Negative Feedback as Driver of Spatio-Temporal Patterns in Plants: A Unified Multiscale Model

Characteristic spatio-temporal patterns linked with dispersal processes have been observed at varying scales in plants. These range from dispersal kernels visible at the individual level to patterns emerging through interactions between plants at the population or community level. Generally, each pattern has been studied separately and has given rise to a specific set of potential explicative theories. We focus our demonstration on three patterns reported in plants at different scales. (i) At the individual scale, the Janzen-Connell hypothesis, which is one popular explanation for the maintenance of tree biodiversity in tropical environments, relies on a characteristic seedling recruitment curve that is believed to stem from host-specific predation. (ii) At the population scale, several types of vegetation patterns (e.g. stripes, labyrinths, spots) have been observed, especially in arid regions, and have been hypothesized to originate from either individualscale intraspecific interactions or water limitations. Among these, rings and socalled “fairy circles” have been the most challenging to explain. (iii) At community level, small-scale species turnover have been observed to produce counter-intuitive temporal variations in species richness (i.e. stable scale-independent species diversity, but cumulative values increasing in time at every scale). This has been assumed to come from species mobility and nice exploitation. Here, we show that these phenomena can be rendered under a unique theory relying on the concept of species-specific negative-feedback. We present a simple hybrid model coupling an Individual-Based approach, in which each plant is recognized as a separate entity, with a spatial System Dynamics submodel endering accumulation of autotoxic compounds in the soil. Through spatially explicit simulations, we expose how this model is able to reproduce the three behaviours mentioned previously.