Seed mix performance and species network roles as a framework to select candidate key resources for pollinator habitat

Abstract Current native bee declines have been attributed in part to loss of habitat and floral resources. Mitigation approaches include trying to establish or enhance bee habitat by planting wildflowers, which is often done by using seed mixes. Most assessments of plant performance in pollinator seed mixes are based on the abundance and diversity of visitors they support, which might be key elements of community interactions such as supporting specialized links within plant–pollinator networks or changes in core position within networks across seasons. Therefore, the selection of “candidate key pollinator resources” could be usefully extended to include assessment of species network roles. In two different experiments, we independently assessed the performance of a set of 28 native California wildflowers, first in seed mix applications over three years and second in standardized mono‐specific plantings of the same plant species. Within seed mixes, plant species' performance according to maximum floral area and phenological coverage clearly differed with certain taxa found to dominate, or only establish short term or not be competitive at all. Only a small set of 5 species persisted with high performance over the whole three years. Based on mono‐specific plot studies quantifying species network roles, as opposed to simply abundance‐diversity metrics, we identified network core species and species supporting ecologically specialized pollinators. Both species roles are characteristic in natural plant–pollinator networks and therefore could present key characteristics for choosing plants with the aim of restoring mutualistic plant–pollinator networks with wildflower habitat plantings. Plant species' network roles changed over the seasons within a year arguing for the inclusion of seasonality when choosing candidate key pollinator resources for plant mixes. Only a subset of species in networks proved to be successful in seed mix applications where plants needed to perform under competition. The results emphasize the need to evaluate wildflower plantings and prioritize species of high performance for multiple criteria in current and future applications.