This dataset contains counts of pollinators visiting different varieties of oilseed rape (OSR). Data were collected from four trial sites in the UK in May 2012. The trial sites comprised of 20 varieties (plots) replicated in three blocks on each farm but only 2 of the blocks at each site were used for pollinator observations. Pollinator observations were also only made where there were greater than 30 percent of OSR plants in flower in the plot and only when weather conditions were within standardised limits. For each plot per site a six minute observation period was made during which the number of pollinators within the following taxon groups were counted: bumblebees to the species level, solitary bees identified to general body forms (Lasiglossum to genus level; Osmia separated to bicolour and rufa; Andrena separated to body forms typical of dorsata, carantonica, nigroaenea, haemorrhoa, fulva, flavipies, nitida, cineraria, bicolour and minuta), large hoverflies (> 12 mm), small hoverflies (< 11 mm), and Bibionidae. Each variety was observed for two separate six minute periods to reduce the impacts of minor fluctuations in weather that may reduce pollinator observations within single six minute periods. The dataset was collected as part of a project which aimed to identify key pollinators for OSR and identify if there are feeding preferences for individual varieties.
Publication date: 2014-05-07
All field based data collection was undertaken by either experts in bee identification or were experienced field ecologists. Consistency of identification between individuals undertaking recording was checked. To ensure quality of data entry standardised data entry sheets and a common set of unique abbreviations for species names were used. A common nomenclature was used to ensure that consistent species names were used and to avoid situations where taxonomic revision cause false duplications of species within the data set. Data entry was based on transcribing paper copies of field data sheets and was undertaken by a single individual. Following data entry, values entered were double checked to ensure accidental typing errors were removed. Graphing of raw values was used to identify potential out-of-range values. As data on the OSR varieties under testing is considered commercially sensitive codes rather than names are used for OSR varieties.