Human activities are expected to result in a diversity of directional or stochastic constraints that affect species either directly or by indirectly impacting their resources. However, there is no theoretical frame- work to predict the complex and various effects of these constraints on ecological communities. We developed a dynamic model that mimics the use of different resource types by a community of competing species. We investigated the effects of different environmental constraints (affecting either directly the growth rate of species or having indirect effects on their resources) on several biodiversity indicators. Our results indicate that (i) in realistic community models (assuming uneven resource requirements among species) the effects of perturbations are strongly buffered compared to neutral models; (ii) the species richness of communities can be maximized for intermediate levels of direct constraints (unimodal response), even in the absence of trade-off between competitive ability and tolerance to constraints; (iii) no such unimodal response occurs with indirect constraints; (iv) an increase in the environmental (e.g., climatic) variance may have different effects on community biomass and species richness (Lorrillière et al., 2012).
A theoretical model of specialist/generalist interactions provide a general reflection on the definition of resilience at the community level. Resilience is an indicator of the capacity of recovery of an ecological system. It can be estimated by measuring the resistance of a system to various disturbances (or its ability to maintain its equilibrium state despite the initial disturbance) or by measuring the time required to return to baseline after a temporary disturbance. This resilience is an indicator of mechanistic effects of disturbance on ecological systems. This indicator suffers from its dependence on the parameter used to define the system state (species richness, biomass or functional indicators). This is why I have been working to develop an integrative indicator of this resilience which introduced the original concept of meta-resilience.