News from
Office of Communications
22 Chambers St., Suite 201
Princeton, New Jersey 08544-5264
Telephone 609-258-3601; Fax 609-258-1301

For immediate release: December 9, 2002

Contact: Steven Schultz, (609) 258-5729, sschultz@princeton.edu

Stung by success: Intensive farming may suppress pollinating bees
Study shows native bee species provide valuable services when allowed to flourish

PRINCETON, N.J. -- Intensive, industrial-scale farming may be damaging one of the very natural resources that successful crops require: pollinating bees. A study by Princeton scientists found that native bee populations decline dramatically as agricultural intensity goes up.

In farms studied in and around the Sacramento Valley in California, concentrated farming appeared to reduce bee populations by eliminating natural habitats and poisoning them with pesticides, the researchers reported.

U.S. farmers may not have noticed this effect because historically they have achieved their harvests with the help of imported bees rented from beekeepers. These rented bees, however, are in decline because of disease and heavy pesticide use.

The study, to be published this week in an online edition of the Proceedings of the National Academy of Sciences, found that native bees are capable of doing a lot more pollinating than previously thought. But it would take careful land use to take advantage of that capacity, the researchers concluded, because current high-density, pesticide-dependent agriculture cannot support native bees.

"This is a valuable service that we may actually be destroying through our own land management practices," said Princeton ecologist Claire Kremen, who co-wrote the study with Neal Williams, a postdoctoral researcher, and Robbin Thorp of the University of California-Davis.

Suppressing the many species of native bees and relying on just a few species of imported ones may be unnecessarily risky, said Kremen. Farmers who use managed bee populations -- that is, most commercial farmers -- depend on fewer than 11 species out of the 20,000 to 30,000 bee species worldwide. Other researchers have estimated that $5 billion to $14 billion worth of U.S. crops are pollinated by a single species of bee, the European honey bee Apis mellifera.

"Right now we are really very dependent on that species," said Kremen. "If something happened to that species and we haven't developed other avenues, we could really be in great difficulty."

The researchers spent two years examining watermelon farms located at varying distances from oak woodlands and chaparral habitats that are native to the Sacramento Valley. They also looked at land that was farmed conventionally (with pesticides) and organically (without pesticides). They focused on watermelon because it requires a lot of pollen and multiple bee visits to produce marketable fruit.

The research required painstaking work. Kremen and Williams first put fine mesh bags on watermelon flower buds, so that when the flowers opened they had no pollen. They then removed the bags, put the freshly opened flowers on the ends of sticks and presented them in front of bees to tempt them to land. For each of about 20 species of native bees that frequented the flowers, they determined the median number of pollen grains deposited in each visit.

Then, in each of their selected locations, the researchers watched watermelon flowers over long periods and recorded how many of each kind of bee visited. They found that native bee visits dropped off dramatically in the farms that were distant from natural habitats and that used pesticides. "We could then multiply the number of visits by the number of grains deposited per visit and sum that up for all the species and figure out how much pollen the watermelon plants were receiving," said Kremen.

"We found that, where it still flourished, the native bee community could be sufficient to provide the pollination service for the watermelon," Kremen said, adding that the result is likely to apply to a variety of other species. Farmers began renting bees many years ago to improve yields and became dependent on them as the size and concentration of farms increased. Typically, farmers whose lands are located near natural habitat don’t bother to rent bees, presumably because they receive sufficient pollination from the natural community, said Kremen.

One interesting finding, said Kremen, was that the mix of native bees providing the pollination was very different in the two years of the study. In one year, a few strong pollinators accounted for most of it, while in the other, many species contributed.

"That says something about the need for long-term studies and also argues for the need to maintain diversity," said Kremen.

The research fits into a broader question that Kremen and others are studying regarding the relation between biodiversity and what ecologists call "ecosystem services," the economic benefits that natural systems provide to people but that are not normally accounted for in the marketplace. Scientists need carefully collected data to quantify the value of biodiversity, Kremen said.

Kremen is now working on follow-up studies to determine what parts of the natural landscape are critical for native bees and what parts of the man-made agricultural landscape also may support native bees.

"Ultimately, we should be able to come up with a plan for restoring this natural service across the agro-natural landscape," she said.