July 04, 2023

Local or imported queens?

In Best practices 18 min. read

Previous article
Written by
Nico Coallier
Next article

Beekeepers commonly accept that local queens are better adapted to local conditions and will therefore survive and perform better than imported queens. However, few studies have really examined this assumption. Furthermore, it is unclear what is considered imported versus local. In this first blog, we will explore if we can find an association between the origin of a queen and the survival of a colony. We’ll do so by reviewing the literature and then comparing it with the Nectar dataset.

What’s a good queen?

Queen breeders define a good queen as a queen that shows heritable and measurable traits (Maucourt and al. 2020). The main traits measured and used by breeders are honey production, gentleness, overwintering ability, hygienic behavior, mite resistance, brood pattern and bees being disease free which represent 69% of the surveyed beekeepers in the Canadian Honey Bee Queen Bee Breeders' Reference Guide. Given that survivorship is a high quality data point in Nectar’s dataset, we will use overwintering ability (i.e. winter survival) as the main indicator of the quality of a queen. 

While colony survival is not strictly dependent on the queen, in our dataset colonies are managed in similar ways and commercial beekeepers often try to make each hive similar in terms of grading which ease management and also reduce bias in our analysis. We assume colonies are managed in the same way in order to simplify the current analysis but future analysis will include management practices. A good queen is therefore defined as a queen that increases the probability of survival of a colonie over winter

Why would a local queen be better?

The general reasoning is that local queens are better adapted to local habitat (revivalqueenbees , Meixner and al. 2014, Büchler and al. 2014). By selecting queens that present desired traits such as honey production, hygienic behavior or high overwintering success, honeybees should present higher success in a similar context or environment. 

In a perfect world, where we could really control the cross-over between different populations and create multiple distinct lineages, this might work... However, honeybee genetic selection presents many challenges due to their biological cycle. Between 6 and 10 days after the queen is born, she will go on her mating flight (The biology of the honeybee). During this mating flight which occurs once in her life, she will mate with multiple drones. She can typically mate with between 10 and 77 drones leading to thousands of possible gene combinations (Withrow and Tarpy 2018). While few studies have looked into the impact of the origin of the queen on the survival of the colony, a few studies have looked at a couple hundred colonies in Europe and concluded that local queens lived longer than the imported ones. We haven’t found any study done in North America addressing the same question. The genetic diversity of the honeybee in North America is much smaller than in Europe (Munoz and Rua 2020; Whitfield and al. 2005) due to the large scale minimally regulated importation of bees to North America and the fact that Europe, not North America, is within the species’ natural range. Therefore, the importance of local queens in Europe can’t be generalized worldwide. While there are observations and discussions about the importance of local queens in Canada, we need to quantify their importance in order to avoid the potential negative impact of this practice. Indeed, there are however observations of declining genetic diversity in Europe which could be due to introgression (Munoz and Rua 2020; Themudo and al. 2020) and American beekeepers should be aware of the risk of importing genetics into their environment. 

Queens’ open mating systems can be considered a cause for another argument against imported queens. Imported queens are affecting the local gene pool which could result in undesired effects. While the sub-population of honeybees are more genetically distinct in Europe, this is not true for North America where the USDA found 93.79 percent of U.S. honey bees belonged to the North Mediterranean C lineage. The corruption of the local gene pool may be less important in colder climates like Canada where feral honey bees can’t survive without human intervention and thus, no true natural selection process takes place. With the open mating system, queen breeders may not be selecting the traits they are quantifying since 50% of the gene is coming from the drones that are generally not controlled. Therefore any traits present in the local population can be selected such as their efficiency towards certain crops or activities which are not the desired traits selected by the breeder. 

The freshness of the queen and the conditions of transport may impact the queen's health in different ways. For example, the number of days the queen was caged seems to be associated with higher quality queens (between 0 , 2 and 4 days) where queens surprisingly are of higher quality when caged 4 days (Wu and al. 2018) Another study looked at the temperature range and variation during transportation of queens and its effect on queen failures. While they didn’t find clear evidence of increased queen failure in function of temperature, they did observe slightly higher temperature and lower temperature variance for queen that failed (Withrow and al. 2019). Finally, some beekeepers I interact with have suggested that freshness seemed to be an important factor of queen quality, because queens that spend a long time in transport result in lower quality queens. While I haven’t found research looking at this question, temperature stress may explain the observed lower queen quality (Rousseau and al. 2018; Pettis and al. 2016) but it should only affect introduction rate and not survival of the colony.

There is little scientific evidence that clearly shows a link between the origin of the queen breeder and the quality of the queen once successfully introduced in a colony. We will explore that question using data from more than 3 000 colonies collected during the 2022-2023 season. Hives are located in the province of Quebec in Canada and come from 3 different beekeeping operations (2 commercial and 1 sideliner). We don’t expect the following study to answer the question is local queen better than imported one for the survival of your colony but rather question our data in order to better understand if the origin of the queen is an important factor of hive survival.  

What does the data say?

I chose to focus on Quebec beekeepers because the data quality was high and the proportion of imported to locally bred queens was even. In future studies, we will be able to expand to a larger dataset across all North America (over 100,000 hives). For the current dataset, we’ve only kept hives where the queen introduced survived the whole summer to avoid mixed effects of multiple queens within the same colony. After filtering out hives that could lead to bias and bad interpretation of the data, we ended up with a dataset of 2 950 hives.

For this study, 40% of the colonies have an imported queen while 60% have a queen purchased from a local breeder or made in-house by the beekeeper. The origin of queens is distributed this way:


Looking at the raw overwintering survival of each region, we observe similar survival for each region. However, homemade queens seem to perform worse (-11.28% survival relative to the average survival versus a -0.8 to +4.18% difference with the mean). This is likely explained by the high variation in the quality of care given to queens by beekeepers that are not focused on queen breeding, which is likely due to a lack of time in most cases. We will therefore remove the homemade queen from the rest of the analysis in order to really focus on the original question. Doing so, we observe very similar survival between local and imported queens (0.6% and -0.4% difference with the average population respectively).

Since we don’t observe an effect of the origin of the queen on the survival of these colonies, it may be that the distance between the local breeder and the beekeeper has an effect but the actual origin being not granular enough, the effect isn’t grasped. So let’s take the straight line distance between the average location of the hive and the origin of the queen, which we’ll assume is the headquarter of the breeder. Doing so, we observe no effect of the distance on the survival rate.


We didn’t have the information about the specific location of Italian breeders so we’ve used the middle of Italy as the origin for these queens

Later in the post, we will return to the subject of the origin’s effect in survival modeling . For now, distance between the breeder and the beekeeper doesn’t seem to have an effect on the quality of the queen. The other main difference between local and imported queens resides in the availability of queens for beekeepers in colder climates at specific times of year. Local queens in Quebec are not available until late May which might lead to survival difference of the colony since age of the queen is known to be a strong driver of the colony performance (Akyol and al. 2016 ; Ricigliano and al. 2018; Oberreiter & Brodschneider 2020). As expected, imported queens are on average 15.5 days older than local ones at wintering. Queen age here is referred to the number of days wince the queen was introduced in the hive.

As a beekeeper , I wouldn’t think a roughly two weeks difference in queen age would be a strong driver of performance in my colonies. However, it may be that local queens could outperform imported ones if they could be introduced earlier. This raises the question, is the relation between queen age and survival rate the same for imported versus local queens?

It seems that imported queens show no relation relative to their age while a positive slope is observed for local queens where older queen's lead to higher survival. Local queens surpassed the imported queen performance after 92 days. While some studies have looked at the effect of queen age with yearly difference (Akyol and al. 2016 ; Ricigliano and al. 2018; Oberreiter & Brodschneider 2020), we haven’t found any studies looking at the effect of age in terms of smaller time step (i.e. days, weeks or months) of the queen on the survival of the colonies. Another way to look at it is that the queen age is directly linked to the introduction date of that queen in the colony and it may be that there is a period in the season that leads to higher performing colonies. Let’s look at the effect of the introduction date on the resulting survival of these colonies.

It seems that early introduction leads to higher survival for local queens while imported queens show higher variation throughout the season but we don’t see a clear effect of the introduction date on the resulting survival of the colony. Indeed, we observe 461 colonies with imported queens introduced prior to July 1st and 309 colonies with local queens introduced prior to July 1st, these colonies show a survival difference with the average population of 0.4% for the imported one and of 7.2% for the local one. Colonies with local queens introduced early in the season seem to lead to higher survival. However when excluding colonies where the queen was introduced prior to the first local queen introduced, we lose that effect and both populations show similar survival. In Quebec, colonies pollinate blueberries in late May.his may explain why these early introductions lead to higher mortality, therefore the effect of the queen origin might not be the driver here.

In order to consider all the differences in management between both groups, we’ve built a shallow learning model adapted to survival modeling and provided the model with multiple feature such as : temperature variation at the location of the hive, precipitation sum at the location of the hive, average wind speed at the location of the hive, feeding round count, formic round count, oxalic round count, proportion of yard classified as honey, pollination and holding yard,  grade of the hive at wintering, queen origin and finally the beekeeper in order to consider the different management strategies. Doing so, we can abstract the context in which each colony lived and extract a less biased queen origin effect on the survival function of each group. In layman's terms, we will model the effect of queen origin while considering the difference between the groups so that we can observe a true queen origin effect on the survival function. The best model was evaluated mainly with the concordance index which reached 92.2% on unseen colonies! The feature that showed the strongest importance are the following:


We’ve split our data into a training and validation set (80%-20%) and tuned our model on 100 folds to get the best hyperparameters.

We see that the age of the queen is the fifth most important driver of survival while the top 3 is completed with the variation in temperature across the season and sum of rain across the season. We’ve looked at the reaction of both groups relative to temperature and rain and didn’t find differences in their relation with them. There is a positive correlation with temperature and survival and a negative relationship between sum of rain and survival. We will explore the impact of climatic conditions in a later blog post.

Let’s answer the question about the effect of queen origin on survival functions by extracting both functions from the model.

‍In the graphic above we see that the effect of the origin of the queen is not significant and slightly positive with respect to the imported queens. Confounding factors such as the age of the queen and the difference in management between colonies in the spring versus summer and autumn managements may explain this effect and one shouldn’t conclude that imported queens are better as of now. We need to collect more data!

What’s coming to BeeTrack and what are the next steps

Here, we’ve explored the effect of the queen origin on about 3000 colonies in Quebec and didn’t find the expected effect. However, we found evidence that management practices have a stronger impact on the resulting performance of the colony than the origin of the queen. While the general logic of artificially selecting queens that are better adapted to the climate of a region seems strong, in practice few breeders use insemination techniques or exclude their population from neighboring colonies leading to genetic mix with local bees. Therefore, the traits selected may be more influenced by the management practices of that region instead of truly being adapted to the habitat they are in. Selection is driven by the strongest effect, in other words, the strongest effect on mortality drives the selection of honeybees and it seems that the management practice has more importance on the survival than the genetics of these colonies. Therefore, we could expect imported queens to outperform local ones when beekeepers manage bees in similar ways as the region of origin (ex: large scale pollination). We did see a pattern in the data where imported queens started to perform better than local queens when they occupied mainly pollination yards while we saw the opposite effect in honey yards. This may be due to traits being selected for pollination by breeders in regions where pollination is the main activity (ex: California).

In the data, it seems that local queens are less variable in terms of their effect on survival and that early introduction of local queens could result in better performing hives. Currently, beekeepers have started to explore banking techniques so beekeepers in colder climates could be accessing local queens early in the season (Rousseau & Giovenazzo 2021). It would be interesting to compare the performance of these early introduced banked queens with imported ones since extensive pollination events occur early in the season and bias the analysis of performance of queens.

Coming to BeeTrack is better queen tracking and analytics. We’ve only scratched the surface of what’s possible with large scale data collection on queens. While survival of the colony is one interesting trait, other meaningful traits such as the strength of the colony, honey production, the acceptance rate and the requeening rate could show bigger differences between local and imported queens. I’m really excited to dig into these new data points and challenge more of the observational knowledge we, beekeepers,  base our decisions on! Finally, larger datasets and more advanced modeling techniques using deep learning approaches will allow us to better understand how queens interact with their habitat and generalize higher concepts. I firmly believe there is no such thing as the best queen but just the right queen for your needs!

For now, I would encourage beekeepers to support their local queen breeders since we haven’t found strong evidence that imported queens are better. By supporting your local breeder, you can not only develop relationships where you can influence the selected traits but also reduce the risk of disease and pests spreading while supporting your local economy and reducing the risk of polluting your local genetic diversity!

About Nectar

Nectar started is a precision beekeeping company based in Montreal, Canada and offers BeeTrack, a traceability and management platform for commercial beekeepers. 

Interested in learning more about Nectar and our solutions? Click Here.


  1. Hatjina, F., Bienkowska, M., Charistos, L., Chlebo, R., Costa, C., Dražić, M. M., Filipi, J., Gregorc, A., Ivanova, E. N., Kezic, N., Kopernicky, J., Kryger, P., Lodesani, M., Lokar, V., Mladenović, M., Panasiuk, B., Petrov, P. P., RaŠic, S., Skerl, M. I. S., … Wilde, J. (2014). A review of methods used in some European countries for assessing the quality of honey bee queens through their physical characters and the performance of their colonies. In Journal of Apicultural Research (Vol. 53, Issue 3, pp. 337–363). International Bee Research Association. https://doi.org/10.3896/IBRA.
  2. Akyol, E., Yeninar, H., Korkmaz, A., & Çakmak, I. (2008). An observation study on the effects of queen age on some characteristics of honey bee colonies. Italian Journal of Animal Science, 7(1), 19–25. https://doi.org/10.4081/ijas.2008.19
  3. Rousseau, A., & Giovenazzo, P. (2021). Successful indoor mass storage of honeybee queens (Apis mellifera) during winter. Agriculture (Switzerland), 11(5). https://doi.org/10.3390/agriculture11050402
  4. Alburaki, M., Madella, S., Lopez, J., Bouga, M., Chen, Y., & vanEngelsdorp, D. (2023). Honey bee populations of the USA display restrictions in their mtDNA haplotype diversity. Frontiers in Genetics, 13. https://doi.org/10.3389/fgene.2022.1092121
  5. Withrow, J. M., Pettis, J. S., & Tarpy, D. R. (2019). Effects of Temperature during Package Transportation on Queen Establishment and Survival in Honey Bees (Hymenoptera: Apidae). Journal of Economic Entomology, 112(3), 1043–1049. https://doi.org/10.1093/jee/toz003
  6. Maucourt, S., Fortin, F., Robert, C., & Giovenazzo, P. (2020). Genetic parameters of honey bee colonies traits in a canadian selection program. Insects, 11(9), 1–16. https://doi.org/10.3390/insects11090587
  7. Rousseau, A., Houle, É., & Giovenazzo, P. (2020). Effect of shipping boxes, attendant bees, and temperature on honey bee queen sperm quality (Apis mellifera). Apidologie, 51(5), 724–735. https://doi.org/10.1007/s13592-020-00756-3
  8. Espregueira Themudo, G., Rey-Iglesia, A., Robles Tascón, L., Bruun Jensen, A., da Fonseca, R. R., & Campos, P. F. (2020). Declining genetic diversity of European honeybees along the twentieth century. Scientific Reports, 10(1). https://doi.org/10.1038/s41598-020-67370-2
  9. Oberreiter, H., & Brodschneider, R. (2020). Austrian COLOSS survey of honey bee colony winter losses 2018/19 and analysis of hive management practices. Diversity, 12(3). https://doi.org/10.3390/d12030099
  10. Bixby, C., Guarna, M. M., & Hoover, S. E. (2019). Canadian Honey Bee Queen Breeders’ Reference Guide. In Canadian Association of Professional Apiculturists Publication (Vol. 55).
  11. Whitfield, C. W., Behura, S. K., Berlocher, S. H., Clark, A. G., Johnston, J. S., Sheppard, W. S., Smith, D. R., Suarez, A. v, Weaver, D., & Tsutsui, N. D. (n.d.). Thrice Out of Africa: Ancient and Recent Expansions of the Honey Bee, Apis mellifera. www.sciencemag.org
  12. Muñoz, I., & de la Rúa, P. (2021). Wide genetic diversity in Old World honey bees threaten by introgression. Apidologie, 52(1), 200–217. https://doi.org/10.1007/s13592-020-00810-0
  13. Büchler, R., Costa, C., Hatjina, F., Andonov, S., Meixner, M. D., le Conte, Y., Uzunov, A., Berg, S., Bienkowska, M., Bouga, M., Drazic, M., Dyrba, W., Kryger, P., Panasiuk, B., Pechhacker, H., Petrov, P., Kezić, N., Korpela, S., & Wilde, J. (2014). The influence of genetic origin and its interaction with environmental effects on the survival of Apis Mellifera L. Colonies in Europe. Journal of Apicultural Research, 53(2), 205–214. https://doi.org/10.3896/IBRA.
  14. Ricigliano, V. A., Mott, B. M., Floyd, A. S., Copeland, D. C., Carroll, M. J., & Anderson, K. E. (2018). Honey bees overwintering in a southern climate: Longitudinal effects of nutrition and queen age on colony-level molecular physiology and performance. Scientific Reports, 8(1). https://doi.org/10.1038/s41598-018-28732-z
  15. Pickett, K. L., Suresh, K., Campbell, K. R., Davis, S., & Juarez-Colunga, E. (2021). Random survival forests for dynamic predictions of a time-to-event outcome using a longitudinal biomarker. BMC Medical Research Methodology, 21(1). https://doi.org/10.1186/s12874-021-01375-x
  16. Meixner, M. D., Büchler, R., Costa, C., Francis, R. M., Hatjina, F., Kryger, P., Uzunov, A., & Carreck, N. L. (2014). Honey bee genotypes and the environment. In Journal of Apicultural Research (Vol. 53, Issue 2, pp. 183–187). International Bee Research Association. https://doi.org/10.3896/IBRA.
  17. Guda, A.-Z. G., Elshemy, A. A.-E., & El- Enany, Y. E. (2023). EFFECT OF REARING SEASON ON QUEEN AND THE REPRODUCTION OF <i>APIS MELLIFERA CARNICA</i> AND A <i>MELLIFERA LIGUSTICA</i> Indian Journal of Entomology. https://doi.org/10.55446/IJE.2023.885
  18. Withrow, J. M., & Tarpy, D. R. (2018). Cryptic “royal” subfamilies in honey bee (Apis mellifera) colonies. PLoS ONE, 13(7). https://doi.org/10.1371/journal.pone.0199124
  19. Pettis, J. S., Rice, N., Joselow, K., van Engelsdorp, D., & Chaimanee, V. (2016). Colony failure linked to low sperm viability in honey bee (Apis mellifera) queens and an exploration of potential causative factors. PLoS ONE, 11(2). https://doi.org/10.1371/journal.pone.0147220

About the author

Nico Coallier

Nico Coallier is a Principal Data Scientist at Nectar Technologies, leading the development of models for beekeeping management. With a background in deep reinforcement learning and ecology, Nico has held various advisory and leadership roles in data science, AI, and technology across multiple companies. With a strong focus on project lifecycle management, Nico has a track record of implementing AI-driven solutions and guiding teams towards efficient workflows and successful project outcomes. From founding a beekeeping company to contributing to scientific research, Nico's diverse experience showcases a passion for data-driven innovation and problem-solving.

Join a thriving community of beekeepers making a difference.

Request a demo