Citizen scientists help track bee populations.

Stories about the honeybee crisis and colony collapse disorder (CCD) keep turning up in the news (at least here in California, where we grow so many big cash crops like almonds that rely on honeybees to pollinate them). But it turns out that getting to the bottom of CCD is made more difficult by the the gaps in biologists' knowledge about the wild bee populations. (A lot of the bees pollinating food crops are commercially kept rather than wild.)

But, as reported in an article in the September-October 2008 issue of American Scientist [1], the Great Sunflower Project is enlisting the efforts of citizen scientists to fill in some of those gaps.

Conservation biologist Gretchen LeBuhn and colleagues at San Francisco State University launched the project last spring. In it, volunteers plant sunflowers and then regularly monitor the blooms for bees, which they report to the researchers.

The trick, it seems, was working out what sorts of observations would be informative and practical to get from the non-scientist volunteers (who, after all, probably have other things they need to do besides sitting amongst the sunflowers all day waiting for bees):

Before beginning the project, LeBuhn and her team did trial runs to find out what observations people could perform reliably over several months. They first had volunteers sit for 30 minutes and count the bees that came to their sunflowers. But this was frustrating all around: When lots of bees came at once, they were too difficult to count. And when none came in half an hour, LeBuhn says, "people felt like they had let me down." She had to figure out "how to communicate that they hadn't failed" -- a quandary the biologist in the field is unlikely to encounter.

After all, observing an absence of bees in a particular interval of time is still finding out something important about the system you're trying to understand.

LeBuhn decided to have participants measure how long it took for five bees to visit their sunflowers, observing two times each month for a maximum of 30 minutes each time. "People can take their cup of coffee out in the morning," she says, "and if they see five bees in five minutes, they're done. If they see none in 30 minutes, they're still done, and they have the most important data" because the lack of bees indicates spots where populations may be in trouble.

Of course, I'm wondering about all the variables within the two observation periods each month that might affect the number of bees visiting the sunflowers, from temperature to precipitation to the amount of sunlight. Do bees have a time of day when they're more likely to be out carousing with sunflowers? Is that daily rhythm more likely to coincide with a morning cup of coffee or a twilight glass of wine?

I suspect the volunteers are noting the date and time of their observations, in case the researchers need to take such factors into account.

In addition to tracking the time to observe five bees (or the lesser number that may show up by the 30 minute cutoff), the citizen scientists also report what kind of bees they've observed, classifying them as honeybees, bumblebees, carpenter bees, green bees, or "other". They are given key traits that should help them distinguish between these different kinds of bees, and they are also asked whether they are confident of their classification of the bees they've observed.

If confident and less-confident participants return similar results, the data even from those with low confidence are more likely to be good.

To be safe, LeBuhn and colleagues will also be using statistical models to identify less reliable data.

The design of this project, and the involvement of citizen scientist-participants across the lower 48 states, is pretty cool. But what does all of this have to do with CCD, which is, after all, a problem that primarily affects the hives of commercial honeybee keepers? In part, the data may give an indication of "whether native bees are filling in as 'insurance policies' in areas where honeybees are gone." Besides, tracking the populations of those other bees is important in itself -- they may still fill an important ecological niche as pollinators, even if their vomit is not as tasty.

_______
[1] Anna Lena Phillips, "Of Sunflowers and Citizens," American Scientist (September-October 2008) Vol. 96, No. 5, 375-376.

More like this

Thanks for posting this! I've signed up to participate in the project next year.

I haven't read the article but apart from time of day, temperature and other weather conditions and location, I can immediately think of one huge confounding factor that I sincerely hope she's taking into account: What other sources of nectar and pollen are near each location? Honeybees memorise locations of food sources and communicate them to their nestmates, which means that if there's a large, good source such as a rapeseed field, that's where the bees are going to be. A lack of bees at sunflowers nearby doesn't mean there are no bees, it just means they couldn't care less about an appetizer when they have a whole buffet waiting round the corner. This should pertain less to solitary bees which have no friends to tell them where the good stuff is, but once they found it they won't be seeking out the sunflowers on the next trip, so it'll still be a factor.

This is a truly great thing. One of the nice features of (much) nineteenth century science was that it went out of its way to involve ordinary citizens in the actual scientific process, drawing on them to discover weather patterns, new species, etc. (Herschel, I think, discusses and advocates this somewhere in A Preliminary Discourse on the Study of Natural Philosophy.) Science was often seen not as something done by isolated scientists but by the whole community. It's nice that there is still research being done on this model, even though (as you suggest) it can be tricky to coordinate given all the variables that can affect observations.