Friends with benefits? A holistic approach to diffuse mutualism in plant-pollinator interactions
Basic data for this project
Type of project: Participation in other joint projects
Duration: 01/11/2021 - 31/10/2024
Description
Most flowering plant species benefit from the pollination services of animals, where the animal helps transport pollen between flowers and individual plants, thereby supporting seed and fruit production. In turn, plants provide nutritional resources to pollinators, include nectar (which serves as a source of carbohydrates) and pollen (which serves as a source of protein and fat). This plant-pollinator mutualism is essential to supporting terrestrial ecosystems and is also vital for human agricultural production. Declines in pollinator populations due to anthropogenic change (including habitat loss) endangers these ecosystems and threatens food security. Generalist pollination systems in which flowering plant species uses a broad spectrum of pollinators for pollination services, and pollinator species visit diverse plant species to meet their nutritional needs, can create robust and resilient plant-pollinator communities. Designing and restoring habitats with these generalist systems can serve to reduce or reverse pollinator decline, since multiple pollinator species will be supported. Yet, how such generalist and diffuse mutualisms are formed in an ecological community remains a mystery - primarily due to the lack of tools supporting high throughput monitoring of pollinator visitation patterns and the lack of plant systems where pollinator-attractive traits can be precisely genetically controlled. This interdisciplinary project combines computer vision with plant genetics and pollinator behavioral ecology to identify the mechanisms mediating plant-pollinator interaction networks in generalist pollination systems. In addition to generating novel, broadly accessible monitoring tools and experimental plant systems, this study will provide the first comprehensive characterisation of the plant traits that attract and reward different pollinator species and the pollinator behavioral strategies which optimise pollinator nutrient acquisition in ecological communities. By linking plant genetics, pollinator health and quantitative behavioral data, this project will generate novel concepts and approaches to mitigate pollinator decline in agricultural, urban and natural ecosystems.
Keywords: Machine Learning; Artificial Intelligence; Computer Vision; Insect Behaviour; Sustainable Development