Migrating bats at the southern North Sea Approach to an estimation of migration populations of bats at the southern North Sea

Full text

2016/2017 - Technical Report Zoogdiervereniging (Dutch Mammal
Society) in collaboration with Wageningen Marine Research
Migrating bats
at the southern North Sea
Approach to an estimation of migration populations of bats at
the southern North Sea
Limpens, H.J.G.A., S. Lagerveld, I. Ahlén, D. Anxionnat,
T. Aughney, H.J. Baagøe, , L. Bach, P. Bach, J.P.C. Boshamer,
K. Boughey, T. Le Campion, M. Christensen, T. Douma,
M.-J. Dubourg-Savage, J. Durinck, M. Elmeros, A-J. Haarsma,
J. Haddow, D. Hargreaves, J. Hurst, E.A. Jansen,
T.W. Johansen, J. de Jong, D. Jouan, J. van der Kooij,
E.-M. Kyheroinen, F. Mathews, T.C. Michaelsen, J.D. Møller,
G. Pētersons, N. Roche, L. Rodrigues , J. Russ, Q. Smits ,
S. Swift, E.T. Fjederholt, P. Twisk, B. Vandendriesche &
M.J. Schillemans.

Rapport nr.:
2016.031
Datum uitgave:
Maart 2017
Status
Definitief 08-03-2017
Authors/contributors:
Limpens, H.J.G.A., S. Lagerveld, I. Ahlén, D. Anxionnat, T.
Aughney, H.J. Baagøe, , L. Bach, P. Bach, J.P.C. Boshamer, K.
Boughey, T. Le Campion, M. Christensen, T. Douma, M.-J.
Dubourg-Savage, J. Durinck, M. Elmeros, A.-J. Haarsma, J.
Haddow, D. Hargreaves, J. Hurst, E.A. Jansen, T.W. Johansen,
J. de Jong, D. Jouan, J. van der Kooij, E.-M. Kyheroinen, F.
Mathews, T.C. Michaelsen, J.D. Møller, G. Pētersons, N. Roche,
L. Rodrigues , J. Russ, Q. Smits , S. Swift, E. T. Fjederholt, P.
Twisk, B. Vandendriesche & M.J. Schillemans
Kwaliteitscontrole:
Marcel Schillemans
Productie:
Steunstichting VZZ, in rapport vermeld als de
Zoogdiervereniging
Bezoekadres:
Postadres:
Toernooiveld 1
6525 ED Nijmegen
Postbus 6531
6503 GA Nijmegen
Tel.: 024 7410500
secretariaat@zoogdiervereniging.nl
www.zoogdiervereniging.nl
Gegevens opdrachtgever:
Wageningen Marine Research
Ankerpark 27
1781 AG Den Helder
Contactpersoon opdrachtgever
Michaela Scholl
De Steunstichting VZZ is onderdeel van de Zoogdiervereniging
Dit rapport kan geciteerd worden als:
Limpens, H.J.G.A., S. Lagerveld, I. Ahlén, D. Anxionnat, T. Aughney, H.J. Baagøe, , L. Bach, P. Bach, J.P.C.
Boshamer, K. Boughey, T. Le Campion, M. Christensen, J.J.A. Dekker, T. Douma, M.-J. Dubourg-Savage, J.
Durinck, M. Elmeros, A.-J. Haarsma, J. Haddow, D. Hargreaves, J. Hurst, E.A. Jansen, T.W. Johansen, J. de
Jong, D. Jouan, J. van der Kooij, E.-M. Kyheroinen, F. Mathews T.C. Michaelsen, J.D. Møller, G. Pētersons, N.
Roche, L. Rodrigues , J. Russ, Q. Smits , S. Swift, E.T. Fjederholt, P. Twisk, B. Vandendriesche & M.J.
Schillemans, 2017. Migrating bats at the southern North Sea - Approach to an estimation of migration
populations of bats at southern North Sea . Rapport 2016.031. Zoogdiervereniging (Dutch Mammal Society),
Nijmegen/ Wageningen Marine Research .
Migrating bats at the southern North Sea
Approach to an estimation of migration populations of bats at
the southern North Sea

Contents
Summary ........................................................................................................... 1
Reading guide .................................................................................................... 3
1. Introduction ................................................................................. 4
1.1 Context ................................................................................................... 4
1.2 Aim of the study ....................................................................................... 5
2. Method ......................................................................................... 6
3. Design process ............................................................................. 8
3.1 Participants .............................................................................................10
3.2 Choice of relevant species for the estimator ................................................10
3.3 B] Basic regional data and estimates ..........................................................11
3.4 A] The flow model approach ......................................................................15
4. Results: Prototype estimator of populations of migrating bats .. 23
4.1 Feedback basic regional information and data: ............................................23
4.2 Feedback information and data on flow model .............................................23
4.3 Results A] flow model based on regional data and estimates B] .....................26
5. Conclusions and discussion ........................................................ 29
5.1 B] regional data and estimates ..................................................................29
5.2 A] The flow model approach ......................................................................30
5.3 How to interpret or use the outcome ..........................................................31
6. Knowledge gaps and priority research questions ....................... 33
7. References ................................................................................. 34
7.1 References reporting and basic data ...........................................................34
7.2 Literature basic regional data ....................................................................36
7.3 Websites regional data .............................................................................38

8. Annexes ..................................................................................... 39
Annex 1:Questionnaire to collect (available) basic regional information
and data ................................................................................................ 40
Annex 2: Assessment of relevant species .............................................. 46
Annex 3: Summary feedback B] basic regional information and data .... 51
Annex 5: Summary of the complex geographic interconnection between
distribution, mating/reproduction and migration of the Nathusius’
Pipistrelle .............................................................................................. 59
Annex 6: Summary feedback on A] flow model ..................................... 60
Annex 7: Analysis of the reaction of the prototype estimator (the flow
model A]) through using a range of settings for the number of
juveniles/female and satellite males/male. .......................................... 63
Annex 8: Preliminary estimates for migrating Nathusius’ pipistrelle over
the southern North Sea - histograms of test runs status January 2017 67

Estimating the migration populations of bats on SNS
1
Summary
Bats experience a fatality risk at wind turbines. Since some species migrate not
only over land, but also over sea, offshore turbines might pose a threat. To be
able to assess the impact of potential added mortality at sea, it is necessary to
know or estimate the population size of bats, including the portion of the
population crossing the sea. Potentially relevant species are Nathusius’ Pipistrelle
(Pipistrellus nathusii), Common Noctule (Nyctalus noctula) and Particoloured Bat
(Vespertilio murinus).
The current study aims at developing a prototype estimator for migrating
populations of bats. This is based on data, or estimates, regarding the size and
bandwidth of source populations, population dynamical factors defining such
populations, and factors defining migration fluxes. Acknowledging the rareness of
such data, a flow model is constructed targeting a preliminary estimate for the
southern North Sea (SNS). However, the approach can be adapted for use in
other regions/study areas as well.
The model is based on available information and data regarding bat species in
the different countries bordering the SNS. This includes countries further off,
which might be ‘source countries’ of bats that eventually might fly on the SNS.
The approach to the estimation of basic data, as well as the flow model, was a
cyclical process together with members of a design team from these countries.
Together with other contributing colleagues from their region, information and
data were brought together. Information on the non-availability of data, or
knowledge gaps, are an important part of the result.
The flow model consists of a series of interconnected excel sheets incorporating
the basic data for the countries and fluxes. In the current model, parameters
used per country are: population size of males, females and juveniles (M, F and
J), the percentage of migrating individuals (M, F and J), percentage migrating
towards different connected countries in the west/southwest direction,
percentage migrating over land/sea, as well as generic factor parameters: basic
population dynamical factors such as J/F and satellite males/male.
Currently the model focuses on the Nathusius’ Pipistrelle, because this is the
species where some information and data are available. Even for this species
quantitative basic estimates on source populations were only available for RO
Ireland, UK (specifically for England and northern Ireland) and the Netherlands.

Estimating the migration populations of bats on SNS
2
The model produces a preliminary estimate for bats crossing the SNS of roughly
40.000 individuals with a bandwidth between 100 and 1.000.000 individuals.
The accuracy of this outcome can (and must) be improved through assessment
of (more accurate) data and/or estimates per country/region to improve the
different factor components per country, to define the now generic factor
components as components per country, and to incorporate mortality during
migration.

Estimating the migration populations of bats on SNS
3
Reading guide
In chapter 1, introduction, the context of bats over the southern North Sea and
aim of the study, targeting at an estimator for source populations and migrating
populations, are explained.
Note: To facilitate easy reading for the non-bat-specialist, the content of the
chapters is focused on the estimator, where background information is provided
in the annexes. This leads to some repetition of text in the Annexes.
Note: some of the literature references will be found in the more elaborate text
on specific topics in the annexes.
Chapter 2, method, deals with the process of designing a prototype estimator
with concurrence of a team of bat experts in the relevant north west European
region. The species to use for the design are selected. The methods for
estimating basic input parameters, as well as a flow model for the in and
outfluxes of migration connecting the countries in the relevant, area are
discussed.
Chapter 3, results, describes the approbation of the design team and other
contributors to the approach. The current concrete state of the model is given.
The information used for input, and available/non-available data for the different
countries is described.
In chapter 4, conclusions and discussion, an analysis of the current state of the
estimator is given. Knowledge gaps are identified. The output values in the
current state of the model are interpreted.
Chapter 5 lists the knowledge gaps and priority research questions.

Estimating the migration populations of bats on SNS
4
1. Introduction
The main research question is to estimate migrating populations of bats over the
southern North Sea. Also estimating the number of bats of the source population
from where the migrating bats originate, is an important research goal.
1.1 Context
Different studies in Europe and the USA reveal bat fatalities at wind turbines on
land and show the potential of high numbers of fatalities (Brinkmann et al. 2011,
Voigt et al. 2015). Fatalities are occurring from direct collisions with the rotor
blades as well as through barotrauma as a result of the low pressure and
dynamics in pressure in the air turbulence near the rotor blades (Brinkmann et
al. 2011, Lehnert et al. 2014, Voigt et al. 2015). Some bat species are observed
at sea at the southern North Sea (SNS), which might reflect migration as well as
foraging (Lagerveld et al. 2014a, Peterson et al. 2014). Although carcasses are
not expected to be found at offshore turbines, there is no reason not to expect
collisions and/or barotrauma with respect to bats and wind turbines at sea.
Bats are frequently recorded in the study area: the southern North Sea (study
area as defined in Leopold et al. 2014). Observers of bird migration at the Dutch
coast regularly report bats flying in from sea and there have also been offshore
observations during ship-based surveys (e.g. Ahlén et al. 2009, Hobbs 2014,
Lagerveld et al. 2014b). In addition, bats have been found on oil and gas
platforms, ships and remote islands (Ahlén et al. 2007, Walter et al. 2007,
Boshamer & Bekker 2008, Petersen et al. 2014, Rydell et al. 2014). Studies with
passive acoustic recorders in 2012, 2013, 2014 and 2015 off the Dutch coast
revealed that bats are recorded at every location where a bat detector was
installed offshore (Jonge Poerink et al. 2013, Lagerveld et al. 2014a, 2015 &
2016).
The most common species in the study area is Nathusius’ Pipistrelle (Pipistrellus
nathusii). Common Noctule (Nyctalus noctula) and Particoloured Bat (Vespertilio
murinus) also probably occur regularly at the southern North Sea. Other species
like Common Pipistrelle (Pipistrellus pipistrellus), Northern Bat (Eptesicus
nilssonii), Serotine Bat (Eptesicus serotinus) and Leisler’s Bat (Nyctalus leisleri)
have all been observed at the southern North Sea, but are likely to be more
occasional visitors or vagrants (Hüppop & Hill 2016, Lagerveld et al. 2014a,
Leopold et al. 2014).
Nathusius’ Pipistrelle, Common Noctule and Particoloured Bat are migratory
species which may cover large distances during migration (e.g. Hutterer et al.
2005, Roer 1995). Most offshore bat activity occurs during autumn from late

Estimating the migration populations of bats on SNS
5
August until early October and to a lesser extent from late March until late May.
Records in June and July are very scarce. The observed pattern of occurrence in
combination with the species composition strongly suggests that offshore bat
activity is caused by migrants (Hüppop & Hill 2016, Leopold et al. 2014).
Most bat activity occurs during nights with low wind speeds, high atmospheric
pressure and no rain. Therefore, it seems unlikely that bat activity at sea is
predominantly caused by individuals blown offshore (Lagerveld et al. 2014b).
This corresponds with the findings of Ahlén et al. (2007, 2009) who observed
that migrating bats aggregate at coastal locations and wait for favourable
conditions to cross over the Baltic Sea. However, Hüppop & Hill (2016) show that
migration occurs both with tailwind and headwind.
Several ringing recoveries of Nathusius’ Pipistrelles have shown that they are
able to successfully cross the North Sea1. In addition these recoveries (n=4)
show that the migration direction of these individuals is roughly from east
northeast to west southwest.
Leopold et al. 2014 concluded that, in the light of the increasing area and
numbers of wind turbines on the southern North Sea, negative effects on bat
populations, at least for the Nathusius’ Pipistrelle (Pipistrellus nathusii), and
possibly also for the Common Noctule (Nyctalus noctula) and the Particoloured
Bat (Vespertilio murinus) cannot be excluded.
1.2 Aim of the study
The ultimate aim of this study is to be able, in the future, to assess the number
of bats crossing the SNS, in comparison to the number of bats residing in the
various countries.
To be able to do so, population estimates (paragraph 2.1 and 2.4) for the
different regions, as well as a practically applicable numerical model for the
migration flux (paragraph 2.1 and 2.5) are needed. This study aims at
developing a prototype estimator, regarding national/regional estimates and the
migration flux and assessing the availability of the necessary data. The process
of developing such a prototype estimator will highlight knowledge gaps. In
principle, the approach targets all bat species that regularly occur on the North
Sea. Availability of data will determine which species practically can be used for
the first steps towards estimating migrating populations of bats.
1 see e.g. (http://www.bats.org.uk/pages/national_nathusius_pipistrelle_project.html).

Estimating the migration populations of bats on SNS
6
2. Method
To be able to estimate the number of individuals of a bat species that might be
migrating across the SNS, information is needed regarding the factors defining
the population size and their movements.
We worked with an approach where we try to A] model the migration flux, based
on B] available data. The available data consist of either quantitative or
estimated information of the parameters defining the population dynamics, as
well as the migration, for the different regions in the relevant geographical
population/migration area for the species (Figure 1).
Figure 1: principle of combination of B] regional basic data as input for A] the flow model
Combined, the components of the basic regional data and the flow model build
the basis for the assessment of the migration pattern of the relevant species in
Europe. All different components of the model and estimations will have their
own specific and probably large uncertainty intervals.

Estimating the migration populations of bats on SNS
7
Leading questions2 for the structure of model are listed below.
• What are the source populations of bats potentially migrating over the SNS?
• What are the population sizes of the source populations? What quantitative
information is available on numbers of different roost types (maternity,
summer, mating and hibernation roosts), numbers using these (networks) of
roosts? How is this reflected in numbers of males, females and offspring?
• What are the approximate directions, areas of migration and landscapes used
for migration of the source populations?
• What is the number of animals, or parts of the source populations that migrate
through these areas and landscapes? What is the number of individuals
(males, females, offspring) that do not take part in migration?
• Is there a more or less random migration flow across (the southern North)
Sea, or is there an – observable - differentiation between following the coast
lines and migrating over sea? What might be the size of sub-populations
migrating across the sea and those following the coast line?
• What might be the overall size of the population migrating over the SNS?
The construction of a flow model and discussions regarding the input and
outcome of several test runs are an effective way of producing first estimates as
well as of detecting knowledge gaps. This is why regional bat specialists were
invited to take part in the model development. They were asked to provide
feedback on the structure of the flow model and as a source of information, to
help estimate and/or give their expert judgement regarding chosen values.
Both the A] flow model and B] estimates for basic regional input should be
usable in an iterative approach regarding the estimates. Chosen or estimated
values of one parameter influence the outcome of others and provide feedback
that can be validated. This process starts with expert judgement and in the end
makes clear what research is needed for further ground truthing.
In working with this approach it is logical to think from the view point of the flow
model, which needs to be based on basic regional data. In the development of
this integrated approach, however, the development of the approach to B]
assessing basic regional information, preceded the development and testing of A]
the flow model. Therefore in our reporting we start with B] (paragraph 2.4)
followed by A] (paragraph 2.5).
2 In this study we are not necessarily providing answers to all of these questions, or answering them
completely.

Estimating the migration populations of bats on SNS
8
3. Design process
The design process is summarized in Table 1.
A number of European bat researchers was contacted and invited to take part in
the team for the first design and development of the estimator B] for the basic
regional data, and later on A] the flow model. They were also asked to bring
together available data and information via their network in their regions.
Together we have been designing and developing a prototype estimator.
In a first loop, a draft approach for the B] estimation of the basic regional data
and a first sketch of the flow model were developed. They were presented in the
form of a PowerPoint (e.g. illustration of approach geographically) and
Questionnaire (Excel-table for collecting information (see Annex I). These first
sketches were discussed with the members of the design team. Comments and
remarks were processed.
The approach to the B] estimation of the basic regional data and the
questionnaire was then distributed, directly and/or via the members of the
design team, to an extended number of bat experts3 in their region(s) and
countries.
The first round of feedback and the first basic information from the participants
were then used to specify the design of A] the flow model approach.
Next, the basic data and information on non-availability of data, received from
the participants, were used as first input for the flow model. The flow model – in
the form of a series of interrelated excel sheets – was then used to collect new
feedback on the flow model approach and on the chosen estimates.
Note: designing and building of the model, as well as gathering input, involved
repeated feedback (Table 1). The main ingredients of this cyclical process, more
specifically of the design process, e.g. relevant species, participants, premises for
the model, selection of countries/regions to take into the model etc., are
presented in the following paragraphs of the method description.
3 The data and information brought together will be made available to all participants. All contributors are co-
authors of this technical report.

Estimating the migration populations of bats on SNS
9
Table 1: Time line of the design process
2016
05 Recruitment of participants design team
05 / 06 Design of B] estimator approach for basic regional data
+ first sketch of flow model
Development of B] questionnaire for collation of basic reginal data
 
Collecting + processing feedback of members of design team
06 / 07 Recruitment of extended number of regional experts
 Distributing information on B] approach to basic regional data to all participants
 
08 / 11 Collecting input and data from all experts,
including information on (non)availability of basic data
 
10 / 12 Design of A] flow model (based on already available/non-available information)
11 / 12 Test runs of flow model with
2016/2017
12 / 01  Distributing information on A] approach to flow model to design team
 
Collecting + processing feedback of members of design team
 Distributing information on A] approach to flow model to all participants
2017
 
01 Collecting input from all participants regarding general and country/region specific
estimates which are used as input to the flow model.
2017  
01 Draft report on constructed flow model and outcome.

Estimating the migration populations of bats on SNS
10
3.1 Participants
Using our network of European bat scientists and conservationists from BatLife
Europe, the Eurobats Advisory Committee and the regular European Bat
Research Symposia we established the following expert team of participants in
the process.
Expert colleagues participating in the first loop of feedback for designing and
gathering of data:
Hans Baagøe, Lothar Bach, Katherine Boughey, Marie-Jo Dubourg-Savage, Eric
Jansen, Jeroen van der Kooij, Sander Lagerveld, Herman Limpens, Tore
Michaelsen, Gunārs Pētersons, Niamh Roche, Luisa Rodrigues, Jon Russ, Marcel
Schillemans, Esben Terp Fjederholt, Bob Vandendriessche;
Expert colleagues participating in the second loop of feedback on flow model and
basic data:
Ingemar Ahlén, Tina Aughney, Diane Anxionnat, Petra Bach, Jan Boshamer,
Thomas Le Campion, Morten Christensen, Julie Dahl Møller, Jasja Dekker, Theo
Douma, Jan Durinck, Morten Elmeros, A-J Haarsma, John Haddow, Daniel
Hargreaves, Johanna Hurst, Thomas Johansen, Johnny de Jong, Dorothee
Jouan, Eeva-Maria Kyheröinen, Fiona Mathews, Susan Swift, Peter Twisk;
3.2 Choice of relevant species for the estimator
Based on the assessment of relevant species for the estimator, presented in
Annex II, three species might be relevant for the development and use of the
flow model estimator A]. These are the Nathusius’ Pipistrelle (Pipistrellus
nathusii), the Common Noctule (Nyctalus noctula) and the Particoloured Bat
(Vespertilio murinus).
Technically the approach of the flow model is applicable to all three species.
However, the available basic information B] in the different countries is rather
poor. This is the case for the most common of the species, the Nathusius’
Pipistrelle, and even more so for the other species.
Therefore, in this study the flow model is developed using data on the Nathusius’
Pipistrelle only. We expect that the structure of the model and knowledge gaps
deduced from working with the Nathusius’ Pipistrelle, will be representative of
the situation of other species.

Estimating the migration populations of bats on SNS
11
3.3 B] Basic regional data and estimates
Input for the basic regional information, data and estimates were gathered
through a detailed questionnaire (see annex II, III and questionnaire) and
literature research.
The approach to the estimation of the size of populations and migration
movements tries to use different and independently assessable components.
These are components for which data and information might either be currently
available, or of which the numbers might currently be reliably estimated.
A basic approach to the estimation of the size of a source population, e.g. a
maternity population in one of the Baltic States, is described in Table 2.
In the same way, the size of a target population, e.g. the population of
advertising males, (clusters of) mating roosts, in the Netherlands, might be
estimated as described in Table 3.
The same approach can be used for summer roosts and hibernation sites. The
validity of the outcome is of course dependent on the available data and the sub-
estimates that we can deduct from them.

Estimating the migration populations of bats on SNS
12
Table 2: Example of the estimation of a source population. The numbers in the table are
just examples for the calculation.
Approach to quantification of source population for a specific region
Maternity colonies (network of roosts) / summer roosts
A
estimate # known maternity colonies:
15 - 35 - 60
roughly 1/3
B
estimate # unknown maternity colonies:
30 - 70 - 120
roughly 2/3
C
estimate average # individuals/colony:
100 - 150 - 200
preferable foraging grounds for the species: Wetlands, shores, broadleaved forest,….
D
estimate average area ([foraging] home
range) one colony? 8 – 12 – 16 km2
E
estimate available area qualified habitat
(roosts and foraging)?
500 - 1500 -2500
(A+B)*C

source population :
4500 – 16.000 – 36.000
Cross checks
B*C
E/D*C
Is B*C roughly similar to E/D*C ? 3000 – 10.500 – 24.000 6.250 – 18.750 – 31.250
If the numbers chosen for this example would be true numbers/sub-estimates, this comparison
indicates the need for more information and adjustment of the estimates.
Is (A+B)*D roughly similar to E ?
(A+B)*D
E
360 – 1.260 – 2.880
500 - 1500 -2500
If the numbers chosen for this example would be true numbers/sub-estimates, this comparison
indicates a reasonable similarity.

Estimating the migration populations of bats on SNS
13
Table 3: Example of the estimation of a target population. The numbers in the table are
just examples for the calculation.
Approach to quantification of target population for a specific region
A
estimate # investigated
clusters of mating roosts:
30 - 50 - 70 roughly ½ ?
B
estimate total # unknown
clusters of mating roosts/:
30 - 50 - 70
roughly ½ ?
C
estimate average # individuals/known
mating roosts/clusters of:
20 - 40 - 60
males
(A+B)*C = # males
1,200 – 4.000 – 8.400
-
D
Estimate # of visiting females
3.600 – 12.000 – 25.200
3 f/m?
Assumption: 1 female visits 2 male sites in
NL region (turnover):
1.800 – 6.000 - 12.600
preferable foraging grounds for the species: Wetlands, shores, broadleaved forest,….
E
estimate average area ([foraging] home
range) one mating roost?
1 - 2 - 4 km
2
F
estimate available area qualified habitat
(roosts and foraging)?
200 - 500 - 800 km
2
Cross checks
(A+B)*C
F/E*C
Is (A+B)*C roughly similar to F/E*C ?
1,200 – 4.000 – 8.400
10 – 1.000 – 12.000
If the numbers chosen for this example would be true numbers/sub-estimates, this comparison
indicates the need for more information and adjustment of the estimates.

Estimating the migration populations of bats on SNS
14
Via the questionnaire, the input of expert team and contributors, personal
comments of participants and literature, basic relevant information on ecological
and population dynamics parameters was brought together. The questionnaires
for the different regions assessed the (availability of) data and possible estimates
for the different parameters for this specific region, as well as knowledge gaps
and data deficiency.
For the basic regional data information was sought on the parameters as listed in
Table 4.
Table 4: Components for estimate used in questionnaire
Population (estimated) regional population size;
Migrating part of .. (estimated) size of migrating part of this population;
Migration direction (expert judgement on) direction(s) of the migration;
Direction/ part of .. plus part of population(s) using this/these direction(s)
Migration area (expert judgement on) area of migration zone(s);
Number(s) of roost
(estimated) number of known and unknown maternity
colonies/groups in the area; likewise for summer colonies/roosts,
(clusters of) mating roosts and hibernacula;
Number(s) of
individuals / roost
(estimated) number of individuals in maternity colonies/groups in
the area; likewise for summer colonies/roosts, (clusters of) mating
roosts and hibernacula;
Home range / roost
(estimated) size of the area used by one maternity colony;
likewise
for summer colonies/roosts, (clusters of) mating roosts and
hibernacula;
Migration landscape (expert judgment) landscape types used for migration;
Migration over sea (expert judgment) existence of migration over sea; along
coastline;
Front - funnelling (expert judgment) existence of funnelled migration and/or
migration over a broader front;
The information used consists of estimated values (basic estimate[s] + lower and
upper values) for specific source or target populations in the different (regions
of) countries around the southern North Sea, such as ‘the number of maternity
roosts in the source population in Lithuania’, ‘the % of migrating males or
females’, or ‘the numbers of (clusters of) mating roosts in the Netherlands, a
target area’.

Estimating the migration populations of bats on SNS
15
3.4 A] The flow model approach
The basic idea of the migration flow model is to assess data of the population
flowing into and out of a specific country or geographical area, and linking this to
the neighbouring countries or area (the size of the movement can be called flow
or flux). For this study, this is captured in a series of interrelated excel-tables
which are available with this document (Figure 2).
Figure 2: schematic image of the flow model: [the population present in the ‘summer
season’], plus [influx], minus [outflux], for males, females and offspring (only outflux),
equals the hibernating population.
For the flow model we work with the principle that the summer population, plus
immigration, minus emigration, for males, females and offspring, equals the
hibernating population (see Table 5).

Estimating the migration populations of bats on SNS
16
Table 5: components of the flow model A] for the migration population of Pipistrellus nathusii
relevant for the estimation of the population migrating on the southern North Sea.
summer
population + reproduction +
immigration - emigration =
hibernation
population
influx Outflux
resident females females # females % females # females
resident males males # males % males # males
+ floating population
= SM/M ratio* # juveniles % juveniles # juveniles
J/F ratio
# are % from
different
neighbouring
countries
% of # to
neighbouring
countries
Ratio f/m =
approx. 1 ratio migrating / non
migrating
ratio different
countries
factors per country
% flow to .. (over sea)
% flow to .. (over land)
bandwidth migration out of country - males
- females Juveniles follow J/F ratio**
= Estimate (lower, upper) per country Expert judgement + Input expert team contributors
= General parameter (lower, upper) , set for all countries
= Result calculation by model
** SM/M ratio = satellite males/male ratio
*J/F ratio = juvenile/female ratio
Working with this flow model A] needs a series of premises (for more detail see
annex IV), which of course will lead to a simplification of reality. Numbers in and
numbers out are estimates, where possible based on the basic regional data B]
from the questionnaires, and will come with an uncertainty interval. The
premises used are based on literature and the feedback from the expert team
and contributors. For more detail see annexes IV (premises) and V (summary
complex geographic interconnection between distribution, mating/reproduction
and migration of the Nathusius’ Pipistrelle).

Estimating the migration populations of bats on SNS
17
Premises4 regarding the migration flow model are:
a) Our focus is on the Nathusius’ Pipistrelle.
b) Our focus is on information related to the autumn migration period.
c) We work with a main migration direction from east/northeast to
west/southwest for the autumn migration. However a smaller part of the
population may have more west or more southwest directions. Directions
south, following western coast lines and river valleys, and even
south/southeast, along the east coast of Denmark are also (input Hans J.
Baagøe & Terp Fjederholt) observed. Drift cannot be excluded (Hüppop & Hill
2016; extended references in Annex IV).
Figure 3: Main migration direction, and other observed directions (see text).
d) We work from the concept of a broader migration front rather than narrow
corridors. At the same time the existence of funnelling along the coast and
along river valleys is a fact. This may also lead to a more north to south
migration direction on parts of their route (questionnaires and literature:
Ahlén 1997, Ahlén et al. 2007, 2009; Brinkmann et al. 2011; Lagerveld et al.
2014a).
4 The premises for the model are a first result of the design process with the expert team members: input from
expert team and literature. For all premise(s): premise = interpretation of the available knowledge in a way
that allows for calculation in the estimation.

Estimating the migration populations of bats on SNS
18
e) Larger areas of water will be crossed (see e.g. Ahlén 1997, Ahlén et al. 2007,
2009), where there is no other (logical) choice, such as 1) at the south points
of isles, 2) at sites where the coastline turns land inward (direction SE and
E), and 3) in situations where the coastline for longer distances deviates from
the general migration direction (e.g. longer stretches of north south
coastlines).
f) Based on the selection of relevant species and the developed approach,
source and target areas of bats migrating above and/or along the coast of
the southern North Sea are identified (see also: (www.grida.no6, Limpens &
Schulte 2000). We incorporated Norway, UK (Scotland, Northern-Ireland,
England and Wales), the Republic of Ireland, Finland, Sweden, Estonia,
Latvia, Lithuania, north western part of Poland, Denmark, the northern part
of Germany, the Netherlands and Belgium in the flow model.
Migrating bats from these countries might be arriving at the southern North
Sea. Populations migrating from/though Norway would end up in the
northern North Sea outside our study area. Norway is included, however,
because their outflux may be an influx to Scotland and Northern Ireland.
Migrating bats from more southern or south eastern countries in Europe
would end up more to the south of the current study area.
g) Finland is ‘purely source country’ in the sense that there is no influx of
migrating bats from neighbouring countries. Norway may or may not have an
influx from Sweden, but an outflux from Norway does not add to a flux over
the current study area. The western territory of Russia, bordering to the
Baltic States and Belarus, is only incorporated as a source area. Other
countries incorporated in the flow model will have an influx and outflux of
migrating bats.
6 http://www.grida.no/graphicslib/detail/nathusius-pipistrelle-distribution-and-migration_18cb

Estimating the migration populations of bats on SNS
19
Figure 4: Nathusius’ Pipistrelle distribution and migration (From collection: Living Planet:
Connected Planet, Rapid Response Assessment, Riccardo Pravettoni, UNEP/GRID-Arendal, 2011).
Ad Figure 4: Note: We use this illustration because it gives a quit good overview, but since 2011
new data is available. Please note that no maternity groups are ever discovered in Norway, nor
were any of the captured Nathusius’ Pipistrelles females (J. v.d. Kooij, pers. com.). A possible flux
from Sweden to Norway (J. v.d. Kooij, pers. com.) is not depicted in this illustration. Also all areas
without breeding, in the sense of occurrence of maternity sites, are qualified as hibernation area.
The fact is that many sites with (clusters of) mating roosts are found in both the maternity and
hibernation area.
h) We work on the premise that the majority of the animals from the northern
(Scandinavia) and north-eastern European maternity regions (Baltic States,
western Poland and western Russia) will hibernate in western Europe, in the
present case in middle-western Europe.
i) We work on the premise that the number of females born is equal to the
number of males, and that there is an equal survival for both sexes.
j) The ratio between territorial males and satellite males is unknown. It is
unclear whether there would be different ratios for different zones in the
migration direction. We work with a large uncertainty interval:
lower-central-upper: 0.2–2–5, and we also calculate with different settings:
3–4–5, 2–3–4, 1-2-3 and 0.2-1.2-2.2 satellite males per territorial male
(pers. com. Peter Lina). Working with a range of settings provides insight
into the effect of this component on the outcome.
k) We work on the premise that yearling males for the larger part will take part
in migration, as does the larger part of yearling and adult females. In
comparison adult territorial males will, for a larger part, stay behind (e.g.

Estimating the migration populations of bats on SNS
20
Boshamer & Bekker 2009, Lina pers. comm.). This is incorporated in the
bandwidth for migration of the separate groups per country in the model.
We work with the assumption that at least a part of the non-resident male
population, that is the population that does not occupy their own mating
roost/territory, also migrates.
l) We work on the premises (e.g. Lina pers. comm. Limpens & Schulte 2000,
Petersons 1990, 2004, Petersons & Lapina 1990) that:
- in all geographic regions, more females will take part in migration than
males;
- the more west/southwest in Europe we get, the less males take part in
migration, and thus the more males stay behind (establishing a territory or
only hibernating)
- in the autumn, in the northeast of Europe most females will migrate, while
more to the southwest a larger part will stay behind (hibernate).
This effect is accounted for by estimating the bandwidth of which part will
migrate out of the specific country (Table 6).
Table 6: Example table: Estimated bandwidth of the part of the population that will migrate, per
country. Values presented in this table are the values used in the first draft in the excel sheets for
the flow model. Note: Current values with input of design team and other contributors are in the
excel sheets for the flow model.
bandwidth migration out of
male low
male
male up
fem low
fem
fem up
Finland
0,250
0,750
1,000
0,300
0,900
1,000
Estonia
0,125
0,600
0,800
0,300
0,900
1,000
Sweden
0,250
0,750
1,000
0,300
0,900
1,000
Lithuania
0,125
0,500
0,750
0,300
0,800
0,950
Denmark
0,100
0,400
0,600
0,200
0,800
1,000
Poland / north west
0,100
0,400
0,600
0,200
0,800
1,000
Belarus
0,125
0,600
0,800
0,300
0,750
0,850
Latvia
0,125
0,600
0,800
0,300
0,700
0,800
Germany / north
0,050
0,250
0,500
0,250
0,750
1,000
Netherlands
0,050
0,250
0,500
0,250
0,750
1,000
Belgium
0,050
0,250
0,500
0,250
0,750
1,000
Norway
0,25
0,75 1
0,3
0,9 1
UK Scotland
0,25
0,75 1
0,3
0,9 1
UK Northern Ireland
0,1
0,5 0,9
0,3
0,9 1
UK England/Wales
0,25
0,75 1
0,3
0,9 1
Republic of Ireland
0
0
0
0
0
0

Estimating the migration populations of bats on SNS
21
m) There are very few data on the fecundity of bats. Available data show J/F
ratios of around 0.6 to 0.8 (e.g. O’Shea et al. 2011, Schaub et al. 2007). We
work on the premise that every adult female that is part of the western
European migrating population will on average have 0,7 offspring per year.
We calculated with the following variability: (min)-(central)-(max): 0.6-0.7-
0.8, 0.5-0.6–0.7 and 0.4–0.5–0.6 juveniles per female. Working with a
range of settings provides insight into the effect of this component on the
outcome.
Gathering of data and input for the A] migration flow model
Estimates and information for the flow model (figure 2) was first sought after
through input in the questionnaires / completed excel sheets for (B] available
regional data. In the next step, renewed input was gathered through distribution
of the excel file with the A] migration flow model, comprising the interconnected
excel tabs per country. Here the different members of the design team were
asked to comment on the chosen preliminary values for the components (see
excel sheet flow model).

Estimating the migration populations of bats on SNS
22
Figure 5: example illustrating the tables used to collect data for the flow model.

Estimating the migration populations of bats on SNS
23
4. Results: Prototype estimator of populations of
migrating bats
4.1 Feedback basic regional information and data:
All respondents currently involved in the project agreed on the approach for
regional/national estimates, but not everybody had the data or information to
provide estimates. Input was given for such data as was available (see Annex III
summary feedback basic regional information and data).
Relatively elaborate data and estimates, reported via the questionnaire, were
received from the UK, differentiated for Northern Ireland, Scotland, Wales and
England, and for the Republic of Ireland and the Netherlands. This is 6 out of 16
units (UK is divided in the 4 countries).
Limited data, and data for smaller regions/sites within countries were received
for Germany, Norway, Sweden, Denmark, Finland, Latvia and Belgium. These
respondents were not able to present estimates (7/16). Data for Poland, Estonia
and Lithuania are currently not available (3/16).
For the Netherlands also data and input for smaller regions/sites within the
country were received. For France and Portugal, which are both not within the
current range of the model, no estimates could be given. For France data and
input for smaller regions/sites within north-western French Départements were
shared.
4.2 Feedback information and data on flow model
The general opinion of the members of the expert team and the other regional
contributors is a consent to the approach of the flow model (see Annex VI:
summary feedback on flow model).
More specifically, the respondents for Norway, the UK, the RO Ireland, Latvia,
Denmark / northern Germany, Belgium and the Netherlands consented to the
(prototype) approach. This is 13 out of 16 units currently in the flow model.
Respondents for Finland and Sweden refrained from comment (2/16).
Respondents for France and Portugal, not within the current range of the model,
respectively refrained from comment and consented. Currently there is no input
for Estonia, Belarus, Lithuania and north-western Poland (4/16).
Feedback in the sense of information or comments on input and output of the
model was received from Latvia, Denmark, Norway, northern Germany, the
Netherlands, Belgium, the UK and the RO Ireland (11/16 ). More quantitative
input, in the sense of suggestions for alterations to the used estimates in the

Estimating the migration populations of bats on SNS
24
flow model were received for northern Germany, the Netherlands, Belgium, the
UK and the RO Ireland (8/16).
Although there is a consent to the approach, many of the contributors hesitate to
give estimated values for those factors in the excel sheet (see e.g. table 7)
where their regional expert input is needed (see Annex IV). This is due to the
lack of basic regional data and differences among countries regarding the
availability of systematic surveys of occurrences and distribution of bats. Also a
lack of experience with expert judging of possible occurrences of bats in
landscapes will have played a role. The values used in the flow model are
predominantly expert judgement estimates carried out by Herman Limpens.
These were based on such input as was available, and adapted where
contributors were able to react for their region on the original value in the model
(also see accompanying excel sheet, annexes III and VI).

Estimating the migration populations of bats on SNS
25
Table 7: overview estimated summer population per country = source population(s) for migration.
M= males, MS= Male satellites, F= Females, J= Juveniles
central
lower
upper
Male + sat
fem
juv
male
fem
Juv
male
fem
Juv
Finland
1.500
5.000
3.000
150
50
20
30.000
50.000
40.000
Estonia
900
1.800
1.800
36
300
120
18.000
30.000
24.000
Belarus
3.000
10.000
6.000
120
1.000
400
6.000
100.000
80.000
Latvia
1.500
5.000
3.000
60
500
200
30.000
50.000
40.000
Lithuania
1.500
3.000
1.800
60
300
120
30.000
30.000
24.000
Poland north-western
4.500
10.000
6.000
120
1.000
400
60.000
100.000
80.000
Sweden
300
1.000
600
12
10
4
60.000
10.000
8.000
Denmark
1.500
2.500
1.500
60
250
100
30.000
25.000
20.000
Germany northern
48.000
5.000
3.000
4.800
500
200
384.000
50.000
40.000
Netherlands
55.350
100
60
3.816
0
0
340.200
1.000
800
Belgium
30.000
50
30
1.200
25
10
24.000
100
80
sub total
148.050
43.450
26.790
10.434
3.935
1.574
1.012.200
446.100
356.880
Norway
252
5.000
3.000
25
50
20
2.505
5.000
4.000
UK Scot
150
0
0
6
0
0
3.000
0
0
UK Northern Ireland
3.000
8.000
4.800
120
2.000
800
60.000
14.000
11.200
UK England/Wales
6.000
5.000
3.000
600
2.000
800
30.000
10.000
8.000
RO Ireland
3.200
10.000
6.000
700
2.000
800
9.000
18.000
14.400
sub total
12.602
28.000
16.800
1.451
6.050
2.420
104.505
47.000
37.600
Total
160.652
71.450
43.590
11.885
9.985
3.994
1.116.705
493.100
394.480
M+MS+F+J central
M+MS+F+J lower
M+MS+F+J upper
sub Relevant for SNS
218.290
15.943
1.815.180
sub Northwest/West SNS
57.402
9.921
189.105
Total
275.692
25.864
2.004.285
Calculated with the following settings and bandwidth
Estimate
lower
upper
Number of juv per female per year
0,60
0,40
0,80
Number of satellite males (adults!) per territorial male
2,00
0,20
5,00
Number of Males per Female (not yet used as variable)
1,00
0,90
1,10

Estimating the migration populations of bats on SNS
26
4.3 Results A] flow model based on regional data and
estimates B]
The premise regarding migration directions leads to‘ countries where an
out/influx can be expected’ (i.e. from Poland to Germany), and ‘countries where
out/influx is very unlikely to occur’ (i.e. from Sweden to Poland; see Figure 6).
Combined with available basic regional data and estimates B] for these countries,
and the feedback cycle with the expert team (updated until 20 01 2017), this
resulted in the current structure of the flow model (Figure 6 and Table 5). This
flow chart is a representation of the series of interconnected excel sheets per
country, which are, in the concrete sense, the flow model A] (see excel sheets
flow model).
Figure 6: schematic image of the flow model of the migration of Nathusius’ Pipistrelle from
east/northeast to west/south-west, for that part of the population that may be expected to partially
also migrate over the southern North Sea (between European and UK).
For the sake of simplicity, in this schematic approach Kaliningrad is processed in the transfer from
Lithuania to Poland. General autumn migration direction is predominantly east/northeast to
west/southwest, with a smaller part of the population moving in more west or more southwest
directions.
Note: a flux from Sweden to Norway is not incorporated. We were not able to process this input
from Norway at this stage. The outflux from Norway, however, will be north of our current study
area SNS.
We ‘confronted’ our knowledge and expert judgement (or best informed guess)
with its effect on the flow and population sizes. This interaction resulted in the
currently used estimates in the flow model (status January 2017), where settings
were chosen that are for now more or less acceptable for all participants.

Estimating the migration populations of bats on SNS
27
A range of different settings for juveniles/female and satellite males/male is used
to assess the reaction of the model on differences in the setting (Annex VII). The
result, through using the flow model with the minimum, maximum and most
central settings out of the tested range of settings, is presented in table 87, and
Figures 7 (linear) and 8 (logarithmic). This output represents the current
preliminary estimate for the number of Nathusius’ Pipistrelle potentially migrating
across the southern North Sea. We give the output in the numerical table as well
as the linear and logarithmic histogram, to provide different images to assess the
preliminary outcome.
Table 8: the current preliminary estimate for the number of Nathusius’ Pipistrelle potentially
migrating across the southern North Sea.
lower
central
upper
male
80
9.428
163.504
female
43
19.226
521.399
Juv
17
11.536
417.119
total
140
40.190
1.102.022
settings
juvenile/female: 0,4 - 0,6 - 0,8
satellite males/male: 0,2 - 2 - 5
(current lowest and highest settings tested)
When we look at the outcome using the range of settings for J/F against the
range for SM/M8 (Table 1, Annex VII) for a series of runs, the highest central
estimate produced for the total number of bats is around 50.000 individuals.
7 Note: The output presented in tables 8 reflects the status of January 2017 and the accuracy of the current
input.
8 Satellite males / (territorial) males

Estimating the migration populations of bats on SNS
28
Figure 7: the current preliminary estimate for the number of Nathusius’ Pipistrelle potentially
migrating across the southern North Sea (linear scale).
Figure 8: the current preliminary estimate for the number of Nathusius’ Pipistrelle potentially
migrating across the southern North Sea in a logarithmic scale.
-50000 50000 150000 250000 350000 450000 550000
male low
male
male up
fem low
female
fem up
Juv low
Juv
Juv up
juv/female 0,4 - 0,6 - 0,8
sat 0,2 - 2 - 5
110 100 1000 10000 100000
male low
male
male up
fem low
female
fem up
Juv low
Juv
Juv up
juv/female 0,4 - 0,6 - 0,8
sat 0,2 - 2 - 5

Estimating the migration populations of bats on SNS
29
5. Conclusions and discussion
5.1 B] regional data and estimates
Very few colleagues have been able to produce concrete data for larger areas.
Data on occurrence and distribution resulting from systematic survey projects
(Broekhuizen et al. 2016, Limpens et al. 1997, Limpens & Roschen 1996, 2002)
are not widely available. Very few countries have data resulting from systematic
surveys of the whole of their territory. There are also few data on occurrence and
distribution on a smaller scale, allowing for expert judgement or estimating how
many roosts (maternity, summer, mating, hibernation) might be present, and
what kind of numbers of bats actually use them.
It is necessary to stimulate research regarding the data needed to answer the
questions formulated in the ‘Questionnaire to collect (available) basic regional
information and data’ (see annex II). Even when this is done for some of the
elements/factors (such as J/F, satellite males/male), or only specific landscapes
and/or countries, such information can be incorporated in the flow model and
enhance the accuracy of the output. We recommend prioritising of research into
factors that can be used for ground truthing of the model.
At this stage (status January 2017) we have a preliminary output and we have to
work with large uncertainty intervals for the different elements comprising the
flow model. Regarding the basic data, there are differences between countries
leading towards different bandwidths of the uncertainty intervals for the
countries.
The outcome or the estimation of the migration population on the southern North
Sea is a result of testing the model and its basic regional input and data. More
accurate input of such data is needed to produce a more accurate estimate with
a lower bandwidth.
At this stage, using the combination of the basic regional data with the
developed flow model was an important first step in ground truthing of the
outcome of the flow model. We need to work on the availability and accuracy of
the input data.

Estimating the migration populations of bats on SNS
30
5.2 A] The flow model approach
The constructed model A] was consented by the expert team members (and
contributors) and considered to be a good first prototype estimator.
Uncertainty regarding the output (A] based on B], status January 2017) was
rather focused on the non-availability of basic regional data or estimates for
most countries/regions, regarding:
- the population size of males, females and juveniles,
- the %’s of migrating individuals (Males, Females, Juveniles) as a whole,
and/or
- the different proportions of migration towards different connected countries
in the west/southwest direction, and
- the basic population dynamical factors such as J/F and satellite males per
male.
This uncertainty is also reflected in the fact that many contributors preferred to
approach the estimates, for the basic input in the model, in a conservative way.
In practice this means estimating the regional female and male population sizes
‘not too high’, because underestimating ‘feels like’ a lesser mistake than
overestimating. In itself this ‘feeling’ may well be untrue, since bats generally
prove to be more abundant than intuitively assumed. Therefore, at the same
time others feel that the total migrating population is probably underestimated.
These signals are conflicting and are an important qualitative cue towards the
process and state of the estimator (status January 2017) we need to
acknowledge: the prototype estimator needs accurate input and needs ground
truthing.
General factors in the flow model are now J/F (3 settings), satellite
males/territorial male (4 settings), male/female ratio in offspring (1 setting). The
model will improve when these factors can be made adaptable per
country/region and when ‘settings/values’, can easily be adapted.
Mortality is a factor that may be included in an enhanced version of the flow
model, preferably differentiated for M, F and J per country or region.
Bandwidth
For ‘the part of the population that migrates’, we already work with a bandwidth
for the different groups, males, females and juveniles. However, the percentages
defining migration from one country to another (via land or sea) are not yet
implemented with a bandwidth, and not yet differentiated for males, females and

Estimating the migration populations of bats on SNS
31
juveniles per country. The percentage for migration over sea is estimated per
country depending on actual direction of the coastline in relation to the premises
of a predominantly E/NE to W/SW autumn migration direction. The way these
factors impact the migration flux in the model, can and need to be improved
through implementing these bandwidths.
The flow model A] (status January 2017) is constructed in the form of a series of
interrelated sheets in an excel table, and processed with concrete values. The
model uses different settings, for ecological and population dynamical
parameters, almost all a with specific confidence bandwidths. We recommend to
assess the pros and cons of converting the model to a computer language (e.g.
R?).
For Germany and Poland it is needed to investigate how to – quantitatively -
differentiate between the ‘general’ west/southwest directed migration, and the
part of the population that follows a more southerly directed path along the river
valleys. The same is necessary for the more southeast directed pathway along
the eastern coastline of Denmark and the German Federal country Schleswig-
Holstein. And indeed more insight into a flux from Sweden to Norway is needed.
The output of the flow model A], the current resulting estimate, is a best practice
estimate and naturally goes with rather high uncertainty intervals.
We stress that bandwidth value(s) should not be misinterpreted in the sense that
they indicate the possibility of e.g. high numbers of individuals, where added
mortality automatically results in a low impact, or where high numbers
automatically result in high risk. Assessment of fatality risk in different studies
concerning wind turbines on land (e.g. Brinkmann et al. 2011) also reveal
relatively large bandwidths. In order to be able to adequately interpret the added
mortality as caused by offshore wind farms, insight is urgently needed on land
and at sea into the components that would allow assessment of impact on: a)
the numbers of fatalities, b) the relation between abundance and risk/numbers of
fatalities, as well as c) the size of the population.
We need more detail in data and ecological and population dynamical
parameters. This would provide more and more accurate basic regional data, as
well as more accurate estimates of general parameters. This would improve the
(reliability of the) outcome of the flow model.
5.3 How to interpret or use the outcome
The bandwidth of the preliminary estimate (see paragraph 3.3 and annex VII) is
between 100 and a 1.000.000 Nathusius’ Pipistrelles migrating on the southern

Estimating the migration populations of bats on SNS
32
North Sea, with a central estimate of around 40.000 individuals. This is the total
for males, females plus juveniles and was calculated with the minimum value of
the range of settings used for the lower limit and the maximum value of the
range of settings for the upper.
The members of the design team as well as the other contributors agree – for
now – to such values. Nevertheless we feel that more ground truthing and
assessment of the population factors is needed to confirm these figures.
Comparing the estimate of the total (males+females+juveniles) population
migrating on the SNS to the estimate of the total for the summer population of
the potential source, is another way to try and get a sense of the functioning of
the prototype estimator (Tables 8 and 9). The estimate of the central total
number of bats for the potential source populations is about 5 to 6 times higher
than the central total number of bats for the population migrating on the SNS.
This illustrates that accuracy is still likely to be low, because it is difficult to
imagine, but in itself not impossible, that between 17 and 20% of the source
population would migrate on the SNS. In the current state, it is unclear whether
this is predominantly a result of inaccurate estimates for the source population,
the general parameters regarding J/F and SM/M, or the chosen values for the
fluxes to different countries.
The estimate of the central total number of bats for the potential source
populations in Latvia, Lithuania, Poland and Sweden resulting from the flow
model, is about 9 to 10 times higher, than the estimate for these source
populations given in Leopold et al. (2014) as a total for the population
approaching the SNS while on migration. Both figures are estimates. In the
current study, however, a more comprehensive approach is used.
The now developed estimator and especially the current outcome is far from
perfect. However, this prototype gives direction and insight into the work and
data needed to achieve better estimates.

Estimating the migration populations of bats on SNS
33
6. Knowledge gaps and priority research questions
Organize an international programme focused on ground truthing of the factors
relevant for the flow model and estimates. Many of the general parameters and
country specific parameters are largely unknown.
Important actions would be: stimulate active and systematic survey of
occurrence and distribution and abundance of the relevant species, their different
roost types and the – average - numbers in such roost types, and preferably in
relation to the landscape. For an estimate for the SNS this is most urgently
needed in the range of the countries used in this study.
In relation to the urgency regarding improving population estimates of migrating
bats, the surveying should preferably be done in the form of a statistical effective
sample of e.g. grid cells (per country/landscape) in order to allow for quantitative
extrapolation, and thus deliverance of the basic estimates for the different
countries.
Investigate, per country/region, the basic ecological and population dynamical
factors (such as J/F, satellite males/territorial male, mortality, % migrating, %
migrating in different directions/to different countries, abundances per
landscape, all for M, F and J) and the variation in such factors.
Develop methods – e.g. acoustical methods - to compare abundance of bat
species on land and on see.
Organize telemetry of relevant species during migration, to verify the current
estimates – deduced from available information - regarding migration directions,
and quantitative differentiation between such directions. Work together with the
different states in the region around the southern North Sea.
Examine the situation (weather, landscape, coastline, ..) in which individuals
may stop following a coastline and start crossing open sea.
Investigate differences in (quantitative) migration parameters for the autumn
and spring migration (abundance/acoustic activity; spread of
individuals/abundancy in space and time [M, F and J], occurrence of funnelling,
..).

Estimating the migration populations of bats on SNS
34
7. References
7.1 References reporting and basic data
Note: The items in the literature list for reporting and basic data are referred in the
report text and the annexes.
Ahlén I., 1997. Migratory behaviour of bats at south Swedish coasts. Zeitschrift für
Säugetierkunde 62: 375-380.
Ahlén, I., H. J. Baagøe & L. Bach, 2009. Behavior of Scandinavian bats during migration
and foraging at sea. - Journal of Mammalogy, 90(6):1318–1323.
Ahlén I., L. Bach, H.J. Baagøe & Pettersson J. 2007. Bats and offshore wind turbines
studied in Southern Scandinavia. Swedish EPA, Report 5571, Stockholm.
Boshamer, J.P.C. & J.P. Bekker, 2008. Nathusius’ pipistrelles (Pipistrellus nathusii) and
other species of bats on offshore platforms in the Dutch sector of the North Sea. Lutra
51: 17-36.
Brinkmann, R., O. Behr, I. Niermann & M. Reich (eds.), 2011. Entwicklung von Methoden
zur Untersuchung und Reduktion des Kollisions-risikos von Fledermäusen an Onshore-
Windenergieanlagen. - Umwelt und Raum Bd. 4, 40-115, Cuvillier Verlag, Göttingen.
Broekhuizen, S., K. Spoelstra, J.B.M. Thissen, K.J. Canters & J.C. Buys, 2016. Atlas van
de Nederlandse zoogdieren - deel 12 serie Nederlandse fauna. 300 pp. ISBN:
9789050115346Cryan, P.M. & A.C. Brown, 2007. Migration of bats past a remote
island offers clues toward the problem of bat fatalities at wind turbines, Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.05.019
Dürr, T. & L. Bach, 2004. Fledermäuse als Schlagopfer von Windenergie-Anlagen - Stand
der Erfahrungen mit Einblick in die bundesweite Fundkartei. Bremer Beiträge für
Naturkunde und Naturschutz Band 7. Bremen, BUND Landesverband Bremen e.V.:
253-263.
Dürr, T., 2010. Auszug aus der bundesdeutschen Schlagopferkartei, Stand 05.03.2010.
Schriftl. Mitt.
Furmankiewicz, J. & M. Kucharska, 2009. Migration of bats along a large river valley in
southwestern Poland. Journal of Mammalogy 90: 1310-1317.
Hobbs, M., 2014. North Sea Ferry Bat Migration Research Report. BSG Ecology.
Hutterer, R., T. Ivanova, C. Meyer-Cords & L. Rodrigues, 2005. Bat migrations in Europe:
a review of banding data and literature. Naturschutz und Biologische Vielfalt 28: 1–
176.
Hüppop, O. & R. Hill, 2016. Migration phenology and behaviour at a research platform in
the south-eastern North Sea. Lutra 59 (1/2):5-22.
Jarzembowski, T., 2003. Migration of the Nathusius’ pipistrelle Pipistrellus
nathusii (Vespertilionidae) along the Vistula Split. Acta Theriol 48: 301.
doi:10.1007/BF03194170
Jonge Poerink, B., S. Lagerveld & H. Verdaat, 2013. Pilot study bat activity in the Dutch
offshore wind farms OWEZ and PAWP (2012). IMARES report C026/13.
Kurta, A., 2010. Reproductive timing, distribution, and sex ratios of tree bats in Lower
Michigan. - Journal of Mammalogy, 91(3):586–592.
Lagerveld, S., B. Jonge Poerink, R. Haselager & H. Verdaat, 2014a. Bats in Dutch
offshore wind farms in autumn 2012. - Lutra 57(2): 61-69
Lagerveld, S., B. Jonge Poerink & H. Verdaat, 2014b. Monitoring bat activity in offshore
wind farms OWEZ and PAWP in 2013. IMARES Report C165/14.
Lagerveld, S., B. Jonge Poerink, P. de Vries & M. Scholl, 2015a. Bat activity at offshore
wind farms LUD and PAWP in 2015. IMARES Report C001/2016.
Lagerveld, S. B. Jonge Poerink & P. de Vries, 2015b. Monitoring Bat activity at the Dutch
EEZ in 2014. IMARES Report C094/15

Estimating the migration populations of bats on SNS
35
Landesamt für Umwelt Land Brandenburg, 2016. Zentrale Fundkartei über Anflugopfer an
Windenergieanlagen (WEA).
Lehnert, L.S., S. Kramer-Schadt, S. Schönborn, O. Lindecke, I. Niermann & C.C. Voigt,
2014. Wind farm facilities in Germany kill noctule bats from near and far. PLoS ONE
9:e103106
Leopold M.F., M. Boonman, M.P. Collier, N. Davaasuren, R.C. Fijn, A. Gyimesi, J. de Jong
R.H. Jongbloed, B. Jonge Poerink, J.C. Kleyheeg-Hartman, K.L. Krijgsveld, S.
Lagerveld, R. Lensink, M.J.M. Poot, J.T. van der Wal & M. Scholl 2014. A first approach
to deal with cumulative effects on birds and bats of offshore wind farms and other
human activities in the Southern North Sea. IMARES Report C166/14
Limpens, H.J.G.A., 2001. Beschermingsplan Vleermuizen van Moerassen. Rapport
2001.05 Vereniging voor Zoogdierkunde en Zoogdierbescherming, Arnhem, in
opdracht van ExpertiseCentrum LNV Onderdeel Natuurbeheer. 84 pp.
Limpens, H.J.G.A., K. Mostert & W. Bongers, 1997. Atlas van de Nederlandse
vleermuizen; onderzoek naar verspreiding en ecologie. - KNNV Uitgeverij, 260 pp.
Limpens, H.J.G.A. & R. Schulte, 2000. Biologie und Schutz gefährdeter wandernder
mitteleuropäischer Fledermausarten am Beispiel von Rauhhautfledermäusen
(Pipistrellus nathusii) und Teichfledermäusen (Myotis dasycneme). - Nyctalus (N.F.)
7(3):317-327.
Limpens, H.J.G.A. & A. Roschen, 1996. Bausteine einer systematischen
Fledermauserfassung, Teil 1: Grundlagen. - Nyctalus (N.F.) 6, Heft 1, S. 52-60.
Limpens, H.J.G.A. & A. Roschen, 2002. Bausteine einer systematischen
Fledermauserfassung. Teil 2 - Effektivität, Selektivität, und Effizienz von
Erfassungsmethoden. Nyctalus (N.F.) 8/2:155-178.
Lina, P.H.C., 1990. Long-distance recoveries of Nathusius' pipistrelles Pipistrellus nathusii
found or banded in The Netherlands. Lutra 33: 45±48.
Lundberg, K., 1989. Social organisation and survival of the Pipistrelle bat (Pipistrellus
pipistrellus), and a comparison of advertisement behaviour in three polygynous bat
species. Dissertation, Department of animal ecology, University of Lund, Sweden.
Masing, M.V., 1988. Long-distance flight of Pipistrellus nathusii banded or recaptured in
Estonia. Myotis 26: 159–164.
Masing, M.V., 2011. How many bats migrate along Estonian coasts during late summer?
p 37 in XIIEuropean Bat Research Symposium, Vilnius, Lithuani a, 22–26 August 2011
Lithuanian Society for Bat Conservation, Vilnius, Lithuania, 100 pp.
Mitchell-Jones, A.J., G. Amori, W. Bogdanowicz, B. Kryštufek, P.J.H. Reijnders, F.
Spitzenberger, M. Stubbe, J.B.M. Thissen, V. Vohralík, and J. Zima (eds), 1999. The
Atlas of European Mammals. Academic Press, London.
Niederfriniger, O., G. Rallo, C. Violani & B. Zava, 1991. Ringed Nathusius’ bats,
Pipistrellus nathusii, recovered in N Italy (Mammalia, Chiroptera). Atti della Societa
Italiana di Scienze naturali e del Museo Civico di Storia naturale di Milano 131: 281–
284.
Oldenburg, W., & H. Hackethal, 1989. Zur Migration von Pipistrellus nathusii (Keyserling
u. Blasius). Nyctalus (N. F.) 3: 13–16.
O’Shea, T.J., L.E. Ellison & T. R. Stanley, 2011. Adult survival and population growth rate
in Colorado big brown bats (Eptesicus fuscus). - Journal of Mammalogy, 92(2):433–
443.
Petersen, A., J-K Jensen, P. Jenkins, D. Bloch & F. Ingimarsson, 2014: A review of the
occurrence of bats (Chiroptera) on islands in the North East Atlantic and on North Sea
installations’ Acta Chiropterologica, 16, 169–195, 2014. doi:
10.3161/150811014X683381: 169-195.
Peterson, T.S., S.K. Pelletier, S.A. Boyden & K.S. Watrous, 2014. Offshore acoustic
monitoring of bats in the Gulf of Maine. Northeastern Naturalist 21(1):86 – 107.
Petersons, G., 1990. Die Rauhhautfledermaus Pipistrellus nathusii (Keyserling u. Blasius,
1839) im Lettland: Vorkomen, Phänologie und Migration. Nyctalus (N. F.) 3: 81–98.

Estimating the migration populations of bats on SNS
36
Petersons, G., 2004. Seasonal migrations of north-eastern populations of Nathusius’
pipistrelle Pipistrellus nathusii (Chiroptera) Myotis 41/42: 29-56.
Petersons, G. & A. Lapina, 1990. [The results of migration studies of Pipistrellus
nathusii in Latvia]. [In: The bats. Proceedings of the fifth bat research conference in
the USSR, Moscov, 1990. P. P. Strelkov and V. A. Rodionov, eds]. Pensa: 73–76. [In
Russian]
Roer, H., 1995. 60 years of bat-banding in Europe. Results and task for future research.
Myotis 32–33: 251–261.
Rydell, J., L. Bach, M.-J. Dubourg-Savage, M. Green, L. Rodrigues & A. Hedenström,
2010. Bat mortality at wind turbines in northwestern Europe. - Acta Chiropterologica,
12(2):261–274, PL ISSN 1508-1109 © Museum and Institute of Zoology PAS doi:
10.3161/150811010X537846.
Rydell, J., L. Bach, P. Bach, L.G. Diaz, J. Furmankiewicz, N. Hagner-Wahlsten, E.-M.
Kyheröinen, T. Lilley, M.V. Masing, M.M. Meyer, G. Petersons, J. Šuba, V. Vasko, V.
Vintulis & A. Hedenström, 2014. Phenology of migratory bat activity across the Baltic
Sea and the South-Eastern North Sea. Acta Chiropterologica, 16 (1): 139-147.
Schaub, M., O. Gimenez, A. Sierro & R. Arlettaz, 2007. Use of Integrated Modeling to
Enhance Estimates of Population Dynamics Obtained from Limited Data - Conservation
Biology Volume 21, No. 4, 945–955.
Šuba, J., G. Petersons & J. Rydell, 2012. Fly-and-forage strategy in the bat Pipistrellus
nathusii during autumn migration. Acta Chiropterologica 14: 379–385.
Voigt, C.C., A.G. Popa-Lisseanu, I. Niermann, S. Kramer-Schadt, 2012. The catchment
area of wind farms for European bats: A plea for international regulations. Biological
Conservation 153 (2012) 80–86.
Voigt, C.C. & L.S. Lehnert, G. Petersons, F. Adorf & L. Bach, 2015. Wildlife and renewable
energy: German politics cross migratory bats. - Eur J Wildl Res.
Voigt, C.C., O. Lindecke, S. Schönborn, S. Kramer-Schadt & D. Lehmann, 2016. Habitat
use of migratory bats killed during autumn at wind turbines. Ecological Applications,
26(3):pp. 771–783.
Walter, G., H. Matthes & M. Joost, 2007. Fledermauszug über Nord- und Ostsee —
Ergebnisse aus Offshore-Untersuchungen und deren Einordnung i das bisher bekannte
Bild zum Zuggesschehen. Nyctalus (N.F.), 12.
7.2 Literature basic regional data
Note: The items in the literature list for regional basic data are received via the
questionnaires and are input for the estimates in the excel sheets per country. These are
not specifically referred to in the report text.
Ahlén, I., 2011a. Fladdermusfaunan i Sverige. Arternas utbredning och status.
Kunskapsläget 2011. Fauna och flora 106: 2-19.
Ahlén, I., 2011b. Pipistrellus nathusii, trollpipistrell. Artdatabanken, SLU, Uppsala.
http://artfakta.artdatabanken.se/taxon/100111
Baagøe, H. J., 2001. Danish bats (Mammalia: Chiroptera): atlas and analysis of
distribution, occurrence, and abundance. Steenstrupia 26:1–117.
Baagøe, H.J., 2007: Troldflagermus Pipistrellus nathusii (Keyserling & Blasius, 1839). In
Baagøe. H.J. & T.S. Jensen, 2007: Dansk Pattedyratlas. Gyldendal. s. 66-69
Baagøe, H.J., 2011: Bornholms flagermus- status 2010. Natur på Bornholm Nr. 9, 2011,
s 22-30. BugBook Publishing.
Baagøe, H., & D. Bloch, 1994: Bats (Chiroptera) in the Faroe Islands. Frödskaparrit, pp.
83-88. Thorshavn.

Estimating the migration populations of bats on SNS
37
Baagøe, H.J. & E.T. Fjederholt, 2014: Dværgflagermus (Pipistrellus pygmaeus)- første
sikre fund fra Bornholm – og lidt om de to andre Pipistrellus-arter. Natur på Bornholm.
Nr. 12, 2014, s 8-12. BugBook Publishing. [English summary: Soprano bat
(Pipistrellus pygmaeus) – first positive records from Bornholm.]
Battersby, J. (Ed) & Tracking Mammals Partnership, 2005. UK Mammals Species Status
and Population Trends. First Report by the Tracking Mammals Partnership.
JNCC/Tracking Mammals Partnership, Peterborough.
Borkenhagen, P., 1993. Atlas der Säugetiere Schleswig-Holsteins. Landesamt für
Naturschutz. 129 pp.
Broekhuizen, S., K. Spoelstra, J.B.M. Thissen, K.J. Canters & J.C. Buys (eds.), 2016.
Atlas van de Nederlandse zoogdieren - herkenning, verspreiding & leefwijze. Deel 11
serie Nederlandse fauna. 300 pp. ISBN: 9789050115346
Elmeros, M, J.D. Møller & H.J. Baagøe, 2015: Part A Bat studies 2013 in: Therkildsen
O.R. & M. Elmeros 2015: First year post- construction monitoring of bats and birds at
wind turbine test center Østerild.
FEBI, 2013. Fehmarnbelt Fixed Link EIA. Fauna and Flora – Bats of the Fehmarnbelt Area
– Baseline Volume I Report No. E3TR0016 Report: 65 pages
FEBI, 2013. Fehmarnbelt Fixed Link EIA. Fauna and Flora – Impact Assessment - Bats of
the Fehmarnbelt Area. Report No. E3TR0017. Report: 66 pages
Gerell, R., 1987. Flyttar svenska fladdermöss? Fauna och Flora 82:79–83.
Isaksen, K., P.O. Syvertsen, J. van der Kooij & H. Rinden, (red.), 1998. Truete pattedyr i
Norge: faktaark og forslag til rødliste. – Norsk Zoologisk Forening. Rapport 5. (182 s.)
Isaksen, K. og Olsen, K.M. 2007. Kartleggingsarbeid utført av Norsk Zoologisk Forenings
flaggermusgruppe. – Fauna 60 (3–4): 176–182.
Isaksen, K., K.M. Olsen & P.O. Syvertsen, 1993. Kartlegging av pattedyrenes utbredelse i
Norge – en prosjektorientering. – Fauna 46 (1): 3–9.
Isaksen, K. (Red.), M. Klann, J. Van Der Kooij, , T.C. Michaelsen, , K.M. Olsen, , T.
Starholm, C.F. Sunding, , M.F. Sunding & P.O. Syvertsen, 2009. Flaggermus I Norge.
Kunnskapsstatus Og Forslag Til Nasjonal Handlingsplan. Norsk Zoologisk Forening.
Rapport 13. 124 S. (Isbn 978-82-7857-014-2).
Kapteyn, K., 1995. Vleermuizen in het landschap; over hun ecologie, gedrag en
verspreiding. Schuyt & Co., Haarlem.
Limpens, H.J.G.A., K. Mostert & W. Bongers, 1997. Atlas van de Nederlandse
vleermuizen; onderzoek naar verspreiding en ecologie. - KNNV Uitgeverij, 260 pp.
National Report on the Implementation of the Agreement on the Conservation of Bats in
Europe: United Kingdom (1999).
National report on the implementation of the Agreement on the Conservation of
Populations of European Bats: United Kingdom 2010-2013.
National report on the implementation of the Agreement on the Conservation of
Populations of European Bats: 2010-2013: the Netherlands.
Roche, N., T. Aughney, N. Kingston, D. Lynn & F. Marnell, 2015. Records for Nathusius’
pipistrelle (Pipistrellus nathusii) in Ireland from a car-based bat monitoring scheme.
Irish Naturalists’ Journal 34(2): 83-88.
Russ, J.M., 1999. The Microchiroptera of Northern Ireland: community composition,
habitat associations and ultrasound. PhD thesis. The Queen’s University of Belfast.
Russ, J.M., 2014. Nathusius’ pipistrelle in Great Britain & Ireland. Available at:
http://www.nathusius.org.uk (accessed 10th September 2016).
Russ, J.M.. 2012. British Bat Calls: A Guide to Species Identification. Pelagic Publishing,
Exeter, UK.
Russ, J.M., A.M. Hutson, W.I. Montgomery, P.A. Racey & J. R. Speakman, 2001, The
status of Nathusius' pipistrelle (Pipistrellus nathusii Keyserling & Blasius, 1839) in the
British Isles. J. Zool., Lond. 254:91-100
Syvertsen, P.O., P. Shimmings & K. Isaksen, 1996. The Norwegian Mammal Fauna:
Status And Atlas Mapping. Hystrix, (Ns.) 8 (1-2): 91-95

Estimating the migration populations of bats on SNS
38
Skov, H., M. Desholm, S. Heinänen & T.W. Johansen, 2015: Kriegers Flak Offshore Wind
Farm Environmental Impact Assessment Technical background report Birds and bats.
A report prepared for Energinet.dk as part of the EIA for Kriegers Flak Offshore Wind
Farm. The report is prepared by Danish Center for Environment and Energy (DCE) at
Aarhus University and DHI in collaboration with NIRAS. January 2015
Verkem, S., J. De Maeseneer, B. Vandendriessche, G. Verbeylen, & S. Yskout,
(Ed.) 2003. Zoogdieren in Vlaanderen: ecologie en verspreiding van 1987 tot 2002.
Natuurpunt Studie, JNM Zoogdierenwerkgroep: Mechelen. ISBN 90-77507-01-9. 451
pp.
7.3 Websites regional data
http://www.batmap.de/web/start/karte;jsessionid=4E67E0846C15AF90FC4E6A0031AED
D8C
https://www.verspreidingsatlas.nl/zoogdieren
http://www.nathusius.org.uk/Distribution.htm
http://www.bats.org.uk/pages/nathusius_pipistrelle.html#Distribution

Estimating the migration populations of bats on SNS
39
8. Annexes
I) Annex 1 : Questionnaire to collect (available) basic regional information
and data
II) Annex 2: Assessment of relevant species
III) Annex 3: Summary feedback basic regional information and data
IV) Annex 4: Premises regarding the migration flow model
V) Annex 5: Summary complex geographic interconnection between
distribution, mating/reproduction and migration of the Nathusius’
pipistrelle
VI) Annex 6: Summary feedback on flow model
VII) Annex 7: Analysis of the reaction of the prototype estimator (the flow
model) through using a range of settings for the number of
juveniles/female and satellite males/male.
VIII) Annex 8: Preliminary estimates for migrating Nathusius’ pipistrelle over
the southern North Sea - histograms of test runs January 2017

Estimating the migration populations of bats on SNS
40
Annex 1:Questionnaire to collect (available) basic regional information
and data
Tables 1 to 6 present the questions used to assess B] basic regional data for the
estimator, as well as information regarding the (non)-availability of such data for
the specific states, nations or regions.
Table 1: Questionnaire to collect data on bat maternity roosts/colonies.
MATERNITY COLONIES
occurrence
is the species present in your country?
can you give an already existing estimate of the population size (EHD/EUROBATS
reporting e.g. all individuals year round)?
please add sources and publications
maternity colonies/roosts (colony = network of roosts)
Are there maternity colonies (network of roosts)?
can you give an estimate of how many maternity colonies exist?
can you give an estimate of the number or proportion of 'investigated' maternity
colonies?
can you give an estimate of the average number of individuals (females) in one colony
(network of roosts)? If only the number of total individuals (females and offspring) is
known, please indicate so under remarks
can you indicate whether the number of juveniles per female differs from 1? -if known
population estimates / your expert interpretation of above information
estimate of population (females) based on numbers of maternity roosts and average
number in maternity roosts
Does this number provide a good estimation of the population number (of females) in
your expert opinion? If not: what would be a good number?
Area's if possible
can you describe preferable foraging grounds for the species?
can you give an estimate of the average area ([foraging] home range) used by one
colony?
can you give an estimate (hectares or %tage region/country) of the available area that
could be qualified as habitat (roosts and foraging) for the species?
please add a sketch/sketches on a map where appropriate?
please add sources and publications

Estimating the migration populations of bats on SNS
41
Table 2: Questionnaire to collect data on bat summer roosts.
Summer roosts
Does the information below pertain to maternity roosts as well? If distinction
between summer and maternity roosts can be made, please use the two
separate
sheets.
summer roosts (individual roosts, network not known)
can you give an estimate of how many summer roosts exist?
can you give an estimate of the number or proportion of investigated summer roosts?
please add sources and publications
can you give an estimate of the average number of individuals in one summer roost?
can you indicate whether the number of individuals includes juveniles
population estimates / your expert interpretation of above information
estimate of population (all individuals) based on numbers of summer roosts and average
number in summer roosts
Does this number provide a good estimation of the population number (of females) in your
expert opinion? If not: what would be a good number?
Area's if possible
can you give an estimate of the average area ([foraging] home range) of a summer roost?
can you give an estimate (hectares or %-age region/country) of the available area that
could be qualified as habitat (roosts and foraging) for the species?
please add a sketch/sketches on a map where appropriate?
please add sources and publications

Estimating the migration populations of bats on SNS
42
Table 3: Questionnaire to collect data on bat mating roosts and clusters.
Mating roosts and clusters
Occurence
are there mating sites/territorial males?
are there clusters of mating sites/clusters of territorial males?
Mating roosts and/or clusters
can you give an estimate of how many (clusters of) mating sites exist?
can you give an estimate of the number or
proportion of investigated (clusters of) mating
sites/territorial males?
can you give an estimate of the average number of individuals in one mating site/cluster?
can you give an estimate of the average turnover of females in one mating site/cluster?
can you estimate the average number of sites where 1 female could be counted?
population estimates / your expert interpretation of above information
estimate of population (all individuals) based on numbers of mating roosts and average
number in (clusters of) mating roosts
Does this number provide a good estimation of the population number in your expert
opinion? If not: what would be a good number?
Area's if possible
can you give an estimate (hectares or %tage region/country) of the available area that
could be qualified as habitat (roosts and foraging) for the species?
please add a sketch/sketches on a map where appropriate?
please add sources and publications

Estimating the migration populations of bats on SNS
43
Table 4: Questionnaire to collect data on bat hibernation roosts.
Hibernation
hibernation roosts
are there hibernation roosts known?
can you give an estimate of how many hibernation roosts might exist?
can you give an estimate of the number or proportion of investigated hibernation roosts?
can you give an estimate of the average number of individuals in one hibernation roost?
population estimates / your expert interpretation of above information
estimate of population (all individuals) based on numbers of hibernation sites and average
number in hibernation sites
Does this number provide a good estimation of the population number in your expert
opinion? If not: what would be a good number?
area's if possible
can you give an estimate of the available area that could be qualified as habitat
(hibernation roosts) for the species?
please add a sketch/sketches on a map where appropriate?
please add sources and publications

Estimating the migration populations of bats on SNS
44
Table 5: Questionnaire to collect data on bat migration, part 1.
migration
Occurence
is there evidence of migration for the species?
what is the evidence?
Pathways?
what is the landscape/are the landscape structures used for migration?
what evidence is there for migration over land?
what evidence is there for migration along the sea shore?
what evidence is there for migration over sea?
Direction
is there an idea on the direction(s) of the migration?
what evidence is there to support this notion?
please add a sketch on a map where appropriate?
would you recognize one general direction or distinct directions?
if there is one general direction: what is the general direction of migration?
if there are distinct directions: what are the
possible different directions that might be
distinguished? Use the below cells to indicate appropriate %-age per direction
South
South East
East
North East
North
North West
West
South West

Estimating the migration populations of bats on SNS
45
Table 6: Questionnaire to collect data on bat migration, part 2.
migration
Size
can you give an estimate of what %
-age of your national/regional population would be
migrating?
please add a sketch/sketches on a map with %-ages where appropriate?
Distance (if known)
if one general direction: can you give an estimate of the migration distance?
can you give an estimate of the migration distance per specific migration direction?
South
South East
East
North East
North
North West
West
South West
please add sources and publications
funnel or broad front
is migration funnelled, or more like a broad front?
are there areas where 'funnelling' would be the right description?
are there areas where 'a broader front' would be the right description? Can you add those
to the map?
can you give an estimate of the available area that could be qualified as habitat migration
for the species?
please add a sketch/sketches on a map with %-ages where appropriate?
please add sources and publications

Estimating the migration populations of bats on SNS
46
Annex 2: Assessment of relevant species
Based on Leopold et al. 2014, updated with new information, an assessment of
the relevance of species to work with in the design of the prototype estimator is
performed. The relevance of a species is deduced from
- the occurrence of species offshore at sea,
- the abundance or (relative) numbers that might be present offshore
(expert judgement: abundance offshore is related to abundance onshore),
- the fatality risk that might arise from the species behaviour (such as
migration, flight height, accumulation around wind turbines), as well as
- the feasibility of assessing basic regional information B] necessary for the
flow model A].
Species potentially present on the (southern North) sea.
So far, 11 European bat species have been observed above sea in the north
western regions in Europe (Ahlén et al. 2009, Boshamer & Bekker 2008, Jonge
Poerink et al. 2013, Lagerveld et al. 2014a,b, 2015a,b. Petersen et al. 2014,
Walter et al. 2007).
These are: Daubenton’s Bat (Myotis daubentonii), Pond Bat (Myotis dasycneme),
Nathusius’ Pipistrelle (Pipistrellus nathusii), Common Pipistrelle (Pipistrellus
pipistrellus), Soprano Pipistrelle (Pipistrellus pygmaeus), Leisler’s Bat (Nyctalus
leisleri), Common Noctule (Nyctalus noctula), Northern Bat (Eptesicus nilssonii),
Serotine (Eptesicus serotinus), Particoloured bat (Vespertilio murinus) and Long-
eared Bat (Plecotus auritus).
Since 1988 collection of data from found offshore platforms – roughly
representing a sample for the southern North Sea - revealed Nathusius’
Pipistrelle (32x), Noctule (2x), Northern Bat (2x), Serotine (1x) and Particoloured
Bat (5x), with the highest incidence for Nathusius’ Pipistrelles. For Nathusius’
Pipistrelles and most other bat species (with the exception of the Noctule) no
bias towards platforms closer to the shore was observed. Bats were recorded at
distances of 60-80 km from the shore (Boshamer & Bekker 2008; Boshamer
pers. comm.). These are distances of 1/3 to 1/2 of the distance between the
southwest of the Netherlands and the UK, and larger than the distance between
Calais and the UK.

Estimating the migration populations of bats on SNS
47
Table 1: bats recovered from off shore oil platforms between 1988 and 20149 (after
Boshamer & Bekker 2008, pers. comm. )
#
%
Nathusius’ Pipistrelle
32
76
Noctule
2
5
Northern Bat
2
5
Serotine
1
2
Particoloured Bat
5
12
42
100
In (pilot) studies using real time recorders in wind parks on the southern North
Sea, Nathusius’ Pipistrelle, Common Pipistrelle, Noctule and probably
Particoloured Bat were recorded off shore. The majority of recordings could be
attributed to the Nathusius’ Pipistrelle (Jonge Poerink et al. 2013, Lagerveld et
al. 2014a,b, 2015a, b). Common Pipistrelle, Pond Bat and Daubenton’s Bat were
recorded on a site on the beach (Lagerveld et al. 2015b). In some recordings
from these studies the species comprising the ‘Nyctaloid’-group with the genera
Nyctalus, Vespertilio, and Eptesicus, could not be identified to the level of
species.
Based on ringing, long distance migration during which seas potentially have to
be crossed, is known for the Nathusius’ pipistrelle, the Noctule and the Parti-
coloured bat (Hutterer et al. 2005, Roer 1995). These data predominantly reflect
east/north east towards west/south west migration directions. (Hutterer et al.
2005, Jarzembowski 2003, Limpens & Schulte 2000, Limpens 2001, Lina 1990,
Massing 1988, Mitchell-Jones et al. 1999, Niederfriniger 1991, Oldenburg &
Hackethal 1989, Petersons 1990, Petersons & Lapina 1990, Roer 1995, Russ et
al. 2001, Voigt et al. 2012, 2016). Although ringed individuals from the larger
ringing schemes from Latvia and northern Germany would not need to cross the
sea, observations on e.g. Nathusius’ Pipistrelle and Noctule in southern Sweden
clearly indicate that the sea is crossed (Ahlén et al. 2009).
In 2013 the first Nathusius’ Pipistrelle ringed in the UK (ringed 2012 by Daniel
Hargreaves10 Blagdon lake/Somerset) was recovered on mainland Europe in the
Netherlands (December 2013 by Teddy Dolstra). Two ringed Nathusius’
Pipistrelles, one from Latvia and one from Lithuania, were recovered in Sussex
and Kent. These findings clearly showed that Nathusius’ Pipistrelles do cross the
North Sea.
Assessment of relevancy of different species
9 Due to changes in personnel, in the last few years the contact to those people from this industry willing to
collect found bats was lost.
10 http://www.bats.org.uk/pages/national_nathusius_pipistrelle_project.html

Estimating the migration populations of bats on SNS
48
On the off shore platforms 75% (N=42) of the observations are Nathusius’
pipistrelle, with 2 and 5 % for the Noctule and Party-coloured bat (Boshamer &
Bekker 2008; Boshamer pers. comm.).
The data from the real time recorders on off shore sites are even more skewed
towards the Nathusius’ Pipistrelle, with over 95% of acoustic recordings for
Nathusius’ Pipistrelle and only few Nyctaloids, and/or Noctules and Particoloured
Bats (Jonge Poerink et al. 2013, Lagerveld et al. 2014a,b, 2015a,b). Differences
in detection range of a species echolocation (e.g. peak frequencies, loudness,
pulse length, duty cycle), in relation to their flight style (flight height,
directionality of flight, hunting versus commuting), bias these data.
Data on ringed bats crossing to the UK are only known for the Nathusius’
Pipistrelle (Hargreaves pers. Comm./UK national Nathusius’ pipistrelle project).
In fatality searches in onshore wind farms, the Noctule and Nathusius’ Pipistrelle
are the most commonly found species with between 30 and 50% of casualties for
the Noctule and between 25 and 35% for Nathusius’ Pipistrelle (e.g. Brinkmann
et al. 2011, Dürr/LU Land Brandenburg 2016, Dürr & Bach 2004, Rydell et al.
2010).

Estimating the migration populations of bats on SNS
49
Table 2: indication of relevance for assessment of impact of off shore wind turbines on
species. The number of dots indicate a relative incidence for that label/row.
species observed
off shore 
Nathusius’ pipistrelle
Noctule
Parti-coloured bat
Common pipistrelle
Soprano pipistrelle
Daubenton’s bat
Pond bat
Leisler’s bat
Serotine
Northern bat
Brown long-eared bat
observation on
off shore oil platform





acoustic observation
off shore on southern
North Sea



banded individuals
potentially crossing
sea




banded individuals
crossing between UK
and mainland

casualties on land











The above data are all biased in various ways, but can still roughly be used to try
to deduce which species is relevant for impact assessment in relation to the
development of off shore wind parks (table 2).
The prevalence of the Nathusius’ Pipistrelle both as a fatality at wind turbines on
land, as in observations on (the southern North) Sea, and migration between UK
and mainland Europe, indicate this species as a relevant species in the
assessment of impact of off shore wind turbines on the species (Table 2).
The Noctule has an even higher incidence of fatality in onshore wind farms, but is
much less prevalent in the available (acoustic) data from the southern North
Sea. The Particoloured Bat has a lower incidence of fatality on land, but is also
much rarer than e.g. the Noctule and Nathusius’ Pipistrelle. Compared to the
Noctule, observations of the Particoloured Bat offshore are little higher on the oil
platforms, and of comparably low level in acoustic observations.
For both of the latter two species the evidence of, and expected (potential)
presence on the southern North Sea (ringing, acoustic, off shore oil platforms),
are much higher than the rest of the other species known to occur at sea. For
these two species this indicates that it is worth trying to assess the impact of
offshore wind turbines on the species (Table 2).

Estimating the migration populations of bats on SNS
50
The Common Pipistrelle can reveal a high incidence of fatality at onshore wind
farms, in situations where turbines and maternity roosts are in relatively close
proximity. The species is not observed on the oil platforms nor in the acoustic
data from the southern North Sea. For the other species known to potentially
occur over sea, there are low to accidental observations of fatalities at wind
turbines on land, and no acoustic data from the southern North Sea. Although a
fatality of such a species at a wind turbine at sea cannot be 100% ruled out,
based on current knowledge the fatality risk will be low (table 2).
Systematic data on occurrence and distribution (e.g. Limpens & Roschen 1996,
2002) of a species, such as resulting from national mapping projects, are not
available for all national territories around the southern North Sea.
Data directly related to the offshore situation (oil platforms, acoustic studies
offshore and on islands of the coast) are scarce, and certainly on this large
geographical scale. The available data are predominantly of Nathusius’ Pipistrelle.
Current data collection concerned with wind turbines offshore, on islands
off the coast and in wind farms close to the coast is steadily increasing the
available information.
Due to the relative lack of data for the Noctule and Particoloured Bat, we focus
on the Nathusius’ Pipistrelle.

Estimating the migration populations of bats on SNS
51
Annex 3: Summary feedback B] basic regional information and data
Table 1: overview feedback B] basic regional information and data
consent approach
basic estimates
elaborate data
and estimates
limited data not
including estimates
Finland
1
1
Estonia
Belarus
Latvia
1
1
Lithuania
Poland north-western
Sweden
1
1
Denmark
1
1
Germany north
1
1
Netherlands
1
1
1
Belgium
1
1
Norway
1
1
UK Scot
1
1
UK Northern Ireland
1
1
UK England/Wales
1
1
UK England/Wales
1
1
RO Ireland
1
1
France
1
1
Portugal
1

Estimating the migration populations of bats on SNS
52
Figure 1: overview feedback basic regional information and data (data processed till 20
Jan 2017)

Estimating the migration populations of bats on SNS
53
Table 2: summary feedback B] basic regional information and data
Regional data and estimates 
General
input
approach
Data / estimate s
Germany
Lothar & Petra Bach
x
Not able to estimate through lack of data
Johanna Hurst
x
Outside model region; lack of data
UK
Kathrine Boughey
x
Input via Jon Russ
Jon Russ,
x
Estimates England, Estimates Northern Ireland, no data
available for Scotland and Wales
John Haddow, Susan Swift, Daniel Hargreaves, Fiona
Mathews
x
Input via Jon Russ
RO Ireland
Niamh Roche
x
Data on ROI
Tina Aughney
x
Input via Niahm Roche
Scandinavian region
Norway
Jeroen van der Kooij
x
No countrywide data, overview of such data as are
available
Tore Chr Michaelsen
x
Input local observation points, no countrywide data
Sweden
Johnny de Jong
x
No countrywide data, overview of such data as are
available
Denmark
Hans Baagoe + Esben Terp
Fjederholt
x
No countrywide data, overview of such data as are
available
Ingemar Ahlén, Jan Durinck, Morten Christensen,
Morten Elmeros, Thomas W. Johansen and Julie Dahl
Møller
x
Input via Baagoe and Fjederholt
Finland
Eeva-Maria Kyheröinen
x
No countrywide data, overview of such data as are
available
Baltic region
Latvia
Gunars Peterson
x
No countrywide data, overview of such data as are
available
Estonia,
Lithuania,
Belarus, Poland
Belgium
Bob Vandendriessche
x
No countrywide data, overview of such data as are
available, estimate based on road migration front.
SMITS Quentin
France
Marie-José Dubourg-
x
No countrywide data
Thomas Le Campion, Dorothee Jouan, Diane Anxionnat
x
Input local observation points, no region wide data
Iberic region
Portugal
Luisa Rodrigues
x
Outside model region; lack of data
Nederland
Herman Limpens Eric Jansen,
x
Estimates for the Netherlands
Peter Twisk
x
No input
Jasja Dekker, A-J Haarsma, Jan Boshamer, Theo
Douma
x
Input local observation points
Marcel Schillemans

Estimating the migration populations of bats on SNS
54
Annex 4: Premises regarding the migration flow model
Note: This annex is a more elaborate version of the description of the premises
described in paragraph 2.5.
a) In this study, due to the relative lack of data for the Noctule and Parti-
coloured bat, we will focus on the Nathusius’ pipistrelle. Although of course
there are differences between the species, we think that developing and
testing the approach for the Nathusius’ pipistrelle, will make future
application for other species easier. The knowledge gaps identified for the
Nathusius’ pipistrelle will only be larger for the other species.
b) In our interpretation we focus on information related to the autumn
migration period.
c) Information from the questionnaires and other contributions of the
respondents/participants in this study, indicate a main migration direction
from east/north east to west/south west. However a smaller part of the
population may have more west or more southwest directions. This is in
accordance with literature (Hutterer et al. 2005, Jarzembowski 2003,
Limpens & Schulte 2000, Limpens 2001, Lina 1990, Masing 1988, Mitchell-
Jones et al. 1999, Niederfriniger 1991, Oldenburg & Hackethal 1989,
Petersons 1990, Petersons & Lapina 1990, Roer 1995, Russ et al. 2001,
Voigt et al. 2012, 2016).
At the same time migration in more southward direction may be observed
along coastlines (e.g. along the Dutch province of North-Holland) and
through river valleys (input L. & P. Bach, J. Hurst) such as the Rhine valley
and the Elbe valley, and even in south/southeast direction along the east
coast of Denmark (input Baagøe and Terp Fjederholt). Drift cannot be
excluded (Hüppop & Hill 2016).
We use the premises of an east/north east to west/south west migration
direction in our model, although this is most likely a simplification of reality.
Applying this assumption results in countries where exchange (influx/outflux)
can be expected (i.e. from Poland to Germany) and countries where
out/influx is very unlikely to occur (i.e. from Sweden to Poland).

Estimating the migration populations of bats on SNS
55
Figure 1: Main migration direction, and other observed directions (see text).
d) Almost independent from the geographical location, data from studies related
to wind turbines on land in Europe, reveal peaks of observations of
Nathusius’ Pipistrelles in autumn (questionnaires and literature: Ahlén 1997,
Ahlén et al. 2007, 2009; Brinkmann et al. 2011; Lagerveld et al. 2014a).
This suggests a broader migration front rather than narrow corridors.
However, the relative numbers observed in such studies, as well as visual
observation in the landscape, suggest the existence of funnelling along the
coast and along river valleys. This may also lead to a migration direction
following the coast line or river valley in a more north to south direction.
e) Larger areas of water will be crossed, probably mostly at sites where there is
no other (logical) choice (see e.g. : Ahlén 1997, Ahlén et al. 2007, 2009),
e.g. in situations such as in
- Falsterbo, or the south points of Isles like Gotland and Öland in Sweden,
where all directions other than back (back north/north-east) lead to open
water,
- Westkapelle (and the west point of other islands) in the south west of the
(province of Zeeland in the) Netherlands, where the coastline turns land
inward (direction SE and E), and following the coastline would mean a large
deviation of the general migration direction,
- and probably also in situations where the coastline deviates from the
general migration direction for long distances (long periods of flying), such as
found along the N/S coast lines of the west coast of Denmark and the
province of North-Holland in the Netherlands.
f) Based on the selection of relevant species and the developed approach,
source and target areas of bats migrating above and/or along the coast of
the southern North Sea are identified. Calculating from the northern and

Estimating the migration populations of bats on SNS
56
north-eastern most of the countries around the southern North Sea, and
based on a rough interpretation of the geographical migration pattern of the
Nathusius’ Pipistrelle (www.grida.no12, Limpens & Schulte 2000; see fig. 3)
we constructed a flow model for the setting given in fig. 6.
Figure 2: Nathusius’ pipistrelle distribution and migration (From collection: Living Planet:
Connected Planet, Rapid Response Assessment, Riccardo Pravettoni, UNEP/GRID-Arendal, 2011).
Ad figure 2: Note: We use this illustration because it gives a quit good overview, but since 2011
new data is available. Please note, however, that no maternity groups are ever discovered in
Norway, nor were any of the captured Nathusius’ pipistrelles females (J. v.d. Kooij, pers. Com.). A
possible flux from Sweden to Norway is not taken into account (J. v.d. Kooij, pers. Com.). Also all
areas without breeding, in the sense of occurrence of maternity sites, are qualified as hibernation
area. The fact is that many sites with (clusters of) mating roosts are found in both the maternity
and hibernation area.
We incorporated Norway, UK (Scotland, Northern-Ireland, England and
Wales), the Republic of Ireland, Finland, Sweden, Estonia, Latvia, Lithuania,
north western part of Poland, Denmark, northern part of Germany, the
Netherlands and Belgium in the flow model.
The migrating bats (the ‘migration flow of individuals’) of these countries
might be arriving at the southern North Sea, where populations migrating
from/though Norway would end up in the northern North Sea. Norway is
included, however, because their outflux may be an influx to Scotland and
Northern Ireland. Those migrating from/through Luxemburg, the middle and
12 http://www.grida.no/graphicslib/detail/nathusius-pipistrelle-distribution-and-migration_18cb

Estimating the migration populations of bats on SNS
57
south of Germany, middle and east of Poland, Czech Republic et cetera would
end up more to the south of the current study area, in the direction of
France, Switzerland, Italy, and Spain, or even more south in central southern
Europe.
g) In this study, some of the countries will be regarded as being ‘source
countries’ (e.g. Finland or Norway), where based on the distribution of the
species, in our model we assume that there is no influx of migrating bats
from neighbouring countries (also see fig. 6). Norway, however is no direct
source fot the SNS. Russia, more specific the western bordering territory, is
only incorporated in the model as a source area, although undoubtedly there
will be migration from more eastern Russian territory. Others will be
countries where there is an influx and outflux of migrating bats (e.g.
Denmark). Norway
h) We work on the premise13 that the majority of the animals from the northern
(Scandinavia) and north-eastern European maternity regions (Baltic States,
western Poland and western Russia) will hibernate in western Europe in the
present case in middle-western Europe.
i) We work on the premise that the number of females born equal the number
of males, and that there is an equal survival for both sexes.
j) Males are territorial in the mating season. The ratio between territorial males
and satellite males is unknown. It is unclear whether there would be different
rations for different zones in the migration direction. There is some evidence
that the number of satellite males, not occupying their own territory, is about
4/5 of the population of males (e.g. Lundberg 1989, Lina pers. comm). In the
calculation of the flow model this proves to be a factor with a large influence
on the outcome. Therefore it is necessary to validate and update this factor.
Here we try to compensate for this uncertainty by working with a large
uncertainty interval as well as different settings within this interval. In the
flow model we calculated with the following variability in this parameter:
lower-central-upper: 0.2–2–5, and a range of different settings: 3–4–5, 2–
3–4, 1-2-3 and 0.2-1.2-2.2 satellite males per territorial male (pers. com.
Peter Lina).
k) We might work on the premise that all young males take part in migration
males, as do all adult and young females, where all adult males do not take
part in migration. There is, however, evidence that some of the adult males
do take part in migration. Boshamer & Bekker (2008) find some adult males
13 For all premise(s): premise = interpretation of the available knowledge in a way that allows for calculation in
the estimation.

Estimating the migration populations of bats on SNS
58
on off shore platforms. Lina (pers. comm.) finds adult males with non-
swollen buccal glands, together with females and the resident male in bat
boxes. It is unclear whether these are ‘resident’ satellite males, or migrating
males. We therefore also work with the assumption that at least a part of the
the non-resident male population, the population that not occupies their own
mating roost/territory, also migrates.
l) We work on the premise that
- in all geographic regions, more females will take part in migration than
males;
- the more south/southwest in Europe we get, the less males take part in
migration, the more males stay behind (territorial and/or, hibernation)
- in the autumn, in the northeast of Europe most females will migrate,
while more to the south west a larger part will stay behind (hibernate).
m) There are very few data on the fecundity of bats. Available data present J/F
ratios of around 0.6 to 0.8 (e.g. O’Shea et al. 2011, Schaub et al. 2007).
We work on the premise that every adult female that is part of the western
Europeans migrating population will on average have 0,7 offspring per year.
In the flow model we calculated with the following variability in this