Hotspots of damage by antagonists shape the spatial structure of plant–pollinator interactions

Abstract

The balance between mutualistic and antagonistic plant-animal interactions and their spatial variation results in a highly dynamic mosaic of reproductive success within plant populations. Yet, the ecological drivers of this small-scale heterogeneity of interaction patterns and their outcomes remain virtually unexplored. We analyzed spatial structure in the frequency and intensity of interactions that vertebrate pollinators (birds and lizards) and invertebrate antagonists (florivores, nectar larcenists, and seed predators) had when interacting with the insular plant Isoplexis canariensis, and their effect on plant fitness. Spatially autocorrelated variation in plant reproductive success (fruit and viable seed set) emerged from the combined action of mutualists and antagonists, rather than reflecting the spatial pattern of any specific animal group. However, the influence of antagonists on plant fitness was stronger primarily due to the florivores' action on earlier reproductive stages, consuming and damaging floral structures before the arrival of pollinators. Our results indicate that the early action of antagonists creates hotspots of increased plant damage, where the effects of later acting mutualists are not translated into increased reproductive benefits. We foresee the potential for antagonists to shape the intra-population mosaics of plant fitness in situations where antagonists outnumber mutualists, when their interactions occur before those of mutualists, and when mutualists can detect and avoid damaged plants while foraging. Severely damaged plants in antagonistic hotspots might be excluded from the mating network and render a limited production of viable seeds, reducing both the growth rate of the plant population and the effective population size.

Functional consequences of plant‐animal interactions along the mutualism‐antagonism gradient

“…Our study shows that even at small population scales, interacting animal assemblages vary considerably across individual plants, independently of spatial autocorrelation effects (Rodríguez‐Rodríguez et al. ). However, this variation is not randomly distributed within the population, as illustrated by the presence of interaction motifs.…”

“…We monitored tagged individuals during the flowering season (May–September 2008) until just prior to the dispersal of seeds (see Rodríguez‐Rodríguez et al. for details).…”

“…Both mutualistic and antagonistic interaction strengths were standardized (see Rodríguez‐Rodríguez et al. and Appendix ).…”

“…Finally, we obtained viable seed set per fruit as the averaged P vs calculated across all fruits analyzed for each plant (see Rodríguez‐Rodríguez and Valido , Rodríguez‐Rodríguez et al. , for details).…”

Complementary Effects of Species Abundances and Ecological Neighborhood on the Occurrence of Fruit-Frugivore Interactions

“…Identifying these spatial patterns could be valuable, as they may establish a template for the distribution of the ecological outcomes of pairwise interactions (e.g., higher plant recruitment derived from the activity of seed dispersers; Blendinger et al, 2008;Hampe et al, 2008). Indeed, the spatio-temporal patterns of pairwise interactions may condition relevant processes such as density-dependent mortality, genetic exchange, and ultimately the population dynamics of interacting species (Rodríguez-Rodríguez et al, 2015).…”