Flowering phenology has been recognized as an important factor shaping plant-pollinator networks because they impact resource availability for pollinator communities. Phenology may be constrained by phylogenetic relationships as closely related taxa are likely to have similar flowering times, impacting flowering patterns at the community level and plant-pollinator interactions. Despite the perils imposed by climate change and anthropogenic disturbances, little information exists on the reproductive ecology of upper montane tropical forests. We described the flowering phenology and the flower visitor community of the herbaceous flora of the upper montane forests of Costa Rica. We specifically asked if phylogeny influences flowering patterns. We also studied the plant-visitor network and asked how changes in floral resources between the dry and wet season structured visitor community.
Between 2013 and 2018, we recorded monthly flowering phenology of 70 herbaceous species in an upper montane forest at 3100 m asl. For all species, we also collected and identified all flower visitors. We used circular statistics to describe yearly flowering patterns. Phylogenetic relationships resolved from Phylomatic were correlated with differences in flowering peaks. The diversity and similarity of the flower visitor assemblage was compared across different times of the year. Plant-flower visitor networks were created and compartmentalization was used to determine if modularity differed between the dry and wet seasons.
Results/Conclusions
The flowering schedules of the herbaceous flora of upper montane forests in Costa Rica showed distinct seasonality with only 10 species flowering continuously. Over two thirds of all species had their flowering peak in the dry season (Nov-Apr). We found a significant correlation between flowering peaks and phylogenetic distance with closely related taxa having similar flowering schedules. Pollinator networks revealed significant nestedness and specificity. Modularity increased in the dry season compared to pollinator interactions in the rainy season, despite insect communities being similar throughout the year. This suggests that during the dry season higher resource availability may increase specificity of flower visitors due to lower competition or higher insect density which reinforces plant-animal interactions. Our results suggest that phenology may significantly shape the modularity of ecological networks. Changes in phenological patterns as a result of climate change may reduce flower resources or decouple plant pollinator interactions, which may have detrimental impacts on population dynamics of plants and their pollinators. Species loss due to the reduction of natural protected areas may also reduce insect communities and the ecosystem services they provide as pollinators.