Content uploaded by Lothar Bach
Author content
All content in this area was uploaded by Lothar Bach on Apr 08, 2017
Content may be subject to copyright.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
1
BAT DEATHS AND WIND TURBINES - A REVIEW OF CURRENT KNOWLEDGE, AND OF THE
INFORMATION AVAILABLE IN THE DATABASE FOR GERMANY
Tobias DÜRR & Lothar BACH
Abstract
The present paper deals with bat mortality in Germany due to the action of wind turbines (up
to now, 10 species in Germany and 15 in Europe) which is known at the present. According to
the studies conducted in Brandenburg and Saxony it appears that most dead bats are found at
the end of summer and during the autumn migration, a fact that is not statistically confirmed
by the irregular checks carried out throughout the year and which should be considered only
as an indication. About 93% of all the victims were found between the dispersal of the
breeding colonies and the end of the autumn migration (August- September). 89% of the
fatalities were around turbines whose towers were less than 100 m from wooded landscape
features (woods and hedgerows). Because of the heterogeneity of the data, no difference in
effect has been noted among the diverse types of wind turbines - all can be responsible for the
death of bats. It appears obviously that the siting of the turbines has a direct influence on the
number of bats killed. The possible reasons for this mortality are discussed here, but they are
still largely unexplained and need more specific research. To this end a protocol for future
comparative studies is presented.
1 Introduction
Until the 1990s only isolated dead birds were found under wind turbines. Up to that time wind
parks were only set up in open landscape near the coast, and nobody thought of the potential
risk for bats. To start with then, the connection between bat mortality and wind turbines was
only a well-founded supposition (Bach et al 1999, Rahmel et al. 1999), but the risk for these
animals has now been proved through numerous findings in Germany and elsewhere. The first
evidence of bat mortality due to wind turbines was published in Australia in the 70s (Hall &
Richards 1972). Much later Osborne et al (1996) mentioned bat bodies found under wind
turbines, this time in Minnesota (USA). After these discoveries, made by chance during
research into the death of birds, the studies were intensified and now most data about bat
mortality comes from the USA (Anonymous 1999, Johnson et al. 2000, Johnson in press,
Johnson et al. 2003, Keeley et al. 2001).
Until recently there were very few studies in Germany about bat behaviour (Rahmel et al.
1999, Bach 2001, 2002) and collisions with wind turbines. The publication by Vierhaus (2000)
of the first dead bat found in Germany in 1998 at first went unnoticed. In 1999 Heddergott
started systematically to research death by collision in Thuringia. His data has yet to be
published. Only a summary is available and the original data is not accessible and cannot be
verified. But the mortality figures published in 2001 by Dürr triggered an avalanche. Since
2002 all published and known data on birds and bats are recorded in a national databank for
mortality by wind turbines which is situated at the Landesamt Brandenburg (Staatliche
Vogelschutzwarte).
In this paper we attempt to analyse the possible relationship between known victims and
the numerous parameters of the wind turbines involved (axle height, rotor diameter and total
height). We also try to establish a possible relationship between the distance of the WT from
the nearest wooded feature that bats might use as a foraging habitat, or as a linear landmark
for orientation while commuting or migrating.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
2
2 Bat fatalities at wind turbines in Germany and Europe
The German databank shows that up to now 207 bats from 10 species have been killed by wind
turbines in 8 different states in Germany (Tab.1)
Tab.1: Bats killed by WTs in Germany (BB = Brandenburg, SAH = Saxony-Anhalt, SN = Saxony, TH =
Thuringia, SH = Schleswig-Holstein, BY = Bavaria, NRW = North Rhine-Westphalen, RP = Rhineland-
Palatine, NDS = Lower Saxony (Vauk et al. 1990) (state as on 31.08.2004).
Species BB SAH SN TH SH BY NRW RP NDS Total
Nyctalus noctula Noctule bat 39 1 20 54 3 1 1 118
Nyctalus leisleri Leisler’s bat 5 1 1 3 10
Eptesicus serotinus Serotine bat 2 2 1 1 6
Vespertilio murinus Parti-coloured bat 1 7 8
Myotis myotis Greater mouse-eared bat 7 7
Myotis daubentonii Daubenton’s bat 1 1
Plecotus austriacus Grey long-eared bat 1 1
Pipistrellus pipistrellus Common pipistrelle 15 2 6 2 25
Pipistrellus nathusii Nathusius’ pipistrelle 17 1 23 2 1 44
Pipistrellus pygmaeus Soprano pipistrelle 1 1
Pipistrellus sp. Pipistrelle sp. 4 14 18
Chiroptera sp. Chiroptera sp. 2 2 4
TOTAL 85 5 59 70 18 1 3 2 0 243
The information from Thuringia (70 victims from 6 species) goes back to M. Heddergott’s
findings in Eichsfeld from 1999 to 2001. At present this data is still insufficiently documented,
but the circumstances of Heddergott’s findings leave no doubt as to its veracity. The 7 dead
Greater Mouse-eared bats (Myotis myotis) are the only individuals of this species found in
Germany under wind turbines and up to now the only dead Myotis spec. killed at European
wind farms1. Unfortunately this data was orally communicated and up to now it has been
impossible to see the dead animals. According to Dürr, the data is perfectly believable, but we
are convinced that the figures for the Greater Mouse-eared bat must be considered with
caution in the absence of the bodies.
The findings from Schleswig-Holstein come exclusively from unsystematic checks at
several wind farms in late summer and autumn 2003 by M. & M. Göttsche; they concern 3
Noctule bats (Nyctalus noctula) and a large number of Pipistrellus spec. We still do not have
the precise identification results of one particular specimen. Data is also missing for those
German states not mentioned in the table and for those with low figures. This missing data
may be simply due to lack of registration2. In Lower Saxony and Schleswig-Holstein a total of
9 wind farm sites along the coast were checked in 1989 and 1990 (Vauk et al. 1990).
It seems there will be a large number of victims wherever the siting of the wind turbines is
in conflict with the foraging airspace of some high flying bats (Eptesicus serotinus,
Vespertilio murinus, Nyctalus noctula, Nyctalus leisleri and also to a lesser extent Pipistrellus
pipistrellus and Eptesicus nilssonii) as described by Rahmel et al. (1999). For the Northern
bat (Eptesicus nilssonii) mortality is still unproved in Germany and no doubt is related to the
very limited distribution area of the species in this country and the lack of corresponding
1 At the date of the translation, other dead Myotis spec. have been found
2 At the date of translation two more investigations are finished or in progress: Brinkmann et al. (2005) Baden-Württemberg, Hermanns
mündl. Mecklenburg-Vorpommern
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
3
studies in the area. But Ahlén (2002) considered this species to be the most common victim
around wind turbines in Southern Sweden. In Germany the migratory species Nyctalus
noctula (47,8%) and Pipistrellus nathusii (19,3%) were found particularly often. Together
they represent two thirds (67,1%) of all victims. On two sites Vespertilio murinus have been
killed by wind turbines (3,9%). Isolated findings (each one 2,4%) concern Nyctalus leisleri
and Eptesicus serotinus and one Pipistrellus pygmaeus and Plecotus austriacus. The
percentage of Pipistrellus pipistrellus is surprisingly high (9,8%) and if we consider the total
of all individuals belonging to the genus Pipistrellus, they represent 31,5% of all findings.
Tab. 2: Bat fatalities at wind farms in Europe (X = species found but no figure available)
Species Germany Sweden Spain Total
Nyctalus lasiopterus Greater noctule X >1
Nyctalus noctula Noctule 118 1 X >100
Nyctalus leisleri Leisler’s bat 10 - - >5
Eptesicus serotinus Serotine 6 - X >5
Eptesicus nilssonii Northern bat - 8 - 8
Vespertilio murinus Parti-coloured bat 8 1 - 9
Tadarida teniotis European free-tailed bat X
Myotis myotis Greater Mouse-eared bat 7 7
Myotis daubentonii Daubenton’s bat 1 1
Pipistrellus pipistrellus Pipistrelle bat 25 1 X +1 >22
Pipistrellus nathusii Nathusius’ pipistrelle 44 5 - 45
Pipistrellus pygmaeus Soprano pipistrelle 1 1 X 2
Pipistrellus kuhlii Kuhl’s pipistrelle X >1
Pipistrellus sp. Pipistrelle spec. 18 17
Hypsugo savii Savi’s pipistrelle X+1 >1
Plecotus austriacus Grey Long-eared bat 1 1
Chiroptera sp. Chiroptera spec. 4 30 14 48
TOTAL 243 47 22 277
In the rest of Europe there are very few results of studies and most are unpublished (Tab. 2).
In Spain, up to the year 2000, about 20 bats were found dead on one wind farm (Alcade 2003,
Artazcoz pers. comm ), mainly Pipistrellus kuhlii but also Pipistrellus pipistrellus, P.
pygmaeus, Hypsugo savii and Eptesicus serotinus as well as Nyctalus noctula and N. leisleri.
Lekuona (2001) made available the results of studies of different wind farms in Navarra
(Spain), revealing the death of one Pipistrellus pipistrellus at Izco Airbar wind farm and one
Hypsugo savii at Salajones wind farm. Dubourg-Savage (pers. comm.) announced that in
France also the first dead bats had been reported under wind turbines, but precise data was not
then available.
On Gotland (Sweden), about 30 dead bats were found in 1999 under one solitary wind
turbine. They were probably Nyctalus noctula. The only other survey of different wind farms
and solitary wind turbines in Southern Sweden in 2002 resulted in findings of dead Eptesicus
nilssonii, Pipistrellus nathusii, Nyctalus noctula, Pipistrellus pygmaeus and Vespertilio
murinus (Ahlén 2002), the Northern bat (E. nilssonii) being the most common victim. Apart
from these findings, no other data from Europe has reached us.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
4
3 Frequency of findings
Since 2001 wind turbine checks are carried out in Brandenburg and the results recorded in the
central databank of the Staatliche Vogelschutzwarte (Tab. 3). The majority of wind farms
were monitored for a minimum of 2 years, and some for 3 years, checked at random dates
throughout the year. However the majority of checks took place between the end of March
and the end of May and between the end of July and the end of October. The intervals
between the controls were irregular and varied between yearly, weekly (March to October)
and fortnightly (November to February). They averaged out at no more than 5 per WT per
year. The following wind farms were surveyed (see also Annex 1): Bredow, Bückwitz,
Eichfelder Damm, Jacobsdorf, Klessen-Görne, Ketzin, Lietzow, Markau, Michelsdorf, Nackel,
Nahmitz, Nauen, Netzen, Prützke, Sieversdorf, Tremmen, Wernikow, Wernitz, Zachow.
Tab. 4 presents Förster’s analysis (2004) of the number of dead bats found. Taking all the
surveys into consideration, we find higher mortality than in Brandenburg (Tab. 3). The
maximum was 11 bats found under just one wind turbine. These high figures can be explained
by the fact that the resources put into the survey were more than twice those in Brandenburg.
In addition the majority of checks took place only during the season when most fatalities
occur (Tab. 3 + 4). Unlike Brandenburg, in Saxony no searching for victims was done during
the six winter months when bat activity is reduced to a minimum. On some wind farms in
Brandenburg (some findings) or in Saxony (Förster 2004), during certain years at least, no
dead bats were found at all during random checks. These checks were for the most part not
continuous (Brandenburg) or took place only during a few weeks or months (Saxony). The
studies in Brandenburg and Saxony made it very likely that dead bats are missed, by the mere
fact that they did not cover the whole activity cycle of the bats. This was also partly the case
for the studies in coastal areas. The data must therefore be considered with great care. But
even on wind farms where dead bats have been found, animals were not lying under every
wind turbine. Other factors probably play a role, such as the activity of predators, the type of
WT, or their distance from the nearest wooded feature, etc. (see below).
To get a better idea of the problem, in September and December 2003 the Staatliche
Vogelschutzwarte of Brandenburg made tests on several wind farms to evaluate how long it
would take for 1-day chicks to disappear if left on the ground (Dürr 2004). Predators, human
and agricultural activity contributed to the rapid disappearance of most chicks within a week
(83% in September, 42% in December). But these results have to be interpreted with caution,
knowing the taste of carnivores for chick flesh. Many bats were found mummified and despite
their state of decomposition had been unnoticed or spurned by the carnivores. We know of no
single case of a bat showing traces of being bitten by a mammal, though there were signs of
insect activity. On the other hand the teeth marks of mammal were regularly observed on dead
birds. Chick flesh may therefore be compared to small bird flesh, but not to the flesh of bats
with its strong smell.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
5
4 Seasonal distribution of findings
Fig.1 shows the seasonal distribution of 49 dead bats found in Brandenburg. Here we must
take into account the fact that the search for dead bats is not evenly distributed throughout the
year. For the purpose of this study, we have analysed data from the central mortality file
concerning 1367checks for a total of 201 wind turbines. During the winter months from
December to February no bat has been found during 156 checks. In spring (March to May)
271 checks were carried out and 3 bats (0,011 ind./control) found. In summer (June to August)
484 controls revealed 27 dead bats (0,056 ind./control). In autumn (September to November)
455 checks were carried out and 19 bats (0,042 ind/control) were found. Therefore for the
period under consideration (2001-2003) 0,04 bat per check was found. The numbers differ
slightly from the data in Tab. 3 which only shows information coming from the studies
realised directly by the Staatliche Vogelschutzwarte.
5 Distance between wind turbines and wooded structures
To determine if landscape features around wind turbines play a role in bat mortality, the
distance of wind turbines responsible for bat fatalities from forest edges, hedgerows and/or
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
6
other types of wooded structures has been analysed (fig. 2). Unfortunately we do not have
data for WT sites with no bat mortality (see above). It is clear that most animals (36
individuals, 77%) were found under turbines which were less than 50 m from trees. As each
bat species reacts differently to landscape features, there should be differences in the
correlation between fatalities under the turbine and its distance from the nearest trees, and
data for the 44 victims should therefore be presented by species. It is however necessary to
take into account the fact that the sampling number per species is very low. As expected,
mortality of Pipistrellus spec. was mostly near trees and there were only very few differences
between Pipistrellus nathusii and Pipistrellus pipistrellus. But for the Noctule bat, mortality
at wind turbines occurred also at a mean distance of 200 m and as far as 600 m from woods.
A surprising finding was that of a Serotine bat (Eptesicus serotinus) killed by a WT which
was 700 m away from the nearest wood.
6 Influence of the type of wind turbine
As the different types of wind turbines (axle height, rotor diameter) may have a different
effect on the occurrence of bat fatalities, this will be analysed in detail. A total of 113
fatalities have been analysed: 51 from Brandenburg, 59 from Saxony, 2 from Rhineland-
Palatinate, and 1 from North Rhine-Westphalia. As complete information was not available in
every case, slight differences occur in the size of samples (number of WTs checked, number
of checks per WT). For the purposes of this analysis we must also take into account the fact
that the season, the exact position and the site environment, sometimes unknown to the
authors, play an important part in the results of the findings and therefore in the classification
of the wind turbine types. For this reason the data must be considered as no more than a rough
outline! In general an average number of 1 dead bat was found for every 25 checks.
7 Rotor axle height
The distribution of bat fatalities for the 292 checked wind turbines (2982 controls) with a
rotor axle between 31 and 110 m in height (fig.3) shows that bat mortality occurs at all types
of wind turbine with an axle height of over 51 m. No mortality occurred when the rotor axle
was lower than 50 m.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
7
8 Rotor diameter
The measurements of the rotor diameter of 291 wind turbines were noted for 2981 checks.
The distribution of victims shows that deaths occur with nearly every size of rotor between 21
and 90 m (fig. 4). Only for the largest rotors was no victim recorded, but only 6 controls for
these were carried out (and it takes 25 checks to find 1 dead bat, see above).
9 Discussion
For 3 years now the phenomenon of bat mortality through wind farms has been discussed in
Germany. Up to now it has not been possible to identify the real reasons for these losses.
Reasons mentioned in discussion are
- the siting of the wind farm
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
8
- the immediate surroundings of the wind turbines
- the type of wind turbine
- the season when dead bats are found
- the reason for the collisions.
It is probably a combination of several factors which contributes to mortality. We will try here
to shed some light on these different factors.
The first wind farms were built along the coast at the end of the 1980s. They consisted of
wind turbines which in most cases had a rotor axle height of under 40 m. The first intensive
study of existing wind farms was made by VAUK et al. (1990), commissioned by the BMFT.
The aim of this work was to measure the possible effects on bird populations but mammals
and even bats had also to be taken into account, but to a lesser extent. From this point of view
9 wind farms were partially studied for 17 months. It was found that the wind turbines were
always responsible for bird fatalities. But as these wind farms were situated in open landscape
without hedges (except in Bredstedt, North Friesland, where a hedge came close to wind
turbines), there was no hint of negative impact on bats and none was found dead or injured.
Nor were any dead bats found during other studies of bird mortality, no dead bat was
discovered (e.g. Bremerhaven [SCHERNER 1999] where 2 wind turbines were placed at the
confluence of the Lune and the Weser, near a clarification plant). But things changed when
wind turbines started to appear inland in habitats with more varied and richer landscape
(RAHMEL et al. 1999).
At the same time intensive studies were carried out in the USA, looking for bird fatalities
on 14 wind farms of different sizes. The results showed that unexpectedly there were even
higher numbers of dead bats than there were dead birds (JOHNSON et al 2000, KEELEY 2001).
It was estimated that, depending on the siting, there are between 5 and 319 dead bats/wind
farm/year. Even more meaningful is the estimated mortality per turbine on 7 wind farms
which fluctuates from 0.07 to 10.0, with a mean value of about 2.42 dead bats/WT/year
(JOHNSON in press). In general bat mortality concerned 61% of Lasiurus cinereus and 17% of
Lasiurus borealis, two high-flying species that can be compared with the European species of
the genera Nyctalus. The four other species found were less numerous and were bats which
usually forage in heights less than 10 m (BARCLAY 1984, FENTON & BELL 1979). More recent
American studies underline that species usually known to be low-flying ones are also often
victims of collision with WTs (BOONE, pers. comm.). These studies have revealed 475 dead
bats at just one wind farm of 44 turbines from April to November 2003 (weekly checks).
BOONE estimated the number of bats which have been killed there during 2003 of about 3000
individuals. Results of comparable studies carried out over a period of at least 1 year are
being processed in Germany at the moment (MÖCKEL and WIESMER, pers. comm.). The
analysis of German studies that we presented here show that in Europe also highflying species
are not the only bats affected, although they represent the majority of the victims, but that
species which forage near structures, such as Long-eared bats (Plecotus spec.) are also
concerned.
Do the features of the surrounding landscape therefore play a part in the risk of collision
for bats? A first analysis of the available data shows that dead Pipistrellus nathusii and
Pipistrellus pipistrellus usually appeared around wind turbines which were close to
hedgerows. In the case of Nyctalus, the results are, as expected, less obvious; the mean
distance between turbines with Noctule bat fatalaties and the nearest copse was about 197 m
(max. 600 m). These results were confirmed by a swedish study (AHLÉN 2002), but major
research is needed to confirm these statements. After a careful interpretation of the rather
sparse data available, a distance of 150 m between the tower base of the wind turbines and the
copse features would clearly lead to a decrease in bat mortality, as 89% of the recorded
findings are related to turbines built less than 100 m from hedgerows or forest edges.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
9
Another factor that might affect bat mortality is the type of wind turbine. Previously it had
been thought that especially small turbines could have a bigger impact on bats (RAHMEL et al
1999). Today this seems no longer to be the case. The data analysed for this study shows that
bats are also killed by tall WTs with almost any length of blade. As no other statistically
comparable studies are available, no conclusion can be drawn here about the effects that
different types of wind turbines may have. On the other hand, it may be supposed that the
siting of the wind farms is crucial. For example, VAUK et al. (1990) could not find dead bats
(see above) during their surveys on wind farms with small turbines that were situated along
the coast in large open areas, but bat mortality was not expected there due to the bareness of
the landscape. It is the development of taller wind turbines which has allowed the wind
industry to develop inland sites where the landscape is more variable and rich of structures.
This error of judgement is probably due to the fact that at present little is known about the
flight heights of bats when they are out of range of ultrasonic detectors. In addition, there is
our lack of knowledge about how collisions occur, whether if they happen during foraging,
commuting or even searching for a roost within the turbine (see below). In the american list of
victims there were also some bat species whose ecology could be compared to the European
Myotis mystacinus/brandtii (e.g. Little brown bat, Myotis luciferus) or to Myotis nattereri (e.g.
Long-eared bat, Myotis septentrionalis), which are not thought to be a species which forage
typically high in the sky.
All European and American studies have one point in common: most bat fatalities were
found during the migration period in late summer and more especially during the period of
autumn migration (ALCADE, pers. comm., JOHNSON et al. 2000, JOHNSON et al. 2003,
HEDDERGOTT pers. comm.). Dead bat findings in the USA were almost synchronous to the
German findings (JOHNSON et al. 2003). But up to this moment this has to be regarded with
caution as for many German studies it was not always obvious if the period of research
covered the full season of bat activity or was concentrated during the autumn migration, as in
Saxony (TRAPP et al. 2002, FÖRSTER 2004). In 1367 checks in Brandenburg (see above)
which include every month except January, dead bats were found from the beginning of May
to the end of November. The number of fatalities clearly increased at the beginning of August
and reached a peak at the end of August. After mid-September only, occasional findings were
recorded. Beside the temporal aspect, it is evident that in Germany also the species that were
most often found are migratory bats (AHLÉN 1997, AHLÉN et al. 2002, JOHNSON in press,
PETERSONS 1990).
During the last decades an enormous amount of knowledge has been obtained on bird
migration, especially about migratory routes, physiology and orientation, but in the case of
bats there are very serious gaps in our knowledge. Apart from the work on the ability of
homing of bats (GRIFFIN 1970, MERKEL 1980) and a study from the 1970s on possible
receptors in bats for magnetic fields (BUCHLER & WASILEWSKI 1985), there are neither
studies about the orientation of migrating bats, flight heights nor migration physiology. There
is also only a rudimentary knowledge of migration routes (AHLÉN 1997, AHLÉN et al. 2002,
PETERSONS 1990).
Up to this day the reason for collisions with wind turbines or other man-made structures
(radio masts etc.) is not understood (OSBORN et al. 1996, JOHNSON et al. 2003). According to
VAN GELDER (1956) bat collisions with such structures occur during migration and
“unfriendly” weather, like bird mortality. However the “bad weather theory” could not be
confirmed later (CRAWFORD & BAKER 1981), and studies on wind farms brought no evidence
of this (JOHNSON et al 2003, AHLÉN 2002). On the other hand during the studies on American
wind plants, many bats have been killed by wind turbines but not by the weather towers of the
respective wind plants (JOHNSON, e-mail). It seems, that during migration, bats are less likely
to navigate by ultrasound, preferring other methods of orientation (CRAWFORD & BAKER 1981,
EKLÖF 2003, GRIFFIN 1970, MUELLER 1966, VAN GELDER 1956). It is assumed that during
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
10
migration, bats orientate by visual means, though this is unconfirmed. On a wind farm in
Wyoming, for example, only 20 ultrasonic signals of Lasiurus cinereus have been recorded at
wind turbines, but the species represented 88% of the victims. On the other hand in Minnesota
the mortality rate could be correlated to bat activity deduced from echolocation. Off course
visual orientation allows them to see the turbines in time, but is their eyesight good enough, in
the prevailing light conditions, to spot the rotating blades and identify it as a danger?
Furthermore, during migration the animals cross regions that they do not know as well as their
summer foraging habitats. The combination of these phenomena implies then that obstacles
such as wind turbines will be perceived, either too late or not at all. But then the question
arises why the percentage of dead bats is so much higher than the percentage of dead birds?
The results presented here, as well as the list of species found in Europe and in the USA
show that migration is not the only reason for bat mortality, as a series of non-migratory
species is also to be found among the victims, and fatalities occur also, even to a lesser extent,
at the beginning of the summer. It means that other factors must play a role. With a thermal
imaging camera, AHLÉN (2002) observed on Gotland that e.g. Northern bats (Eptesicus
nilssonii) were foraging in autumn around rotors and nacelles. During foraging one Northern
bat was fatally hit by a blade. The reason for this behaviour was the presence around the
nacelle of warm air which attracted insects from the much cooler surrounding environment.
This phenomenon is to be expected mainly in the autumn. Bats are real opportunists when it
comes to feeding and will take advantage in the short term of this abundant food supply. We
cannot agree with HENSEN’S theory (in press) that this phenomenon only occurs on warm
nights, thus explaining the high mortality rate in August. There is no particular concentration
of insects round the nacelle on warm nights, which means that bats have no incentive to
forage around it. The explanation is rather that during warm nights flying insects reach higher
altitudes and get stuck on the slowly rotating blades. The nacelle temperature should not have
much influence on the mortality of insects on the blades as these sweep continuously through
the air. It is possible that this behaviour plays a part in areas where bats congregate in autumn.
During detector controls at one wind plant in the USA, similar behaviour has been noted, but
no dead bat has been found. At all other wind farms there was evidence of bats passing
through but none of foraging bats (JOHNSON in press).
The main cause of death to be discussed is direct collision with the blades. This is borne
out by the numerous broken bones and wounds on the patagium shown by the animals that we
have examined. These wounds are caused by collision with a blade or by air turbulence (see
AHLÉN 2002); they are not bruises inflicted by moving parts inside the nacelle (see below).
Neither the bat’s ability to orientate by sight nor by ultrasound can realy explain the many
recorded fatalities. In principle bats can detect wind turbines. They can react to movements in
the air. They have been seen taking evasive action when pursued by hobbies or peregrine
falcons (Falco subbuteo and Falco peregrinus) (SÖMMER & HAENSEL 2003). But they may
underestimate the speed of the blades and therefore the danger. This estimation would become
more difficult with the increase of rotor diameter as the speed of the blade tips increases
accordingly. This may be another reason for the high mortality rate found with longer blades.
It is possible that bats perceive only one blade in movement without noticing the next one
coming from above or from underneath.
Another kind of mortality, not yet studied in detail but involving a large number of deaths
by implosion, seems to be caused by low pressure fields (e.g. when a blade passes close to the
tower, or as a result of the huge difference in pressure between the front and the rear of the
blades). In the findings from Saxony published by TRAPP et al. (2002), about 20% of the bats
had very oily fur. This could also be explained by contact with the hydraulic oil of the turbine,
or possibly by the presence of oil particles in the air around the nacelle. No other “very oily”
bats have been recorded in other studies. Up to now it has been assumed that bat implosion in
low pressure fields would cause the “body fat” to ooze through the skin pores, thus giving the
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
11
animals an oily appearance. The results of the analysis of the Saxony animals by ZINKE in the
Oberlausitz Museum at Kamenz make poisoning by hydraulic oil very improbable. This
analysis shows that death was due to low pressure fields (FÖRSTER, pers. comm.). On the
other hand quite a lot of animals with no external wounds show internal haemorrhaging
whose origin could also be explained by the fact that the animals came too close to low
pressure fields. This phenomenon obviously awaits a definitive explanation!
Another kind of accident could result from bats trying to find a roost within the nacelle
(AHLÉN 2002, HENSEN in press). Using credible technical parameters HENSEN launched a
debate based on the fact that the majority of fatalities occurred during the least windy period
(August). Thus Noctule bats (Nyctalus noctula) looking for a roost inside the nacelle seem
indeed to succumb to injuries inflicted by the gearbox of the turbine. But the use of the
nacelle as a roost seems still to be an exceptional occurrence as only a single finding of a
bruised bat has been reported (see above). It is however possible that searching for a roost in
wind turbines ends in death because, strangely enough, the majority of fatalities start when
colonies disperse and the search begins for temporary roosts.
Apart from these obvious cases, we have little information on the reasons for bat mortality.
This is true especially for all estimates of the impact on bat populations. It is even more
difficult because we do not know whether the victims are resident bats or migrating ones. And
in the latter case we know very little about the phenomenon of bat migration and even less
about their numbers (see above).
10 Conclusion
The unsatisfactory state of our knowledge of the real factors which result in bat mortality
demands immediate investigation to establish these factors. This can only be achieved by a
close co-operation with the wind turbine industry (manufacturers and power generating
companies). We must investigate more closely not only the siting but also the weather
conditions and the technical parameters of the different wind farms. All the findings must be
documented as precisely as possible, all the information being forwarded to the central
databank. To set up a uniform method of documenting the victims, we have formulated a
questionnaire (.ppt-file) which may be obtained from the Staatliche Vogelschutzwarte of
Brandenburg.
JOHNSON et al (2003) found 54% of the victims within a radius of 10 m, 43% within 10-
20m and only 0.5% > 50 m. The data from Germany are slightly different (90% of all victims
< 30 m). This could be due to the type of wind turbines installed in Germany as they are much
taller and have appreciably longer blades.
11 Proposal for checking bat mortality at wind turbines
As shown above, bat mortality occurs around all types of wind turbines and in various regions
of Germany and Europe. Until now, as the victims were found mainly by accident and/or
during semi-systematic studies, no definite conclusion can be drawn about the influence of
particular types of wind turbine, their distance from wooded areas etc. However, such
conclusions must be reached if we are going to be able to deal with the problem of bat
mortality and e.g. to react in case of impact regulations. It is only if the problem of bat
mortality is studied systematically that we will be able to look for solutions.
All this means that in future we should spend less time on unsystematic and occasional
collecting of dead bats under wind turbines – that bat mortality exists has already been proved
– but systematic studies using consistent methods must be realised in the different regions of
Germany. We must also launch some experimental studies to find out how bats recognise the
presence of WTs and investigate the phenomenon of bat migration, especially their migratory
routes.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
12
How should a research project of bat mortality be set up? The aim is not only to document
the death of bats, but also to find out if the following factors have an influence:
- the different types of turbines
- the general environment (distance from copes, water, waste dumps, foraging habitats,
etc.)
- the siting of the turbines (migration route, proximity of roosts and foraging habitats,
etc.)
- the season,
Above all the aim is to find out if WT may have an influence on bat populations. In each case
it is important, in addition to the collecting of victims, to evaluate bat activity around the wind
farm. Only then will it be really possible to compare different studies.
Species BB SAH SN TH SH BY NRW RP NDS Total
Nyctalus noctula Noctule bat 39 1 20 54 3 1 1 118
Nyctalus leisleri Leisler’s bat 5 1 1 3 10
Eptesicus serotinus Serotine bat 2 2 1 1 6
Vespertilio murinus Parti-coloured bat 1 7 8
Myotis myotis Greater mouse-eared bat 7 7
Myotis daubentonii Daubenton’s bat 1 1
Plecotus austriacus Grey long-eared bat 1 1
Pipistrellus pipistrellus Common pipistrelle 15 2 6 2 25
Pipistrellus nathusii Nathusius’ pipistrelle 17 1 23 2 1 44
Pipistrellus pygmaeus Soprano pipistrelle 1 1
Pipistrellus sp. Pipistrelle sp. 4 14 18
Chiroptera sp. Chiroptera sp. 2 2 4
TOTAL 85 5 59 70 18 1 3 2 0 243
12 Draft protocol for studying bat mortality (search for victims)
12.1 General remarks
When it comes to choosing the study areas one should ascertain that bat data collected before
the construction of the turbines is available for the region. The studies must account all
regions of Germany and the most diverse habitats (open land, landscapes with hedges, forest)
and all the different types of turbines.
The method used for each study must be described in detail. This means also describing
very carefully the study area, the types of wind turbines, the siting of each and the method
and intensity of the search (person or person + dog) etc It should be noted that the use of
different persons during the search phase may result in discrepancies in discovery rates. And
it is important that all studies cover the whole “summer season” from April to October.
12.2 Profile of the study
Bat activity
Bat activity must be monitored according to the method described by RAHMEL et al. (2004)
for the impact assessment. This means 7 all-night surveys (April-September) to assess
resident bat populations and 16 half-night surveys to study the bat migration. Automatic
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
13
recording units (“automatical bat registration boxes”) must also be used (see RAHMEL et al.
2004).
Search for victims
From April to mid-October all wind turbines must be checked once a week in the early
morning looking for dead bats. Each study area should have a minimum of 10-15 turbines. At
least 5 WTs must be checked every morning, the radius of the search area being the radius of
the blades + 50 m. For each wind turbine, individual protocols should be written with site
description, findings, etc.. If possible searches should be combined with nights when bat
activity is being monitored.
13 Acknowledgements
For the communication of their findings data we are particularly obliged to T. Blohm, R.
Brinkmann, D. Dolch, A. Ewert, A. Fölling, F. Förster, M & M. Göttsche, R. Günter, M.
Heddergott, M. Heiss, M. Hirt, L. Ittermann, L. Manzke, R. Möckel, G. Nessing, M. Putze, R.
Reifenrath, P. Schonert, H. Trapp and P. Wiesner. We express our thanks to D. Dolch (Nature
Conservation Station of Zippelsförde) who verified a great number of findings from
Brandenburg. We also thank the services for the protection against emissions in
Brandenburg/Havel, Cottbus, Frankfurt/Oder, Neuruppin and Wünsdorf for their excellent
support in our search for the data related to the types of different wind turbines. Without
forgetting P. Burkhardt and U. Rahmel for their comments on our script.
14 References
AHLEN, I. (1997): Migratory behaviour of bats at south Swedish coasts. - Z. Säugetierk. 62:
375-380.
AHLEN, I. (2002): Fladdermöss och fäglar dödade av vindkraftverk. - Fauna och Flora
97(3):14-22
ALCALDE, J.T. (2003). Impacto de los parques eolicos sobre las poblaciones de murciélagos.
- Barbastella 2: 3-6.
ANONYMUS (1999): Abstract of windpower-related mortality at Buffalo Ridge, Minnesota
1996-1999. - unveröff. Manuskript: 4 S..
BACH, L. (2001): Fledermäuse und Windenergienutzung – reale Probleme oder Einbildung?
- Vogelkdl. Ber. Niedersachs. 33: 119-124.
BACH, L., R. BRINKMANN, H. LIMPENS, U. RAHMEL, M. REICHENBACH & A.
ROSCHEN (1999): Bewertung und planerische Umsetzung von Fledermausdaten im
Rahmen der Windkraftplanung. - Bremer Beiträge für Naturkunde und Naturschutz 4:
162-170.
BARCLAY, R.M.R. (1984): Observations on the migration, ecology and behaviour of bats at
Delta Marsh, Manitoba. – Can. Field Nat. 98:331-336.
BRINKMANN, R., H. SCHAUER-WEISSHAHN & F. BONTADINA (2005):
Untersuchungen zu möglichen betriebsbedingten Auswirkungen von Windkraftanlagen
auf
Fledermäuse im Regierungsbezirk Freiburg – unveröff. Gutachten
BUCHLER, E.R. & P.J. WASILEWSKI (1985): Magnetic remanence in bats. – in:
KIRSCHVINK, J.L., D.S. CRAWFORD, R.L. & W.W. BAKER (1981): Bats killed at a
north Florida Television tower: a 25-year record. – J. Mammal. 62: 651-652.
DÜRR, T. (2001): Fledermäuse als Opfer von Windkraftanlagen. – Naturschutz und
Landschaftspflege in Brandenburg 10: 182.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
14
EKLÖF, J. (2003): Vision in echolocating bats. – Dissertation an der Univ. Göteborg: 106 S..
FENTON, M.B. & G.P. BELL. (1979): Echolocation and feeding behaviour in four
species of Myotis (Chiroptera). - Can. J. Zool. 57: 1271-1277.
FÖRSTER, F. (2004): Windkraftanlagen und Fledermausschutz in der Oberlausitz.
Manuskript zum Tagungsband der Fachtagung „Kommen die Vögel und Fledermäuse
unter die Wind(räder)?“, 17./18.11.2003, TU Dresden: 15 S.
GRIFFIN, D.R. (1970): Migration and homing in bats. – in: WIMSATT, W.A. (Hrsg.):
Biology of bats. – Academic Press, New York: 233-264.
HALL, L.S. & G.C. RICHARDS (1972): Notes on Tadarida australis (Chiroptera:
Molossidae). - Australian Mammalogy 1: 46.
HENSEN, F. (im Druck). Gedanken und Arbeitshypothesen zur Fledermausverträglichkeit
von Windenergieanlagen.
JOHNSON, G.D., W.P. ERICKSON, M.D. STRICKLAND, M.F. SHEPHERD & D.A.
SHEPHERD (2000): Avian monitoring studies at the Buffalo Ridge, Minnesota Wind
Resource Area: Results of a 4-year study. – unveröff. Bericht der Northern States Power
Company, Minnesota: 262 pp.
JOHNSON, G.D., W.P. ERICKSON, M.D. STRICKLAND, M.F. SHEPHERD & D.A.
SHEPHERD (2003): Mortality of bats at a Large-scale wind power development at
Buffalo Ridge, Minnesota. – Am. Midl. Nat.150: 332-342.
JOHNSON, G.D. (in pres.): What is known and not known about impacts on bats? –
Proceedings of the avian interactions with wind power structures, Lackson Hole, Wyoming,
KEELEY, B.W., S. UGORETZ & D. STRICKLAND (2001): Bat Ecology and Wind Turbine
Considerations. – in: NATIONAL WIND COORDINATING COMMITTEE (Hrsg.)
National Avian-Wind Power Planning Meeting 4, Washington, D.C.: 135-146.
KUGELSCHAFTER, K. & A. ROSCHEN (2004, i. d. Bd.): Wieso verunglücken
Fledermäuse an Windkraftanlagen - eine theoretische Annäherung an eine komplexe
Thematik - Bremer Beiträge für Naturkunde und Naturschutz, Band
LEKUONA, J. M. (2001): Uso del Espacio por la Avifauna y Control de la mortalidad de
Aves y Murciélagos en Los Parques Eolicos de Navarra durante un Ciclo anual.
Direccion General de Medio Ambiente Departmento de Medio Ambiente, Ordenacion
del Territorio y Vivienda. Gobierno de Navarra: 1-147.
MERKEL, F.W. (1980): Orientierung im Tierreich. - Grundbegriffe der modernen Biologie
15, Gustav Fischer Verlag, Stuttgart.
MUELLER, H.C. (1966): Homing and distance-orientation in bats. – Z. Tierpsychol. 23: 403-
421.
OSBORNE, R.G., K.F. HIGGINS, C.D. DIETER & R.E. USGAARD (1996): Bat collisions
with wind turbines in Southwestern Minnesota. - Bat Research News 37: 105-108.
PETERSONS, G. (1990): Die Rauhhautfledermaus, Pipistrellus nathusii (Keyserling u.
Blasius, 1839), in Lettland: Vorkommen, Phänologie und Migration. - Nyctalus 3: 81-98.
RAHMEL, U., L. BACH, R. BRINKMANN, C. DENSE, H. LIMPENS, G. MÄSCHER, M.
REICHENBACH & A. ROSCHEN (1999): Windkraftplanung und Fledermäuse.
Konfliktfelder und Hinweise zur Erfassungsmethodik. – Bremer Beiträge für Naturkunde
und
Naturschutz, Band 4: 155-161.
RAHMEL, U., R. BRINKMANN, H. LIMPENS, & A. ROSCHEN (2004, i. d. Bd.):
Windenergieanlagen und Fledermäuse – Hinweise zur Erfassungsmethodik. – Bremer
Beiträge für Naturkunde und Naturschutz, Band
SCHERNER, E. R.. (1999): Windkraftanlagen und „wertgebende Vogelbestände“ bei
Bremerhaven: Realität oder Realsatire? – Beitr. Naturk. Niedersachsen 52 (4): 121-156.
Bremer Beiträge für Naturkunde und Naturschutz, 2004, Volume 7, pp. 253-264
15
SÖMMER, P. & J. HAENSEL (2003): Fledermäuse als Beute von Taggreifvögeln –
überraschende neue Befunde besonders für die beiden schnellsten deutschen Falkenarten.
Nyctalus (N.F.), Berlin (9) 1: 61-78.
TRAPP, H., D. FABIAN, F. FÖRSTER & O. ZINKE (2002): Fledermausverluste in einem
Windpark der Oberlausitz. – Naturschutzarbeit in Sachsen 44: 53-56.
VAN GELDER, R.G. (1956): Echo-location failure in migratory bats. - Transactions of
Kansas Academy of Science 59: 220-222.
VAUK, G., M. BÖTTGER, T. CLEMENS, G. GROTE, G. HARTMANN, E. HARTWIG, C.
LAMMEN u. E. VAUK-HENTZELT (1990): Biolog.-ökol. Begleituntersuchung zum
Bau und Betrieb von WEA. Endbericht. NNA-Ber. (3) Sonderheft: 3-124.
VIERHAUS, H. (2000): Neues von unseren Fledermäusen. - ABU Info 24 (1): 58-60.
