In Southern California, a region with distinctive yet fragmented and invaded native flora and the richest native bee fauna in the United States, both native and non-native plants are frequently used as ornamentals in public and private landscaping. Data-supported ornamental plant selection based on pollinator usage can supplement increasingly fragmented habitats by providing abundant floral resources throughout the season. But little work has been done in this region to evaluate the relative attractiveness of the most common ornamentals to pollinators. In addition, the numerically dominant visitor to many plants is the Western Honey Bee (Apis mellifera). This introduced species often out-competes native bees for floral resources, and its visitation patterns can differ sharply from those of native visitors. In this study, we examined visitors to common ornamental perennials in the Southern California landscape to determine: (1) native and non-native plants’ relative attractiveness to a variety of visitors, including the honey bee; and (2) how foraging patterns differ between visitor taxa and between plant species.
Results/Conclusions
We found that native plants received significantly more visitors per minute on a standardized experimental plant area than did non-native ornamentals (p < 0.001; Poisson generalized linear model). Not only were native plants more frequently visited by non-Apis pollinators, they also received a greater diversity of visitor taxa, including a richer suite of native bees. However, visitation rates within native status groups differed by as much as a factor of 7, suggesting that attractiveness is more complex than native status. In our behavioral analysis, non-Apis visitors differed significantly from honey bees in the number and rate of flowers visited (p < 0.001; Gaussian linear model), as well as differing significantly in visitation pattern among individual taxa. A weak correlation between plant native status and visitor behavior in a hotter, drier year (p = 0.006) was not supported by the data as a whole. Our results suggest that including a data-driven selection of both native and non-native ornamental perennials in the developed landscape can diversify the assemblage of native pollinator fauna, reduce the impact of honey bee landscape foraging dominance, and may support an increased variety of floral-visitor foraging behaviors.