In temperate climates, winter can be an arduous time for eusocial insects. Survival of honey bee colonies during winter depends on a delicate balance between hive thermoregulation, managing the food reserves, and timing the onset of the new worker bee generation. Winter survival is influenced by several factors, most notably colony size, Varroa mites infestation levels and the availability of stored food. Importantly, the climatic conditions and floral resources of the previous foraging season can also impact honey bee health and colony strength before hibernation. This study, conducted across Europe, examines how landscape composition and weather conditions affect winter survival of honey bee colonies. It uses pollen diversity as a proxy for flower resource quality and available foraging days as a climatic variable to understand their causal relationships to winter survival. We found that landscapes with higher percentages of agricultural areas increased pollen diversity collected by honey bees in autumn, whereas higher percentages of semi-natural areas increased the diversity during summer. Spring and autumn pollen diversity was the main driver for winter survival success, emphasizing the importance of diverse flower resources for colony health. While we did not find a statistically significant effect of weather on winter survival, trends suggest potential influences, warranting further research to confirm and clarify the role of seasonal foraging on colony health. Our study highlights the critical role of including floral resource diversity and weather conditions, in a comprehensive framework for studying honey bee hibernation. It suggests that increasing plant diversity around apiaries and implementing agricultural practices that enhance floral resources can significantly improve winter survival, with honey bee colonies benefiting even in landscapes with higher agricultural activity, distinct from the needs of other pollinators.

The study quantified landscape composition around apiaries using QGIS with a 2-km buffer and land use data from the CORINE 2018 Land Cover map (European Environment Agency and European Environment Agency 2019), classifying habitats into four categories. Pollen diversity, used as a proxy for floral resource diversity, was assessed through monthly pollen collection from March to November 2022 using pollen traps, followed by DNA metabarcoding for species identification (Sickel et al.,2015). Weather data were obtained using local temperature sensors and precipitation records retrieved using the R package climate (Czernecki et al., 2020), with seasonal foraging conditions calculated based on temperature and precipitation thresholds. Statistical analysis employed DAGs to explore relationships between landscape composition, floral diversity, weather conditions, and honey bee overwintering success. Two GLMMs were fitted: one examining the effect of landscape composition on pollen diversity and another assessing the influence of foraging days and floral resource diversity on winter survival. Analyses were conducted in R using various statistical and visualization packages, ensuring model validity through diagnostic checks. All statistical analyses were implemented using R v.4.3.1 (R Core Team 2023). Both models were fitted using the glmmTMB package (Brooks et al., 2017). The package DHARMa (Hartig 2022) allowed us to evaluate the models’ performance. Model coefficients from the GLMM were generated using the packages broom.mixed (Bolker & Robinson 2022) and emmeans (Lenth 2023). The graphs were generated using R packages ggplot (Wickham 2016) and ggeffects (Lüdecke 2018). Other packages used are specified in the script.

REFERENCES

Bolker, B., & Robinson, D. (2022). Broom.mixed: Tidying methods for mixed models. https://CRAN.R-project.org/package=broom.mixed

Brooks, M. E., Kristensen, K., van, K. J., Magnusson, A., Berg, C. W., Nielsen, A., Skaug, H. J., Maechler, M., & Bolker, B. M. (2017). glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. 9. https://doi.org/10.32614/RJ-2017-066

Czernecki, B., Głogowski, A., & Nowosad, J. (2020). Climate: An r package to access free in-situ meteorological and hydrological datasets for environmental assessment. 12, 394. https://doi.org/10.3390/su12010394

European Environment Agency, & European Environment Agency. (2019). CORINE Land Cover 2018 (vector), Europe, 6-yearly - version 2020_20u1, May 2020. European Environment Agency. https://doi.org/10.2909/71C95A07-E296-44FC-B22B-415F42ACFDF0

Hartig, F. (2022). DHARMa: Residual diagnostics for hierarchical (multi-level / mixed) regression models. https://CRAN.R-project.org/package=DHARMa

Lenth, R. V. (2023). Emmeans: Estimated marginal means, aka least-squares means. https://CRAN.R-project.org/package=emmeans

Lüdecke, D. (2018). Ggeffects: Tidy data frames of marginal effects from regression models. 3, 772. https://doi.org/10.21105/joss.00772

R Core Team. (2023). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/

Sickel, W., Ankenbrand, M. J., Grimmer, G., Holzschuh, A., Härtel, S., Lanzen, J., Steffan-Dewenter, I., & Keller, A. (2015). Increased efficiency in identifying mixed pollen samples by meta-barcoding with a dual-indexing approach. BMC Ecology, 15(1), 20. https://doi.org/10.1186/s12898-015-0051-y

Wickham, H. (2016). ggplot2: Elegant graphics for data analysis. https://ggplot2.tidyverse.org