Pollination networks and plant local adaptation: the importance of serving the pollinator community in restoration

The need to consider present and future species interactions can make it difficult to ensure the long‐term effectiveness of a restoration effort. We designed an experimental restoration garden in an arid system to investigate the relationship between seed source population and plant‐pollinator interactions. Seedlings planted in the garden were sourced from eight populations of each of seven forb species. For 18 months, we tracked flower visitation by insects. We used network analysis to examine the roles of plant species and populations. In the whole‐garden network, taller plants as well as those that differed more in temperature of the source environment from the garden were more generalized, receiving visitation from a higher subset of the local visitor community. We detected no other relevant network patterns consistent across the full garden. However, when within‐species networks were constructed, plants from different populations exhibited significant differences in metrics of centrality and generalization for four out of seven plant species. Plants sourced from populations that exhibit similar climatic conditions to those found in the experimental garden, as well as plants from populations that displayed traits indicative of greater pollinator reward (e.g., greater plant size and greater floral abundance), were linked to increased centrality and generalization in networks. Our results suggest that it may be possible to select restoration plantings by seed source to bolster the diversity and frequency of pollinator interactions in a site, and further, that considering how plant local adaptation serves pollinators is likely to be important in restoration success.