Global warming is becoming one of the major threats to the persistence of life on Earth. Understanding how microbial communities respond to such threat is of paramount importance for ecological processes, such as ecosystem services and disease control. Here we show that warming increases the differences among microbial interactions; and as a consequence, it increases the heterogeneity to the limits at which microbial communities can tolerate environmental changes.
We use a structural stability framework to measure thermal tolerance of microbial communities. To test the structural stability in microbial communities under warming, we performed an experimental manipulation of competing protist species in monoculture and two and three species mixtures at different temperatures (six levels, each constant, modelled as a linear effect). Species compete for the same food resource (bacteria) and space. To obtain interaction coefficients we fitted a gradient of Lotka-Volterra type models to data using linear moving towards more nonlinear responses.
Using the carrying capacities observed from monocultures and the inferred interaction matrix, we estimated the feasibility domain of each community. The structural fitness difference for each community shows the distance between the observed vector of carrying capacities and the vector of carrying capacities located at the geometric centroid of the feasibility domain. Species biomasses are even at the centroid. Thus, moving away from the centroid pushes the community to a more uneven state. The relationship between evenness and the distance from the centroid depends on the shape and size of the feasibility domain. More similar communities show a negative homogeneous pattern, but this pattern weakens among heterogeneous communities.
Results/Conclusions
Using a microcosm experiment, we show that under a control temperature, the tolerance limits are similar for all communities. However, under thermal stress, these limits change among communities. We demonstrate theoretically and experimentally that this change is due to an increase in the variability in the strength of species interactions among communities.
We found that increasing temperature makes communities more heterogeneous in their responses (losing the negative relationship between evenness and the distance to the centroid) compared to control temperature, where communities are more homogenous. This implies that warming affects communities differently, changing species interactions and therefore changing their limit of tolerance to environmental perturbations.
Our findings provide new concepts and tools to understand the response and tolerance of microbial communities to environmental changes.