Topological keystone species complexes in ecological interaction networks

Advanced techniques of network analysis allow the quantification of the indirect interactions and the topological importance of components in ecological interaction networks. In current conservation biology, considering indirect causal effects, identifying keystone species and outlining multispecies approaches begin to be high priority goals. We make an attempt to connect these issues within a network context. Our main interest is to determine the positionally most important set of n nodes in a network, to analyse whether a set of nodes for small n is a subset of another for larger n, and to quantify this nestedness. We apply the KeyPlayer software, a novel tool for network analysis in ecology, introduced originally in mathematical sociology. Topological keystone species complexes are defined, we illustrate the use of this method in a case study and analyse a database of 9 plant-pollinator interaction networks. Our main conclusion is that multispecies approaches may give results very different from single species analyses.