Advances in automatic identification of flying insects using optical sensors and machine learning.

Worldwide, farmers use insecticides to prevent crop damage caused by insect pests, while they also rely on insect pollinators to enhance crop yield and other insect as natural enemies of pests. In order to target pesticides to pests only, farmers must know exactly where and when pests and beneficial insects are present in the field. A promising solution to this problem could be optical sensors combined with machine learning. We obtained around 10,000 records of flying insects found in oilseed rape (Brassica napus) crops, using an optical remote sensor and evaluated three different classification methods for the obtained signals, reaching over 80% accuracy. We demonstrate that it is possible to classify insects in flight, making it possible to optimize the application of insecticides in space and time. This will enable a technological leap in precision agriculture, where focus on prudent and environmentally-sensitive use of pesticides is a top priority.

Monitoring Neonicotinoid Exposure for Bees in Rural and Peri-urban Areas of the U.K. during the Transition from Pre- to Post-moratorium.

Concerns regarding the impact of neonicotinoid exposure on bee populations recently led to an EU-wide moratorium on the use of certain neonicotinoids on flowering crops. Currently, evidence regarding the impact, if any, the moratorium has had on bees' exposure is limited. We sampled pollen and nectar from bumblebee colonies in rural and peri-urban habitats in three U.K. regions: Stirlingshire, Hertfordshire, and Sussex. Colonies were sampled over three years: prior to the ban (2013), during the initial implementation when some seed-treated winter-sown oilseed rape was still grown (2014), and following the ban (2015). To compare species-level differences, in 2014 only, honeybee colonies in rural habitats were also sampled. Over half of all samples were found to be contaminated ( n = 408), with thiamethoxam being the compound detected at the highest concentrations in honeybee- (up to 2.29 ng/g in nectar in 2014, median ≤ 0.1 ng/g, n = 79) and bumblebee-collected pollen and nectar (up to 38.77 ng/g in pollen in 2013, median ≤ 0.12 ng/g, n = 76). Honeybees were exposed to higher concentrations of neonicotinoids than bumblebees in 2014. While neonicotinoid exposure for rural bumblebees declined post-ban (2015), suggesting a positive impact of the moratorium, the risk of neonicotinoid exposure for bumblebees in peri-urban habitats remained largely the same between 2013 and 2015.

Oilseed rape (Brassica napus) as a resource for farmland insect pollinators: quantifying floral traits in conventional varieties and breeding systems.

Oilseed rape (OSR; Brassica napus L.) is a major crop in temperate regions and provides an important source of nutrition to many of the yield-enhancing insect flower visitors that consume floral nectar. The manipulation of mechanisms that control various crop plant traits for the benefit of pollinators has been suggested in the bid to increase food security, but little is known about inherent floral trait expression in contemporary OSR varieties or the breeding systems used in OSR breeding programmes. We studied a range of floral traits in glasshouse-grown, certified conventional varieties of winter OSR to test for variation among and within breeding systems. We measured 24-h nectar secretion rate, amount, concentration and ratio of nectar sugars per flower, and sizes and number of flowers produced per plant from 24 varieties of OSR representing open-pollinated (OP), genic male sterility (GMS) hybrid and cytoplasmic male sterility (CMS) hybrid breeding systems. Sugar concentration was consistent among and within the breeding systems; however, GMS hybrids produced more nectar and more sugar per flower than CMS hybrid or OP varieties. With the exception of ratio of fructose/glucose in OP varieties, we found that nectar traits were consistent within all the breeding systems. When scaled, GMS hybrids produced 1.73 times more nectar resource per plant than OP varieties. Nectar production and amount of nectar sugar in OSR plants were independent of number and size of flowers. Our data show that floral traits of glasshouse-grown OSR differed among breeding systems, suggesting that manipulation and enhancement of nectar rewards for insect flower visitors, including pollinators, could be included in future OSR breeding programmes.

Bumblebee flight distances in relation to the forage landscape.

1. Foraging range is a key aspect of the ecology of 'central place foragers'. Estimating how far bees fly under different circumstances is essential for predicting colony success, and for estimating bee-mediated gene flow between plant populations. It is likely to be strongly influenced by forage distribution, something that is hard to quantify in all but the simplest landscapes; and theories of foraging distance tend to assume a homogeneous forage distribution. 2. We quantified the distribution of bumblebee Bombus terrestris L. foragers away from experimentally positioned colonies, in an agricultural landscape, using two methods. We mass-marked foragers as they left the colony, and analysed pollen from foragers returning to the colonies. The data were set within the context of the 'forage landscape': a map of the spatial distribution of forage as determined from remote-sensed data. To our knowledge, this is the first time that empirical data on foraging distances and forage availability, at this resolution and scale, have been collected and combined for bumblebees. 3. The bees foraged at least 1.5 km from their colonies, and the proportion of foragers flying to one field declined, approximately linearly, with radial distance. In this landscape there was great variation in forage availability within 500 m of colonies but little variation beyond 1 km, regardless of colony location. 4. The scale of B. terrestris foraging was large enough to buffer against effects of forage patch and flowering crop heterogeneity, but bee species with shorter foraging ranges may experience highly variable colony success according to location.