Content uploaded by James F. Dwyer
Author content
All content in this area was uploaded by James F. Dwyer on Nov 02, 2018
Content may be subject to copyright.
The Wilson Journal of Ornithology 130(3):600–614, 2018
Documenting and reducing avian electrocutions in Hungary: a conservation
contribution from citizen scientists
Iva´n Demeter,
1
Ma´rton Horva´ th,
1
Ka´roly Nagy,
1
Zolta´n G¨or¨ogh,
1
P´
eter T´
oth,
1,4
Ja´nos Bagyura,
1
Szabolcs Solt,
1
Andra´s Kova´ cs,
3
James F. Dwyer,
2
* and Richard E. Harness
2
ABSTRACT—Electrocutions of birds on power structures is a global conservation concern that has not been thoroughly
reported in all areas where it occurs. Here we provide information from citizen scientists describing 3,400 avian carcasses of
at least 79 species found at the bases of 57,486 electrical pylons in Hungary. Of these carcasses, 3%were found at the bases
of pylons retrofitted to reduce electrocution risk. On average, one carcass was found per 15 nonretrofitted pylons surveyed
compared one carcass per 89 retrofitted pylons, an 83%difference in frequency. Electrocutions included 4 species of
conservation concern in Hungary: Red-footed Falcons (Falco vespertinus), European Rollers (Coracias garrulous), Saker
Falcons (Falco cherrug), and Eastern Imperial Eagles (Aquila heliaca). Only 3 of 104 (3%) electrocutions involving these
species occurred on retrofitted pylons. Across birds of various sizes (small 25 cm long, medium 26–49 cm long, and large
50 cm long), differences in electrocution frequencies on nonretrofitted and retrofitted pylons were smallest for small birds,
apparently because small birds could walk across the unprotected gaps in coverage directly below energized conductors. In
this study, citizen scientists documented the breadth of the electrocution problem in Hungary but were not trained to record
detailed pylon-specific configuration details. Rather, each pylon surveyed was categorized into one of 8 general
configurations. Pylons with terminal connections were the most dangerous, accounting for 8%of pylons and 24%of
electrocutions. Future mitigation may benefit from professional scientists conducting detailed analyses of how electrocutions
occurred on retrofitted pylons. Received 28 February 2017. Accepted 22 May 2017.
Key words: corvid, eagle, electrocution, mortality, power line, pylon, raptor.
A madarakat ´
er}
oa´ram¨ut ´
esek dokumeta´la´sa ´
es visszaszor´
ıta´ sa Magyarorsza´ gon ¨
onk´
entesek k¨
ozremu22 k¨
od´
es´
evel
ABSTRACT (Hungarian)—A madarakat ´
er}
oa´ ram¨
ut´
es egy globa´lis term ´
eszetv´
edelmi probl´
ema, amelynek dokumenta´la´sa azonban nem
minden ter¨
uleten t¨ort´
enik meg, ahol a probl´
ema el}
ofordul. Ebben a cikkben ¨onk ´
entesek bevona´sa´val gyu22jt ¨ott informa´ci´
okat mutatunk be
minimum 79 mada´rfaj 3,400 tetem ´
er}
ol, amelyek ¨osszesen 57,486 elektromos oszlop k ¨ornyezet´
eben ker¨
ultek el}
o. A tetemek 3%-a olyan
oszlopokna´l volt, amelyeken ma´r t¨ort´
entek mada´rv´
edelmi technikai beavatkoza´sok. ´
Atlagosan 15 felm´
ert a´talak´
ıta´s n´
elk¨
uli oszlopra jutott egy
tetem, m´
ıg az a´talak´
ıtott oszlopok eset´
eben minden 89 oszlopra jutott egy a´ldozat, ami 83%-os gyakorisa´gbeli elt ´
er´
est jelent. Az a´ram ¨
ut´
es
n´
egy olyan fajt is ´
erint, amelyek a magyar term´
eszetv´
edelem sza´ma´ra kiemelt fontossa´g ´
uak, ´
ugymint: a k´
ek v´
ercse (Falco vespertinus), a
szalak´
ota (Coracias garrulus), a kerecsens´
olyom (Falco cherrug)´
es a parlagi sas (Aquila heliaca). Az ezeket a fajokat ´
erint}
o 104 a´ram ¨
ut´
es
k¨oz ¨
ul csak 3 eset (3%)t¨ort ´
ent a´talak´
ıtott oszlopokon. A madarak m´
eret´
et tekintve (kicsi 25 cm, k ¨ozepes 26-49 cm ´
es nagy 50 cm
testhossz) az a´talak´
ıtott ´
es a´talak´
ıta´s n´
elk¨
uli oszlopok k¨oz ¨ott mutatkoz ´
oa´ram ¨
ut´
es-gyakorisa´gbeli elt ´
er´
es a kis testu22 madarak eset´
eben volt a
legkisebb, val´
osz´
ınu22 leg az ´
ert, mert a kis testu22 madarak a´t tudnak s´
eta´lni a burkolatlan elemek k ¨oz¨ott ´
es az a´ram alatt l´
ev}
o vezet´
ekek alatt. A
felm´
er´
esekbe bevont ¨onk´
entesek dokumenta´ lta´ k a probl´
ema magyarorsza´gi nag ysa´ grendj´
et, de a vizsga´lt oszlopok konfigura´ci´
os
saja´tossa´gainak r ¨ogz´
ıt´
es´
ehez nem rendelkeztek megfelel}
ok
´
epzetts´
eggel. Ehelyett minden egyes felm´
ert oszlopot 8 oszlopkateg ´
oria´ba
soroltak. A legvesz´
elyesebbnek a fesz´
ıt}
o elemekkel ella´tott oszlopok bizonyultak, amelyek az ¨osszes oszlop 8%-a´t teszik ki, de az a´ram ¨
ut´
esek
24%-a´´
ert felel}
osek. A j¨ov}
obeli mada´rv´
edelmi beavatkoza´sok sza´ma´ ra hasznos lehet a ma´r mada´rv´
edelmi szempontb´
ol a´talak´
ıtott oszlopokon
tapasztalt a´ram ¨
ut´
esek tudoma´nyos m ´
odszerekkel t¨ort´
en}
or
´
eszletes felta´ra´sa.
Kulcsszavak: a´ram ¨
ut´
es, elektromos ta´vvezet ´
ekek, mortalita´s, oszlopok, ragadoz ´
o, sasok, varj´
uf´
el´
ek.
Documentation of avian electrocutions on
overhead electric infrastructure began almost
immediately after construction of the world’s first
overhead electric power systems (Hallinan 1922,
Lano 1927). Electrocutions were reported sporad-
ically thereafter through the mid-20th century
(Anderson 1933, Marshall 1940, Dilger 1954,
Dickinson 1957) before becoming a prominent
conservation concern in the 1970s (Olendorf 1972,
Miller et al. 1975, Bijleveld and Goeldin 1976). As
overhead electric systems spread, so too did avian
electrocutions, and despite 40 years of research
and mitigation efforts, electrocutions remain an
international conservation concern (Bevanger
1998, Prinsen et al. 2011, Servicio Agr´
ıcola y
Ganadero 2015). Electrocution is now often
included in lists of anthropogenic impacts to
declining raptor species. For example, Eagle Owls
(Bubo bubo) in Italy (Sergio et al. 2004), Egyptian
Vultures (Neophron percnopterus) in Egypt (An-
gelov et al. 2013), Golden Eagles (Aquila
chrysaetos) in the United States (USFWS 2016),
1
MME BirdLife Hungary, Budapest, Hungary
2
EDM International, Inc., Fort Collins, CO, USA
3
3300 Koszor´
u u. 46, Eger, Hungary
4
Current address, Agria ¨
Okosziszt´
ema Kft, 3300 Eger,
R´
ozsa Ka´roly u. 12/b, Hungary
* Corresponding author: jdwyer@edmlink.com
600
Spanish Booted Eagle (Aquila pennata) in Spain
(Martinez et al. 2016), and Tasmanian Wedge-
tailed Eagles (Aquila audax fleayi) in Australia
(Bekessy et al. 2009) are all thought to be
declining due in part to electrocution mortality.
Although concerns are global, avian electrocu-
tion research has been disseminated primarily from
studies in Germany (Haas et al. 2005), South
Africa (Van Rooyen et al. 2003, Boshoff et al.
2011, Jenkins et al. 2013), the Iberian Peninsula
(Guil et al. 2011, L´
opez-L´
opez et al. 2011, Moreira
et al. 2017), and the United States (Harness and
Wilson 2001, Lehman et al. 2007, Dwyer et al.
2016). These centers of electrocution research
contribute to overall perceptions of avian electro-
cution but do not reflect the global nature of the
problem, either in terms of species affected or of
the electrical configurations, materials, or local
perceptions involved. Increasingly, research pro-
grams outside of these areas are facilitating
broader understanding of avian electrocution. For
example, the overhead electric systems of India
and Mongolia are adjacent to, and engineered
similarly to, the electrical systems of China, and
each of these systems have been implicated in
large numbers of avian electrocutions (Purevdorj
and Sundev 2012, Dixon et al. 2013, Harness et al.
2013). Consolidated with studies documenting
avian electrocutions in nearby Kazakhstan (Lasch
et al. 2010), Russia (Karyakin et al. 2009,
Barbazyuk et al. 2010), and Russian Siberia
(Goroshko 2011), the potential exists for avian
electrocution risk across much of Asia.
Avian electrocution may also be more prevalent
across Europe than is currently recognized. For
example, in addition to electrocutions in Germany
(Haas et al. 2005), Italy (Rubolini et al. 2001),
Portugal (de Sousa 2017, Moreira et al. 2017), and
Spain (Ferrer 2012), .200 electrocutions of
raptors, corvids, passerines, and other birds were
documented in 6 Important Bird Areas in Bulgaria
(Demerdzhiev et al. 2009, Gerdzhikov and De-
merdzhiev 2009, Demerdzhiev 2014). Power line
configurations in Bulgaria are similar to those in
Croatia, Hungary, and Slovakia (MAVIR 2016), so
concerns identified in Bulgaria may occur region-
ally. Avian electrocution in Hungary first became
apparent in 1980 (Bagyura et al. 2004) and is
either particularly problematic or particularly well
documented, with an estimated 30,000–100,000
birds killed annually (Horva´th et al. 2008, 2011;
T´
oth 2010). In 1991, Birdlife Hungary developed
crossarm insulators designed to reduce electrocu-
tion risk, and from 1994 through 2002, .20,000
pylons were fitted with these insulators. This
design was used as the sole solution to avian
electrocution through the late 2000s when covers
began to be installed on jumpers (the small wires
linking terminal connections and energized equip-
ment). Despite the high numbers of avian
electrocutions in Hungary and the novel mitigation
strategies used to address these incidents, data
from Hungary have not been widely reported in
English-language journals. This omission limits
the ability of the scientific and conservation
communities to learn from mitigation actions
applied in Hungary or to consider potential
regional effects of electrocution mortality in
Hungary.
To address this communication gap, our primary
objective was to document and report as many
electrocuted birds of as many species as possible
from Hungary. Our secondary objective was to
compare mortality frequencies on pylons with and
without retrofitting measures. In doing so, we
compared null hypotheses that no differences in
electrocution rates exist between nonretrofitted and
retrofitted pylons, either overall or when consid-
ering specific pylon types or birds of different
sizes. Some of the data included were previously
released in local reports written in Hungarian
(Horva´th et al. 2010; Supplement S1), but here we
offer a more extensive dataset than was previously
released, provide new analyses of these data, and
compare our findings to avian electrocutions
reported elsewhere.
Methods
Study area
We conducted our study throughout Hungary
(Fig. 1), an Eastern European country character-
ized by mixed habitats from forested mountains to
open agricultural fields and grasslands. The Great
Hungarian Plain east of Budapest is one of the
most extensive floodplains of Europe, supporting
numerous breeding, migratory, and wintering
avian species. Because much of the area was
grassland and open agricultural land or cropland,
electric pylons were often the tallest structure on
the landscape and consequently were attractive
perches to many bird species.
601Demeter et al.Avian electrocutions in Hungary
Electric power is delivered within Hungary via
one national transmission company, Hungarian
Transmission System Operator Company Ltd.
(MAVIR), and 5 regional 22 kV distribution
service providers. These companies are responsi-
ble for ~700,000 pylons supporting 55,000 km of
overhead electric lines (Horva´th et al. 2010).
Distribution pylons are constructed primarily of
reinforced concrete vertical components and
grounded metal arms. Roughly 560,000 of these
pylons are in rural areas, where avian impacts have
historically been associated primarily with electro-
cutions of raptors and corvids (Horva´th et al.
2010).
Carcass survey protocol
We designed a survey protocol used by citizen
scientist volunteers throughout Hungary from
January 2004 through December 2014. In the
context of avian research, citizen scientists are
nonprofessional ornithologists, such as bird watch-
ers, volunteer nest monitors, and bird banders
(Cooper et al. 2015). Although nonprofessionals,
citizen scientists often possess deep expertise
drawn from years of personal engagement and
study and can contribute important data to studies
of avian ecology, facilitating access to broad-scale
patterns not visible to individual researchers. For
example, studies of raptor migration and raptor
responses to climate change have relied on data
from citizen scientists (Hussell and Inzunza 2008,
Paprocki et al. 2017). To our knowledge, studies of
raptor electrocution have not focused on citizen
science, presumably because the technical aspects
of overhead electrical systems were thought to
preclude meaningful contributions by nonprofes-
sionals.
In this study, we coordinated with citizen
scientists to search for avian carcasses at the bases
of pylons in Hungary because leveraging the
availability of citizen scientists met our primary
objective of documenting the greatest number and
widest range of electrocutions possible. The
approach allowed us to identify a broad distribu-
tion of electrocuted species and to compare broad
categories of pylon types (described later). Citizen
science data can be relatively coarse, however,
sometimes precluding researcher’s ability to rig-
Figure 1. Locations of pylons surveyed and avian carcasses found in Hungary from 2004 to 2009. Reproduced from Horva´th
et al. (2010); Supplement S1.
602 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
orously measure all variables of potential interest
(Ryder et al. 2010). In our study, use of citizen
science data maximized detection of carcasses but
limited our ability to identify pylon-specific
attributes that might have contributed to a more
detailed understanding of technical specifications
associated with the electrocutions we report here.
Consequently, in addressing our secondary objec-
tives, our analyses are relatively coarse compared
to studies such as Lehman et al. (2010) undertaken
by professional scientists working in more com-
pact study areas with expertise in both ornithology
and overhead electric system.
Citizen scientists were instructed to search for
the carcasses of electrocuted birds at the bases of a
line of pylons of the citizen’s choosing and to
record the pylon configuration and any carcasses
found for each pylon searched. Citizen scientists
generally focused surveys in areas where they
knew or suspected avian electrocutions might
occur, especially where IUCN Red Listed threat-
ened bird species were believed to be present.
When surveys led to the detection of the
electrocuted carcasses of species of special
concern or the detection of large numbers of
electrocuted birds of any species, citizen scientists
sometimes repeated surveys of the same pylons
within and across years. Thus, some areas known
to be particularly dangerous were surveyed more
frequently, which could be problematic if infer-
ences were made between line segments or
habitats, but in this study, inferences were made
exclusively to pylon types. All pylon types
surveyed were distributed throughout surveyed
areas, seasons, and years and were surveyed
without regard to configuration, eliminating po-
tential bias resulting from citizen scientists prefer-
entially surveying particular pylon types.
This nonrandom approach facilitated documen-
tation of the scope of the electrocution problem in
terms of species involved but simultaneously
prevented us from reporting an estimate for total
electrocution mortality in Hungary because sam-
pled areas were not selected randomly. Despite the
nonrandom approach, because citizen scientists
surveyed continuous lines of pylons without
regard to configuration and reported pylon config-
urations for all pylons surveyed, we were able to
evaluate differences in electrocution rate by pylon
configuration.
We defined 8 pylon configurations to which
citizen scientists could assign surveyed pylons. All
pylons were constructed with steel arms mounted
on steel or steel-reinforced concrete vertical
components. For each configuration, pylons were
identified as nonretrofitted or retrofitted, creating
16 possible categories (Fig. 2–4). Nonretrofitted
pylons included no retrofitting measures designed
to mitigate avian electrocution risk; retrofitted
pylons included covers on crossarms or jumpers
(short wires connecting energized equipment)
designed to reduce the possibility of simultaneous
contact with energized wires and grounded
structural components (Horva´th et al. 2010).
Pylons without jumpers were retrofitted by
covering crossarms; pylons with jumpers were
retrofitted by covering jumpers. Importantly, none
of the pylons evaluated would have been consid-
ered fully retrofitted according to APLIC (2006)
standards because in all cases, some phase-to-
phase or phase-to-ground contact points persisted
after retrofitting. Even minor errors in retrofitting
can lead to electrocutions (Dwyer et al. 2017), so
this study provides information on incomplete
retrofitting, which has never before been specifi-
cally quantified in a scientific study.
In cases where pylons included aspects of
multiple configurations, pylons were categorized
based on the most complex configuration appro-
priate. For example, switches required wires to
terminate on each side of a pylon so the utility
could energize or de-energize discrete line sec-
tions. In these cases, the configuration was
identified as a switch tower even though terminal
connections were present. In other cases, where
terminal connections existed without switches, the
pylons were categorized according to terminal
connections.
Each carcass found was identified to species
when possible or to the most precise taxonomic
category possible otherwise. We also estimated the
body length of each species found based on
previously published values (Mullarney et al.
1999). Information on body length is informative
because avian electrocution is usually associated
with large birds, particularly raptors, and retrofit-
ting often specifically targets raptors (APLIC
2006, Lehman et al. 2010, Dwyer et al. 2015).
Electrocution has been the primary cause of death
in previous studies of avian carcasses found in
association with pylons in Hungary (Bagyura et al.
603Demeter et al.Avian electrocutions in Hungary
2004, T´
oth 2010, Fidl´
oczky et al. 2014). Conse-
quently, we assumed electrocution to be the
primary cause of death for carcasses reported in
this study.
Statistical analyses
To evaluate differences in electrocution risk on
nonretrofitted pylons compared to retrofitted
pylons, our null hypothesis assumed electrocuted
birds would be found below pylons in proportion
to the frequency with which pylons occurred in
surveys. For example, if 3.5%of pylons surveyed
were switch towers without avian-friendly insula-
tion, then 3.5%of carcasses should be attributable
to pylons of this configuration. If carcasses were
found in proportion to pylon configuration for all
configurations, then we would not reject our null
hypothesis. Our alternate hypothesis would be
supported if any ratio of pylons to carcasses was
not proportional (Dixon et al. 2013, Dwyer et al.
2013, Harness et al. 2013). We evaluated our
hypothesis for all carcasses together and for small,
medium, and large birds separately. To evaluate
possible differences in retrofitting effects for birds
of various sizes, we defined small birds as 25 cm
long, medium-sized birds as 26–49 cm, and large
birds as 50 cm. Only carcasses identified to
species were included in this portion of our
analyses.
We used 3 chi-square (v
2
)goodnessoffittestsof
independence to analyze our data. Because we
conducted 3 tests of the same data, we implemented
a Bonferroni correction to reduce the likelihood of
Type I error, adjusting our critical level from a¼
0.050 to a¼0.017. Thus, we considered Pvalues
of 0.017 statistically significant. In our first v
2
test, we compared proportions of nonretrofitted and
retrofitted pylons where carcasses were or were not
found. This coarse analysis met recommendations
for v
2
tests but omitted information on pylon
configurations. Our second v
2
test distinguished 8
pylon configurations, each with 2 treatments (non-
Figure 2. (a) Tangent, 1 insulator per Ø, 1 level; (b) tangent, 1 insulator per Ø, 2 level; (c) tangent, 1 insulator per Ø, 2 level,
retrofitted with covers on the crossarm; (d) tangent, 2 insulators per Ø, 2 levels. Photo credit P. T´
oth. Ø denotes phase.
604 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
retrofitted and retrofitted). This resulted in viola-
tions of recommended procedures for v
2
testing
because 4 observed values were ,5. All of these
values were for retrofitted pylons and were due to
relatively few retrofitted pylons existing for some
configurations, compounded by relatively few
carcasses found at the bases of retrofitted pylons.
Despite these violations, we proceeded with the v
2
test because our first analysis provided support for
overall conclusions regarding nonretrofitted and
retrofitted pylons. This approach enables readers
concerned with v
2
best practices to choose to accept
or disregard the second test. Our third test compared
proportions of small, medium, and large electrocut-
ed birds on nonretrofitted and retrofitted pylons.
Previous models of electrocution risk on
wooden poles in the United States and concrete
pylons in India (with configurations similar to
pylons in Hungary) indicated that relative levels of
risk could be quantified if numbers of conductors,
numbers of jumpers, presence of grounding, and
habitat were evaluated for each structure (pole or
pylon) surveyed (Dwyer et al. 2013, Harness et al.
2013). Our dataset did not include these metrics
because our citizen scientists were trained to
record pylon tops into 1 of 16 categories, not to
quantify pylon-top components. To facilitate
comparison of our data to model results in other
studies, we quantified typical numbers of ener-
gized and exposed conductors, jumpers, and
terminal connections for each configuration. This
simplified approach has been used successfully in
other electrocution research where model results
were intended to be accessible to nonprofessionals
(Dwyer et al. 2013, Harness et al. 2013). We used
these components to create a risk score for each
configuration (Table 1), which omitted some of the
nuance of more detailed models but facilitated
general comparisons to other studies. It also made
our results more accessible to conservationists who
Figure 3. (a) Tangent, 2 insulators per Ø, 1 level, retrofitted with covers on the crossarm; (b) terminal connection, 1 level; (c)
terminal connection, 1 level retrofitted with covers on jumpers; (d) terminal connection, 2 levels. Photo credit P. T´
oth. Ø
denotes phase.
605Demeter et al.Avian electrocutions in Hungary
may not have, or need, advanced technical
understanding of pylon components. For analysis,
we fitted a line between risk score and mean
number of carcasses found per pylon configura-
tion. We used the R
2
value for a best-fit line to
identify the proportion of variation in carcasses per
pylon explained by pylon configuration.
Results
From January 2004 through December 2014,
citizen scientists surveyed 57,486 pylons, includ-
ing 10,372 retrofitted pylons. Citizen scientists
identified exactly 3,400 avian carcasses of at least
79 species (Supplement S1), including 4 species of
conservation concern: Saker Falcon (Falco cher-
rug), Red-footed Falcon (Falco vespertinus),
Eastern Imperial Eagle (Aquila heliaca), and
European Roller (Coracias garrulous). Searchers
found 117 carcasses of at least 17 species at the
bases of retrofitted pylons (Supplement S2). Thus,
97%of carcasses were found under nonretrofitted
pylons, even though these pylons made up only
82%of pylons surveyed, indicating an 83%
difference from an average of one carcass found
per 15 nonretrofitted pylons to one carcass found
per 89 retrofitted pylons.
Some configurations were more dangerous than
others when considering pooled nonretrofitted and
retrofitted configurations (v
2
¼459, df ¼1, P,
0.001; Table 2) or when comparing all 16
configuration*retrofitting categories (v
2
¼3272, df
¼15, P,0.001; Table 3). For example, non-
retrofitted terminal connection 1-level and 2-level
pylons were the most dangerous configurations
(Fig. 5) given the differences between the propor-
tion of the overhead electric system composed of
these 2 configurations (8%) and the proportion of
carcasses found dead at the bases of pylons with
these configurations (24%). Overall, more carcasses
(36%) were found under Tangent, 1 insulator per
phase, 2-level pylons than any other configuration,
Figure 4. (a) Switch tower; (b) switch tower retrofitted with covers on the upper crossarm; (c) transformer pylon; (d)
intersection pylon. Photo credit P. T ´
oth. Ø denotes phase.
606 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
but this was the most common pylon type surveyed
(48%). Thus the configuration posed less risk per
pylon than more complicated configurations with
jumpers and terminal connections.
Carcass sizes ranged in length from 12 to 119
cm, and we found differences in the proportions of
small, medium, and large birds under nonretrofit-
ted and retrofitted pylons (v
2
¼44.65, df ¼5,
Table 1. Pylon configurations and risk scores assigned during surveys of overhead electric distribution systems in Hungary,
January 2004 through December 2014. Ø denotes phase.
Pylon configuration
Number of
Risk score FigureOverarm conductors Jumpers Terminal connections
Tangent, 1 insulator per Ø, 1 level 3 0 0 3 1a
Tangent, 1 insulator per Ø, 1 level, retrofitted 0 0 0 0
a
Tangent, 1 insulator per Ø, 2 level 3 0 0 3 1b
Tangent, 1 insulator per Ø, 2 level, retrofitted 1 0 0 1 1c
Tangent, 2 insulators per Ø, 1 or 2 levels 3 3 0 6 1d
Tangent, 2 insulators per Ø, 1 or 2 levels, retrofitted 1 3 0 4 2a
Terminal connection, 1 level 0 3 6 9 2b
Terminal connection, 1 level, retrofitted 0 0 6 6 2c
Terminal connection, 2 levels 0 3 6 9 2d
Terminal connection, 2 levels, retrofitted 0 0 6 6
b
Switch tower 3
a
6 6 15 3a
Switch tower, retrofitted 0 0 6 6 3b
Transformer 0 3 3 6 3c
Transformer, retrofitted 0 0 3 3
b
Corner or intersection 3 3 3 9 3d
Corner or intersection, retrofitted 0 0 3 3
b
a
Not illustrated. Crossarm covers are identical to those of Tangent, 1 insulator per Ø, 2 level, retrofitted.
b
Not illustrated. Jumper covers are identical to those of Terminal connection, 1 level, retrofitted.
Table 2. Summary statistics from v
2
test for avian carcasses found under all pylon types during surveys in Hungary, January
2004 through December 2014. Ø denotes phase.
Pylon configuration
Pylons surveyed Carcasses found v
2
results
Risk
scoreNumber Percent Observed Expected
Percentage
deviation
Standardized
residuals
Nonretrofitted
Tangent, 1 insulator per Ø, 1 level 4,748 8.3 20 272 93 15 3
Tangent, 1 insulator per Ø, 2 level 27,805 48.4 1,174 1,584 26 10 3
Tangent, 2 insulators per Ø, 1 or 2 levels 4,621 8.0 494 262 89 14 6
Terminal connection, 1 level 2,547 4.5 419 147 185 22 9
Terminal connection, 2 levels 1,918 3.3 353 108 227 24 9
Switch tower 1,978 3.5 238 115 107 11 15
Transformer 1,162 2.0 101 64 58 5 6
Corner or intersection 2,335 4.1 357 134 166 19 9
Retrofitted
Tangent, 1 insulator per Ø, 1 level 1,569 2.7 0 88 100 90
Tangent, 1 insulator per Ø, 2 level 7,098 12.3 56 403 86 17 1
Tangent, 2 insulators per Ø, 1 or 2 levels 647 1.1 27 36 25 1.5 4
Terminal connection, 1 level 332 0.6 19 20 506
Terminal connection, 2 levels 184 0.3 8 10 20 16
Switch tower 169 0.3 2 10 80 36
Transformer 142 0.2 1 7 86 23
Corner or intersection 231 0.4 4 13 69 23
Total
a
57,486 100.0 3,273 3,273 — —
a
Excludes 127 carcasses for which pylon configuration was not recorded.
607Demeter et al.Avian electrocutions in Hungary
P,0.0001). Smaller birds benefitted slightly less
from retrofitting than did medium and large birds
(Table 4). Across configurations, counts of con-
ductors, jumpers, and terminal connections ex-
plained 61%of the variation in electrocution rates
when considering all configurations, and 85%of
the variation in electrocution rates when consider-
ing all configurations other than switch towers
(Fig. 6), which incorporated elements of many
other configuration types.
Discussion
Our primary objective in this study was to
investigate the possibility that avian electrocution
Table 3. Summary statistics from v
2
test for avian carcasses of all sizes found under all types of nonretrofitted pylons pooled
and all types of retrofitted pylons pooled during surveys in Hungary, January 2004 through December 2014. Compare to
Table 1 for analyses of various pylon types and Table 3 for analyses of various carcass sizes.
Status of pylons
Pylons surveyed Carcasses found v
2
results
Number Percent Observed Expected Percentage deviation Standardized residuals
Nonretrofitted 47,114 82 3,156 2,686 18%9
Retrofitted 10,372 18 117 587 80%19
Total
a
3,273 3,273 — —
a
Excludes 127 carcasses for which pylon configuration was not recorded.
Figure 5. Comparisons of the average number of carcasses per pylon on nonretrofitted and retrofitted pylons surveyed in
Hungary, January 2004 through December 2014.
608 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
may pose a conservation concern in Hungary. The
3,400 birds of 17 species we found in association
with electrical pylons met this goal. Carcasses
were not formally necropsied in this study, but
because in Hungary carcasses found in association
with pylons have consistently been identified as
electrocuted (Bagyura et al. 2004, T ´
oth 2010,
Fidl´
oczky et al. 2014), even if some individuals
included in these data died of other causes, overall
numbers are sufficient to justify conservation
concern. Because pylon configurations in Hungary
are similar to those in nearby Croatia, Slovakia,
and Bulgaria (MAVIR 2016), and because avian
electrocutions occur widely in Europe and Asia,
the electrocutions reported here should be viewed
as informative to a conservation concern of
regional importance. Specifically, birds migrating
through or wintering in Eastern Europe may be
encountering potentially dangerous pylons
throughout much of their migratory or wintering
Table 4. Summary statistics from v
2
test for carcasses of
various sizes (small 25 cm long, medium 26–49 cm long,
large birds 50 cm long) found under nonretrofitted and
retrofitted pylons during surveys in Hungary, January 2004
through December 2014.
Status of carcass
Carcasses found v
2
results
Observed Expected
Percentage
deviation
Standardized
residuals
Nonretrofitted
small 217 189 þ15%þ2
medium 1,454 1,233 þ18%þ6
large 1,343 1,139 þ18%þ6
Retrofitted
small 14 42 67%4
medium 50 271 82%13
large 46 250 82%13
Total
a
3,124 3,124 — —
a
Excludes 127 carcasses for which pylon configuration was not recorded, and
22 carcasses which could not be identified to species.
Figure 6. Comparisons of risk score and carcasses per pylon on nonretrofitted and retrofitted pylons surveyed in Hungary,
January 2004 through December 2014.
609Demeter et al.Avian electrocutions in Hungary
ranges, potentially impacting populations breeding
well north of our study area.
Conservation concern is especially relevant for
threatened and endangered species because elec-
trocutions could potentially contribute to or
exacerbate declines of listed species. Our data
included 4 species of concern: the vulnerable and
declining Eastern Imperial Eagle (Demeter et al.
2005, Kova´cs et al. 2008, Birdlife International
2016a), the declining European Roller (Birdlife
International 2016b), the globally endangered and
decreasing Saker Falcon (Birdlife International
2016c), and the near threatened and declining Red-
footed Falcon (BirdLife International 2016d). For
2 of these species, electrocution is known to be a
primary agent of mortality in Hungary, causing
21%of the mortality of Eastern Imperial Eagles
(Horva´th et al. 2011) and 7–10%of juvenile
mortality of the Saker Falcon (Nagy and Demeter
2006, Prommer et al. 2012, Kova´cs et al. 2014).
Considering these observations, future electrocu-
tion research in Hungary may benefit from
transitioning to more focused before-after-con-
trol-impact studies to shift electrocution research
from correlative to causative and to focus
mitigation on species that need it most.
Our secondary objectives in this study were to
evaluate null hypotheses that all pylons were
equally dangerous, and that retrofitting was
equally effective for birds of all sizes. Both null
hypotheses were rejected. Rather, consistent with
numerous other studies and with models of
electrocution risk (Bevanger 1998, Lehman et al.
2007, Ferrer 2012), avian electrocutions were
disproportionately associated with more complex
configurations. Our risk score was particularly
effective in demonstrating these effects; lower risk
scores were associated with fewer carcasses per
pylon type and higher risk scores with more
carcasses per pylon type. Similar risk scoring may
be useful in other situations where personnel are
not fully trained in the technical aspects of power
line equipment. Avian electrocutions on non-
retrofitted and on retrofitted pylons were especially
associated with terminal connections. Terminal
connections often included jumpers that passed
over and were adjacent to grounded crossarms.
Studies precisely modeling electrocution risk as a
Figure 7. A Kestrel (Falco tinnunculus) perched on a grounded pylon-top directly below an energized distribution jumper.
Photo P. T ´
oth.
610 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
function of pole-top and pylon-top components
have found the presence of grounding in proximity
to energized jumpers such as these increases
electrocution risk (Janss and Ferrer 1999, Dwyer
et al. 2013). This risk can be mitigated in part by
routing jumpers below crossarms (APLIC 2006),
provided doing so does not create safety concerns
for utility personnel climbing poles. Importantly,
our findings are consistent with those of electro-
cution studies elsewhere (Spain, United States,
India), even though our mechanism of identifying
pylon types was relatively coarse.
Electrocution retrofitting was less effective for
small birds than for larger birds, possibly because
crossarm covers did not extend below conductors,
allowing small birds to stand on grounded cross-
arms directly below energized conductors (Fig. 7).
Medium and large birds would not fit in these
locations, perhaps introducing a counterintuitive
physiological constraint that reduced electrocution
risk. Covering conductors with snap-on covers, a
common mitigation method for protecting raptors
from contact with primary conductors, may further
reduce electrocution risk for small birds.
Although retrofitted pylons were associated with
fewer carcasses, electrocutions persisted on retro-
fitted pylons because retrofitting did not fully
separate energized from grounded components on
any pylon type. Even small, subcentimeter gaps in
retrofitting can allow electrocutions to occur
(Dwyer et al. 2017), and the retrofitting measures
used in Hungary included such gaps. Future
retrofitting measures in Hungary should involve
strategies to create the full 152 cm wide by 102 cm
tall phase-to-ground separations recommended to
reduce avian electrocution risk (APLIC 2006,
Dwyer et al. 2015). To meet this goal, retrofitting
mitigation should continue to include jumper
covers but should also include covering conduc-
tors and terminal connections to eliminate exposed
energized components adjacent to grounded com-
ponents (APLIC 2006, Dwyer et al. 2015).
Alternative pylon-top designs may also reduce
avian electrocution risk.
Previous studies of raptor electrocution have not
overtly relied on citizen science data, although
many have benefitted from observations contrib-
uted initially by concerned citizens and then
incorporated into datasets by researchers (e.g.,
Lehman et al. 2010). Consistent with Dwyer et al.
(2013) and Harness et al. (2013), our findings
suggest that nonprofessionals can be actively
involved in documenting avian electrocutions.
Future studies of electrocution should carefully
consider benefits and drawbacks of including
citizen scientists in addressing specific research
questions, and when possible capitalize on the
wealth of interest and energy volunteers have the
potential to bring to conservation science.
Supplemental Material
Previously released data and analyses in Hun-
garian (Horva´th et al. 2010 [Supplemental Material
S1]), together with detailed lists of carcasses found
(Supplemental Material S2, Tables S1 and S2) are
available in an online Supplement. The authors are
solely responsible for the content and functionality
of these materials. Queries (other than absence of
the material) should be directed to the correspond-
ing author.
Acknowledgments
We thank the hundreds of volunteers who conducted
power line surveys in support of this study (Horva´th et al.
2010; Supplemental Material S1). We thank Birdlife
Hungary (MME), the LIFE program of the European Union
(LIFE02NAT/H/8627, LIFE05NAT/H/000122, LIFE06-
NAT/H/000096, LIFE09NAT/HU/000384), the Norwegian
and EEA Financial Mechanism, and EDM International for
funding this research. We are also grateful for retrofitting
efforts implemented by the five 22kV distribution companies
operating in Hungary during our studies: ELMU
22
-´
EM ´
ASZ,
E. ON- ´
ED ´
ASZ, E. ON-D´
ED ´
ASZ, E. ON-TIT ´
ASZ, and
EDF-D ´
EM ´
ASZ. Two anonymous peer reviewers provided
insights which greatly improved the clarity of this work.
Literature cited
Anderson AH. 1933. Electrocution of Purple Martins.
Condor. 35:204.
Angelov I, Hashim I, Oppel S. 2013. Persistent electrocution
mortality of Egyptian Vultures Neophron percnopterus
over 28 years in East Africa. Bird Conservation
International 23:1–6.
[APLIC] Avian Power Line Interaction Committee. 2006.
Suggested practices for avian protection on power
lines: the state of the art in 2006. Washington (DC) and
Sacramento (CA): Edison Electric Institute, APLIC,
and the California Energy Commission.
Bagyura J, Szitta T, S´
andor I, Viszl´
o L, Firm´
anszky G, et al.
2004. A review of measures taken against bird
electrocution in Hungary. In: Chancellor RD, Meyburg
B-U, editors. Budapest (Hungary): Raptors Worldwide
world working group on birds of prey and owls and
MME/BirdLife; p. 423–428.
611Demeter et al.Avian electrocutions in Hungary
Barbazyuk EV, Bakka SV, Barashkova AN, Semenov AR,
Smelyanskiy IE. 2010. The outcomes of preliminary
monitoring for death of raptors and other bird species
through electrocution in the Eastern Orenburg district,
Russia. Raptor Conservation. 20:40–47. Russian.
Bekessy SA, Wintle BA, Gordon A, Fox JC, Chisholm R, et
al. 2009. Modelling human impacts on the Tasmanian
Wedge-tailed Eagle (Aquila audax fleayi). Biological
Conservation. 142:2438–2448.
Bevanger K. 1998. Biological and conservation aspects of
bird mortality caused by electricity power lines: a
review. Biological Conservation. 86:67–76.
Bijleveld MFIJ, Goeldin P. 1976. ´
Electrocution d’un couple
de Buses Buteo buteo `
a Jongny [Electrocution of a pair
of Buteo buteo buzzards in Jongny]. Nos Oiseausx.
33:280–281.
BirdLife International. 2016a. Aquila heliaca.IUCNRedList
of threatened species 2016: e.T22696048A90410809;
[accessed 22 Dec 2016]. http://dx.doi.org/10.2305/
IUCN.UK.2016-3.RLTS.T22696048A90410809.en
BirdLife International. 2016b. Coracias garrulus.IUCNRed
List of threatened species 2016: e.T22682860A92965129;
[accessed 22 Dec 2016]. http://dx.doi.org/10.2305/IUCN.
UK.2016-3.RLTS.T22682860A92965129.en
BirdLife International. 2016c. Falco cherrug.IUCNRedList
of threatened species 2016: e.T22696495A90562101;
[accessed 22 Dec 2016]. http://dx.doi.org/10.2305/
IUCN.UK.2016-3.RLTS.T22696495A90562101.en
BirdLife International. 2016d. Falco vespertinus.IUCNRed
List of threatened species 2016: e.T22696432A84476145
[accessed 22 Dec 2016]. http://dx.doi.org/10.2305/IUCN.
UK.2016-3.RLTS.T22696432A84476145.en
Boshoff AF, Minnie JC, Tambling CJ, Michael MD. 2011.
The impact of power line-related mortality on the Cape
Vulture Gyps coprotheres in a part of its range, with an
emphasis on electrocution. Bird Conservation Interna-
tional. 21:311–327.
Cooper CB, Bailey RL, Leech DI. 2015. The role of citizen
science in studies of avian reproduction. In: Deeming
DC, Reynolds SJ, editors. Nests, eggs, and incubation.
Oxford (UK): Oxford University Press; p. 208–220.
Demerdzhiev DA. 2014. Factors influencing bird mortality
caused by power lines within special protected areas
and undertaken conservation efforts. Applied Zoology.
214:411–423.
Demerdzhiev DA, Stoychev SA, Petrov TH, Angelov ID,
Nedyalkov NP. 2009. Impact of power lines on bird
mortality in southern Bulgaria. Acta Zoologica Bulgar-
ica. 61:175–183.
Demeter I, Horv´
ath M, Kov´
acs A. 2005. Conservation of the
Imperial Eagle in the Carpathian Basin 2002–2005:
layman’s report of the LIFE02NAT/H/8627 project.
Budapest (Hungary): MME BirdLife-Hungary.
de Sousa JDBB. 2017. Eletrocuss˜
ao de aves em apoios da
rede el´
etrica e m´
etodos de corre¸c˜
ao [Electrocution of
birds in electrical networks and methods of correction]
[master’s thesis]. Lisbon (Portugal): University of
Lisbon. Department of Animal Biology; [accessed 26
Apr 2017]. https://www.researchgate.net/publication/
316241102_Eletrocussao_de_aves_em_apoios_da_
rede_eletrica_e_metodos_de_correcao
Dickinson LE. 1957. Utilities and birds. Audubon. 59:54–
55, 86–87.
Dilger WC. 1954. Electrocution of parakeets at Agra, India.
Condor. 56:102–103.
Dixon A, Maming R, Gunga A, Purev-Ochir G, Batbayar N.
2013. The problem of raptor electrocution in Asia: case
studies from Mongolia and China. Bird Conservation
International. 23:520–529.
Dwyer JF, Harness RE, Donohue K. 2013. Predictive model
of avian electrocution risk on overhead power lines.
Conservation Biology. 28:159–168.
Dwyer JF, Harness RE, Gerber BD, Landon MA, Petersen P,
et al. 2016. Power pole density informs spatial
prioritization for mitigating avian electrocution. Journal
of Wildlife Management. 80:634–642.
Dwyer JF, Harness RE, Eccleston D. 2017. Avian
electrocutions on incorrectly retrofitted power poles.
Journal of Raptor Research. 51:293–304.
Dwyer JF, Kratz GE, Harness RE, Little SS. 2015. Critical
dimensions of raptors on electric utility poles. Journal
of Raptor Research. 49:210–216.
Ferrer M. 2012. Aves y tendidos el´
ectricos: del conflicto a la
soluci´
on [Birds and power lines: from conflict to
solution]. Sevilla (Spain): Endesa S.A. and Fundaci´
on
Migres.
Fidl´
oczky J, Bagyura J, Nagy K, T ´
oth P, Szitta T, Haraszthy
L. 2014. Bird conservation on electric-power lines in
Hungary: nest boxes for Saker Falcon and avian
protection against electrocutions: project report. Slovak
Raptor Journal. 8:87–95.
Gerdzhikov GP, Demerdzhiev DA. 2009. Data on bird
mortality in ‘‘Sakar’’ IBA (BG021), caused by
hazardous power lines. Ecologia Balkanica. 1:67–77.
Goroshko OA. 2011. Bird electrocution in the Daurian
Steppe (south-eastern trans-Baikal Region), Russia.
Raptor Conservation. 21:84–99. Russian.
Guil F, Fern´
andez-Olalla M, Moreno-Opo R, Mosqueda I,
G´
omez ME, et al. 2011. Minimizing mortality in
endangered raptors due to power lines: the importance
of spatial aggregation to optimize the application of
mitigation measures. PLoS ONE. 6:e28212. doi:10.
1371/journal.pone.0028212
Haas D, Nipkow M, Fiedler G, Schneider R, Haas W,
Sch¨
urenberg B. 2005. Protecting birds from power-
lines. Nature and Environment, No. 140. Strasbourg
(France): Council of Europe Publishing.
Hallinan T. 1922. Bird interference on high tension electric
transmission lines. Auk. 39:573.
Harness RE, Juuvadi PR, Dwyer JF. 2013. Avian electro-
cutions in western Rajasthan, India. Journal of Raptor
Research. 47:352–364.
Harness RE, Wilson KR. 2001. Electric-utility structures
associated with raptor electrocutions in rural areas.
Wildlife Society Bulletin. 29:612–623.
Horv´
ath M, Demeter I, Bagyura J, Kov´
acs A, Lov´
aszi P, et
al. 2010. Madarak ´
es l´
egvezet´
ekek [Birds and overhead
lines]. Budapest (Hungary): Magyar Mad ´
artani ´
es
Term´
eszetv´
edelmi Egyes¨
ulet [Birdlife Hungary]. Hun-
garian.
Horv´
ath M, Nagy K, Papp F, Kov´
acs A, Demeter I, et al.
2008. Magyarorsz´
ag k¨oz´
epfesz¨
ults´
egu22 elektromos ve-
612 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
zet´
ekh´
al´
ozat´
anak mad´
arv´
edelmi szempont´
u´
ert´
ekel´
ese
[Assessment of the Hungarian medium-voltage electric
grid from bird protection point of view]. Budapest
(Hungary): Magyar Mad´
artani ´
es Term´
eszetv´
edelmi
Egyes¨
ulet [Birdlife Hungary]. Hungarian.
Horv´
ath M, Szitta T, Fat´
er I, Kov´
acs A, Demeter I, et al.
2011. Population dynamics of the Eastern Imperial
Eagle (Aquila heliaca) in Hungary between 2001 and
2009. Acta Zoologica Bulgarica Suppl. 3:61–70.
Hussell DJT, Inzunza ER. 2008. Long-term monitoring: the
raptor population index in principle. In: Bildstein KL,
Smith JP, Inzunza ER, Veit RR, editors. State of North
America’s birds of prey. Cambridge (MA): Nuttall
Ornithological Club; and Washington (DC): American
Ornithologists Union; p. 151–163.
Janss GFE, Ferrer M. 1999. Mitigation of raptor electrocu-
tion on steel power poles. Wildlife Society Bulletin.
27:263–273.
Jenkins AR, de Goede KH, Sebelle L, Diamond M. 2013.
Brokering a settlement between eagles and industry:
sustainable management of large raptors nesting on
power infrastructure. Bird Conservation International.
23:232–246.
Karyakin IV, Nikolenko EG, Vazhov SV, Bekmansurov RH.
2009. Raptor electrocution in the Altai region: results
of surveys in 2009, Russia. Raptor Conservation.
16:45–64. Russian.
Kov´
acs A, Demeter I, Fat´
er I, Bagyura J, Nagy K, et al.
2008. Current efforts to monitor and conserve the
Eastern Imperial Eagle Aquila heliaca in Hungary.
AMBIO. 37:457–459.
Kov´
acs A, Williams NP, Galbraith CA. 2014. Saker Falcon
Falco cherrug Global Action Plan (SakerGAP),
including a management and monitoring system, to
conserve the species. Abu Dhabi (United Arab
Emirates): Coordinating Unit - CMS Raptors MOU,
CMS Technical Series No. 31, Raptors MOU Technical
Publication No. 2.
Lano A. 1927. Great Blue Heron (Ardea herodias)
electrocuted. Auk. 44:246.
Lasch U, Zerbe S, Lenk M. 2010. Electrocution of raptors at
power lines in Central Kazakhstan. Raptor Conserva-
tion. 18:95–100.
Lehman RN, Kennedy PL, Savidge JA. 2007. The state of
the art in raptor electrocution research: a global review.
Biological Conservation. 136:159–174.
Lehman RN, Savidge JA, Kennedy PL, Harness RE. 2010.
Raptor electrocution rates for a utility in the Inter-
mountain Western United States. Journal of Wildlife
Management. 74:459–470.
L´
opez-L´
opez P, Ferrer M, Madero A, Casado E, McGrady
M. 2011. Solving man-induced large-scale conserva-
tion problems: the Spanish Imperial Eagle and power
lines. PLoS ONE. 6:e17196.
Martinez JE, Zuberogoitia I, Jimenez-Franco MV, Manosa S,
Calvo JF. 2016. Spatio-temporal variations in mortality
causes of two migratory forest raptors in Spain.
European Journal of Wildlife Research. 62:109–118.
Mullarney K, Svensson L, Zetterstrom D, Grant PJ. 1999.
Princeton field guides: birds of Europe. Princeton (NJ):
Princeton University Press.
Marshall W. 1940. ‘‘Eagle guard’’ developed in Idaho.
Condor. 42:166.
MAVIR. 2016. Seventh MAVIR conservation conference.
7–9 Nov 2016, Budapest, Hungary.
Miller AD, Boeker EL, Thorsell RS, Olendorff RR. 1975.
Suggested practices for raptor protection on power
lines. Washington (DC): Edison Electric Institute, and
Provo (UT): Raptor Research Foundation.
Moreira F, Encarna¸c ˜
ao V, Rosa G, Gilbert N, Infante S, et al.
2017. Wired: impacts of increasing power line use by a
growing bird population. Environmental Research
Letters. 12:024019:1–7.
Nagy S, Demeter I. 2006. Saker Falcon: European single
species action plan. Bern Convention: T-PVS/Inf
(2006) 2 revised.
Olendorff RR. 1972. Eagles, sheep and power lines.
Colorado Outdoors. 2:3–11.
Paprocki N, Oleyar D, Brandes D, Goodrich L, Crewe T,
Hoffman SW. 2017. Combining migration and winter-
ing counts to enhance understanding of population
change in a generalist raptor species, the North
American Red-tailed Hawk. Condor: Ornithological
Applications. 119:98–107.
Prinsen HAM, Boere GC, P´
ıres N, Smallie JJ. 2011.
Review of the conflict between migratory birds and
electricity power grids in the African-Eurasian region.
Bonn (Germany): AEWA Technical Series. CMS
Technical Series No. XX; [accessed 24 Jan 2017].
http://www.cms.int/en/document/review-conflict-
between-migratory-birds-and-electricity-power-grids-
african-eurasian-regi-1
Prommer M, Bagyura J, Chavko J, Uhrin M. 2012.
Migratory movements of Central and Eastern European
Saker Falcons (Falco cherrug) from juvenile dispersal
to adulthood. Aquila. 119:111–134.
Purevdorj A, Sundev G. 2012. The assessment of high risk
utility lines and conservation of globally threatened
pole nesting steppe raptors in Mongolia. Ornis
Mongolica. 1:2–12.
Rubolini D, Bassi E, Bogliana G, Galeotti P, Garavaglia R.
2001. Eagle Owl Bubo bubo and power line interac-
tions in the Italian Alps. Bird Conservation Interna-
tional. 11:319–324.
Ryder TB, Reitsma R, Evans B, Marra PP. 2010.
Quantifying avian nest survival along an urbanization
gradient using citizen- and scientist-generated data.
Ecological Applications. 20:419–426.
Sergio F, Marchesi L, Pedrini P, Ferrer M, Penteriani V.
2004. Electrocution alters the distribution and density
of a top predator, the Eagle Owl (Bubo bubo). Journal
of Applied Ecology. 41:836–845.
Servicio Agr´
ıcola y Ganadero. 2015. Gu´
ıa para la
evaluaci´
on del impacto ambiental de proyectos e´
olicos
ydel
´
ıneas de transmisi´
on el´
ectrica en aves silvestres y
murci´
elagos [Guide for the evaluation of the environ-
mental impact of wind projects and electric transmis-
sion lines on wild birds and bats]. 1st edition. Santiago
(Chile): Ministerio de Agricultura. Spanish.
T´
oth P. 2010. A madarakat ´
er}
o´
aram ¨
ut´
esek nagys´
a-
grendj´
enek becsl´
esi hib´
aja – a d¨ogev}
ok hat´
asa [Bias
in the estimation of electrocution rates – scavenger
613Demeter et al.Avian electrocutions in Hungary
removal effect]. Heliaca. 8:95–103. Hungarian with
English summary.
[USFWS] US Fish and Wildlife Service. 2016. Bald and
Golden eagles: population demographics and estimation
of sustainable take in the United States, 2016 update.
Washington (DC): US Fish and Wildlife Service,
Division of Migratory Bird Management; 102 p.
Van Rooyen C, Vosloo H, Harness RE. 2003. Watch the
birdie! IEEE Industry Applications Magazine. 9:55–
60.
614 The Wilson Journal of Ornithology Vol. 130, No. 3, September 2018
