Pollination service delivery is complex: urban garden crop yields are best explained by local canopy cover and garden scale plant species richness.
Abstract
Pollination is an important ecosystem service to agriculture, however, the factors influencing pollination in urban food gardens are poorly understood. We investigated how features within urban environments, including floral resources and canopy cover, impacted (a) flower visitation and seed production of a model plant (Brassica rapa) and (b) total yields of produce from urban gardens in Sydney, Australia. Floral visitation to model plants was dominated by a few common species, with honeybees responsible for 53.4% of visitation and native stingless bees and two hoverfly species accounting for another 37.1% of visitors; 28 other species were collectively responsible for less than 10% of visitation. While insect pollinators were abundant, the relationship between floral visitation and model plant reproductive output was weak and not always positive. The pollination success of model plants was only marginally impacted by environmental features, while pollinator visitation had a small negative impact. However, total crop yields across urban gardens were strongly influenced by local canopy cover and garden scale plant species richness. Synthesis and applications. The factors influencing pollination service delivery to urban gardens are complex; pollinator visitation, richness and/or floral resource availability may not always reflect positive, plant reproductive outcomes. Yields of garden crops, whether pollinator dependent or not, were more strongly influenced by surrounding environmental variables, including broad scale measures of canopy cover and local plant diversity, than pollination-related metrics. This implies policies to support pollinator richness will likely differ to those supporting crop yields in urban gardens and will require assessment and management of external environmental factors.