Global patterns of plant taxonomic and functional diversity on landslides are modulated by climate and topography

Background/Question/Methods

Disturbance is a key ecosystem process that contributes to the re-organization of biological communities and ecosystem functions. The extent of change experienced by a community, however, will depend on the type, magnitude, frequency of the disturbance, existing environmental conditions, and species' ability to cope with disturbance. In this regard, both the pool of species and traits may be important to understand how plant communities resist and recover from disturbance. In mountainous areas prone to landsliding traits that confer “mechanical” resistance to slope failure may characterize intact plant communities whereas traits that allow plants to persist under stressful and nutrient-limited conditions plant communities developing on landslides. Here we ask: 1) How do patterns of plant taxonomic and functional diversity on landslides vary globally? And 2) To what extent are these patterns of diversity modulated by environmental and topographic conditions? We reviewed 593 research articles and used 28 together with unpublished data to assemble a data set of 100 studies on plant presence/absence on landslides and extracted functional traits from available databases (TRY, BIEN, TR8). We characterized alfa and beta taxonomic and functional diversity and developed Boosted Regression Tree Models relating these metrics with a number of climatic and topographic variables.

Results/Conclusions

Taxonomic alpha diversity varied greatly at the family (1-70), genus (1-156) and species (1-198) level. The two first axis of a principal coordinate analysis-PcoA separated sites according to their floristic composition and identified sites biogeographically different but with similar composition. Regression tree analyses showed that temperature of the coldest and wettest quarter (mean R2= 0.825, p<0.005) and mean diurnal range (R2= 0,85, mean SE=0.031, p<0.005), are important parameters for alpha diversity. Beta diversity was related to the mean, intra and inter-annual cloud cover (mean R2 = 0.938, p<0.005). Functional diversity, on the other hand was higher on landslides (mean FRic = 6.18, FEve = 0.6) than paired undisturbed forests (mean FRic = 5.4, FEve = 0.57). Our study is the first one to document global patterns of plant taxonomic and functional diversity on landslides which may have a great potential on habitat restoration based on appropriate groups of species. In addition, The finding a relationship between diversity and climatic variables may help understand how the recovery of landslides may be affected under scenarios of climate change. We consider that our results will provide information for a sustainable and adaptive management of diversity on slopes in the face of global changes.