Content uploaded by Marcello D’Amico
Author content
All content in this area was uploaded by Marcello D’Amico on Feb 14, 2022
Content may be subject to copyright.
Risk of bird electrocution in power lines: a framework for
prioritizing species and areas for conservation and impact
mitigation
L. D. Biasotto
1,2
, F. Moreira
3
, G. A. Bencke
4
, M. D’Amico
5,6
, A. Kindel
1,2
&
F. Ascens~
ao
7
1 Programa de P
os-Graduac
ß
~
ao em Ecologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
2N
ucleo de Ecologia de Rodovias e Ferrovias (NERF), Departamento de Ecologia, Universidade Federal do Rio Grande do Sul, Porto
Alegre, Brazil
3 REN Biodiversity Chair, CIBIO/InBIO –Centro de Investigac
ß
~
ao em Biodiversidade e Recursos Gen
eticos, Laborat
orio Associado,
Universidade do Porto, Vair~
ao, Portugal & CIBIO-ISA, Institute of Agronomy, University of Lisbon, Lisbon, Portugal
4 Museu de Ci^
encias Naturais, Departamento de Biodiversidade, Secretaria de Meio Ambiente e Infraestrutura, Porto Alegre, RS, Brazil
5 Theoretical Ecology and Biodiversity Modelling Group - THEOECO, CIBIO-InBIO (University of Porto and University of Lisbon), Tapada da
Ajuda Campus, Lisbon, Portugal
6 Department of Conservation Biology, Do~
nana Biological Station CSIC, Spain
7 Centre for Ecology, Evolution and Environmental Changes (cE3c), Faculdade de Ci^
encias, Universidade de Lisboa, Lisboa, Portugal
Keywords
avian conservation; electrocution hazard;
power lines; linear infrastructures; risk
assessment; electric utility; pylon
management; impact mitigation.
Correspondence
Larissa D. Biasotto, Programa de P
os-
Graduac
ß
~
ao em Ecologia, Universidade
Federal do Rio Grande do Sul, Av. Bento
Gonc
ßalves 9500, CEP 91501-970, CP 15007,
Porto Alegre, RS, Brazil.
Email: larissabiasotto@hotmail.com
Present address
N
ucleo de Ecologia de Rodovias e Ferrovias,
Departamento de Ecologia, Universidade
Federal do Rio Grande do Sul, Porto Alegre,
RS, Brazil
Editor: Jeff Johnson
Associate Editor: Stefano Canessa
Received 07 January 2021; accepted 04
August 2021
doi:10.1111/acv.12736
Abstract
Electrocution on power lines is an important human-related cause of bird mortality
and an important conservation issue worldwide. Besides impacts on bird popula-
tions, electrocutions cause power outages, resulting in damage to power line net-
work integrity. However, there is a general lack of knowledge on the risk of bird
electrocution, especially in developing countries. Generating information over large
scales without resorting to local mortality data can be useful for the development
of regional management strategies, particularly in countries where electrocution is
poorly documented. Here, we developed a framework to model the risk of bird
electrocution as an interaction between the species-specific exposure to power lines
(pole density within a species distribution range) and susceptibility (morphological
and behavioral traits associated with electrocution hazards). We applied this fra-
mework to Brazil, identifying 283 species that face a risk of electrocution, of
which 38 were classified as higher risk, mostly raptors (76%). The Pantanal (a
large wetland biome) concentrates the greatest cumulative susceptibility due to the
high number of species vulnerable to electrocution (i.e. large species using power
lines for perching or nesting), while the Atlantic Forest region has a higher risk for
electrocution, due to the spatial overlap between the presence of vulnerable species
and high exposure to power lines. Furthermore, our study identified spatial patterns
of bird electrocution, highlighting priority areas for electrocution susceptibility and
electrocution risk to be further investigated, and where measures to mitigate bird
electrocutions should be applied on new and existing power lines. Our framework
allows a preliminary assessment aimed at identifying areas of higher risk of elec-
trocution, to highlight species vulnerable to this threat and to improve power line
routing. This approach can be replicated to other understudied areas of the world
where the same information is available.
Introduction
Access to electricity is essential to meet basic human needs,
boost economic growth and foster development. Today,
almost all regions of the globe are crossed by transmission
networks, which are continually expanding to underserved
regions (Jenkins, Smallie & Diamond, 2010). However,
despite their benefits to humans, power lines can pose a seri-
ous threat to birds by causing direct mortality due to colli-
sion with wires and electrocution (Bernardino et al., 2018;
Biasotto & Kindel, 2018; D’Amico et al., 2018). Such mor-
tality can have negative effects on population dynamics and
Animal Conservation (2021) – ª2021 The Zoological Society of London 1
Animal Conservation. Print ISSN 1367-9430
demography, thus potentially affecting the persistence of spe-
cies over time. Population-level impacts caused by power
lines have been demonstrated, for example, for the Cape vul-
ture Gyps coprotheres (Boshoff et al., 2011), Bonelli’s eagle
Aquila fasciata (Hern
andez-Mat
ıas et al., 2015) and Lud-
wig’s bustard Neotis ludwigii (Shaw et al., 2015), and for a
number of other raptor species as reviewed by Slater, Dwyer
& Murgatroyd (2020). Bird electrocutions may also affect
human populations, as they can cause power outages (Bur-
gio, Rubega & Sustaita, 2014; Reed et al., 2014) and wild-
fire ignitions (Guil et al., 2018), resulting in economic losses
for companies and customers (Maricato et al., 2016). Thus,
assessing the risk of bird electrocution is relevant from both
biodiversity conservation and economic perspective.
Identifying both species and areas most prone to electro-
cution events is essential to mitigate the impacts of existing
power lines and to improve the planning of new energy cor-
ridors. However, the systematic collection of bird-
electrocution data along power lines has been primarily per-
formed at a local scale, and mainly for the identification of
high-risk structures (Tint
o, Real & Ma~
nosa, 2010; Guil
et al., 2011; Dixon et al., 2017). Upscaling these surveys
from the local scale to larger management areas (at the
regional or country level) can be difficult to achieve due to
the high amount of resources and time required. The alterna-
tive of multiple surveys conducted on a local scale, if not
coordinated, may fail to identify where mitigation and con-
servation efforts are most needed by overlooking areas and
species with higher electrocution risk (Dwyer et al., 2016).
Therefore, generating sound information at large scales with-
out performing extensive fieldwork would be highly useful
for identifying those areas where to develop site-based man-
agement strategies (D’Amico et al., 2019). This can be par-
ticularly important in countries where electrocution events
are poorly documented and where power line grids are
expanding rapidly (Eccleston & Harness, 2018).
Here, we suggest a framework for an initial risk assess-
ment, at a large scale, using already available data on bird
species and power line networks. The framework is based on
the combination of two types of species-specific information,
namely exposure to power lines and susceptibility to electro-
cution which, when combined, provide an integrated measure
of the electrocution risk (Fig. 1). Exposure is defined as the
likelihood for the individuals of a target species to encounter
an electric pole and is assumed to increase proportionally
with increasing pole density (Dwyer et al., 2016). That is,
when a species inhabits a geographic range with a high pole
density, that species has high overall exposure, and conse-
quently the probability of electrocution occurring increases
(Dwyer et al., 2020). Species susceptibility to electrocution
is mainly influenced by intrinsic behavioral and morphologi-
cal traits (Bevanger, 1998; Janss, 2000). Birds using power
line structures as poles and wires for perching (Prather &
Messmer, 2010) and nesting (Morelli et al., 2014; Moreira
et al., 2018) are supposed to be more susceptible. Suscepti-
bility also increases with body size (Dwyer et al., 2015), as
electrocution occurs when a bird simultaneously touches
two-phase conductors or one conductor and a ground wire
device on a pole (Janss, 2000). Similar frameworks estimat-
ing risk as a combination of exposure and susceptibility have
been previously proposed to address other impacts related to
linear infrastructures, such as roadkills (Visintin, van der Ree
& McCarthy, 2016; Morelli, Benedetti & Delgado, 2020)
and collision with power lines (D’Amico et al., 2019).
Our framework allows a preliminary assessment of poten-
tial electrocution risk for whole bird communities over broad
geographic scales (regional to continental, but potentially
also applicable at a global scale). We applied this framework
to Brazil, the largest megadiverse country, and for which no
specific study or standardized assessment of bird electrocu-
tion has been conducted to date. With rare exceptions, bird
electrocution has been largely neglected in the Neotropics
(Galmes et al., 2017). In Brazil, all information consists of a
few published casualty records and some anecdotal reports
available in the scientific or gray literature. However, the
country has a high species richness in taxonomic or func-
tional groups known to be affected by electrocutions else-
where, and there is no reason to suspect that such casualties
do not occur in a similar way in Brazil. Electrocution also
represents a threat to mammals, a group for which fatality
events are far better documented in Brazil (Lokschin et al.,
2007; Correa et al., 2018). Together, this evidence suggests
that bird electrocution, although currently underreported, is a
real issue that potentially affects many species in Brazil.
Therefore, it is necessary to increase our ability to under-
stand potential electrocution risk patterns across Brazilian
regions. Here, we aimed to answer the following questions:
(1) How is the overall bird susceptibility to electrocution
spatially distributed? (2) How is the potential overall risk of
bird electrocution spatially distributed? (3) Which species
may face a higher risk of electrocution and, therefore, should
receive special attention for conservation and mitigation?
Materials and methods
Study area
Brazil has a large land area (>8.5 M km
2
), about half of
South America. Both human population growth and urban-
ization rates are unevenly distributed across the country and
demand an extensive and growing energy network (MME,
2019). Thousands of kilometers of power lines traverse the
six Brazilian biomes (Fig. 2), regions with distinct environ-
ments defined according to the prevailing climate and vege-
tation type: Amazon (tropical rainforests), Atlantic Forest
(coastal rainforests), Caatinga (seasonally dry forests), Cer-
rado (tropical savannas), Pampa (subtropical/temperate grass-
lands) and Pantanal (tropical wetlands) (IBGE, 2019). This
environmental heterogeneity affords an exceptional diversity
of birds, accounting for almost 20% of the world’s bird spe-
cies richness (Jetz, Thomas & Joy, 2012).
Power line information
We used the georeferenced database on the distribution of
poles at the country level (updated through November 2018)
2Animal Conservation (2021) – ª2021 The Zoological Society of London
Assessing the risk of bird electrocution in power lines L. D. Biasotto et al.
Figure 1 A general framework to assess the risk of bird electrocutions. Exposure is measured as the density of electrical infrastructure
within the species distribution and susceptibility is related to the morphological and behavioral traits associated with a higher risk of electro-
cution. This approach allows identifying species and areas with higher electrocution risk.
Figure 2 Workflow to identify target species and areas for bird electrocution prevention/control in eight steps. (1) Identification of geo-
graphic areas with high pole density –overall exposure map. (2) Obtaining species-specific exposure maps by overlaying the species distri-
bution areas with the overall exposure map. (3a) Insertion of the wing-length value in the raster maps of the species distribution to generate
the species-specific susceptibility maps. (3b) Cumulative map of susceptibility by summing up all species-specific susceptibility maps. (4a)
Species risk maps obtained by multiplying species-specific susceptibility and exposure maps. (4b) Identification of areas with higher risk of
bird electrocution by multiplying the overall exposure and overall susceptibility maps. (5) Extraction of median value within species risk
maps. (6) Identification of higher-risk species.
Animal Conservation (2021) – ª2021 The Zoological Society of London 3
L. D. Biasotto et al. Assessing the risk of bird electrocution in power lines
available from the Brazilian Electricity Regulatory Agency
(ANEEL). We selected structures from medium-voltage
power lines (1–44 kV, n=30 665 490 poles) as these are
the most likely to cause bird electrocutions, because the dis-
tance between the electrical components (wire-wire and pole-
wire) matches the wingspan of several bird species (APLIC,
2006; Lehman, Kennedy & Savidge, 2007; Eccleston & Har-
ness, 2018). Throughout the text, we apply the term ‘pole’
to designate all the structures used to support medium-
voltage power lines in Brazil, regardless of their material or
specific configuration, as there is no information available
that allows us to explicitly distinguish between the different
structures (such as wood poles or pylons).
Bird species information
We used the spatial data available from BirdLife Interna-
tional and Handbook of the Birds of the World (www.bird
life.org/datazone; BirdLife v10, 2017) to obtain georefer-
enced polygons corresponding to the geographic range of
each species. We excluded marine, insular, extinct and
vagrant species according to the latest checklist of Brazilian
birds published by the Brazilian Ornithological Records
Committee –CBRO (de Piacentini, 2015). We reconciled the
BirdLife and CBRO datasets for taxonomic inconsistencies
(e.g. use of different scientific names for the same species or
the adoption of divergent species limits for certain taxa).
Whenever a species in BirdLife was treated as two or more
distinct species on the CBRO checklist, we treated them as a
single species in our analyses. This resulted in a working list
of 1668 regularly occurring continental species (87% of the
Brazilian species).
Wingspan is often selected as an indicator of morphologi-
cal susceptibility to electrocution in birds (Bevanger, 1998),
but it is not available for most Brazilian species. We used
wing length as a proxy because this measure often represents
the overall body size better than other univariate traits (e.g.
Wiklund, 1996), and because it correlates well with wing-
span in a wide variety of bird groups (see Supporting Infor-
mation, Figure S1). For the species for which we could not
obtain accurate wing-length data in the literature (n=30, all
passerines), we estimated missing values using a linear
model relating ‘wing length’~‘body length’(see Supporting
Information, Figure S2).
To assess the behavioral susceptibility to electrocution, we
classified each species according to its use of poles and
wires for perching or nesting. First, we searched for evidence
of perching or nesting on cables and poles in the online pho-
tographic archive of Brazilian birds WikiAves (www.wikiave
s.com.br). This citizen-science platform currently encom-
passes over 3 million photographic records of Brazilian birds
from all over the country. For each species, photographs
were inspected until unequivocal evidence of perching and/or
nesting on poles or wires was found (if any). This meant
that, to be accepted, a record had to clearly show one or
more individuals of the species perched on a power wire or
pole, or with an active nest built on a pole or its associated
structures (transformers, cable supports, etc.). To facilitate
the search for images of nesting behavior, we filtered pho-
tographs with content ‘nest’or main actions ‘incubating’,
‘caring for/feeding offspring’or ‘building nest’using the
website built-in advanced search tool. For species poorly rep-
resented in the WikiAves database (e.g. rare or restricted-
range species) and lacking evidence of the use of power line
structures, the potential for perching/nesting on poles or
wires was assessed by one of us (G.A.B) based on his exten-
sive field experience with Brazilian birds and using evidence
accumulated for phylogenetically related species sharing sim-
ilar behavior and habitat preferences.
Modeling framework and analysis
The framework here proposed allows obtaining independent
and complementary maps of susceptibility, exposure and risk
as defined in Fig. 1. We used raster information with the
same extent (Brazilian territory) and a 50 950 km resolu-
tion (n=3419 cells). This resolution is suitable for our pur-
poses since 76% of the 98 Brazilian energy companies
manage areas larger than 2500 km²(equivalent to our pixel
of 50 950 km). In addition, this resolution is recommended
for continental or broad-scale analysis (Hawkins, Rueda &
Rodr
ıguez, 2008). Therefore, it can be assumed as an appro-
priate scale for regional mitigation planning and manage-
ment. We conducted the analyses at the biome level because
each Brazilian biome contains a distinct bird assemblage and
has unique drivers affecting biodiversity (Souza et al., 2020).
These drivers (land-use and land-cover changes) directly
influence the rates of urbanization and expansion of the
energy network. There are also several legal and conserva-
tion planning instruments that have been developed and/or
are applied at the biome level and need to be complied by
energy companies. The delimitation of Brazilian biomes fol-
lowed IBGE (2019).
We started by building an overall exposure map based on
the electricity pole distribution at the country level (see
Power line information section), described as the density of
medium voltage poles per unit area (50 950 km grid cell
or km²) (Fig. 2, step 1). The value of pole density was log
transformed to reduce the importance of the few cells con-
taining very large urban areas and correspondingly very high
pole densities. We derived species-specific exposure maps
(Fig. 2, step 2) for the subset of behaviorally susceptible
species (i.e. those species identified as using cables and/or
poles; see Bird species information section) by clipping the
overall exposure map with the polygon of distribution of
each species.
Similarly, we built species-specific susceptibility maps
(Fig. 2, step 3A) by using the wing length (in mm) as a sus-
ceptibility indicator and assigning this value to each pixel of
each individual species map. We then obtained the overall
susceptibility map for Brazil by summing all species-specific
susceptibility maps (Fig. 2, step 3B). We chose to work with
the cumulative susceptibility because we assume that all spe-
cies using power line structures have some risk of electrocu-
tion, even the small ones. Thus, this metric indicates the
accumulated susceptibility for each pixel, considering only
4Animal Conservation (2021) – ª2021 The Zoological Society of London
Assessing the risk of bird electrocution in power lines L. D. Biasotto et al.
those species that use power line structures. From this over-
all susceptibility map, we estimated the median value of sus-
ceptibility for each biome.
Using these information layers, we further derived
species-specific maps of electrocution risk by multiplying the
species morphological susceptibility (i.e. wing length) and
exposure maps (Fig. 2, step 4A). To obtain the map of over-
all electrocution risk in Brazil (Fig. 2, step 4B), we multi-
plied the overall exposure map by the overall susceptibility
map, from which we also extracted the median risk for each
biome. The species-specific maps of electrocution risk were
used to rank species according to their median risk (within
their respective distribution ranges) (Fig. 2, step 5). To iden-
tify the subset of species potentially most affected by elec-
trocution (i.e. with higher median risk values), we split the
ranking into different classes using the method proposed by
Jiang (2013) for data with heavy-tailed distributions. The
method partitions the class intervals and so establishes the
number of classes through an iterative multistep approach.
The first step splits data values around the mean into two
parts (head and tail); the next step splits the above-average
values again into head and tail by the new mean, and so on
until the head values are no longer heavy-tailed. Each mean
corresponds to the upper limit of a class. This approach
resulted in three classes, interpreted by us as species with
‘higher’,‘intermediate’and ‘lower’electrocution risk,
respectively (Fig. 2, step 6). We used this objective classifi-
cation to reduce arbitrariness in the assignment of classes, as
we do not have data to establish a clear relationship between
risk level and priority level, which should ideally be defined
on a local or case-by-case basis. We considered higher-risk
species to be of greatest concern for conservation and miti-
gation actions at the national scale.
Finally, the overall exposure map, the species-specific
maps, and the final susceptibility map were normalized to 0–
1 values according to the minimum and maximum values in
each map, with 0 being assigned to the grid cells with the
lowest observed value and 1 to the grid cells with the high-
est observed value. This facilitates comparisons since all
maps are on the same scale and ensured that the values of
the input maps (exposure and susceptibility) had equal
weights when multiplied to produce the risk map. We per-
formed all spatial analyses in R environment (R Core Team,
2020) and the R packages ‘sf’(Pebesma, 2018), ‘fasterize’
(Ross, Sumner & Al, 2020), ‘raster’(Hijmans & van Etten,
2020), ‘classInt’(Bivand, 2020) and ‘rgdal’(Bivand et al.,
2020). Map layouts were made using ArcGIS 10.3 (ESRI,
2015).
Results
Pole density across the country ranged from 0 to 260 poles/
km
2
(Table 1). Extensive areas in the Amazon (more than
half of the biome’s land surface) and Pantanal, and a few
areas in the Cerrado, had very low pole densities (Fig. 3a).
The Atlantic Forest, Pampa, Caatinga and the rest of the
Cerrado had considerably higher and more homogeneous
pole densities (Fig. 3a).
We found direct evidence of the use of power lines for
perching and/or nesting for 242 bird species, and we further
identified another 41 species that potentially use power line
structures, totaling 283 species (Supporting Information,
Table S1). Pantanal was the biome with the highest cumula-
tive bird susceptibility to electrocution (Fig. 3b). Some areas
of Cerrado adjacent to Pantanal also showed high cumulative
susceptibility (Fig. 3b). At the other extreme, the Amazon
showed the lowest median of cumulative susceptibility
(Fig. 3b). The overall pattern does not change when only
species with direct evidence of the use of power line struc-
tures are included in the analyses (242 species) (see Support-
ing Information, Figure S3). The Atlantic Forest showed the
highest electrocution risk (median values), followed in order
by Cerrado, Pantanal, Pampa, Caatinga and Amazon
(Fig. 4).
We identified 38 species with higher electrocution risk
(13% of the analyzed species) (Fig. 5), having a median risk
ranging between 0.25 and 0.57. Most of the higher risk spe-
cies were raptors (76%). The three species with the highest
risk values were the jabiru Jabiru mycteria (a very large
stork), the black-chested buzzard-eagle Geranoetus
melanoleucus and the crowned eagle Urubitinga coronata.
The risk within each species range was highly variable
across species, but some species such as the black-chested
buzzard-eagle and the crowned eagle had a small interquar-
tile range over high-risk values, denoting a high pole density
across their entire distribution ranges (Fig. 5). Of the remain-
ing species, 51 were classified as ‘intermediate risk’(18%;
0.10–0.25) and 194 as ‘lower risk’(67%; 0.00–0.10) (Sup-
porting Information, Figure S4).
Discussion
Electrocution is probably one of the main causes of human-
induced bird mortality (Loss, Will & Marra, 2015; Slater
et al., 2020), but for many regions in the world, there is still
no information on the extent of this impact. We provided a
framework that allows a preliminary assessment of the spa-
tial distribution of bird exposure and susceptibility, which
combined provide the potential risk of electrocution. The
framework is based on available information, including the
Table 1 Pole density of medium voltage power lines (1–44 kV) in
the six Brazilian biomes. Biomes are arranged by the median pole
density
Biome
Range pole
density/km²
Mean (SD) pole
density/km²
Median pole
density/km²
Atlantic
forest
0.5–259.8 13.1 19.8 9.7
Pampa 0.5–68.3 6.0 7.8 4.5
Caatinga 0.1–25 4.4 3.3 3.6
Cerrado 0.0–63.8 3.8 4.7 2.6
Pantanal 0.0–5.4 0.6 1.0 0.1
Amazon 0.0–48.4 0.8 2.3 0.0
Animal Conservation (2021) – ª2021 The Zoological Society of London 5
L. D. Biasotto et al. Assessing the risk of bird electrocution in power lines
Figure 3 Summary results for exposure and susceptibility to electrocution, for 283 species with behavioral characteristics associated with a
higher risk of electrocution: (a) overall exposure map based on pole density of medium voltage power lines (1–44 kV) (raw values of pole
density were log transformed to reduce the high relative weight of a few areas in the Atlantic Forest, (see Table 1); (b) overall susceptibility
map based on distribution area and wing length. In both panels, pixel values were normalized to a 0–1 scale. Boxplots represent the
interquartile range (IQR; box), the median (vertical bar), the 1.5 9IQR interval (whiskers) and the outliers (dots).
Figure 4 Overall risk of bird electrocution relating power pole density to bird susceptibility map. Boxplots represent the interquartile range
(IQR; box), the median (vertical bar), the 1.5 9IQR interval (whiskers) and the outliers (dots).
6Animal Conservation (2021) – ª2021 The Zoological Society of London
Assessing the risk of bird electrocution in power lines L. D. Biasotto et al.
density of electrical infrastructures within a species’range,
and morphological and behavioral traits associated with a
higher risk of electrocution.
The application of the framework to Brazil indicated the
Atlantic Forest as the biome with the highest risk of bird
electrocution, a condition resulting from a high pole density
combined with a high number of susceptible species occur-
ring therein. The Atlantic Forest concentrates more than 60%
of the Brazilian human population (Scarano & Ceotto,
2015), so it is not surprising that it has the highest pole den-
sity (Ribeiro et al., 2009). The most cost-effective method to
reduce bird electrocution in areas that already accommodate
such a dense electrical network is the adoption of mitigation
measures in strategic sites, including the retrofitting of wood
poles (Dwyer, Harness & Eccleston, 2017) or pylons (Che-
vallier et al., 2015; Dixon et al., 2019) whose design poses
a risk of electrocution. In areas classified as higher risk,
under our framework, such as the Atlantic Forest, a more
focused investigation can be applied to search for data on
fatalities or power system failures assigned to electrocutions
and to evaluate the relationship between landscape composi-
tion and electrocution events (e.g. Hern
andez-Mat
ıas et al.,
2020). The pole location and configuration are important in
determining the risk of mortality at the local scale (Mojica
et al., 2018) and to inform the best mitigation interventions.
Our approach indicates where electric companies and conser-
vation biologists can focus their attention to perform detailed
assessments aimed at identifying high-risk poles. For exam-
ple, it is known that a combination of high pole density,
prey abundance and few natural perches can increase electro-
cution risk for some raptor species (P
erez-Garc
ıaet al.,
2011).
Our results also showed that the Pantanal and surrounding
areas concentrate the highest cumulative susceptibility (i.e.
the highest concentration of species vulnerable to electrocu-
tion). Although this biome and the Amazon currently experi-
ence high rates of agricultural expansion, they still have
relatively few energy infrastructures (Souza et al., 2020),
including power lines. However, the Pantanal can become a
central area for future projects of infrastructure expansion
that will connect the Amazon to central Brazilian regions
(Hyde, Bohlman & Valle, 2018; Vilela et al., 2020). Our
results indicate that the routing of new transmission lines
through the Pantanal can lead to a significant increase in the
risk of electrocution for susceptible species. Consequently,
we recommend that these new facilities are carefully planned
Figure 5 Variation in electrocution risk for 38 bird species with higher risk of electrocution. Boxplots represent the interquartile range (IQR;
box), the median (vertical bar), the 1.5 9IQR interval (whiskers) and the outliers (dots). Gray boxes represent raptor species. Common
names in bold represent globally threatened species according to the IUCN Red List of Threatened Species (BirdLife International, 2021).
Animal Conservation (2021) – ª2021 The Zoological Society of London 7
L. D. Biasotto et al. Assessing the risk of bird electrocution in power lines
and, if it is not possible to avoid the most sensitive areas,
the new power lines should at least be built with appropriate
mitigation measures and considering the behavior and mor-
phology of the most susceptible species.
Thirty-eight bird species were ranked as higher risk of
electrocution in our study. Specific guidelines for protection
against electrocution should be delineated for these species,
which should further receive special attention during the
environmental licensing of new power lines. We found that
birds of prey predominate among our higher risk species. In
particular, the globally threatened crowned eagle, white-
collared kite Leptodon forbesi and white-necked hawk
Amadonastur lacernulatus are of special concern because the
electrocution of even a few individuals can potentially cause
serious population-level impacts. Raptors are top predators
and are well known for using cables and poles to perch (Sla-
ter et al., 2020). Overall, they have a delayed maturity, low
fertility rates and smaller population sizes, thus electrocution
events may have severe implications for their population
dynamics and persistence over time (Eccleston & Harness,
2018; De Pascalis et al., 2020; Slater et al., 2020). In our
ranking, the black-chested buzzard-eagle and the endangered
crowned eagle occupied the top positions. These results are
in line with studies performed in neighboring semiarid areas
of central Argentina, where the former species accounts for
numerous electrocution records (Ibarra & De Lucca, 2015)
and the latter has been shown to be disproportionately
affected by electrocution considering its low population
density (Galmes et al., 2017; Sarasola, Galmes & Watts,
2020).
Likewise, parrots (order Psittaciformes), represented in our
higher-risk class by the endangered lear’s macaw Anodor-
hynchus leari and hyacinth macaw Anodorhynchus hyacinthi-
nus, stand out for their curiosity and social behavior when
interacting with power line structures (Seibert, 2006). They
often peck the structures and interact closely with pole ele-
ments like jumpers and energized wires, and most commonly
in small flocks. Galmes et al. (2017) reported two of the
three species of parrots in Argentina as victims of electrocu-
tion, suggesting that this is indeed an important mortality
factor for this group, at least in certain regions. In general,
parrots are long-lived birds, and many species have naturally
restricted ranges and are threatened with extinction (Berkun-
sky et al., 2017). We strongly recommend that particular
attention be paid to these threatened macaws and parrots, for
example by carefully considering the pole design to be used,
ideally tailored for both raptors and for this group, since
measures to mitigate the electrocution of raptors may not be
sufficient or suitable for parrots.
Further developments of the framework
We acknowledge some limitations of our framework and
reinforce that it should be considered a first approach to
identify regions and species that need more attention and
should be targeted for a more in-depth data collection pro-
cess. We highlight some features that could be refined to
improve the effectiveness of the framework.
First, some authors argue that the use of pole density as a
single surrogate for bird exposure to electrocution can com-
promise the results because this approach ignores habitat
availability and configuration in hazard pole arrangement
(P
erez-Garc
ıaet al., 2017; Hern
andez-Lambra~
no, S
anchez-
Agudo & Carbonell, 2018). In our study, we used pole den-
sity as a proxy for exposure to electrocution because there is
no systematic information on electrocution hazards for any
part of the Brazilian territory (and for many other countries),
thus preventing approaches that identify the most problematic
pole configurations. Yet, it has been shown that the probabil-
ity of electrocution increases with pole density (Guil et al.,
2011, 2015; Dwyer et al., 2020). We strongly recommended
that future replications of our framework include information
about pole/pylon configuration whenever available, in order
to refine the electrocution risk as it is influenced by ground-
ing, pole and crossarms material, conductor and wire
arrangement, insulators, exposed jumpers and other technical
elements (Tint
oet al., 2010; Dwyer et al., 2017).
Second, the use of the wing length and perching/nesting
behaviors as the sole predictors of susceptibility may seem
reductionist. These traits, however, can be easily obtained,
allowing a preliminary assessment of a vast pool of species
as ours, similarly to previous approaches that looked at the
risk of bird collision with wires and wind turbines (Beston
et al., 2016; Santangeli et al., 2018; D’Amico et al., 2019).
Our framework also does not integrate information on spe-
cies abundance, or species-specific patterns of frequency and
temporal use of the infrastructures. Resident species, for
example, can have higher temporal exposure to poles than
migratory species. On the other hand, a migratory species
that spends its non-breeding period in low-risk areas might
at least theoretically be significantly affected if it passes
through high-risk areas during migration. While the inclusion
of such information would greatly improve the ability to
identify higher risk areas at finer resolutions or with larger
confidence, we currently have no information about these
species’traits. However, this information may be available at
smaller scales or for some target species or groups.
Third, although we have explicitly indicated species that
should receive special attention, we recommend caution
when interpreting the terms ‘intermediate’and ‘lower’risk,
as our rating provides a value that can only be used for
comparisons within our species pool. All 283 species ana-
lyzed face some risk of electrocution, due to their use of
energy structures. Some smaller species, despite their size,
show gregarious nesting behavior or build bulky nests of
sticks. The monk parakeet Myiopsitta monachus, for example
builds large communal nests that greatly increase the risk of
electrocution and power outages (Burgio et al., 2014).
Fourth, we were unable to verify or validate our results
because bird electrocutions are under-reported and understud-
ied in Brazil. We hope that our results will stimulate hypoth-
esis testing and encourage electrocution surveys aimed at
higher-risk areas and species to further improve the frame-
work. Finally, the grid resolution used here is rather coarse
for mitigation planning and routing at local scales. For
example, it is not appropriate to indicate the most suitable
8Animal Conservation (2021) – ª2021 The Zoological Society of London
Assessing the risk of bird electrocution in power lines L. D. Biasotto et al.
corridors for new power lines or to identify specific poles to
be mitigated. However, our spatial scale is applicable for
planning the expansion of the energy network, especially in
countries where there is a long distance between the power
plants and the main energy-consuming regions, such as Bra-
zil (Cardoso J
unior, Magrini & da Hora, 2014). The out-
comes of our framework can be ‘fine-tuned’following a
systematic process of data refinement. The bird distribution
and pole configuration databases are key elements to be
improved for fine-scale studies.
Conclusion
Our framework allows recognizing target regions and species
to receive special attention in bird electrocution assessments,
fostering win-win situations that will simultaneously prevent
bird fatalities and power outages. We recommend refining the
scale in higher-risk regions and for higher-risk species. The
method proposed here can be replicated in any geographic
area of the world where bird electrocutions are poorly studied
and where information on birds and power lines is available.
Acknowledgments
We thank Brazil’s Access to Information Law and ANEEL for
making power line spatial data available to this study. We thank
Luis Borda de Agua and Vinicius A. G. Bastazini for reading a
previous version of the paper. We also thank the associate editor
and two anonymous reviewers for helpful suggestions that
greatly improved the paper. This study was financed by the Coor-
denac
ß
~
ao de Aperfeic
ßoamento de Pessoal de N
ıvel Superior
(88882.345595/2019‐01) –Brazil (CAPES) –Finance Code 001
through Portal de Periodicos´and also scholarships granted to
LDB under the Programa de Excel^
encia Acad^
emica (PROEX).
MD and FA were funded by Fundac
ß
~
ao para a Ci^
encia e
Tecnologia (CEECIND/03798/2017 and CEECIND/03265/
2017, respectively). FM was supported by the Portuguese
Science and Technology Foundation (FCT), through contract
IF/01053/2015.
Authors’ contributions
Conception and design were conducted by LB, FM, MD,
AK and FA. Data collection was performed by LB and
GAB. Data preparation and analysis were performed by LB
and FA. The first draft of the paper was written by LB. All
authors commented and critically reviewed this and subse-
quent versions of the paper. All authors read and approved
the final paper.
References
APLIC - Avian Power Line Interaction Committee. (2006)
Suggested practices for raptor protection on power lines:
the state of the art in 2006. Washington D.C and
Sacramento C.A: Edison Electric Institute, APLIC, and the
California Energy Comission, 1–204.
Berkunsky, I., Quillfeldt, P., Brightsmith, D.J., Abbud, M.C.,
Aguilar, J., Alem
an-Zelaya, U., Arambur
u, R.M. et al.
(2017). Current threats faced by Neotropical parrot
populations. Biol. Cons. 214, 278–287.
Bernardino, J., Bevanger, K., Barrientos, R., Dwyer, J.F.,
Marques, A.T., Martins, R.C., Shaw, J.M., Silva, J.P. &
Moreira, F. (2018). Bird collisions with power lines: state of
the art and priority areas for research. Biol. Cons. 222,
1–13.
Beston, J.A., Diffendorfer, J.E., Loss, S.R. & Johnson, D.H.
(2016). Prioritizing avian species for their risk of
population-level consequences from wind energy
development. PLoS One 11, e0150813.
Bevanger, K. (1998). Biological and conservation aspects of
bird mortality caused by electricity power lines: a review.
Biol. Cons. 86,67–76.
Biasotto, L.D. & Kindel, A. (2018). Power lines and impacts
on biodiversity: a systematic review. Environ. Impact
Assess. Rev. 71,110–119.
Bivand, R. (2020). ‘classInt’Choose Univariate Class
Intervals. R Package 1–11. Available at: https://r-spatial.
github.io/classInt/
Bivand, R., Keitt, T., Rowlingson, B., Pebesma, E., Sumner,
M., Hijmans, R., Rouault, E. & Bivand, M.R. (2020).
Package “rgdal”. Available at: https://cran.rproject.org/web/
packages/rgdal/index.html
Boshoff, A.F., Minnie, J.C., Tambling, C.J. & Michael, M.D.
(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 Conserv. Int. 21,311–
327.
BirdLife International. (2021) IUCN Red List for birds.
Available at: http://www.birdlife.org
BirdLife International and Handbook of the Birds of the
World. (2021) Bird species distribution maps of the world.
Version 2017.2. Available at: http://datazone.birdlife.org/spec
ies/requestdis
Burgio, K.R., Rubega, M.A. & Sustaita, D. (2014). Nest-
building behavior of Monk Parakeets and insights into
potential mechanisms for reducing damage to utility poles.
PeerJ 2, e601.
Cardoso J
unior, R.A.F., Magrini, A. & da Hora, A.F. (2014).
Environmental licensing process of power transmission in
Brazil update analysis: case study of the Madeira
transmission system. Energy Pol. 67, 281–289.
Chevallier, C., Hern
andez-Mat
ıas, A., Real, J., Vincent-Martin,
N., Ravayrol, A. & Besnard, A. (2015). Retrofitting of
power lines effectively reduces mortality by electrocution in
large birds: An example with the endangered Bonelli’s
eagle. J. Appl. Ecol. 52, 1465–1473.
Corr^
ea, F.M., Chaves,
O.M., Printes, R.C. & Romanowski,
H.P. (2018) Surviving in the urban–rural interface: Feeding
and ranging behavior of brown howlers (Alouatta guariba
clamitans) in an urban fragment in southern Brazil. Am J
Primatol,80,1–12. https://doi.org/10.1002/ajp.22865
Animal Conservation (2021) – ª2021 The Zoological Society of London 9
L. D. Biasotto et al. Assessing the risk of bird electrocution in power lines
D’Amico, M., Catry, I., Martins, R.C., Ascens~
ao, F.,
Barrientos, R. & Moreira, F. (2018). Bird on the wire:
landscape planning considering costs and benefits for bird
populations coexisting with power lines. Ambio 47,
650–656.
D’Amico, M., Martins, R.C.,
Alvarez-Mart
ınez, J.M., Porto,
M., Barrientos, R. & Moreira, F. (2019). Bird collisions
with power lines: prioritizing species and areas by
estimating potential population-level impacts. Divers.
Distrib. 25, 975–982.
De Pascalis, F., Panuccio, M., Bacaro, G. & Monti, F. (2020).
Shift in proximate causes of mortality for six large
migratory raptors over a century. Biol. Cons. 251, 108793.
de Piacentini, V.Q., Aleixo, A., Agne, C.E., Maur
ıcio, G.N.,
Pacheco, J.F., Bravo, G.A., Brito, G.R.R., Naka, L.N.,
Olmos, F., Posso, S., Silveira, L.F., Betini, G.S., Carrano,
E., Franz, I., Lees, A.C., Lima, L.M., Pioli, D., Schunck, F.,
do Amaral, F.R., Bencke, G.A., Cohn-Haft, M., Figueiredo,
L.F.A., Straube, F.C. & Cesari, E. (2015). Annotated
checklist of the birds of Brazil by the Brazilian
Ornithological Records Committee/Lista comentada das aves
do Brasil pelo Comit^
e Brasileiro de Registros Ornitol
ogicos.
Rev. Bras. Ornitol. 23,91–298.
Dixon, A., Rahman, M.L., Galtbalt, B., Bold, B., Davaasuren,
B., Batbayar, N. & Sugarsaikhan, B. (2019). Mitigation
techniques to reduce avian electrocution rates. Wildl. Soc.
Bull. 43, 476–483.
Dixon, A., Rahman, M.L., Galtbalt, B., Gunga, A.,
Sugarsaikhan, B. & Batbayar, N. (2017). Avian
electrocution rates associated with density of active small
mammal holes and power-pole mitigation: Implications for
the conservation of Threatened raptors in Mongolia. J. Nat.
Conserv. 36,14–19.
Dwyer, J.F., Gerber, B.D., Petersen, P., Armstrong, W.E. &
Harness, R.E. (2020). Power pole density and avian
electrocution risk in the Western United States. J. Raptor
Res. 54, 93.
Dwyer, J.F., Harness, R.E. & Eccleston, D. (2017). Avian
electrocutions on incorrectly retrofitted power poles. J.
Raptor Res. 51, 293–304.
Dwyer, J.F., Harness, R.E., Gerber, B.D., Landon, M.A.,
Petersen, P., Austin, D.D., Woodbridge, B., Williams, G.E.
& Eccleston, D. (2016). Power pole density informs spatial
prioritization for mitigating avian electrocution. J. Wildl.
Mgmt. 80, 634–642.
Dwyer, J.F., Kratz, G.E., Harness, R.E. & Little, S.S. (2015).
Critical dimensions of raptors on electric utility poles. J.
Raptor Res. 49, 210–216.
Eccleston, D.T. & Harness, R.E. (2018). Raptor electrocutions
and power line collisions. In Sarasola, H.J, Grande, J.M &
Negro, J.J. (Eds). Birds of prey: 273–302. Cham: Springer
International Publishing.
ESRI. (2015). ARCMAP 10.3.1. Redlands, CA: ESRI.
Galmes, M.A., Sarasola, J.H., Grande, J.M. & Vargas, F.H.
(2017). Electrocution risk for the endangered Crowned
Solitary Eagle and other birds in semiarid landscapes of
central Argentina. Bird Conserv. Int. 28, 403–415.
Guil, F.,
Angels Colomer, M., Moreno-Opo, R. & Margalida,
A. (2015). Space–time trends in Spanish bird electrocution
rates from alternative information sources. Glob. Ecol.
Conserv. 3, 379–388.
Guil, F., Fern
andez-Olalla, M., Moreno-Opo, R., Mosqueda, I.,
G
omez, M.E., Aranda, A., Arredondo,
A., Guzm
an, J., Oria,
J., Gonz
alez, L.M. & Margalida, A. (2011). Minimising
mortality in endangered raptors due to power lines: the
importance of spatial aggregation to optimize the application
of mitigation measures. PLoS One 6, e28212.
Guil, F., Soria, M.
A., Margalida, A. & P
erez-Garc
ıa, J.M.
(2018). Wildfires as collateral effects of wildlife
electrocution: an economic approach to the situation
in Spain in recent years. Sci. Total Environ. 625,
460–469.
Hawkins, B.A., Rueda, M. & Rodr
ıguez, M.
A. (2008). What
do range maps and surveys tell us about diversity patterns?
Folia Geobot. 43, 345–355.
Hern
andez-Lambra~
no, R.E., S
anchez-Agudo, J.
A. & Carbonell,
R. (2018). Where to start? Development of a spatial tool to
prioritise retrofitting of power line poles that are dangerous
to raptors. J. Appl. Ecol. 55, 2685–2697.
Hern
andez-Mat
ıas, A., Ma~
nosa, S., Rollan,
A., Bosch, R., Tint
o,
A. & Real, J. (2020). Using multi-scale spatial prioritization
criteria to optimize non-natural mortality mitigation of target
species. Glob. Ecol. Conserv. 23, e01082.
Hern
andez-Mat
ıas, A., Real, J., Par
es, F. & Pradel, R. (2015).
Electrocution threatens the viability of populations of the
endangered Bonelli’s eagle (Aquila fasciata) in Southern
Europe. Biol. Cons. 191,110–116.
Hijmans, R.J. & van Etten, J. (2020). Package raster:
geographic data analysis and modeling. Version: 1.9-55. R
66. In Ecological applications. Vol. 25, No. 1. Stralberg, D.
et al. (Ed.). Vienna: Foundation for Statistical Computing.
Available at: https://cran.rproject.org/web/packages/raster/
index.html
Hyde, J.L., Bohlman, S.A. & Valle, D. (2018). Transmission
lines are an under-acknowledged conservation threat to the
Brazilian Amazon. Biol. Cons. 228, 343–356.
Ibarra, J. & De Lucca, E. (2015).
Aguilas moras (Geranoaetus
melanoleucus), v
ıctimas de electrocuci
on en Luj
an de Cuyo,
Mendoza, Argentina. N
otulas Faun
ısticas 176,1–7.
IBGE - Instituto Brasileiro de Geografia e Estat
ıstica. (2019)
Biomas e sistema costeiro-carinho do Brasil: compat
ıvel
com a escala 1:250.000.–Coordenac
ß
~
ao de Recursos
Naturais e Estudos Ambientais, 168.
Janss, G.F.E. (2000). Avian mortality from power lines: a
morphologic approach of a species-specific mortality. Biol.
Cons. 95, 353–359.
Jenkins, A.R., Smallie, J.J. & Diamond, M. (2010). Avian
collisions with power lines: a global review of causes and
mitigation with a South African perspective. Bird Conserv.
Int. 20, 263–278.
10 Animal Conservation (2021) – ª2021 The Zoological Society of London
Assessing the risk of bird electrocution in power lines L. D. Biasotto et al.
Jetz, W., Thomas, G.H., Joy, J.B., Hartmann, K. & Mooers,
A.O. (2012). The global diversity of birds in space and
time. Nature 491, 444–448.
Jiang, B. (2013). Head/tail breaks: a new classification scheme
for data with a heavy-tailed distribution. Prof. Geogr. 65,
482–494.
Lehman, R.N., Kennedy, P.L. & Savidge, J.A. (2007). The
state of the art in raptor electrocution research: a global
review. Biol. Cons. 135, 459–474.
Lokschi, L.X., Rodrigo, C.P., Hallal Cabral, J.N. & Buss, G.
(2007) Power lines and howler monkey conservation in
porto alegre, Rio Grande do Sul, Brazil. Neotrop. Primates,
14,76–80. https://doi.org/10.1896/044.014.0206
Loss, S.R., Will, T. & Marra, P.P. (2015). Direct mortality of
birds from anthropogenic causes. Annu. Rev. Ecol. Evol.
Syst. 46,99–120.
Maricato, L., Faria, R., Madeira, V., Carreira, P. & de Almeida,
A.T. (2016). White stork risk mitigation in high voltage
electric distribution networks. Ecol. Eng. 91,212–220.
MME - Minist
erio de Minas de Energia. (2019). Plano
decenal de expans~
ao de energia 2029. Brasil: Empresa de
Pesquisa Energ
etica.1–382.
Mojica, E.K., Dwyer, J.F., Harness, R.E., Williams, G.E. &
Woodbridge, B. (2018). Review and synthesis of research
investigating golden eagle electrocutions. J. Wildl. Mgmt.
82, 495–506.
Moreira, F., Martins, R.C., Catry, I. & D’Amico, M. (2018).
Drivers of power line use by white storks: a case study of
birds nesting on anthropogenic structures. J. Appl. Ecol. 55,
2263–2273.
Morelli, F., Beim, M., Jerzak, L., Jones, D. & Tryjanowski, P.
(2014). Can roads, railways and related structures have
positive effects on birds? –A review. Transp. Res. D
Transp. Environ. 30,21–31.
Morelli, F., Benedetti, Y. & Delgado, J.D. (2020). A
forecasting map of avian roadkill-risk in Europe: a tool to
identify potential hotspots. Biol. Cons. 249, 108729.
Pebesma, E. (2018). Simple features for R: standardized
support for spatial vector data. RJ.10, 439–446. https://doi.
org/10.32614/RJ-2018-009.
P
erez-Garc
ıa, J.M., Botella, F., S
anchez-Zapata, J.A. &
Mole
on, M. (2011). Conserving outside protected areas:
edge effects and avian electrocutions on the periphery of
special protection areas. Bird Conserv. Int. 21, 296–302.
P
erez-Garc
ıa, J.M., DeVault, T.L., Botella, F. & S
anchez-
Zapata, J.A. (2017). Using risk prediction models and
species sensitivity maps for large-scale identification of
infrastructure-related wildlife protection areas: the case of
bird electrocution. Biol. Cons. 210, 334–342.
Prather, P.R. & Messmer, T.A. (2010). Raptor and corvid
response to power distribution line perch deterrents in Utah.
J. Wildl. Mgmt. 74, 796–800.
R Core Team. (2020). R: a language and environment for
statistical computing. Vienna: R Foundation for Statistical
Computing. Available at: https://www.R-project.org/
Reed, J.E., McCleery, R.A., Silvy, N.J., Smeins, F.E. &
Brightsmith, D.J. (2014). Monk parakeet nest-site selection
of electric utility structures in Texas. Landsc. Urban Plan.
129,65–72.
Ribeiro, M.C., Metzger, J.P., Martensen, A.C., Ponzoni, F.J. &
Hirota, M.M. (2009). The Brazilian Atlantic Forest: how
much is left, and how is the remaining forest distributed?
Implications for conservation. Biol. Cons. 142,
1141–1153.
Ross, N., Sumner, M. & Al, E. (2020). Package ‘fasterize’.
Santangeli, A., Butchart, S.H.M., Pogson, M., Hastings, A.,
Smith, P., Girardello, M. & Moilanen, A. (2018). Mapping
the global potential exposure of soaring birds to terrestrial
wind energy expansion. Ornis Fenn. 95,1–14.
Sarasola, J.H., Galmes, M.A. & Watts, B.D. (2020).
Electrocution on power lines is an important threat for the
endangered chaco eagle (Buteogallus coronatus)in
Argentina. J. Raptor Res. 54, 166.
Scarano, F.R. & Ceotto, P. (2015). Brazilian Atlantic forest:
impact, vulnerability, and adaptation to climate change.
Biodivers. Conserv. 24, 2319–2331.
Seibert, L.M. (2006). Social behavior of psittacine birds. In:
Manual of parrot behavior (Luescher, A.U., Ed.). Ames,
Iowa: Blackwell Publishing, 43–48.
Shaw, J.M., Jenkins, A.R., Allan, D.G. & Ryan, P.G. (2015).
Population size and trends of Ludwig’s Bustard Neotis
ludwigii and other large terrestrial birds in the Karoo, South
Africa. Bird Conserv. Int. 26,69–86.
Slater, S.J., Dwyer, J.F. & Murgatroyd, M. (2020).
Conservation letter: raptors and overhead electrical systems.
J. Raptor Res. 54, 198–203.
Souza, C.M., Shimbo, J.Z., Rosa, M.R., Parente, L.L.,
Alencar, A.A., Rudorff, B.F.T., Hasenack, H. et al. (2020).
Reconstructing three decades of land use and land cover
changes in Brazilian biomes with Landsat archive and earth
engine. Remote Sens. 12, 2735.
Tint
o, A., Real, J. & Ma~
nosa, S. (2010). Predicting and
correcting electrocution of birds in Mediterranean areas. J.
Wildl. Mgmt. 74, 1852–1862.
Vilela, T., Harb, A.M., Bruner, A., Da Silva Arruda, V.L.,
Ribeiro, V., Alencar, A.A.C., Grandez, A.J.E., Rojas, A.,
Laina, A. & Botero, R. (2020). A better Amazon road
network for people and the environment. Proc. Natl. Acad.
Sci. USA 117, 7095–7102.
Visintin, C., van der Ree, R. & McCarthy, M.A. (2016). A
simple framework for a complex problem? Predicting
wildlife–vehicle collisions. Ecol. Evol. 6, 6409–6421.
Wiklund, C.G. (1996). Body length and wing length provide
univariate estimates of overall body size in the merlin.
Condor 98, 581–588.
Supporting information
Additional supporting information may be found online in
the Supporting Information section at the end of the article.
Animal Conservation (2021) – ª2021 The Zoological Society of London 11
L. D. Biasotto et al. Assessing the risk of bird electrocution in power lines
Figure S1. Linear regression model relating wingspan (de-
pendent variable) and wing length (independent variable) for
129 species of Brazilian birds belonging to 36 taxonomic
families.
Figure S2. Linear model relating bird wing length ~body
length measures.
Figure S3. Map of electrocution susceptibility for 242 spe-
cies of Brazilian birds that had their use of poles or wires
for perching or nesting confirmed by examination of WikiA-
ves photographs.
Figure S4. Classes and class intervals of bird electrocution
risk as defined by the application of the clustering algorithm
scheme of Jiang (2013) for data with a heavy-tailed distribu-
tion.
Table S1. Information on susceptibility and risk of electrocu-
tion for 283 Brazilian bird species.
12 Animal Conservation (2021) – ª2021 The Zoological Society of London
Assessing the risk of bird electrocution in power lines L. D. Biasotto et al.
