Impacts of mass-flowering crops on wildflower pollination network structure and function

Background/Question/Methods: The importance of pollination as an ecosystem service for increasing crop yields and maintaining wild plant communities has become increasingly evident in recent years. With around one-third of global crop production and 60-90% of wild flowering plant species benefiting from animal-mediated pollination, pollination as an ecosystem service has been valued globally at over 500 billion dollars. Concerns about how global crop production will be impacted if the world’s most important crop pollinator, Apis mellifera, declines have brought the value of diversified pollination services into the spotlight. In particular, over the last decade there has been growing interest in the role of wild insects as pollinators of commercially grown crops worldwide. While pollinator spillover from natural habitats to crop systems has been relatively well documented, much less is known about the effects of agricultural pollinator spillover/depletion in natural habitats within agricultural landscapes. Our project aims to explore the effects that mass-flowering crops have on wildflower pollination network structure and function in highly fragmented agricultural landscapes of SW-Western Australia. Specifically, we assess the pollen load composition of pollinators in these landscapes, and create site-specific pollination networks to determine how mass-flowering crops alter network structure.

Results/Conclusions: Preliminary results indicate that crop identity (being adjacent to insect attracting mass-flowering crops – canola, versus non-insects attracting cereal crops – wheat) significantly alters pollination network structures in remnant wildflower communities (anosim – p = 0.002). We analysed the following mutualistic network metrics in order to determine the process and extent of network differences: connectance, network specialisation, linkage density, and vulnerability. Pollination networks in wildflower remnants adjacent to canola are less connected (glm – p = 0.0239) and less specialised (glm – p = 0.0435) than those adjacent to wheat fields. Additionally, canola adjacent remnant networks have higher linkage densities (glm – p = 0.00906) and greater vulnerability scores (glm – p = 0.00181). These results are consistent with the expectations that wheat adjacent communities display higher specialisation resulting in a greater degree of ecological functionality in terms of plant reproductive success. Given the short-term needs of annual wildflower communities, the importance of specialisation for ensuring conspecific pollen transfer may be more essential for annual wildflower persistence than the long-term needs of increased generalisation and linkage density for improving system-level robustness. This study provides valuable insights into how mass-flowering crops alter pollinator-plant network dynamics in remnant patches of vegetation with high conservation value.