Seasonal dynamics in a cavity-nesting bee-wasp community: Shifts in composition, functional diversity and host-parasitoid network structure

Abstract

Ecological communities are composed of species that interact with each other forming complex interaction networks. Although interaction networks have been usually treated as static entities, interactions show high levels of temporal variation, mainly due to temporal species turnover. Changes in taxonomic composition are likely to bring about changes in functional trait composition. Because functional traits influence the likelihood that two species interact, temporal changes in functional composition and structure may ultimately affect interaction network structure. Here, we study the seasonality (spring vs. summer) in a community of cavity-nesting solitary bees and wasps ('hosts') and their nest associates ('parasitoids'). We analyze seasonal changes in taxonomic compostion and structure, as well as in functional traits, of the host and parasitoid communities. We also analyze whether these changes result in changes in percent parasitism and interaction network structure. Our host and parasitoid communities are strongly seasonal. Host species richness increases from spring to summer. This results in important seasonal changes in functional composition of the host community. The spring community (almost exclusively composed of bees) is characterized by large, univoltine, adult-wintering host species. The summer community (composed of both bees and wasps) is dominated by smaller, bivoltine, prepupa-wintering species. Host functional diversity is higher in summer than in spring. Importantly, these functional changes are not only explained by the addition of wasp species in summer. Functional changes in the parasitoid community are much less pronounced, probably due to the lower parasitoid species turnover. Despite these important taxonomic and functional changes, levels of parasitism did not change across seasons. Two network metrics (generality and interaction evenness) increased from spring to summer. These changes can be explained by the seasonal increase in species richness (and therefore network size). The seasonal shift from a bee-dominated community in spring to a wasp-dominated community in summer suggests a change in ecosystem function, with emphasis on pollination in spring to emphasis on predation in summer.

Fig 1. Effect of season on host-parasitoid networks.

Spring (A) and summer (B) host-parasitoid network (data from 14 sites lumped together). Numbers correspond to species names in S4 Table. Width of grey bands denotes interaction strength (number of cells parasitized). Width of red and yellow bars indicates host abundance (number of cells produced). Note different scales (number of cells) in spring and summer networks.

Fig 2. Effect of season on community composition.

Nonmetric multidimensional scaling (NMDS) of (A) host and (B) parasitoid community composition in each season and site. Dots represent seasons (blue: spring; red: summer). Numbers represent species codes (S4 Table; red: bees; orange: wasps; black: parasitoids). Polygons encompass all sites within a season. Ellipses represent 0.95% confidence intervals. Only two of the three dimensions obtained in the analyses (k = 3) are displayed.

Fig 3. Effect of season on host nest diameter.

Distribution of cavity diameters used by hosts for nest construction in spring (n = 719 nests) and summer (n = 768 nests). (X-squared = 272.45, df = 6, P < 0.0001).