Anthropogenic activity has already modified the composition of frugivore assemblages across ecosystems worldwide. As a result, animal-dispersed plant species are deprived of dispersal interactions when their dispersal vectors are totally or partially extirpated as previously recorded for defaunated forests. Alternatively, plant species gain dispersal ability after the reintroduction of formerly extinct frugivores (e.g. rewilding) or when dispersal vectors reach extra-range locations (e.g. invasive species). The gain and the loss of plant dispersal ability determine the probability that propagules would eventually reach distant sites, i.e., the frequency of long distance dispersal (LDD) events that underlie forest regeneration and expansion dynamics across managed landscapes. Therefore, a robust and reliable quantification of the plant dispersal ability as a function of key animal traits, such as body size, is urgently needed if we are to forecast their chances to cope with environmental change. This study first applies statistics of extremes functions to quantify to what extent dispersal kernels shift when frugivore assemblages change under the following scenarios: (i) random defaunation; and (ii) preferential extirpation of large body-sized frugivorous. Secondly, I apply a spatial-explicit eco-evolutionary model to forecast distribution range shifts of animal-dispersed species with modified assemblage of frugivorous vertebrates
Results/Conclusions
Preliminary results showed that our forecasting ability greatly improved by adding animal correlates that works as a proxy of their ability to provide long distance services, such as body size. In a context of partially defaunated habitats, where large-sized frugivorous vertebrates are extirpated the expansion rate of plant populations dropped and a sizeable proportion of suitable distant habitats remained unoccupied, as oppose to scenarios with a full assemblage of frugivorous. Overall, this study shows that statistics of extremes coupled with spatial explicit eco-evolutionary models represent a suitable and novel approach to forecast the fate of animal-plant species that inhabit defaunated habitats and to quantify cascading consequences of modifying trophic interactions for animal-dispersed species.