The project

The general aim of RESILIENCE is to understand how different scales of biological organisation, organisms, functional community structure, metacommunity, and their interactions, drive community re-assembly and multifunctional resilience in neotropical ecosystems, following drought events that range from the current norm to extreme events and predictions of the Intergovernmental Panel on Climate Change (IPCC).


The impacts of severe drought events on ecosystem functions are far from being understood, as are the mechanisms which underlie functional resilience after the disturbance has passed. This topic is of utmost importance in tropical regions, where climate change models forecast significant changes in water availability due to increasing frequency and intensity of drought events. Only a handful of studies have examined how drought can affect multiple functions in tropical systems. Because such studies focused on the immediate outcome of drought (ignoring the recovery and resilience trajectories), we don’t know how organism traits and ecological mechanisms mediate the post-drought trajectory of ecosystem functions. Metacommunity theory predicts that immigration from source patches should prevent extinction in sink populations, but we know nothing of how habitat patch size and distance to source populations interactively mediate ecosystem resilience to drought. 


Our specific hypotheses are: (1) Multifunctionality will shift to alternative states at lower drought intensity if source patches are not available to prevent extinctions; (2) as drought intensity increases, the driving factors underlying ecosystem resilience will shift from organism tolerance to resistance, and from functional community structure to metacommunity dynamics; and (3) once we account for the negative effect of distance from source patches on recolonization rates, larger patches will be more attractive to immigrants and will undergo faster resilience than smaller habitat patches. 


The experiments take place in French Guiana near the Petit Saut Dam. Over the past 10 years, we have established a 25ha forest plot, where we control the distribution of the tank-bromeliad Lutheria (Vriesea) splendens. These settings offer unique opportunities to manipulate natural ecosystems in a primary forest understory.

Left: the understory; right: the bromeliad Lutheria (Vriesea) splendens. Photos: J.F. Carrias


We manipulate drought and metacommunity dynamics at the level of an entire, spatially-discrete ecosystem (the natural microcosm formed by rainwater-filled leaves of tank bromeliads and their microbial-faunal communities), to separate the roles of in situ recovery (tolerance, resistance forms) versus immigration on the resilience of key ecosystem functions under different drought scenarios. 

Rainshelters over individual bromeliads

Insect exclusion