This page summarizes my latest research.
Wildlife populations in managed ecosystems
The fluctuations of species abundances are rarely analyzed simultaneously in spite of their connection across space and time. To have the possibility to analyze the variation of taxon-related species at the same time, I developed a Bayesian multi-species model that estimates taxon-level parameters related to dynamics, abundance, detection probability, and environmental stochasticity. From this model, I analyzed the variability of populations of a community of raptors composed of accipiters, buteos, and owls in Georgia and Florida (USA). I showed that migration pulses of the accipiter, buteo, and fledging owl’s dispersal drive variation of raptor’s abundances. I also demonstrated that large-scale climatic processes, such as the North Atlantic Oscillation (NAO), influenced the variation in abundances of raptors. When the NAO was in a positive phase, the abundance of raptors decreased, and inversely. The approach developed in this study aimed to facilitate the modeling of species-specific effects of environmental variation and guild-level dynamics that could be used for ecosystem-based conservation measures. This study is published in Ecological Informatics.
Community dynamics and variable environment
A better understanding of how species coexist within a dynamic community is indispensable for considering the consequences of global change on ecosystem functioning over short and long-term temporal scales. To this aim, I applied a multivariate stochastic community dynamics model to describe the fluctuations in the abundances of a freshwater community over 28 years (1990-2017) in Lake Atnsjøen (Norway) during the ice-free period. I showed that the community dynamics was driven by environmental variability in spring. In contrast, community-level ecological heterogeneity was highest in the autumn. The community returned faster toward equilibrium when the ecological heterogeneity was the highest over the autumn. Moreover, the community responded to the long-term warming of water temperature by decreasing species diversity and increasing abundances. These can have consequences on the stability and functioning of the ecosystem. This study is published in Oecologia.
Predicting abundances of interacting harvested species
Species are part of complex food webs, making identifying species interactions challenging. Nevertheless, knowing how species interact is essential to better understand the consequences of harvest. Considering the different life history stages of the interacting species might improve the identification of interactions. To analyze how life history stages affect the interaction between harvested related species, I used Atlantic haddock (Melanogrammus aeglefinus) and cod (Gadus morhua) living in the Barents Sea as a study model. First, I developed a life-cycle state-space model that considers haddock-cod interactions, climate variability, observation errors, and the stochasticity of population dynamics. I showed that the predator-prey interaction at the earliest stages turns into a competitive interaction around the age of 2 to 3 years. Then, I set up a hindcasting approach based on the estimation of life cycle state-space models that include interactions between both species, which, from the definition of different harvesting scenarios, allows for tracking population-level responses of the two species to harvest. I demonstrated that an increase in cod harvest, which led to decreasing abundance of cod, was associated with an increase in the abundance of haddock, suggesting compensatory dynamics of both species. This approach leads forward the analyses of the effects of harvest and climate in multi-species systems by considering the comprehension of complex ecological processes to facilitate the sustainable use of natural resources. This study is published in Ecology and Evolution.