Despite decades of research, there is intense debate about how the relationship between diversity and disturbance regulates local diversity. Intermediate disturbance hypothesis (IDH) predicts a humped-shaped pattern, with colonisation and competitive exclusion acting as the outcome of diversity. However, while several reports support the IDH, the consistency of the humped-shaped pattern is often questioned. Previous studies including meta-analyses rarely consider interactive effects of disturbance frequency (time since last disturbance) and intensity on plant species diversity. In forest landscapes, intermediate disturbance such as insect outbreaks or selective logging could occur during early, mid and late sucessional stages (i.e., levels of frequency), creating openings in the canopy that affect species diversity. For instance spruce budworm outbreaks may occur at late successional stage in the boreal forests and selectively kill Abies balsamea, Picea glauca or Picea mariana leaving other trees, which could result in coexistence of early and late successional species. Here, we propose a conceptual framework that integrates the combined effects of frequency and intermediate intensity of disturbance on plant species diversity in forest landscapes. We also conducted meta-analyses to test the applicability of IDH for the relationships between species richness, Shannon's index or evenness, and disturbance frequency or intensity at the global scale. We examined the relationships between the response variables (i.e., species richness, Shannon’s index, evenness) and predictors such as frequency and intensity of disturbance across terrestrial landscapes using boosted regression trees.
Results/Conclusions
Conceptually, we show that species diversity could peak at low, intermediate and high intensity disturbance depending on frequency. The interaction between frequency and intensity of disturbance could explain why different patterns of diversity response to disturbance exist in terrestrial landscapes contrary to what the IDH predicts. We found evidence in support of the IDH for the relationship between species richness, and frequency or intensity. Model for richness explained 29 of mean total deviance. Richness peaked at intermediate frequency and intensity of disturbance for all terrestrial landscapes. Shannon's index and evenness also peaked at intermediate frequency of disturbance, supporting the IDH; however, much weaker relationships existed than those for richness. The available evidence generally supports the IDH, consequently, we recommend managing for intermediate disturbance frequency in order to enhance the maintenance of plant species diversity in terrestrial landscapes.