Decreases in bird numbers are most rapid in areas that are most heavily polluted with neonicotinoids, suggesting that the environmental damage inflicted by these insecticides may be much broader than previously thought. See Letter p.341
The debate over the environmental risks posed by neonicotinoid insecticides has raged since the late 1990s, when French beekeepers began blaming the chemicals for losses of honeybee colonies. The discussion has focused closely on bees, particularly the risks posed by the use of neonicotinoid treatments on flowering crops that bees visit. But on page 341 of this issue, Hallmann et al.1 provide strong evidence that this debate may have missed the bigger picture. Analysing long-term data sets on bird populations in the Netherlands, the authors demonstrate that regional patterns of population decline in insect-eating birds are neatly predicted by levels of neonicotinoids detected in environmental samples. In other words, birds have declined faster in places with more neonicotinoid pollution.
Dozens of papers have been published on the effects of neonicotinoids on bees and, following a review of the evidence, the European Food Safety Authority declared in 2013 that neonicotinoids posed an “unacceptable risk” to the insects. Shortly afterwards, the European Union voted in favour of a two-year moratorium on the use of three widely used neonicotinoids on flowering crops. It has already been suggested that the impacts of these chemicals are likely to extend far beyond bees2, but Hallmann and colleagues' study is the first to provide direct evidence that the widespread depletion of insect populations by neonicotinoids has knock-on effects on vertebrates.
Neonicotinoids are neurotoxins that are exceptionally toxic to insects but much less so to birds3. Because of this, the observed bird declines are unlikely to be due to direct toxicity. As Hallmann et al. argue, it is much more plausible that the effects are the result of a depletion of the birds' food — insects. However, it is worth noting that none of the bird species studied would ordinarily eat bees in any quantity.
Hallmann and colleagues essentially infer cause and effect from correlation, but this is made more convincing because they consider a range of other measures of land use that are known to affect bird and insect populations, but found none that predicted bird declines as powerfully as environmental neonicotinoid concentration. Of course, an experimental, manipulative approach to test cause and effect would be more compelling, but that would be almost impossible on a realistic scale, with replication, in organisms as highly mobile as birds, and in any case would face severe ethical issues.
How might neonicotinoids, most of which are applied as seed dressings to arable crops, come to have such widespread impacts on the environment? The insecticides' intended mechanism of action is that the dressing should dissolve around the seed, be absorbed by the growing seedling and spread through its tissues, protecting all parts of the crop from herbivorous insects. However, only approximately 5% of the active ingredient is taken up by the crop4 (Fig. 1). A little is lost as toxic dust that blows away and may affect flying insects or be deposited on non-target vegetation5, but most enters the soil and soil water. The half-life of neonicotinoids in soil varies with soil type, but can exceed 1,000 days, such that they can accumulate over time. The consequences of this accumulation for soil fauna and soil health are poorly understood2.
The chemicals can also be washed from soils into waterways, where they are likely to affect aquatic insects6, which are key sources of food for both birds and fish. And they can be taken up by the roots of hedgerow plants, where they will have the same systemic action as in crops, spreading through the leaves and flowers. Non-target herbivorous insects such as grasshoppers, beetles, shield bugs and the caterpillars of butterflies, moths and sawflies will all be exposed through this route, and these form the food supply for a broad range of predatory insects, birds and some mammals, such as shrews and bats.
The persistent nature of neonicotinoids and their high solubility in water mean that such broad contamination is also probable with other methods of application, such as foliar sprays or soil drenches. Given these manifold routes of spread, it is perhaps not surprising that, after 20 years of steadily increasing use, there is now evidence that neonicotinoids are having broad effects through the food chain — as shown by Hallmann et al. and by a recent meta-analysis7 of studies on the ecosystem effects of systemic pesticides.
The European two-year moratorium came into effect in December 2013, but it is designed to protect bees from exposure only to mass-flowering crops. As such, neonicotinoids are still used as seed dressings on other major crops, such as wheat and barley, and they are still widely sprayed in horticulture and sold for use in gardens and public areas. Hence, impacts on birds and other insectivores might be expected to continue. Elsewhere in the world, the emerging evidence for environmental harm has not yet resulted in any new restrictions on their use.
The story is reminiscent of Rachel Carson's Silent Spring8, published in 1962. She wrote: “These sprays, dusts, and aerosols are now applied almost universally to farms, gardens, forests, and homes — nonselective chemicals that have the power to kill every insect, the 'good' and the 'bad', to still the song of birds and the leaping of fish in the streams ...”
Carson was describing the environmental devastation caused by the over-reliance on and overuse of organochloride insecticides such as DDT (dichlorodiphenyltrichloroethane) in the 1950s and 1960s, which led to major problems with outbreaks of pesticide-resistant pests, widespread contamination of the environment and knock-on effects through the food chain, including chronic poisoning of people. She would undoubtedly think that we seem to have learnt little from our past mistakes.