétés pour canaliser l’adaptation des populations phytopathogènes : comment résoudre le compromis efficacité – durabilité - adoption · 2022
• Restauration intégrée de la Biodiversité pour Limiter les Risques Epidémiques dans les Socio-Ecosystemes · 2021

Publications HAL récentes repérées

• Mechanistic-statistical model for the expansion of ash dieback · 2026

  Résumé de la publication : Hymenoscyphus fraxineus is an invasive forest fungal pathogen that induces severe dieback in European ash populations. The spread of the disease has been closely monitored in France by the forest health survey system. We have developed a mechanisticstatistical model that describes the…

• Periodic KPP equations: new insights into persistence, spreading, and the role of advection · 2026

  Résumé de la publication : We focus on the persistence and spreading properties for a heterogeneous Fisher-KPP equation with advection. After reviewing the different notions of persistence and spreading speeds, we focus on the effect of the direction of the advection term, denoted by $b$. First, we…

• Identification of ecoclimatic indicators of esca disease through 20 years of large-scale vineyard monitoring in France · 2026

  Résumé de la publication : Plant disease incidence depends on host and pathogen phenology, and the abiotic environment.Plant health and productivity are, therefore, strongly influenced by global environmental changes, particularly for perennial crops, such as grapevines. Esca is a vascular disease causing significant damage to vines globally.…

• Artificial intelligence for reconstructing invasion and expansion routes of pest populations: from ABC-RF to neural methods · 2026

  Résumé de la publication : Understanding the origins, large-scale spread routes and demographic dynamics of harmful organisms, including invasive species, pests, vectors and pathogens, is essential for guiding surveillance and management. The models describing these processes are often too complex for their likelihoods to be evaluated directly.…

• Combining machine learning quantile regression and Gaussian random fields: a general framework for modeling and simulating space-time processes · 2026

  Résumé de la publication : Modeling and simulating space-time random fields while accounting for possibly misaligned covariates is crucial for many applications requiring uncertainty quantification and risk assessment. To achieve this, we propose a flexible framework that couples machine-learning quantile regression with space-time Gaussian random fields.In this…