Evolved digital ecosystems: Dynamic steady state, not optimal fixed point

Traditional models of ecosystems often assume that the species composing an unperturbed ecosystem become fixed so that only the relative abundances of the species change over time. Such ecosystems are said to have reached an optimal fixed point. However, recent work has suggested that neutral evolutionary processes can significantly alter the species composition of an ecosystem, allowing the ecosystem to exist in a dynamic steady state. Here, we investigate the stability of ecosystems and the nature of the equilibrium that forms using the digital evolution platform Avida, tracking evolving ecosystems over thousands of generations. We find that the communities that form are remarkably stable, and do not experience a significant loss of diversity in the long run even in experimental treatments where the communities suffer catastrophic population bottlenecks. When diversity rebounds, ecological communities are reconstituted in a different form than the one that was destroyed, but this difference is comparable to the difference the system would have accumulated if it had been left untouched. Thus, digital ecological communities exist in a dynamic steady state, which ultimately eliminates the effect of historical disturbances.