Long-term organic farming and floral diversity promotes stability of bee communities in agroecosystems


Increasing bee diversity promotes pollination services on farms. Yet, given the high turnover in pollinator communities, without knowledge of how pollinator communities assemble, it is difficult to conserve or increase bee diversity. Thus, a mechanistic understanding of factors mediating pollinator community assembly could promote pollinator conservation measures. To assess the determinants of pollinator community assembly and structure, we surveyed bee communities and floral resources on 36 farms ranging from 0 to 43?years in organic production. We used niche-based and stochastic species abundance models to characterise the mechanisms driving community assembly, and an additive partition of beta diversity to evaluate resource and species turnover (i.e. community structure). We then used statistical models to assess whether resource turnover or time in organic production altered community assembly and beta diversity, and a jackknife analysis to assess the sensitivity of top models to resource and species identity. We show that bee communities on farms that practiced organic methods for longer assembled by niche-based rather than stochastic processes and had less turnover in bee species across years. Because our model of niche-based processes assumes resource use, these results indicate bee communities reflect underlying species-specific resource preferences (e.g. floral and/or nesting resources) and that longer periods of organic management reduced dissimilarity mediated by species replacement. Our jackknife approach then examined the role of species identity effects in beta diversity, showing changes in floral resources increased dissimilarity driven by bee species loss, but only in landscapes simplified by urbanisation. This jackknife analysis then indicated that landscape resource replacement which was not driven by particular landscape classes, mediated bee species replacement wherein dissimilarity was driven by a generalist native bee, suggesting bee life history (e.g. flexibility in resource use) and landscape complementarity, rather than identity, underlie patterns identified using the beta diversity equations. Our results show bee communities assemble by niche-based processes, evidenced by collinearity in resource and bee species turnover. Because niche-based assembly indicates ecosystem health, farmers who adopt crop diversification and practice organic methods for longer may promote pollinator diversity and stability leading to improved pollination in farms.

Functional Ecology