Bridging the Gap between Field Experiments and Machine Learning: The EC H2020 B-GOOD Project as a Case Study towards Automated Predictive Health Monitoring of Honey Bee Colonies.

Honey bee colonies have great societal and economic importance. The main challenge that beekeepers face is keeping bee colonies healthy under ever-changing environmental conditions. In the past two decades, beekeepers that manage colonies of Western honey bees (Apis mellifera) have become increasingly concerned by the presence of parasites and pathogens affecting the bees, the reduction in pollen and nectar availability, and the colonies' exposure to pesticides, among others. Hence, beekeepers need to know the health condition of their colonies and how to keep them alive and thriving, which creates a need for a new holistic data collection method to harmonize the flow of information from various sources that can be linked at the colony level for different health determinants, such as bee colony, environmental, socioeconomic, and genetic statuses. For this purpose, we have developed and implemented the B-GOOD (Giving Beekeeping Guidance by computational-assisted Decision Making) project as a case study to categorize the colony's health condition and find a Health Status Index (HSI). Using a 3-tier setup guided by work plans and standardized protocols, we have collected data from inside the colonies (amount of brood, disease load, honey harvest, etc.) and from their environment (floral resource availability). Most of the project's data was automatically collected by the BEEP Base Sensor System. This continuous stream of data served as the basis to determine and validate an algorithm to calculate the HSI using machine learning. In this article, we share our insights on this holistic methodology and also highlight the importance of using a standardized data language to increase the compatibility between different current and future studies. We argue that the combined management of big data will be an essential building block in the development of targeted guidance for beekeepers and for the future of sustainable beekeeping.

Pesticide risk assessment: honeybee workers are not all equal regarding the risk posed by exposure to pesticides.

Toxicological studies in honeybees have long shown that a single pesticide dose or concentration does not necessarily induce a single response. Inter-individual differences in pesticide sensitivity and/or the level of exposure (e.g., ingestion of pesticide-contaminated matrices) may explain this variability in risk posed by a pesticide. Therefore, to better inform pesticide risk assessment for honeybees, we studied the risk posed by pesticides to two behavioral castes, nurse, and forager bees, which are largely represented within colonies and which exhibit large differences in their physiological backgrounds. For that purpose, we determined the sensitivity of nurses and foragers to azoxystrobin (fungicide) and sulfoxaflor (insecticide) upon acute or chronic exposure. Azoxystrobin was found to be weakly toxic to both types of bees. However, foragers were more sensitive to sulfoxaflor than nurses upon acute and chronic exposure. This phenomenon was not explained by better sulfoxaflor metabolization in nurses, but rather by differences in body weight (nurses being 1.6 times heavier than foragers). Foragers consistently consumed more sugar syrup than nurses, and this increased consumption was even more pronounced with pesticide-contaminated syrup (at specific concentrations). Altogether, the stronger susceptibility and exposure of foragers to sulfoxaflor contributed to increases of 2 and tenfold for the acute and chronic risk quotients, respectively, compared to nurses. In conclusion, to increase the safety margin and avoid an under-estimation of the risk posed by insecticides to honeybees, we recommend systematically including forager bees in regulatory tests.

Pesticide risk assessment in honeybees: Toward the use of behavioral and reproductive performances as assessment endpoints.

The growing gap between new evidence of pesticide toxicity in honeybees and conventional toxicological assays recommended by regulatory test guidelines emphasizes the need to complement current lethal endpoints with sublethal endpoints. In this context, behavioral and reproductive performances have received growing interest since the 2000s, likely due to their ecological relevance and/or the emergence of new technologies. We review the biological interests and methodological measurements of these predominantly studied endpoints and discuss their possible use in the pesticide risk assessment procedure based on their standardization level, simplicity and ecological relevance. It appears that homing flights and reproduction have great potential for pesticide risk assessment, mainly due to their ecological relevance. If exploratory research studies in ecotoxicology have paved the way toward a better understanding of pesticide toxicity in honeybees, the next objective will then be to translate the most relevant behavioral and reproductive endpoints into regulatory test methods. This will require more comparative studies and improving their ecological relevance. This latter goal may be facilitated by the use of population dynamics models for scaling up the consequences of adverse behavioral and reproductive effects from individuals to colonies.

Ecology of Varroa destructor, the Major Ectoparasite of the Western Honey Bee, Apis mellifera.

Varroa destructor is the most important ectoparasite of Apis mellifera. This review addresses the interactions between the varroa mite, its environment, and the honey bee host, mediated by an impressive number of cues and signals, including semiochemicals regulating crucial steps of the mite's life cycle. Although mechanical stimuli, temperature, and humidity play an important role, chemical communication is the most important channel. Kairomones are used at all stages of the mite's life cycle, and the exploitation of bees' brood pheromones is particularly significant given these compounds function as primer and releaser signals that regulate the social organization of the honey bee colony. V. destructor is a major problem for apiculture, and the search for novel control methods is an essential task for researchers. A detailed study of the ecological interactions of V. destructor is a prerequisite for creating strategies to sustainably manage the parasite.