Keeping pollinators safe in our cropping systems


Farmers have known for centuries that insect pollinators play a role in increasing the yield of many crops, but our knowledge of how to manage pollination for greatest yield benefit has not kept pace with other advances in agronomy. As a result, we typically pay great attention to inputs such as water, fertiliser, and pesticides, but usually have an ad hoc approach to pollination, even when growing crops known to benefit from insect pollination. This attitude is beginning to change, as research all around the world (including Australia) has demonstrated that for many crops the role of pollinators in achieving maximum yield is more important than had been appreciated.

For example, people have often assumed that if a crop is self-fertile then insect pollinators will not be important. However, experiments commonly show that even self-fertile crops, such as many Brassica species (e.g. canola), yield more when insects increase the rate of pollination.

Another reason for the recent interest in the role of pollinators in agriculture is because of fears that pollinator populations are in decline around the world. As usual, the truth is a bit more complicated. The number of managed honeybee hives is declining in some countries and increasing in others. One driver of decreases is the increasing frequency of certain honeybee diseases, and there is an emerging view that diseases may be driven by stresses to bees in highly agricultural environments. Australia has not experienced dramatic declines in managed bee numbers, but nevertheless does experience problems associated with certain diseases. 

Much less is known about the status of wild pollinators because most of them are never monitored. However, the pattern of change reflects the general pattern for global biodiversity, with some species known to be in decline and some known to have gone extinct. On the other hand, many bees are well adapted to quite open environments and can prosper in agricultural landscapes as long as they have places to reproduce, and flowers are available to collect food from during their period of flight. These resources can be provided by patches of non-crop land in the landscape (including roadsides, scattered trees, fence lines, uncropped paddocks, bush blocks etc). Crops with flowers attractive to insects play a role in providing food, but obviously only during the flowering period, which might be quite limited.

Risk of exposure to insecticides is one of the challenges for pollinators in agricultural environments. Wise use of chemical insecticides has always been a difficult balancing act. On the one hand, reducing pest damage to crops is obviously of great benefit to growers, supporting yield and reducing risk. On the other hand, insecticides can cause unintended harm to beneficial insects, including pollinators. Beekeepers can reduce the risk by moving hives away from crops during periods of insecticide use; a strategy which relies on good communication between growers and local beekeepers. Wild pollinators are more vulnerable because they live in the same area for their whole life cycle, and have no one to look out for them.

Integrated pest management approaches aim to balance the benefits (pest control) with the costs (including loss of beneficials), but decisions regarding what is the acceptable balance depend both on our knowledge of the system (such as ‘what is the field exposure level for insect species x, y and z?’) and on changing views regarding what kind of negative impacts are considered acceptable. These issues have come to a head in Europe in recent years, where public concern about bees led ultimately to the European Union establishing a moratorium on a number of neonicotinoid insecticides. 

Here I focus on our knowledge of the system rather than investigating what created public concern in Europe, except to say that losing the support of the broader community is a risk that Australian growers need to consider, even if they see the problem differently to those who speak out against insecticide use.

The risk that an insecticide will have a substantially negative effect on the population of beneficial insects depends on both the direct toxicity and the rate of field exposure. Toxicity is relatively well understood because it can be examined in the laboratory, but field exposure is much more complicated and difficult to study. It depends on factors including how long the active chemical remains in the environment and how the insect moves and feeds in its environment. Both these things vary from place to place and from time to time. So, while neonicotinoids are known to be toxic to bees (like most insecticides) they can be applied in ways that greatly reduce field exposure. 

Seed-coated systemic insecticide, for example, will have a high concentration when the plant is small, and then much reduced concentration when the plant is flowering and attracting pollinators. This application strategy has the potential to reduce the exposure of pollinators, compared with insecticides that are sprayed, and may therefore leave a toxic residue on the crop or be vulnerable to drift. But, being a new strategy, new and different kinds of exposure have emerged.

Some bee kills in Europe were attributed to accidents where an insecticidal dust was released from the seed coating during pneumatic seeding. This problem was solved, but is a reminder that significant impacts can come from unexpected incidents. Most of the ongoing debate, however, surrounds the issue of ‘sub-lethal impacts’. This refers to the idea that while we have traditionally focused on the rate at which a toxin kills on exposure, there may in fact be impacts of ‘sub-lethal’ exposures that do not kill immediately, but which lead to progressively declining health. Social bees (including managed honeybees) might be particularly vulnerable because the hive is a ‘superorganism’ where the success or failure of the colony depends on the sum of the behaviours of many individuals and the complex ways in which they interact. In the case of neonicotinoids, experiments show that sub-lethal doses can affect the behaviour of individual bees in ways that effect foraging. 

But to know whether or not this explains other problems with bee health or crop pollination we need answers to these further complex questions:

  • Are these effects common at ‘typical’ levels of field exposure? That is: how many individuals might be exposed and at what concentration?
  • How do effects on individuals affect colony-level health?
  • Are any negative effects widespread enough to general explain bee health problems, or are they rare?

These three questions are now the focus of researcher attention. Highly controlled experiments are useful for better understanding mechanisms of insect response, but the only way to understand real outcomes in the field is to study them at full scale on working farms. To date, few of these studies exist, but two important experiments have been published recently.

A Swedish research group (Rundlof et al. 2015) examined bee population responses in fields of oilseed rape where fields without insecticide were paired with fields that had systemic neonicotinoid seed treatment. They detected decreases in wild bee density, and reduced colony growth of wild bumblebees, but no negative effects on managed honeybee hives. A similar study from Canada (Cutler et al. 2014) also found no negative effects on managed honeybees, but did not collect data on wild bees. Both these studies address only ‘within-season’ effects on bees. 

More research is needed to determine if there are any cumulative effects over multiple seasons, keeping in mind that honeybee colonies in particular can persist for years, over which time they will be exposed to a wide range of food resources, potentially including a range of different crop species under a range of different insecticide use strategies. For example, bee hives in Australia might move over a matter of months from canola, to almonds, to apples, to bushland eucalyptus and so on.

A recent review of scientific research on the effects of neonicotinoids on bees (Lundin et al. 2015) found that although there has been an explosion of interest, the available data remains focused on studies of honeybees (i.e: one species), in Europe and North America, and on only a limited number of crops. We have relatively little understanding of effects on other species (like the 20,000 wild bee species globally) or in other climates and cropping systems, such as those in Australia. Regulators have to work with the level of available knowledge and monitor new discoveries as they come to light. The Australian Pesticides and Veterinary Medicines Authority (APVMA) reviewed knowledge on neonicotinoids in Australia in February 2014, and concluded that there was no need to follow the European example by further restricting allowable uses.

Their report also encouraged more research and surveillance, and better product stewardship. Because new practices are diverse and constantly evolving, and natural systems are complex, we will never be in a situation where knowledge is complete, but instead must make decisions on the basis of calculated risk.


Protecting pollinators in Australian agriculture begins with careful use of insecticides that stays within the registered uses determined by the regulator. When weighing up merits of an insecticide use, make sure to think about the economic and environmental costs as well as benefits, and be aware that some important crops for Australian grains growers receive a benefit from pollinators, even though we have not traditionally focused on pollinator management for yield.

Considering all pesticide use through an integrated pest management approach will help minimise the need for pesticides in many situations. Wild pollinators also need flowering resources to feed on and locations in which to nest, and so benefit when little patches of habitat are preserved throughout the farmed landscape, such as from scattered trees, roadsides, fence lines and so on. By protecting pollinator habitat and by working with beekeepers to avoid problems, growers can enhance their reputation with the broader public. Meanwhile, we need to learn more about field exposure of pollinators to insecticides to ensure that we can reduce conflicts and keep crop pollinators doing their good work.

Useful resources

Cutler CG, Scott-Dupree CD, Sultan M, McFarlane AD, Brewer L 2014 A large-scale field study examining effects of exposure to clothianidin seed-treated canola on honey bee colony health, development, and overwintering success. PeerJ 2:e652; DOI 10.7717/peerj.652

Lundin O, Rundlöf M, Smith HG, Fries I, Bommarco R 2015 Neonicotinoid insecticides and their impacts on bees: a systematic review of research approaches and identification of knowledge gaps. PloS One, DOI: 10.1371/journal.pone.0136928

Rundlof M, Andersson GKS, Bommarco R, Fries I, Hederstrom V, Herbertsson L, Jonsson O, Klatt BK, Pedersen TR, Yourstone J, Smith HG 2015 Seed coating with a neonicotinoid insecticide negatively affects wild bees Nature 521, 77-80
Only available by subscription

Australian Pesticides and Veterinary Medicines authority (APVMA) 2014 Overview Report: Neonicotinoids and the health of honey bees in Australia

BeeAware website

Contact details

Saul A. Cunningham
CSIRO Box 1700 Canberra, ACT, 2601
02 6246 4356