Dryness, wetness and temporary flooding reduce floral resources of plant communities with adverse consequences for pollinator attraction

1. Climate change alters precipitation regimes worldwide and is regarded as a major threat for pollinators and pollination services. Yet, not much is known on how wetter as well as drier conditions affect food resources for pollinators and pollinator attraction in a multi-species community context. It is unclear how community shifts under changed hydrological conditions might affect pollinators. 2. This mesocosm study addresses existing research gaps by investigating effects of drought, wetness and temporary flooding on floral resources and pollinator attraction within a plant community, initially including nine insect- and six wind-pollinated species. Floral resources were assessed over three growing seasons monitoring community descriptors (percentage of herbs and herbaceous species richness, presence of flowering insect-pollinated individuals), floral traits (flower height, size and weight, floral sugar content of one key species) and proxies for plant fitness (pollinator attraction, seed and flower weight of one key species). 3. Wetness and especially dryness decreased the species richness and biomass proportion of insect-pollinated herbaceous species and flowering herbaceous plant individuals occurred less frequently. Permanent wetness decreased the floral sugar content of the key species Trifolium pratense. Consequently, wet, temporary flooded and especially dry communities were visited by pollinators less often, and active pollinators spent less time within wet and flooded communities, while not at all visiting dry communities. Seed and flower weight of the obligate xenogamous T. pratense in flooded and dry communities was decreased, indicating negative consequences for plant fitness caused by a lack of pollination. 4. Synthesis: While dryness had negative effects for floral resources when looking at both, community descriptors and floral traits, negative effects of wetness and temporary flooding were mostly caused by a decrease of insect-pollinated herbaceous species. The study thus indicates that shifts in plant community composition are decisive for a predictive understanding of plant-pollinator interactions under environmental change, but have been neglected in past research. Changing precipitation patterns will adversely affect floral resources and pollinator attraction in agriculturally used temperate grassland, which might have widespread negative consequences for pollination services and food security in coming decades.water= hydrological conditions under which plant communitis grew from 2016 until 2019 L=waterlogged D=dry M=mesic PL=temporary flooding species: herbaceous plant species Am= Achillea millefolium Bp= Bellis perennis Cb= Crepis biennis Cj= Centaurea jacea Ga= Galium mollugo agg. Pl= Plantago lanceolata Td= Trifolium dubium Tp= Trifolium pratense Vc= Vicia cracca pot pot number no. visitors number of insects touching flowers of herbaceous species time spent seconds that insect stayed on flowers of a certain pot time spent without rest seconds without insects resting on flowers for more than 60 s visitor kind of visitor: BB= bumblebee, B= bee, D=dipteran species (visitor 1-4; one pot was never visited by more than four pollinators, time [s] 1 refers to time visitor 1 spent in the pot (and so on) flow number of flowers touched by respective visitor species richness number of different insect-pollinated species in community present was plant species present in community (1=yes, 0=no) occurring in flowering state did species flower in a given year (1=yes, 0=no) abundance [%] abundance of herbaceous species in %, based on biomass production data (does not add up to 100, because graminoid species were present in communities as well) biomass production [g] total biomass production of one pot in grams herb weight weight of herbaceous species from one pot grass weight weight of graminoid species from one pot singly potted: results of floral traits for plants that grew singly in pots in 2018 NA means that data were not available or could not be measured (e.g. no floral traits could be measured when flowering species was not present) Funding provided by: Deutsche ForschungsgemeinschaftCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001659Award Number: WA 3442/2-1Data was collected in a pot experiment with sown grassland communities, as described in the article