Know more

To understand ecological communities you need to make a choice a priori: to construct simplified and qualitative theories, or realistic simulations that require an exhaustive empirical knowledge. But this opposition is not inevitable. Simple behaviors can emerge spontaneously from complex interactions. These behaviors are collective and generic: not very sensitive to the individual traits of the species, and determined by aggregate properties at the community level.

Predict diversity, functioning and stability

The authors of the article apply this idea to complex simulations covering a wide range of types and structures of interaction (competition, mutualism, predation, hierarchies and functional groups). They combine these features into large simulated communities, and show that diversity, function, and stability can be fully predicted by some statistical properties of the community.

Between order and disorder

This emerging simplicity is based on the mathematical intuition of disorder. Orderly and predictable interactions, such as the hierarchy of a food chain, give a clear and distinct role to each species. On the other hand, complex interactions, which interfere with each other, blur these roles; they leave room for collective effects, similar to crowd behavior. Ecological communities lie between the hierarchy and the crowd, and even complex simulations can be broken down into these two elements: simple ordered patterns (mechanistically comprehensible), and disordered collective effects (predictable statistically).

An approach that provides a powerful framework for predicting the consequences of assemblages of ecosystem modeling and quantifying the added value of more detailed models and measurements.