Neonicotinoids linked to decline in population of some birds – experts respond

SMC UK: Research from the Netherlands published today in the journal Nature found that populations of farmland birds fell most sharply in areas where pollution from neonicotinoids was highest.

Screen Shot 2014-07-10 at 4.36.36 pm
Starlings: among the birds thought to be most affected

Neonicotinoids are used to kill insects that can damage crops. Among the birds most affected are starlings, tree sparrows and swallows.

Our colleagues at the UK Science Media Centre gathered the following commentary from experts as well as the Before the Headlines analysis of the research featured below.

Dr Dave Moore, Senior Scientist at CABI, comments: 

“The paper once again demonstrates that effective alternatives to chemical pesticides are urgently needed. If a fraction of the money spent on chemical insecticide research was spent on developing biological pesticides, we could have environmentally safe methods of delivering control that would have many benefits over the chemicals. Not least of the benefits would be increased control of pests by the biodiversity killed as collateral damage by chemicals. Biological pesticides can be incredibly cheap to develop and can deliver control that is as effective, though with different characteristics, as chemicals.”

Dr Raimund Grau, Senior Ecotoxicologist at Bayer CropScience, comments: 

“The hypothesis of the authors and their conclusions primarily build on data sets which were already published by van Dijk et al. (2013). This referenced publication used the same imidacloprid surface water data set to claim declines in Dutch surface water invertebrates and was recently rebutted in another publication by Vijver and van der Brink (2014).  The published rebuttal criticised both the method used and the conclusions reached in this analysis. In particular, the measurement of imidacloprid residues and macro-invertebrate monitoring were not even taken from the same locations, nor at the same time.  The study of Van Dijk et al. (2013) has also been evaluated by the authorisation board in The Netherlands, Ctgb. Ctgb concluded that this study cannot be used to show a causal relationship between the concentration of imidacloprid and the number of observed species.  Therefore, if this conclusion that imidacloprid causes declines in Dutch surface water invertebrates is not reliable, then drawing the additional conclusion that the same imidacloprid water residues are responsible for bird population declines that feed on these aquatic invertebrates, is even less plausible.

“The basis for the correlation of “concentration in water” leading to “decline in food resource, as the investigated species live on emerging insects” is not substantiated. Most of the birds on the list are not foraging to a large extent on insects emerging from water bodies. Some of the species showing significant negative correlations have a diet that would be dominated by invertebrates other than those with aquatic life stages e.g. skylarks predominantly feed on ground dwelling Coleoptera, spiders etc., Mistle thrush on earthworms, snails etc., Starling on leatherjackets and other soil invertebrates.

“No attempt is made to account for other possible sources of the reported “decline” like change of habitat. Two of the authors (van Turnhout and Foppen)[1] in 2010 concluded “that trophic mismatches (consequent on climate change) may have become a major cause for population declines in long-distance migrants in highly seasonal habitats.” This conclusion was for forests but agricultural areas are even more seasonal. Therefore this known aspect should have been considered in an unprejudiced research project, especially when in this study too all the exclusively insectivorous species with a significantly negative trend are long-distant migrants.

“Overall, this is a correlational study of potential indirect effects of imidacloprid on birds and does not test causality. The strong bias of the publication is evident e.g. by a sentence like this ‘However, as our results are correlative, we cannot exclude other trophic or direct ways in which imidacloprid may have an effect on the bird population trends. Food resource depletion may not be the only or even the most important cause of decline. Other possible causes of decline include trophic accumulation of this neonicotinoid …’. Other potential causes of a bird population decline are not even mentioned.”

[1] Both,C. et al.,Avian population consequences of climate change are most severe for long-distance migrants in seasonal habitats Proc. R. Soc. B. 2010 277 1685 1259-1266doi:10.1098/rspb.2009.1525

Prof Charles Godfray FRS, Oxford Martin School, Oxford University comments:

“The strength of this work is that it brings together for the first time datasets on bird populations and neonicotinoid concentrations in the environment to ask very important questions.  But as the authors acknowledge it is a correlative study and there will be debate about the detailed statistical methodologies used (for example whether all confounding variables have been accounted for and whether spatial covariance in the data has been adequately dealt with).  What I take from this paper and other work on the effects of neonicotinoids on insect pollinators is the enormous importance of setting up large, replicated field experiments in real agricultural landscapes to get much harder data on the effects of this class of insecticide on all elements of biodiversity.”

Declared interests

Prof Godfray is Chair of Defra’s Pollinator Expert Advisory Group

Dr Grau’s, employer, Bayer CropScience, manufactures neonicotinoids including Imidacloprid

Before the Headlines

Title, Date of Publication & Journal

Declines in insectivorous birds are associated with high neonicotinoid concentrations

Nature, 9 July 2014

Claim supported by evidence?

The paper supports the claim that use of neonicotinoid pesticides is associated with declines in the population of some bird species


· The study used standardised data on bird populations collected over a long period of time
· A limitation is results are from a single country (The Netherlands)
· As with all observational data, correlation cannot prove causation (though here, a causal relationship seems likely)

Study Conclusions

The study found that areas of The Netherlands with high concentrations of imidacloprid in surface water high greater rates of decline in a variety of bird species than areas with low concentrations of imadacloprid.

The association was found in appropriate statistical analyses and was highly significant overall (P < 0.0001). In areas with high imadacloprid concentrations, bird populations declined by 3.5% on average annually.

In addition to the significant overall effect, statistically significant inverse relationships between imidacloprid and population growth in individual species were found in 6 of the 15 species studied (after appropriate statistical adjustment for multiple comparisons). None of the species showed the opposite effect.

The authors suggest that the relationship between neonicotinoid concentrations and bird decline is causal. Although observational data like this can never prove causality, the claim of causality here seems plausible. This is partly because of the careful way the authors investigated alternative explanations for the relationship (see below), and partly because the results are exactly what we would expect from biological plausibility.

There is good evidence that use of neonicotinoids harms insect populations (see references below). The birds observed in this study are insectivores, and so it is to be expected that if insect populations decline, then insectivorous bird populations would decline as a result.

It is also possible that imidacloprid may have direct toxic effects on birds, though the authors suggest that the loss of food source when insect populations decline is a more likely explanation for the effects seen in birds. However, their study does not provide any data to distinguish between those possibilities.


Data on bird populations were collected in a standardised way over a long period of time in The Netherlands, so the study had a robust dataset to track bird populations.

The authors were careful to exclude alternative explanations for the association between neonicotinoids and bird population decline. They found no such relationship between location and population decline in a time period before imidacloprid was used, thus ruling out a long-term trend. They also adjusted for land use in a multivariate analysis.

A limitation is that the research is from a single country, so we cannot be sure that the results would apply outside The Netherlands. However, it seems reasonable to believe that countries with similar insect and bird populations would behave similarly.


There is abundant literature on the harmful effects of neonicotinoids on beneficial insects. A small selection is provided below:

Krupke CH, Hunt GJ, Eitzer BD, Andino G, Given K (2012) Multiple Routes of Pesticide Exposure for Honey Bees Living Near Agricultural Fields. PLoS ONE 7(1): e29268. doi:10.1371/journal.pone.0029268

Cresswell, James E. “A meta-analysis of experiments testing the effects of a neonicotinoid insecticide (imidacloprid) on honey bees.” Ecotoxicology 20.1 (2011): 149-157.

Henry, Mickaël, et al. “A common pesticide decreases foraging success and survival in honey bees.” Science 336.6079 (2012): 348-350.

Alaux, C., Brunet, J.-L., Dussaubat, C., Mondet, F., Tchamitchan, S., Cousin, M., Brillard, J., Baldy, A., Belzunces, L. P. and Le Conte, Y. (2010), Interactions between Nosema microspores and a neonicotinoid weaken honeybees (Apis mellifera). Environmental Microbiology, 12: 774-782. doi: 10.1111/j.1462-2920.2009.02123.x

Any specific expertise relevant to studied paper (beyond statistical)?

I have no professional expertise in this area beyond statistical, though I have been following the literature on neonicotinoids and insects for some time as a matter of personal interest.

Before The Headlines is a service provided to the SMC by volunteer statisticians: members of the Royal Statistical Society (RSS), Statisticians in the Pharmaceutical Industry(PSI) and experienced statisticians in academia and research. A list of contributors, including affiliations, is available at

* ‘Declines in insectivorous birds are associated with high neonicotinoid concentrations’ by Caspar Hallmann et al was published in Nature at 6pm UK time on Wednesday 9 July.