Content uploaded by Pascal Malkemper
Author content
All content in this area was uploaded by Pascal Malkemper on Aug 15, 2019
Content may be subject to copyright.
Discussion
Risk to pollinators from anthropogenic electro-magnetic radiation
(EMR): Evidence and knowledge gaps
Adam J. Vanbergen
a,e,
⁎, Simon G. Potts
b
, Alain Vian
c
, E. Pascal Malkemper
d
,
Juliette Young
a,e
,ThomasTscheulin
f
a
Agroécologie, AgroSup Dijon, INRA, Univ. Bourgogne Franche-Comté, F-21000 Dijon, France
b
Centre for Agri-Environmental Research, School of Agriculture, Policy and Development, Reading University, RG6 6AR, UK
c
IRHS, Université d'Angers, Agrocampus-Ouest, INRA, SFR 4207 QuaSaV, 49071 Beaucouzé, France
d
Research Institute of Molecular Pathology (IMP), Campus-Vienna-BioCenter 1, 1030 Vienna, Austria
e
Centre for Ecology & Hydrology, Bush Estate, Penicuik, Edinburgh EH26 0QB, UK
f
Department of Geography, University of the Aegean, University Hill, GR-81100, Greece
HIGHLIGHTS
•Anthropogenic electromagnetic radia-
tion (light, radiofrequency) is perceived
to threaten pollinators and biodiversity.
•Potential risks are artificial light at night
(ALAN) and anthropogenic radiofre-
quency electromagnetic radiation
(AREMR).
•We assessed the quantity and quality of
evidence, and the level of consensus, to
distil key messages for science and pol-
icy.
•ALAN can alter pollinator communities
and functions, although this remains to
be well established.
•Evidence of AREMR impactsis inconclu-
sive due to a lack of high quality, field-
realistic studies.
•Whether pollinators and pollination
face a threat from the spread of ALAN
or AREMR remains a major knowledge
gap.
GRAPHICAL ABSTRACT
abstractarticle info
Article history:
Received 25 June 2019
Received in revised form 2 August 2019
Accepted 6 August 2019
Available online 07 August 2019
Editor: Henner Hollert
Worldwide urbanisation and use of mobile and wireless technologies (5G, Internet of Things) is leading to the
proliferation of anthropogenic electromagnetic radiation (EMR) and campaigning voices continue to call for
the risk to human health and wildlife to be recognised. Pollinators provide many benefits to nature and human-
kind, but facemultiple anthropogenic threats.Here, we assess whether artificial light at night(ALAN) and anthro-
pogenic radiofrequency electromagnetic radiation (AREMR), such as used in wireless technologies (4G, 5G) or
emitted frompower lines, represent an additional and growing threat to pollinators. A lack of high quality scien-
tific studiesmeans that knowledge of the risk to pollinators from anthropogenic EMR is either inconclusive, un-
resolved, or only partly established. A handful of studies provide evidence that ALAN can alter pollinator
communities, pollination and fruit set. Laboratory experiments provide some, albeit variable, evidence that the
Science of the Total Environment 695 (2019) 133833
⁎Corresponding author at: Agroécologie, AgroSup Dijon, INRA, Univ. Bourgogne Franche-Comté, F-21000 Dijon, France.
E-mail addresses: adam.vanbergen@inra.fr (A.J. Vanbergen), s.g.potts@reading.ac.uk (S.G. Potts), alain.vian@univ-angers.fr (A. V ian), pascal.malkemper@imp.ac.at (E.P. Malkemper),
jyo@ceh.ac.uk (J. Young), t.tscheulin@geo.aegean.gr (T. Tscheulin).
https://doi.org/10.1016/j.scitotenv.2019.133833
0048-9697/© 2019 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
honey bee Apismellifera and other invertebrates can detect EMR, potentially using it for orientation ornavigation,
but they do not provide evidence that AREMR affects insect behaviour in ecosystems. Scientifically robust evi-
dence of AREMRimpacts on abundance or diversity of pollinators (or other invertebrates) are limited to a single
study reporting positiveand negative effectsdepending on the pollinator groupand geographical location. There-
fore, whether anthropogenic EMR (ALAN or AREMR) poses a significant threat to insect pollinatorsand the ben-
efits they provide to ecosystems and humanity remains to be established.
© 2019 Elsevier B.V. All rights reserved.
Keywords:
Pollinators
Invertebrates
Electromagnetic
Anthropogenic EMR
ALAN
EKLIPSE
1. Pollinators and pollination under threat
Global insect biodiversity is under threat from multiple anthropo-
genic drivers of global environmental change, which in turn jeopardize
the many benefits people obtain from this component of nature
(Hallmann et al., 2017;IPBES, 2016;Potts et al., 2016;Sánchez-Bayo
and Wyckhuys, 2019). Insect pollinators are particularly high on the sci-
ence and policy agenda worldwide and there exists a comparatively
strong evidence base on their values, status and trends, and the threats
they face (IPBES, 2016). Known pressures impacting pollinators and
pollination services include land-use change, intensive agricultural
(and other land) management, use and misuse of pesticides, climate
change, pests and pathogens, alien invasive species and potentially in-
teractions between these different drivers (Brown et al., 2016;IPBES,
2016;Vanbergen et al., 2013).
Environmental pollution presents a further potential risk to pollina-
tors and pollination although its impact is much less studied according
to the Intergovernmental Science-Policy Platform on Biodiversity and
Ecosystem Services
1
(IPBES, 2016). One form of pollution that presents
a potential risk to wildlife and that has grown significantly since the
mid-20th century (Fig. 1) is the global spread of anthropogenic electro-
magnetic radiation (EMR: radio waves, microwaves, infrared, visible
light, ultraviolet, X-, and gamma radiation) (Balmori, 2015;Bandara
and Carpenter, 2018;Grubisic et al., 2018;Russell, 2018). With a focus
on human health, the WHO has evaluated the risk (e.g. the International
EMF project since 1996
2
) from non-ionizing anthropogenic EMR (up to
300 GHz). Although finding no major public health concerns, the WHO
acknowledges uncertainties around chronic exposure and new
technologies
2
. Currently, neither the WHO nor the OECD with its eco-
nomic and development focus (OECD, 2012), have considered the cur-
rent or future indirect risks from anthropogenic EMR to the natural
environment, which provides diverse values and benefits to
humankind.
The global proliferation of both artificial light at night (ALAN) and
anthropogenic radiofrequency electromagnetic radiation (AREMR)
utilised in mobile and smart wireless technologies (Fig. 1) is ongoing
with increasing urbanisation and the worldwide launch of next genera-
tion wireless technologies e.g. 5G, smart grids and the ‘Internet of
Things’(Bandara and Carpenter, 2018;Bin Zikria et al., 2018;
Macgregor et al., 2015;Russell, 2018). What remains unclear is the ex-
tent that ALAN or AREMR represent a threat to insect pollinators and
the benefits they provide to nature and humankind (Fig. 2).
2. Artificiallightatnight(ALAN)
The potential risk from ALAN to nocturnal pollinators and pollina-
tion was noted by the academic community (Macgregor et al., 2015)
and mentioned in the IPBES assessment of pollinators and pollination
as “a driver clearly affecting nocturnal species and growing in importance
due to urbanization”. The IPBES also noted that compared to other
drivers “Its effect is still scarcely studied”and called for further studies
“to evaluate the extent of light pollution effects on nocturnal pollinators”
(IPBES, 2016). Since this global assessment, further studies have been
published that show how ALAN can disrupt pollinator communities
and plant reproduction. Knop et al. (2017) demonstrated by field exper-
iment how artificial light modified the architecture of nocturnal plant-
pollinator communities and reduced visitation rates to plants by 62%
leading to a 13% drop in the fruit set of a focal plant species (Cirsium
oleraceum, Asteraceae) (Knop et al., 2017). Other recent studies have
demonstrated how artificial street lighting reduces local abundance
and species richness of moths and their rates of feeding and pollen
transport (Grubisic et al., 2018;Macgregor et al., 2017;van
Langevelde et al., 2017). Together such studies illustrate the potential
for ALAN to modify pollinator foraging and pollination function in
ways that have the capacity to jeopardize pollinator populations, either
directly or indirectly. Moreover, analysis of combined nocturnal and di-
urnal plant-pollinator networks suggested that these nocturnal impacts
of light pollution transmitted to the diurnal network through trophic in-
teractions connecting nocturnal to diurnal species (Knop et al., 2017).
This highlights the potential for the effects of ALAN to exacerbate the
overall anthropogenic pressure on diurnal pollinators (Potts et al.,
2016;Vanbergen et al., 2013). Therefore, there is emerging, albeit in-
complete, evidence that ALAN represents a potentially growing impact
on pollinators and pollination as global urbanisation proceeds apace
(Figs. 2 & 3).
3. Anthropogenic radiofrequency electromagnetic radiation
(AREMR)
Campaigning voices continue to perceive and call for the threat from
anthropogenic EMR to both human health
3
and wildlife
4
to be
recognised and evaluated. Aside from ALAN, the IPBES report (IPBES,
2016;Potts et al., 2016) did not consider other sources and wavelengths
of anthropogenic EMR. This was because it was judged at that time
(publications up to July 2015) there was insufficient data for an evi-
dence assessment, with only a few studies showing how bees utilise
magnetic fields (Clarke et al., 2013;Gould et al., 1978;Hsu and Li,
1994) and fewer still on potential effects of AREMR (Favre, 2011;
Greenberg et al., 1981). Similarly, AREMR (and ALAN) were not identi-
fied as a risk in a 2016 horizon scan of future threats and opportunities
for pollinators and pollination (Brown et al., 2016). Moreover, likely due
to the continuing lack of scientific publications, anthropogenic EMR as a
driver of biodiversity change remained unassessed by the IPBES during
its most recent regional (2018) and global (2019) assessments.
5
How-
ever, a 2018 horizon scan focussed on biodiversity conservation, natural
capital and ecosystem services pointed to the potential, but unstudied,
risk to wildlife of non-ionizing radiation from 5G mobile phones and
wireless transmission infrastructure (Sutherland et al., 2018).
Therefore, the perception remains that AREMR (in addition to ALAN)
poses a current and growing ris k to pollinators and pollination (Balmori,
1
https://www.ipbes.net/
2
https://www.who.int/peh-emf/project/en/
3
https://www.emfscientist.org/
4
https://www.buglife.org.uk /news-and-even ts/news/could-our-obsession-with-
mobile-technology-destroy-wildlife
5
https://www.ipbes.net/assessment-reports
2A.J. Vanbergen et al. / Science of the Total Environment 695 (2019) 133833
2015). Studies have shown the honey bee (Apis mellifera)isabletode-
tect magnetic fields physiologically (Gould et al., 1978;Hsu and Li,
1994;Kirschvink and Kirschvink, 1991;Lambinet et al., 2017;Liang
et al., 2016) and potentially use this capacity for orientation, navigation
and foraging (Fig. 2). Furthermore, bees use electric fields of the same
magnitude as commonly encountered AREMR for intraspecific(within
hive) and interspecific (plant-pollinator) communication,in the context
of foraging on floral resources (Clarke et al., 2013;Greggers et al., 2013).
Therefore, there is the possibility that AREMR could disrupt these phys-
iological functions, ultimately affecting bee health and survival.
In October 2016, the EU (H2020) EKLIPSE
6
project (Watt et al.,
2019), in response to a request from the UK charity Buglife, organised
a foresight activity identifying the state of and gaps in knowledge
concerning the emerging issue of anthropogenic EMR (excluding
ALAN) impacts on wildlife. This involved a scientific literature search
(ISI Web of Knowledge & Google Scholar, completed by July 2017) to
gather a representative, but not exhaustive, set of relevant peer-
reviewed papers published from 2000 onwards, coincident with the
onset of the proliferation of mobile technologies (Fig. 1) (details in
Malkemper et al., 2018). A scientific expert working group was con-
vened to assess this evidence (Malkemper et al., 2018) and, following
a stakeholder web conference, a summary of the current evidence and
knowledge gaps was produced (Goudeseune et al., 2018). Here we sum-
marise and update these reports with assessment of new studies to un-
derstand what the evidence base is for a risk to pollinators and
pollination from the global spread of AREMR.
The EKLIPSE report confirmed the sparseness of the literature and
scarcity of data regarding anthropogenic EMR impacts on wildlife
(Malkemper et al., 2018). Of an initial 147 scientificpapersorreviews
identified, further scrutiny of their relevance to the topic of anthropo-
genic EMR emissions and its effect (or lack of) on wildlife reduced the
list of citations to 82 papers (97 including reviews). These included 39
studies on various invertebrate groups including Drosophila fruit flies,
beetle or ant species, but also a managed pollinator species
(A. mellifera) and, in a few cases, wild pollinator communities
(Malkemper et al., 2018). Each study was assessed and scored according
to their scientific and technical quality (0 = irrelevant or very poor
6
EKLIPSE isfunded to develop a new, self-sustaining support mechanism for evidence-
based and evidence-informed policy on biodiversity and ecosystem services by assessing
knowledge,research gapsand emerging issues in response to requestsfrom policymakers,
civil society and science actors. http://www.eklipse-mechanism.eu/
Fig. 1. Typical maximum daily exposure to radiofrequency electromagnetic radiation from anthropogenic and natural power flux densities in comparison with International Commission
on Non-Ionizing Radiation Protection (ICNIRP) safety guidelines. Sources of anthropogenic radiofrequency electromagnetic radiation levels are illustrated for different periods.
Source: Bandara P, Carpenter DO. Planetary electromagnetic pollution: it is time to assess its impact. The Lancet Planetary Health 2018; 2: e512–e514. Reproduced under license from
Elsevier.
3A.J. Vanbergen et al. / Science of the Total Environment 695 (2019) 133833
Fig. 2. The level of scientific knowledge about the impact on pollinators and pollination of natural (a) and anthropogenic (b: ALAN; c: mobile; d: electrical infrastructure) sources of
electromagnetic radiation. Based on the available evidence from journal publications, the impact on different aspects of pollinator biology and pollination services are assessed as being
positive, negative, neutral or variable (idiosyncratic or contrasting). The level of confidence (quantity, quality and consensus) in this evidence is expressed according to the four-box
model adopted from the IPBES (see Fig. 3 for details).
Fig. 3. Position of key messages in relation to the level of confidence (quantity,quality and consensus) in the evidence base using a four-box model for the qualitative communication of
certainty (IPBES, 2016). Confidence increasestowards the top-rightcorner as shown by the increased strengthof shading. Terms are: Wellestablished - comprehensive meta-analysis or
other synthesis or multiple independent studies that agree; Established but incomplete - general agreement although only a limited number of studies exist but no comprehensive
synthesis and/or th e studies imprecisely address the question; Unresolved - multiple independent studies exist but conclusions do not agree; In conclusive - limited evidence,
recognising major knowledge gaps.
4A.J. Vanbergen et al. / Science of the Total Environment 695 (2019) 133833
science; 1 = minimal with only some utility; 2 = medium - some lim-
itations or caveats, 3 = excellent). Here, we also distilled key messages
for scientists and decision-makers to evaluate our assessment of the
published evidence and the level of the potential problem. To communi-
cate the level of certainty in knowledge, we attached a degree of confi-
dence to each key message using a qualitative ‘four-box model’
(adopted from the IPBES, 2016) that shows the assessment of the quan-
tity, quality and level of expert consensus on the evidence (Fig. 3).
3.1. Acute exposure to EMR in laboratory experiments
The highest quality research (score = 3) assessed in the EKLIPSE re-
port were laboratory experiments testing the fundamental biological
responses of the bumblebeeBombus terrestrisor ot her model insect spe-
cies (i.e. Drosophila flies, locusts, plant-hoppers) to naturally occurring
electromagnetic fields and their exclusion or experimentally-imposed
electromagnetic treatments closely mimicking nature. Thesefew scien-
tifically rigorous laboratory experiments showed how insects can detect
and may orientate using electromagnetic fields and the effects, or lack
of, on behaviour, cell development, and physiological function (Bae
et al., 2016;Sutton et al., 2016;Tomanová and Vácha, 2016;Wan
et al., 2014)(Fig. 2). Of these experiments, there was little evidence of
exposure to EMR leading to damage or effects on individual develop-
ment, or reproduction in these model invertebrate species (Bae et al.,
2016;Wan et al., 2014;Wyszkowska et al., 2016;Zhang et al., 2016).
The most convincing finding was, as with birds (Engels et al., 2014)
and mammals (Malkemper et al., 2015), that the magnetic sense of in-
vertebrates appears to be affected by AREMR in the MHz-range
(Tomanová and Vácha, 2016;Vácha et al., 2009), although the extent
to which insects are dependent on their magnetic sense for successful
navigation remains unknown. Crucially, whilst providing some mecha-
nistic basis for testing hypothesis on EMR impacts, these laboratory
studies do not provide any evidence about impacts of AREMR on inver-
tebrates in ecosystems. Moreover, the effects observed tended to be
complex, variable in direction or effect size, and onlysometimes adverse
(Fig. 2).
Since publication of the EKLIPSE report, Shepherd et al. (2018)
produced an appropriately controlled and analysed set of laboratory
experiments testing the impact of experimental AREMR treatments
on the cognitive and motor abilities of the honey bee (Apis mellifera).
They exposed honey bees to extremely low frequency electromag-
netic fields (50 Hz) replicating modelled estimates of fields gener-
ated by overhead power transmission cables at ground level (range
20–100 μT EMF) or proximate to the conductor (7000 μTEMF)
(Shepherd et al., 2018). Acute laboratory exposure to levels of 20 to
N100 μT EMF had a clear negative impact on worker bee learning
and memory as measured with a standard test (Proboscis Extension
Reflex) (Shepherd et al., 2018). An increase in wingbeat frequency in
tethered flight experiments was seen at the highest EMF (7000 μT),
but because pollinators mostly forage at ground level it is unlikely
that bees or other pollinators would routinely flycloseenoughto
conductors to be exposed to such levels. There were also significant,
but relatively weak, deterrence effects on the flight rate and number
of feeding worker bee foraging on a sugar source at 100 μT, which is
at the higher end of levels that pollinators would encounter foraging
in ground vegetation (Shepherd et al., 2018). These effects on bee
cognition and behaviour are in response to experimental EMF fields
that exceed predicted field-realistic exposure at ground level (5–15
μT) (Burda et al., 2009) where the bulk of bee activity occurs. None-
theless, this study provides a basis for future research to conduct
semi-field experiments that corroborate and extend this line of en-
quiry by adding more biological realism.
Hitherto, laboratory investigations into thephysiological or develop-
mental responses of invertebrates focussed on short-term or acute ex-
posure to experimental sources of EMR. No studies have examined
effects on invertebrates of long-term or chronic exposure to sources of
anthropogenic EMR (Malkemper et al., 2018). This is potentially a
more realistic type of exposure and may reveal sub-lethal effects as ob-
served in pesticide impacts on pollinators (reviewed in: Godfray et al.,
2014;Godfray et al., 2015;IPBES, 2016). Experiments that examine
the potential multifactorial interplay (Vanbergen et al., 2013)between
EMR exposure and other environmental stressors (e.g. pathogens, envi-
ronmental pollutants or chemicals) (IPBES, 2016) affecting pollinator
health and reproduction would also be valuable.
3.2. Field and semi-field experiments and surveys
Overall, the EKLIPSE assessment found that there is a dearth of
scientifically robust evidence of EMR impacts on invertebrates from
field or semi-field situations. Most available field studies conducted
to date are dominated by deeply flawed investigations (scored 0 or
1) that are zero or under-replicated or anecdotal and consequently
provide little meaningful data on which to judge the risk of exposure
to anthropogenic EMR (Malkemper et al., 2018). There were a very
restricted number of more robust field studies albeit with some re-
maining problems associated with scale and scope (score = 2). A
single honey bee hive experiment provided some indication that
very close proximity to AREMR (900 MHz) can affect honey bee col-
ony acoustic behaviour (worker bee piping) associated with
swarming or disturbance (Favre, 2011). Whilst this experiment
(score = 2) had some design strengths (included negative and
sham controls) it was, given the natural variability of honey bee col-
onies, under-replicated with only five colonies tested (although
therewere12runs)(Favre, 2011). Another more recent honey bee
experiment (random frames of bee larvae from eight colonies split
and assigned to AREMR (925–960 MHz) or control treatments with
two runs) showed that exposure increased mortality during pupa-
tion and reduced hatching rate of the new queens, however, this
did not translate into reduced subsequent mating success or colony
size (Odemer and Odemer, 2019). Importantly, the AREMR exposure
in these studies was not field-realistic because the emission source
was from mobile phones placed inside the hives (Favre, 2011;
Odemer and Odemer, 2019). This was an acute and highly artificial
level of exposure, acknowledged as a worst case scenario by
Odemer and Odemer (2019), and a feature shared with studies eval-
uated as being of the poorest scientificquality(score=0or
1) (Malkemper et al., 2018).
In field realistic settings, a well-executed and robustly analysed
survey (score = 2) of wild pollinator communities around 10 mobile
phone antennas using high frequencies (800–2600 MHz) and dis-
tributed across two Aegean islands revealed complex effects on in-
sect abundance (Lázaro et al., 2016). This analysis revealed a
complex correlation between the variable anthropogenic electric
field (range 0.01–0.67 V·m
−1
) and insect abundance, measured at
distance intervals (50, 100, 200 and 400 m) from the antenna, but
these were contingent on the insect taxon and sometimes varied
with geographical (island) location. Greater exposure to EMR was
related negatively (hoverflies; wasps), positively (underground
nesting wild bees and bee flies) or uncorrelated (butterflies) to
abundance and had no effect on species richness (wild bees,
hoverflies) (Lázaro et al., 2016).
Another similar field study (Vijver et al., 2014) of four phylogeneti-
cally distant invertebrate taxa (Collembola, Heteroptera, Hymenop-
teran parasitoid and Drosophila melanogaster) found no effects of
controlled (Faradaycages, blind recording) exposure to EMR from a mo-
bile antenna on reproductive capacity subsequently measured in the
laboratory. Caveats to this experiment (score = 2) are that the field ex-
posure was very brief (48 h) and whilst well replicated at the observa-
tion level (number of caged individuals) it was un-replicated at the
treatment level (single exposure site) limiting its capacity for
generalisation.
5A.J. Vanbergen et al. / Science of the Total Environment 695 (2019) 133833
4. Overall quality of studies of AREMR effects on pollinators and
other invertebrates
Overall, of the primary research reviewed the highest quality studies
(score = 3; 23% of the papers) reported on the fundamental nature
of interactions between invertebrates and naturally occurring electro-
magnetic fields. Such studies were always laboratory based, well repli-
cated and controlled (Malkemper et al., 2018). The next tier of studies
(score = 2; 38%) mostly comprised laboratory studies focussed on an-
thropogenic EMR, such as frequencies or wavelengths produced by mo-
bile phone masts. These were very mixed with respect to scientific
quality: sometimes replication appeared at a reasonable and appropri-
ate level, but often a lack or underreporting of the design, replication
or methods meant that the study could not be evaluated properly
(Malkemper et al., 2018). The few field studies in this tier of study
(score = 2; 38%), reported negligible or contrasting effects on behav-
iour or abundance. The remaining field and laboratory studies (score
= 0 or 1; 39%) were anecdotal or flawed from the perspective of scien-
tific design, such as having very low or non-existent levels of replication,
pseudoreplication, highly unrealistictreatments, or sometimes a combi-
nation of all flaws (Malkemper et al., 2018). Consequently, no meaning-
ful information can be gleaned from such studies.
5. Key messages and evidence confidence ratings
•ALAN can modify nocturnal pollinator communities and foraging be-
haviour to change pollination and plant reproduction [Figs. 3 & 2:
established but incomplete]. These nocturnal changes may transmit to
and alter diurnal pollinator communities and pollination via species
interaction networks [Fig. 3:established but incomplete]. ALAN will in-
crease in prominence asa driver of change as global urbanisation pro-
ceeds [Fig. 3:established but incomplete].
•Naturally occurring EMR is detectable by invertebrate physiological
mechanisms governing orientation or movement [Figs. 3 & 2:
established but incomplete]. AREMR has the potential to effect such
physiological mechanisms, but robust scientific evidence is currently
equivocal or lacking [Figs. 3 & 2:inconclusive].
•The limited number of well-executed laboratory experiments show
that EMR can affect behaviour or reproduction of the honey bee Apis
mellifera and other model insect species (e.g. Drosophila
melanogaster), but effects are variable, negligible or inconsistent be-
tween studies [Figs. 3 & 2:unresolved] and do not necessarily translate
into AREMR impacts on pollinators (or other invertebrates) in ecosys-
tems.
•Current evidence of impacts of AREMR on pollinators (or other inver-
tebrates) from field- or semi-field studies is very limited with only a
single scientifically robust ecological study of impacts on wild pollina-
tors (Lázaro et al., 2016), which reported positive and negative effects
depending on the pollinator group and geographical location [Figs. 3 &
2:inconclusive].
6. Future research recommendations
To evaluate properly the potential threat to pollinators and other in-
vertebrates from exposure to anthropogenic EMR requires further re-
search. In particular, new research is required to assess the hitherto
unstudied effects on pollinators and other biodiversity of emerging
AREMR technologies and infrastructure (e.g. 5G, ‘Internet of Things’)
(Bandara and Carpenter, 2018;Bin Zikria et al., 2018;Russell, 2018).
Importantly the overall scientific quality of investigations must im-
prove if we are to obtain a realistic picture of the level of risk
(Makinistian et al., 2018). Future studies must be hypothesis driven,
based on a sound theoretical framework that allows for testable predic-
tions of the experimental or survey outcomes. Good study design is
obviously essential, but seemingly overlooked in many instances. For
example, replication at the level of the treatment (i.e. source of expo-
sure) is necessary for effects to be generalised and appropriate controls
are essential or, where not possible in some field situations, otherwise
ensuring sufficient replication to allow statistical partitioning of effects
of measured EMR levels from other correlates. Studies must also report
sufficiently detailed technical information (e.g. EMR wavelength, fre-
quency and duration of exposure), instruments or methods used in ex-
periments, and any environmental covariates (e.g. weather) to ensure
reproducibility, comparability and to facilitate future syntheses
(Makinistian et al., 2018).
Studies must maximise the level of biological or ecological realism.
As with research into pesticide impacts on pollinators (Godfray et al.,
2014;Godfray et al., 2015;IPBES, 2016), exposure to anthropogenic
EMR must be field-realistic mimicking accurately and not exceeding
wavelengths and frequencies encountered by pollinators in the field.
As seen with other driver combinations (Godfray et al., 2014;
González-Varo et al., 2013;Vanbergen et al., 2013), assessments of
chronic exposure and the potential for additive or synergistic effects
arising from exposure to single or multiple sources of ALAN/AREMR or
in combination with other stressors (e.g. pesticides, pathogens, nutri-
tional deficits) need testing to evaluate the overall levelof risk from an-
thropogenic EMR. To understand field-realistic exposure and effects
also requires consideration of pollinator species traits, such as nesting
habits, foraging or dispersal behaviour and sociality, that will govern
the level of impact of different sources of anthropogenic EMR alongside
other drivers (e.g. land-use) (Potts et al., 2016;Vanbergen et al., 2013).
Measuring pollinator responses to EMR exposure at different levels of
biological organisation (species, population, community) and resulting
change in pollination services (plant reproduction or crop yield) over
the longer-term and, ideally, pre- and post-exposure would be espe-
cially valuable. For such field-realistic studies to be conclusive, would
require their implementation across geographical regions in different
semi-natural or anthropogenic ecosystems.
Interdisciplinary collaborations bringing together engineers, physi-
cists, ecotoxicologists and biologists to test hypotheses about biological
impacts at field-realistic exposure are more likely to avoid the concep-
tual or technical pitfalls that confound studies (Makinistian et al.,
2018) and provide insights into the potential effects of anthropogenic
EMR on pollinators and other insects.
7. Conclusion
Anthropogenic EMR emissionsare proliferatingbut the extent that it
is a risk to pollinators and pollination is currently unclear. There is some
recently published evidence that ALAN may alter pollinator communi-
ties and functions, although there is a need for further high quality stud-
ies whose results align before we can conclude that ALAN is a major
driver of change in pollinators and pollination. Our current knowledge
of the impact of AREMR on pollinators (and other invertebrates) is in-
conclusive or unresolved and hindered by the scarcity of high quality
scientific studies. Most experiments and field studies suffered from
poor scientific method, underreporting of scientificortechnicaldetails
that obstruct their assessment, and use of highly unrealistic exposure
to AREMR sources. The extent that anthropogenic EMR (ALAN or
AREMR) represents a significant threat to insect pollinators and the
benefits they provide to nature and humankind therefore remains to
be clearly established.
Acknowledgements
This article arose from an EKLIPSE foresight activity (EKLIPSE grant
agreement number 690474, European Union's Horizon 2020 research
and innovation programme) following a request by the UK charity
Buglife (The Invertebrate Conservation Trust https://www.buglife.org.
uk/) to produce an evidence assessment relating to the impacts of
6A.J. Vanbergen et al. / Science of the Total Environment 695 (2019) 133833
Electromagnetic Radiation (EMR) on invertebrates and other wildlife.
Thanks to Lise Goudeseune (Belgian Biodiversity Platform) and Estelle
Balian (http://fea-l.eu/contact) for their work in searching the scientific
literature, coordinating and organising the expert group, stakeholder
conferences and reporting.
References
Bae, J.E.,Bang, S., Min, S., Lee,S.H., Kwon, S.H., Lee, Y., et al.,2016. Positive geotactic behav-
iors induced by geomagnetic field in Drosophila. Mol Brain 9, 55.
Balmori, A., 2015. Anthropogenic radiofrequency electromagnetic fields as an emerging
threat to wildlife orientation. Sci. Total Environ. 518-519, 58–60.
Bandara,P., Carpenter, D.O., 2018. Planetary electromagnetic pollution: it is time to assess
its impact. The Lancet Planetary Health 2, e512–e514.
Bin Zikria, Y., Kim, S.W., Afzal, M.K., Wang, H.X., Rehmani, M.H., 2018. 5G Mobile services
and scenarios: challenges and solutions. Sustainability 10.
Brown, M.J.F., Dicks, L.V., Paxton, R.J., Baldock, K.C.R., Barron, A.B., Chauzat, M.-P., et al.,
2016. A horizon scan of future threats and opportunities for pollinators and pollina-
tion. PeerJ 4, e2249.
Burda, H.,Begall, S., Cerveny,J., Neef, J., Nemec, P.,2009. Extremely low-frequency electro-
magnetic fields disruptmagnetic alignment of ruminants. Proc. Natl. Acad.Sci. U. S. A.
106, 5708–5713.
Clarke, D., Whitney,H., Sutton, G., Robert, D., 2013.Detection and learning of floral electric
fields by bumblebees. Science 340, 66–69.
Engels, S., Schneider, N.L., Lefeldt, N., Hein, C.M., Zapka, M., Michalik, A., et al., 2014. An-
thropogenic electromagnetic noise disrupts magnetic compass orientation in a mi-
gratory bird. Nature 509, 353.
Favre, D., 2011. Mobile phone-induced honeybee worker piping. Apidologie 42, 270–279.
Godfray, H.C.J., Blacquière, T., Field, L.M., Hails, R.S., Petrokofsky, G., Potts, S.G., et al., 2014.
A restatement of the natural science evidence base concerning neonicotinoid insecti-
cides and insect pollinators. Proc. R. Soc. B Biol. Sci. 281.
Godfray, H.C.J., Blacquière, T., Field, L.M., Hails, R.S., Potts, S.G., Raine, N.E., et al., 2015. A
restatement of recent advances in the natural science evidence base concerning
neonicotinoid insecticides and insect pollinators. Proc. R. Soc. Lond. B Biol. Sci. 282.
González-Varo, J.P., Biesmeijer, J.C., Bommarco, R., Potts, S.G., Schweiger, O., Smith, H.G., et
al., 2013. Combined effects of global change pressures on animal-mediated pollina-
tion. Trends Ecol. Evol. 28, 524–534.
Goudeseune, L., Balian, E., Ventocilla, J., 2018. The Impacts of Artificial Electromagnetic Ra-
diation on Wildlife (Flora and Fauna). Report of the web conference. A report of the
EKLIPSE project. .
Gould, J.L., Kirschvink, J.L., Deffeyes, K.S., 1978. Bees have magnetic remanence. Science
201, 1026–1028.
Greenberg, B., Bindokas, V.P., Frazier, M.J., Gauger, J.R., 1981. Response of honey bees, Apis
mellifera L, to high-voltage transmission lines. Environ. Entomol. 10, 600–610.
Greggers, U., Koch, G.,Schmidt, V., Durr,A., Floriou-Servou,A., Piepenbrock,D., et al., 2013.
Reception and learning of electric fields in bees. Proc. R. Soc. B Biol. Sci. 280.
Grubisic, M., van Grunsven, R.H.A., Kyba, C.C.M., Manfrin, A., Holker, F., 2018. Insect de-
clines and agroecosystems: does light pollution matter? Ann. Appl. Biol. 173 ,
180–189.
Hallmann,C.A., Sorg, M., Jongejans, E., Siepel, H., Hofland,N.,Schwan,H.,etal.,2017.More
than 75 percent decline over 27 years in totalflying insect biomass in protected areas.
PLoS One 12, e0185809.
Hsu, C.-Y., Li, C.-W., 1994. Magnetoreception in honeybees. Science 265, 95–97.
IPBES, 2016. In: Potts, S.G., Imperatriz-Fonseca, V.L., Ngo, H.T. (Eds.), The Assessment Re-
port of the Intergovernmental Science-Policy Platform onBiodiversity and Ecosystem
Services on Pollinators, Pollination and Food Production. Secretariat of the Intergov-
ernmental Science-Policy Platform on Biodiversity and Ecosys tem Services, Bonn,
Germany.
Kirschvink, J.L., Kirschvink, A.K., 1991. Is geomagnetic sensitivity real - replication of the
Walker-Bitterman magnetic conditioning experiment in honey bees. Am. Zool. 31,
169–185.
Knop, E., Zoller, L., Ryser, R., Erpe, C.G., Horler, M., Fontaine, C., 2017. Artificial light at
night as a new threat to pollination. Nature 548, 206.
Lambinet, V., Hayden, M.E., Reigl, K., Gomis, S., Gries, G., 2017. Linking magnetite in the
abdomen of honey be es to a magnetoreceptive function. Proc . R. Soc. B Biol. Sci.
284, 20162873.
van Langevelde, F., van Grunsven, R.H.A., Veenendaal, E.M., Fijen, T.P.M., 2017. Artificial
night lighting inhibits feeding in moths. Biol. Lett. 13.
Lázaro, A., Chroni, A., Tscheulin, T., Devalez, J., Matsoukas, C., Petanidou, T., 2016. Electro-
magneticradiation of mobiletelecommunicationantennas affects theabundance and
composition of wild pollinators. J. Insect Conserv. 20, 315–324.
Liang, C.-H., Chuang, C.-L., Jiang, J.-A., Yang, E.-C.,2016. Magnetic sensing through the ab-
domen of the honey bee. Sci. Rep. 6, 23657.
Macgregor, C.J., Pocock, M.J.O., Fox, R., Evans, D.M., 2015. Pollination by nocturnal Lepi-
doptera, and the effects of light pollution: a review. Ecological Entomology 40,
187–198.
Macgregor, C.J., Evans, D.M., Fox, R., Pocock, M.J.O., 2017. The dark side of street lighting:
impacts on mothsand evidence for the disruption of nocturnal pollen transport. Glob.
Chang. Biol. 23, 697–707.
Makinistian, L., Muehsam, D.J., Bersani, F., Belyaev, I., 2018. Some recommendations for
experimental work in magnetobiology, revisited. Bioelectromagnetics 39, 556–564.
Malkemper, E.P., Eder, S.H.K., Begall, S., Phillips, J.B., Winklhofer, M., Hart, V., et al., 2015.
Magnetoreception in the wood mouse (Apodemus sylvaticus): influence of weak
frequency-modulated radio frequency fields. Sci. Rep. 5.
Malkemper, E.P., Tscheulin, T., Vanbergen, A.J., Vian, A., Balian, E., Goudeseune, L., 2018.
The impacts of artificial electromagnetic radiation on wildlife (flora and fauna). Cur-
rent knowledge overview: A background document to the web conference. A Report
of the EKLIPSE Project http://www.eklipse-mechanism.eu/.
Odemer, R., Odemer, F., 2019. Effects of radiofrequency electromagnetic radiation (RF-
EMF) on honey bee queen devel opment and matin g success. Sci. To tal Environ.
661, 553–562.
OECD, 2012. Environmental Outlook to 2050: The Consequences of Inaction. p. 350.
Potts, S.G., Imperatriz-Fonseca, V., Ngo, H.T., Aizen, M.A., Biesmeijer, J.C.,Breeze, T.D., et al.,
2016. Safeguarding pollinators and their values to human well-being. Nature 540,
220–229.
Russell, C.L., 2018. 5 G wireless telecommunications expansion: public health and envi-
ronmental implications. Environ. Res. 165, 484–495.
Sánchez-Bayo, F., Wyckhuys, K.A.G., 2019. Worldwide decline of the entomofauna: a re-
view of its drivers. Biol. Conserv. 232, 8–27.
Shepherd, S., Lima, M.A.P., Oliveira, E.E., Sharkh, S.M., Jackson, C.W., Newland, P.L., 2018.
Extremely low frequency electromagnetic fields impair the cognitive and motor abil-
ities of honey bees. Sci. Rep. 8.
Sutherland, W.J., Butchart, S.H.M., Connor, B., Culshaw, C., Dicks, L.V., Dinsdale, J., et al.,
2018. A 2018 horizon scan of emerging issues for global conservation and biological
diversity. Trends Ecol. Evol. 33, 47–58.
Sutton, G.P., Clarke, D., Morley, E.L., Robert, D., 2016. Mechanosensory hairs in bumble-
bees (Bombus terrestris) detect weak electric fields. Proc. Natl. Acad. Sci. U. S. A.
113, 7261–7265.
Tomanová, K., Vácha, M., 2016. The magnetic orientation of the Antarctic am phipod
Gondogeneia antarctica is cancelled by very weak radiofrequency fields. J. Exp. Biol.
219, 1717–1724.
Vácha, M., Puzova, T., Kvicalova, M., 2009. Radio frequency magnetic fields disrupt
magnetoreception in American cockroach. J. Exp. Biol. 212, 3473–3477.
Vanbergen, A.J., the Insect Pollinators Initiative, 2013. Threats to an ecosystem service:
pressures on pollinators. Front. Ecol. Environ 11, 251–259.
Vijver, M.G.,Bolte, J.F., Evans, T.R., Tamis, W.L., Peijnenburg, W.J.,Musters, C.J., et al., 2014.
Investigating short-term exposure to electromagnetic fields on reproductive capacity
of invertebrates in the field situation. Electromagn Biol Med 33, 21–28.
Wan, G.J., Jiang, S.L., Zhao, Z.C., Xu, J.J., Tao, X.R., Sword, G.A., et al., 2014. Bio-effects of
near-zero magnetic fields on the growth, development and reproduction of small
brown planthopp er, Laodelphax striatellus and bro wn planthopper, Ni laparvata
lugens. J. Insect Physiol. 68, 7–15.
Watt, A., Ainsworth, G., Balian, E., Cojocaru, G., Darbi, M., Dicks, L., et al., 2019. EKLIPSE:
engaging knowledge holders and networks for evidence-informed European policy
on biodiversity and ecosystem services. Evidence & Policy 15, 253–264.
Wyszkowska, J., Shepherd, S., Sharkh, S., Jackson, C.W., Newland, P.L., 2016. Exposure to
extremely low frequency electromagnetic fields alters the behaviour, physiology
and stress protein levels of desert locusts. Sci. Rep. 6, 36413.
Zhang, Z.Y., Zhang, J.,Yang, C.J., Lian, H.Y.,Yu, H., Huang, X.M., etal., 2016. Coupling mech-
anism of electromagnetic field and thermal stress on Drosophila melanogaster. PLoS
One 11, e0162675.
7A.J. Vanbergen et al. / Science of the Total Environment 695 (2019) 133833
