From agricultural soils to the human gut, all microbial communities are exposed to multiple pulse perturbations that vary in their intensity and frequency over time. While a growing body of anecdotal evidence suggests that the combined effects of multiple pulses may depend on their temporal dynamics, a general theory for predicting how the order and interval of pulses influences microbial composition remains undeveloped. The overarching objective of this work is therefore to establish a predictive framework for understanding how the temporal dynamics of pulse perturbations affects microbial communities. Given two pulses (e.g. antibiotics, phages or fertilizers) and a community of interacting species, how should the pulses be arranged if we want to maximize or to minimize their joint effect?
We first use graphical approaches to illustrate that the two key ingredients determining the worst (or best) timing of pulses are: (i) the similarity of pulses in terms of how they impact the different species, and (ii) the intrinsic dynamics of the community. We apply our theory to a range of ecological models, starting with classic Lotka-Volterra models and consumer-resource models and then moving on to more complex models representing bacteria-phage dynamics and cross-feeding interactions. Numerical simulations reveal that community dynamics are pivotal in determining the worst-timing of pulses, and that the same two pulses may have additive, antagonistic, or synergistic combined effects depending on their timing.
Furthermore, we present preliminary results from experimental tests of our theory where bacterial communities in chemostats are exposed to multiple antibiotics and invasions under different temporal regimes. Our findings have broad implications for the management and conservation of natural microbial systems facing multiple perturbations, as well as for developing strategies to manipulate microbial communities in medical and other applied settings.