The All Party Parliamentary Group on Agroecology (APPGA) “A Countryside fit for Pollinators”

Meeting date 12th February 2013

Caroline Lucas MP is the chair today.  She opened the meeting by saying that pollinators are massively central to our whole agricultural system both in terms of the economy and in terms of food sustainability and security.  The services that pollinators provide would cost the UK $1.8 billion annually if agriculturalists had to pollinate by hand.  She reported that two bumble bee species are already extinct and that between 1985-2005 managed honey bee colonies declined by over half.  Reasons for decline are complex and not yet fully understood, but are thought to be due to a combination of factors, which include climate change, parasites, herbicides and pesticides, and in particular a new class of pesticides called neonicotinoids.

The APPGA group is concerned as a whole about intensive farming practices and the growing evidence that neonicotinoids harm bees, so they are promoting best practice to government and farmers for the restoration of farmlands and ecosystems, and a reduction of pesticides to protect biodiversity.

Dave Goulson is Professor of Biology and Environmental Sciences, University of Stirling. He recently gave powerful evidence to the Environment Audit Committee about bee populations decline and below is a transcript of his talk given to the APPGA.

Farmland Wildlife and Systemic Pesticides

Thank you very much indeed for the invitation to speak here.  I do work mainly on bumblebees, but I’m not going to be talking too much about bees today.  I’m trying to cover a broader area relating to pesticides and wildlife generally.  I come from a conservation background and I’ve spent the last twenty years not working on pesticides, but trying to work out how to conserve bumblebees, primarily.  I guess that gives me an inherent bias, but as a scientist you aim to be objective and impartial when you are assessing the evidence.   The last couple of years I’ve been working on pesticide related issues, and however hard one struggles to remain impartial, I suspect that you might find what I am going to say in the next twenty minutes slightly one-sided, so do bear that in mind. I have done my best to present everything as factually and accurately as I can, but it’s up to you to decide whether it’s entirely fair or not.  We shall see.

Britain is really lucky.  We have amazing data sets. Long term data sets, on the population of some animals and plants, much better than any other country, particularly for birds and butterflies, classic long term data sets that have been going on for decades, which allows us to see which species are increasing and decreasing and so on. We have slightly less strong data set for moths, and definitely not so good ones for bees and some families of beetles.

And it’s apparent by looking at these data that farmland wildlife is declining rather rapidly, more so than in the woodlands and uplands, the vast majority are declining. So here are some examples I picked from the internet.  Linnets, farmland butterflies and farmland moths, and you can see they are all doing rather badly.  The question is why? Well, you probably know most of the answers why, at least broadly.  It’s very easy to explain why farmland wildlife declined between 1945-1990.  Essentially, during the Second World War, you can blame Adolf Hitler if you like, we needed food, food security then really was important in the early 1940’s because we hadn’t got access to imports, so government policy was designed to encourage farmers to maximize food production at the expense of everything else.  They brought lots of new land into farming, they ploughed up ancient hay meadows, drained marshes, ripped out hedgerows, and so on, all of which was exactly what we needed if we wanted lots of food.  It’s slightly odd that we continued with those practices up until 1990, despite the fact that food security was slightly less of an issue and war had long since ceased.  That also coincided with the period when lots of artificial chemicals were brought into farming  - pesticides and fertilizers -  this is when the  organochloride insecticides became mass manufactured during the war to control mosquitoes and the organophosate chemicals came out of Germany where they were developing nerve agents. So, the second half of the 20th century was pretty bad for farmland wildlife generally.

What puzzles me is why farmland wildlife is still declining, because we have stopped doing most of these things, we’re not increasing the intensification of farming anymore as far as I can see.  In fact quite the reverse, we now invest lots of money in paying farmers to adopt schemes that are designed to boost biodiversity further, something like £400m a year to farmers for schemes like planting hedges, planting strips of flowers, and so on, specifically intended to boost farmland biodiversity, and yet it isn’t working.  And we have to ask ourselves why, what’s going wrong?  The answer is probably complicated, partially to do with the fact that many agro-environmental schemes aren’t terribly effective,  but I’m coming round to the view that a little part, and I‘m not quite sure how little, may be to do with neonicotinoid insecticides. I guess that’s why many of you are here and that’s what I really want to focus on today.

So I’ll give you a brief introduction to neonicotinoid insecticides.  They are a group of insecticides that were developed in the eighties and introduced in the early nineties, and have become the most widely used insecticides in the world. This slide shows the UK use of neonicotinoids in 2010, five different compounds are licensed for use in the UK and clothiandinin is by far the most widely used at the moment.  They are mainly used in seed dressings, so about 90 percent of the use in the UK is applied to seed before they’re sown. The farmer buys treated seeds, for example oilseed rape seeds and they are sown, usually in the autumn and as the plant germinates and grows, the compound is water soluble, it dissolves around the seed and the seedling sucks up the neonicotinoid and it spreads through all its tissues and it protects the plant against insect herbivores, very effectively, particularly in the early stages of the growth of the plant, which is exactly what the farmer wants.

Neonicotinoids are also used as sprays, largely sprayed onto strawberries and raspberries and various other horticultural crops.  Sometimes they are used as soil drenches to kill things like leatherjackets (larvae of the European Crane fly or Daddy Long Legs) in turf.  And they are widely sold for gardening use.  Any of us could go out into a local garden centre and buy some Ultimate Bug Killer or a range of other products, which are based on neonicotinoids, which are advertised for use on your roses, or your vegetables or whatever.

So that just shows you the increase in use over time.  They were first used in the UK in 1994, steadily increasing year on year.  Those are the different proportion of the different compounds.  I should say that we are very lucky in the UK, because DEFRA provide all this data on the website for anyone to see.  It’s actually almost impossible or extremely hard and none of us have succeeded in getting similar data in any other country in the world, but anyone can see the data for the UK, which is nice, well done DEFRA.

There have been lots of studies of the toxicity of neonicotinoids on various different organisms, and I’ve picked out a very few to give you some idea.  So toxicity studies usually calculate either a LD50 or an LC50.  Now, an LD50 is the dose that will kill half the animals in your test in a single one off dose, either painted onto the animal, or given it in a single meal.  The LC50 is the concentration that will kill it, either in the concentration of food that it’s consuming or more commonly the concentration in the environment in which it’s living.  So if it’s an aquatic insect, a may fly for example, then the testers add it to the water and they see what concentration will kill may flies, so this gives you some idea.  So, for example, rats are fairly resistant, it doesn’t seem to have much impact relative to other organisms, this is particularly imidacloprid I’m focussing on here, but the others are similar. So it takes 177mg of imidicloprid to kill half the rats in your test, for example.  A grey partridge is somewhat more susceptible, 5 mg per partridge.  But then if you look at honey bees, it’s 4 nanograms.  The difference there is that there are milligrams and nanograms, and there is a million-fold difference in units being used there.  So it’s much, much more toxic to insects than to vertebrates, even allowing for the larger size of a partridge compared to a honey bee, partridges are about a thousand times as heavy as a honey bee, but it takes a millionth as much imidicloprid to kill the honey bee than it does the partridge.  So it’s about a thousand times more toxic to insects than vertebrates taking into account their size. For aquatic invertebrates such as the shrimp or the may fly, the amounts that kill them are in the region of 1-7, and some other species 10-15 part per billion (ppb) in the water.  Just for a comparison with other insecticides, I’ve put down there on the left hand side, the LD50 for honey bees and for two other well known insecticides, so imidicloprid is 5 nanograms per honey bee, dimethoate, which is an organophosphate, fairly nasty stuff is considerably less toxic, it takes a lot more to kill a bee.  And just for comparison, DDT is about 5 000 times less toxic to honey bees than imidacloprid.

So they are quite poisonous, but of course, that’s exactly what you want, they are insecticides.  So it would be kind of odd if they weren’t poisonous to insects.  They are much more poisonous to insects than they are to mammals, including us presumably.  Ok, so they are mostly, in the UK, 90% applied as seed dressings, and they’ve always been billed as being a really good way of targeting the crop because the old way of applying pesticides generally was to spray them from a boom on a tractor and that inevitably risks them drifting on the wind, contaminating hedgerows, being breathed in by the farmer and so on.  The idea of painting them onto the seed and then planting them in the ground so they all get sucked up by the crops sounds brilliant, it’s all targeted at the crop and only the crop is protected, you haven’t got your insecticide swilling around in the environment.  Except, it turns out that it isn’t quite as brilliant as that.  So a small amount of dust is created during the drilling and that’s been shown to be extremely toxic to insects, bees that happen to be flying nearby at the time.  And, presumably, that could drift, just as the spray from tractor boom can drift. But it’s in the region of maybe 1% or so that blows off as dust, maybe less, depending on exactly how you sow the seeds.  But a small amount is lost that way.  The really remarkable thing is the studies of the amount that is then drawn up by the crops, suggests that on average about 2% of the amount put on the seed ends up in the plant – the target crop.  Which then leaves the question, where’s the rest of it?  So in the region of 96% is going into the ground on the seed, but it’s not being sucked up by the plant.  Some of it, it seems, probably gets leached out in water, they are water soluble, there is evidence from some countries for neonicotonoids ending up in steams and so on. But it seems that the majority of it probably stays in the soil.

How long does it last in the soil?  There have been loads of studies of the persistence of neonicotinoids in the soil.  Some studies suggest that they have a half life (the time it takes for half of the compound to disappear, to break down, to degrade or whatever) - a few studies have shown half lives of less than a hundred days, but the vast majority are of between the region of 200-500 days, so about a year say, for half of the compound to disappear.  And one or two studies, it seems to depend on soil type, I’m not quite sure, suggest that sometimes the compounds have half lives in excess of 1000 days.  One study showed a half-life of 6000 days, so that the chemical is almost sitting there permanently in that situation.

Now if the half life is about a year, that means we would expect them to accumulate over time if we use them every year.  Now that’s something that certainly the pesticide companies to my knowledge have always denied is the case.  They seem to suggest that they  break down rather quickly in the environment and they’ve been a little bit evasive on this point, I would say, in the past.  But certainly the data that’s been published would suggest that these compounds should accumulate.  The big question is, do they?  Well, it turns out, at least from some studies conducted by Bayer themselves in the early 1990’s, which they do.  This is taken from an online document that anyone can get hold of, a draft assessment of a report on imidacloprid in 2006, [1] and it shows studies done in East Anglia by Bayer scientists in the early 1990’s where they simply sowed a seed treated winter wheat crop every year in the same field for six years in a row, and the day before they sowed the following year they measured the amount of imidacloprid in the soil.  And you can see that after one year in the top left trial that’s at Bury St Edmonds at a low application rate, there was still 11ppb of imidacloprid in the soil a year after it was used.  I should explain.  There were two different sites -  Bury St Edmonds on the left and Wellesbourne on the right - and they had two different application rates with a doubling of the amount on the seeds.  The details don’t really matter.  What’s clear to me and I think to anybody is that these things do accumulate in soil year on year, if you keep applying them, you’ll get more and more.  In fact the concentrations for the higher application rate are getting up to 50-60ppb, which bear in mind that the amounts needed to kill aquatic invertebrates are of less than 10ppb.  The amounts needed in the crop to kill the crop pests are typically about 10ppb.  So if you’ve got 50-60ppb in your soil, what’s that going to do to things that live in and around the soil?  All the soil inverts that are important for maintaining soil health and ultimately for maintain farming productivity?

What I find most alarming of all is that the study was done by Bayer scientists, submitted to the EU regulators as part of a huge bundle of documents presumably regarding the safety of these compounds.  The regulator wrote this (about these graphs) on the same page of the document, there is no doubt that this is what he was talking about, “Long term field observation trials of imidacloprid in the soil for its repeated use as a seed treatment for six consecutive years have confirmed that the compound has no known potential for accumulation in the soil.”  And as a result of that it was all decided it was perfectly fine, they didn’t look into it any more and they continued to sell imidacloprid.  I really struggle to work out what’s going on here.  There are only two explanations here, one is that the person who wrote this was an idiot! And the other explanation is much more unpleasant.  And I’ll leave you to work out which one it might be.

Ok.  Slight change of tack on soil for a second.  One way of using this compound is to add them to irrigation waters, so they are commonly added to irrigation water for vines as a soil drench and are then sucked up by the roots of the plants.  And one treatment of imidicloprid to the vines is sufficient to control pests through the growing season and I can give you references to these things if you want them.  These are all studies that have been published.  Sometimes they are injected into trees to protect the trees against pests.  Often ornamental trees in urban areas.  So this guy [on the slide] is injecting a tree and that tree will then be essentially toxic to any herbivores. And the idea behind this is in urban areas things like these green fly which release honeydew, which might land on a car, and you end up with a sticky roof to your car, which is obviously an appalling thing to happen so much better to inject the tree and make it entirely toxic to all insects.

An application can control insect pests for four years in a tree.  So they can accumulate in soil and they can persist in plants for a long time it would seem.  So what does that mean about the environment?  If you look at this picture here, you’ve got arable fields which presumably have been exposed in some years, perhaps not in every year, but quite regularly to additional doses of neonicotinoids in crops, and you’ve got hedgerows and these roots of these hedgerow trees and the shrubs all going into the field.  Are they not absorbing neonicotinoids?  I don’t know.  Nobody’s looked and I would very much like to, but haven’t yet had the funding to do so.  But it seems to me to be quite possible that the soil is all full of neonicotinoids.  There is a French study where they survey the land of arable fields and found a range of concentrations, some exceeding 100ppb in France.  Nobody has done that in the UK.  It would love to do it (I’m not sure if love is the right word), just to see how much is out there and what it’s getting into and how much is getting into the hedgerows, shrubs and trees and if it is, couldn’t that explain why lots of insects in farmlands are declining.  It could, but I don’t know.  I don’t have any data on this, nobody does so far as I can tell, but it strikes me as something that really does need looking into.

Everyone has always assumed that vertebrates are not affected by these compounds. Apart from one or two people who I would describe perhaps as being a tiny bit eccentric, but on the whole people haven’t claimed that neonicotinoids are directly impacting on invertebrates because they are much less toxic to vertebrates as I told you at the beginning.  That said, the LD50 for a grey partridge is 5mg, so if it eats 5mg it’s got a 50% chance of dying.  If it eats more than that then the odds of it dying go up very rapidly.  That’s the equivalent of eating 5 treated maize seeds or 6 sugar beet seeds or 32 oilseed rape seeds and given that a grey partridge eats about 25g of seeds a day, it might be able to consume 600 maize seeds in one day if they were left lying around.

Now there’s a study from the States, which suggests that on average about half a percent, sometimes up to one percent of seeds during drilling are left on the surface, spilled by the farmer or the unevenness of the ground means that the drilling operations didn’t get all the seeds properly buried.  And inevitably if the farmer’s sowing for example 800 000 seeds of oilseed rape per hectare, a few are going to end up on the surface.  And it seems to me entirely plausible that a grey partridge might stumble across 5 maize seeds when it’s wandering around a field in the autumn just after it’s been sown, or for that matter a few dozen oilseed rape seeds.  And again, I have absolutely no evidence that grey partridge or any other granivorous farmland bird or for that matter voles or whatever that might eat grain are actually being poisoned by these compounds, but equally, I can’t find any evidence that says they’re not.  And it strikes me as highly plausible that this is a route which could be causing problems for farmland birds and small mammals.

There have been quite a few statements put out by the agro-chemical industry in defence of these compounds and I’ll go through them as quickly as I can.  One is that they say bees and plants are still unwell in France, where these compounds have been largely banned or they are now very restricted.  Well that isn’t really true.  For example thiamethoxam was recently banned for use on oilseed rape.  But farmers just used clothiniadin instead, so it doesn’t mean that there is any reduction in the amount of neonicotinoids being used.  Partial bans of one compound on one crop just mean that the farmers use a different compound.  There is no crop in France in which all neonicotinoids are banned.  So I strongly suspect that the use hasn’t declined at all in France.

The second one they always say is that bees in Australia are all really healthy and they use loads of neonicotinoids in Australia.  Therefore it cannot be anything to do with neonicotinoids.  I just took these off the website, these first two quotes from the website of the Australasian Bee Keeper Journal.  I’ve no idea how credible they are, but it seems to me that the Australians aren’t quite so convinced that their bees are as healthy as the agro-chemical industry is.  For example, “beehive numbers have deteriorated to 50% less of what they were 20 years ago.  The use of neonicotinoids is the single biggest threat to the survival of beekeeping”, says an Aussie beekeeper.  So perhaps they are not quite as healthy in Australia as we have been led to believe.

It’s often claimed that varroa has been widely demonstrated to be the big killer in bees and therefore it can’t be anything to do with pesticides.  It’s a bit of a nonsense argument.  For one, varroa only affects honey bees, not bumble bees for example, many of which are also declining, but also it’s a bit like saying eating fatty food cannot be bad for you because smoking is much worse.  It doesn’t really make any sense.  They often say that lab studies are all nonsense and most of the studies done to date which have shown impacts of pesticides on bees are at least partially done in the lab, partially done in the lab because that’s the only way one can control exposure of the bees.  In actual fact, most of the best science ever done has been done in the lab.  There’s nothing fundamentally wrong with lab studies and as you will eventually discover when the results are announced, the recent DEFRA field study ran into exactly this problem.  If you try to do everything in the field, it’s actually extremely difficult to control where the bees feed and what they are exposed to.  You end up with no controls.

Anyway, where are the industry trials demonstrating that neonicotinoids are safe?  Surely it should be up to them to produce field trials showing they’re not having an impact rather than up to me to prove that they are?

The final myth about neonicotinoids is that, if they were withdrawn, there would be enormous economic ramifications for the EU.  There was recently a document published by the Humbolt Forum for Food and Agriculture, which is funded by Bayer and Syngenta exclusively, which claimed that it would cost the European Union 17billion euros and 50 000 jobs if neonicotonoids were withdrawn.  It wasn’t really based on any evidence.  I’ll have to skip over the details of that.  But I tried to dig around to find out what evidence there was and I’m not saying there wasn’t some truth in it, but it’s very hard to get to the bottom of what the ramifications would be if they were withdrawn.  One example I’ve come up with here and you can actually look at the use over time of neonicotinoids which I showed you earlier and they’ve obviously greatly increased, there were none used prior to 1994.  If you look at oilseed rape yields, a crop which has gone from being not at all treated with neonicotinoids to 100% treated with neonicotinoids, there is no significant increase in yield at all over time.  So farmer’s yields have not increased with the increase from not using neonicotinoids to using neonicotinoids.  What that tells you I’m not entirely sure, but it certainly suggests that oilseed rape yields wouldn’t drop if farmers went back to what they were doing before.

I really struggled to find any good examples of studies close to home where they have compared the use of neonicotinoids with alternative means of controlling pests.  But there’s a really nicely done study from Brazil on soya beans [2], not terribly relevant here, which actually shows that although nearly all farmers use neonicotinoids they would get a higher yield per hectare if they went back to integrated pest management which, when I was at university, was what was taught as the gold standard of pest management, which largely boils down to monitoring your pest problems and only using a pesticide when you absolutely have to rather than chucking them on all the time.  The whole concept of a seed dressing means that you have to apply it every year before you know whether or not there is going to be a pest problem, which for me is the fundamental problem with neonicotinoid seed dressings.

So far as I can see the evidence is not clear with regard to the economic benefits of using neonicotinoids.  But there is a remarkable lack of data one way or the other.  Some other issues I just want to flag up:  I think it’s fundamentally wrong that agrochemical companies conduct their own safety tests and then submit the data to a regulator.  It’s very easy to bias the outcome of an experiment if you’ve got $3.5billion at stake. So that seems to me to be daft.  It also seems daft to me that these studies are never made available for public scrutiny.  They go to regulators and as I suggested earlier with that accumulation study my faith in the regulators is less than 100%.  So the system doesn’t seem to be terribly robust.  The idea that we [scientists] should be expected to produce evidence that these compounds are harmful before any action can be taken seems to have got things back to front.  Surely industry should demonstrate clearly that they are safe, should produce the field trials that they have done in all their detail to convince us that these compounds are safe.  Or are they expecting us to do the reverse, when we don’t have any funding to do that?

One thing I find very worrying is that much agronomic advice given to farmers in the UK comes from people who have very strong vested interests in selling them chemicals.  I had a meeting last summer, 2012, with a company called Abagrii who employ 400 full time agronomists to travel the country, they said they had about 40% of the UK market for advising farmers on what crops to grow, which varieties to grow and which pesticides to apply to them and so on.  And they openly admitted that 90% of their profit came from the mark-up on the pesticides they stock and sell on to the farmers.  So imagine if you went to the doctor and the doctor worked on commission from the pharmaceutical industry and he would get a hundred pounds for prescribing you drug Y and ten pounds if he prescribed you drug X and nothing if he told you there was nothing wrong with you, go home, you wouldn’t be very happy.  But that’s exactly what we have in farming.  A system that’s basically set up for farmers to be encouraged to use products they don’t necessarily need and to use more than they need.

And finally, why on earth do we sell highly toxic compounds to gardeners?  There is no economic case.  I could bang on about this for half an hour.  Toronto has banned use of all pesticides in gardens and they still have nice gardens.  Does it matter if there are a few aphids on someone’s roses? It really doesn’t.  We should encourage suburban areas to be oases for wildlife not places where gardeners can wander round without even bothering to read the label and chuck toxic chemicals all over everything.

This is my last slide and this is just to show you bee Armageddon. This is South West China, Sichuan, where they now have to hand pollinate their apple trees because they have used so many pesticides there are no insects and no bees.  And they send their children up to do the higher flowers.  Now they don’t always wear their fancy local costume, I suspect that may be for the cameras.  They can probably afford to do this in China where they have cheap labour for a valuable crop, but can you imagine a farmer in this country hand pollinating his oilseed rape?  It doesn’t seem to be viable to me.

This talk was transcribed by Sam Burcher MSc (4,995w)

(photo Dave Goulson and Caroline Lucas (c) Sam Burcher 2013

References:

1. Pesticide use and crop yields –           Defra (2012a).https://secure.fera.defra.gov.uk/pusstats.  Accessed 20/1/1          Defra (2012b). http://www.defra.gov.uk/statistics/foodfarm/food/cereals/cerealsoilseed/     Accessed 20/1/13.
2. Brazilian study showing neonicotinoids less effective than IPM:
3. Bueno, A.D., Batistela, M.J., Bueno, R.C.O.D., Franca-Neto, J.D., Nishikawa, M.A.N., Liberio, A. (2011). Effects of integrated pest management, biological control and prophylactic use of insecticides on the management and sustainability of soybeans. Crop Protection, 30, 937-945.