March 26, 2019

expert reaction to the research on Britain’s pollinators

Research published in Nature Communications presents evidence of widespread loss of pollinating insects in Britain.

Paul Pearce-Kelly, Senior Curator of Ectotherms at ZSL (Zoological Society of London) said:

“This research provides further evidence that the majority of insect populations are experiencing major population declines. While the study confirms that some species appear to have increased in distribution (most likely in response to agri-environmental improvement initiatives), the majority of species included in the study demonstrate significant declines.

“The study suggests that improved habitat provision is an essential conservation action but in fact environmental conditions for the majority of species studied is insufficient to prevent their worrying decline. These are highly regarded researchers using robust methodology, utilising the best available data. They clearly explain the constraints and limitations of the study and their conclusions are well balanced and measured.

“Invertebrates are at the very core of our natural world, pollinating important food crops and sustaining ecosystems while acting as a food source themselves for wildlife such as birds, amphibians, reptiles, fish and small mammals. It is essential that we prevent the net loss of biodiversity now – as the connection between wildlife health and human health is intrinsically linked. We are already starting to see the impacts that the decline in invertebrates is having on both wildlife and humans.”

Prof Tom Oliver, Professor of Applied Ecology at the University of Reading, said:

“This study uses state-of-the-art analysis techniques that make the most of Britain’s substantial legacy of natural history observations by members of the public.

“The result that common bumblebees associated with crops such as oil seed rape are increasing is good news. Adequate investment in standardised monitoring schemes to provide additional confidence in these results, and to understand which particular landscape-conservation measures most help pollinators, is still urgently needed.

“The result that solitary bees are declining is very worrying — these species are important for the pollination of fruit such as apples and, crucially, for the pollination of hundreds of wildflower plants which form an essential part of our natural heritage and the beauty of the British landscape.”

Dr Scott Hayward, Lecturer in the School of Biosciences, University of Birmingham, said:

“The paper provides a comprehensive analysis of researcher and citizen science data, setting out the state of bees and hoverflies in the UK.  It clearly illustrates the value of looking at data across long time series, for example with some bee species showing quite large declines in the past 5 years, but actually having higher abundance/site occupancy than in 1980.

“The overall declines in both bees and hoverflies are very concerning, and it is interesting to note a steady decline in hoverflies since 1987, but a much more recent decline (since 2007) in bees.  The reasons for the biggest declines in upland species aren’t that clearly set out in this brief communication, but displacement by/competition from other species moving to upland sites as a result of climate warming could be one cause.

“The study discusses conservation measures, such as the planting of legumes in flower-rich field margins as part of agri-environment schemes, however another important consideration is the greatly increased commercial provision of certain bee species which could account for some of the changes in assemblage evenness noted.  For example, Bombus terrestris provision has increased massively since 1980.  Honey bees are not referred to directly, but there are both wild, feral and managed honey bee populations.  Even some solitary bee numbers are being supplemented to some extent by semi-commercial practises where ‘bee hotels’/nest boxes are collected at the end of each season and overwintered under more controlled conditions in order to then sell ‘provisioned hotels’, particularly to orchard growers, the following year.

“Finally, it isn’t clear if data were collected across all months of the year.  While spatial patterns of species distribution change clearly emerge from this study, it is important to note that temporal changes in periods of pollinator activity (phenology) have also shifted as a result of climate change.  This could mean that some species are being missed as part of pollinator surveys if the timing of sampling doesn’t cover all months of the year.  The implications of changing phenology/temporal distribution can be as important as changes in spatial distribution, for example leading to loss of synchrony with certain flowering plants.”

Professor Nigel Raine, Rebanks Family Chair in Pollinator Conservation at the University of Guelph, Canada said:

“Declines of insect pollinators could affect the critical ecosystem services they provide that support global food production and the reproduction of most wild flowering plant species. This study provides important insights into changes in the likelihood that 353 species of bees and hoverflies were found in individual 1km squares across the UK over a 33-year period (1980-2013).”

“Results from this study reveal a third of 353 wild pollinator species decreased over a 33-year period (1980-2013), while another tenth of these species increased. Overall occupancy by solitary bee species declined by 32%, while occupancy by eusocial bees (predominantly bumblebees) increased by 38%. Trends indicating that bumblebee occupancy has increased could be the result of widespread implementation of agri-environment schemes on farms targeted at supporting these important crop pollinators. If so, then developing and implementing similarly effective conservation schemes to support solitary bees in agricultural landscapes could also help to reverse their declines.”

“Supporting pollinators means minimising key environmental stress factors affecting them, such as habitat loss, pesticide use and climate change. This study confirms that rare pollinator species continue to be at greatest risk of decline, and that different conservation strategies are likely needed to support these species at risk compared to common species needed to pollinate many agricultural crops.”

Dr Lynn Dicks, NERC Research Fellow at the University of East Anglia, said:

“This is an extremely important, high quality analysis of how wild bees and hoverflies have been doing in the UK, since 1980. The message is clear. Overall, wild bees and hoverflies are still declining. A group of common bee species, such as red-tailed and buff-tailed bumblebees, are increasing their ranges. This may be good news for farmers because these bees do around half of our crop pollination, working alongside managed honey bees to provide a value of more than £600 million per year to the UK economy, through better crop yields and quality. These common species are perhaps responding positively to agri-environmental actions to increase flower-rich habitats in farmland, supported by Government since 2003. However, it’s really important to note that the analysis is based on records of distribution (WHERE the bees are found), not counts of their actual numbers. The total numbers of even some common bumblebees, such as the garden bumblebee, are declining nationally, according to analysis of count data collected by volunteers for the Bumblebee Conservation Trust, from 2010-2017 (Comont & Dickinson 2018).

“This increase in some common species is against a background of overall loss of diversity. Every square kilometre in the UK has lost an average of 11 species of bee and hoverfly, between 1980 and 2013, according to the new analysis. Species living in uplands are doing the worst. Solitary bees, which mainly live in burrows in the ground or in small holes in hedgerows, trees or walls, are mostly declining. And for bees, the steepest declines have been since 2000.

“This pattern of biodiversity loss is happening everywhere we look. Some common species increase, while many more species decline, and the decliners tend to be more specialised in their needs, associated with particular habitats or food sources for example. It’s a process of homogenisation and leaves us with a natural world that is far poorer and less resilient to change.”

Dr Richard Comont, Science Manager at the Bumblebee Conservation Trust and Honorary Research Fellow at the University of Worcester, said:

“This paper is a vital assessment of the state of Britain’s pollinator populations.  Understanding how different species are doing is the first step to focusing action on species most at need, and this study marks an essential move away from some previous studies which assessed pollinator trends by counting honeybee hives, towards examining trends in the 1500+ wild species which carry out the lion’s share of pollination.  The 715,392 biological records analysed make it the largest assessment ever undertaken for a major pollinator group in Britain. 

“This new analysis matches what we see in the long-term distribution trends for bumblebee species, with a small number of species doing well, and appearing in gardens, parks and farmland, but several species in decline.  Although 10% of all species analysed are increasing, including the Tree Bumblebee and Ivy Bee (both new to Britain in 2001 and rapidly spreading across the UK), the 33% of species in decline (largely the continuing decline of already-rare species) are cause for serious concern and provide clear evidence that we are not currently doing enough to support British wildlife.

“The paper rightly points out that studying species distribution and occupancy only tell half the story.  Abundance is the missing piece of the puzzle, and it is good to see a call for schemes carrying out standardised monitoring of pollinator abundance, such as the Bumblebee Conservation Trust’s BeeWalk scheme, active from 2010. Recent analysis of The BeeWalk data shows a different pattern – approximately half our bumblebee species are increasing in abundance, half decreasing. These results complement, rather than contradict, each other: abundance changes are a precursor to distributional changes.

“It is important to note that the datasets analysed only run to 2013.  This means that it doesn’t cover the hot summers of 2014 or 2018, or recent extreme weather events such as last winter’s Beast from the East.  It does mean that the dataset ends on two years which are likely to have been bad for bees and hoverflies: 2012 had the wettest summer since 1912, and 2013 the coldest spring since 1962: other flying insects, such as butterflies, did very poorly in both years regardless of underlying population trends. If the same is true of bees and hoverflies, the trend endpoints are likely to be lower than they might have been in more ‘normal’ years.

Prof Dave Goulson, Professor of Biology, University of Sussex, said:

“Previous studies have described declines in UK butterflies, moths, carabid beetles, bees and hoverflies. If further evidence were needed, this new study confirms that declines in insects are ongoing. The new study focusses on Britain’s hoverflies and wild bees, with about 350 species included. Overall, this new analysis suggests that roughly three times as many hoverflies and bees are declining as are increasing, with an overall net loss of species ranges of 2.7 million km2 in 32 years, equating to an average loss of about 11 species from each km2 across the UK since 1980.

“It is important to recognize that these data are not direct measures of changes in actual populations of these insects; such measures don’t yet exist.  Instead, the authors attempt to infer what is happening over time and at a national scale from ad hoc recording of sightings of species (over 700,000 records over 33 years), and what these tell us about the changing distributions of these species.

“Upland and northern species seem to have been worst hit, which the authors suggest might be the effects of climate warming. If so, then we can expect far more rapid declines of these species in the future, for climate change has barely got started.

“There is some good news in the study; the most common crop pollinators have, on average, shown small increases in range. This might be due to the sowing of wildflower strips by farmers, which we know boost numbers of some common bumblebee species, or it might simply be because much more oilseed rape is grown now than in 1980. It might also reflect the simplification of the landscape, so that the tougher, more generalist species such as the buff-tailed bumblebee come to dominate pollinator communities, while the more specialized species disappear. 

“The authors note that southern species seem to have declined more rapidly since 2006. This was the year that clothianidin was first used in the UK, a neonicotinoid insecticide which quickly became the most popular neonicotinoid used in British farming until it was banned in 2018. Most of the use of this chemical was in the southern and central lowlands of Britain.”

Prof Phil Stevenson, Senior Research Leader at the Royal Botanic Gardens, Kew, said:

“It is becoming increasingly apparent that pollinator diversity is critical for food production and also maintaining our natural landscapes ensuring reproduction in crops and non-food plants alike.  We cannot rely on honeybees to deliver this ecosystems service alone.   Evidence of the pollinator declines reported by Powney and colleagues are crucial in helping to support the development of environmental policy that safeguards these important ecosystem service providers.  It is particularly interesting to see that the generalist bumble bee Bombus terrestris is increasing in numbers too as this may be a signal of a reduction of plant diversity rather than a reduction of nectar and pollen resources as a whole.  Landscape management practise needs to focus on the diversity of floral resources to provision food for all pollinators providing a diversity of macro and micro nutrients and the necessary resources for specialist pollinators which are otherwise being pushed out of our increasingly uniform landscapes.”

Mr Norman Carreck, Laboratory of Apiculture and Social Insects, University of Sussex, said:

“It is widely recognised that bee populations around the world have been in decline for many years, but actually quantifying the declines is difficult, and organised monitoring schemes are very costly to implement. However, volunteer recording schemes have extremely valuable long term presence and absence data which can be used to quantify these declines. It can be concluded from this important study that there is no single cause of pollinator decline, although long term changes in land use are almost certainly the most important factor. Loss of rare habitats is clearly a key driver for the decline of rare species, but there is some cause for optimism that agri environment schemes may be improving the situation in agricultural land, but there is clearly much more to be done.”

Dr Roy van Grunsven, ecologist at the Dutch Butterfly Association, said:

“It’s an interesting article that does give insight in the trends in distribution of these important pollinators.  There is no standardised monitoring scheme for these groups, but the method used is valid for this type of data (opportunistically collected data) and gives reliable results.

“The results agree with what we know from moths and butterflies and from other countries. It is mostly bad news: species are overall declining in distribution and there is homogenisation of communities. The eusocial species are increasing (overall), this is likely a result of targeted efforts. This is however a very small group and a group that tends to be more general; many of the solitary bees and are relatively large and mobile with a larger home range. This also means that we rely more and more on a smaller set of pollinating species which is a risky development. 

“There are a few things that can influence the results, e.g. they could only use grid cells with enough data. As the data is gathered by volunteers these are likely the more interesting cells (often nature reserves and areas with management schemes?). It is therefore likely that the impact is underestimated.

“Furthermore there are two important considerations. The authors describe the decline in distributions since 1980. This is not the overall decline in these species. First of all, most of the decline has probably happened before 1980, and it wasn’t all peachy back then. There is very little data before 1980 so it’s a reasonable starting point and refraining from drawing conclusions on the period before is the purest way to discuss this. The second point is that it looks at distribution, occurrence in grid cells. This means that a decline in abundance is not recorded as long as a species still occurs. Going from flowery meadows full of bees to intensive agriculture with a few individuals of those species in a road verge does not result in a change in distribution but of course is a huge change in pollination services. 

“It is very likely that the number of bees and hoverflies declined a lot more than the distribution. The 25% decline in distribution in bees and 24% in hoverflies since 1980 are worrying but likely reflect a much stronger decline in numbers.

“Overall I think it is a valid and valuable study that confirms what many suspect.”

‘Widespread losses of pollinating insects in Britain’ by Gary D. Powney et al. was published in Nature Communications at 4pm UK TIME on Tuesday 26 March 2019.

Declared interests

Dr Dicks: I work in partnership with the following companies and NGOs. Those that my research group has actually received financial support from in the last 10 years are marked with an asterisk.

Waitrose*

Cool Farm Alliance

Primafruit Ltd

Labrunier Group (Brazil)

Fauna and Flora International

UNEP-World Conservation Monitoring Centre

Bumblebee Conservation Trust*

Highway England*

Hutchinsons Ltd*

BerryWorld Group*

Dr Comont – My work at the Bumblebee Conservation Trust involves working with the Centre for Ecology & Hydrology on the ongoing UK Pollinator Monitoring Scheme, and some of my biological records are likely to have made it into the dataset, but I have had no involvement with this paper.

Mr Carreck: Nothing to declare.

No others received.