Review: vanEngelsdorp et al. (2008). PLoS ONE, 3: e4071.

Written: February 1, 2009

Posted: 02/02/09

Word count: 724

Question: Are there any patterns in the recent bee die-offs?

Answer: Some, but they are complicated, which is what makes finding the cause(s) so difficult

It is often said that there is a special place in hell reserved for three groups of dishonest people: liars, damned liars, and statisticians. While certainly true for the first two, and those who employ statistics have been known—on occasion—to fudge the numbers, it gives number crunchers and other bean counters a pretty unfair reputation. Statistics, in its most basic form, is the application of mathematics and probability theory to make sense of raw data so that we can make understand it all. Without it, much of what we do in society would be practically baseless.

Historically, it has been exceptionally difficult to get any meaningful statistics about beekeeping. Oh sure, we have very good numbers on how many acres of upland cotton is planted in Georgia, the average yield of apples per acre in upstate New York, and how many head of cattle there are in western Texas. But how many beehives are there in North Carolina? No idea. The “official” statistics, collected and reported by the National Agricultural Statistics Service (NASS), says that there are 15,000 managed honey bee colonies in the state. I can account for that many by thinking of only a few dozen large-scale beekeepers, so these numbers are way off (we often quote roughly 100,000 colonies in the state, but that is more a questimate than a statistic). Nationally, this huge shortcoming of understanding our industry has long been highlighted as an impediment to helping solve many of the large-scale problems that have befallen the apiculture industry.

It is for this reason that when a study compiles some new statistics, particularly those with much more realistic numbers than the NASS reports, we can really hope to identify patterns that might help answer some of the bigger questions that we are currently asking. This is what a new study, lead by Dennis vanEngelsdorp of the Pennsylvania Department of Agriculture, has recently accomplished. They surveyed beekeepers all across the country, asking them detailed questions about many things but centered around winter mortality (over the 2007-2008 season). In doing so, they got information on 324,571 (13.3%) of the 2.44 million beehives in the nation, therefore the patterns that they found are very insightful indeed.

They found several important patterns in their statistics. First, there was a total loss of 35.8% of colonies, which is an increase of 11.4% compared to year prior. This is indeed a stunning figure: could you imagine if more than a third of Holstein cattle died every year without any explanation?! Second, sixty percent of all colonies that were reported dead in their survey died without a noticeable number of dead bees in or around the hive. Since this is the hallmark symptom of Colony Collapse Disorder, this opens the possibly that many (even a majority) of these beehives suffered from CCD. Third, large-scale operations (those with over 50 colonies) were more likely to have this symptom, suggesting that a contagious condition and/or different management practices may be causal factors. Fourth, when the beekeepers were asked what they suspected for their mortality, they responded that poor queens (31%), starvation (28%), varroa mites (24%), CCD (9%), and poor weather (9%) were to blame in that ranked order. Finally, within the state of Pennsylvania (where some of the best numbers are known because of mandatory registration of beehives), losses decreased in warmer regions in the state, indicating that ambient temperature over winter may be an important factor.

Even with these important statistics, they were still not able to detect any strong pattern as to why so many colonies are dying. In essence, there are many, many different reasons for colony mortality, which makes identifying the underlying cause much more difficult. Nonetheless, this study will help serve as a baseline for future comparisons and help guide future research into more mechanistic studies of mortality. Clearly, exceptions need to be made for the heavenly fate of some statisticians.

Reference

vanEngelsdorp, D., J. Hayes, R. M. Underwood, and J. Pettis. (2008). A Survey of honey bee colony losses in the U.S., fall 2007 to spring 2008. PLoS ONE, 3: e4071. doi:10.1371/journal.pone.0004071