The influence of intensity of interspecific competition on foraging strategy for the three bees
Our study indicated that the intensity of interspecific competition (estimated by niche overlap) was not correlated with bee abundance across the sampling dates. The increasing bee abundance did not accelerate the intensity of interspecific competition while the mitigation of intensity of interspecific competition was not caused by a decrease of bee abundance. Moreover, we found that the foraging strategy was significantly correlated with competition intensity for all the three bees (Figs. 3 and 4). We conclude that for these bumble bees, the primary factor influencing the temporal change of foraging strategy might be the abundance of interspecific instead of intraspecific individuals.
The mainly pollen-seeking bee, B. festivus , foraged more on nectarless flowers than those with nectar during our investigations. However, its preference for reward type (pollen vs. nectar) did not remain stable. Our results indicated that the intensity of interspecific competition significantly influenced the foraging strategy of the bee. The bee mainly visited nectarless flowers to collect pollen grains on the days with low competition intensity. As competition intensity increased, the bees tended to forage for nectar. Although pollen provides important and necessary nutrimental requirement for bumble bee (Roulston & Cane, 2000; Goulson & Darvill, 2004; Woodard, 2017; Woodard et al., 2019), nectar may provide more energy than pollen (Baker & Baker, 1983). The increase of intensity of interspecific competition may decrease the availability of floral rewards, which may also reduce the net energy intake for a single foraging trip. In this circumstance, a shift to foraging for nectar may benefit colony development.
Although we did not detect niche differentiation in proboscis length and flower use for the two mainly nectar-seeking bumble bees (B. lepidus and B. friseanus ), because they foraged on a wide range of flowering plants, the application of the index of composite flower depth uncovered that such niche differentiation might occur. The fact that the ratios of worker proboscis length for the species pairs are 1.18 (festivus :friseanus ) and 1.25 (friseanus :lepidus ), close to the ratio of 1.2 predicted as a minimum for coexisting species (Inouye, 1978), suggests that proboscis length might play some role in the resource partitioning. When the intensity of interspecific competition was low, the flowers chosen by B. friseanus presented a larger composite flower depth than those visited by B. lepidus , the bee with relatively shorter proboscis length (Fig. 4). The individuals of B. lepidus had a much longer proboscis length than the composite flower depth of the plants they visited. Although long-tubed flowers may prevent a short-proboscid pollinator from removing nectar (unless they resort to nectar robbing), long-proboscid pollinators may retain the option of visiting short-tubed flowers for nectar (Nilsson, 1988; Johnson et al., 2017; Vlašánková et al., 2017). Our study suggests that the match of proboscis length and flower depth may not be important in the niche differentiation in proboscis length and flower use at our study site.
However, the bees’ foraging preferences were significantly influenced by intensity of interspecific competition. As the competition intensity increased, B. friseanus , the bee with relatively longer proboscis, shifted to visit plants with shorter flower depth, and the bees with shorter proboscis (B. lepidus ) tended to visit plants with longer flower depth. When the competition intensity reached its highest point, the ratio of the composite flower depth to proboscis length was in the range of 0.5 to 0.6 for both the two bees. However, the plant assemblages visited by the two bees were still highly variable except for the most dominant plant, Polygonum macrophyllum (Table S3). When the competitor bee species reduced the availability of floral resources, enhancement of foraging efficiency in terms of net energy intake would be important for the coexisting bees. The match of proboscis length and flower depth might highly enhance the foraging efficiency, e.g. reducing the handling time on a flower (Inouye, 1980; Ranta & Lundberg, 1980). The temporal dynamic in plant preference for the two mainly nectar-seeking bees in response to competition intensity contributed to our understanding of niche differentiation in proboscis length and flower use.
In conclusion, our study indicated that the community context of foraging strategy for bumble bees not only responded to intensity of interspecific competition, but also depended on preference for reward type (pollen vs. nectar). Moreover, the change of foraging strategy in response to competition intensity depended on the species. For the mainly pollen-seeking bee, B. festivus , when the intensity of competition with the other two bees (B. friseanus and B. lepidus ) increased, it was forced to adjust its foraging preference between plants with and without nectar but not to change to target plants with different flower depths. At relatively high abundance, the latter two species did not respond to intensity of competition withB. festivus at all. The two bees mainly sought nectar, and an increase in competition intensity impelled them to shift their flower preferences to enhance the foraging efficiency. Our findings suggest that in a community, intensity of interspecific competition, relative bee abundance, and reward preference (pollen vs. nectar) might collectively influence the niche differentiation in flower use for a bumble bee species. The coexistence of multiple bumble bee species in a community should be achieved by adjusting foraging strategy in response to intensity of interspecific competition for floral resources.
Acknowledgements We thank Min Lv, De-Xin Liu, Yong-Deng He, Si-Zhen Liu and Yi-Fan Ma for field assistance and Yan-Hui Zhao and Huan Liang for helping in insect identification. This work was supported by the National Natural Science Foundation of China (Nos. 31970253 and 31770255 to CFY, 31800194 to ZMY).