Using DNA metabarcoding to investigate honey bee foraging reveals limited flower use despite high floral availability

Abstract

Understanding which flowers honey bees (Apis mellifera) use for forage can help us to provide suitable plants for healthy honey bee colonies. Accordingly, honey DNA metabarcoding provides a valuable tool for investigating pollen and nectar collection. We investigated early season (April and May) floral choice by honey bees provided with a very high diversity of flowering plants within the National Botanic Garden of Wales. There was a close correspondence between the phenology of flowering and the detection of plants within the honey. Within the study area there were 437 genera of plants in flower during April and May, but only 11% of these were used. Thirty-nine plant taxa were recorded from three hives but only ten at greater than 1%. All three colonies used the same core set of native or near-native plants, typically found in hedgerows and woodlands. The major plants were supplemented with a range of horticultural species, with more variation in plant choice between the honey bee colonies. We conclude that during the spring, honey bees need access to native hedgerows and woodlands to provide major plants for foraging. Gardens provide supplementary flowers that may increase the nutritional diversity of the honey bee diet.

Figure 1. Survey area illustrating habitat types and the number of plant genera in flower per m² for April and May.

Maps created in QGIS v2.8.4. www.qgis.org.

Figure 2. Plants found within honey collected in April and May from three honey bee colonies, along with the status and habitat of the recorded plants.

(A–C) denote the three hives. For April and May the presence of plants within the honey sample is shaded in black for each hive. The proportion of DNA reads (%) for each plant is then given for all of the hives together, for April, May and both months combined. Supplementary Data SI 1 provides the proportions per hive. Use: whether honey bees use the plant for nectar N or pollen P according to Howes, brackets denote that this plant is considered to be used infrequently for this purpose. Status of plant: native N, horticulture H, or both B. Plants that are designated as ‘both’ include native plants along with their horticultural relatives and varieties. These plants can be described as native and near-native. Habitat of the plant within the survey area: woodland, hedgerows and scrub W, grassland G, horticulture planting H. Growth form: woody tree or shrub W, herbaceous H, bulb or corm B. Taxa richness and diversity (Simpson’s Diversity Index) are provided for each colony separately and combined for each month.

Figure 3. The location of plants in flower during April and May within the study site and proportion of that plant’s DNA (%) found within the honey samples using DNA metabarcoding.

Maps are shown for all the plants with a DNA percentage of greater than 5% in the honey samples, with the results for all three hives combined. Maps created in QGIS v2.8.4. www.qgis.org.