As stubborn facts ruin their narrative that neonicotinoid pesticides are causing a honeybee-pocalypse, environmental pressure groups are shifting to new scares to justify their demands for “neonic” bans.
Honeybee populations and colony numbers in the United States, Canada, Europe, Australia and elsewhere are growing. It is also becoming increasingly clear that the actual cause of bee die-offs and “colony collapse disorders” is not neonics, but a toxic mix of predatory mites, stomach fungi, other microscopic pests, and assorted chemicals employed by beekeepers trying to control the beehive infestations.
Naturally, anti-pesticide activists have seized on a recent study purporting to show that wild bee deaths in Britain have been correlated with neonic use in oil seed rape fields (canola is a type of OSR). In a saga that has become all too common in the environmental arena, their claims were amplified by news media outlets that share many activist beliefs and biases – and want to sell more subscriptions and advertising.
(Honeybees represent a small number of species that humans have domesticated and keep in hives, to produce honey and pollinate crops. Many are repeatedly trucked long distances, to pollinate almond and other crops as they flower. By contrast, thousands of species of native or wild bees also flourish across the continents, pollinating plants with no human assistance.)
The recent Center for Ecology and Hydrology study examined wild bee population trends over an 18-year period that ended in 2011. It concluded that there was a strong correlation between population and distribution numbers for multiple species of British wild bees and what study authors called their “measure of neonic dose” resulting from the pesticide, which is used as a seed coating for canola crops.
The study is deeply flawed, at every stage – making its analysis and conclusions meaningless. For example, bee data were collected by amateur volunteers, few of whom were likely able to distinguish among some 250 species of UK wild bees. But if even one bee of any species was identified in a 1-by-1 kilometer area during at least two of the study period’s 18 years, the area was included in the CEH study.
This patchy, inconsistent approach means the database that formed the very foundation for the entire study was neither systematic nor reliable, nor scientific. Some species may have dwindled or disappeared in certain areas due to natural causes, or volunteers may simply have missed them. We can never know.
There is no evidence that the CEH authors ever actually measured neonic levels on bees or in pollen collected from OSR fields that the British wild bees could theoretically have visited. Equally relevant, by the time neonics on seeds are absorbed into growing plant tissue, and finally expressed on flecks of pollen, the levels are extremely low: 1.3–3.0 parts per billion, the equivalent of 1–3 seconds in 33 years.
(Coating seeds ensures that pesticides are incorporated directly into plant tissue – and target only harmful pests that feed on the crops. It reduces or eliminates the need to spray crops, which can kill birds, bats and beneficial insects that are in the fields or impacted by accidental “over-sprays.” Indeed, numerous field studies on two continents have found no adverse effects from neonics on honeybees at the hive level.)
A preliminary U.S. Environmental Protection Agency risk assessment for one common neonic sets the safe level for residues on pollen at 25 ppb. Any observable effects on honeybee colonies are unlikely below that. Perhaps wild bees are more susceptible. However, at least two wild bee species (alfalfa leaf cutters and miner bees) are thriving in areas where OSR/canola fields are widespread, and the CEH study found reduced numbers of certain wild bees that do not collect pollen from oil seed rape.
Perhaps most important, the CEH authors appear to have assumed that any declines in wild bee numbers were due to neonicotinoid pesticides in OSR fields, even at very low doses. They discounted or ignored other factors, such as bee diseases, weather and land use changes.
For instance, scientists now know that parasitic Varroa destructor mites and phorid flies severely affect honeybees; so do the Nosema ceranae gut fungus, tobacco ringspot virus and deformed wing virus. Under certain circumstances, those diseases are known to spread to bumblebees and other wild bees.
Significant land development and habitat losses occurred in many parts of Britain from 1930 to 1990, causing wild bee populations todecline dramatically. Thankfully, they have since rebounded – during the same period that neonic use was rising rapidly, replacing older insecticides that clearly are toxic to bees! The CEH team also failed to address those facts.
To compensate for these shortcomings (or perhaps to mask them), the CEH researchers created a sophisticated computer model that supposedly describes and explains the 18 years of wild bee data.
However, as any statistician or modeler knows, models and output are only as good as the assumptions behind them and data fed into them. Garbage in/Garbage out (GIGO) remains the fundamental rule. Greater sophistication simply means more refined refuse, and faster computers simply generate faulty, misleading results more rapidly.
The CEH models are essentially “black boxes.” Key components of their analytical methodologies and algorithms have not been made public and thus cannot be verified by independent reviewers.
However, the flawed data gathering, unjustified assumptions about neonic impacts, and failure to consider the likely effects of multiple bee diseases and parasites make it clear that the CEH model and conclusions are essentially worthless – and should not be used to drive or justify pesticide policies and regulations.
As Prime Minister Jim Hacker quipped in the theatrical version of the British comedy series Yes, Prime Minister: “Computer models are no different from fashion models. They’re seductive, unreliable, easily corrupted, and they lead sensible people to make fools of themselves.”
And yet studies like this constantly make headlines. That’s hardly surprising. Anti-pesticide campaigners have enormous funding and marvelous PR instincts. Researchers know their influence and next grant can depend on issuing studies that garner alarmist headlines and reflect prevailing news themes and imminent government actions. The news media want to sell ads and papers, and help drive public policy-making.
The bottom line is fundamental: correlation does not equal causation. Traffic lights are present at many intersections where accidents occur; but that does not mean the lights caused most or all of the accidents. The CEH authors simply do not demonstrate that a neonic-wild bee cause-effect relationship exists.
The price to society includes not just the countless dollars invested in useless research, but tens of billions in costs inflicted by laws and regulations based on or justified by that research. Above all, it can lead to “cures” that are worse than the alleged diseases: in this case, neonic bans would cause major crop losses and force growers to resort to older pesticides that clearly are harmful to bees.
There is yet another reason why anti-pesticide forces are focusing now on wild bees. In sharp contrast to the situation with honeybees, where we have extensive data and centuries of beekeeper experience, we know very little about the thousands of wild bee species: where they live and forage, what risks they face, even how many there really are. That makes them a perfect poster child for anti-neonic activists.
They can present all kinds of apocalyptic scenarios, knowing even far-fetched claims cannot be disproven easily, certainly not in time to address new public unease amid discussions about a regulatory proposal.
The Center for Ecology and Hydrology study involved seriously defective data gathering and analytical methodologies. More troubling, it appears to have been released in a time and manner calculated to influence a European Union decision on whether to continue or rescind a ban on neonicotinoid pesticides.
Sloppy or junk science is bad enough in and of itself. To use it deliberately, to pressure lawmakers or regulators to issue cures that may be worse than alleged diseases, is an intolerable travesty.