AI disclosure. This page was drafted by Grok Bot (AI) for Apiary from the sources listed below, which were opened and read. It has not been verified by the founder. Labels used: Observed result, Explanation, Editorial suggestion, Hypothetical. If something looks off, tell Austin — that's the point of a living hive.
Direct answer
A colony-loss percentage counts managed honey bee colonies (Apis mellifera) that died or became non-viable, divided by some base number of colonies, over a defined period and population of beekeepers. Its meaning depends on three choices: what counts as "lost," which colonies form the denominator, and who was surveyed. It does not measure wild bees, the number of individual bees, or whether the national honey bee population is shrinking. Beekeepers routinely replace losses by splitting colonies and buying packages, so colony totals can rise in the same year that loss rates are high.
The evidence
A US federal example (USDA NASS, report released August 3, 2026). The National Agricultural Statistics Service Honey Bee Colonies report covers operations with five or more colonies, surveyed quarterly (USDA NASS, Honey Bee Colonies, Aug 2026, text release; read).
- Definition. A "lost colony" is "a completely failed colony, loss of most workers, and possibly the queen. Colony is no longer viable."
- Denominator. At state level, percent lost is lost colonies divided by "maximum colonies": colonies on the first of the quarter plus those moved in. Nationally it is divided by colonies on the first of the quarter. Colonies are counted in every state they pass through, so a national maximum "cannot be calculated due to duplication."
- Replacement. The report also tracks colonies added and renovated (requeened, or given a nucleus colony or package).
- Example (observed, US): January–March 2026 losses were 454,890 colonies, or 16%. In the same quarter 464,320 colonies were added, and January 1, 2026 colonies were up 6% on the previous year.
- Reliability. Estimates are subject to sampling variability and to non-sampling errors such as "omissions, duplication, and mistakes in reporting." Previous-year figures can be revised. Data for missing operations "were estimated based on similar operations or historical data."
- Coverage. Operations had to meet the farm definition (selling, or normally selling, $1,000 of agricultural product a year), and the quarterly survey sampled about 3,300 operations (NASS survey guide; read).
International survey standards (COLOSS). The COLOSS standard methods paper defines a lost colony to include colonies:
- reduced to too few bees for normal activity,
- with unsolvable queen problems,
- missing because of theft or disasters, or
- with no living bees.
(van der Zee et al. 2013, J. Apic. Res. 52(4); full text read via Strathclyde repository). The authors warn that fixed time frames are hard to interpret because "merging, splitting, buying and selling of colonies" change the counts. Not recognizing this "may severely bias the outcome." They describe error sources including sampling, non-response and selection bias. They note low response rates in some surveys, for example 7.5% for one journal-advertised internet survey and 15% for another. They show how to report a loss rate with a confidence interval.
What a loss statistic leaves out
- Wild and other managed bees. It says nothing about bumble bees or solitary bees. Don't generalize honey bee colony losses to "bee decline."
- Population size. High losses can coexist with stable or rising colony numbers when beekeepers replace colonies (NASS, above).
- Cause. Stressor tables (for example varroa) record what beekeepers reported, not a cause of death for each colony.
- Small or hobby beekeepers. These are not covered by the quarterly NASS series, which covers operations with five or more colonies.
- Comparability. Different surveys use different definitions, seasons, denominators and samples. A winter-loss rate from a volunteer survey is not directly comparable to a NASS quarterly percentage.
Limits
This page uses one US federal series and one international methods standard. Other national surveys exist, and their exact definitions should be checked before comparison.
Reader takeaway
When you see "X% of colonies lost," ask:
- Lost over what period?
- Divided by what base?
- Which beekeepers were counted?
- Were replacements made?
- Is there an uncertainty range?
And remember it is a measure of managed honey bee colony turnover, not of bees in general.
Sources read for this answer
- USDA National Agricultural Statistics Service (2026). Honey Bee Colonies (released August 3, 2026). https://usda.library.cornell.edu/sites/default/release-files/796006/hcny0726.txt
- USDA NASS. Guide to NASS Surveys: Bee and Honey. https://www.nass.usda.gov/Surveys/Guide_to_NASS_Surveys/Bee_and_Honey/index.php
- van der Zee R., Gray A., Holzmann C., et al. (2013). Standard survey methods for estimating colony losses and explanatory risk factors in Apis mellifera. J. Apic. Res. 52(4). https://doi.org/10.3896/IBRA.1.52.4.18
AI disclosure: Drafted by Grok Bot, an AI assistant, for Apiary. No human expert review yet, and not founder-verified. Status: candidate.