Can Thermal Imaging Using a FLIR Camera Reliably Detect Queenright Status Within a Colony?

Click Here if you listened. We’d love to know what you think. There is even a spot for feedback! Read along below!

Students of the University of Montana Master Beekeeping Class, Summer 2025

By: Christine Crick (Boise, ID), Jeremiah Bennett (Monroe, GA), Luke Harvey (Chillicothe, IL), Jason Howell (Jackson, OH), Annalie Neve (Syracuse, UT), Scott Taylor (Springfield, OH)

NOTE: For ease of analysis and enhanced visual clarity, all graphs and hyperlinked figures in this document are interactive. Readers may double-click on any chart, graph, or embedded hyperlink to open a larger version for closer examination. This feature is intended to support detailed inspection of data and improve accessibility of complex visual information.

Abstract
This study investigated the use of thermal imaging technology as a non-invasive method to detect queenright status in honey bee (Apis mellifera) colonies. Queenright colonies, those containing a healthy, actively laying queen, are central to hive productivity and social structure. Traditional methods of assessing queen presence often rely on intrusive inspections, which may disrupt colony function and place the queen at risk. Our study explored whether consistent thermal differences could be detected between queenright and queenless colonies using FLIR imaging technology in a real-world beekeeping setting.

Honey bee colonies generate and regulate heat through metabolic activity, particularly in the brood nest. This dynamic can be visualized using radiometrically calibrated thermal cameras. Previous research by Shaw et al. (2010) demonstrated that pre-dawn infrared imaging could reliably estimate colony population size by detecting heat signatures correlated with internal brood presence, without opening the hive. These findings offer early validation for non-invasive methods and support further exploration of thermal cues linked specifically to queenright status.

Introduction
The status of the queen is fundamental to the stability and productivity of a honey bee colony. A queenright colony is one with a healthy, mated queen actively producing eggs and pheromones, consistent brood patterns, and social cohesion. As the sole reproductive female, the queen’s ability to lay fertilized eggs ensures the ongoing replenishment of the worker population. Equally critical is her production of queen mandibular pheromone (QMP), which regulates worker behavior, suppresses ovary development in other females, and maintains social order. A decline in the queen’s fertility or presence, exhibited in absent pheromone production, can lead to disorganization, reduced productivity, or even colony collapse. Because of this, a colony’s success is inextricably linked to the condition of its queen, making accurate and timely assessment of queen health a cornerstone of effective hive management.

Challenges in Defining a Queenright Colony
Defining a “queenright” colony is more complex than it appears. Although commonly understood as a colony with an established queen, confirming this status often relies on invasive or subjective criteria (Caron & Connor, 2013; Tew, 2015). As defined scientifically, a truly queenright colony typically displays a consistent production of royal pheromones, cohesive and organized worker behavior, viable sperm stored in the queen’s spermatheca, the laying of both fertilized (worker) and unfertilized (drone) eggs, and the uninterrupted continuation of normal social activity within the hive. Because these indicators are often subtle, biochemical, or behavioral, they are difficult to confirm without opening the hive. Conventional inspection methods, while informative, may disrupt brood development or even jeopardize the queen through accidental injury or loss. These risks underscore the importance of developing non-invasive tools capable of reliably assessing queen presence and colony stability in the field.

For the purposes of this study and within our capabilities to measure, we have defined queenright as demonstrating a healthy colony, with a mated queen present and the presence of eggs and brood.

Rationale for Thermal Imaging
This study investigates whether thermal imaging can serve as a tool to measure queenright status in a colony. Honey bee colonies generate and regulate heat through metabolic activity, particularly in the brood nest. Thermal imaging has been used in other fields to detect physiological and environmental anomalies, and its use in apiculture could offer a non-invasive method of detecting shifts in colony structure and behavior, especially following queen loss. This research aims to determine whether consistent thermal differences can be identified between queenright and queenless colonies using FLIR camera technology in a real-world, multi-location study.

Materials and Methods
This study was conducted across six geographically diverse regions within the United States, with all participants beginning data collection on June 23, 2025. Each participant monitored three queenright colonies and maintained one undrawn-frame control hive, resulting in a total of 18 queenright colonies and 6 control hives.

Thermal imaging was performed using a combination of four FLIR C3-X and two FLIR One Edge cameras. To ensure thermal consistency, all devices were powered on at least 20 minutes before use to allow for calibration stabilization. Cameras were manually set to a standardized temperature range, and images were captured from both the front and back of each hive at a fixed distance of three feet.

Control hives were established with undrawn frames to serve as a broodless thermal baseline. All hives followed the Langstroth design, but construction materials varied, including pine or cedar hive bodies that were either painted, waxed, or left untreated. No supplemental insulation was added. One observer had hive bodies in their apiary that included styrofoam and plastic. Early evidence showed drastically different surface temperatures. The group rationalized that due to obvious insulation factor, and small representative sample number, these material designs were omitted from our study. Bottom boards were solid or screened, and lids were telescoping (with or without inner covers) or migratory, with variations recorded.

From June 23 to July 6, daily thermal images and weekly physical inspections confirmed queen presence. On July 7, one queen was intentionally removed from each participant’s apiary to simulate queen loss, generating six queenless colonies. Twenty-four hours post-removal, on July 8, thermal images were collected again to detect any immediate temperature changes. Imaging continued daily through July 25, following consistent local schedules.

Dr. Jerry Bromenshenk notes that infrared cameras measure only surface temperatures and do not directly capture internal cluster heat. He recommends using multiple imaging angles to minimize clustering bias. Despite limitations, high-quality, factory-calibrated cameras — such as the FLIR Exx series — paired with analytical tools like FLIR ResearchIR can yield quantitative insights into hive thermal behavior (Bromenshenk, 2016).

All thermal images and corresponding metadata — including ambient temperature, humidity, hive construction details, and site-specific variables — were logged in a shared Google spreadsheet. Image data were analyzed using FLIR ResearchIR software to quantify differences among control, queenright, and queenless colonies.

Results
A total of 24 hives were monitored across six U.S. locations between June 23 and July 25, 2025. Each participant tracked three initially queenright colonies and one control hive. On July 7, one colony per participant underwent queen removal. Thermal images and environmental metadata were collected consistently throughout the study period.

Delta Trends Over Time by Observer
To assess whether the thermal shift in queenless colonies occurred immediately after queen removal or evolved gradually, we plotted average delta (defined as Hive Thermal Average minus Control or Ambient) values over time for each observer, broken out by queen status. These time series graphs indicated that the change in delta was not immediate following queen removal. Instead, queenless hives typically showed either flat or decreasing delta trends over time. For instance, Jason’s queenless hives displayed a pronounced decline in delta values (slope = -0.419 Δ/day), while Annalie’s and Chris’ showed slight negative slopes (-0.021 and -0.063 Δ/day, respectively). In contrast, queenright hives tended to maintain stable or slightly increasing deltas over time. Overall, the time-series analysis suggests that if queenlessness impacts thermal dynamics, it does so in a gradual and variable manner rather than through an immediate shift.

Group Delta Distribution by Queen Status
To determine whether queen status has a measurable thermal signature across colonies, a two-sample t-test was conducted on the delta values for all observations. When aggregated across all six observers (Annalie, Chris, Jason, Jeremiah, Luke, and Scott), queenless hives showed a statistically significant difference in delta values compared to queenright hives (T = 2.30, p = 0.022). This result is visualized in the boxplot titled Group Delta Distribution by Queen Status, which reveals a higher median delta in queenless colonies. These findings suggest that, in aggregate, queenless hives tend to exhibit greater thermal deviation from ambient or control temperatures.

Per-Observer Delta Distributions
To evaluate consistency across field sites, delta values were analyzed for each observer using two-sample t-tests and visualized using individual boxplots grouped by queen status. Of the six participants, only one observer’s dataset showed a statistically significant difference between queenless and queenright hives (p = 0.012). All other observers had p-values greater than 0.19, indicating no significant differences. Despite some variation in median and range across boxplots, the lack of consistent statistical significance at the individual level underscores substantial environmental and methodological variability between sites. This suggests that while group-level trends are detectable, local field conditions may obscure the thermal effects of queenlessness when analyzed in isolation.

External Thermal Trends in Relation to Ambient Temperature
To evaluate whether hive condition influences thermal regulation, we plotted the relationship between ambient temperature and the external surface temperature of hives across all participating study sites. The resulting scatterplot (Figure X) distinguishes data points by hive type — queenright, queenless, and control — using color-coded markers. This visual representation helps clarify thermal behavior under varying environmental conditions and colony statuses.

The chart reveals a clear positive correlation between ambient and external hive temperatures across all hive types. Notably, queenright colonies consistently maintain higher external temperatures than both queenless and control hives, especially at moderate ambient temperatures (70–85°F). This suggests that queenright colonies exhibit active thermoregulation, likely driven by brood-rearing and internal colony activity. Queenless colonies, while showing some thermogenic behavior, generally fall in an intermediate range, indicating reduced metabolic output in the absence of a queen. In contrast, control hives (which contained no bees) show external temperatures closely mirroring ambient conditions, underscoring their lack of internal heat generation. These distinctions in thermal response provide compelling evidence that FLIR thermal imaging may serve as a viable, non-invasive method to infer queenright status in managed colonies.

Hive Thermal Averages by Queen Status
Analysis of thermal delta values across all participating observers revealed a statistically significant difference between queenright and queenless hives (p = 0.022). On average, queenless hives exhibited higher delta values, suggesting greater thermal variation or loss. However, this trend was not consistent across individual observers; only one participant’s hive showed a statistically significant difference at the individual level. This indicates that while queenlessness may be associated with increased thermal delta on a broad scale, local variability — such as environmental conditions, hive construction, or measurement timing — limits the reliability of using thermal delta alone to determine queen status at the individual apiary level.

While overall trends support the hypothesis that queenright hives exhibit distinct thermal signatures, particularly in stability and clustering, the study did not find statistically significant differences (p < 0.05) in delta or average temperature values within the short 2–3 week post-removal window. However, visual trends suggest promise for thermal imaging as a non-invasive tool when combined with stricter controls on environmental variables and longer observation periods.

Discussion Defining ‘Queenright’ for Beekeepers
Despite the importance of queen status in colony health, the study team was unable to find any specific studies that clearly define what scientifically constitutes “queenright” in a hive, especially as it pertains to the thermal assessment of a queenright colony. Most references rely on indirect indicators — such as brood presence or pheromone output—that are either subjective or difficult to assess without invasive inspection.

Thermal Imaging Practicality and Hive Material Considerations
The field application of thermal imaging is subject to several practical variables, particularly hive construction materials. As noted by Bromenshenk (2015, 2016), surface emissivity differs among hive types — painted versus unpainted wood, foam, plastic, and metal components each reflect and retain heat differently. These variations can significantly impact the accuracy of thermal readings, especially when comparing colonies housed in non-uniform structures. For consistent interpretation, either standardization of hive materials or correction for emissivity differences is essential. It is important to note that hive components — including bodies, covers, and bottom boards — are manufactured from a wide range of materials, which our study did not attempt to control or categorize. Future research should evaluate how thermal imaging behaves across hive types such as styrofoam, plastic, or insulated designs, to better understand how structural material influences heat signatures.

In addition, Bromenshenk emphasizes the importance of multi-angle image capture, consistent camera calibration, and standardized use of analytical software (e.g., FLIR ResearchIR) to minimize error and maximize utility. These insights are critical for translating thermal imaging from controlled experimental use to real-world beekeeping practices, particularly in large-scale or remote operations where non-invasive queen and brood assessment may offer major time and labor savings (Bromenshenk, 2016).

Interpreting Thermal Patterns Through Brood Physiology
The relevance of brood-stage thermoregulation to colony-level thermal signatures is supported by the findings of Debnam et al. (2025), who demonstrated that honey bee eggs, larvae, and pupae are each maintained at slightly different yet consistently elevated temperatures relative to the surrounding brood nest. Their study highlights that nurse bees and heater bees play a critical role in regulating these microclimates, adjusting their behavior to sustain stage-specific thermal needs. This precision thermoregulation suggests that any disruption in brood development — such as that caused by queen loss and the cessation of egg-laying — could alter the hive’s thermal profile in detectable ways. While Debnam et al. used modified hives and internal imaging to assess brood-stage temperatures, their findings provide foundational support for our study’s external thermal imaging approach. The distinct thermal patterns generated by developing brood lend credence to the hypothesis that colony-level thermal shifts may occur as early as 24 hours after queen removal, reflecting both the absence of new brood and changes in nurse bee activity.

Limitations and Uncontrolled Variables
Despite efforts to standardize equipment settings and imaging protocols, several uncontrolled variables may have influenced the consistency and comparability of thermal data across sites.

First, environmental interference introduced inconsistency in image capture. Grass, weeds, or debris in front of hives may have artificially altered surface readings by reflecting or absorbing heat, especially in close-range imaging. Similarly, light exposure varied significantly, with sunlight versus shade producing drastic differences in observed surface temperatures. The orientation of hive entrances relative to the sun’s path (east–west vs. north–south) was not standardized, potentially affecting heat retention and external surface values.

Second, the hive design and imaging scope may have introduced variability. Supers, if included in the thermal frame but not fully built out or occupied by brood, could have artificially lowered thermal averages. Similarly, inclusion or exclusion of hive entrances and landing boards may have altered readings, particularly when guard bee clustering or bearding was present. These external bee activities often occur during heat stress or dusk and may not have been noticed during low-light photography.

Lastly, image acquisition timing likely impacted thermal consistency. Some hives were photographed close to sunset, when wooden surfaces may have retained and radiated heat absorbed during the day. In contrast, pre-dawn imaging would reduce solar heat interference and provide a more accurate representation of colony-generated warmth.

This recommendation is supported by Shaw et al. (2010), who demonstrated that radiometrically calibrated long-wave infrared imaging can reliably estimate colony population by correlating surface radiance with the number of occupied brood frames. Their findings identified pre-dawn imaging as the most accurate condition for non-invasive hive assessment, due to the increased contrast between internal hive heat and cooler ambient conditions. The authors concluded that such techniques could reduce colony disruption while maintaining diagnostic reliability in population monitoring (Shaw et al., 2010).

Together, these findings and limitations emphasize the need for standardized imaging protocols that consider hive materials, image timing, and environmental variables to improve the precision of thermal diagnostics in field settings.

Conclusion & Benefits of Our Research
This study evaluated whether thermal imaging using FLIR technology can serve as a reliable, non-invasive tool for detecting queenright status in honey bee colonies in a real-world beekeeping setting. At the group level, queenless hives exhibited measurably greater thermal deviation from ambient or control values compared to queenright colonies, supporting the hypothesis that queen presence influences internal hive temperature stability. Time-series analysis revealed that these thermal differences emerged gradually rather than immediately following queen removal, suggesting a delayed physiological or behavioral shift within affected colonies.

Although group-level trends reached statistical significance (p = 0.022), only one observer’s dataset showed significant differences at the individual level. This inconsistency highlights the influence of environmental variation, hive construction, and imaging conditions on thermal readings. Control hives, which lacked bee populations, closely tracked ambient temperatures, validating the sensitivity of FLIR devices in detecting biologically generated heat. However, the limited post-removal observation period and uncontrolled site-level variables constrained the strength of individual-level inferences.

Our findings are consistent with laboratory work by Grodzicki et al. (2020), who demonstrated that queenless groups of workers exhibit increased locomotor activity and disrupted thermal preferences. While their research was limited to small, controlled environments, our results extend these insights to full colonies in field conditions, suggesting that queen absence may alter collective thermoregulation. Similarly, Delaplane and Harbo (1987) found that queen loss leads to longer-term declines in worker survival, colony defense, and honey gain. Although our study focused on early thermal indicators rather than productivity outcomes, the observed thermal disruption may represent a physiological precursor to the broader colony-level deterioration they described.

Taken together, these findings indicate that thermal imaging has potential as a minimally disruptive diagnostic tool for detecting queen status. If optimized, this method could allow for earlier identification of queen failure or supersedure without opening the hive, thereby reducing colony disturbance and beekeeper labor. As queen loss remains a leading contributor to colony collapse, tools that enable passive and early detection offer a major step forward for both commercial and small-scale apiculture. Our results suggest that thermal imaging may not only assist in diagnostics but also support timely requeening decisions, potentially improving overwintering success and overall colony resilience. Having learned from this study, we now know that if we were to conduct it again, we should incorporate longer monitoring windows, standardized hive materials, and early-morning imaging protocols to reduce environmental interference and strengthen the utility of thermal data in field diagnosis.

To summarize, the study group currently believes that FLIR imaging may only be successful if performed by an advanced-level beekeeper, and needs to be implemented with a dedicated control hive, uniform materials of hive bodies, regimented, expressed pre-dawn image acquisition, and uniform positioning of their hives directionally to the sun. This leads to our conclusion that we would not recommend FLIR imaging to assess for queenright colonies for the average beekeeper at this time.

References
Caron, D. M., & Connor, L. J. (2013). Honey bee biology and beekeeping (Rev. ed.). Wicwas Press.
Tew, J. E. (2015). The Beekeeper’s Problem Solver: 100 Common Problems Explored and Explained. Quarry Books.
Shaw, J. A., Nugent, P. W., Johnson, J., Bromenshenk, J. J., Henderson, C. B., & Debnam, S. (2010). Longwave infrared imaging for noninvasive assessment of honey bee colony population. Optics Express, 19(1), 399–408. https://doi.org/10.1364/OE.19.000399
Bromenshenk, J. J. (2016, May 20). Professional IR Cameras. Bee Culture. https://beeculture.com/professional-ir-cameras-2/
Bromenshenk, J. J. (2015, December 21). Infrared: The Next Generation in Colony Management. Bee Culture. https://beeculture.com/infrared-the-next-generation-in-colony-management/
Delaplane, K. S., & Harbo, J. R. (1987). Effect of queenlessness on worker survival, honey gain, and defense behavior in honeybees. Journal of Apicultural Research, 26(1), 37–42. https://bees.caes.uga.edu/content/dam/caes-subsite/bee-program/images/research-archives/research-archives/DelaplaneandHarboqueenlessness1987.pdf
Grodzicki, P., Piechowicz, B., & Caputa, M. (2020). The effect of the queen’s presence on thermal behavior and locomotor activity of small groups of worker honey bees. Insects, 11(8), 464. https://doi.org/10.3390/insects11080464
Debnam, S. E., McCormick, M. B., Seibold, C., Callaway, R. M., & Woods, H. A. (2020). Honey bee eggs, larvae, pupating juveniles, and pupae develop at slightly different temperatures and are all warmer than the brood nest. Journal of Thermal Biology, 94, 102759. https://doi.org/10.1016/j.jtherbio.2020.102759

Appendices
LINKS TO GDRIVE IMAGES AND CHARTS

Christine Crick Giltner is a full-time entomology student at the University of Idaho, specializing in melittology and the primary author of a six-member University of Montana Master Beekeeper research team. Her work focuses on organic beekeeping standards, swarm derived genetics, and advancing naturally resilient European honey bee populations.

Like this article? Purchase a copy of this issue here.

Subscribe to Bee Culture Magazine here.