Disentangling direct and indirect effects of habitat fragmentation on wild plants' pollinator visits and seed production. Issue 5 (31st March 2020)

• Record Type: Journal Article
• Authors: Lázaro, Amparo; Fuster, Francisco; Alomar, David; Totland, Ørjan
• Is Part Of: Ecological applications. Volume 30:Issue 5(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-03-31
• Subjects: landscape composition -- patch complexity -- patch isolation -- patch size -- plant reproduction -- pollinator evenness -- pollinator richness -- visitation rates; Ecology -- Periodicals; Environmental protection -- Periodicals; Biology, Economic -- Periodicals; 577.05
• DOI: 10.1002/eap.2099
• Languages: English
• ISSNs: 1051-0761

Abstract

Abstract: Habitat fragmentation threatens plant and pollinator communities, as well as their interactions. However, the effects of landscape fragmentation on the pollination of wild plant species are not well understood yet, partly because there are many correlated features in fragmented landscapes (e.g., decreased patch size, increased isolation, and patch complexity) whose influences are difficult to disentangle. Using a structural equation modeling approach, we assessed the direct and indirect effects of landscape fragmentation (patch size, isolation and complexity, percentage of surrounding land in forest) on the abundance, functional‐group richness, and evenness of pollinators of 24 habitat fragments within an agricultural landscape in Southern Norway. In addition, we studied how these variables affected visitation rates (visits per flower) and seed production (seed set, seed mass) in the four most abundant plant species in the area. Flower abundance was higher in larger and complex patches and decreased with the percentage of forest in the surroundings, while flower richness increased with patch complexity. We found a direct negative relationship between patch complexity and the overall number of pollinator visits that the habitat fragments received. Apart from this direct landscape effect, pollinator visits were mostly affected by the floral communities, with overall flower abundance and richness increasing both total number of pollinator visits and pollinator‐group richness, and flower richness having an additional negative influence on pollinator‐group evenness. Interestingly, we did not find any direct link between visitation rates and reproductive success for any of the study plant species. Instead, several landscape variables directly affected species seed production, although the effects of landscape on seed production were highly species specific. Patch complexity had a negative effect on seed production in two of the four focal species, while other components of the landscape had species‐specific effects. Increasing fragmentation of agricultural landscapes affects pollination interactions at the community level and the reproduction of wild plants. However, understanding the effects of fragmentation on seed production requires going beyond estimating visitation rates, since landscape effects on plant reproduction are not always related to overall interaction frequencies.