In silico substrate dependence increases community productivity but threatens biodiversity

The critical role that biodiversity plays in ecosystem functioning has motivated many studies of the mechanisms that sustain biodiversity, a notable example being cyclic competition. We extend existing models of communities with cyclic competition by incorporating variable community evenness and resource dependence in demographic processes, two features that have generally been neglected. In this way, we align previous approaches more closely with real-world microbial ecosystems. We demonstrate the existence of a trade-off between increasing biomass production and maintaining biodiversity. This supports experimental observations of a net negative biodiversity effect on biomass productivity, due to competition effects suffered by highly productive species in diverse communities. Our results also support the important role assigned by microbial ecologists to evenness in maintaining ecosystem stability, thus far largely overlooked in in silico approaches.

Figure 1

Mechanisms of interactions, following Reichenbach et al. [2]. Whether or not an interaction occurs depends on the individuals' internal substrate levels.

Figure 2

Comparison of average biomass yield for the different scenarios. Each curve represents the mean of 200 simulations.

Figure 3

Comparison of average final evenness for the different scenarios. Each curve represents the mean of 200 simulations.

Figure 4

Comparison of pattern formation for resource-dependent demographic processes, for two different substrate thresholds, after 500 generations. The box in the center indicates the size of the substrate source. Top: ${10}^{-16}$ g substrate threshold. Bottom: $2\times {10}^{-15}$ g substrate threshold. Left to right: Substrate-limited process—reproduction, competition, mobility.

Figure 5

Comparison of average final evenness for different initial evenness values, for the symmetric competition limitation scenario. Each curve represents the mean of 200 simulations.