Dominance of non-wetland-dependent pollinators in a plant community in a small natural wetland in Shimane, Japan

Abstract

Many wetland plants rely on insects for pollination. However, studies examining pollinator communities in wetlands remain limited. Some studies conducted in large wetlands (> 10 ha) have suggested that wetland-dependent flies, which spend their larval stage in aquatic and semi-aquatic habitats, dominate as pollinators. However, smaller wetlands surrounded by secondary forests are more prevalent in Japan, in which pollinators from the surrounding environment might be important. Additionally, information regarding floral traits that attract specific pollinator groups in wetland communities is scarce. Therefore, this study aimed to understand the characteristics of insect pollinators in a small natural wetland (2.5 ha) in Japan. We examined the major pollinator groups visiting 34 plant species and explored the relationship between the flower visitation frequency of each pollinator group and floral traits. Overall, flies were the most dominant pollinators (42%), followed by bees and wasps (33%). Cluster analysis indicated that fly-dominated plants were the most abundant among 14 of the 34 target plant species. However, 85% of the hoverflies, the most abundant flies, and 82% of the bees were non-wetland-dependent species, suggesting that these terrestrial species likely originated from the surrounding environment. Therefore, pollinators from the surrounding environment would be important in small natural wetlands. Flies tend to visit open and white/yellow flowers, whereas bees tended to visit tube-shaped flowers, as in forest and grassland ecosystems. The dominance of flies in small wetlands would be due to the dominance of flowers preferred by flies (e.g., yellow/white flowers) rather than because of their larval habitats.

Keywords: Bumblebee; Diptera; Hover fly; Muscoid fly; Pollination syndrome; Pollinator community.

Fig. 1

Entomophily types in Akana Wetland target plants. This figure presents the results of a hierarchical cluster analysis using Ward’s method and the Shannon-Wiener diversity index was calculated based on the order-level pollinator composition of each plant species. The plant species were classified into five entomophily types: fly-type (Diptera-dominated type), bee-type (Apoidea-dominated type), coleoptera-type (Coleoptera-dominated type), generalist-type (both flies and other insects were abundant), and fly/bee-type (Diptera-Apoidea-mixed type). The numbers in parentheses indicate the total number of captured pollinators for each plant

Fig. 2

Flower morphology (open or tube-shaped) and the visitation frequency of flies and bees. We Tested for significant differences using generalized linear mixed models (GLMMs: Poisson error distribution). The details of GLMMs are shown in Table 1

Fig. 3

Flower color (blue-violet, white and yellow) and visitation frequency of flies and bees. We tested for significant differences using generalized linear mixed models (GLMMs) with a Poisson error distribution. More information on the GLMMs is presented in Table 1

Fig. 4

Ordination plot of non-metric multidimensional scaling (NMDS) based on the flower visitation frequency of different pollinator groups (bumblebees, other bees, muscoid flies, hover flies, coleoptera, and butterflies) for each plant species. This analysis explored the relationship between visitation frequency with flower size, density, shape (open or tube-shaped), and color (blue-violet, white/yellow). Only significant relationships from this analysis are presented. Detailed information on the NMDS is presented in Table 2