Acciona Photo Simulations of turbines for St. Lawrence Wind Complex off Millen Bay
Richards Bay Wind Energy Facility - CESNET Bay Wind Energy Project TK06.… · Bio3 owns a Merlin...
Transcript of Richards Bay Wind Energy Facility - CESNET Bay Wind Energy Project TK06.… · Bio3 owns a Merlin...
Richards Bay Wind Energy Facility
Bat Community Monitoring
August 2013
Final Report (pre-construction phase)
In collaboration with
1 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
E X E C U T I V E S U M M A R Y
The main results of the bat community pre-construction monitoring programme of the Richards
Bay Wind Energy Facility are presented in this report. Vehicle based transects (with a manual
echolocation detector and recorder), static detection at ground level and height, bat live trapping
with mist-netting and harp trapping, as well as bat roost searches and inspection were
implemented during the monitoring surveys conducted between May 2012 and April 2013.
These methodologies resulted in the confirmation of occurrence of 21 bat species and the
potential occurrence of 15 additional species in the study area. Among the confirmed species is a
fruit-eating bat, Wahlberg's epauletted fruit bat (Epomophorus wahlbergi), with reproduction
confirmed in a roost located in the southern portion of the study area. Six of the confirmed
species are considered to be of conservation concern, classified as “Near Threatened” by the
South Africa Red List: Anchieta's pipistrelle (Hypsugo anchietae), Lesser long-fingered bat
(Miniopterus fraterculus), Natal long-fingered bat (Miniopterus natalensis), Welwitsch's myotis (Myotis
welwitschii), Temminck's myotis (Myotis tricolor) and Geoffroy's horseshoe bat (Rhinolophus clivosus).
An additional suspected species in the study area is classified as “Vulnerable”, the Large-eared
giant mastiff bat (Otomops matiensseni). Bat activity in the study area seems to be higher during
spring and summer months, when food availability is higher. The results obtained in this report
suggested that bat activity was generally higher at heights below 7 to 30 m in relation to heights of
60 or 80 m, which was confirmed with statistical significance. Nonetheless, the placement of rotor
height higher than 60 m could contribute mitigating the potential collision risk identified. Among
the environmental variables with significant importance to explain bat activity in the study area,
wind speed and air temperature are considered those with more influence on the critical night
periods where bats are more susceptible to potential impacts by wind turbines. Several confirmed
bat roosts were identified by NSS field team in the study area providing roosting for at least eight
species. These were spread throughout the study area, yet the northern part of the wind energy
facility site presented a higher density of roosts. It is of note that the proposed Richards Bay wind
energy facility is located within the broader vicinities of five important roost caves, occupied by
migratory species such as Miniopterus natalensis.
Both the analysis of bat activity and environmental features in the study area led to the
classification of the study area as a medium to high sensitivity area for bats, especially in the
northern area due to its environmental features. This area is classified as a high sensitivity area
and is located in the Endangered Zululand Coastal Thornveld, with remaining patches of
appropriate foraging habitat, such as native vegetation, contributing to the occurrence of higher
bat activity areas and several roosting areas. The proposed site for the Richards Bay Wind Energy
Facility is located in an area considered to be of medium to high sensitivity for bats, especially in
the northern section of the site where bat activity proved to be higher.
Considering the potential impacts of collision fatalities of bat species occurring in the area it was
important to analyse their risk of collision with wind turbines. This analysis has shown that 4
confirmed species have a potential high risk of collision with wind turbines and 9 others have
2 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
medium to high potential collision risk. These species may be affected by the operational phase of
this project and it will be necessary to implement mitigation measures, suggested in this report, to
reduce the probability and significance of such impacts on local bat communities.
The analysis of potential negative impacts conducted at this stage of the pre-construction
monitoring programme highlighted the occurrence of high to moderate significance impacts,
mostly caused by the possibility of mortality of bats due to collision with wind turbines and/or
barotraumas, and also caused by bat displacement due to habitat loss. Among the mitigation
measures proposed, relocating the most sensitive turbines and adopting blade feathering are
considered very important to reduce likelihood of the impacts impact.
3 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
T ECHNI CAL TEAM
The bat monitoring programme was established and implemented by NSS – Natural Scientific
Services CC which was appointed by NEA Renewable Energy (Pty) Ltd. All the baseline data, field
methodology approach (except when stated otherwise) and data collection are therefore solely
the responsibility of NSS. Bio3 Lda, in collaboration with Savannah Environmental Pty (Ltd), was
appointed by NEA Renewable Energy (Pty) Ltd to compile the final report of the pre-construction
monitoring programme based on NSS baseline data.
The technical team responsible for this final report is presented in Table 1.
Table 1 – Technical team
Technician Qualifications Role on project
Joana Marques Masters in Ecology and Environmental management
Degree in Environmental Biology Report compilation
Mario Carmo
Masters in Natural Resources Management and Conservation (under
conclusion)
Degree in Biology
Report compilation
Dr. M. Corrie
Schoeman
Evolutionary Ecologist at University of KwaZulu-Natal
Ph.D (University of Cape Town)
B.Sc Honours summa cum laude (University of Cape Town)
B.Sc with distinction (University of Cape Town)
Scientific advisor on Bat
ecology, distribution and
identification
Ricardo Ramalho PhD: Environmental studies
BSc Honours Degree in Biological Sciences Technical coordination
Mike Cohen D Sc (University of Pretoria) Wildlife Management
Registered Professional Zoologist (SACNASP 401917/83)
Report review
Scientific supervision
Hugo Costa
Degree in Biology Applied to Animal Resources – terrestrial derivation
Masters in Impact Assessment
Technician Specialised in Environment
Coordination
Jo-Anne Thomas
BSc Honours Degree in Natural Science
Masters of Science degree in Natural Science
Registered Professional Natural Scientist (Pr.Sci.Nat)
Coordination
Report review
Report delivered in 28th of August 2013.
Bio3 is an international consultancy and research company that reconciles human development
with biodiversity management and conservation.
Founded in 2005, Bio3 have more than 8 years of experience, have conducted the
ecological/biological assessments of over 440 projects, mostly related to ecological impact
4 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
assessment, biological monitoring, biodiversity management planning. Bio3 also executes applied
research studies.
Bio3 is an international expert and market leader on Ecological Baseline Studies and Biodiversity
Monitoring Surveys, with an emphasis on renewable energy projects. With over 80 Wind Farm
projects executed and 10 Hydropower, Bio3’s clients include large renewable energy companies,
such as EDIA, EDP Renewables, ENEOP2, ENERSIS, GALP Energia, GENERG, IBERWIND, REN
and Ventinveste. Bio3’s complete portfolio can be consulted at www.bio3.pt/en.
As a result of a R&D work that began in 2005 and in 2011 Bio3 was considered one of the most
innovative companies in Portugal having been integrated into COTEC’s Innovation PME Network.
Bio3 owns a Merlin Avian Radar System among many other technologies for Bird and Bat
monitoring at Wind Farms and other types of projects. The two most relevant R&D projects
from Bio3 are:
Wind & Biodiversity (2011-2015): In February 2011, Bio3 initiated the Wind & Biodiversity
project, in co-promotion with two research units of the University of Aveiro, the Associated
Laboratory CESAM – Centro de Estudos do Ambiente e do Mar (Centre for Environmental and
Marine Studies) and IEETA – Instituto de Engenharia Electrónica e Telemática de Aveiro (Institute
of Electronics and Telematics Engineering of Aveiro). This project aims to develop the
technologies and know-how that will allow the development of effective strategies to reconcile
wind farms with biodiversity and focus mainly on the impacts on bat and bird communities.
Biologist-dog (2008-2012): Bio3 provides services using trained dogs to detect birds and bats’
corpses at wind farms and around power lines, in order to accurately quantify the real impacts of
human infrastructure on biodiversity. This service results from a R&D project started in 2008
with the establishment of a pioneer protocol between Bio3 and the Portuguese Public Security
Police K9 Special Unit.
The team appointed by Bio3 for this project is coordinated by Ricardo Ramalho. He has a PhD in
Environmental Sciences and holds an Honours Bachelor of Science degree in Biology. He has 8
years of experience consulting in the environmental assessment field, with special emphasis on
fauna and flora assessments of large projects such as wind farms, power lines, roads and hydro-
power facilities. His key focus is on bat and bird communities’ impact assessment and monitoring.
He has managed more than 25 Ecological monitoring studies, 15 EIA projects and 15 SEA
projects. Has participated in several R&D and ecological monitoring projects and has authorship
on more than 10 scientific publications. Has experience in international ecological monitoring and
EIA assessments and has been the technical coordinator on 23 bat monitoring programmes in
wind energy facility projects in South Africa, on the Western, Eastern and Northern Cape and
Free State Provinces.
C I T A T I O N
Recommended citation when using this report as a reference: Bio3 and Savannah Environmental
(2013). Richards Bay wind energy facility – Bat community monitoring. Pre-construction phase. Final
report.
5 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
C O P Y R I G H T
This report was compiled for NEA Renewable Energy Pty (Ltd) by Bio3 Lda. and Savannah
Environmental Pty (Ltd). The contents of the report are Bio3 and Savannah Environmental
intellectual property. These should not be reproduced or used by third parties without prior
written consent.
6 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
P R E F A C E : B A T S A N D W I N D T U R B I N E S
Currently wind is considered worldwide as one of the most promising renewable energy sources.
Wind farm infrastructures in operation do not produce any carbon emission. Even considering
the total carbon emissions caused by the installation of wind turbines and ancillary infrastructure
and its maintenance, this is the source of energy with the lowest emissions developed to date
(EWEA, 2011). For this reason, it is considered that the expansion of wind power contributes
positively to the reduction of climate change caused by increasing human energy needs.
Wind power has grown exponentially in the last decade and it is one of the main alternative
energy sources to fossil fuels (Gsänger and Pitteloud, 2013). Its development in South Africa has
just started and by the end of 2012 only 10 MW were installed in the country (Gsänger and
Pitteloud, 2013). Because of the growing demand for electricity in South Africa and concerns
about climate change, the South African government has set targets to produce 10 000 GWh of
renewable energy in 2013 whereas 1 850 MW are expected to be provided by wind energy.
South Africa, is the country of the African continent with the largest CO2 emission, it is also
considered to represent one of the fastest growing wind energy industry markets (Mukasa et al.,
2013).
This energy source is however not free from environmental impacts. The installation of wind
energy facilities around the world has revealed some issues regarding wildlife conservation
(Eichhorn and Drechsler, 2010), specially related to bird (Barrios and Rodriguez, 2004; Drewitt
and Langston, 2008) and bat communities (Johnson et al., 2003; Barclay et al., 2007; Arnett et al.,
2011). Beyond the birds and bats, habitat loss affects all existing biodiversity (Kikuchi, 2008).
The impact on natural populations is not only due to direct mortality caused by collisions and
barotrauma1, the latter affecting bats only (Baerwald et al., 2008). Impact on natural populations
may also be caused by the disturbance effect, barrier effects and habitat loss (Drewitt and
Langston, 2006). These impacts, especially mortality, have become a source of major concern
among a number of stakeholder groups (Estep, 1989; Erickson et al., 2002). Results obtained
during several international monitoring studies indicated that wind farms were responsible for the
decrease in population of some species’ (Hunt, 2002; Carrete et al., 2009) although many other
studies revealed that these impacts were not important when compared to those originating from
other man-made infrastructures (Erickson et al., 2001; Drewitt and Langston, 2008). Nevertheless,
the potential for wind farms to affect bat populations should not be underestimated (Hunt, 2002;
Madders and Whitfield, 2006).
1 Barotrauma is used in the present report referring to bat deaths due to tissue damage to air- containing structures
caused by rapid or excessive pressure change close to the rotating wind turbine blades surface. Death is usually caused
by pulmonary barotrauma where lungs are damaged due to expansion of air in the lungs that is not accommodated by
exhalation (Baerwald et al., 2008).
7 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Extensive research has been conducted internationally regarding bats and wind farms (e.g. Arnett
and Kunz, 2008; Baerwald et al., 2008; Horn, 2008; Arnett et al., 2011). However, not much
research has been conducted on these matters in South Africa until recently. Research about
seasonal and daily movement patterns of bat species and what the potential impacts of the
development of multiple wind energy facilities and thousands of turbines across the country might
be has been lacking and has begun only recently.
Also, information regarding bat distribution, seasonal and daily movements and migration is very
limited for South African bat communities. Therefore, the need to evaluate the potential effects
and interactions between bats and wind energy facilities is more relevant in South Africa, since the
countries’ experience in wind energy generation has been extremely limited to date and wind
energy developments are currently under expansion. Until recently only eight wind turbines had
been constructed, 3 at a demonstration facility at Klipheuwel in the Western Cape, in 2002 and
2003, 4 at a site near Darling, and 1 at Coega near Port Elizabeth. Moreover, to date only a 1 year
preliminary study assessing bird and bird fatalities has been completed in South Africa and the
results published, reporting bat and bird fatalities produced by wind energy facilities (Doty and
Martin, 2013). This study was undertaken at a pilot turbine installed in the Coega Industrial
Development Zone, Port Elizabeth, Eastern Cape, where a total of 18 bat fatalities were recorded
over a 12 month period. Another short pilot study (over a 2 month period, covering solely a bat
migration period) was conducted in the experimental Darling wind energy facility where only one
bat fatality was recorded (Aronson et al., 2013). The potential impacts of wind turbines on South
African bat communities is still largely unknown, due to a lack of research on bats in the country
and a poor level of knowledge on bat abundance, locations of roost sites, and both foraging and
migratory behaviour. Therefore, data collection and further investigations are needed. Pre- and
post-construction monitoring at wind energy facilities can go some way to filling these gaps and
promoting the sustainability of wind energy developments in South Africa.
The Guidelines for Surveying Bats in Wind Farm Developments (Sowler & Stoffberg, 2012) were
developed in collaboration with the Endangered Wildlife Trust (EWT). These guidelines provide
technical guidance for consultants to carry out impact assessments and monitoring programmes
for proposed wind energy facilities, in order to ensure that pre-construction monitoring surveys
produce the required level of detail for authorities reviewing environmental authorisation
applications. These guidelines outline basic standards of best practice and highlight specific
considerations relating to the pre-construction monitoring of proposed wind energy facility sites
in relation to bats.
8 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
INDEX PAGE
1. I N T R O D U C T I O N .............................................. 10
1.1. O B J E C T I V E S O F T H E M O N I T O R I N G P R O G R A M M E ....................... 10
1.2. S C O P E O F W O R K A N D T E R M S O F R E F E R E N C E ......................... 11
1.3. L E G A L F R A M E W O R K ........................................... 12
1.4. P R O P O S E D W I N D E N E R G Y F A C I L I T Y A N D S U R V E Y A R E A ................. 13
1.5. S U M M A R Y O F T H E EIA ........................................ 20
2. M O N I T O R I N G P R O G R A M M E D E S C R I P T I O N ............................ 22
2.1. D E S K T O P P R E P A R A T O R Y W O R K ................................... 22
2.2. F I E L D S U R V E Y S ............................................. 24
2.3. I M P A C T S E V A L U A T I O N ......................................... 38
2.4. A S S U M P T I O N S A N D L I M I T A T I O N S ................................. 43
3. R E S U L T S A N D D I S C U S S I O N ..................................... 46
3.1. S P E C I E S P R E S E N T A T T H E S I T E ................................. 46
3.2. F I E L D S U R V E Y S ............................................. 52
3.3. S P A T I A L- T E M P O R A L A C T I V I T Y .................................. 60
3.4. U S E O F R O O S T S ............................................. 90
4. P O T E N T I A L S E N S I T I V E A R E A S O N T H E W I N D E N E R G Y F A C I L I T Y ....... 94
4.1. T U R B I N E S E N S I T I V I T Y A N A L Y S I S ................................ 97
5. P O T E N T I A L I M P A C T S A S S E S S M E N T ................................ 99
5.1. I N T E R A C T I O N S B E T W E E N W I N D E N E R G Y F A C I L I T I E S A N D B A T S ........... 99
5.2. I M P A C T A S S E S S M E N T - C O N S T R U C T I O N P H A S E ...................... 109
5.3. I M P A C T A S S E S S M E N T - O P E R A T I O N A L P H A S E ....................... 113
5.4. I M P A C T A S S E S S M E N T - D E C O M M I S S I O N I N G P H A S E ................... 116
9 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5.5. M I N I M I S A T I O N A N D/ O R M I T I G A T I O N M E A S U R E S ..................... 117
6. I M P A C T S T A T E M E N T ......................................... 121
7. C O N C L U S I O N S A N D R E C O M M E N D A T I O N S ............................ 123
7.1. M A I N R E S U L T S O F T H E P R E- C O N S T R U C T I O N M O N I T O R I N G P R O G R A M M E ...... 123
7.2. R E C O M M E N D A T I O N S F O R T H E N E X T P H A S E S O F T H E P R O J E C T ............ 125
7.3. S U I T A B I L I T Y O F T H E M O N I T O R I N G P R O G R A M M E ..................... 126
8. R E F E R E N C E S .............................................. 128
9. A P P E N D I C E S .............................................. 136
9.1. A P P E N D I X I - F I G U R E S
9.2. A P P E N D I X II - S A M P L I N G P O I N T S D E S C R I P T I O N
9.3. A P P E N D I X III - S U M M A R Y O F T H E N U M B E R O F R E C O R D I N G S A N A L Y Z E D P E R
S P E C I E S
9.4. A P P E N D I X IV - C O L L I S I O N R I S K A N A L Y S I S FO R T H E B A T S P E C I E S O C C U R R I N G
A T T H E S I T E
9.5. A P P E N D I X V – B R I E F D E S C R I P T I O N O F B A T S P E C I E S W I T H O C C U R R E N C E I N
R I C H A R D S B A Y S I T E
9.6. A P P E N D I X VI – S U M M A R Y O F T H E I N F O R M A T I O N R E C E I V E D F O R T H E
C O M P I L A T I O N O F T H I S R E P O R T
9.7. A P P E N D I X VII – P O T E N T I A L B A T C O L L I S I O N R I S K W I T H W I N D T U R B I N E S F R O M
A C C O R D I N G L Y T O S O W L E R A N D S T O F F B E R G (20 12)
9.8. A P P E N D I X VIII – T U R B I N E I M P A C T A N A L Y S I S R E S U L T S
9.9. A P P E N D I X IX – P R O P O S E D B A T M O N I T O R I N G P R O G R A M M E
10 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
1. I N T R O D U C T I O N
This document is the final report of the bat community pre-construction monitoring programme
(comprising the monitoring period from May 2012 to April 2013) at the Richards Bay Wind
Energy Facility site being developed by NEA Renewable Energy (Pty) Ltd.
In this report the data collected during the one-year monitoring surveys conducted between May
2012 and April 2013 is presented. A detailed characterisation of bat communities within this
period is the primary objective of this report and will contribute to the general year-round
evaluation during the pre-construction phase. The purpose of this monitoring was to undertake a
general characterisation of bat communities, and allow the establishment of a baseline scenario for
the pre-construction phase. Simultaneously the data gathered during the pre-construction phase
of the development will contribute towards determining potential impacts of the construction and
operation of the wind energy facility on the bat community.
1.1. O BJECT IVES OF T HE M ONITO RING P ROG RAMME
The main objective of the Richards Bay Wind Energy Facility (WEF) monitoring programme is to
characterise the bat community in the area, and assess its potential impacts.
The specific objectives of the bat monitoring programme are:
a) Establish the baseline reference and characterization of the bat communities occurring at
the development area (e.g. species occurrence, activity and distribution);
b) Identify the potential changes in the bat community present within Richards Bay wind
energy facility site and the eventual exclusion effect caused by the projects’ presence
and/or operation (avoidance of the wind facility area during the operational phase of the
project);
c) Assess the use of roosts in the wind energy facility development footprint and its
immediate vicinity;
d) Quantify bat fatalities associated with the wind energy facility during the operation phase
of the project and determine the species affected2;
e) Identify potential impacts from the wind energy facility on the bat community and
propose adequate monitoring, mitigation or, if unavoidable, compensation measures.
In order to achieve the objectives of the bat monitoring programme an experimental protocol
was established, covering the wind energy facility site, and hence comply with the main
requirements of the “South African Good Practice Guidelines for Surveying Bats in Wind Farm
2 Despite one of the main specific objectives of the bat monitoring programme is the determination of bat mortality
associated with the wind energy facility, this goal will only be achieved during the operational phase of the project.
11 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Developments” (Sowler and Stoffberg, 2012) and the major indications from the Environmental
Impact Assessment (EIA) report of the Richards Bay Wind Energy Facility (CES, 2012). The bat
monitoring programme was established and implemented by NSS – Natural Scientific Services CC
which was appointed by NEA Renewable Energy Pty (Ltd). All the baseline data, field methodology
approach (except when stated otherwise) and data collection are therefore the sole responsibility
of NSS. Bio3 Lda/Savannah Environmental Pty (Ltd) was appointed by NEA Renewable Energy Pty
(Ltd) to conduct the data analysis and compilation of the present report, including only one field
survey with the main objective of site recognition and habitat characterization.
In order to accomplish the above mentioned objectives, the monitoring work of the community
of bats shall include the following tasks:
• Sampling of ultrasound in the wind energy facility site and in a control area – to be
conducted during pre-construction, construction and operation phases. This task will
provide data that will enable to accomplish Objectives a) and b);
• Bat carcass searches around the turbines - to be conducted during the operation
phase. This task will provide data that will enable Objective d) to be accomplished:
• Searcher efficiency and carcass removal (by scavengers or decomposition) trials -
during operation phase. This task will provide data that will enable Objective d) to be
accomplished;
• Inventory, search, inspection and monitoring of shelters in the area surrounding the
wind energy facility – during pre-construction and operation phases. This task will
provide data that will enable Objective c) and complementary compliance with
Objective b) to be accomplished.
All the above methodologies will enable the accomplishment of Objective e).
The results of the pre-construction monitoring will contribute to the establishment of the
baseline situation henceforth allowing the accomplishment of all the objectives stated in future
phases of the project. More specifically, this phase will contribute to the characterization of the
bat community present in the study area and evaluate bat habitat use at project location. The
assessment of potential bat fatalities associated with the Richards Bay wind energy facility will be
subject of the monitoring programme to be implemented during the operational phase of the
development.
1.2. S COPE OF W ORK A ND T ERMS OF RE FERENC E
The following assessment was conducted according to the specialist terms of reference and was
based solely on the methodological design and field data collected by NSS:
Review of international literature and experience relating to operational wind farms -
including other facilities around the world;
Description of the affected environment and determine the bat species present in the
future impact site;
12 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Identification of species of special concern and assessment of potential effects of the
development on the bat community;
Assessment of how the bat community will be affected by the proposed development,
listing, describing and evaluating potential impacts;
Mapping of sensitive areas in and around the proposed wind energy facility site;
Recommendations for relevant mitigation measures which will allow the reduction of
negative effects and maximization of the benefits associated with any identified positive
impacts;
Proposal of a suitable monitoring programme for the evaluation of the impacts expected
during the operational phase of the development, if considered necessary.
1.3. L EGAL FRAME WORK
There are no permit requirements dealing specifically with bats in South Africa. It is considered
best practise for bat monitoring to be undertaken on wind energy facility sites, in order to fulfil
the requirements outlined by the South African Good Practice Guidelines for Surveying
Bats in Wind Farm Developments (Sowler and Stoffberg, 2012). Legislation dealing with
mammals applies to bats and includes the following:
National Environmental Management: Biodiversity Act, 2004 (Act 10 of 2004):
The National Environmental Management: Biodiversity Act (Act 10 of 2004) (NEMBA) provides
for listing threatened or protected ecosystems, in one of four categories: critically endangered
(CR), endangered (EN), vulnerable (VU) or protected. The Act calls for the management and
conservation of all biological diversity within South Africa.
NEM:BA also deals with endangered, threatened and otherwise controlled species, under the
ToPS Regulations (Threatened or Protected Species Regulations). The Act provides for listing of
species as threatened or protected, under one of the following categories:
Critically Endangered: any indigenous species facing an extremely high risk of extinction in
the wild in the immediate future.
Endangered: any indigenous species facing a high risk of extinction in the wild in the near
future, although it is not a critically endangered species.
Vulnerable: any indigenous species facing an extremely high risk of extinction in the wild
in the medium-term future; although it is not a critically endangered species or an
endangered species.
Protected species: any species which is of such high conservation value or national
importance that it requires national protection. Species listed in this category include,
among others, species listed in terms of the Convention on International Trade in
Endangered Species of Wild Fauna and Flora (CITES).
13 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
A ToPS permit is required for any activities involving any ToPS-listed species. A number of bat
species are listed as critically endangered, endangered, vulnerable and protected in terms of
Regulations published under this Act.
KwaZulu-Natal Nature Conservation Management Amendment Act, No. 5 of 1999:
The primary purpose of this Act is to provide for the appointment of honorary officers; to
provide for the conservation of plants and animals; to provide for the control of hunting; to
provide a procedure for the issue and enforcement of permits; and to provide for matters
incidental thereto. The Fourth Schedule provides a list of specially protected indigenous animals,
including some bat species. Protected indigenous animals, including a number of bat species, are
listed in the Fifth Schedule.
IUCN Red List of Threatened Species
The International Union for the Conservation of Nature (IUCN) Red List of Threatened Species
ranks plants and animals according to threat levels and risk of extinction, thus providing an
indication of biodiversity loss. This has become a key tool used by scientists and conservationists
to determine which species are most urgently in need of conservation attention. In South Africa,
a number of bats are listed on the IUCN Red List.
1.4. P ROPOS ED W IND ENERG Y FA CILIT Y AND SUR VEY ARE A
The Richards Bay wind energy facility proposed layout includes 39 wind turbines, distributed
across an area of approximately 5950 ha. Since the final wind turbine specifications were not
defined at the date that this report was compiled, the following characteristic ranges were
considered: hub height may vary between 95m and 120m and rotor diameter may be 90m to
120m. The proposed wind energy facility site is to be located in the KwaZulu-Natal Province,
approximately 1km northeast of Empangeni, across 31 farm portions. The study area includes the
proposed wind energy facility, to the northeast of Empangeni, and is bisected to the south by the
national road N2 (refer to Figure 39 – Appendix I).
The wind energy facility site is located within the Zululand Coastal Thornveld and Maputaland
Coastal Belt ecosystems, characteristic of the two biomes that divide the wind energy facility area,
the Indian Ocean Coastal Belt and the Savannah Biome (Mucina and Rutheford, 2006) (Figure 1).
Zululand Coastal Thornveld is primarily characterised by landscapes with wooded bushes
dominated by Themeda triandra and bush, with Phoenix reclinata and Gymnosporia senegalensis. This
vegetation unit is classified as Endangered according to the National Biodiversity Assessment
(Mucina and Rutheford, 2006; Driver et al., 2012) (Figure 2). Maputaland Coastal Belt consists of a
coastal plain composed of pockets of various forest types (separated into different vegetation
units), thickets, bushes, extensive timber plantations and cane fields. The Conservation Status of
this unit is considered Vulnerable (Driver et al., 2012) (Figure 2).
14 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Despite its potential vegetation, the study area is characterized mainly by transformed habitats,
composed by extensive cultivated areas of sugar cane plantation3, with some patches of
eucalyptus plantations and native vegetation (Photograph 1). Remnant patches of natural Coastal
Belt Forest and Coastal Thornveld remain on site and these patches may constitute a relevant
biotope for some of the bat species occurring on site. As a humanized area, some houses and
other infrastructures are present, surrounded by some vegetation (usually clumps of exotic and
indigenous trees). The area is bisected by several water lines which are associated with riverine
vegetation. A few dams are also present. The wind energy facility area is not protected according
to any statutory conservation areas, since it is highly transformed, mostly by cultivation of sugar
cane.
Photograph 1 – Photographs of the sugarcane plantations (top left), eucalyptus plantation (top right),
riverine vegetation (bottom left), and native vegetation (bottom right) at the Richards Bay wind energy
facility site.
The local geology is mostly formed by Letaba Formation Basalts of the Karoo Super Group and
Quaternary sediments of marine origin. The soils are mainly black with a high clay content and
depth in the range 200-300 mm or yellowish and argillaceous redistributed sands. Soils are
nutritionally very poor and well leached, except in the interdune depressions where organic-rich
soils are sometimes found.
3 Sugarcane growth cycle during the monitoring period: Spring – terrain preparation and plantation; Summer – growing
season; Autumn – full growth attained; Winter – harvesting season.
15 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
According to Mucina and Ruderford (2006) the climate in the region where the project is
proposed is characterized mainly by high humidity levels and high average temperatures with very
infrequent frost. Rainfall occurs mostly in summer although some rain may occur during the
winter, especially in the Zululand Coastal Thornveld ecosystem, with maximum annual
precipitation between 800 and 1500 mm, generally higher towards the coastal areas. Mean
maximum and minimum monthly temperature for Lake St. Lucia Research Centre are 35.3°C
(January) and 5.5°C (June).
Figure 1 – Ecosystems present in the study area and its surroundings (within a 50km radius) (adapted from
Mucina and Rutherford, 2006).
16 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 2 – Conservation Status of the Ecosystems present in the study area and its surroundings (within a
50 km radius) (adapted from Mucina and Rutherford, 2006).
The Richards Bay wind energy facility is located approximately 10 km north of the Richards Bay
Game Reserve, 18 km east of the Ongoye Forest Reserve and 25km south of the Hluhluwe-
Umfolozi Park, all classified as Important Bird Areas (IBA) (refer to Figure 4).
Richards Bay Game Reserve was in the past an extensive papyrus swamp on the Mhlatuze River
floodplain that has been largely drained and the land converted for sugar cane cultivation. The
remnant natural vegetation in the Richards Bay area is fragmented and disturbed, and is
surrounded by industrial sites and roads. Vegetation alongside the Richards Bay–Empangeni road
consists of extensive stands of papyrus, with tall, dense, coarse grass and forbs at the edge.
Ongoye Forest Reserve is drained by the Umhlatuzana River and its tributaries to the north, and
the tributaries of the Umlalazi River to the south. The open wind-exposed ridges of the reserve
hold extensive patches of bushland. Rocky outcrops often have bush clumps. Some of the valleys
hold open woodland while stream bank woodland develops into hygrophilous forest with many
liana species that make the forest edge almost impenetrable.
17 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
The Hluhluwe–Umfolozi Park (HUP) has a subtropical climate with temperatures that range from
an average minimum of 13°C to an average maximum of 33°C. Most rain falls in summer
(October–March) and winters are mild and dry. Several rivers flow permanently within this area,
along with other seasonal streams and ephemeral rivers. The landscape is mostly composed of
bush land to parkland, with well-developed woodlands.
The current natural areas surrounding the study area are characterized by mild temperatures and
permanent rivers with well wooded forests which may represent important areas for bats,
concerning, for example, feeding and roosting sites. Although the study area is not well preserved
and none of these nature reserves’ protection status is related with bat conservation, it is
bisected by several streams with well-developed vegetation which connects small patches of
wooded habitats. These types of habitats may be potentially used by clutter feeder bats, which
forage close to vegetation. Additionally several houses and small infrastructures associated with
the sugar cane plantations occur spread throughout the area, providing potential roosting
structures. Considering these factors, the Richards Bay study area is considered to have potential
for utilization by several bat species.
Richards Bay wind energy facility site is also located in the broader vicinities of at least five
important caves, well known as roosting locations for bat communities, i.e.: Mission Rocks Caves
(87km to the north of the site), Hlatikula Forest Reserve (154km to the south of the site), Border
Cave (183km to the south of the site) Sibudu Cave (135km to the south of the site) and
Doornhoek Tunnel (~180km to the south of the site) (refer to Figure 3). Natal long-fingered bat
(Miniopterus natalensis) has already been observed at two of these caves, i.e. Mission Rocks Caves
and Hlatikula Forest Reserve (CES, 2012). This species is known to migrate several kilometres
between winter and summer roosts. Accordingly to CES (2012) personal observations have also
detected the presence of Geoffroy's horseshoe bat (Rhinolophus clivosus) at Sibudu Cave
Archaeological site, and of Sundevall’s leaf-nosed bat (Hipposideros caffer), Darling's horseshoe bat
(Rhinolophus darlingi) and Egyptian tomb bat (Taphozous perforatus) at Border Cave. Mission Rocks
is also known to provide shelter for thousands of individuals of Egyptian rousette (Rousettus
aegyptiacus) especially during winter, since the number of individuals present in these caves
decreases dramatically in spring (from approximately 5000 to 300 individuals), and increases at the
beginning of winter (Monadjem et al., 2010). A colony of 150 individuals of Bushveld horseshoe
bat (Rhinolophus simulator) is known in Doornhoek Tunnel (near Pietermaritzburg), where females
stay during winter and males stay throughout the year (Monadjem et al., 2010).
18 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 3 – Bat caves located in the vicinities of Richards Bay wind energy facility.
19 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 4 - Nature reserves surrounding Richards Bay wind energy facility site.
20 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
1.5. S UMMAR Y OF THE EIA
NSS conducted a preliminary Bat Impact Assessment in April 2012 at the request of Coastal and
Environmental Services (CES), the Environmental Assessment Practitioner (EAP) for the EIA
process being conducted for the WEF. The Environmental Impact Assessment report identified
the presence of ten bat species at the Richards Bay wind energy facility. According to the previous
likelihood of occurrence analysis, all species were expected to be present on site. One of those
identified species has conservation status of “Data Deficient” according with the South African
Red List, i.e. the Sundevall's leaf-nosed bat (Hipposideros caffer), however considered to have a low
risk of collision with wind turbines and fatality. Another two species were considered to have a
high collision risk – the Little free-tailed bat and Angolan free-tailed bat (Chaerephon pumilus and
Mops condylurus) – four were considered to have medium-high collision risk – Wahlberg's
epauletted fruitbat, Cape serotine, Banana bat and Yellow-bellied house bat (Epomophorus
wahlbergi, Neoromicia capensis, Neoromicia nana and Scotophilus dinganii) – and two were considered
to have medium collision risk– Schlieffen's twilight bat and Dusky pipistrelle (Nycticeinops
schlieffeni and Pipistrellus hesperidus) (CES, 2012).
Preliminary results from the EIA indicated higher bat activity near water bodies and bush areas,
and higher activity appeared to occur in the younger shorter sugar cane fields and wet bare soil,
than in the taller older fields. In the EIA study the whole site was considered sensitive due to a
great diversity of bat species. However, the northern section of the site falls within the
Endangered Zululand Coastal Thornveld ecosystem and appears to have a higher variable
topography, as well as a greater amount of natural vegetation patches and riparian vegetation.
This area was therefore considered to be of higher sensitivity than the southern section of the
site (Figure 5) (CES, 2012).
The probable fatality of foraging and migrating bats was identified as a very high significance impact
without mitigation and a moderate significance impact with mitigation. Nonetheless, the study
indicated that there are ways for these impacts to be mitigated in order to reduce the impacts:
e.g. turbine sitting, turbine dimensions adjustments and wind curtailment, the success of which
should be assessed through long-term monitoring (CES, 2012).
With these preliminary findings, the pre-construction monitoring methodology was designed to
cover 12 months. Several methodologies were used, i.e.: static monitoring at height and ground
level, manual surveys, roost inspection, roost surveys and mist-netting.
21 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 5 - Bat Sensitivity areas identified in the Richards Bay wind energy facility EIA report (from CES,
2012).
22 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
2. M O N I T O R I N G P R O G R A M M E D E S C R I P T I O N
The methodology used for the present monitoring programme was developed by bat specialists,
NSS, in order to comply with the South African Good Practice Guidelines for Surveying Bats in Wind
Farm Developments (Sowler and Stoffberg, 2012) and the main findings from the “Bat Impact
Assessment for the Richards Bay Wind Energy Facility” (CES, 2012).
2.1. D ESKTO P PRE PARAT ORY W O RK
Field surveys and other preparatory work was conducted by the NSS team.
The Bio3 team conducted a desk-top survey targeting a collection of all readily available
information to provide a better evaluation of all the conditions present in the study area.
Therefore, available data sources were consulted (Table 2) to assess which species could occur
and their likelihood of occurrence in the different habitats present within the proposed area for
the Richards Bay wind energy facility. In order to evaluate and interpret the field data,
bibliographic references and bat specialists (both local and international) were consulted
concerning all available information, such as: possible migration routes, patterns of bat activity
throughout the year in the study area, presence of known roosts surrounding the study area that
might be important for bats occurring at Richards Bay site, as well as, local or regional
echolocation variation in the sound parameters, or other type of information that could be
relevant for the contextualization of the importance of the study area for bats occurring in South
Africa, particularly, in the KwaZulu-Natal Province. This information was evaluated and validated
by a local bat specialist (i.e. Professor Corrie Schoeman from University of KwaZulu Natal) taking
into consideration the list of species with potential occurrence at the site and their ecological
requirements.
Potential roosting sites and potential important areas for bats were identified, in a preliminary
stage, by means of a desktop survey, taking into consideration the 1:50 000 maps of South Africa,
aerial imagery and any other relevant information overlaid in a GIS system.
Table 2 below includes, but is not limited to, the list of data sources and reports consulted and
taken into consideration for the compilation of this report, in varying levels of detail. Other
references were consulted for particular issues (these are detailed in section 6).
23 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 2 – Main data sources consulted. The international references and guidelines used to support the
methodological approach and result analysis are presented.
Typ
e
Name Reference Detail of information
Data
so
urc
es
Bats of Southern and Central Africa Monadjem et al. (2010)
National level
African Chiroptera Report 2012 African Bats (ACR, 2012) National level
Caves and Caving in the Cape http://www.darklife.co.za/Caves/ Regional level
Assessment of bat and avian mortality at a
pilot wind turbine at Coega, Port Elizabeth,
Eastern Cape, South Africa
Doty and Martin 2013 Regional level
The Vegetation of South Africa, Lesotho
and Swaziland Mucina and Rutherford 2006 National level
Global List of Threatened Species IUCN 2012 International level
Renewable Energy Application Mapping –
Report version I CSIR, 2013 National level
Gu
ideli
nes
an
d o
ther
inte
rn
ati
on
al re
fere
nces
Wind energy development and Natura
2000 European Commission 2010 International level
Directrices para la evaluación del impacto
de los parques eólicos en aves y
murciélagos
Atienza et al. 2011 International level
Comprehensive Guide to Studying Wind
Energy/Wildlife Interaction Stickland et al. 2011
International level
Methodological approach and analysis
U.S. Fish and Wildlife Service Land-Based
Wind Energy Guidelines USFWS 2012 International level
Hundt, 2012 Methodological approach Bat surveys: Good practice
guidelines, 2nd edition
South African Good Practice Guidelines for
Surveying Bats in Wind Farm
Developments
Sowler and Stoffberg 2012 Methodological approach
Good Practice Wind Project www.project-gpwind.eu/ International level
Species occurrence
The probability of occurrence of bat species in the study area was evaluated according with
several criteria, as described below. The distribution maps used to evaluate species occurrence
were the ones included in Monadjem et al. (2010) and ACR (2012). In this evaluation, species
which are known not to occur in the study area were not considered. The likelihood of
occurrence of bat species in the Richards Bay study area was characterised as:
24 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
High probability – the species has been historically confirmed on, or near the site
within the last 20 years; and the habitat present on site is suitable for the species
preferences.
Moderate probability – the species is within the higher probability modelled
distribution of potential occurrence according to Monadjem et al. (2010); and the species
has been historically confirmed in the area within the past 20-50 years; and/or the habitat
is adequate for the species requirements.
Low probability – the species is within the lower probability modelled distribution of
potential occurrence according to Monadjem et al. (2010); and the species has been
historically confirmed in the study area more than 50 years ago; and/or the habitat
present in the site is adequate for the species preferences.
The utilization of these two sources of information may cause some differences in the evaluation
on the probability of a species occurrence, since ACR (2012) presents a compilation of records of
the species and Monadjem et al. (2010) presents a modelled distribution of the species based on
several factors, such as previous records and habitat conditions. Nonetheless both types of
information were considered and evaluated according with the type of biotopes present at the
Richards Bay wind energy facility study area. The output of this evaluation was validated by
Professor Corrie Schoeman (University of KwaZulu Natal), a bat specialist with a broad
knowledge on South African bat species identification and distribution. Hence, species which are
known not to occur in the study area were not considered and the likelihood of occurrence was
adjusted according to this local specialist expertise and knowledge.
2.2. F IELD SURV EYS
Surveys of the bat community monitoring programme conducted by NSS included the
implementation of several field techniques adjusted to the specific characteristics of the study
area. Manual echolocation surveys were conducted during the surveys from May 2012 to April
2013, through vehicle transects; static echolocation surveys at ground level and at rotor height;
roost searches and inspections to any potential structure considered having any potential as a bat
roosting location were conducted, as well as live-trapping with mist-netting and harp-trapping.
2.2.1. S amp lin g P eri od
The bat communities pre-construction monitoring programme was implemented throughout a
one-year period (12 consecutive months) as recommended by the South African Good Practice
Guidelines for Surveying Bats in Wind Farm Developments (Sowler and Stoffberg, 2012). The pre-
construction monitoring programme was implemented by NSS from May 2012 until April 2013. In
this report the results from May 2012 to April 2013 surveys are presented and analysed, these
included all the surveys conducted, as described in Table 3, covering the following seasons: late
autumn, winter, spring, summer and early autumn. Considering that static detection (at ground
25 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
level and at height) was conducted almost continuously throughout one complete year, and that
the manual detection surveys were conducted three times every season, it is considered that the
sampling periods were adequate for the study area, exceeding the requirements of the South
African Good Practice Guidelines for Surveying Bats in Wind Farm Developments (Sowler and
Stoffberg, 2012).
Table 3 – Schedule of bat monitoring field work conducted by NSS at the Richards Bay wind energy
facility. The days of the month when field surveys were conducted are indicated in each cell. * Vehicle
transect surveys could not be conducted due to a fire in the study area.
Year Season Date
Passive
detection (Vehicle
transect surveys)
Static detection
Bat live trapping
Roost Inspection /
Monitoring
2012
Autumn 27th April to 31st May X
29th and 30th May X
Winter
1st to 31st June X
1st to 31st July X
18th and 19th July X
1st to 31st August * X
Spring
1st to 30th September X
13th September X X X
1st to 31st October X
24th October X X X
1st to 30th November X
29th and 30th November X
Summer
1st to 31st December X
1st, 2nd and 4th December X
2013
1st to 31st January X
19th to 22nd January X
1st to 28th February X
18th to 20th February X X X
Autumn
1st to 31st March X
25th and 26th March X
1st April to 1st May X
1st and 2nd May X X X
In order to evaluate the wind energy facility location, the Bio3 team also conducted field work in
February 2013 (Table 4). This visit to the Richards Bay wind energy facility site had the objective
of producing accurate vegetation cartography of the area proposed for the wind energy facility, as
26 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
well as inspecting some of the previously identified sites of interest for bats through desk-top
survey.
Table 4 – Schedule of bat monitoring field work conducted by Bio3 at the Richards Bay wind energy
facility.
Year Season Date Cartography Roost Inspection
2013 Summer 4th to 5th February X X
2.2.2. W eat her co ndi tio ns
The surveys were conducted under mild weather conditions, with the minimum averaged
temperatures recorded in the winter months, with approximately 19.1ºC (Figure 6). While wind
speed conditions were generally low, with average values approximate to 7 m/s, humidity was
considerably high throughout the year, with the minimum values recorded also during the winter
season.
(a)
0
5
10
15
20
25
30
0
2
4
6
8
10
May
june
July
August
Septe
mber
Octo
ber
Novem
ber
Decem
ber
January
Febru
ary
Marc
h
Apri
l
Autumn Winter Spring Summer Autumn
Tem
peratu
re (
ºC)
Win
d s
peed
(m
/s)
Average Wind speed (m/s) - above 60m Average Wind speed (m/s) - below 60m
Average Temperature (ºC) - above 60m
27 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
(b)
Figure 6 - Average weather conditions verified during the surveys conducted at Richards Bay wind energy
facility (data from anemometer masts installed in the site): average wind speed and average temperature per
month (a); average wind speed and average humidity per month (b).
2.2.3. E val uat ed Par ame ter s
To characterize the bat community present in the study area the following parameters where
evaluated for the Richards Bay wind energy facility:
Species Richness;
Activity Index;
Location and use of roosts within and around the site;
Type of utilization of the study area by bats.
2.2.4. D ata co lle cti on tec hn iqu es and me th ods ad opt ed by NSS
Bats are usually divided into two main groups: echolocating and non echolocating bats, the former
that usually uses high evolved ultrasound echolocation to navigate, forage and communicate
(Schnitzler and Kalko, 2001) and the latter that uses mostly the vision for orientation, to navigate
and search for food sources (Monadjem et al., 2010). Non echolocating bats are commonly known
as fruit bats (feeds mainly on fruits), whereas echolocating bats are known as insectivorous bats
(insects are their main food resource). The different flight and echolocation inter-specific
0
2
4
6
8
10
0
20
40
60
80
100M
ay
june
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Win
d s
peed
(m
/s)
Hu
mid
ity (
%)
Average humidity (%) Average Wind speed (m/s) - above 60m
Average Wind speed (m/s) - below 60m
28 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
characteristics are directly related to differences in species’ foraging habitats (Schnitzler and Kalko
2001).
Tracking the conservation status of insectivorous bat populations through the abundance and
distribution of echolocation calls has the potential to offer a more efficient alternative to trapping
or visual sampling methods for bat survey and monitoring programs (Walters et al., 2012). The
detection, recording and analysis of ultrasounds is very useful in the detection and identification of
different bat species, since these mammals are nocturnal and, in the majority of species, they emit
ultrasound calls to guide them, and to detect prey, as well as to communicate. Details pertaining
to the collection techniques adopted by NSS are described in Sections 2.2.4.1; 2.2.4.2; 2.2.4.3;
2.2.4.4; 2.2.4.5; and 2.2.4.6 below. All techniques adopted by NSS are in accordance with the
requirements of the South African Good Practice Guidelines for Surveying Bats in Wind Farm
Developments (Sowler and Stoffberg, 2012). Assumptions and limitations of the techniques
adopted by both NSS (data collection) and Bio3 (data analysis) are presented in chapter 2.3.1.
In the following sections, methodologies and field work conducted by NSS to the date of
compilation of the present report are presented.
2.2.4.1. Ec holo cati ng b at s peci es: Manual detection ( Vehicle
Transect)
The manual detection of ultrasounds was conducted through four driven transects, using a
Wildlife Acoustics® EchoMeter 3 (EM3) handheld ultrasonic bat detector sampling at 384kHz. This
equipment allows recording bat calls in compressed files (.WAC) that are later converted into
files suitable for analysis (.WAV or .ZC files) using Wildlife Acoustics Kaleidoscope® software.
The detector also recorded the transect route and the geographical position coordinates of each
triggered event (when a bat echolocation is detected). This information enables the subsequent
visualization of bat passes distribution throughout the transects.
Four transects were established, intended to be representative of the biotopes present at the
study area, which are mainly agriculture areas, small areas of native vegetation and patches of
eucalyptus plantations (Figure 40 – Appendix I). Each transect was characterised according to:
minimum distance to the future wind turbines, existing biotope, minimum distance to a water
source and minimum distance to known roosts.
The driven transects were conducted once a month, when possible. Starting at sunset, the vehicle
was driven at 10-20 km/h along a set route, whilst the microphone was set up on the top of the
car facing the opposite direction of the car movement. Each route was driven for approximately
1.5 hours. The direction of the route was alternated monthly to avoid bias on the peak bat
activity recording period only at one section of the route.
29 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
2.2.4.2. Echolocating bat s pecies: Stati c dete ction
Static detection was performed by means of a Wildlife Acoustics® Song Meter SM2BAT
Ultrasonic Recorder (http://www.wildlifeacoustics.com) installed at ground level (7-10m) and at
rotor height (30-80m) (Table 5). This system has known effective recording, stereo/ dual
channels, lower costs, weatherproof casings and omni-directional weather resistant microphones.
The detectors were powered by a 12V 7 Amp/hour battery and solar panel. The equipment was
scheduled to optimise the possibility of recording the most bat passes and the least interference
noise. As advanced settings the static detectors were configured with:
Microphone bias – Off;
Analog high-pass filter (HPF) – 1000Hz;
Gain – 36dB;
Sample rate – 384000Hz;
Compression – WAC 0;
Mic gain – 0;
Digital high-pass filter – fs/244;
Low-pass filter (LPF) – Off;
Trigger Level – 12 SNR;
Trigger window – 1second;
Division ratio – 16.
One of the six static detector locations (RB1) consisted of two microphones installed at different
heights and therefore the advanced configurations were adjusted to match this different scenario:
Microphone bias – Off;
Analog high-pass filter (HPF) - 1000Hz;
Gain – 36dB;
Sample rate – 192000Hz;
Compression – WAC 0;
Mic gain – 0;
Digital high-pass filter – fs/125;
Low-pass filter (LPF) – Off;
Trigger Level 12 SNR;
Trigger window 1second;
Division ratio – 16.
The detectors were set to automatically record bat calls every day, from sunset to sunrise. When
the recorder was in recording mode it triggers a recording every time an ultrasonic call is
detected. However due to power supply and data recording limitations, the detectors were
programmed to record at certain intervals of time: start at sunset, record for 2 hours after
sunset, pause for 30 min, record for 30min, repeat the 30min cycles until 2 hours before sunrise
when the detector recorded continuously for 2 hours.
During the monitoring surveys conducted to date, six bat detectors were set in the study area
(Figure 40 – Appendix I) at different time periods, as stated in Table 5. The difference in the time
periods of each detector collecting data in the study area was due to constraints in the
monitoring programme and to security concerns.
4 Filters ultrasound frequencies bellow 16 kHz. 5 Filters ultrasound frequencies bellow 16 kHz.
30 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 5 – Timeline of the static detectors used in the Richards Bay wind energy facility study area. Inside
cells is the height at which the detectors were placed.* Detector presented technical problems and data
was not collected. # Bio3 only receive recordings until 8 August 2012.
Month RB1 RB2 RB3 RB4 RB5 RB6
April/May 7m; 30m
June 7m; 30m
July 10m; 60m 10m 10m
August 10m; 60m 10m# 10m
September 10m; 60m 10m
October 10m; 60m
November 10m; 60m
December 10m; 60m 10m; 80m 10m 10m
January -* 10m; 80m 10m 10m
February 10m; 60m 10m; 80m 10m 10m
March 10m; 60m 10m; 80m 10m 10m
April 10m; 60m *;80m 10m 10m
Due to security concerns, on the 30th of April 2012, only one static monitoring detector (RB1)
was set on the Meteorological Mast, with two microphones - one at approximately 7m (RB1
bottom) and one at approximately 30m (RB1 top) (Photograph 2). The RB1 detector was
equipped with a stereo recording option, with a two-channel sample rate card. The use of two
channels allows the monitoring to take place at two different heights on the Met Masts with only
one detector, however it does not allow the identification of bats species echolocating at
frequencies higher than 96kHz. The microphones were set at this height due to strong winds at
the time of the installation. The microphones were later moved to higher heights, as provided in
the bat monitoring proposal, 10m and 60m (within rotor sweep height) respectively, in mid July
2012 where they stayed for the remaining monitoring period.
The single static station method was followed for the months of May, June and part of July 2012,
however, based on the “remarkable results” from the static detector, an additional two static
monitoring stations were set at Richards Bay (RB2 and RB3), in July 2012. However RB3 detector
was stolen in August 2012 and the solar panel and battery of the RB2 were stolen in September
2012. Therefore these stations were no longer valid and were replaced by the below described
stations.
In November 2012, RB4 was installed in a new 87m meteorological mast erected at Richards Bay,
with two independent SM2 bat detector in order to record on a one-channel 384kHz 16-bit
31 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
sample rate card to detect whether high frequency bats are flying at height. The detectors and
microphones were set at 10m and approximately 80m respectively on the new mast. This option
was selected due to the recording of high frequency bats at the RB2 and RB3 stations. Since
recording in stereo does not allow for the detection of species echolocating higher than 96kHz,
this option was the most adequate to record bat species that echolocate at high frequencies. This
station is located very close to the abandoned location of RB3 station.
Additionally, a third and fourth 10m static monitoring station were set at RB5 and RB6. These
stations were erected on Spiderbeam 10m telescopic masts with the mono 384kHz SM2BAT
detector option, including a one-channel 384kHz 16-bit sample rate card (Photograph 2). These
detectors were set up with only a left channel for a single ultrasonic microphone, but can record
bats at higher frequencies. The left channel ultrasonic microphone connected to these detectors
was erected at 10m on temporary aluminium masts. Strips of bristle brush and/ or Perspex spikes
were inserted on the microphone connector to reduce the risk of birds perching and damaging
the microphones. To reduce the risk of theft, power was supplied via internal D-cell batteries
which allowed about two weeks of operation autonomy. However this method of powering the
bat detectors has its limitations, since power may end before the team´s next visit, leading to gaps
in information collection.
Photograph 2 – Example of installation in the field of the static detectors on Spiderbeam 10m telescopic
masts (left) and on a meteorological mast (right).
Each static monitoring sampling point was characterised according to: minimum distance to the
proposed wind turbine locations, slope, dominant orientation, biotope (Table 6), minimum
32 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
distance to a water source and minimum distance to known roosts. The equipment automatically
recorded the temperature at each recording event.
Table 6 – Characterization of the percentage of each biotope present in a 500 m radius from the static
detector location.
Static detector
location
Biotope (%)
Native vegetation
Riverine vegetation
Sugar cane plantation
Urban areas
RB1 0 0 100 0
RB2 1.2 2.8 90.0 5.9
RB3 0 8.5 91.5 0
RB4 0 4.7 95.3 0
RB5 0 0 95.3 4.7
RB6 0.9 16.5 80.9 1.7
2.2.4.3. Non echolocating b ats
Fruit bats belong to the Pteropodidae family, and are characterized by the absence of evolved
echolocation, being dependent on eyesight for perception of the environment (Monadjem et al.,
2010). Only one South African fruit bat species, the Egyptian rousette (Rousettus egyptiacus), is
considered to have a rudimentary echolocation navigation systems, based on tongue clicks (Griffin
et al., 1958; Waters and Vollrath, 2003). The South African fruit bats feed on the fruits, flowers
and nectar of a wide range of indigenous trees, as well as on domestic or commercial fruit trees
(Monadjem et al., 2010). To determine the occurrence of fruit-eating bat species in the study area,
searches were directed to potential roosting sites suitable to these species during daytime. These
species occurrence was also accessed with the mist nest trapping methodology.
2.2.4.4. Roost searches, in spection and monito ring
Structures that could potentially provide roosting locations for bats (caves, mines, abandoned
buildings, bridges, large trees, banana leaves, water towers, culverts, rock crevices, etc.) were
identified in the study area and its surroundings, within an area of 30 km radius, by means of a GIS
based desktop study and during the fieldwork visits to the area. Discussions with local farmers
were also carried out attempting to identify important bat roosting locations in the study area.
The potential roosting locations identified were then inspected in the subsequent surveys through
the monitoring programme in order to record evidence of bats presence and occupation (such
live bats roosting, guano6 accumulation, bat corpses or insect remains). Additional information
was also recorded: season, the individual’s activity rate, presence of progeny, degree of human
disturbance and type of roost.
6 Name given to bat droppings.
33 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
During fieldwork, the location of each roost prospected was recorded with a handheld GPS, and
was photographed as well. The occupation rate, species present and conservation status were
determined to each roost inspected.
Roost surveys are particularly important for fruit-eating bats since acoustic monitoring techniques
do not detect the presence of these species.
2.2.4.5. Live-trapping
Based on the roost prospection and search, certain roost sites and foraging areas were selected
for two techniques of live-trapping bats: mist-netting and harp trapping (Figure 40 – Appendix I).
The captures performed have been done under legal permits (KZN Ezemvelo Wildlife permit). All
bats captured in live traps, were removed safely from the nets, put into black handling bags and
hung safely in a tree for later processing. At the end of the trapping session, bags were weighed
with and without the bat with a hanging scale to determine the mass of the bats, the forearm
length of the bats were measured and the bats were photographed. All bats were released at the
point of capture and the release call of each bat recorded with an EM3 detector on release.
2.2.4.6. Occasional observa tions
During field team technicians dislocations in the area all observations of bats were registered (e.g.
visual observations of individuals). This data may contribute to the confirmation of occurrence of
species difficult to capture by recordings, or the particular difficult species to capture on nets.
This information could be useful in the identification in loco of roosts or important feeding areas.
2.2.4.7. Habitat cartograph y
In order to properly assess the relationship between bat activity and the area conditions, a
mapping of vegetation and biotopes present was specifically produced by the Bio3/Savannah team
based on military maps, satellite images and one field visit to the wind energy facility site. The
cartography included all farms portions that comprise the wind energy facility site. During the
habitat cartography the previous assessment of the habitats that could compose the study area
through desk-top survey was confirmed and adjusted in order to better represent the reality of
the study area.
Cartography was performed by two observers that travelled by car through the whole the wind
energy facility area, marked on military maps and through geographic information systems
identified the correct habitats present in the study area. This marking was made with the
simultaneous help of a GPS connected to a computer with Geographic Information System
software in order to avoid errors of terrain interpretation.
34 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
2.2.5. D ata an aly sis an d c ri ter ia
2.2.5.1. Ultra-sounds analy sis
Acoustic monitoring produces huge amount of data, therefore the call data was recorded by the
SM2BAT as a compressed format (*.WAC files) that was later converted using Wildlife Acoustics
Kaleidoscope® Software to *.WAV files to allow species acoustic identification by expert
technicians. Using the same software, it was conducted an acoustic scrubbing for filtering non-
biological noise such as rain, wind, birds and insects, false triggers or anthropogenic noise. With
this operation it was intended to eliminate periods of rain or wind, long periods of noise with low
frequencies, within the audible frequencies. It is however necessary to consider that the software
is not perfect and that biological noise is highly variable. Therefore, whenever considered
necessary, a manual scrubbing was performed using a software developed specifically to address
this issue (by the IEETA – Institute of Engineering and Telematics of Aveiro University in
Portugal). This software allowed an expedited visualisation of the recordings and was used as a
complementary scrubbing to Kaleidoscope® tool software, assuring that all activity recorded was
considered.
Identification of bat species through analysis of echolocation calls is a very time consuming task7,
as specialized technicians have to go through each call, extract the necessary acoustic parameters
with specific software and then identify the species using a reference echolocation call library for
South African bats. Considering the amount of data produced it was necessary to conduct a sub-
sampling methodology of the overall calls recorded by the static detectors. This sub-sampling
methodology was intended to estimate the proportion of bats that belong to a certain species,
among the total bat calls recorded. Since the surveys were conducted throughout time and in
several different locations, a simple random sampling would not be suitable. Therefore, the
adequate method applied was a stratified random sampling (Cochran, 1977), using as factors the
sampling location and survey.
Since sampling the total number of recordings was made with the purpose of
analysing the population species composition in the study area, sub-sampling results were
only used when analysing the species present at the site. In all remaining analysis the total
amount of data collected by the detectors was considered (both from manual and static
surveys).
The total size of the sample was calculated according to the following equation (Cochran, 1977):
(
)
7 We estimate that one specialized technician can identify, on average, 30 echolocation recordings during a working day
(8 hours).
35 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Where: n = number of elements of the sample; P = estimated proportion of the interest characteristic (bat species);
= critical value associated to the degree of confidence; e = maximum error of estimation.
The number of elements of the sample for each of the considered factors was obtained through
proportional affectation, using the equation (Cochran, 1977):
(
)
Where: ni = number of elements of the sample in the factor; Ni = number of elements in the factor; n = number of
elements of the sample; N = number of elements of the population.
With the number of elements to analyse in each of the factors (location and survey), resulting
from this process of stratified random sampling, the recordings for analysis were randomly
selected through a random algorithm. The randomly selected recordings were then processed by
a specialized technician, considering the several parameters that allow the identification of bat
species. One of the characteristics of echolocation pulses that have to be considered for the
identification of bat species is the shape of echolocation pulses - frequency modulation (FM),
quasi-constant frequency (QCF) and constant frequency (CF) (Altringham, 1996; Russo and Jones,
2002). However most of the bats use a combination of both FM/QCF (Altringham, 1996), where
the initial part of the pulse uses frequency modulation, and the end of the presents almost a
constant pulse frequency. Further characteristics of the pulses are used for the species
identification such as the frequency of maximum energy (FMaxE), pulse duration, initial and final
frequencies, bandwidth, interval between pulses, shape of the pulse, among others (Fenton and
Bell, 1981).
The analysis of the recorded calls was performed using Audacity 2.0.0 – Cross-Platform Digital Audio
Editor, from Dominic Mazzoni. Through the analysis of pulse characteristics, the identification of
detected species was possible. The reference values used were the ones presented in ACR
(2012), Pierce (2012), Gauteng and Northern Regions Bat Interest Group (2012), Monadjem et al.
(2010), Hauge (2010), Kopsinis (2009) and Taylor et al. (2005). This acoustic echolocation
parameters reference table was reviewed and adjusted where necessary by professor Corrie
Schoeman in order to use the most accurate reference parameters as possible, considering the
limitations of the current knowledge on South African bats echolocation (refer to section 2.3.1).
To effectively use echolocation as a means of surveying bats, it is important that we can reliably
identify the species detected. Even with their similar sensory aims, many bat species have evolved
a species-specific echolocation call structure (O’ Farrell et al., 1999; Simmons et al., 1979)
providing the potential to use their echolocation calls to identify bats to species level (O’ Farrell
1997; O’ Farrell et al., 1999; Sattler et al., 2007). However, these call structures are extremely
flexible and may depend on various factors including habitat structure, foraging strategy, age,
gender, morphology, and the presence of other conspecifics (Bell and Fenton, 1987). As different
species face similar sensory challenges, call convergence has led to overlap in frequencies and call
shapes used, by some species making it difficult to distinguishing between some calls (Preatoni et
al., 2005).
36 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
In spite of this kind of problems faced through bioacoustics, on some recordings the identification
was only possible to the level of genus, family or to some phonic groups with very similar acoustic
identification parameters. If the species was identified through recording analysis and its
occurrence in the study area is considered plausible, then it was classified as Confirmed in the
study area. If a species could not be confirmed through recordings analysis, due to uncertainty
with the call parameters obtained, and could only be identified as a group of species, its
occurrence in the study area was considered as Possible (e.g. if the parameters obtained in a
recording are coincident with call parameters from different species and none of them was
confirmed in other recordings, then all these species are considered possible, if the habitat is
suitable). When the pulses recorded were too weak, and no diagnose parameters could be
obtained, the identification was only up to the level where the specialists felt to have a high
degree of confidence they were not making any inaccurate identification (family, gender, family
group or species group).
Through call analysis it was also possible to identify the occurrence of different bat behaviours
according to different types of pulses, such as echolocation pulses (searching phase and feeding
buzz8) or social calls.
2.2.5.2. Spatial-temporal a nalysis
The results from the surveys undertaken (between May 2012 and April 2013) were analysed
separately and compared. The information was collected through the automatic scrubbing
performed by the Kaleidoscope® Software, as well as the manual scrubbing, as described in the
previous section (2.2.5.1). For the study area, and for each sampling location, the species
identified were listed together with their conservation status and distinctive behaviour.
Space and time use of the site was also studied. The number of bat passes at each sampling point
allowed the determination of Activity Indexes (number of bat passes9 / unit of time) for manual
and static detection:
8 Feeding buzz: when a bat identifies a potential prey it starts to approach the insect prey. In this process it will increase
the rate of its echolocation pulses and each pulse will become shorter until it is difficult to distinguish between different
pulses. This method of increasing its echolocation resolution while homing in on its prey is referred to as a feeding
buzz.
9 For the calculation of the above parameters it was necessary to define a “bat pass”. There is a standard widely used
definition of bat pass: two call notes from one bat not separated by more than 1 second (White and Gehrt, 2001; Gannon et
al., 2003). However, this is not very consensual since the duration and frequency of call notes vary according with the
species present. In South Africa, and considering the species present, the current possible definition of bat pass is that
of a sequence of ≥1 echolocation calls where the duration of each pulse is ≥2 ms (Weller and Baldwin, 2011).
Single call fragments do not apply, only complete pulses were considered for the analysis. Where there is a gap between
pulses of >500ms in one file, this then represents a new bat pass.
37 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Average number of bat passes/sampling point (static detection) or grid cell (manual
detection);
Average number of bat passes/hour;
The calculation of the activity index, has defined by Miller (2001), is performed by counting the
number of periods of time where a certain species was recorded. This method could be applied in
areas of high species diversity, where files contain calls from more than one species. Considering
that in Richards Bay the analysis of the ultra-sounds revealed that this was not the case, a simpler
approach was considered, by calculating the number of bat passes per hour10, as the
activity index, for each of the sampling points.
Notice however that the activity index does not provide an absolute number of individuals,
indicating solely a relative index of abundance (Hayes, 2000; Kunz et al., 2007). An analysis of the
activity index for the recording time period was also performed in order to evaluate the variation
of activity throughout time, and which periods have higher bat activity.
Considering the information recorded from the manual transects a mapping of the study area
utilization was also performed, considering the total number of passes and the number of
confirmed species recorded by each grid cell, within a previously created grid of 500x500 m.
These parameters were also analysed in terms of environmental factors, such as environmental
conditions (air temperature, humidity and wind speed), biotope, and illuminated lunar fraction.
The same parameters were also analysed on a spatial basis, according to each sampling point
location. For comparison and statistical analysis purposes the characterization of the biotope
where each static detector was installed was determined by the percentage of each biotope (e.g.
sugar cane plantation, riverine vegetation, native vegetation, urbanized area, natural vegetation)
within a buffer of 500m centred on the proposed turbine location, rather than considering just
the biotope on the exact location of the detectors. From our point of view, this will allow a more
accurate analysis providing a more realistic representation of bat utilization of the area.
A Generalized Linear Mixed Model (GLMM) was carried out to test possible relations between
bat activity in the study area (number of passes) and the environmental conditions recorded while
conducting the fieldwork. Two separate analyses were conducted based on the information from
the static detection surveys (static detectors) and the manual detection surveys (car transects).
For static detection analysis, a total of 10 predictor variables (fixed effects) were initially
considered: air temperature, wind speed, lunar fraction, air humidity, biotope (including sugar
cane plantation, native vegetation, urbanized areas, riverine vegetation), height of the detector and
season of the year. Sample site “point” and “hour” were included in the model as a random effect
(Bolker et al., 2008).
10 Number of bat passes per hour determined as the average number of passes recorded during the number of hours of each sampling
night.
38 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
For the manual detection (car transects), 10 predictor variables (fixed effects) were initially
considered in the analysis: air temperature, wind speed, lunar fraction, air humidity and biotope
(including sugar cane plantation, native vegetation, urbanized areas, riverine vegetation, ploughed
fields and eucalyptus plantation). Sample site “grid cell” was included in the model as a random
effect (Bolker et al., 2008). The significance of each covariable initially assessed using univariate
models, and the final model included covariates whose p-value for the Wald test was less was than
0.2 (Hosmer and Lemeshow, 2000). Also, to measure collinearity between explanatory variables
the Spearman's rank correlation coefficient p was calculated, to ensure that the covariables were
not highly correlated. The selection of the final model was based on the Akaike's Information
Criterion (AIC) through a backward stepwise procedure. Finally, adjustment measures (goodness-
of-fit) were estimated and the residuals for the final model were analysed. All calculations were
performed with R software (R Development Core Team, 2012).
2.3. I MPACT S EVA LUATI ON
2.3.1. T urb ine Se nsi tiv ity A nal ysi s
The location of each turbine was assessed in relation to the environmental features considered
more relevant for the bat community, accordingly to the results from the pre-construction bat
monitoring programme. With this systematic approach turbines were groups accordingly to their
most likely potential impacts on bats. Seven different criteria were used in this classification (Table
7).
For this analysis it was considered the presence of sensitive features within a 500 m buffer from
each of the proposed wind turbine location. Considering the type of sensitive features identified
in the surrounding area of each turbine, the turbines were then grouped in three major groups:
Roosting related sensitive turbines – turbines that were located within buffer areas from
a roost or a reproduction roost were grouped in this category. The presence of a high
activity index may also indicate the proximity of a roost, as bats tend to more active in
the proximities of their roosting locations;
Habitat related sensitive turbines – turbines which were located within bat important
areas buffer, buffer areas of native vegetation, dams and riverine vegetation were
grouped in this same category. The presence of a high activity index near these locations
may also be a consequence of the importance that these features have for bats, as they
may spend larger amounts of time foraging at these locations;
Roosting and Habitat sensitive turbines – were considered the turbines that may affect
both roost related features and habitat related features.
When turbines did not include the presence of any sensitive features within the 500 m buffer
defined, they were considered as “Non sensitive”.
39 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 7 – Criteria used for the turbine risk assessment. P – present.
No. Criteria Description
Ro
ost
rela
ted
featu
res
Hab
itat
rela
ted
featu
res
Gen
era
l
sen
siti
ve
featu
res
1 Roosts Presence of roosts within 500m buffer from wind
turbine P
P
2 Buffer around roosts Presence of roosts buffer area within 500m buffer
from wind turbine P
P
3 Buffer around
reproduction roosts
Presence of reproduction roosts protection buffer
area within 500m buffer from wind turbine P
P
4 Buffer around bat
features
Presence of bat important features buffer area (as
defined in section 4) within 500m buffer from wind
turbine
P P
5 Riverine vegetation
and dams
Presence of dams and riverine vegetation within
500m buffer from wind turbine P P
6 Native vegetation Presence of native vegetation within 500m buffer
from wind turbine P P
7 High Activity areas
Presence of high activity areas (more than 30 bat
passes/hour) identified during the bat monitoring
programme within 500m buffer from wind turbine
P P P
2.3.2. I mpa cts Ev alu ati on
The direct, indirect and cumulative impacts on bats resulting from the Richards Bay Wind Energy
Facility and associated structures identified through the pre-construction monitoring programme
results were quantified and classified, for the different development phases of the project:
construction, operation and decommission.
Five factors need to be considered when assessing the significance of impacts, namely:
1. Relationship of the impact to temporal scales - the temporal scale defines the significance of
the impact at various time scales, as an indication of the duration of the impact.
2. Relationship of the impact to spatial scales - the spatial scale defines the physical extent of
the impact.
3. The severity of the impact- the severity/beneficial scale is used in order to scientifically
evaluate how severe negative impacts would be, or how beneficial positive impacts would
be on a particular affected system (for ecological impacts) or a particular affected party.
The severity of impacts can be evaluated with and without mitigation in order to
demonstrate how serious the impact is when nothing is done about it. The word
‘mitigation’ means not just ‘compensation’, but also the ideas of containment and remedy .
For beneficial impacts, optimization means anything that can enhance the benefits.
However, mitigation or optimization must be practical, technically feasible and
economically viable.
40 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
4. The likelihood of the impact occurring - the likelihood of impacts taking place as a result of
project actions differs between potential impacts. There is no doubt that some impacts
would occur (e.g. loss of vegetation), but other impacts are not as likely to occur (e.g. vehicle
accident), and may or may not result from the proposed development. Although some
impacts may have a severe effect, the likelihood of them occurring may affect their overall
significance.
Each criterion is ranked with scores assigned as presented in Table 8 to determine the overall
significance of an activity. The criterion is then considered in two categories, e.g. effect of the
activity and the likelihood of the impact. The total scores recorded for the effect and likelihood
are then read from the matrix presented in Table 9, to determine the overall significance of the
impact (Table 10). The overall significance is either negative or positive. The environmental
significance scale is an attempt to evaluate the importance of a particular impact.
Negative impacts that are ranked as being of “Very High” and “High” significance will be
investigated further to determine how the impact can be minimised or what alternative activities
or mitigation measures can be implemented. These impacts may also assist decision makers i.e.
lots of High negative impacts may bring about a negative decision.
For impacts identified as having a negative impact of “Moderate” significance, it is standard
practice to investigate alternate activities and/or mitigation measures. The most effective and
practical mitigations measures will then be proposed.
For impacts ranked as “Low” significance, no investigations or alternatives will be considered.
Possible management measures will be investigated to ensure that the impacts remain of low
significance.
The significance scale is an attempt to evaluate the importance of a particular impact. This evaluation
needs to be undertaken in the relevant context, as an impact can either be ecological or social, or
both. The evaluation of the significance of an impact relies heavily on the values of the person making
the judgment. For this reason, impacts of a social nature need to reflect the values of the affected
society.
Cumulative Impacts
Cumulative Impacts affect the significance ranking of an impact because it considers the impact in
terms of both on-site and off-site sources. For example, pollution making its way into a river
from a development may be within acceptable national standards. Activities in the surrounding
area may also create pollution which does not exceed these standards. However, if both on-site
and off-site activities take place simultaneously, the total pollution level at may exceed the
standards. For this reason it is important to consider impacts in terms of their cumulative nature.
Seasonality
Although seasonality is not considered in the ranking of the significance, if may influence the
evaluation during various times of year. As seasonality will only influence certain impacts, it will
41 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
only be considered for these, with management measures being imposed accordingly (i.e. dust
suppression measures being implemented during the dry season).
Ranking of Evaluation Criteria
The assessment of each impact was undertaken according to the parameters in Table 8, Table 9
and Table 10.
Table 8 – Parameters considered in the classification of impacts on bats.
11 In certain cases it may not be possible to determine the severity of an impact thus it may be determined: Don’t know/Can’t know.
EF
FE
CT
Temporal scale Score
Short term Less than 5 years 1
Medium term Between 5 and 20 years 2
Long term Between 20 and 40 years (a generation) and from a human
perspective almost permanent. 3
Permanent Over 40 years and resulting in a permanent and lasting change that
will always be there 4
Spatial Scale
Localised At localised scale and a few hectares in extent 1
Study area The proposed site and its immediate environs 2
Regional District and Provincial level 3
National Country 3
International Internationally 4
Severity11 Benefit
Slight / Slightly
Beneficial
Slight impacts on the affected
system(s) or party(ies)
Slightly beneficial to the affected
system(s) or party(ies) 1
Moderate / Moderately
Beneficial
Moderate impacts on the
affected system(s) or party(ies)
An impact of real benefit to the
affected system(s) or party(ies) 2
Severe / Beneficial Severe impacts on the affected
system(s) or party(ies)
A substantial benefit to the
affected system(s) or party(ies) 4
Very Severe / Very
Beneficial
Very severe change to the
affected system(s) or party(ies)
A very substantial benefit to the
affected system(s) or party(ies) 8
LIK
EL
IHO
OD
Likelihood
Unlikely The likelihood of these impacts occurring is slight 1
May Occur The likelihood of these impacts occurring is possible 2
Probable The likelihood of these impacts occurring is probable 3
Definite The likelihood is that this impact will definitely occur 4
42 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 9 - The matrix that will be used for the impacts and their likelihood of occurrence.
Lik
eli
ho
od
Effect
3 4 5 6 7 8 9 10 11 12 13 14 15 16
1 4 5 6 7 8 9 10 11 12 13 14 15 16 17
2 5 6 7 8 9 10 11 12 13 14 15 16 17 18
3 6 7 8 9 10 11 12 13 14 15 16 17 18 19
4 7 8 9 10 11 12 13 14 15 16 17 18 19 20
43 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 10 - Ranking matrix to provide an Environmental Significance.
Environmental Significance Positive Negative
LOW
An acceptable impact for which mitigation is desirable but not
essential. The impact by itself is insufficient even in combination with
other low impacts to prevent development.
These impacts will result in either positive or negative medium to
short term effects on the social and/or natural environment.
4 - 7 4 – 7
MODERATE
An important impact which requires mitigation. The impact is
insufficient by itself to prevent the implementation of the project but
which, in conjunction with other impacts may prevent its
implementation.
These impacts will usually result in either positive or negative
medium to long term effect on the social and/or natural
environment.
8 – 11 8 – 11
HIGH
A serious impact which, if not mitigated, may prevent the
implementation of the project.
These impacts would be considered by society as constituting a
major and usually long term change to the natural and/or social
environment and result in severe negative or beneficial effects.
12 – 15 12 – 15
VERY HIGH
A very serious impact which may be sufficient by itself to prevent
the implementation of the project.
The impact may result in permanent change. Very often these
impacts are immitigable and usually result in very severe effects or
very beneficial effects.
16 - 20 16 - 20
2.4. A SSUMP TIONS AND LIMI TATION S
The field work was conducted by a different team (NSS) which implemented all the described
methodologies. For conducting the present report and analysis it was assumed that all the field
data was made available for Bio3 analysis and field methodologies were correctly implemented by
NSS. Therefore, the results presented in this report are limited to the information sent and made
available by the team responsible for the data collection. The Bio3 team made all efforts to clearly
understand the data organization, field techniques implemented and its implications for the
subsequent analysis. However, it is assumed that some misinterpretations could have occurred as
some specific information related to several factors (e.g. field personal observations, details
pertaining the organization of data, field experience perception or even organization and potential
data errors that could have been corrected after the data was made available to our team) could
have influenced data analysis, interpretation and presentation. A summary of the received
information where the present report was based is summarized in Appendix IV.
It is important to note that NSS team found some limitations while conducting the field work,
which may influence the data analysis:
Field surveys conducted in September and October 2012 were disturbed by extreme
rainfall and slippery roads, making transects surveys difficult to perform;
44 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
August survey was disrupted by the burning of sugar cane which is a common practice by
farmers. Despite being common and presumably controlled, this practice was considered
a potential threat to the team’s safety and the equipment integrity;
Field surveys were constrained by security issues. The field teams faced several difficulties
while trying to secure the detectors and their power supply, leading to the placement of
fewer static detectors in the study area than the optimal number. This difficulty led to the
theft of several pieces of equipment, and subsequently of data losses, placed in
Spiderbeam 10m telescopic mats that could easily be dismantled. These obstacles were
overcome by working closely with farmers and farm watch security by selecting the most
secure locations.
Some of the conclusions of the present study are constrained by the lack of baseline information
concerning bat ecology and distribution in South Africa, such as: detailed information of species
migration and dispersal, known roosts, existing populations, bioacoustics synthesized information,
among other factors. This lack of published and peer-reviewed information also leads to
difficulties in the ultrasound acoustic identification of less studied species, since very few or no
reliable references may exist for the study area or its surroundings. Many published papers refer
to acoustic parameters that relate to bats that were first captured and recordings were then
made through hand release or in other special circumstances (e.g. trapped in cages, in habitats
different the ones present in the studied area) that can misrepresent the reference sound
parameters collected. This type of information may be very important when analysing recordings
collected in the field because differences from the natural echolocation parameters of the species
may occur in relation to the reference echolocation parameters, leading to incorrect or uncertain
identifications. In order to overcome these echolocation problems, a southern African bat
specialist was contacted to standardize all data collected from scientific references.
Despite the lack of information published concerning bats in South Africa, it is important to
consider and review the information already available. This is very important for data accuracy
since some publications refer to some uncertainty associated with the data, due to records that
were not finally confirmed or not collected by proper bat specialists or species which may have
been incorrectly identified, as it was confirmed during the compilation of Monadjem et al. (2010).
Therefore, an effort was made in this report to engage with Professor Corrie Schoeman, a local
bat specialist, in order to confirm and verify the information collected from several bibliographic
sources, and therefore present the most accurate results regarding the potential species
occurring at the site.
The large number of recordings produced through static recordings lead to a great effort in
desktop analysis hours (e.g. audio files conversion, scrubbing, compilation and call identification).
Due to this constraint, a sampling procedure had to be implemented for the recordings that were
analysed for species identification as it was not viable to identify every single recording collected
in the field. Therefore, and in order to make the best possible evaluation, it was assumed that all
recordings analysed represented a good indication of the bat population in study, with a minimal
margin of error of approximately 1.8%. More details on this subject are explained in section 2.2.5.
45 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
It was also noticed that some of the recordings collected through static detection had a poor
quality (blurry recording, noise masking the bat pulses, only segments of bat pulses recorded) not
allowing the identification to the species level with certainty in some of the those recordings. This
limitation may influence the results presented in this report, leading to the identification of fewer
species than those that really occur in the study area. Nonetheless, it was assumed that the best
approach would be to consider the families identified with certainty, as a good measurement of
the possible species present, rather than present results that do not have a good degree of
certainty associated. During the conducted call identification, the level of detail of each
identification was determined to the level possible with certainty, so in some cases it was possible
to identify groups of species, in some another cases up to the genus level.
Another limitation is related with the survey techniques implemented and the limitations
intrinsically linked to the field methodologies. Species that echolocate at high frequencies are
more difficult to capture through ultrasound detection, since ultrasounds with very high
frequencies do not travel very long distances (compared to low frequency ultrasounds) (Limpens
and McrAcken, 2002). Therefore these species would have to be very close to the ultrasound
detector to be captured.
The experimental design of the current monitoring programme did not consider the assessment
of bat activity and diversity in a similar control area. The existence of a control area is mostly
used in a BACI (Before, After, Control and Impact) approach of Impact Assessment, where the
data from two sites is used with two objectives: a primary objective of assessing an impact before
and after the impact begins; and a secondary objective of assessing a control and impact area
which will allow determining if the effects verified during the subsequent phases of the project
(construction and operation) are related with the development itself or with natural stochasticity
of environmental variables affecting bat populations (Smith et al., 1993). Despite the second
objective it is very important to accurately validate the identified potential impacts (and identify
any additional impacts), and it is not considered essential to determine the baseline scenario to
establish the benchmark for the local bat communities. It is considered that the current
methodological approach is suitable to provide baseline information regarding bat communities on
the wind energy facility site. However the absence of a control area to compare the impact area
will not allow an accurate and expedited assessment of effects during operation such as
displacement and disturbance, if changes before and after the impact occur in the study area. It is
therefore recommended that a control area(s) be established as soon as possible.
46 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3. R E S U L T S A N D D I S C U S S I O N
3.1. S PECIE S PRE SENT AT THE SITE
During bat monitoring surveys conducted at the Richards Bay wind energy facility site between
May 2012 and April 2013, a total of approximately 373 000 recordings (from manual and static
detection) were obtained. For the identification of bat species occurring at the area and bat
communities’ characterization, a sub-sampling of the total recordings was conducted.
Consequently, a total of 5050 recordings from static detection and 801 recordings from manual
detection were analysed (with 95% of confidence and an approximate error of 1.8% in the
estimates). The results from the recordings analysis allowed the identification of the family or
species of the individuals in 98.7% of the recordings analysed. The remaining records had weak
pulses or very low volume and in those cases the identification of the individuals to the family
level was not possible, being classified as Unidentified.
3.1.1. D esk top re vie w
3.1.1.1. Species with poten tial occurren ce at the site
According to Monadjem et al. (2010), a total of about 67 species of bats may occur in South
Africa. Through the analysis of species probability of occurrence at the study area, it was
concluded that 35 bat species may possibly occur in the vicinity of site, which corresponds to 52%
of overall species in the country (Table 11). This high percentage of species is justified by the
favourable climate of the area, presence of important biotopes for bats, such as riparian
vegetation and wooded areas, high number of water sources and the presence of potential
roosting sites.
From the 35 bat species considered to have potential occurrence in the area, 23 are considered
to have a high probability of occurrence, 6 with medium and the remaining 6 with low probability
of occurrence.
Regarding the 23 species considered with high probability of occurrence at the site, only 1 is
considered endangered in South Africa (Swinny’s horseshoebat – Rhinolophus swinnyi) but such
species is considered to have a potential low collision risk with wind turbines (Sowler and
Stoffberg, 2012). From these 23 species, 4 species are perceived as a potential high risk of
collision with wind turbines – i.e. Chaerephon pumilus, Mops condylurus, Tadarida aegyptiaca and
Taphozous mauritianus – all with populations considered as being Least Concern in South Africa
(Table 11). 12 of the species with high probability of occurrence at the site have a medium /
medium-high potential collision risk with wind turbines, most considered of Least Concern and
only 3 considered as Nearly Threatened in South Africa – Miniopterus fraterculus, Miniopterus
natalensis and Myotis tricolor.
47 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
From the total list of 35 species with potential occurrence in the area, 4 are of conservation
concern in South Africa, namely: Cloeotis percivali, Kerivoula argentata, Rhinolophus blasii and
Rhinolophus swinnyi. With the exception of R. Swinnyi, the remaining species are considered to have
a low potential of occurrence in the site. A total of 11 species out of those 35 are considered to
have a South African population classified as Nearly Threatened, 6 with low potential collision risk
and 5 with medium-high potential collision risk (Table 11).
A brief description of the species with potential to occur in the study area is presented in
Appendix V.
48 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 11 – List of species with possible occurrence at the Richards Bay wind energy facility study area. Legend (IUCN, 2012 and South Africa Red List): CR – Critically
Endangered; EN – Endangered; VU – Vulnerable; NT – Near Threatened; LC – Least Concerned; NE – Not Evaluated; Data Deficient; Flight height: LH – Low Height
(generally below 2 meters); MH – Medium Height (generally between 2 and 10 meters); HH – High Height (generally above 10 meters); n.a. – information not available.
Family Species Common name
IUC
N
So
uth
Afr
ica R
ed
Lis
t
Roosts Habitat preferences Foraging habits
Type of flight
Foraging habits
Flight height
Risk of collision
(Sowler and
Stoffberg,
2012)12
Probability
of
occurrence
HIPPOSIDERIDAE
Cloeotis percivali Percival's short-
heared trident bat LC CR
Caves, possible in
narrow crevices Woodland. Clutter forager n.a. Low Moderate
Hipposideros caffer Sundevall's leaf-nosed
bat LC DD
Caves, sinkholes and
cavities
Savannah woodland,
riparian vegetation;
avoid open areas.
Clutter forager n.a. Low High
NYCTERIDAE
Nycteris hispida Hairy silt-faced bat LC NT
Dense bush, also
houses, hollow trees
and caves
Wide variety of
vegetation, avoids arid
zones.
Clutter forager LH Low Low
Nycteris thebaica Egyptian silt-faced bat LC LC
Caves, burrows,
culverts and trunks
of large trees.
Savannah and Karoo
biomes; avoids open
grassland.
Clutter forager LH Low High
PTEROPODIDAE
Eidolon helvum African straw-
coloured fruit bat NT NE Trees Forest n.a. n.a. Medium-High High
Epomophorus crypturus Peter's epauletted
fruit bat LC DD
Dense foliage of tall
trees
Forest; wooded
gardens in urban areas n.a. n.a. Medium-High Low
Epomophorus wahlbergi Wahlberg's
epauletted fruit bat LC LC
Dense foliage of tall
trees Forest n.a. n.a. Medium-High High
Rousettus aegyptiacus Egyptian rousette LC LC Caves (gregarious) Forest Clutter forager MH, HH Medium-High High
MINIOPTERIDAE
Miniopterus fraterculus Lesser long-fingered
bat LC NT Caves Mountain areas.
Clutter-edge
forager MH, HH Medium-High High
Miniopterus natalensis Natal long-fingered
bat LC NT Caves
Savannas and bushlands;
montane grassland.
Clutter-edge
forager MH, HH Medium-High High
VESPERTILIONIDAE Eptesicus hottentotus Long-tailed serotine LC LC Caves, rock crevices Woodland; rocky
outcrops, miombo
Clutter-edge
forager MH, HH Medium High
12 The detailed criteria of potential collision risk from Sowler and Sttofberg (2012) are presented in Appendix VII.
49 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Family Species Common name
IUC
N
So
uth
Afr
ica R
ed
Lis
t
Roosts Habitat preferences Foraging habits
Type of flight
Foraging habits
Flight height
Risk of collision
(Sowler and
Stoffberg,
2012)12
Probability
of
occurrence
woodland and granitic
hills.
VESPERTILIONIDAE
Glauconycteris variegata Variegated butterfly
bat LC NT Dense foliage
Savannah, open
woodland, riparian
forest.
Clutter-edge
forager n.a. Medium Moderate
Kerivoula argentata Damara woolly bat LC EN Foliage, bird nests,
also buildings
Woodland and riparian
vegetation.
Clutter-edge
forager n.a. Low Moderate
Kerivoula lanosa Lesser woolly bat LC NT Bird’s nests, specially
weavers Riparian vegetation. Clutter forager n.a. Low Moderate
Myotis tricolor Temminck's myotis LC NT Caves Mountains. Clutter-edge
forager MH, HH Medium-High High
Myotis welwitschii Welwitsch's myotis LC NT Foliage (bananas);
caves
Mountains; woodland-
fores; mosaic
vegetation
Clutter-edge
forager MH Medium-High Moderate
Hypsugo anchietae Anchieta's pipistrelle LC NT - Riparian vegetation. Clutter-edge and
clutter forager n.a. Low High
Neoromicia capensis Cape serotine LC LC
Under the bark of
trees, foliage,
buildings
Semi-arid areas to
mountain areas, forests
and savannah.
Clutter-edge
forager MH Medium-High High
Neoromicia nana Banana bat LC LC
Foliage (bananas and
other plants),
buildings
Well-wooded habitats,
riparian forest.
Clutter-edge
forager MH Medium-High High
Neoromicia zuluensis Zulu serotine LC LC - Woodland savannah,
riparian vegetation
Clutter-edge
forager MH Medium-High Moderate
Nycticeinops schlieffeni Schlieffen’s twilight
bat LC LC
Crevices in rock and
houses
Low-lying savannas,
well-wooded places,
riparian vegetation,
drainage lines.
Clutter-edge
forager n.a. Medium Low
Pipistrellus hesperidus Dusky pipistrelle LC LC
Narrow cracks in
rock, exfoliating
rock, under bark of
dead trees
Well-wooded habitats,
riparian forest.
Clutter-edge
forager MH Medium High
Scotophilus dinganii Yellow-bellied house
bat LC LC
Buildings, hollow
trees
Savannah, avoids open
habitats and bushlands
and Karoo scrubs.
Clutter-edge
forager n.a. Medium-High High
Scotophilus viridis Green house bat LC LC Hollow trees,
buildings
Hot savannah, avoids
open habitats.
Clutter-edge
forager n.a. Medium-High High
50 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Family Species Common name
IUC
N
So
uth
Afr
ica R
ed
Lis
t
Roosts Habitat preferences Foraging habits
Type of flight
Foraging habits
Flight height
Risk of collision
(Sowler and
Stoffberg,
2012)12
Probability
of
occurrence
RHINOLOPHIDAE
Rhinolophus blasii Blasius's horseshoe
bat LC VU Caves and mines
Savannah and
woodland. Clutter forager LH Low Low
Rhinolophus capensis Cape horseshoe bat LC NT Caves and mines Fynbos and succulent
Karoo biomes. Clutter forager LH Low Low
Rhinolophus clivosus Geoffroy's horseshoe
bat LC NT Caves and mines
Savannah, woodland
and riparian forest. Clutter forager LH Low High
Rhinolophus darlingi Darling's horseshoe
bat LC NT
Caves and mines,
also in culverts and
cavities in piles of
boulders
Savannah and
woodland. Clutter forager LH Low High
Rhinolophus simulator Bushveld horseshoe
bat LC LC
Caves and mines,
also in culverts and
small caverns
Savannah woodland and
riparian forests. Clutter forager LH Low High
Rhinolophus swinnyi Swinny's horseshoe
bat LC EN Caves and mines
Savannah woodland and
mountain forest. Clutter forager LH Low High
MOLOSSIDAE
Chaerephon pumilus Little free-tailed bat LC LC
Narrow crack in
rock and trees, also
in buildings and
crevices
Wide range of habitats Open-air forager HH High High
Mops condylurus Angolan free-tailed
bat LC LC
Narrow crevices in
rock, caves, hollow
trees, also houses
and bridges
Semi-arid to mesic
habitats; wide range of
habitats.
Open-air forager HH High High
Tadarida aegyptiaca Egyptian free-tailed
bat LC LC
Caves, rock crevices,
under exfoliating
rocks, hollow trees
and behind the bark
of dead trees, also
buildings
Wide variety of
vegetation (desert,
semi-arid scrub,
savannah, grassland and
agricultural land);
avoids forests.
Open-air forager HH High High
EMBALLONURIDAE
Taphozous mauritianus Mauritian tomb bat LC LC Rock faces, tree
trunks, walls Savannah woodland. Open-air forager HH High High
Taphozous perforatus Egyptian tomb bat LC - Crevices, hags in
sandstone, building
Savannah woodland;
open habitats. Open-air forager HH High Low
51 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3.1.1.2. Daily dispersion m ovements and known migration routes
Bat migration and dispersion behaviours and distances covered are not very well documented as
yet in South Africa. There is a lack of information in South Africa regarding the distribution and
abundance of bats as the migratory habits and migration routes of bats through the country are
not yet clearly understood. Much research is needed in this subject. However, there is some
evidence that some species reveal long-distance migration and seasonal movements within South
Africa. For example, Natal Long-fingered Bat (Miniopterus natalensis) is known to migrate up to
260 km (Van der Merwe, 1975 in Monadjem et al., 2010) between summer maternity caves and
those ones used during mating and hibernation period during the winter months. Temminck´s
Myotis (Myotis tricolor) may undertake similar seasonal migrations (Monadjem et al., 2010),
however details of this species are not well known. The frugivorous bat, Egyptian rousette
(Rousettus aegyptiacus) is a gregarious cave-dweller, also thought to move distances of between 50
km to 500 km (Monadjem et al., 2010).
Not much information is available regarding South African bat species’ home ranges and daily
dispersion movements (mainly to forage). Non-migrating bats will require movement around its
essential homing area: e.g. to forage, drink, and search for mates or search new roosting
locations. Some bat species will have daily roosts and night roosts (that they use for shorter
periods while foraging in an area) (Monadjem et al., 2010). Daily dispersion will depend on several
factors including the species, the habitat, weather conditions and food availability. Nevertheless,
based on the available information for South Africa and/or international references regarding
similar species somewhere else in the world, most bats species will cover, in general, less that
5km from their roosting location per night. Nevertheless, some species have been recorded to
travel longer distances, such as Epomophorus wahlbergi that was radio tracked to move more than
13 km between roosting and feeding sites per night (Monadjem et al., 2010); Nycteris thebaica that
has been recorded (only one individual) to move up to 9km (Monadjem, 2005); Rosettus
aegyptiacus was radio tracked up to 24km flying from a roosting cave to a feeding area (Jacobson
et al. 1986 in Monadjem et al. 2010).
3.1.1.3. Known roosting loc ations
As previously mentioned, several roosting caves are known in KwaZulu-Natal Province, including
roosting locations of migratory species such as Miniopterus natalensis, Myotis tricolor or Rousettus
aegyptiacus. These migratory species may possibly occur at Richards Bay wind energy facility study
area. However, since migration movements are not very well known for most of the South
African bat species, and as stated above, the majority of the species do not move long distances
to forage, it cannot be assured that the individuals of Miniopterus natalensis, Myotis tricolor or
Rousettus aegyptiacus that may be detected in the study area are related with the populations that
roosts at the caves mentioned in section 1.4.
52 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3.2. F IELD SURV EYS
The implementation of all of the previously indicated sampling methods resulted in the
confirmation of the presence of 21 bat species, including 20 echolocation bat species and 1 fruit-
eating bat species. These will be discussed in chapter 3.2.1, 3.2.2 and 3.2.3.
3.2.1. E cho loc ati on bat sp ec ies co nfi rme d at the si te
The results of all the methodologies implemented during the pre-construction bat monitoring
programme at Richards Bay wind energy facility site resulted in the identification of 20
echolocation bat species in the study area (Table 12). Six of these species are classified as “Near
Threatened”, according with the South African Red List, i.e.: Hypsugo anchietae, Miniopterus
fraterculus, Miniopterus natalensis, Myotis welwitschii, Myotis tricolor and Rhinolophus clivosus.
Accordingly to NSS indications some other species have been confirmed on site through call
analysis, i.e. Hypsugo anchietae, Rhinolophus simulator, Myotis welwitschii. However, the occurrence
of these species was not confirmed by the Bio3 team through the analysis of the collected data.
This may be due to several reasons, one of which is the different methods used by both teams for
the ultrasonic identification of bat species, as NSS used an automated process based on zero
crossing file type and Bio3 manually processed the wav files which contain more information than
the later file type. Another possible reason may arise from the implementation of the sub-
sampling method (due to the large number of recordings collected) by Bio3. Since random sub-
sampling was applied in Bio3 analysis there is the possibility of not selecting recordings of
determined species which are very rare or not abundant in the area, which seems to be the case
of the previous described species.
Table 12 – List of confirmed species at the Richards Bay wind energy facility site, between May 2012 and
April 2013, through all methodologies implemented. *species additionally confirmed by the NSS team only.
Common name Scientific name
Conservation status
Risk of
collision
(Sowler &
Stoffberg,
2012)
Method of Detection
Global
(IUCN,
2012)
National
(Monadjem
et al. 2010) Man
ual
surv
eys
Sta
tic
surv
eys
Ro
ost
insp
ecti
on
Liv
e-
trap
pin
g
Egyptian silt-faced bat Nycteris thebaica LC LC Low * *
Long-tailed serotine Eptesicus hottentotus LC LC Medium X X
Sundevall's leaf-nosed
bat Hipposideros caffer LC DD Low X
Lesser long-fingered bat Miniopterus
fraterculus LC NT Medium-High X X
Natal long-fingered bat Miniopterus natalensis LC NT Medium-High X X
Temminck's myotis Myotis tricolor LC NT Medium-High X
Anchieta's pipistrelle Hypsugo anchietae LC NT Low * *
Cape serotine Neoromicia capensis LC LC Medium-High X X *
Banana bat Neoromicia nana LC LC Medium-High X X *
53 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Common name Scientific name
Conservation status
Risk of
collision
(Sowler &
Stoffberg,
2012)
Method of Detection
Global
(IUCN,
2012)
National
(Monadjem
et al. 2010) Man
ual
surv
eys
Sta
tic
surv
eys
Ro
ost
insp
ecti
on
Liv
e-
trap
pin
g
Schlieffen’s twilight bat Nycticeinops
schlieffeni LC LC Medium * *
Dusky pipistrelle Pipistrellus hesperidus LC LC Medium X *
Yellow-bellied house
bat Scotophilus dinganii LC LC Medium-High X * *
Green house bat Scotophilus viridis LC LC Medium-High X *
Little free-tailed bat Chaerephon pumilus LC LC High X X * *
Angolan free-tailed bat Mops condylurus LC LC High X X
Egyptian free-tailed bat Tadarida aegyptiaca LC LC High X X
Egyptian tomb bat Taphozous
mauritianus LC LC High *
Geoffroy's horseshoe
bat Rhinolophus clivosus LC NT Low X
Bushveld horseshoe bat Rhinolophus simulator LC LC Low * *
Welwitsch's myotis Myotis welwitschii LC NT Medium-High * *
Additionally one species was considered suspected to occur in the study area, following NSS team
indications. This species is considered as “Vulnerable”, according with the South African Red List,
i.e.: Otomops martiensseni (Monadjem et al., 2010). Otomops martiensseni is also a NEM: Biodiversity
Act, 2004: Threatened and Protected Species (TOPS) listed species. According with the IUCN,
this is also the only species of conservation concern: the Otomops martiensseni is globally
considered as “Near Threatened” (IUCN, 2012).
This species is known to occur in large numbers in several buildings in the Durban area, where it
roosts communally. Although this is a relatively little studied species, foraging distances are
generally larger than 3km away from their roosts (Fenton et al., 2002). Considering that the
species is only known to roost and to occur in the Durban area, located at approximately 100km
southwest of the study area, this species was not considered as having a likely occurrence at the
Richards Bay study area following the parameters defined in section 2.1, and the information’s
provided by the field team were not enough to consider its occurrence in the study area as
confirmed. It’s potential occurrence in the area cannot be, however, excluded being considered
unlikely to roost in the study area until field work provides evidence to the contrary.
When analysing the records obtained through manual or static detection surveys in situations
where the species’ identification level was not possible, the outcome was the identification of the
corresponding groups of species, the genus or the family. Several sets of species with similar
acoustic parameters were identified in the study area and grouped as per Appendix III.
Table 13 indicates the number of records from both static and manual detection per species
analysed for the sub-sampling made. The Tadarida aegyptiaca was the most frequent species in the
study area and was recorded at least 548 times through static detection, representing 40.2% of
54 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
the total number of detections. Chaerephon pumilus, Miniopterus natalensis and Neoromicia capensis
were the species recorded more frequently while conducting manual detection surveys.
Table 13 – Number of recordings analysed with the sub-sampling method from both static and manual
detection per species identified by Bio3. Conservation Status according to the South Africa Red Book: LC –
Least concern; NT – Near Threatened (Monadjem et al., 2010); * Higher potential risk of collision with
wind turbines.
Common name Scientific name Conservation
status
Manual
detection
Static
detection Total %
Egyptian free-tailed bat Tadarida aegyptiaca* LC 23 613 636 33,2
Banana bat Neoromicia nana LC 39 452 491 25,6
Little free-tailed bat Chaerephon pumilus* LC 131 92 223 11,6
Cape serotine Neoromicia capensis LC 42 152 194 10,1
Natal long-fingered bat Miniopterus natalensis NT 75 74 149 7,8
Temminck's myotis Myotis tricolor NT 1 98 99 5,2
Yellow-bellied house bat Scotophilus dinganii LC 0 41 41 2,1
Angolan free-tailed bat Mops condylurus* LC 1 29 30 1,6
Long-tailed serotine Eptesicus hottentotus LC 13 16 29 1,5
Lesser long-fingered bat Miniopterus fraterculus NT 1 17 18 0,9
Dusky pipistrelle Pipistrellus hesperidus LC 0 3 3 0,2
Green house bat Scotophilus viridis LC 0 2 2 0,1
Geoffroy's horseshoe bat Rhinolophus clivosos NT 0 1 1 0,1
In spite of the conservation status of the species confirmed in the study area, it is important to
analyse their presence in the study area bearing in mind the potential risk caused by the project
for any of these species. Therefore, it is of note that 4 of the species with confirmed presence in
the study area present a high risk of collision with wind turbines, i.e.: Chaerephon pumilus, Mops
condylurus, Tadarida aegyptiaca and Taphozous mauritianus (all considered to have populations of
Least Concern both nationally and internationally). This is caused mostly by their flight type and
foraging behaviour, since these species forage in open areas and may fly at high altitudes,
potentially coincident with the rotor swept area (Sowler and Stoffberg, 2012). There are records
of mortality of species from Tadarida sp. genus on wind farms elsewhere in the world (Eurobats,
2013).
From the remaining confirmed species, 3 are considered to have a medium potential collision risk
with wind turbines, i.e.: Eptesicus hottentotus, Nycticeinops schlieffeni and Pipistrellus hesperidus all
considered to have populations of Least Concern both nationally and internationally. These 3
species are clutter-edge foragers (Monadjem et al., 2010) and have specific morphologic and
acoustic adaptations to allow the required manoeuvrability refined acoustic echolocation in order
to hunt for its insect preys while avoiding colliding with the background vegetation (e.g. short and
broad wings that facilitate slow, manoeuvrable flight) (Schnitzler and Kalko, 2001).This means that
this species will forage primarily around vegetation (clutter) associated with either forested areas
55 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
or tall bushes and it is not expected to fly higher than 2 to 10m above or far from the vegetation
clutters. This means that its absolute flight height (distance from the individuals to the ground) will
depend mainly on the height of the vegetation, and on its foraging areas. There are records of
mortality with wind turbines of species within the same genus in Europe (Eurobats, 2013;
Appendix IV).
There are 7 species present considered to have a medium-high potential risk of collision with
wind turbines, i.e.: Miniopterus fraterculus, Miniopterus natalensis, Myotis tricolor (the latter
considered Near Threatened in South Africa), Neoromicia capensis, Neoromicia nana, Scotophilus
dinganii and Scotophilus viridis. All these species are clutter-edge foragers.
There are 3 species considered to have a low potential collision risk with wind turbines, i.e.:
Rhinolophus clivosus, Nycteris thebaica and Hipposideros caffer (Table 12).
It is possible to assume that the species that are expected to be mostly affected by the wind
energy facility could potentially be the open air foragers as most of these species, additionally to
having behaviours that pose higher risks, they were amongst the more abundant species in the
area (e.g. Tadarida aegyptiaca and Chaerephon pumilus, both considered to have populations of
Least Concern both nationally and internationally) (Table 13). Clutter edge foragers may be
affected by the wind energy facility and collisions are likely to occur as well, because some of
these species are within the genus that has records of mortality in other parts of the world.
Nevertheless, it is considered that if mortality due to collision with wind turbines occurs with
these medium and medium-high risk species it should be at a lesser extent and it will depend
mainly on the habitats where turbines will be sited (higher if turbines are sited closer to forested
or high vegetation areas and lower if turbines are to be sited in open areas).
Nevertheless, bats also have to move from their foraging areas to the roosting areas and even the
clutter and clutter-edge foragers may fly over open areas to accomplish this. Therefore, these
movements will pose a potential higher collision risk with wind turbines.
At least 2 of the confirmed species in the study area, through ultrasound analysis, are migrant
species, known to migrate from winter to summer roosts, to distances that can be up to 150km,
i.e.: Miniopterus natalensis and Myotis tricolor (Table 14). It is of note that for the remaining species
there is no information regarding migratory movements, being therefore very difficult to assess
possible migration patterns. Considering that bats can migrate mostly due to reproduction, it is
important to refer to the breeding season of the species present in the study area. For most of
the species, breeding season occurs between March and May, typically during the autumn season,
while births occur between October and December, during spring.
56 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 14 – Species confirmed by the ultrasound analysis in the study area from May 2012 to April 2013;
Status according to the South Africa Red List: LC – Least concern; NT – Near Threatened (Monadjem et
al., 2010); n.a. – information not available.
Scientific name Common name
So
uth
Afr
ica
Red L
ist
Migrant Breeding
season Birth season
Tadarida aegyptiaca Egyptian free-tailed bat LC n.a. August November-December
Neoromicia nana Banana bat LC n.a. May November-December
Neoromicia capensis Cape serotine LC n.a. March-April October-November
Chaerephon pumilus Little free-tailed bat LC n.a. n.a. November; January;
March-April
Miniopterus natalensis Natal long-fingered bat NT
Females migrate to
maternity roosts
seasonally between
roosts up to 150km
March-April October-December
Myotis tricolor Temminck's myotis NT Migrants from winter
to summer roosts April November-December
Scotophilus dinganii Yellow-bellied house
bat LC n.a. n.a. November-December
Eptesicus hottentotus Long-tailed serotine LC n.a. n.a. n.a.
Miniopterus fraterculus Lesser long-fingered
bat NT n.a. May-June November – December
Scotophilus viridis Green house bat LC n.a. n.a. n.a.
Mops condylurus Angolan free-tailed bat LC n.a. n.a. September; May
Pipistrellus hesperidus Dusky pipistrelle LC n.a. n.a. October
Geoffroy's horseshoe bat Rhinolophus clivosos NT n.a. May December
A total of 584 feeding buzz events were detected within the analysed recordings from all surveys
conducted between May 2012 and April 2013 (Table 15). Most of those feeding buzz pulses were
emitted by Molossid species and by a group of two species which could not be distinguished due
to the quality of the recordings made: Chaerephon pumilus and/or Mops condylurus, especially
during summer months (November and December). Tadarida aegyptiaca also presented a high
number of feeding buzz events in the study area, during the same time of the year (November and
December). The bats emit pulses to navigate, to avoid collision with objects and to locate prey.
At first the pulses are spaced to verify the presence of prey and once a potential prey is detected
the interval of emission of pulses decreases and, as the bat gets closer to the prey, the time
between pulses decreases, originating the “buzz”. Those buzzes are identified as feeding buzzes,
corresponding to the moment when the bat is closest to its prey. During a feeding buzz the pulse
frequency gets closer to the audible (Ahlen, 1990; Tupinier, 1996; Briggs and King, 1998). While
an individual is navigating or looking for prey it is also foraging, although no feeding buzzes are
produced. So the feeding buzzes are the confirmation that the bat is using the area to forage, but
the possibility that bats are foraging in the area in the absence of feeding buzz should always be
considered. A large number of passes can also indicate that the area is used as a foraging site.
Feeding buzz calls were higher between November and December, during summer, when food
availability is higher. Since feeding buzzes are indicators of feeding activity it is logical to assume
that in the study area this is the time of the year when bats are more active in foraging and
57 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
hunting activities. Feeding buzzes are events that are also less likely to be recorded, in relation to
navigation pulses, therefore this information needs to be considered with caution.
On the other hand, social calls are pulses emitted at lower frequencies and generally shorter in
duration. The reason why those social calls are emitted is not fully understood. However it is
known that, for some species, the male individuals use these calls to attract females, at roost
entrance, or to repel rival males (Kunz and Fenton, 2003). This type of call is also used by several
species to draw hunting territories and avoid conflict among individuals, especially when prey
densities are very low, since the number of social calls rise when the insect densities diminish, and
its use is being described by some authors as a measure for feeding success (Kunz and Fenton,
2003). Social calls may also be used to promote group cohesion, especially at roost exits, as a way
to defend from predators, and in breeding colonies (Kunz and Fenton, 2003). In general, this type
of call may be used at intra-specific level (Kunz and Fenton, 2003). In the study area, 194 social
calls were identified within the recordings analysed and collected through static and manual
detection, throughout the year, but mostly in December (Table 16). Among the species that were
identified emitting this type of pulse, Chaerephon pumilus, the species group Chaerephon pumilus
and/or Mops condylurus and Tadarida aegyptiaca were the most frequent. More bats are active at
this time of the year (December), due to better weather conditions and higher food availability,
increasing the likelihood to record this type of calls between individuals of the same species.
Table 15 – Feeding buzz pulses identified during static and manual surveys conducted at the Richards Bay
wind energy facility site between May 2012 and April 2013. Considering the data analysed thought the sub-
sampling methodology implemented.
Species
2012 2013
Total
May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr.
Chaerephon pumilus 4 0 2 22 3 0 0 0 0 2 3 3 39
Chaerephon pumilus/Eptesicus
hottentotus 0 0 0 7 0 0 0 0 0 1 0 0 8
Chaerephon pumilus/Mops
condylurus 0 0 0 12 4 2 74 39 0 11 7 2 151
Chaerephon pumilus/Tadarida
aegyptiaca 0 0 0 0 0 0 0 0 4 1 0 0 5
Eptesicus hottentotus 0 0 0 2 0 0 0 1 0 0 0 0 3
Eptesicus
hottentotus/Scotophilus dinganii 0 0 0 1 0 0 2 2 0 0 0 0 5
Hypsugo anchietae/Pipistrellus
sp. 0 0 0 0 0 0 0 0 0 1 0 0 1
Miniopterus natalensis 0 0 1 1 1 0 0 0 0 0 0 1 4
Mops condylurus 0 0 1 8 0 0 0 0 0 0 0 0 9
Mops condylurus/Tadarida
egyptiaca 1 0 0 5 1 0 0 0 0 0 0 0 7
Myotis tricolor 0 0 3 7 0 0 0 0 0 0 0 0 10
Neoromicia capensis 0 0 4 1 0 0 0 0 0 2 1 0 8
Neoromicia
capensis/Nycticeinops schlieffeni 0 0 0 4 0 0 0 0 0 0 0 0 4
Neoromicia nana 1 0 2 5 0 0 0 0 0 0 0 0 8
58 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Species
2012 2013
Total
May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apr.
Pipistrellus hesperidus 0 0 0 1 0 0 0 0 0 0 0 0 1
Rhinolophus clivosus 0 0 0 1 0 0 0 0 0 0 0 0 1
Scotophilus dinganii 0 0 0 0 1 0 0 2 0 0 0 4 7
Tadarida aegyptiaca 1 0 0 4 1 3 48 28 0 0 0 0 85
Emballunoridae/Molossidae 0 0 0 2 0 0 0 0 0 0 0 0 2
Nycteridae/Vespertilionidae 0 0 0 1 0 0 0 0 0 0 0 0 1
Miniopteridae/Vespertilionidae 0 0 3 11 0 0 0 1 0 0 4 3 22
Molossidae 2 0 0 16 22 2 82 45 0 0 0 0 169
Vespertilionidae 0 0 0 4 0 0 0 0 0 0 0 0 4
Unidentified 0 0 2 3 0 1 15 8 0 0 0 1 30
Total 9 0 18 118 33 8 221 126 4 18 15 14 584
Table 16 – Social calls identified during static and manual surveys conducted at the Richards Bay wind
energy facility site between May 2012 and April 2013. Considering the data analysed thought the sub-
sampling methodology implemented.
Species
2012 2013
Total
May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. April
Chaerephon pumilus 0 0 0 0 0 0 1 15 0 12 0 0 28
Chaerephon pumilus/Mops
condylurus 25 2 0 4 12 0 10 20 0 0 0 0 73
Chaerephon pumilus/Tadarida
aegyptiaca 0 0 0 0 0 0 0 1 0 0 0 0 1
Eptesicus hottentotus 0 0 0 0 0 0 0 5 0 0 0 1 6
Eptesicus
hottentotus/Scotophilus
dinganii
0 0 0 0 1 0 0 0 0 0 0 0 1
Hypsugo anchietae/Pipistrellus
sp. 0 0 0 0 3 0 0 0 0 0 0 0 3
Miniopterus natalensis 0 0 1 0 2 0 0 0 0 0 0 0 3
Mops condylurus /Tadarida
aegyptiaca 1 0 0 0 0 0 0 0 0 0 0 0 1
Myotis tricolor 0 0 0 1 0 0 0 0 0 0 0 0 1
Neoromicia
capensis/Nycticeinops
schlieffeni
0 0 0 1 0 0 0 0 0 0 0 0 1
Neoromicia nana 0 0 0 4 0 0 0 0 0 0 0 0 4
Scotophilus dinganii 1 0 0 0 1 0 0 0 0 0 0 0 2
Tadarida aegyptiaca 2 2 0 0 3 1 6 7 0 0 0 0 21
Molossidae 6 0 0 2 7 3 7 2 0 0 0 0 27
Molossidae/Vespertilionidae 0 0 0 0 0 0 2 0 4 3 0 1 10
59 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Species
2012 2013
Total
May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. April
Unidentified 0 0 0 4 2 0 1 5 0 0 0 0 12
Total 35 4 1 16 31 4 27 55 4 15 0 2 194
3.2.2. N on- ech olo cat ion sp ec ies - Fru it- ea tin g b at spe cie s
It is considered as possible the occurrence of at least 4 fruit-eating bat species in the Richards Bay
wind energy facility study area, i.e.: African straw-coloured fruit bat (Eidolon helvum), Peter's
epauletted fruit bat (Epomophorus crypturus), Wahlberg's epauletted fruit bat (Epomophorus
wahlbergi) and the Egyptian rousette (Rousettus aegyptiacus) (Table 11). From these 4 species,
Epomophorus crypturus is considered to have a low probability of occurrence due to its habitat
preferences, since it is associated with dense foliage forests and prefers drier conditions than
those observed in the study area, mostly occupied by sugar cane plantation.
Since fruit-bat species do not echolocate (with the only exception of the Egyptian rousette that
emits a type of rudimentary echolocation pulses or clicks to aid its navigation), their identification
is only possible through direct observation at roosting sites, while flying through the area, finding
roost locations or through live-trapping. During field work conducted as part of the
Environmental Assessment Study, one roost with Wahlberg's epauletted (Epomophorus wahlbergi)
was confirmed and another two roosts with this species were considered possible to occur (CES,
2012). The Wahlberg's epauletted is fairly widespread and abundant in the eastern parts of South
Africa, including KwaZulu-Natal. This fruit bat roosts in small groups in dense foliage in large
trees, being associated with forests and forest-edge habitats, riparian vegetation, and occasionally
wooded gardens. As its name suggests, this species feeds mostly on fruit, nectar, pollen and
flowers, being able to travel several kilometres in the same night to search for fruit trees.
A few roosting locations of Epomophorus wahlbergi were found by NSS field team within the study
area, with the higher recorded occupation in the April surveys. Pups were also confirmed in
February surveys. Additional information on the roosts of this species is presented in section 3.4.
Considering the flight behaviour of this species, the only confirmed fruit bat in the study area, it
may be affected by this development since it is considered to have a medium-high risk of collision
with wind turbines (Sowler and Stoffberg, 2012).
3.2.3. L ive -tr app ing
Live-trapping with mist-netting, harp trapping and grab sampling were used by NSS to confirm the
presence of certain species and to record their calls as well. Therefore, through these methods,
the presence of Scotophilus viridis, Scotophilus dinganii, Nycticeinops schlieffeni, Nycteris thebaica,
Neoromicia capensis, and Chaerephon pumilus were confirmed by NSS in the study area (Table 12).
60 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Notice that all species captured through this method had already been confirmed through ultra-
sound detection with the exception of Nycteris thebaica, Nycticeinops schlieffeni and Scotophilus
viridis. These species had only been previously observed in the study area through roost surveys in
a roost located in Shayamoya region, adjacent to Reding Dam. During the live-trapping surveys
conducted in February and April 2013, no individuals were captured.
3.3. SPATI AL-T EMPOR AL AC TIVITY
Most of the species that can occur in the study area are insectivorous and their annual cycle is
related to the abundance of food resources. Since the insect population increases in spring and
summer, it is expected that the bat activity follows a similar pattern. In KwaZulu Natal the
weather is very mild throughout the year, with temperatures in winter around 17ºC and 32ºC in
the hottest summer month. In this province rain is very frequent, with approximately 1000mm a
year. Therefore, considering this favourable climatic conditions, bat activity is expected all year
round, but with higher expression between September and February, corresponding to spring and
summer seasons. This is due mainly because the pattern of insect availability follows the
temperature and rainfall pattern. The high level of activity will be reflected in a high number of
passes and time usage of the area. The bat’s activity levels are also influenced by other factors,
such as weather, biotope or distance to water sources.
3.3.1. S pat ial an aly sis
In this section bat activity will be analysed in relation to its spatial distribution in the wind energy
facility study area. This analysis is directed towards the identification of higher activity areas,
which will help to determine the potential effects of this proposed project on the bat community
of the study area.
3.3.1.1. Manual detection
Figure 7 represents the total number of passes recorded through the car transects conducted, in
a 500x500 meters grid cell over the study area. The higher activity locations are located along the
grid cells sampled in the northeast quadrant of the study area (Figure 7). Considering the
proposed turbine layout, at least one wind turbine (WTG25) is located in a medium activity grid
cell (between 20 to 30 bat passes) at Scott’s Properties. In the southern part of the study area
another wind turbine (WTG48) is proposed in an area considered of medium activity (between
20 to 30 bat passes). The remaining wind turbines WTGnew2, WTGnew3, WTGnew4, WTGnew
5, WTGnew6, WTG1, WTG3, WTG10, WTG11, WTG14, WTG15, WTG31, WTG33, WTG40,
WTG41, WTG42 and WTG46 are proposed for areas where less bat activity was recorded, or
their location was not sampled by this survey method. Despite the fact that the vehicle transect
did not covered the location of the referred wind turbines, it is possible to assume that the bat
activity at the location of these wind turbines will be similar to that observed at the closest grid
cells. Therefore it would be plausible to say that the WTG3 and WTG31 could also be located in
61 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
an area of medium activity (20 to 30 bat passes), while the remaining turbines should be located in
medium-low or low activity areas. However the static sampling method (through fixed points of
static detection) is intended to sample these possible gaps in knowledge.
Two grid cells present values higher than 40 bat passes for the sum of all surveys conducted.
However, they are located more than 1km from any of the wind turbines proposed, thus not
representing an immediate possible negative effect on bats in the study area.
Figure 7 – Total number of passes recorded in each grid cell (500 x 500 meters), in the wind energy
facility site, between May 2012 and May 2013. Vehicle based transects data.
Bat activity also varies according to biotopes present, since some biotopes provide shelter from
wind, predators and some are prone to increased food availability (Verboom and Huitema, 1997).
It is also worth mentioning that different species also have different biotope preferences
(Monadjem et al., 2010).
The number of passes recorded in each grid cell of the study area overlapping the types of
vegetation found in the wind energy facility area is represented in Figure 8. Most of the study area
is composed by sugar cane plantation that is not very interesting for bats, since it provides almost
62 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
no shelter when it’s harvested and not much more when it’s fully grown. Nonetheless, insect
availability in sugar cane plantation is expected to be high and provide suitable foraging areas for
some bat species. There are some “pockets” of native vegetation composed by some types of
forest vegetation, as well as native bushes (as described in section 1.4) that provide shelter for
tree roosting species and food availability. It is also known that some bat species prefer to hunt
near water sources (Monadjem et al. 2010; Rainho and Palmeirim, 2011; Scott et al., 2010), which
are also present within the Richards Bay wind energy facility site, namely the Reding Dam and the
Okuila River which also have some adjacent riverside vegetation. These areas are usually
attractive for bats as hunting/feeding areas. The locations where the number of bat passes and the
number of confirmed species was higher are the grid cells coinciding with native vegetation of
dams and riverside vegetation which are the ones presenting higher levels of activity (Figure 8).
However the influence of these types of vegetation did not proved to have a significant influence
over bat activity (p>0.05) (refer to section 3.3.5.1). This is probably due to the large percentage
that sugar cane plantations occupies in all sampling points (Table 6), which are masquerading the
influence of other biotopes with lower percentage of occupation but may be relevant for bat
activity.
Nonetheless, considering this apparent relation between the type of vegetation present and the
bat activity, it is important to regard which wind turbines are proposed to be placed in areas
consisting of vegetation with higher interest for bats: riverine vegetation and native vegetation.
Analysing the proposed layout, it is concluded that some wind turbines area located within, or in
the immediate surroundings of, this type of vegetation (e.g. WTGnew 4, WTGnew5, WTG03,
WTG08, WTG09, WTG10, WTG11, and WTG25).
The siting of these wind turbines have already been previously analysed and none was placed in an
area of very high bat activity (more than 40 bat passes). However, it is of note that the location of
wind turbines in areas expected to have high bat activity (30 to 40 bat passes) may result in the
occurrence of negative effects over bat populations in the study area, as for example bat fatalities
during the operational phase of this project. In order to avoid these potential impacts, measures
can be implemented (e.g. habitat management surrounding wind turbines) in order to avoid bat
utilization of these areas, if relevant bat collisions are verified during the subsequent phases of the
project.
63 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 8 – Total number of passes recorded in each grid cell (500 x 500 meters), in the wind energy
facility site, between May 2012 and May 2013 and biotopes present in the study area. Vehicle based
transects data
3.3.1.2. Static detection
Concerning the spatial distribution of bat activity in the study area obtained through static
detection, the average number of passes per hour at each of the static recording locations is
shown in Figure 9. In most of the static detectors the annual average activity index was between
10 and 30 passes / hour with two exceptions: 1) at RB3 less than 10 passes per hour were
recorded, being the minimum values observed during the pre-construction monitoring
programme; and 2) the maximum value was registered at RB2 with activity above 50 passes per
hour (approximately 133 passes per hour).
64 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Most of the sampling points were located in sugar cane plantation areas, the most predominant
biotope at the study site. However RB5 was located near farm houses surrounded by vegetation,
RB6 and RB2 were installed near a Dam with riverine vegetation (Table 6). Nonetheless, all static
detectors were installed in areas dominated by sugar cane plantation in the immediate
surroundings (500 m buffer). This difference in the surrounding biotope of the sampling point
does not seem to explain the variation of activity throughout the area, since sampling points
located in areas with high coverage of sugar cane plantation had a great variation in bat activity
results. Therefore the percentage of occupation of the different types of land use surrounding
each of the static detectors was considered, in an effort to understand if the variation observed
was caused by some slight variation in the land cover. The statistical analysis has shown that the
presence of human-made structures, such as farm houses, in the area surrounding the static
detectors had significant influence over bat activity (p<0.05).
Figure 9 – Average activity index (number of passes per hour) recorded between May 2012 and May 2013
and land use observed at the Richards Bay wind energy facility.
65 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3.3.2. B at act ivi ty at hei gh t v s . gro und l eve l
Besides the analysis of horizontal distribution of bat activity in the study area (at the biotope level)
it is also important to assess differences between bat activity at different heights, i.e.: below rotor
swept area (low height) and at rotor swept area (high height). The average activity index in all
detectors, considering the height that each detector or microphone was placed is presented in
Figure 10. Since RB1 microphones were installed at 7 and 30 meters during the first surveys (May
and June 2012), as previously explained in section 2.2.4.2, no high height records were considered
in such months (since 30 meters was considered below rotor swept area). Nonetheless, it was
observed that until December 2012, bat activity was generally higher below rotor swept area.
However, from January until April 2013, for summer and autumn months, bat activity was similar
or higher at rotor swept area (above 60 meters). A hypothesis to justify this observation may be
related with bat migration, since several important bat caves are known in the surrounding area,
and it is also known that bats migrate at a higher altitude (Barclay et al., 2007) but this needs
further confirmation. The presence of bats flying above 60 meters during this time period (January
until April) may indicate higher activity of bat species that migrate, and may be leaving summer
roosts or bat species that fly at high altitude, such as open-air forager species, that generally
present high risk of collision with wind turbines. On the other hand, this may also be due to some
fault at the RB1 detector microphone placed at lower height that presented some problems from
January onwards due to a violent storm on 25 December 2012. Therefore this data must be
confirmed in subsequent monitoring surveys.
Figure 10 – Average activity index (number of passes/hour) and for each of the heights surveyed, Low (7,
10 and 30m) and High (60 and 80 m), per month. Vertical bars represent standard error. Analysis
considering the total data collected from all the detectors installed on site.
0
20
40
60
80
100
120
May
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Nu
mb
er
of p
ass
es/
ho
ur
Low Height High Height
66 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Observing the activity registered at different heights, obtained through the detectors placed at 80
m and 10 m at the RB4 sampling location (Figure 11), it is possible to verify that most bats were
detected by the detector placed at 10 meters throughout most of the December month.
However in January bat activity was very similar at both sampling heights and from February
onwards most of bat activity was detected at higher heights (Figure 11). However this data was
considered not very reliable since one of the field assistants did not adequately check the ground
level microphone at RB4 by the end of February 2013, and this detector presented problems
through March and April 2013. Therefore bat activity at lower height may be higher than the
recorded values and the absence of data may not be related to real lower activity of bats.
Figure 11 – Average activity index (average number of passes/hour) at the RB4 sampling point, at two
different heights between December 2012 and April 2013. Vertical bars represent standard error. Analysis
considering the total data collected.
The evaluation of bat activity at different heights can also be made using the information collected
by the two microphones of the RB1 static detector (Figure 12). In this detector, bat activity at the
microphone placed at 7m and 10m was generally higher than the activity captured at 30m or 60m.
This happened throughout most of the year. However, during the winter sampling (June and July
2012) activity at both heights was very similar. This similarity may be due to the heights that the
microphones were placed, since during the June and half of July sampling surveys, microphones
were placed at 7 and 30 m as due to weather conditions at the time of installation that did not
allow the placement of the microphones at their definitive positions. Therefore it may be possible
that activity is similar up to 30 m in height, being however lower at higher heights (e.g. 60 m). The
discrepancy in bat activity between different heights seems to be more evident during the end of
spring and summer periods. The absence of data from January onwards may be due to
malfunction of the detector, as explained previously.
0
20
40
60
80
100
December January February March April
Summer Autumn
Avera
ge n
um
ber
of p
ass
es/
ho
ur
10m 80m
67 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
This apparent higher bat activity at lower heights in relation to higher heights was confirmed
through statistical analysis, since this variable class had a significant influence over the amount of
bat passes recorded (p<0.05).
Figure 12 – Average activity index (average number of passes/hour) at the RB1 sampling point, at different
heights in the wind energy facility site between May 2012 and April 2013. Bars represent standard error.
Analysis considering the total data collected.
The activity of the confirmed species through static detection at different heights is represented in
Figure 13: below rotor height (detectors installed up to 30m) and above rotor height (detectors
installed higher than 60m). As the previous analysis revealed, bat activity tends to be higher below
rotor height for all of the species detected. Some species were only detected below rotor height,
while others were also detected while flying above rotor height, i.e.: Eptesicus hottentotus,
Neoromicia capensis, and Tadarida aegyptiaca. The remaining species were only detected flying
below rotor height.
Considering the families of species detected, the Molossidae/Vespertilionidae was detected at
higher altitudes at 21% of the occurrences, being the group with higher detections at higher
altitudes. Miniopteridae/vespertilionidae family was the second group of species with a higher
number of detections at 60m and higher, with 16% of the detections at this height, followed by
the Molossidae family with 8% of the detection at 60m and higher.
Although Eptesicus hottentotus, Neoromicia capensis and Tadarida aegyptiaca were detected by the
detectors installed at 60m and higher (approximately 8% of the activity detected), they were also
detected below rotor height and at a higher frequency (approximately 92% of the activity). This
may be influenced by the way bats emit their sonar pulses, with the activity index below rotor
height of these species possibly being overestimated. Bats emit pulses to navigate at a regular
rhythm, changing its orientation, frequency and duration according with the habitat, type of flight
0
20
40
60
80
100
120
140
160
May
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Nu
mb
er
of p
ass
es/
ho
ur
7m 30m 10m 60m
68 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
and possibly height of flight (Altringham, 2001; Tupinier, 1996). Other studies have also detected a
similar pattern of activity, showing more bat detections at lower height, proposing that bats that
fly higher point their sonar beam downward, while bats that fly at lower altitudes tend to point
their sonar beam forward (Collins and Jones, 2009; Jensen and Miller, 1999). Therefore, the
hypothesis that part of the detections at lower height could be attributed to bats flying at higher
altitude (pointing their sonar beam downward being therefore captured by the lower
microphone) should not be ruled out.
Figure 13 – Average activity index (average number of passes/night) of the confirmed bat species and
families through static detection, at different heights in the wind energy facility site between May 2012 and
April 2013. Vertical bars represent standard error. Analysis considering the sub-sampling method with all
static detectors.
3.3.3. I nfl uen ce of Env iro nm ent al var iab le s
3.3.3.1. Manual detection
Since bat activity depends on environmental conditions, such as temperature, wind speed and
illuminated lunar fraction, as well as on biotope and distance to water, it is important to analyse
possible relations between bat activity and each one of these factors.
0
10
20
30
40
50
Cha
erep
hon p
umilu
s
Epte
sicu
s hot
tento
tus
Min
iopte
rus
frate
rculu
s
Min
iopte
rus
nata
lensis
Mop
s co
ndyluru
s
Myo
tis
tricol
or
Neo
rom
icia
capen
sis
Neo
rom
icia
nana
Pipistr
ellu
s he
sper
idus
Rhin
olop
hus
cliv
osus
Scot
ophilu
s din
ganii
Scot
ophilu
s viridis
Tada
rida
aeg
yptia
ca
Em
ballo
nur
idae
/Ves
per
tilio
nid
ae
Min
iopte
ridae
Min
iopte
ridae
/Ves
per
tilio
nid
ae
Mol
ossida
e
Mol
ossida
e/Ves
per
tilio
nidae
Nyc
teridae
/Ves
per
tilio
nid
ae
Rhin
olop
hid
ae
Ves
per
tilio
nidae
Nu
mb
er
of b
at
pass
es/
ho
ur
Below rotor height (7, 10 and 30m) Above rotor height (60 and 80m)
69 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Higher air temperature values correspond, in general, to a higher availability of insects and
therefore a high food source for insectivorous bats (Rodrigues and Palmeirim, 2007). Therefore, a
directly proportional relation between temperature and bat activity should be expected
(Speakman and Rowland, 1999; Kusch et al., 2004; Müller et al., 2012).
The temperature recorded during the surveys fluctuated between the minimum recorded in July
(winter) and the maximum recorded in February (summer) (Figure 14). In general, the results
from manual detection do not suggest a clear pattern between bat activity and air temperature,
since the increase in air temperature did not always result in any clear increase in the number of
passes, considering that in the study area temperatures seemed not to have a great fluctuation
throughout the year. However, the statistical analysis have shown a significant positive influence of
this variable over bat activity (p<0.05) (refer to section 3.3.5.1).
Figure 14 – Average number of passes/grid cell (500x500m) and average air temperature registered
between May 2012 and May 2013 at Richards Bay wind energy facility site. Bars represent standard error.
Analysis considering the total data collected through all vehicle based transects.
Considering the average air humidity recorded at each of the sampling surveys, it is possible to
notice a direct relationship tendency between the average number of passes and the air humidity
(Figure 15). Therefore, the increase in air humidity seems to be related with the increase in bat
activity in most of the surveys conducted, as supported by the statistical analysis (p<0.05) (refer
to section 3.3.5.1).
When observing Figure 16, where the relationship between bat activity and average wind speed at
each of the surveys conducted is represented, it is not possible to perceive any direct relationship
0
5
10
15
20
25
30
0
1
2
3
4
5
May
July
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
May
Autumn Winter Spring Summer Autumn
Tem
pera
ture
(ºC
)
Nu
mb
er
of
pass
es/
gri
d c
ell
Month/Season
Average number of passes Temperature
70 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
due to this environmental variable. Nonetheless, this variable has a significant negative influence
over bat activity (p<0.05) (refer to section 3.3.5.1).
The illuminated lunar fraction can also influence bat activity since bats are expected to be highly
exposed to predators, such as owls, snakes or genets, and some studies indicate that bat activity
is inversely proportional to illuminated lunar fraction (Lang et al., 2005; Cryan and Brown, 2007;
Esberard, 2007). Nevertheless, there is still some controversy around this issue once some
authors suggest this relationship is dependent on several other factors (Hecker and Brigham,
1999). The illuminated lunar fraction observed during the manual surveys conducted at the wind
energy facility varied between new moon and full moon (Figure 17). When the illuminated lunar
fraction was at its lowest point (close to 0), in July and September, the average number of passes
per grid cell was approximately 1. When the illuminated lunar fraction was at its highest point
(close to 1), in November and March, the average number of passes varied greatly, with almost no
activity in December, and a high peak of activity in March. Therefore from the graphical analysis of
results, the illuminated lunar fraction does not seem to have a clear influence on bat activity in a
larger extent, as confirmed by the statistical analysis which did not consider this environmental
variable as significant to explain bat activity at the study area (p>0.05) (refer to section 3.3.5.1).
Figure 15 – Average number of passes/grid cell (500x500m) and air humidity registered between May 2012
and May 2013 at Richards Bay wind energy facility site. Bars represent standard error. Analysis considering
the total data collected through all vehicle based transects.
0
10
20
30
40
50
60
70
80
90
100
0
1
2
3
4
5
May
July
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
May
Autumn Winter Spring Summer Autumn
Hu
mid
ity (
%)
Nu
mb
er
of
pass
es/
gri
d c
ell
Month/Season
Average number of passes Humidity
71 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 16 – Average number of passes/grid cell (500x500m) and average wind speed registered between
May 2012 and May 2013 for the days when transects where conducted at Richards Bay wind energy facility
site. Bars represent standard error. Considered total data collected through all vehicle based transects.
0
2
4
6
8
10
12
0
1
2
3
4
5M
ay
July
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
May
Autumn Winter Spring Summer Autumn
Win
d s
peed
(km
/h)
Nu
mb
er
of
pass
es/
gri
d c
ell
Month/Season
Average number of passes Wind speed
72 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 17 – Average number of passes/grid cell (500x500m), recorded from May 2012 to May 2013 and
the average illuminated lunar fraction. Fraction illuminated at New Moon is 0,0 and at Full Moon is 1,0.
Bars represent standard error. Considered total data collected through all vehicle based transects.
3.3.3.2. Static detection
Among the studied variables to explain bat activity in the study area, air humidity (Figure 18)
appears, by graphical interpretation of the results, to be the only one directly related with the
average number of passes recorded by hour in each of the surveys conducted. Between May 2012
and April 2013, the fluctuations in air humidity were followed by bat activity. Apparently, when air
humidity was higher, bat activity was also higher, and vice-versa, when air humidity was lower, bat
activity was also lower. However, during February onwards this tendency did not seem to occur.
The absence of influence of this variable over bat activity did not had any significant influence over
bat activity (p>0.05) (refer to section 3.3.5.2).
Wind speed relation with the average number of passes recorded in the study area is shown in
Figure 19. An influence of this environmental variable over bat activity is more difficult to notice
graphically since an increase in wind speed was not reflected in changes over bat activity.
Therefore it is not possible to establish a clear relation of this environmental variable with bat
activity at Richards Bay study area solely based on the visual graphical interpretation of results.
The analysis of the variation of bat activity and hourly wind speed through statistic analysis proved
that this environmental variable had a negative significant influence over bat activity (p<0.05)
(refer to section 3.3.5.2). Although a monthly variation between this variable and bat activity
could not be graphically observed, when analysing bat activity in relation to the wind speed
recorded during the hour at which the pass was recorded, it seems to persist a negative relation,
0
0.2
0.4
0.6
0.8
1
0
1
2
3
4
5M
ay
July
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
May
Autumn Winter Spring Summer Autumn
llu
min
ate
d l
un
ar
fracti
on
Nu
mb
er
of
pass
es/
gri
d c
ell
Month/Season
Average number of passes Illuminated lunar fraction
73 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
with the increase of wind speed reducing the amount of bat passes registered. In Figure 20 the
influence of wind speed over bat activity at different heights is represented. While bat activity
below rotor height does not seem to be very influenced by the increase in wind speed, bat
activity recorded at rotor height suffered a severe decrease with the increase of wind speed at
speeds higher than 3m/s (decrease of approximately 63% of the bat activity).
In Figure 21 it can be observed that the temperature recorded was mostly constant throughout
the year, between 20ºC and 26ºC, decreasing only during the winter months to approximately
19ºC. This variation did not seem to influence bat activity (average number of passes per hour) as
graphically observed, however the analysis of its hourly variation proved the existence of a
significant positive influence over bat activity (p<0.05) (refer to section 3.3.5.2).
Observing the relationship between bat activity and hourly air temperature it is clear an influence
of temperature, proving that the increase of air temperature results in an increase of bat activity.
It is also verified that bats have a maximum temperature to which most of species are
comfortable, approximately 24ºC. Higher temperatures than 24ºC led to a decrease in bat
activity. At 29ºC a peak of activity was verified below rotor height being mostly caused by some
outlier activity, likely due to any event that lead to an increase of insects during this period. It is
observed that activity below rotor height was higher when temperatures reached between 18ºC
and 25ºC, while activity at rotor height reached its peak when temperatures reached 23 to 25ºC
(Figure 22).
In Figure 23 the relationship between bat activity index and illuminated lunar fraction recorded at
the Richards Bay wind energy facility can be observed. A relation between monthly average
activity and average illuminated lunar fraction did not seemed to occur, from the observation of
Figure 23. Nonetheless this environmental variable had a significant positive influence over bat
activity at the study area (p<0.05) (refer to section 3.3.5.2).
74 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 18 – Average activity index (average number of passes/hour) and average air humidity (%) in the
wind energy facility site between May 2012 and April 2013. Bars represent standard error. Analysis
considering the total data collected through all static detectors.
Figure 19 – Average activity index (average number of passes/hour) and average wind speed (m/s) in the
wind energy facility site between May 2012 and April 2013. Analysis considering the total data collected
through all static detectors.
0
20
40
60
80
100
0
20
40
60
80
100M
ay
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Avera
ge h
um
idit
y (
%)
Nu
mb
er
of p
ass
es/
ho
ur
Average number of passes/hour Humidity
0
1
2
3
4
5
6
7
8
9
0
20
40
60
80
100
May
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Avera
ge w
ind
sp
eed
(m
/s)
Nu
mb
er
of p
ass
es/
ho
ur
Average number of passes/hour Wind speed
75 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 20 - Average activity index (average number of bat passes/hour) at rotor height (60m and higher)
and below rotor height (below 60 m) at different wind speed (m/s) in the wind energy facility site between
May 2012 and April 2013. Analysis considering the total data collected through all static detectors.
Figure 21 – Average activity index (number of passes/hour) and average temperature (ºC) in the wind
energy facility site between May 2012 and April 2013. Bars represent standard error. Analysis considering
the total data collected through all static detectors.
0
5
10
15
20
25
30
35
40
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Nu
mb
er
of p
ass
es/
ho
ur
Wind speed (m/s)
Below Rotor Height (7m, 10m, 30m) Rotor Height (60m, 80m)
0
5
10
15
20
25
30
0
20
40
60
80
100
May
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Avera
ge t
em
pera
ture
(ºC
)
Nu
mb
er
of p
ass
es/
ho
ur
Average number of passes/hour Temperature
76 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 22 - Average activity index (average number of bat passes/hour) at rotor height (60m and higher)
and below rotor height (below 60 m) at different air temperatures (ºC) in the wind energy facility site
between May 2012 and April 2013. Analysis considering the total data collected through all static detectors.
Figure 23 – Average activity index (number of passes/hour) and illuminated lunar fraction in the wind
energy facility site between May 2012 and April 2013. Fraction illuminated at New Moon is 0,0 and at Full
Moon is 1,0. Bars represent standard error. Analysis considering the total data collected through all static
detectors.
Considering that wind speed seems to have a significant effect over bat activity, an analysis of the
velocity of the wind to which most species seem to be active is shown in Figure 24. The analysis
0
5
10
15
20
25
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
Nu
mb
er
of p
ass
es/
ho
ur
Temperature (ºC)
Below Rotor Height (7m, 10m, 30m) Rotor Height (60m, 80m)
0
0.2
0.4
0.6
0.8
1
0
20
40
60
80
100
May
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Illu
min
ate
d lu
nar
fracti
on
Nu
mb
er
of p
ass
es/
ho
ur
Average number of passes/hour Illuminated lunar fraction
77 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
of the cumulative activity of the groups of species with high, medium/high and medium height of
flight (as defined in Table 11) shown that about 50% of bat activity occurs up to 6m/s wind speed,
and that approximately 80% of bat activity is detected up to 8m/s wind speed.
Figure 24 - Cumulative bat activity (%) of species with higher altitude flight (HH), medium to higher height
of flight (MH/HH) and medium height of flight (MH) and wind speed (m/s) in the wind energy facility site
between May 2012 and April 2013. Analysis considering the sub-sampling method collected through all
static detectors.
Considering the bat activity only from the static detectors installed at height (top microphone of
RB1 and top detector of RB4) (Figure 25) it is possible to confirm the previous analysis. However
at more than 60 m, bat activity seems to be more conditioned by wind since, in general, 50% of
the activity was observed below 4m/s of wind speed, while 80% of the activity was attained at 7
m/s.
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10 11 12 13 15
Perc
en
tage (
%)
Wind speed (m/s)
HH% MH/HH% MH%
78 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 25 - Cumulative bat activity (%) of species with high altitude flight (HH), medium to high height of
flight (MH/HH) and medium height of flight (MH) and wind speed (m/s) in the wind energy facility site
between May 2012 and April 2013. Analysis considering only the sub-sampling method collected through
static detectors placed at height.
3.3.4. T emp ora l a nal ysi s ( se aso nal ac tiv it y)
3.3.4.1. Manual detection
The number of bat passes detected from manual surveys was generally below 1 pass per grid cell
of 500x500 meters, with the exception of February and during the autumn and winter months,
where the average number of passes per transect exceeded this value (Figure 26). In spite of the
previously referred expectations for the study area (of higher bat activity during spring and
summer months), the peak of activity detected through manual surveys was registered during late
summer and autumn (Figure 26).
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Perc
en
tage (
%)
Wind speed (m/s)
HH% MH/HH% MH%
79 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 26 – Average number of bat passes per grid/cell recorded from May 2012 to May 2013 in the wind
energy facility site. Bars represent standard error. Analysis considering the total data collected through all
vehicle based transects.
3.3.4.2. Static detection
The results obtained through static detection between May 2012 and April 2013 allowed the
analysis of the activity index of bats in the study area (average number of passes per hour) (Figure
27). The average number of passes per hour was higher during late autumn (May 2012), late
winter and early spring (August and September 2012) and summer months (December 2012 and
January 2013), with more than 30 passes per hour, and with less activity in the remaining months.
The maintenance of a moderate level of activity even in winter months can be due to the
proximity of the RB2 detector to a riparian fringe of a water body. These features are very
important for bats to forage any insect prey, therefore these places are considered to be very
likely for higher bat activity (Scott et al., 2010). This higher bat activity than expected for winter,
corresponds to the time frame that the RB2 detector was installed and functioning. Nevertheless,
this is a strong indication that riverine vegetation and water bodies constitute important areas for
the bat community and where higher bat activity should be expected.
This summary of bat activity at the Richards Bay study area seems to indicate that bats are less
active during early winter months, and maintain a rather constant activity during spring, followed
by an increase in activity during the summer months. However statistical analysis has shown a
significant difference between the spring and summer months and the remaining seasons,
indicating that this is an important time of the year for bats (p<0.05). This may be well due to the
0
1
2
3
4
5
May
July
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
May
Autumn Winter Spring Summer Autumn
Nu
mb
er
of
pass
es/
gri
d c
ell
Month/Season
80 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
rain patterns in Richards Bay area, where a summer rain season is characteristic, and bats are
more active during winter, considering there is no great fluctuations in the air temperature
throughout the year. Nonetheless this should be further assessed in future years of monitoring
activities (refer to section 3.3.5.2).
Bat activity indexes in the Richards Bay wind energy facility area were higher in relation to those
recorded by Bio3/Savannah in wind project developments in the Western Cape Province. At
these sites the higher average bat activity index was rarely higher than 20 passes/hour and with a
high seasonal variation, as well as showing a higher activity during the spring and summer months
and lower activity during winter. This is just an indicative comparison, as these values are not
comparable due to the strong dissimilarities between geographical areas and potential bat
diversity.
Figure 27 – Average number of passes/hour recorded between May 2012 and April 2013 at Richards Bay
wind energy facility site. Bars represent standard error. Analysis considering the total data collected
through all static detectors.
Considering the average daily bat activity in the study area, from all the detectors implemented at
the site, some peaks of higher activity can be observed, such as in late April-early May (27 April to
10 May), early to mid to late August (15 to 31 August), and throughout the December and
January months (Figure 28). Considering that these peaks are spread throughout the year, this
may indicate that the study area is used for bats for multiple purposes, not only for feeding during
periods of higher food availability, but possibly for dispersion and/or reproduction as well.
0
20
40
60
80
100
May
June
July
Augu
st
Septe
mber
Oct
ober
Nove
mber
Dece
mber
Januar
y
Febru
ary
Mar
ch
Apri
l
Autumn Winter Spring Summer Autumn
Nu
mb
er
of
pass
es/
ho
ur
81 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 28 – Activity index (average number of passes/hour) in the Richards Bay Wind Energy Facility site between May 2012 and April 2013. Analysis considering the total
data collected through all static detectors.
0
50
100
150
200
250
30027/0
4/2
012
11/0
5/2
012
25/0
5/2
012
08/0
6/2
012
22/0
6/2
012
06/0
7/2
012
20/0
7/2
012
03/0
8/2
012
17/0
8/2
012
31/0
8/2
012
14/0
9/2
012
28/0
9/2
012
12/1
0/2
012
26/1
0/2
012
09/1
1/2
012
23/1
1/2
012
07/1
2/2
012
21/1
2/2
012
04/0
1/2
013
18/0
1/2
013
01/0
2/2
013
15/0
2/2
013
01/0
3/2
013
15/0
3/2
013
29/0
3/2
013
12/0
4/2
013
26/0
4/2
013
Nu
mb
er
of p
ass
es/
ho
ur
82 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
In Figure 29 it is represented the average number of bat passes per hour estimated per each
confirmed species in the study area, per each sampling month. While some species were
observed in almost all months, such as Tadarida aegyptiaca, Neoromicia capensis and Chaerephon
pumilus, others seem to have been detected only in some months. A good example is the
Pipistrellus hesperidus and Scotophilus viridis that were only recorded in August and September; or
Myotis tricolor, only detected between July and September (Figure 29). Miniopterus fraterculus was
detected the most in January and Miniopterus natalensis seem to have been also detected mostly in
September. These three species (Myotis tricolor, Miniopterus fraterculus and Miniopterus natalensis)
are migratory species and tend to move between roosts at the end of winter/beginning of Spring
(from winter to summer roosts), or at the beginning of winter (into winter roosts for
hibernation). The predominance of movements of these species in the study area in the previously
identified months may indicate that the area is used as a passing route, however the data available
is not enough to sustain this hypothesis.
Figure 29 – Average activity index (average bat passes /hour) recorded for each confirmed species
through static detection in the Richards Bay wind energy facility, per month, between May 2012 and April
2013. Analysis considering the total data collected through all static detectors.
Considering that species have different periods of higher activity from sunset to sunrise, it is then
important to analyse which periods have higher activity in order to minimize the impacts of the
0
10
20
30
40
50
60
70
Acti
vit
y in
dex
May June July August September October
November December January February March April
83 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
operating wind energy facility on bats during such periods, if necessary. The average number of
passes per hour recorded within each hour period after sunset, for a period of twelve hours
(while the detectors were recording) between May 2012 and April 2013 is presented in Figure 30.
The high activity periods were observed in the first and second hour after sunset, and then in the
ninth and tenth hour after sunset, each one with more than 40 passes per hour. This pattern has
already been pointed in several studies, where a higher activity in the first two hours after sunset
was found and then decreased until sunrise, especially in open habitats (Meyer et al., 2004;
Brooks, 2009). However, other studies indicate that bat activity may vary greatly during the night,
according with the type of habitat present in the study area, as well as the season (O´Donnel,
2000). In order to evaluate if this pattern is also related with sunrise, Figure 31 shows the average
number of passes per hour, in each hour before sunrise. Comparing both Figure 30 and Figure 31
it can be observed that bat activity between the 2nd and 10th hour after sunset is very similar to
the activity recorded between the 10th and 1st hour before sunrise. This way it is understandable
that the peak of activity recorded in the 9th and 10th hour after sunset is similar to the peak of
activity recorded in the 1st and 2nd hour before sunrise, indicating that this second peak of activity
observed after sunset is related to the proximity of sunrise, as observed by other authors as well
(Hayes, 1997).
Figure 30 – Average number of passes per hour after sunset recorded between May 2012 and April 2013,
in the Richards Bay wind energy facility site. Analysis considering the total data collected through all static
detectors.
0
20
40
60
80
100
120
140
160
180
1 2 3 4 5 6 7 8 9 10 11 12
Nu
mb
er
of
pass
es/
ho
ur
Hours after sunset
84 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 31 - Average number of passes per hour before sunrise recorded between May 2012 and April
2013, in the Richards Bay wind energy facility site. Analysis considering the total data collected through all
static detectors.
Analysing the confirmed species recorded at Richards Bay wind energy facility, between May 2012
and April 2013, and their activity through the night period, it is possible to verify that for the
three Molossid species (Chaerephon pumilus and Tadarida aegyptiaca) two periods of higher activity
seem to emerge: one close to sunset, in the 1st and 2nd hour after sunset; and another close to
sunrise, between the 9th and 12nd hour after sunset (Figure 32). Figure 33 presents a similar
analysis focused on the Vespertilionid species confirmed in the study area. Concerning this
particular group, is not so clear the distinction between the higher bat activities throughout the
night period, being only of notice for N. nana, which presented a peak of activity after the tenth
hour after sunset, very close to sunrise. Considering the few recordings of Rhinolophus capensis
obtained, this species was only detected at the 6th hour after sunset. However this detection does
not indicate a preference of the species for being active at this hour since this may be an event of
chance.
0
20
40
60
80
100
120
13 12 11 10 9 8 7 6 5 4 3 2 1
Nu
mb
er
of
pass
es/
ho
ur
Hours before sunrise
85 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 32 - Average activity index (average number of passes/hour/day) recorded for each confirmed
species of the Molossidae family, in the Richards Bay wind energy facility, per hour after sunset, between
May 2012 and April 2013. Analysis only considering the data analysed for sub- sampling.
0
0.5
1
1.5
2
2.5
1 2 3 4 5 6 7 8 9 10 11 12
Acti
vit
y in
dex
Hours after sunset
Chaerephon pumilus Mops condylurus Tadarida aegyptiaca
86 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 33 - Bat activity index (average number of passes/hour/day) recorded for each confirmed species of the Vespertilionidae and Miniopteridae family, in the
Richards Bay wind energy facility, per hour after sunset, between May 2012 and April 2013. Analysis only considering the data analysed following the sub-
sampling methodology implemented.
0
0.5
1
1.5
2
2.5
3
3.5
1 2 3 4 5 6 7 8 9 10 11 12
Acti
vit
y in
dex
Hours after sunset
Eptesicus hottentotus Miniopterus fraterculus Miniopterus natalensis Myotis tricolor Neoromicia capensis
Neoromicia nana Pipistrellus hesperidus Scotophilus dinganii Scotophilus viridis
87 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3.3.5. S tat ist ica l a nal ysi s
Statistical tests were performed in order to evaluate if environmental variables could influence bat
activity in the study area. The variables tested included air temperature and humidity, wind speed,
illuminated lunar fraction, percentage of each biotope in the area surrounding each static detector
(static detection) or composing each grid cell of 500x500m (manual detection), season of the year
(static detection) and height of the static detector (static detection). This analysis was performed
through GLMM tests and was conducted separately for data from manual surveys and static
surveys.
3.3.5.1. Manual detection
Through the univariate models implemented on to the manual detection results, it was observed
that from the tested variables only the wind speed, air temperature, air humidity and the
humanized areas were significant, since p-value of Wald’s test was superior to 0.2. Therefore, for
the multivariate models only the referred variables were considered. The process of backward
stepwise allowed the selection of all of these variables for the final model (Table 17).
Therefore the model that better explains bat activity in the study area, through manual survey
detection includes 4 out of the 10 variables studied: wind speed, air temperature, air
humidity and humanized areas. However, these variables influence the number of contacts in
different ways:
Variables air temperature, air humidity and humanized areas are positively associated with
the number of bat passes recorded. Therefore the probability to observe more passes
tends to increase with the increase of these variables;
On the other hand the results indicate that the probability of recording more bat basses
is negatively influenced by the increase in the wind speed, existing therefore a negative
influence of this variable over bat activity in the study area.
Table 17 – Generalized Linear Mixed Models results, relating the variables studied with the bat activity
detected through all manual surveys (Vehicle based transects). Significance level (p-value): * p<0.05; **
p<0.01; *** p<0.001.
Response
distribution Coefficients Estimate Std. Error z value Pr(>|z|)
Negative
Binomial
(Intercept) -2.5632 0.6713 -3.8200 0.00013 ***
wind speed -0.2850 0.0388 -7.3500 2.0e-13 ***
air temperature 0.1764 0.0318 5.5500 2.8e-08 ***
air humidity 0.0095 0.0016 5.7800 7.5e-09 ***
humanized areas 3.4240 1.6099 2.1300 0.03343 *
88 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3.3.5.2. Static detection
Through the univariate models implemented to the/on to the manual detection results it was
observed that from the tested variables only the wind speed, air temperature, air humidity,
illuminated lunar fraction, presence of human-made structures, height of the detector and season
had a significant influence over bat activity at Richards Bay site, since p-value of Wald’s test was
superior to 0.2. Therefore for the multivariate only the referred variables were considered. The
process of backward stepwise allowed the selection of all of these variables for the final model
(Table 18).
The model that better explains bat activity in the study area, detected through static detection,
includes 6 out of the 10 variables studied: wind speed, air temperature, lunar fraction,
height of the detector, season and the presence of human-made structures. However
these variables influence the number of contacts in different ways:
On the one hand, variables air temperature and illuminated lunar fraction are positively
associated with the number of passes recorded. Therefore the probability to observe
more passes tends to increase with the increase of these environmental variables (e.g.
nights of higher temperature or with a high fraction of the moon illuminated);
On the other hand, the results indicate that the probability of recording more bat passes
is negatively influenced by the increase in the wind speed, existing therefore a negative
influence of this variables over bat activity in the study area;
The time of the year also proved to have an influence over bat activity since the variable
season had a significant influence over the amount of bat passes recorded. Testing the
four seasons of the year it was possible to assess that autumn presented less activity than
summer and winter, while winter demonstrated less activity than spring and summer.
The results also indicate a positive influence of the presence of human-made structures
over the number of passes recorded. This indicates that the presence of human-made
structures such as farm houses led to a higher amount of bat passes, regarding the other
types of land use evaluated in this analysis. This result may be explained by the utilization
of such structures as roosting locations for many species, which increases the activity
detected throughout the night.
Table 18 – Generalized Linear Mixed Models results, relating the variables studied with the bat activity
detected through all static detector surveys. Significance level (p-value): ** p<0.01; *** p<0.001.
Response
distribution Coefficients Estimate Std. Error z value Pr(>|z|)
Negative
Binomial
(Intercept) -0.345 0.547 -0.63 0.528
Wind speed -0.044 0.006 -7.10 1.2e-12 ***
Air temperature 0.077 0.008 9.69 < 2e-16 ***
Lunar fraction 0.185 0.048 3.78 0.00016 ***
Human-made structures 0.218 0.108 2.01 0.04418 *
Height (Low vs High) 0.196 0.042 4.58 4.6e-06 ***
89 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Response
distribution Coefficients Estimate Std. Error z value Pr(>|z|)
Season (summer vs autumn) 0.162 0.048 3.37 0.00076 ***
Season (winter vs autumn) 0.434 0.071 6.10 1.1e-09 ***
Season (winter vs spring) 0.384 0.060 6.37 1.9e-10 ***
Season (winter vs summer) 0.271 0.075 3.58 0.00034 ***
The results of the models tested for manual and static detection were coincident for most of the
variables tested, indicating that bat activity is mostly influenced by wind speed and air
temperature, among other factors. While air temperature influences bat activity positively, wind
speed revealed a negative influence over the amount of recorded bat passes.
90 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
3.4. U SE OF ROOS TS
Potential Bat roosts within the Richards Bay wind energy facility included buildings (suitable roofs
and cracks in buildings), trees (large densely leaved trees for fruit bats, banana trees for certain
Vesper bats, and crevices in any tree for crevice dweller), water towers, culverts and artificial bat
houses. During the sampling surveys conducted by NSS between May 2012 and April 2013 a total
of 13 confirmed bat roosts were found, providing roosting habitats for at least 8 bat species. 3 of
the 13 confirmed roosts were possibly occupied by Wahlberg's epauletted fruit bat (Epomophorus
wahlbergi), and it is likely that this species uses the area with some intensity, at least for roosting.
These roosts were found at different portions of the wind energy facility: Wilton Park (EW1),
Pension Farm (EW2) and Scott’s Properties (EW3) (Figure 34). During February and April
surveys, roosts were revisited, allowing the observation of some changes in bat abundance: the
roosting location found at Wilton Park (EW1) (in the southern part of the study area) housed a
higher number of individuals in April (55 bats of Epomophorus wahlbergi) than in February (24 bats
of Epomophorus wahlbergi). In February surveys pups were also observed, indicating that
reproduction of this species occurs at the wind energy facility site.
Considering the importance of this location (EW1) for the life cycle of this particular species, it is
important to assess whether or not this species could be affected by the implementation of the
wind energy facility. In Figure 34 it is represented the closest wind turbine (WTG48), located at
approximately 780 meters from this roost. Epomophorus wahlbergi is considered a “Least
Concern” conservation species according to the South Africa Red List and it is a common and
abundant species in South Africa, although with a medium to high risk of collision with wind
turbines. Therefore, any potential impacts over this species, particularly its conservation status,
are not expected to significantly affect the population to a greater scale. Nevertheless, the impacts
of wind energy facilities on these species, especially in South Africa, is still largely unknown and
therefore precautionary measures should be implemented and these potential impacts should be
closely monitored during the next phases of the project, with particular relevance to the
construction phase.
Little free-tailed bat (Chaerephon pumilus) was also found in two other roost locations, near
Dover (CP2) and Ukulu (CP1) properties. The remaining species found in the study area were the
Sundevall's leaf-nosed bat (Hipposideros caffer), Banana bat (Neoromicia nana), Schlieffen's twilight
bat (Nycticeinops schlieffeni), Dusky pipistrelle (Pipistrellus hesperidus) and Yellow-bellied house bat
(Scotophilus dinganii). These species were observed in the Ukulu Properties (HC2, SD1, NN2) and
in the Reding Dam (NS1) and Scott’s Properties (HC1) (Figure 34). All those species are known
to roost in crevices of houses and building roofs, with the exception of the Hipposideros caffer,
which is usually associated with cavities in mines and caves (Monadjem et al. 2010). However a
colony of 35 Hipposideros caffer individuals was found occupying the roof and two bathrooms of
an abandoned house (HC2) (Photograph 3).
91 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Photograph 3 – Individuals of Hipposideros caffer found at the bathroom of an abandoned house at Ukulu
Properties (HC2), during the February survey at Richards Bay wind energy facility site.
Although no reproduction of this species was observed during the February survey, during which
this roost (at Ukulu properties – HC2) was detected, the amount of guano present and the high
number of individuals observed may suggest that this is an important daytime roost for this
species in the study area. Analysing this location (HC2) and the proposed wind turbine
coordinates, it is noticeable that this roost is surrounded by the following wind turbines: WTG17,
WTG13 and WTG14. These wind turbines are located at a variable distance from the roost, at a
minimum distance of 700 meters and a maximum of 1000 meters. Hipposideros caffer is “Data
Deficient”, according to South Africa Red List conservation Status and has a Low risk of collision
with wind turbines. Therefore, it is not expected that the potential effects of potential mortality
due to the wind energy facility over this species may cause a significant negative effect over the
population at a larger scale. Nevertheless, these effects should be carefully monitored.
2 other roosts were also found with individuals from different species, i.e.: Sundevall's leaf-nosed
bat (Hipposideros caffer), Little free-tailed bat and Schlieffen's twilight bat, all cohabiting in the same
roost in Ukulu Properties (HC1); and Mauritian tomb bat (Taphozous mauritianus) and Egyptian
silt-faced bat (Nycteris thebaica), at Shayamoya (TM1). These 2 later species are considered to
have a high likelihood of occurring in the study area but were not confirmed through ultra-sound
detection so far.
It is of note that all the species identified in these roosts have “Least Concern” conservation
status according to the South Africa Red List, and most of them have low to medium probability
of collision with wind turbines (Sowler and Stoffberg, 2012). However, 3 of the species present in
those roosts which are located less than 1000 m from a proposed wind turbine location have
medium to high risk of collision (Epomophorus wahlbergi, Scotophilus dinganii, and Neoromicia nana)
while 1 (Chaerephon pumilus) has high risk of collision with wind turbines. These species were
92 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
identified in the roosts located in the vicinities of wind turbines WTG13, WTG14, WTG17,
WTG40, WTG13, WTG45 and WTG48.
However the presence of wind turbines within a distance of at least 1000 meters from roosts
may not necessarily mean an increase in the risk of bat fatality with wind turbines, since when bat
species forage they tend to look for areas far from their daytime roosts (Brigham, 1991; Heithaus
and Fleming, 1978). Therefore, the major impact that Richards Bay wind energy facility
implementation may have on the bat population in the study area should mostly affect bat roosts
that are used as maternity locations. For Richards Bay study area, the roost located 780 meters
west from the wind turbine WTG48 was found under these conditions. Maternity roosts are
more prone to disturbance or displacement, since bats need certain conditions of temperature
and humidity to reproduce. Therefore, the alteration of any of these parameters may cause the
roost utilization disruption for reproduction. In order to prevent the disturbance of this location,
buffer zones may be used as a protection measure. Studies differ about the buffer radius to be
used, since it may require a specific adjustment depending on the desirable roost protection level
(Jenkins et al., 1998; Ellision et al., 2004). For this particular situation, a 500 m protection buffer
around the farm watch office where Epomophorus wahlbergi was found to reproduce (Wilton Park)
is recommended, with the implementation of some restrictions for the use of the area within this
buffer being suggested in chapter 7.2 (Ellision et al., 2004).
These results do not exclude the possibility of other roosts being present within the site that
have not been discovered yet. Many bats, especially tree-dwelling bats change their roosts
regularly. Many bat species utilize two roosts during a 24 hour cycle, a day roost where they sleep
for all daylight hours, and a night roost, which they use to rest in between foraging flights or to
stop to eat a large insect prey.
93 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 34 – Location of the roosts identified during pre-construction monitoring field surveys at the
Richards Bay wind energy facility site.
94 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
4. P O T E N T I A L S E N S I T I V E A R E A S O N T H E W I N D E N E R G Y F A C I L I T Y
Considering bat activity registered to date in the area, the biotopes present and the number of
species confirmed, it is considered that Richards Bay wind energy facility area is of medium to high
sensitivity to bats in general, particularly due to the northern section of the site falling within the
Endangered Zululand Coastal Thornveld, and where bat activity proved to be higher. This
consideration is in line to the outcome of the EIA report (CES, 2012).
The rivers, water bodies (such as dams) riverine and native vegetation areas were considered as
important areas for bats, though statistical analysis did not show a significant relationship between
this type of vegetation and a higher bat activity (Figure 35). Nevertheless, higher bat activity was
recorded in the static detectors installed close to these features. Vehicle based transects also
recorded higher bat activity in such areas as well. According to bibliographic references rivers and
water bodies, and associated riverine vegetation, are important landmarks for bat orientation
(Serra-Cobo et al., 2000) and preferential locations for bat feeding due to the abundance of
insects in the surroundings (Loyd et al., 2006; Scott et al., 2009; Hagen and Sabo, 2012), especially
emergent insects, such as Diptera and some Lepidoptera, important for many bat species diet. In
the case of the species present or possibly present in the study area many are associated with
riparian vegetation, such as Kerivoula lanosa, Nycticeinops shlieffeni, Neoromicia nana, or Pipistrellus
hesperidus, among other species.
The study area has several rivers of some dimension and with well developed vegetation, such as
the Okula and Mvuzane rivers. In general this type of feature is mostly located in the northern
part of the wind energy facility site, where a greater variety of biotopes is also present, including
several patches of native vegetation. According with recent recommendations from the South
African Bat Assessment Advisory Panel (SABAAP), it was defined a minimum buffer of 200m
around all water features considered to be of importance to bats in the study area, namely the
ones with well-developed riverine vegetation (Figure 35). The most relevant riverine fringes in the
area were identified and were included as sensitive areas for bats, wherein implementation of
wind turbines should be avoided.
As previously mentioned, within the wind energy facility site, mostly in the northern and central
areas, there are some native vegetation patches that include forest vegetation as well as bush
areas. It’s in this type of habitat where bats usually forage, preferentially in forest openings, or
along the vertical or horizontal edge created by the junction of forest stands with other type of
vegetation, roads, lakes and ponds. Therefore, the forest areas were also considered as a sensitive
area for bats, with a 200 m buffer surrounding this feature (Figure 35).
The spatial-temporal analysis results (from both manual and static surveys) showed that there are
some localized areas of high bat activity, with a higher average number of passes per unit of time
in relation to the utilization of the remaining area (for this analysis were considered areas of high
activity as superior to 30 bat passes - Figure 7 and Figure 8). This consideration is motivated by
the difference in the activity index observed in this study, when compared with other study areas
throughout South Africa: in Richards Bay more than 30 bat passes per hour at a single static
95 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
detection location were recorded, while in other wind energy facility sites being monitored in
other areas of the country, activity indexes were lower – on average between 10 and 20
passes/hour in the Western and Northern Cape Provinces. These site areas with higher activity
are easily observed in Figure 35. Most of these areas of high activity are located in the vicinities of
confirmed identified roosts.
Considering the confirmed roosts previously referred to in chapter 3.4, buffers of 500 meters
were established since most of the roosts belonged to a small number of individuals from Least
Concern bat species. This distance took in consideration the recent recommendation from the
SABAAP workshop results, and studies that indicate that the environmental features present
within a 1.5 km of a roost are important to determine its occupation and that the alteration of
the surrounding features of the roost at a minimum distance of 500 m may cause bats to abandon
the roost, or alternatively that habitat management may be implemented within this radius to have
better effects (Jenkins et al., 1998). Following the recommendations of the SABAAP it was also
considered a buffer area of 1000 m around the only reproduction roost found in the study area,
where more than 50 individuals of Epomophorus wahlbergi were found roosting (Wilton Park).
These buffer areas should also be considered as indicative of highly sensitive areas that should not
be disturbed in order to prevent disruptions in two of the most sensitive events for bats:
hibernation or reproduction. However it is of note that these buffer areas do not indicate that
wind turbines cannot be placed within these areas, but rather that disturbance should be avoided
and minimized within. Nevertheless, whenever possible and technically viable it would be
preferable if wind turbines should not be implemented within these areas.
Considering these statements, in general the study area is considered from medium sensitivity in
the southern part, to high sensitivity in the northern part, in terms of potential bat collision risk
with wind turbines, due to the previously referred reasons (Figure 35). The northern area is
characterized by the presence of several patches of native vegetation, streams with associated
vegetation, areas of high bat activity detected during monitoring and a large amount of roosts of
“Least Concern” species which contributes to the classification of the broader northern area as
highly sensitive. In the south part of the study area are also present areas of high bat activity
detected, as well as roosts of “Least Concern” species and a maternity roost. The presence of
these features, associated with the absence of sensitive vegetation with importance for bats
contributed to the classification of this area as medium sensitive. Nonetheless there are areas
considered more sensitive than others, as not all of the area is of sensitive character for bats. The
area of sugar cane plantation is not considered to be specifically associated with higher bat activity
and if no roosts or buffer areas of native or riverside plantation are present, area of sugar cane
plantation should not be considered as an area of high sensitivity. This results in a matrix of
sensitive and non-sensitive (or less sensitive) areas both in the north and south part of the study
area, where turbines are proposed and should not be considered as a relevant problem for the
bat community.
For an optimal turbine layout from the bat impact point of view, any of the above mentioned
buffer areas (Figure 35) for the bat communities should be avoided. Disturbance within the
considered sensitive areas should be avoided both during construction and operation phase of the
96 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
development. The following information is relevant in terms of the final layout proposed for the
Richards Bay wind energy facility:
13 of the 39 wind turbines are located within areas considered to be of higher sensitivity
for bats as for example within areas of higher activity (average activity index above 30
passes/hour), near native or riverside vegetation, or inside buffer areas from roosts:
WTG03; WTG04, WTGnew4, WTG08, WTG09, WTGnew5, WGT10, WTG11,
WTG15, WTG18, WTG25, WGT34 and WTG48 (Figure 35);
o From these 13 wind turbines, 12 are proposed to be sited in areas of riverine
vegetation or native vegetation and its relocation to lower sensitive areas should
be considered, in order to minimize the potential impacts over bats but also in
order to preserve some of the few natural patches of vegetation still not
transformed in the study area: WTG03; WTG04, WTGnew4, WTG08, WTG09,
WTGnew5, WGT10, WTG11, WTG15, WTG18, WTG25 and WGT34.
o If possible wind turbine WTG48 should be relocated southwest in order to avoid
the 1000m buffer. This buffer is also dependent on the maintenance of the
reproduction roosting site by E. wahlbergi in the future. So, this roosting site
should be closely monitored in order to evaluate the continuity and dynamics of
its occupation. If it’s verified that the roost is abandoned, or not used for
reproduction purposes, there is no reason not to install this wind turbine, as the
protection buffer should be reduced to 500 m.
Mitigation measures are proposed to minimize the potential impacts foreseen.
The remaining wind turbines are not located within areas foreseen to be sensitive for bats
and are not expected to represent a potential high collision risk for bats. This should be
however validated during the monitoring programme to be conducted during the
following phases of the project.
In order to minimize potential impacts from the operational phase of the project (see section 0)
some recommendations are proposed in chapter 7.2. Note that these are to be considered only
as recommendations and should only be taken into consideration if possible and technically viable.
97 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 35 – Sensitive areas within the Richards Bay wind energy facility site.
4.1. T URBIN E SEN SITIV ITY AN ALYSI S
The analysis of the wind turbines location in relation to the sensitive areas presented in the
previous section (4) is presented in Appendix VIII. Additionally to the turbines identified in section
4 as being sited within the sensitive areas, this analysis indicated other wind turbines that may
have cause potential impacts during the operational phase of the wind energy facility for being
located at a distance lower than 500 m from a sensitive area. Considering that bats are mobile
and may fly large distances to forage, an analysis of the location of the wind turbine without
considering a broader area may not be sufficient to predict potential impacts. For this reason the
analysis presented below always considers a 500 m buffer area surrounding each turbine, and the
sensitive features identified are located within that surrounding area.
As a result of the turbine impact assessment three groups of turbines with possible impacts over
sensitive areas were identified, as well as an additional group of turbines not located within any
sensitive area:
98 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Roosting sensitive turbines: six wind turbines were identified in this group and are
proposed within the broader area of a roost – WTG 40, 41, 43, 45, 47 and 48. From
these six wind turbines, three were located within the 500 m buffer distance from a least
important roost – WTG 40, 43 and 45 – while the remaining three are located at less
than 1000 m from a confirmed reproduction roost – WTG 41, 47 and 48. No roosts
were located within the broader area of any wind turbine (less than 500 m). From these
turbines, the WTG 41, 47 and 48 are considered to have the higher potential for
negative impacts on bats due to the short distance to a reproduction roost.
Habitat sensitive turbines: 19 wind turbines are proposed within the broader area of
a relevant habitat related feature (e.g. the 200 m buffer surrounding an important bat
area, native vegetation and/or riverine vegetation) – WTG 1, new4, 8, 9, 10, 11, new5, 12,
13, 15, 16, 18, 21, 22, 25, 32, 33, 34 and new1. From these 19 turbines some are
considered to be more likely to cause impacts on bats than others. The turbines
considered more likely to cause significant impacts are the ones that are located in areas
that present both a high activity and are close to areas of riverine vegetation and/or native
vegetation. Since bats tend to prefer this type of areas for forage it will be likely that bats
in the study area are more active surrounding native and/or riverine vegetation, so
turbines placed within these may present a higher risk of impact – WTG new4, 8, 9, 10,
new5, 15, 16, 34.
Roosting and Habitat sensitive turbines: 6 of the proposed wind turbines are
proposed within the proximities of roosts and of riverine and/or native vegetation areas –
WTG 3, 4, 14, 17, 26 and 31. These turbines are more likely to cause impacts than the
remaining ones, since they are close to multiple types of important areas for bats: areas of
roosting and areas of foraging. Among these turbines, three are considered to be more
likely to present impacts on bats during the operational phase of the project, since they
are located among or close to high activity areas – WTG 3, 26 and 31.
Non sensitive turbines: from the 39 turbines of the proposed layout 8 are not
proposed within any of the evaluated sensitive areas – WTG new3, 20, new2, 36, 38, 42,
new6 and 46. This indicates that these locations may present a lower potential to cause
impacts on bats. However any installed within an area that is known to have bat activity is
not free of impacts, so these turbines are not considered as “zero impacts”, but rather as
lower potential impacts prone. Though these 8 turbines are not proposed within any of
the important areas for bats considered in the previous evaluation (section 4) they are
located within sugar cane plantation areas, which are expected to favor the occurrence of
foraging bats due to the presence of insects associated with this type of agriculture.
99 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5. P O T E N T I A L I M P A C T S A S S E S S M E N T
Considering the species with potential occurrence at the Richards Bay wind energy facility, it is
important to conduct an analysis of the main potential effects that the construction and operation
of this development may have on the bat community study area. Assuming that bat monitoring
may not detect all bat species present in the study area in the first year of surveys, this analysis
allows the prediction of the impacts to be considered in future assessments.
5.1. I NTERA CTION S BET WEEN WIND ENE RGY FACILIT IES A ND B ATS
The South African experience of wind energy generation has been extremely limited until recently
since only three small wind energy facilities were installed in the country. To date only eight wind
turbines have been constructed, 3 at a demonstration facility at Klipheuwel in the Western Cape,
4 at a site near Darling, and 1 at Coega near Port Elizabeth. Currently there are other turbines
under construction but none of these wind farms are currently operational. Considering this fact
it is necessary to review and understand the possible interactions between bats and wind energy,
in order to clarify a number of issues involved in the assessment process. Much of the available
literature about this issue is international, mostly from the United States, European Union,
Australia and Canada, where the wind power generation is a more established industry.
While wind energy facilities provide a clean source of energy without long term impacts on the
planet, unlike fossil fuels, the existence of impacts over faunal resources was detected not long
after its first implementation (Kikuchi, 2008; Eichhorn and Drechsler, 2010). Bats are, however,
considered as one of the most affected groups with the implementation of wind energy facilities,
from high bat fatalities detected throughout North America and Europe (Johnson et al., 2003;
Fiedler, 2004; Barclay et al., 2007; Arnett et al. 2008; Arnett et al., 2011; EUROBATS, 2013; Hein
et al., 2013). International literature review and specialist expertise have suggested that the
impacts that wind energy facilities have on bat species often result in fatalities, either caused by
direct collision with the turbine tower, collision with rotation blades or barotrauma13 (Kunz et al.,
2007; Cryan and Barclay, 2009). Impact on natural populations may also be caused by the
disturbance effect, barrier effects and habitat loss (Cryan & Barclay, 2006). These impacts,
especially mortality, have become a source of major concern among a number of stakeholder
groups (Estep, 1989; Erickson et al., 2002). Results obtained during several international
monitoring studies indicated that wind farms were responsible for the decrease in population of
some species’ (Hunt, 2002; Carrete et al., 2009) although many other studies revealed that these
13 Barotrauma is used in the present report referring to bat deaths due to tissue damage to air- containing structures
caused by rapid or excessive pressure change close to the rotating wind turbine blades surface. Death is usually caused
by pulmonary barotrauma where lungs are damaged due to expansion of air in the lungs that is not accommodated by
exhalation (Baerwald et al., 2008).
100 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
impacts were not important when compared to those originating from other man-made
infrastructures (Erickson et al., 2001). Nevertheless, the potential for wind farms to affect bat
populations should not be underestimated (Hunt, 2002).
However, not much research has been conducted on these matters in South Africa until recently.
Research about seasonal and daily movement patterns of bat species and what the potential
impacts of the development of multiple wind energy facilities and thousands of turbines across the
country might be has been lacking and has begun only recently. Also, information regarding bat
distribution, seasonal and daily movements and migration is very limited for South African bat
communities. Most of the established general principles that have been learnt internationally can,
to a certain extent, be applied in South Africa with care to adapt existing international knowledge
to local bat species and conditions.
The Guidelines for Surveying Bats in Wind Farm Developments (Sowler & Stoffberg, 2012) were
developed in collaboration with the Endangered Wildlife Trust (EWT). These guidelines provide
technical guidance for consultants to carry out impact assessments and monitoring programmes
for proposed wind energy facilities, in order to ensure that pre-construction monitoring surveys
produce the required level of detail for authorities reviewing environmental authorisation
applications. These guidelines outline basic standards of best practice and highlight specific
considerations relating to the pre-construction monitoring of proposed wind energy facility sites
in relation to bats.
Considering the pre-construction monitoring programme results it is considered that the main
impacts on bats which may arise from the implementation of Richards Bay wind energy facility can
be:
- Bat fatalities by collision with wind turbines, turbine blades or barotrauma;
- Bat displacement and disturbance due to destruction of feeding areas and/or destruction
of roosts;;
- Mortality of frugivorous bat species due to collision with power lines;
- Reduction of ecosystem services provided by bats;
- Cumulative impacts.
The consequences of bat fatalities or bat species displacement of the study area are beyond the
simple impact on bat populations. Bats provide important services for the human population, such
as food, guano for fertilization, and through arthropod suppression (pest control), forest
regeneration and maintenance via seed dispersal and pollination of a wide variety of plants (Kunz
et al., 2011). Richards Bay wind energy facility is to be established in an area that is strongly
influenced by agricultural practices, such as sugarcane cultivation. Bats may be providing an
important ecological service by consuming insects that may become sugarcane pests (Bakker,
1999; Rutherford and Conlong, 2010). Bats prey on many different species, being able to consume
large quantities of lepidoptera, coleoptera, diptera, homoptera and hemiptera, consuming several
times their own weight per night (Kunz et al., 2011). If this control by predation was to be
influenced by the decline of an important part of the bat population, potential problems (either
ecological and economical) concerning agricultural pests may arise, requiring further investment
on pest control measures.
101 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5.1.1. C oll isi on wit h t urb in es
A number of factors influence the number of bats killed at wind energy facilities. These can be
classified into three broad groupings:
- Bat related variable;
- Site related variables;
- Facility related variables.
5.1.1.1. Bat related variab les
Several hypotheses have been suggested by international research, regarding the possible causes
that lead bats to collide with wind turbines: migration over long distances may be one aspect of
bat biology that influences collisions; bat attraction to wind turbines, through sound, lights or
movement or bat attraction to wind turbines as roosting locations (Cryan & Barclay, 2009).
These hypotheses are supported by the observation of migrant species flying higher above the
ground than other bats and high-flying migrants being less likely to echolocate and detect spinning
turbine blades, due to the absence of predicted obstacles (Barclay et al., 2007; Kunz et al., 2007b).
There have also been recorded greater fatality rates of bats at taller turbines in North America
(Barclay et al., 2007), during fall migration (Johnson et al., 2003). Studies have revealed greater
echolocation activity of migratory species higher above the ground compared with other species
(Baerwald & Barclay, 2009). This higher tendency of migrant bats to collide with wind turbines
have been hypothesised to be due to migrating bats flying at approximately 65 m and that turbines
of 65 m or taller are located within that area (Barclay et al., 2007).
Bats may also be attracted to the sights, sounds, or movements of wind turbines, since some
studies have shown bats possibly being attracted to the ‘‘swishing sound’’ of sticks waved through
the air (Barbour & Davis, 1969 in Cryan & Barclay, 2009), or unknown cues at roosts previously
used by congeners (Constantine, 1958 in Cryan & Barclay, 2009; Downes, 1964 in Cryan &
Barclay, 2009), as well as thermal images of bats apparently chasing moving turbine blades (Horn
et al., 2008).
Hypotheses involving attraction of bats to turbines as roosts seem plausible considering that the
species of bats killed most often by wind turbines tend to rely on trees as their primary natural
roost structures. Many species of bats favour taller trees as roosts (Kalcounis-Rüppell et al., 2005)
and fatalities at turbines appear to be correlated with turbine height (Barclay et al. 2007).
Analysing the collision risk determined by Sowler and Stoffberg (2012) for the 35 species with
potential occurrence at the site (Figure 36) it is noticeable that half of these species (17 species)
have low or medium risk of collision with wind turbines, due to their flight and foraging
behaviour. In this group of species with low or medium risk of collision are included most of the
species with conservation status as Critically Endangered, Endangered, Vulnerable or Near
Threatened which may potentially occur at the site.
102 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
From the 18 remaining species, 13 have medium to high risk of collision with wind turbines since
they have medium to high flight pattern and they are in general clutter-edge foragers, being
possible that these species use the area surrounding the moving blades to forage (particularly if
turbines are implemented in areas with tall vegetation), increasing the possibility of mortality by
collision or barotrauma. Nevertheless, the species confirmed at the site considered to have a low
to medium potential collision risk were mostly detected in the detectors installed at low heights.
The remaining 5 species have high risk of collision with wind turbines - Tadarida aegyptiaca, Mops
condylurus, Chaerephon pumilus, Taphozous mauritianus and Taphozous perforatus. Taphozous
perforatus are not likely to occur in the study area. It is therefore expected that the remaining
species (that are highly likely to occur) will be potentially the most affected ones with the
implementation of this project. These species are particularly prone to collision due to their flight
characteristics as open-air foragers, that allow them to fly at high altitudes and enter the rotor
swept area, increasing the probability of collision. However, all of these mentioned species
present a “Least Concern” conservation status in South Africa and are both widespread and
abundant species. The species confirmed on site, and with a potential high collision risk, were in
fact registered mostly in the detectors installed at higher heights confirming its higher potential
collision risk with wind turbines.
Figure 36 – Representation of the number of species (with potential occurrence in the Richards Bay wind
energy facility) in each category of collision risk (Sowler and Stoffberg, 2012).
Concerning wind energy facility impact over bats, and their fatalities record, it is important to
analyse the species present in the study area, and their predicted risk of collision with any of the
wind turbines. A recent study has detected the first events of bat mortality in South Africa due to
wind turbines operation, with several bat fatalities of two species being discovered: Neoromicia
capensis and Tadarida aegyptiaca (Doty and Martin, 2013; Aronson et al., 2013). Considering the
results of fatality records in Europe and North America, most of the species affected by wind
energy infrastructures are Tadarida sp., Pipistrellus sp., Myotis sp. and Miniopterus sp. (Figure 37;
Appendix IV). The occurrence of 20 species was confirmed in the study area (echolocating and
non-echolocating species), including both species that have already been found dead in a wind
103 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
facility monitored at Eastern Cape: Neoromicia capensis and Tadarida aegyptiaca (Figure 37). For
these 2 species there are also records of fatality events in wind energy facilities in several
countries of Europe for the same genus (Tadarida sp.) or similar genus (Pipistrellus sp. is
ecologically and morphologically similar to Neoromicia sp.) (Figure 37) (EUROBATS, 2013).
Though no records of fatalities have yet been observed for South Africa of Banana bat (Neoromicia
nana) or Dusky pipistrelle (Pipistrellus hesperidus), they are considered to be as potentially affected
as Neoromicia capensis since these three species have similar flight patterns and foraging habits.
Another 3 of the species confirmed at Richards Bay wind energy facility were Eptesicus hottentotus,
Miniopterus fraterculus and Miniopterus natalensis. No records of bat fatalities of these species at
wind energy facilities in South Africa have been recorded to date. However, fatalities of species of
the Eptesicus genus have been detected in wind facilities in Europe and North America (Figure 37)
(EUROBATS, 2013; Arnett et al., 2008, 2013), and of Miniopterus genus in wind energy facility in
Europe (EUROBATS, 2013).
The fact that these or similar species have already been found dead in other wind energy facilities
around the world, indicate the probability of having the same behaviour in South Africa wind
energy facilities. It is likely that the collision with wind turbines is high, being likely as well for the
species with confirmed occurrence in the study area to collide with the proposed wind turbines
on Richards Bay wind energy facility site.
0
10
20
30
40
50
60
% o
f b
ats
kille
d
Fatality in South Africa Fatality in Europe Fatality in North America
104 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 37 – Confirmed bat species in the study area with recorded fatalities in Wind Energy Facilities in
South Africa (Doty and Martin, 2013), Europe – for species with the same or similar genus (EUROBATS,
2013) and North America – for species with the same or similar genus (Arnett et al., 2008).*species with
high risk of collision according to Sowler and Stoffberg (2012).
5.1.1.2. Site variables
Linear landscape features, such as natural forest edges, serve as orientation features along which
foraging and commuting bats may regularly travel (Kunz et al., 2007b; Verboom & Huitema, 1997;
Serra-Cobo et al., 2000). Land cleared during the construction of access roads, turbine
foundations, and power transmission lines might attract bats by mimicking these natural linear
landscape features. Some studies indicate that more bat fatalities are expected to occur at
turbines nearer forest edges, newly created clearings, and other linear landscape features. Thus,
bats that migrate, commute, or forage along linear landscapes (Limpens & Kapteyn, 1991;
Verboom & Spoelstra, 1999; Menzel et al., 2005) may be at increased risk of encountering and
being killed by wind turbines (Kunz et al., 2007b).
5.1.1.3. Facility related v ariables
Considering the wind energy facilities components the turbine characteristics may have influence
over the expected impacts on bats. Studies have shown that the rotor-swept area may not be as
significant as the turbine tower height (Barclay et al., 2007). Bat fatalities are expected to increase
exponentially as turbine height increases, with turbine towers 65 m or taller having the highest
fatality rates. Small rotors may cause high bat fatalities if they are mounted on tall towers (Barclay
et al., 2007). Nevertheless, these studies were based on the older generations wind turbines and
didn’t consider the new generation turbines that are significantly taller.
Considering that the final characteristics of the wind turbines to be installed at Richards Bay wind
energy facility are not yet defined, it is also important to analyse if the identified potential negative
effects over bat communities in the study area may be different for different scenarios. As
mentioned in section 1.4, NEA Renewable Energy is currently evaluating the implementation of
turbines with a hub height ranging between 95 and 120 meters and rotor diameter ranging
between 90 and 120 m.
In spite of the scenarios considered, only bats that fly within rotor swept area are expected to be
affected by the operation of wind turbines (Barclay et al., 2007). Assuming that no turbines will be
implemented within forested or tall shrub surroundings, where clutter and clutter edge bats may
forage and also be affected by turbine blades is considered that only bats that fly at higher heights
should be considered for this impact analysis. In the study area 3 species that have high altitude
flight were confirmed: Chaerephon pumilus, Mops condylurus and Tadarida aegyptiaca. Additionally,
migratory species such as Miniopterus natalensis and Myotis tricolor, also confirmed in the study
area, can fly above 60 m as well (Barclay et al., 2007). Therefore, a scenario where blade tips
sweep the space below 60 m it is considered to cause similar impacts for bat species flying at low
height or high altitude. Regarding species with a higher likelihood of flying at rotor swept area, the
possible scenario will be: the lower tips of the blades are at a higher altitude and/or the rotor
105 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
swept area is smaller. This scenario is considered to have less impact, since the probability of
collision within a potential smaller impact area is expected to be lower.
The pre-construction bat monitoring programme indicate a significant negative relation between
wind speed and bat activity at the rotor swept area, with an approximate 60% decrease of bat
activity for wind speed above 3m/s. Therefore, mortality impact at these height ranges is expected
to be lower at wind speeds higher than 3m/s. Implementation of mitigation measures, such as
feathering, will prevent any potential impacts due to the free rotating blades at wind speeds
bellow 3m/s (refer to section 7.2).
This potential impact significance is highly dependent on the species affected and the areas where
the turbines will be sited. Therefore, providing the mitigation measures proposed (refer to
section 7.2) are implemented, the potential impact significance is considered as moderate at this
stage. These potential impacts are expected to occur only after the wind energy facility becomes
operational. Therefore, the implementation of an operational bat monitoring programme is
considered essential to determine the extent of this impact on bat populations, validation and
adjustment of the proposed mitigation measures will be regarded and if necessary additional
mitigation measures will be considered. The significance and magnitude of such impact is expected
to be higher if endangered or small population species are affected and will dependent on several
factors, such as: the size and extent of the impact on the local and regional populations and
cumulative impacts as well. Once there is not much information regarding these variables for the
South African bat species the monitoring programme continuation should contribute also to a
better understanding of bat population’s dynamics in the area.
5.1.1.4. Habitat loss – des truction, dis turban ce and Bat
displacement from feeding areas
Considering the activity detected at Richards Bay wind energy facility, and the assumption that
this may represent an important area for foraging and feeding activities, it is important to assess
other potential effects over bats, namely if the proposed placement of wind turbines is expected
to affect foraging and feeding behaviours of bats resulting from reduction of areas available for
feeding or increasing the risk of bat fatality while foraging. Bats forage mostly in areas of riverine
or native vegetation, being however registered several areas of high bat activity throughout the
study area, revealing that the sugarcane is also important for this purpose. The analysis of turbine
placement, in relation to the vegetation present and areas of high bat activity is presented in
chapter 5.2. Nevertheless, it is considered that this is a likely impact and will depend mostly on
the areas affected during the construction phase that will be occupied by the turbines. It is
considered of low to medium significance at this stage provided the proposed mitigation measures
are implemented (refer to section 7.2).
The implementation of an adequate monitoring protocol, including the assessment at a similar and
suitable control area, will contribute to determining the extent of this impact.
This potential impact will occur during the construction phase and through the operational life of
the wind energy facility.
106 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5.1.1.5. Disturbance and/or destruction of roo sts
It is also possible to generate impacts over bat populations by affecting existing roosts, such as:
temporary roosts, for daytime use, or more importantly roosts, for reproduction or hibernation
that have an important role in bats life cycle. One reproduction roost was identified at Richards
Bay study area and several other daytime roosts of at least 8 species were also identified by NSS
field team throughout the study area. The reproduction roost identified belongs to a fruit-bat
species, Epomophorus wahlbergi, which is a species that despite of its least concern conservation
status has a medium to high risk of collision with wind turbines due to its flight behaviour.
Individuals may travel large distances to forage therefore increasing the probability to encounter a
wind turbine in their path. Considering that this species is a relatively slow breeder, with only one
to two pups at a time, and a gestation period of five to six months, impacts that may disturb
breeding places may have severe impacts over the species population at the study area. In order
to avoid the disturbance of bat populations in the study area these locations have to be properly
identified and secured, so that the facility construction work does not have impact over these
structures. Further analysis over the prevention recommendations on this subject are detailed in
chapter 5.2 and 7.2. Provided that mitigation measures proposed at this stage are implemented,
this potential impact should be considered of low significance. The continuation of the monitoring
programme in the area will contribute to determining: 1) the importance of the identified roosts
for the bat community; 2) identify other roosts that potentially exist in the area; 3) determine the
effects of the development on the already known roosting locations.
5.1.2. M ort ali ty o f fru giv oro us b at sp ec ies b y co lli sio n
w ith po wer li nes
There are no known mortalities associated with the collision of insectivorous bats with power
lines in Europe or elsewhere in the world (EUROBATS, 2008). However, fruit bat species were
confirmed on the site and there are evidences of mortality of fruit bats with collision or
electrocution with power lines in Australia (http://www.bendigoadvertiser.com.au;
http://bats.org.au/uploads/about-bats/batcarebrisbane_electrocution1.pdf). There are no known
evidences of fruit bat mortality associated with power lines in South Africa.
A short power line section (less than 500m length) is proposed for the south portion of the site
posing a potential impact on the fruit bats at the site, which have a maternity roost also in the
south part of the site. However, there are already several existing power lines (mainly high
voltage and transport lines) in the site vicinities, surrounding the site itself. It is therefore
considered residual the potential negative effect of the installation of such short extension of
power line on the fruit bat species.
However the possibility of fruit bats colliding with wind turbines is a subject still little known,
being therefore an issue that should be taken into consideration during the continuation of the
bat monitoring programme.
107 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5.1.3. R edu cti on of eco sys te m s erv ice s p ro vid ed by bat s
Richards Bay wind energy facility is located within an area mostly occupied by sugar cane
cultivation. As bats play an important role in controlling pests, by consuming several times their
weight on arthropods per night, the maintenance of a healthy ecosystem will provide lower losses
in crop yield and reduce pesticide use by farmers (Kalka et al., 2008; Cleaveland et al., 2006). Bats
can play a significant role in disease control, considering they can prey upon insects that are
diseases vectors (Monadjem et al., 2010). Fruit bats can also play an important role in seed
dispersal and pollination of plants where they feed on (Kunz et al., 2011). It can be assumed that
any significant impact on bat populations that play an important ecological role on the ecosystem
will have also an important impact on the ecosystem services provided by these species.
However, this is an impact difficult to assess. Nevertheless, the mitigation measures proposed for
the other potential impacts identified will be translated on the mitigation of this particular
potential impact as well.
5.1.4. C umu lat ive im pac ts
Cumulative impacts of a development project may be defined as “impacts resulting from incremental
actions from the project, by addition with other past, present or future impacts resulting from other
actions/project reasonable predictable” (Hyder, 1999) and more recently as “additional changes caused
by a proposed development in conjunction with other similar developments or as the combined effect of a
set of developments, taken together” (SNH, 2012). This assumes the knowledge of other projects or
actions whose effects could be added to the ones resulting from the project being assessed. Since
it is not reasonably viable to consider in the analysis all the existing or proposed projects for a
certain region, the analysis should focus on (Masden et al., 2010; SNH, 2012):
- The projects (for which there is information readily available) known for the area and its
surroundings and that could be relevant in terms of the expected impacts, in relation to
the project under assessment;
- The target species more relevant and/or susceptible to the expected impacts.
Even where fatality rates may appear low, it should be given adequate attention to it. The
cumulative effects of several facilities on the same species could be considerable, particularly if
these are located in the same region and impact on the same population of the species. Also most
of the long lived and slow reproducing Red Listed species may not be able to sustain any
additional mortality factors over and above existing factors.
The main activities or projects, relevant for the cumulative impacts analysis, known in the broader
area of the Richards Bay wind energy facility are human activities, namely agriculture and
conversion of natural areas for agriculture as well as other few proposed wind energy facilities.
Human activities:
The study area falls within a transformed habitat for agricultural purposes. Presently the area is
mostly used for sugar cane cultivation. There are however some areas with remaining natural
108 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
vegetation, as for example in the riparian strips or small patches or resilient native vegetation. In
spite of the existing patches of natural vegetation it is expected that the agricultural area can
expand in the future, leading to the conversion of some of the areas of unspoiled natural
vegetation into agriculture land. As this area is additionally under a high human pressure it is
expected that the natural areas could be subjected in the future to some level of degradation.
Other wind energy facilities:
There is at least one other known proposed wind energy facility in the vicinity of the Richards Bay
project (Figure 38), the Hluhluwe wind energy facility. This proposed project is located
approximately 70 km south of the proposed Richards Bay wind energy facility (refer to Figure 38),
with a proposed 60 MW capacity. This project is not yet permitted though.
Considering the distance that separates this wind facility from Richards Bay wind energy facility, it
is not expected that most of the present and possibly occurring species in the study area will be
significantly affected by the resulting cumulative impacts, since most bats usually don’t travel
distances of more than 50 km between summer and winter roosts (Kunz et al., 2007; Monadjem
et al., 2010). Therefore the main concern from the wind facilities located in the broader region
relates to bat species that make medium to long migrations such as Miniopterus natalensis, Myotis
tricolor or Rousettus aegyptiacus. The Mission Rocks Caves is located at a minimum distance of 87
km from the study area. These caves are a known winter roost of Rousettus aegyptiacus and other
migratory species such as Miniopterus natalensis (refer to chapter 1.4). In the study area two of
these species with migratory habits have been confirmed using the Richards Bay wind energy
facility site: Miniopterus natalensis and Myotis tricolor being possible to expect that potential
cumulative impacts result for these species while commuting between roosts. Nevertheless,
baseline information on migration and dispersion of bat species in South Africa is deficient and it is
also possible that the individuals identified where not using the area while on migration, but rather
as a foraging area, this diminishes the probability of impact on these species, from cumulative
impacts. Nevertheless, other wind energy facility may be proposed in the area, as the industry is
rapidly expanding in South Africa and at the time of this report compilation this information is not
known.
109 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 38 – Location of known proposed wind energy facilities in the vicinity of Richards Bay wind energy
facility.
5.2. I MPACT ASS ESSME NT - CON STRUC TION PHASE
During construction phase several activities will occur that might result in impacts on bats. The
hardstand areas construction and installation of the turbines; internal access roads construction;
substation construction, underground cabling installation, installation of overhead power lines and
installation of the workshop area, are all necessary action for the wind energy facility installation
that might affect bat communities.
The main impacts resulting from the construction phase will be negative to bats since it would
cause habitat loss due to the clearance of the working areas, which may lead to the displacement
of bats from feeding areas and disturbance and/or destruction of bats roosts due to the increase
of people and vehicles in the area, high levels of noise and machinery movements.
The study area is predominantly agricultural land (sugar cane plantation) and natural vegetation
(especially in the northern section of the site, falling within the Endangered Zululand Coastal
Thornveld). The area is bisected by several water lines which are associated with riverine
vegetation, as well as some dams and wetlands spread across the site. It is expected that the
affected biotopes will be mostly agricultural areas once most of the turbines are proposed to be
implemented in sugar cane plantation areas. The agricultural lands are biotopes with medium to
low value to bats, as they provide foraging areas with abundant insects during at least a section of
the life cycle of the culture; however the species that can use these are generally not threatened
110 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
being widespread (e.g. Neoromicia capensis). Agricultural areas are easily recovered to their pre-
construction condition, once the construction activities ceases.
The areas of native vegetation and riverine vegetation are considered biotopes with high to
medium value; with the value being dependent on the conservation state of the vegetation. These
areas of natural vegetation are potentially used by threatened species, being also more susceptible
to impact due to their slower regeneration.
The habitat loss causing displacement from feeding areas due to the installation of the wind
energy facility and associated infrastructure are considered impacts of:
moderate to low (with mitigation) significance, if agriculture areas (Impact 1), or
moderate to low (with mitigation) significance, if native and/or riverine vegetation areas
(Impact 2).
The disturbance and/or destruction of roosts due to people and vehicle presence and movements
is considered an impact of moderate to low (with mitigation) significance especially because the
impact is temporary (lasting while the construction lasts) and with a very restricted area of
impact, having therefore a local extent, with a limited probability of happening (Impact 3).
Overall, the impacts during the construction phase are mainly related to habitat destruction.
However, due to the limited extent of the affected areas, and these are located primarily on
transformed habitats, these can be expected to be of medium/low significance, once the mitigation
measures are implemented.
111 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Impact 1: Habitat loss - native and riverine vegetation.
Cause and comment
Destruction of native and riverine vegetation due to platforms construction, workstation and
substation construction, internal access roads construction, and turbines, underground cabling
and overhead power lines installation.
Mitigation and Management
The minimization of these impacts can be achieved through the avoidance of sitting of
infrastructure, especially turbines, in the sensitive areas, in the layout planning phase, or through
the minimization of the affected areas as far as possible in the activities of clearance and removal
of vegetation. Existing roads and infrastructure should be used in order to minimize landscape
changes. If large portions of sensitive areas are affected, measures should be taken to restore
vegetation as soon as possible after construction is completed. Movements of machinery, vehicles
and persons should be restricted to the existing roads and avoid the existing natural areas.
Impact Statement
Impact
Effect Risk or
Likelihood
Overall
Significance Temporal Scale Spatial Scale Severity of
Impact
Operational phase
Without mitigation Long term Localised Severe Probable Moderate
With
mitigation Medium term Localised Moderate May Occur Low
No-Go
Without mitigation N/A N/A N/A N/A N/A
With
mitigation N/A N/A N/A N/A N/A
Impact 2: Habitat loss - agriculture areas (sugarcane plantation).
Cause and comment
Destruction of agriculture areas due to platforms construction, workstation and substation
construction, internal access roads construction, and turbines, underground cabling and overhead
power lines installation.
Mitigation and Management
The minimization of these impacts can be achieved through the avoidance of setting of
infrastructure, especially turbines, in areas of agriculture coincident were high activity was
detected, in a layout planning phase, or through minimisation of the affected area as far as possible
as a result of the activities of clearance and removal of vegetation. The beneficiation of existing
accesses should be conducted strictly to the extent necessary. The area of intervention should be
identified and delimitated prior to the beginning of the work. Movement of machinery, vehicles
112 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
and persons should be restricted to the existing roads and avoid destruction of more agriculture
areas than necessary.
Impact Statement
Impact
Effect Risk or
Likelihood
Overall
Significance Temporal Scale Spatial Scale Severity of
Impact
Operational phase
Without mitigation Long term Localised Slight Definite Moderate
With
mitigation Long term Localised Slight May Occur Low
No-Go
Without mitigation N/A N/A N/A N/A N/A
With
mitigation N/A N/A N/A N/A N/A
Impact 3: Disturbance and/or destruction of roosts
Cause and comment
Disturbance and/or destruction of bat roosts due to the increase of people and vehicles in the
area, and destruction of roost locations.
Mitigation and Management
In order to minimize this impact certain measures can be taken, such as avoid the presence of
people and vehicles in the sensitive areas as much possible; whenever possible schedule activities
in order not to cause disturbance during the breeding season; lower the levels of noise whenever
possible around the sensitive areas; avoid construction works during the night and avoid the
destruction or disturbance of identified roosting sites.
Impact Statement
Impact
Effect Risk or
Likelihood
Overall
Significance Temporal Scale Spatial Scale Severity of
Impact
Operational phase
Without mitigation Long term Study area Severe Probable Moderate
With
mitigation Medium term Localized Moderate May Occur Low
No-Go
Without mitigation N/A N/A N/A N/A N/A
With
mitigation N/A N/A N/A N/A N/A
113 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5.3. I MPACT ASS ESSME NT - OPE RATIO NAL P HASE
The most significant negative potential impacts on bat communities can occur during the
operation phase. These impacts are mostly related with bat mortality due to collision with turbine
blades or barotrauma. The potential collision risk is not the same for all bat species and it varies
according to the species’ habits and ecology. Certain bat habits, such as migration, high flight,
clutter-edge foraging or open air foraging, contribute to species susceptibility to collision
(EUROBATS, 2013; Sowler & Stoffberg, 2012). On the other hand the barotraumas phenomenon
is caused by the change of pressure on the back side of the blade, so the specimens don’t contact
directly with the blades. Once more, the species that flight at blade height are more susceptible to
suffer barotrauma. In South Africa the information about bat behaviour is still scarce so the
measurement of collision risk is sometimes difficult.
From the 35 species with possible occurrence in the study area 5 have high risk of collision with
turbines and 13 have medium to high risk of collision (Sowler & Stoffberg, 2012). Four of the
species that can potentially occur in the study area and have medium to high collision risk are
classified as “Near Threatened” by the South Africa Red List (Friedmann & Daly, 2004):
Miniopterus fraterculus, Miniopterus natalensis , Myotis tricolor and Myotis welwitschi. It is also
important to note that two of the confirmed species have documented collisions with wind
turbines operating in South Africa, Neorimica capensis and Tadarida aegyptiaca (Aronson et al.,
2013; Doty & Martin, 2013), though none of them are species of conservation concern
(Friedmann & Daly, 2004; IUCN, 2012).
Considering the potential risk of fatality of bats at the study area, species of low and medium
ecological value can be target of mortality events in the wind energy facility. However this risk of
mortality may be influenced by the turbine location, being considered that the risk of collision is
higher if turbines are placed within the surrounding area of roosts or within areas that are both
close to roosts and important habitat areas for bats. For both situations the impact was
considered to have a high significance without mitigation measures and moderate significance with
mitigation measures (Impact 4). Bats are prone to be more active surrounding their roosts or
while foraging in favourable areas. While foraging bats will be more curious and inquisitive being
likely to investigate the wind turbines if these are placed within areas of interest to bats, and
therefore cause fatalities due to collision with moving blades or barotraumas.
Impacts caused by turbines placed only in the vicinities of important areas for bats are considered
to be of moderate (with and without mitigation) significance impacts (Impact 4). Turbines which
are not located within the surrounding area of sensitive features will most likely cause low
significance impacts. If no sensitive areas are presents bats occurring in the vicinities of these
turbines will most likely be passing by and are not likely to spend time investigating the wind
turbine. There is however the possibility that, since these turbines are placed in sugarcane
plantation, bats will forage in some times of the year, when insect activity is higher. This may
increase the probability of collision with wind turbines, and therefore it is considered that these
turbines may cause an impact as well, though of a lesser significance (Impact 4).
114 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Impacts on bat communities caused by power lines hardly have been assessed as it is considered
that no impacts of the operation of power lines are expected on insectivorous bats. However,
Fruit bats mortality incidents are known to be caused by electrocution in Australia, especially with
mothers carrying their offspring (Bat Care Brisbane, n/date). An early stage analysis of this
potential impact may suppose a moderate to low significance impact (Impact 5). Further
monitoring on fruit bats movements and reproduction within the study area in will be essential in
the following phases to assess the impacts of power lines on these species.
Considering the possible reduction of bat ecosystem services, this impact may be caused by bat
mortality or bat displacement from the study area. Without the presence of a certain community
of bats, important ecosystem services such as pollination or pest control will be reduced, or in
extreme cases, will cease to be provided and most likely agriculture practices will notice the
economic impact. This impact is considered of moderate to low significance if proper mitigation
measures are implemented (Impact 6).
Overall, the impacts during operational phase are about bat mortality and the significance of those
is medium with the possibility to be considered low if proper mitigation measures are
implemented.
Impact 4: Mortality of bat species due to collision and/or barotrauma with turbines.
Cause and comment
Mortality of bats due to collision with moving turbine blades while foraging or communting and/or
mortality of bats due to barotraumas phenomenon. This impact may be influenced by the
presence of sensitive areas for bats surrounding turbines.
Mitigation and Management
The minimisation of deaths caused by wind turbines can be achieved through the avoidance of
turbines installation in sensitive areas for bats. From the turbine impact analysis, 14 of the 39
turbines are likely to cause impacts. Considering that bat activity in the study area was quite high
and most of the turbines are located within sensitive areas for bats being likely to cause impacts,
is recommended the relocation of the turbines likely to cause higher impact - WTG new4, new5,
3, 8, 9, 10, 15, 16, 26, 31, 34, 41, 47, 48. Some of the collision risk is also motivated by the
presence of several roosts, which should be safeguarded and monitored during the operational
phase. A bat monitoring program should be implemented in order to determine the actual
impacts of the wind energy facility on the bat community, as well as the implementation of
mitigation measures, such as the utilization of red lights in the turbines, instead of white, in order
to minimize insect attraction and bat foraging behaviours near the turbines.
115 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Impact Statement
Turbines
Sensitivity Impact
Temporal
Scale Spatial Scale
Severity of
impact Likelyhood
Overall
Significance
Roosting
sensitive
Without Permanent Study area Severe (Probable) High
With Permanent Localised Moderate (Probable) Moderate
Habitiat
sensitive
Without Permanent Study area Moderate (Probable) Moderate
With Permanent Localised Moderate (Probable) Moderate
Roosting
and
Habitat
sensitive
Without Permanent Study area Severe (Probable) High
With Permanent Localised Moderate (Probable) Moderate
Non
sensitive
Without Long term Localised Slight May Occur Low
With Long term Localised Slight May Occur Low
Impact 5: Mortality of frugivorous bats.
Cause and comment
Mortality of frugivorous bat species due to collision with overhead power lines.
Mitigation and Management
Though not much information exists regarding fruit bats collisions with wind turbines, some
minimization may be achieved by implementing visible wires to increase the visibility of the power
lines. There is, however, no specific solutions known to be tested and effective in mitigating this
potential impact. Therefore, a bat monitoring programme should be implemented in order to
determine the actual impacts power lines on the fruit bat community, as well as recommend on
the necessary implementation of mitigation measures.
Impact Statement
Impact
Effect Risk or
Likelihood
Overall
Significance Temporal Scale Spatial Scale Severity of
Impact
Operational phase
Without mitigation Long term Study area Moderate Probable Moderate
With
mitigation Long term Study area Slight Unlikely Low
No-Go
Without mitigation N/A N/A N/A N/A N/A
With
mitigation N/A N/A N/A N/A N/A
116 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Impact 6: Reduction of ecosystem services provided by bats.
Cause and comment
Reduction of ecosystem services provided by bats due to mortality by collision with wind turbines
or displacement from the study area due to turbines operation and increase of people and
vehicles in the area associated with maintenance activities.
Mitigation and Management
The minimization of this impact is achieved by implementing the measures suggested to reduce
bat mortality by collision with wind turbines and/or barotraumas and bat community disturbance.
These measures include the relocation of some turbines with higher likelihood to generate
impacts, and avoid the presence of people or vehicles within areas important for bats (such as
roosts).
Impact Statement
Impact
Effect Risk or
Likelihood
Overall
Significance Temporal Scale Spatial Scale Severity of
Impact
Operational phase
Without mitigation Long term Regional Severe Probable Moderate
With
mitigation Medium term Study area Slight May Occur Low
No-Go
Without mitigation N/A N/A N/A N/A N/A
With
mitigation N/A N/A N/A N/A N/A
Considering the evaluation conducted so far, for the construction and operational phase of the
project, is considered that the No-Go option (if the project does not go ahead), would be the
most favourable outcome for the bats of the study area. However this outcome would not
improve the current environmental status of the study area, since it would not cause changes in
the community of bats, being this option considered as Neutral or Not Applicable.
5.4. I MPACT ASS ESSME NT - DEC OMMIS SIONI NG PHAS E
During the decommissioning phase it is expected the dismantling of turbines and associated
infrastructure, as well as the dismantling of power lines, which can lead to disturbance of bat
community would be similar to the one resulting from construction phase (refer to Impact 1,
Impact 2 and Impact 3), and is classified as an impact of medium to low significance.
The dismantling of the project will eventually contribute to the removal of all the implemented
structures which would cause negative impacts on the bat community and this may therefore be
considered a positive impact.
117 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
5.5. M INIMI SATIO N AND /OR MITIGA TION MEASU R ES
This section of the report provides recommendations for the minimisation/mitigation measures
for the different phases of the project. The proposed mitigation measures are based on
international standards, author’s expertise and follow the South African Good Practice Guidelines for
Surveying Bats in Wind Farm Developments (Sowler & Stoffberg, 2012).
5.5.1. L ayo ut def ini tio n p ha se
The wind turbine model to be selected should preferably be one where the lower tip of
the blades is at least about 60 m above the ground and/or a smaller rotor diameter should
be preferred against a bigger rotor diameter;
Structures should be designed to reduce the availability of roosting sites (e.g. closed
nacelle without cavities);
Relocation of the following wind turbines is recommended: WTG3; WTGnew4,
WTGnew5, WTG8, WTG9, WGT10, WTG15, WTG16, WTG26, WTG31, WTG34,
WTG41, WTG47 and WTG48. These are located within areas of natural vegetation,
within or in the vicinities of riverine vegetation or areas considered to have high bat
activity, and were identified previously as being the turbines most sensitive and likely to
cause impact (refer to section 4.1); WTG48 is currently at approximately 780 m from a
known maternity roost of E. wahlbergi, and the 1000 m buffer considered should be
implemented when sitting this turbine. Nevertheless, the continuity and importance of
this roost to these species should be further monitored;
Relocation of turbines is recommended to any type of area considered to do not be
sensitive for bats: e.g. at more than 500 m from native vegetation, riverine vegetation, bat
roosts, high activity area (identified in this pre-construction monitoring programme),
more than 1500 m from reproduction roosts;
Road accesses, installation platforms and other infrastructures associated with the
development (power lines, substation, and maintenance buildings) should be designed in
order to avoid the sensitive areas identified (refer to section 4).
5.5.2. C ons tru cti on pha se
Clearing of natural vegetation during construction should be kept to a minimum;
Interventions on the identified sensitive areas should be minimized (refer to section 4).
No borrowing or fill material should be collected from these sensitive areas nor any
deposits should take place;
Implement actions for rehabilitation of the vegetation removed, if the affected areas are of
a considerable size and importance to the bat community of the study area;
Minimize areas of construction to the maximum extent possible;
Sensitive areas located close to intervention areas (areas to be constructed, or used
during the construction phase) should be clearly identified;
Avoid construction works during the night in order to avoid bat community disturbance;
118 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Avoid construction works and/or high levels of activity during critical periods for bat
populations, such as the breeding (April – May for most species) and birth (November –
December for most species) seasons;
Proper training should be provided to all the construction personnel. Everybody working
in the area should be aware of the sensitive areas, be alert to the possible presence of
bats, especially when working close to natural vegetation (particularly trees and banana
trees);
The construction works should be supervised by a bat specialist on site, in order to
further identify any conflict situations between the construction works and bats, and
readily take actions to minimize any identified impacts;
During the construction phase any disturbance within the roosting locations sensitive
areas (Figure 34; Figure 35) should be avoided or, if inevitable, kept to the minimum
necessary levels;
If any building, trees, banana trees, or any structure with potential to provide bat
roosting, needs to be demolished or removed, then it should be conducted a visit, prior
to the commence of the works, by a bat specialist to verify the presence / absence of
bats;
The confirmed and potential roosting locations, native vegetation and the riverside
vegetation should be avoided or, if inevitable, kept to the minimum necessary levels;
The confirmed Epomophorus wahlbergi (close to WTG48) reproduction roost should be
considered a no-go area during July and December to February, the peaks of
reproduction of the species (Monadjem et al., 2010). Since births of this species occur
throughout the year the disturbance within the 1000 m buffer around this roost should
be minimized;
In the case that any confirmed or potential bat roost needs to be affected (e.g. utilization
conversion, demolition, recuperation) a bat specialist should confirm bat occupancy and
the necessary measures to be implemented and minimize the impact should this be
necessary;
Sufficient and adequate drainage should be provided along access roads to prevent
erosion and pollution of adjacent watercourses or wetlands.
No chemical spills or any other material dumps should be conducted within the
intervention area, with special focus on areas nearby riparian vegetation or drainage lines.
All the maintenance of vehicles must be carried out in specially designated areas to
prevent any type of pollution on the area.
Ensure the implementation of a construction and post-construction monitoring
(operation phase) programme to survey bat communities on the wind energy facility and
the impacts resulting from the installed infrastructure (refer to Appendix IX). This
programme should have a minimum duration of 1 year.
5.5.3. O per ati ona l p has e
The occurrence of at least two species considered to have medium and high collision risk with
wind turbines and with recorded fatalities in wind energy facility in South Africa have been
119 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
confirmed in the study area (i.e. Neoromicia capensis and Tadarida aegyptiaca). These species have
high risk of collision due to their flight characteristics. They are open-air foragers, which may fly at
high altitudes, therefore being potentially within the rotor swept area. Since these species are
considered potentially affected by the operational phase of the project a set of measures are
proposed in order to minimize the potential bat fatalities:
• Ensure the implementation of a post-construction monitoring programme (operation
phase) to survey bat communities on the wind energy facility and the impacts resulting
from the installed infrastructure (refer to Appendix IX). This plan should have a minimum
duration of at least three years after the construction of the project;
• The results of the operational phase monitoring programme must be taken into account
for the implementation of further mitigation measures, if necessary, during the
operational phase of the proposed wind energy facility;
• If high mortality risk areas are identified during the operational phase, or a high number
of bat fatalities due to wind turbines are recorded, this should be evaluated by the
designated bat specialists as soon as possible. Subsequent mitigation measures, adjusted
to the risk situation identified, should be then proposed and implemented and may
include, but not restricted to curtailment at specific turbines or blade feathering (Arnett
et al., 2013);
• If turbines are to be lit at night, lighting should be kept to a minimum and should
preferably not be white light. Flashing strobe-like lights should be used where possible14;
• Lighting of wind energy facility (for example security lights) should be kept to a minimum
and should be directed downwards (with the exception of avian security lightning).
• It is recommended to implement blade feathering15, or ensure the blades don’t freely
rotate at wind speeds bellow 3 m/s. This measure is considered to significantly reduce
bat mortality at wind speeds bellow the manufacture cut-in speed (Arnett et al., 2013).
The monitoring programme should have a minimum duration of at least 4 years and be revised
afterwards (at least 1 year during construction and 3 years during operational phase). It should
include both the continuation of the assessment of bat communities in the site, complementing
the information gathered during the pre-construction phase and allowing determining any
exclusion effects over the bat community. The operational phase monitoring programme should
include carcass searches and the determination of correction factors (observer’s efficiency and
carcass removal) in order to accurately determine the impact of the wind turbine on bats and
14 Provided this complies with Civil Aviation Authority regulations
15 Adjusting the angle of the rotor blade parallel to the wind, or turning the whole unit out of the wind, to slow or stop
blade rotation. Normally operating turbine blades are angled perpendicular to the wind at all times (Arnett et al., 2013).
120 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
determine any potential critical area and/or wind turbines. To properly calculate the real
mortality associated to wind energy facilities it is recommended that correction factors are
assessed with a 15 days frequency, moreover it is essential to adopt a fatality estimator that
adjusts the observed casualties with the estimated bias correction terms (Bernardino et al. 2013).
In what concerns carcass searches it is recommended to consider the utilization of trained dogs
to search for bat carcasses, as it is proven to be one of the most effective methodologies to
determine bat fatalities with wind turbines, with detection rates of up to 96% of all the carcasses
present in the field, opposing to the approximately 10 to 20% efficiency of human observers
(Paula et al., 2011) (refer to Appendix IX).
5.5.4. D eco mmi ssi oni ng pha se
The minimization measures in this phase are similar to the ones proposed for the construction
phase since the actions are similar. After the total removal of the wind energy facility
infrastructures the areas occupied by its components should be rehabilitated.
121 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
6. I M P A C T S T A T E M E N T
The proposed site for the Richards Bay Wind Energy Facility is located in an area considered to
be of medium to high sensitivity for bats, especially in the northern section of the site where bat
activity proved to be higher. However, considering the bat species observed and the
environmental features present of importance for bats survival activities, some particular areas
could pose a higher risk of collision for bats, which should be acknowledged and safeguarded.
The current layout for the facility has some interference with the identified sensitive areas, as the
location of some turbines may pose increased risk of bat mortality by collision and/or
barotraumas. The relocation of these turbines is recommended and if not possible they require at
least the implementation of mitigation measures during the construction and operational phase, in
order to reduce the probability and magnitude of impact.
The impacts identified most likely to occur are related with bat mortality due to collision with
wind turbines and/or barotraumas and bat displacement from feeding areas due to habitat loss
during the construction phase. Both impacts were considered to be of a high to moderate
significance, mostly due to their permanent character (lasting as long as the project is
implemented) and to their high probability of occurring. Since this is a medium-sized wind project,
the overall impact on bats will be of a moderate level of significance, which emphasizes on the
need to comply with mitigation and monitoring regimes.
The bat species most likely to suffer the impacts caused by the presence of this infrastructure
include mostly open air foragers and clutter-edge foragers species (e.g. Tadarida aegyptiaca or
Neoromicia capensis), which are known to have collisions with other wind energy facilities in South
Africa (Darling National Demonstration Wind Farm Project, Western Cape; and Coega Industrial
Development Zone, Port Elizabeth, Eastern Cape). During- and post- construction monitoring will
be very important to improve the understanding of the real impact caused by the wind energy
facility on local bat populations on the site.
Considering that this Pre-construction Monitoring Report is based on the results of a one-year
pre-construction bat monitoring programme, the observations made have a certain degree of
consistency and provide adequate baseline information for this assessment and the level of
confidence in this study is high.
A rigorous and well planned monitoring programme is considered to be one of the most effective
measures to be proposed at this stage. The continuation of the monitoring programme will
contribute to: a) complement the information gathered during the pre-construction phase,
increase knowledge about bat communities in the Richards Bay Wind Energy Facility and help
determine any exclusion effects over the bat community, b) verify the potential impacts identified
during the pre-construction phase especially those concerning bat mortality with wind turbines c)
verify the effectiveness of the mitigation measures proposed, implement adjustments or additional
measures if necessary. This will allow proposing mitigation measures, if necessary, adjusted to the
site specificities. This mitigation measures20 must be evaluated in a case by case scenario. An
effective mitigation measures plan is one that shows an accurate determination of the most
122 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
problematic areas and/or wind turbines and the characterization of the environmental variables
with higher influence on bat fatalities (Arnett et al., 2013). Nonetheless the implementation of
such measures should be implemented only if necessary and they should be carefully planned in
order to maximize their efficacy in reducing bat mortality and assure the compatibility of the
development with bat communities’ conservation (Arnett et al., 2010; Arnett et al., 2011).
Although bat mortality may occur, based on pre-construction results, this is expected to affect
mostly common and widespread species. However, if impacts identified in the subsequent phases
of the project are more severe than expected additional mitigation measures16 may be evaluated,
particularly if mortality occurs in levels that compromise the local population’s viability. Such
measures should only be implemented if necessary and they should be carefully planned in order
to find the best trade off in reduction of the collision risk and minimize the loss in revenue
resulting from mitigation.
The bat monitoring programme to implement should have a minimum duration of at least 4 years
and be revised afterwards, accordingly to the results obtained (at least 1 year during construction
and 3 years during operational phase). The operational phase monitoring programme should
include carcass searches and the determination of correction factors (observer’s efficiency and
carcass removal) in order to accurately determine the impact of the wind turbine on bats and
determine any potential critical area and/or wind turbines.
16 The increase in cut-in speed as well as feathering the blades are measures considered effective reducing bat mortality
due to collision and/or barotrauma with wind turbines, enabling to reduce bat fatalities at least by 44% (Arnett et al.,
2010; Arnett et al., 2011). Other studies suggest that variables other than wind speed can be considered for the cut-in
of the blades rotation, such as temperature or time of night (Arnett et al., 2013).
123 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
7. C O N C L U S I O N S A N D R E C O M M E N D A T I O N S
7.1. M AIN R ESULT S OF THE P R E-CO NSTRU CTION MON ITOR I NG P RO GRAMM E
During the surveys conducted by NSS for the pre-construction bat monitoring programme of
Richards Bay wind energy facility the following activities were accomplished:
- Manual survey vehicle based transects;
- Six static sampling points, not all simultaneously but up to four detectors in simultaneous
at different heights (ground level and rotor height);
- Roost searches, identification and visits;
- Live-trapping by mist-netting and harp trapping.
All the mentioned activities were conducted between the end of April 2012 and beginning of May
2013 with the objective to characterize and map the bat activity in the Richards Bay wind energy
facility area to subsequently assess the impact of the proposed wind energy facility.
Upon those surveys 21 bat species were identified and confirmed in the study area (from a total
of 35 species likely to occur in the study area):
- 6 species considered as “Near Threatened” by the South Africa Red List: Hypsugo
anchietae, Miniopterus fraterculus, Miniopterus natalensis, Myotis welwitschii, Myotis tricolor and
Rhinolophus clivosus;
- 14 species classified as “Least Concern” by the South Africa Red List;
- 1 species considered as “Data Deficient” by the South Africa Red List, the Hipposideros
caffer.
Of the remaining species that may occur in the study area but which were not confirmed so far,
Cleotis percivali is classified as “Critically Endangered”, Kerivoula argentata and Rhinolophus swinnyi
are “Endangered” and Otomops martiensseni is “Vulnerable” according with the South Africa Red
List.
The analysis of bat activity in the study area, considering the results from both the manual surveys
and static detection, appears to indicate that bat activity in the study area is higher in spring and
summer, though bats seem to be intensely active throughout the year. These times of the year
are coincident with insects’ emergence, which is the appropriate time for feeding, and it is the
birth period of many species. The existence of high activity peaks in these periods may indicate
that the study area is important for bats as foraging areas. It is of note that Richards Bay site is
located in the broader vicinity of at least five important caves used as roosts (Mission Rock Caves,
Hlatikula Forest Reserve, Border Cave, Doornhoek Tunnel and Sibudu Cave), two of them with a
124 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
species that have been confirmed in the study area and that is known to migrate several
kilometres: Miniopterus natalensis (CES, 2012). Therefore the presence of high peaks of activity in
time periods when bats can migrate to summer or winter roosts can indicate that this type of
roost may be present in the study. However this was not confirmed and no roosts of possible
migratory species were identified in the study area (Miniopterus natalenis and Myotis tricolor) to
date.
Considering bat activity observed through the night in the study area, several peaks of activity
were observed associated with the beginning of sunset or the proximity of sunrise. This was
according to some authors that propose that bat activity can be influenced by several factors
including the type of habitat and the season (Meyer et al., 2004; Brooks, 2009; O´Donnel, 2000).
Using this knowledge, it is possible to reduce the probability of bat fatality by implementing
mitigation measures specific for these critical periods of activity. However, according to the data
collected so far and since one bat species (Otomops martiensseni) with conservation status of
concern and high probability of collision with wind turbines is suspected to occur in the
study area, according with the NSS team indications (refer to section 3.2.1), it is important to at
least consider the continuation of the monitoring programme, during the construction and
operation phases of the development. This will allow to validate the higher sensitive areas
identified, collect more information and identify critical collision risk areas where such mitigation
measures may require implementation and specifically adapted to each different scenario.
The examination of the average number of passes by each of the detectors with microphones
placed at different heights allowed the comparison between the activity closer to the ground
(7/10/30 m) and high in the air, nearly at rotor height (60/80m). This comparison indicated that
recorded bat activity in the study area was generally higher at lower heights. The possibility that
bats at Richard Bay forage and travel at lower height may reduce the collision risk of some
species, as the possibility of cruising the path of moving blades is lower. Some authors have found
that in the U.S. a high number of fatalities were related with the increase in tower height, due to
migratory bats that entered the swept area of blades, which is usually located at 65m (Barclay et
al., 2007). Considering that in the study area at least two species are present and confirmed to
have migratory behaviour, and moreover are classified as “Near Threatened” on South Africa
Red list (Miniopterus natalensis and Myotis tricolor) it is therefore recommended that the
characteristics of the wind turbines to implement in this wind energy facility have a rotor swept
area above 60m, in order to minimize the utilization of the rotor swept area by these species
(that utilizes mostly lower heights), where the probability of fatality is higher.
Statistical analysis showed that air humidity, air temperature, wind speed and illuminated lunar
fraction influence positively or negatively bat activity, according with the environmental variable
considered.
Since fires for management of sugar cane plantations are used by local farmers, it is also important
to consider its effects over bats present in the study area. A fire occurred in August, preventing
the team conducting transect surveys. However the static surveys recorded a small peak in
activity in the first half of the month, and a decrease in the second half of August. This may
125 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
indicate that the fire could be of prejudice to bat activity since insect’s abundance may be affected,
especially during winter.
In the study area, 5 species with possible occurrence are perceived as having a potential high risk
of collision with wind turbines (Sowler and Stoffberg, 2012) due to their behaviour, where 3 have
been already confirmed, all of those with a conservation status of Least Concern (Chaerephon
pumilus, Mops condylurus and Tadarida aegyptiaca). Otomops martiensseni, that was not initially
foreseen to occur in the study area, was suspected to occur, and has as well a high risk of
collision with wind turbines. Their potential high risk of collision is related with their foraging
behaviour as open-air foragers, which promotes the entry of individuals in the turbine blade swept
area, therefore increasing the probability of collision (EUROBATS, 2010). There are also
references to mortality incidents in wind facilities in Europe with several species of Tadarida sp.,
same genus as Tadarida aegyptiaca and USA (Tadarida brasiliensis) for species of the same genus
(EUROBATS, 2013; Arnett et al., 2007).
Another 13 species with possible occurrence have medium-high risk of collision, and 4 species
have medium risk of fatality due to collision with wind turbines (Sowler and Stoffberg, 2012).
From these 17 species (13 with medium-high and 4 with medium risk of collision), 11 have already
been confirmed in the study area. These species are in general clutter-edge foragers with known
wind turbine collisions in Europe and USA, from the same or similar genus, such as Miniopterus
sp., Myotis sp., and Eptesicus sp. (also similar to Neoromicia sp.) (EUROBATS, 2013).
The data collection analysis during this first year of monitoring prior to construction, allowed the
characterization of the bat community present in Richards Bay wind energy facility, and predicts
the potential effects that the implementation of this project may have over bat populations in the
study area. This study showed that the broader northern area of the wind energy facility is highly
used by bats as foraging and roosting areas, being however coincident with the higher
concentration of the proposed wind turbine locations which could lead to situations where wind
turbines are placed in areas considered as sensitive for bats. Therefore, the suggested mitigation
measures should be implemented in order to minimize disturbance of these locations during the
construction phase of the project. During the operational phase of the project, bat fatalities are
expected to be the main negative effect over bat populations, being possibly reduced if the
adequate mitigation measures are implemented (see section 7.2). The implementation of an
adequate monitoring programme during the subsequent phases of the project will contribute to
validation of the predicted impacts, and verify if the mitigation measures proposed and
implemented are adequate and if necessary propose any adjustments in this regard. If other
impacts are identified, then additional mitigation measures can be proposed, where necessary.
7.2. R ECOMM ENDAT ION S FOR TH E NEX T PHA SE S O F T HE PR OJECT
Considering the sensitivity analysis conducted (refer to section 4), some recommendations were
proposed to minimize the potential negative impacts of the wind energy facility, namely during its
construction and operational phase, in order to reduce the risk of bat collisions with the wind
turbines located within sensitive areas (refer to section 5.5). The Richards Bay wind energy facility
126 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
development area was considered of medium to high sensitivity in regard to the bat community,
following the results from the pre-construction bat monitoring programme. No wind turbines are
proposed within the higher sensitive areas identified. However some turbines are to be placed in
the immediate vegetation vicinity areas which are important for bats, and are associated with
areas of higher activity (refer to section 4.1). Some wind turbines are also located in areas
surrounding roosts, including roosts where reproduction was observed. Although the proposed
wind turbines are located more than 500 meters from these sites, it is considered very important
to minimize any noise or perturbation of the roosting sites at least during the construction phase.
This may be achieved by identifying the roosts that are occupied before the construction works,
and implement an area of no-disturbance of at least 500 meters, where the presence of
machinery, workers or particularly noisy activities are expected should be avoided.
Considering the hypothesis that bat fatalities in Richards Bay wind energy facility are expected, it
is proposed that a monitoring programme be implemented during the construction and
operational phases of the project. A well planned and rigorous monitoring programme is one of
the most effective management measures to be implemented at this stage. During the
construction phase, the bat monitoring programme should contribute to a better understanding
of bat communities on the area, and contribute with further data to better access the relationship
between bats and environmental variables. During the operational phase the bat monitoring
programme will contribute to access the real bat mortality associated with the wind energy
facility, verify the efficacy of the proposed and implemented mitigation measures and conduct
adjustments if necessary. The identification of any critical areas or situations should be promptly
evaluated by the bat specialist in order to implement adequate and specific mitigation measures.
7.3. S UITAB ILITY OF T HE M ONITOR ING PROGR AMME
It is considered that the current pre-construction bat monitoring programme is suitable to
project specifications and allowed the accomplishment of the established objectives. During the
proposed monitoring programme of the operational phase it is recommended that throughout
the construction and operational phase of the project the monitoring programme is continued
with some adjustments to better suit the community identified during this first year of monitoring.
The current monitoring programme does not consider the assessment of bat activity in a similar
control area. This would be very important to accurately assess displacement and disturbance of
the bat communities present at the site. Therefore, if possible, manual detection should be
implemented by conducting sampling points covering all the different biotopes present within the
wind energy facility area, throughout the construction and operation phase monitoring
programme. A similar control area should be defined, if possible, and monitored during the same
surveys as the wind energy facility area. As the control area was not monitored during the pre-
construction phase there’s no baseline information on this to be compared with subsequent years.
Nevertheless, the use of a control area will contribute to depict the effects of random variations
in bat activity and species diversity in the wind energy facility area. If possible, monitoring the wind
energy facility during the construction phase will contribute to increase the knowledge on bat
communities in the area, and will allow the monitoring of construction phase effects on
127 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
confirmed bat roosts and, more importantly, will allow as well the establishment of the baseline
situation for the control area to be implemented.
128 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
8. R E F E R E N C E S
ACR. 2012. African Chiroptera Report 2012. AfricanBats, Pretoria. i – xvii, 1- 5902.
Ahlen I. 1990. Identification of bats in flight. Swedish society for Conservation of Nature & the
Swedish Youth Association for Environmental Studies and Conservation, Stockholm.
Atienza J.C., I. Martín Fierro O. Infante, J. Valls y J. Domínguez. 2011. Directrices para la evaluación
del impacto de los parques eólicos en aves y murciélagos (versión 3.0). SEO/BirdLife, Madrid.
Altringhan J. D. 1996. Bats - Biology and Behaviour. Oxford University Press. United Kingdom.
Arnett E.B., Inkley D.B., Johnson D.H., Larkin, R.P., Manes, S., Manville, A.M., Mason, R., Morrison,
M., Strickland, M.D., Thresher, R. 2007. Impacts of wind energy facilities on wildlife and wildlife habitat.
Bethesda, MD: The Wildlife Society.
Arnett E. B., Brown W. K., Erickson W. P., Fiedler J. K., Hamilton B. L., Henry T. H., Jain, A.,
Johnson G. D., Kerns J., Koford R. R., Nicholson C. P., O'Connell T. J., Piorkowski M. D.,
Tankersley JR, R. D. 2008. Patterns of Bat Fatalities at Wind Energy Facilities in North America.
Journal of Wildlife Management, 72(1): 61-78.
Arnett E. B., M. M. P. Huso, J. P. Hayes, and M. Schirmacher. 2010. Effectiveness of changing wind
turbine cut-in speed to reduce bat fatalities at wind facilities. A final report submitted to the Bats and
Wind Energy Cooperative. Bat Conservation International. Austin, Texas, USA.
Arnett E.B., Huso M.P., Schirmacher M.R. and Hayes J.P. 2011. Altering turbine speed reduces bat
mortality at wind-energy facilities. Frontiers in Ecology and the Environment, 9(4): 209-214.
Arnett E. B., Johnson G. D., Erickson W. P. and Hein C. D. 2013. A synthesis of operational
mitigation studies to reduce bat fatalities at wind energy facilities in North America. A report submitted to
the National Renewable Energy Laboratory. Bat Conservation International. Austin, Texas, USA.
Aronson J.B., Thomas A.J. and Jordaan S.L. 2013. Bat fatality at a Wind Energy Facility in the
Western Cape. South Africa. African Bat Conservation News, Vol. 31: 9 – 12.
Baerwald E.F., D´Amours G.H., Klug B.J. and Barclay R.M.R. 2008. Barotrauma is a significant
cause of bat fatalities at wind turbines. Current Biology, 18(16).
Bakker, H. 1999. Sugar Cane Cultivation and Management. Springer. 679pp.
Barclay R.M.R., Baerwald E.F. and Gruver J.C. 2007. Variation in bat and bird fatalities at wind
energy facilities: assessing the effects of rotor size and tower height. Canadian Journal of Zoology,
85: 381-387.
129 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Barrios L. and Rodriguez A. 2004. Behavioural and environmental correlates of soaring-bird
mortality at onshore wind turbines. Journal of Applied Ecology, 41: 72- 81.
Bernardino J. 2008. Estimativas da mortalidade de aves e quirópteros em parques eólicos: Avaliação das
metodologias em uso e propostas para o seu aperfeiçoamento. Tese para a obtenção do grau de
mestre em Ecologia e Gestão Ambiental. Faculdade de Ciências da Universidade de Lisboa.
Bernardino J., Bisp, R., Costa H. and Mascarenhas M. 2013. Estimating bird and bat fatality at wind
farms: a practical overview of estimators, their assumptions and limitations. New Zealand Journal of
Zoology. DOI:10.1080/03014223.2012.758155
Briggs B., and King D. 1998. The bats detective – a field guide for bat detection. BATBOX Ltd, West
Sussex.
Bolker B.M., Brooks M.E., Clark C.J., Geange S.W., Poulsen J.R., Stevens M.H. and White, J.S.
2008. Generalized linear mixed models: a practical guide for ecology and evolution. Trends in
Ecology and Evolution, 24(3): 127-135.
Brigham R.M. 1991. Flexibility in foraging and roosting behaviour by the big brown bat (Eptesicus
fuscus). Canadian Journal of Zoology, 1991, 69(1): 117-121, 10.1139/z91-017
Brooks R.T. 2009. Habitat-associated and temporal patterns of bat activity in a diverse forest landscape
of southern New England, USA. Biodiversity Conservation, 18:529-545.
Carrete M., Sánchez-Zapata J.A., Benítez J.R., Lobón M. and Donázar J.A. 2009. Large scale risk-
assessment of wind-farms on population viability of a globally endangered long-lived raptor. Biol.
Conserv. 142(12): 2954-2961.
CES. 2012. Environmental Impact Assessment for the Richards Bay Wind Energy Facility in KwaZulu-
Natal, South Africa. Coastal and Environmental Services.
Cleveland C.J., Betke M., Federico P., Frank J.D., Hallam T.G., Horn J., Lopez J.D. Mccracken G.F.,
Medellin R.A., Moreno-Valdez A., Sansone C.G., Westbrook J.K. and Kunz, T.H. 2006. Economic
value of the pest control service provided by Brazilian free-tailed bats in south-central Texas.
Frontiers of Ecological Environment 4: 238 - 243.
Cochran W.G. 1977. Sampling Techniques, 3rd ed., John Wiley and Sons Inc., Nova York.
Collins, J., Jones, G. 2009. Differences in bat activity in relation to bat detector height:
implications for bat surveys at proposed wind farm sites. Acta Chirpterologica 11: 343-350.
Colorado Committee of the Western Bat Working Group. Available at
http://www.wbwg.org/colorado/colorado.htm.
Cryan P. M., and Brown A.C. 2007. Migration of bats past a remote island offers clues toward the
problem of bat fatalities at wind turbines. Biological conservation 139: 1-11.
130 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Cryan P.M., and Barclay R.M.R. 2009. Causes of bat fatalities at wind turbines: Hypothesis and
predictions. Journal of Mammalogy, 90(6): 1330-1340.
CSIR. 2013. Renewable Energy EIA Application Mapping. Report Version 1. Commissioned by CSIR
on behalf of the Department of Environmental Affairs. 33p.
Drewitt A.L. and Langston R.H.W 2006. Assessing the impacts of wind farms on birds. Ibis 148:
29-42.
Drewitt A.L. and Langston R.H.W. 2008. Collision Effects of Wind-power Generators and Other
Obstacles on Birds. New York Academy of Sciences. 1134: 233-266.
Doty A.C. and Martin A.P. 2013. Assessment of bat and avian mortality at a pilot wind turbine at
Coega, Port Elizabeth, Eastern Cape, South Africa. New Zealand Journal of Zoology 40(1): 75-80.
Driver A., Sink K.J., Nel J.N., Holness S., Van Niekerk L., Daniels F., Jonas Z., Majiedt P.A., Harris
L. and Maze K. 2012. National Biodiversity Assessment 2011: An assessment of South Africa’s
biodiversity and ecosystems. Synthesis Report. South African National Biodiversity Institute and
Department of Environmental Affairs, Pretoria.
Eichhorn M., and M. Drechsler. 2010. Spatial trade-offs between wind power production and bird
collision avoidance in agricultural landscapes. Ecology and Society 15:10.
Ellison L. E., M. B. Wunder C. A. Jones C. Mosch K. W. Navo, K. Peckham J. E. Burghardt J.
Annear R. West J. Siemers R. A. Adams and E. Brekke. 2003. Colorado bat conservation plan.
Erickson W., Johnson G., Strickland M., Young D., Sernka K. and Good R. 2001. Avian collisions
with wind turbines: a summary of existing studies and comparisons of avian collision mortality in
the United States. National Wind Coordinating Committee. Washington D.C.
Erickson W., Johnson G., Young D., Strickland D., Good R., Bourassa M., Bay K. and Sernka K.
2002. Synthesis and comparison of baseline avian and bat use, raptor nesting and mortality information
from proposed and existing wind developments. West, Inc.
Esberard C. 2007. Influência do ciclo lunar na captura de morcegos Phyllostomidae. Iheringia, Sér.
Zool. 97(1): 81-85.
Estep J. 1989. Avian mortality at large wind energy facilities in California: identification of a
problem. Staff report no. P700-89-001.California Energy Commission, Sacramento.
EUROBATS. 2013. Report of the IWG on Wind Turbines and Bat Populations. 18th Meeting of the
Advisory Committee. EUROBATS, Sofia, Bulgaria.
EWEA. 2011. The European wind energy association. Disponível em http://www.ewea.org/.
Consultado a 8 de Junho de 2011.
131 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Fenton M.B., and G.P. Bell. 1981. Recognition of species of insectivorous bats by
their echolocation calls. Journal of Mammalogy 62:233-243.
Fenton M.B, Taylor P.J., Jacobs D.S., Richardson E.J., Bernard E., Bouchard S., Debaeremaeker
K.R., Ter Hofstede H., Hollis L., Lausen C.L., Lister J.S., Rambaldini D., Ratcliffe J.M. and Reddy E.
2002. Researching little-known species: The African bat Otomops martiensseni (Chiroptera:
Molossidae). Bioiversity and Conservation, 11: 1583 - 1606.
Fiedler J.K. 2004. Assessment of bat mortality and activity at Buffalo mountain windfarm, eastern
Tennessee. Master of Science Degree final Thesis. The University of Tennessee, Knoxville.
Gannon W.L., R.E. Sherwin, and S. Haymond. 2003. On the Importance of Articulating
Assumptions When Conducting Acoustic Studies of Habitat Use by Bats. Wildlife Society Bulletin
31: 45-61.
Gauteng and Northern Regions Bat Interest Group. 2012. Call Parameters of some Southern
African bats. http://www.batsgauteng.org.za/Bat_Calls_SA.pdf (consulted in 04 of January 2013).
Griffin D. R., Novick A. and Kornfield M. 1958. The Sensitivity of Echolocation in the Fruit Bat,
Rousettus. Biological Bulletin. 115: 107 - 113.
Gsänger, S. & Pitteloud, J.D. 2013. Annual Report 2012. World Wind Energy Association. Bonn.
www.wwindea.org
Hagen E.M. and Sabo J.L. 2012. Influence of river drying and insect availability on bat activity along
the San Pedro River, Arizona (USA). The Journal of Arid Environments 84: 1 – 8.
Hauge K. 2010. Bat (Chiroptera) activity and community composition in contrasting agricultural
landscapes and the adjacent Budongo forest reserve, Uganda. Master Thesis in Biology – Biodiversity,
Evolution and Ecology. University of Bergen.
Hayes J. P. 1997. Temporal variation in activity of bats and the design of echolocation-monitoring
studies. Journal of Mammalogy 78:514–524. doi:10.2307/1382902
Hayes J. P. 2000. Assumptions and practical considerations in the design and interpretation of
echolocation-monitoring studies. Acta Chiropterologica 2:225–236.
Hecker K.R. and Brigham R.M. 1999. Does moonlight change vertical stratification of activity by
forest-dwelling insectivorous bats? Journal of Mammalogy, 80(4): 1196-1201.
Hein C. D., J. Gruver, and E. B. Arnett. 2013. Relating pre-construction bat activity and post-
construction bat fatality to predict risk at wind energy facilities: a synthesis. A report submitted to the
National Renewable Energy Laboratory. Bat Conservation International, Austin, TX, USA.
Heithaus E. R. and Fleming T. H. 1978. Foraging Movements of a Frugivorous Bat, Carollia
perspicillata (Phyllostomatidae). Ecological Monographs 48(2): 127-143.
132 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Horn, J. W., Arnett, E. B., Kunz, T. K. 2008. Behavioral Responses of Bats to Operating Wind
Turbines. The Journal of Wildlife Management, 72 (1): 123-131.
Hosmer D. and Lemeshow 2000. Applied Logistic Regression (2. ed). New York. Wiley. IUCN.
2012. The IUCN Red List of Threatened Species. http://www.iucnredlist.org/ (consulted in 04 of
January 2013).
Hundt, L. (2012). Bat surveys: Good practice guidelines, 2nd edition. Bat conservation Trust.
Hunt, W.G. 2002. Golden eagles in a perilous landscape: Predicting the effects of mitigation for
energy-related mortality. California Energy Commission Report P500-02-043F.
Hyder, 1999. Consulting guidelines for the assessment of indirect and cumulative impacts as well as
impact interactions. Brussels: EC DGX1 Environment, Nuclear Safety and Civil Protection.
IUCN. 2012. The IUCN Red List of Threatened Species. http://www.iucnredlist.org/ (consulted in 04
of January 2013).
Jenkins, E. V., Laine, T., Morgan, S. E., COLE, K. R., Speakman, J. R. 1998. Roost selection in the
pipistrelle bat, Pipistrellus pipistrellus (Chiroptera: Vespertilionidae), in northeast Scotland. Animal
Behaviour, 56: 909–917
Johnson, G., Erickson, W., Strickland, M.D., Shepherd, M.F., Shepherd, D.A. & Sarappo, S.A. 2003.
Mortality of bats at a large-scale wind power development at Buffalo Ridge, Minnesota. Am. Midl.
Nat. 50: 332-342.
Kalka M.B., Smith A.R. and Kalko E.K.V. 2008. Bats Limit Arthropods and Herbivory in a Tropical
Forest. Science Magazine 320: 71
Kikuchi, R. 2008. Adverse impacts of wind power generation on collision behaviour of birds and
anti-predator behaviour of squirrels. J. Nat. Conserv. 16: 44-55.
Kopsinis, Y., Aboutanios, E., Waters, D., McLaughlin, S. 2009. Time-frequency and advanced
frequency estimation techniques for the investigation of bat echolocation calls. Acoustical Society of
America. 127(2): 1124–1134.
Kunz T. and Fenton M. 2003. Bat Ecology. The university of Chicago Press, Chicago.
Kunz T.H., Arnett E.B., Cooper B.M., Erickson W.P., Larkin R.P., Mabee T., Morrison M.L.,
Strickland M.D. and Szewczak J.M. 2007. Assessing Impacts of Wind-Energy Development on
Nocturnally Active Birds and Bats: A Guidance Document. Journal of Wildlife Management, 71(8):
2449–2486.
Kunz T. H., Torrez E. B., Bauer Dana, Lobova T. and Fleming T.H., 2011. Ecosystem services
provided by bats. Annals of the New York Academy of Sciences, 1223: 1 - 38.
133 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Lang A. B., Kalko E. K. V., Romer H., Bockholdt, C., Dechmann D. K. N. 2005. Activity levels of
bats and katydids in relation to the lunar cycle. Oecologia 146: 659– 666.
Limpens, H. J. G. A. & G. f. MccrAcken. 2002. Choosing a bat detector: theoretical and practical
aspects. Pp 28-37 in Bat Echolocation Research. Tools, techniques and analysis (R. M. Brigham, E.
K. V. Kalko, G. Jones, S. Parsons, and H. J. G. A. Limpens). Bat Conservation International. Austin,
EE.UU.
Loeb S. C. and Waldro, T.A. 2008. Bat activity in relation to fire and fire surrogate treatments in
southern pine stands. Forest Ecology and Management, 255: 3185 - 3192.
Lloyd A., Law B.S. and Goldingay R.L. 2006, Bat activity on riparian zones and upper slopes in
Australian timber production forests and the effectiveness of riparian buffers, Biological
Conservation. 129: 207 – 220.
Madders, M. & Whitfield, D.P. 2006. Upland raptors and the assessment of wind farm impacts. Ibis.
148: 43-56.
Martínez, J. E., Calvo, J. F., Martínez, J. A., Zuberogoitia, I., Cerezo, E., Manrique, J., Gómez, G. J.,
Nevado, J. C., Sánchez, M., Sánchez, R., Bayo, J., Pallarés, A., González, C., Gómez, J. M., Pérez, P.
& Motos, J. 2010. Potential impact of wind farms on territories of large eagles in southeastern
Spain. Biodivers Conserv 19: 3757–3767.
Masden, E. A., Fox, A. D., Furness, R. W., Bullman, R., Haydon, D. T. 2009. Cumulative impact
assessments and bird/wind farm interactions: Developing a conceptual framework. Environmental
Impact Assessessment Review, 30 (1): 1-7.
Miller B.W. 2001. A method for determining relative activity of free flying bats using a new activity
index for acoustical monitoring. Acta Chiropterologica, 3:93-105.
Meyer C.F.J., Schwarz C.J. and Fahr J. 2004. Activity patterns and habitat preferences of
insectivorous bats in a West African forest–savanna mosaic. Journal of Tropical Ecology, 20:397–407.
Monadjem A. 2005. Survival and roost-site selection in the African bat Nycteris
thebaica (Chiroptera: Nycteridae) in Swaziland. Belgian J Zool 135(supplement): 103–107.
Monadjem A., Taylor P., Cotterill,F.P.D. and Schoeman. 2010. Bats of Southern and Central Africa.
Wits University Press, Johannesburg.
Mucina L., and Rutherford, M.C. (eds) 2006. The Vegetation of South Africa, Lesotho and Swaziland.
South African National Biodiversity Institute, Pretoria. 807 pp. ISBN-13: 978-1-919976-21-1 &
ISBN-10: 1-919976-21-3
Mukasa A. D., Mutambatsere E., Arvanitis Y., and Triki T. 2013. Development of Wind Energy in
Africa. African Development Bank. Working Paper Series No 170. Belvédère.
134 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
O’Donnel, C. F. J. 2000. Influence of season, habitat, temperature, and invertebrate availability on
nocturnal activity of the New Zealand long‐tailed bat (Chalinolobus tuberculatus). New Zealand
Journal of Zoology, 27 (3): 207-221.
O’Farrell, M. J. 1997. Use of echolocation calls for the identification of free-flying bats. Transactions
of the Western Section of the Wildlife Society, 33:1-8.
O’Farrell, M. J. and W. L. Gannon. 1999. A comparison of acoustic versus capture techniques for
the inventory of bat. Journal of Mammalogy, 80:24-30.
O’Farrell, M. J., B. W. Miller, and W. L. Gannon. 1999. Qualitative identification of free-flying bats
using the Anabat detector. Journal of Mammalogy. 80:11-23.
Pierce, M. W. 2012. Assessing bat (Chiroptera) diversity: determinants of assemblage and
ensemble structure at Kwalata Game Ranch, Gauteng, South Africa. Master Thesis in Science.
Faculty of Science, University of the Witwatersrand, Johannesburg.
Preatoni D.G., M. Nodari, R. Chirichella, G. Tosi, L. A. Wauters, and A. Martinonli. 2005.
Identifying bats from time-expanded recordings of search calls: comparing classification methods.
Journal of Wildlife Management. 69:1601–1614.
R Development Core Team (2009). R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-
project.org.
Rainho A. and Palmeirim J.M. 2011. The Importance of Distance to Resources in the Spatial
Modelling of Bat Foraging Habitat. PLoS ONE 6(4): e19227. doi:10.1371/journal.pone.0019227
Russo D. and Jones G. 2002. Use of foraging habitats by bats in a Mediterranean area determined
by acoustic surveys: conservation implications. Biological Conservation 107: 71-81.
Rutherford R.S. and Conlong D.E. 2010. Combating sugarcane pests in South Africa: From
researching biotic interactions to bio-intensive integrated pest management in the field. Proc. Int.
Soc. Sugar Cane Technol., 27: 1-17.
Sattler T., Bontadina F., Hirzel A.H. & Arlettaz R. 2007: Ecological niche modelling of two cryptic
bat species calls for a reassessment of their conservation status. J. Appl. Ecol. 44: 1188 – 1199.
Serra-Cobo J., López-Roig M., Marquès-Lopez T. and Lahuerta E. 2000. Rivers as possible
landsmarks in the orientation flight of Miniopterus schreibersii. Acta Theriologica. 45: 347 - 352.
Scott S. J., McLaren G., Jones G., Harris S. 2010. The impact of riparian habitat quality on the
foraging and activity of pipistrelle bats (Pipistrellus spp.). Journal of Zoology, 280: 371–378.
Simmons J. A., Fenton M. B., and O’Farrell M. J. 1979. Echolocation and pursuit of prey by bats,
Science 203 16–21.
135 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Smith E.P., Orvos D. R., Cairns J. 1993. Impact Assessment using the Before-After-Control-Impact
(BACI) Model: Concerns and Comments. Canadian Journal of Fisheries and Aquatic Sciences, 50:
627-637.
SNH. 2012. Assessing the cumulative impact of onshore wind farm developments. Scottish Natural
Heritage.
Schnitzler H.U. and Kalko E.K.V. 2001. Echolocation by insect eating bats. Bioscience 51, 557 –
569.
Sowler S. and Stoffberg S. 2012. South African Good Practice Guidelines for Surveying Bats in Wind
Farm Developments. Wildlife & Energy Programme of the Endangered Wildlife Trust.
Taylor P.B., Navarro R.A., Wren-Sargent M., Harrison J.A., Kieswetter S.L. 1999. TOTAL CWAC
Report: Coordinated Waterbird Counts in South Africa, 1992-97. Avian Demography Unit, Cape
Town.
Taylor P. J., Geiselman C., Kabochi P., Agwnda B., Turner S., 2005. Intraspecific variation in the
calls of some African bats (Order Chiroptera). Durban Museum Novitates 30: 24-37.
Tupinier Y. 1996. European bats: their world of sound. Societe Linneenne de Lyon, Lyon.
U.S. Fish and Wildlife Service 2012. Land-Based Wind Energy Guidelines. US Fish and Wildlife
Service, Arlinton, VA.
Verboom B. and Huitema H. 1997. The importance of linear landscape elements for the pipistrelle
Pipistrellus pipistrellus and the serotine bat Eptesicus serotinus. Landscape Ecology 2(12): 117-125.
Walters C.L., Freeman R., Collen A., Dietz C., Fenton M.B. 2012. A continental-scale tool for
acoustic identification of European bats. Journal of Applied Ecology 49: 1064–1074.
Waters D.A. and Vollrath C. 2003. Echolocation performance and call structure in the
megachiropteran fruit-bat Rousettus aegyptiacus. Acta Chiropterologica, 5(2): 209 – 219.
Weller, T. J., Baldwin, J. A. 2011. Using Echolocation Monitoring to Model Bat Occupancy and
Inform Mitigations at Wind Energy Facilities. The Journal of Wildlife Management 9999:1-13.
Weller, T. J., Baldwin, J. A. 2011. Using Echolocation Monitoring to Model Bat Occupancy and
Inform Mitigations at Wind Energy Facilities. The Journal of Wildlife Management 9999:1-13.
White, E. P., Gehrt, S.D. 2001. Effects of recording media on echolocation data from broadband
bat detectors. Wildlife Society Bulletin, 29: 974–978.
136 Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9. A P P E N D I C E S
- 1 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.1. A PPEND IX I - FIGUR ES
Figure 39 - Richards Bay wind energy facility framework.
- 2 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
- 3 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 40 – Sampling points and transects location.
- 4 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
- 5 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.2. A PPEND IX II - SA MPL ING POI NTS D ESCRI PTI ON
Point Description Photo
RB1
Biotope: Sugar cane plantation
Minimum distance to proposed turbine location:
300m
Orientation: NE
Altitude: 120m
Slope: 4.00%
Minimum distance to a water source: 400m
Minimum distance to a known roost: 650m
Average temperature: 23.3 ºC
RB2
Biotope: Sugar cane plantation
Minimum distance to proposed turbine location:
1100m
Orientation: N
Altitude: 60m
Slope: 12.90%
Minimum distance to a water source: 30m
Minimum distance to a known roost: 840m
Average temperature: 18.2 ºC
RB3
Biotope: Sugar cane plantation
Minimum distance to proposed turbine location:
200m
Orientation: NE
Altitude: 110m
Slope: 3.99%
Minimum distance to a water source: 200m
Minimum distance to a known roost: 1972m
Average temperature: 20.4 ºC
- 6 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Point Description Photo
RB4
Biotope: Sugar cane plantation
Minimum distance to proposed turbine location:
70m
Orientation: SW
Altitude: 100m
Slope: 3,83%
Minimum distance to a water source: 300m
Minimum distance to a known roost: 670m
Average temperature: 24.3 ºC
RB5
Biotope: Houses and trees around
Minimum distance to proposed turbine location:
600m
Orientation: SE
Altitude: 120m
Slope: 7.35%
Minimum distance to a water source: 250m
Minimum distance to a known roost: 40m
Average temperature: 25.4 ºC
RB6
Biotope: Sugar cane plantation
Minimum distance to proposed turbine location:
1100m
Orientation: S
Altitude: 40m
Slope: 3.57%
Minimum distance to a water source: 90m
Minimum distance to a known roost: 700m
Average temperature: 23.6 ºC
- 7 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.3. A PPEND IX III - SUMMA RY OF THE N UMBER OF RECO RDINGS ANAL YZED PER SPECI ES
Table 19 – Number of recording analysed per species and group of species from manual detection.
Species May July September October November December January February March April Total
Chaerephon pumilus 19 0 1 1 5 49 24 14 12 6 131
Chaerephon pumilus / Tadarida aegyptiaca 9 1 0 1 1 28 11 7 6 3 67
Eptesicus hotentotus 1 0 1 1 2 5 0 1 0 2 13
Eptesicus hotentotus / Scotophilus dinganii 1 0 0 0 0 0 2 0 0 0 3
Mops condylurus 0 0 0 0 0 1 0 0 0 0 1
Miniopterus fraterculus 1 0 0 0 0 0 0 0 0 0 1
Miniopterus natalensis 6 30 12 0 0 2 2 2 4 17 75
Neoromicia capensis 8 3 0 0 2 3 0 7 16 3 42
Neoromicia nana 16 5 2 1 1 2 0 0 8 4 39
Tadarida aegyptiaca 0 11 0 3 0 6 3 0 0 0 23
Miniopteridae 11 2 5 0 1 0 0 1 0 7 27
Miniopteridae / Vespertilionidae 11 1 0 0 0 0 11 6 51 29 109
Molossidae 7 1 0 0 0 2 0 0 0 0 10
Molossidae / Vespertilionidae 4 6 2 0 4 1 65 35 5 6 128
Vespertilionidae 1 0 0 0 4 0 0 0 0 0 5
Unidentified 0 0 1 0 0 0 0 0 0 2 3
Total 95 60 24 7 20 99 118 73 102 79 677
- 8 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 20 – Number of recording analysed per species and group of species from static detection.
Species May June July August September October November December January February March April Total
Chaerephon pumilus 3 0 10 44 17 2 1 4 0 11 0 0 92
Chaerephon pumilus/Eptesicus
hottentotus 4 0 1 18 1 3 0 1 0 2 0 0 30
Chaerephon pumilus/Eptesicus
hottentotus/Mops condylurus 0 0 0 0 0 0 0 6 1 0 0 0 7
Chaerephon pumilus/Mops condylurus 191 56 31 105 107 173 253 260 59 71 21 33 1360
Chaerephon pumilus/Tadarida
aegyptiaca 0 0 0 0 3 0 0 0 0 0 0 0 3
Eptesicus hottentotus 1 1 0 4 5 0 3 2 0 0 0 0 16
Eptesicus hottentotus/Scotophilus dinganii 5 2 0 4 7 5 11 20 1 6 1 1 63
Hypsugo anchietae/Pipistrellus hesperidus 0 0 0 2 3 0 0 0 0 0 0 0 5
Hypsugo anchietae/Pipistrellus sp. 1 1 2 18 14 1 0 67 7 2 0 0 113
Miniopterus fraterculus 0 0 4 5 5 1 0 1 1 0 0 0 17
Miniopterus fraterculus/Neoromicia nana 0 0 0 3 0 0 0 0 0 0 0 0 3
Miniopterus natalensis 0 1 4 15 47 0 6 1 0 0 0 0 74
Miniopterus natalensis/Pipistrellus sp. 0 0 0 0 0 1 0 0 0 0 0 0 1
Miniopterus natalensis/Pispistrellus
hesperidus 0 0 0 3 0 0 0 0 0 0 0 0 3
Mops condylurus 11 0 1 11 4 0 0 0 0 2 0 0 29
Mops condylurus/Tadarida aegyptiaca 14 0 1 29 7 0 0 0 0 2 0 0 53
Myotis tricolor 1 0 18 66 13 0 0 0 0 0 0 0 98
Myotis tricolor/Scotophilus viridis 0 0 0 2 0 0 0 0 0 0 0 0 2
Neoromicia capensis 0 0 58 39 3 10 0 29 2 9 1 1 152
Neoromicia capensis/Nycticeinops
schlieffeni 5 2 0 22 13 3 13 7 0 0 0 0 65
Neoromicia capensis/Nycticeinops
schlieffeni/Scotophilus viridis 0 0 0 3 0 0 0 0 0 0 0 0 3
Neoromicia capensis/Scotophilus viridis 0 0 0 6 0 0 0 0 0 0 0 0 6
- 9 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Species May June July August September October November December January February March April Total
Neoromicia nana 0 0 52 307 89 0 0 2 1 1 0 0 452
Neoromicia zuluensis/Pipistrellus
hesperidus 0 0 0 0 1 0 0 0 0 0 0 0 1
Pipistrellus hesperidus 0 0 0 1 2 0 0 0 0 0 0 0 3
Rhinolophus clivosus 0 0 0 1 0 0 0 0 0 0 0 0 1
Scotophilus dinganii 4 0 0 3 10 0 4 6 4 7 0 3 41
Scotophilus viridis 0 0 0 0 2 0 0 0 0 0 0 0 2
Tadarida aegyptiaca 47 19 14 73 45 62 265 85 1 0 0 2 613
Emballunoridae/Molossidae 0 0 0 77 0 0 0 0 0 0 0 0 77
Miniopteridae 0 0 0 0 0 0 0 2 0 0 0 0 2
Miniopteridae/Vespertilionidae 0 0 48 81 41 177 0 71 3 22 31 13 487
Molossidae 82 13 48 206 48 38 240 279 0 5 0 0 959
Molossidae/Vespertilionidae 0 0 0 14 0 0 0 5 42 17 4 5 87
Nycteridae/Vespertelionidae 0 0 0 29 0 0 0 0 0 0 0 0 29
Vespertilionidae 0 0 1 27 0 1 0 0 0 0 0 0 29
Unidentified 2 0 19 15 4 2 18 10 0 1 1 0 72
Total 371 95 312 1233 491 479 814 858 122 158 59 58 5050
- 10 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.4. A PPEND IX IV - CO LLISIO N RIS K ANA LYSIS FO R THE BAT SPE CIES OCCURR ING A T THE SIT E
Legend: * - Collision known in Europe and USA for species within the same genus (EUROBATS, 2013); x – characteristic attributed to the species; ? – possible characteristic of the species.
Family Specie Common name
Migration or
long
movements
Clutter
forager
Clutter-
edge
forager
Open air
forager
High
flight Low flight
Attracted
by light
Collision
known
(EUROBATS,
2013)
Risk of
collision
(Sowler &
Stoffberg,
2012)
HIPPOSIDERIDAE Cloeotis percivali Percival's short-heared
trident bat x Low
HIPPOSIDERIDAE Hipposideros caffer Sundevall's leaf-nosed
bat x Low
NYCTERIDAE Nycteris hispida Hairy silt-faced bat x Low
NYCTERIDAE Nycteris thebaica Egyptian silt-faced bat x x x Low
PTEROPODIDAE Eidolon helvum African straw-coloured
fruit bat x Medium-High
PTEROPODIDAE Epomophorus crypturus Peter's epauletted fruit
bat x Medium-High
PTEROPODIDAE Epomophorus wahlbergi Wahlberg's epauletted
fruitbat x Medium-High
PTEROPODIDAE Rousettus aegyptiacus Egyptian rousette x Medium-High
MINIOPTERIDAE Miniopterus fraterculus Lesser long-fingered bat x * Medium-High
MINIOPTERIDAE Miniopterus natalensis Natal long-fingeres bat x x * Medium-High
VESPERTILIONIDAE Eptesicus hottentotus Long-tailed serotine x ? * Medium
VESPERTILIONIDAE Glauconycteris variegata Variegated butterfly bat x Medium
VESPERTILIONIDAE Kerivoula argentata Damara woolly bat x Low
VESPERTILIONIDAE Kerivoula lanosa Lesser woolly bat x Low
VESPERTILIONIDAE Myotis tricolor Temminck's myotis x x * Medium-High
VESPERTILIONIDAE Nycticeinops schlieffeni Schlieffen's twilight bat x * Medium
VESPERTILIONIDAE Hypsugo anchietae Anchieta's pipistrelle x x ? * Low
VESPERTILIONIDAE Neoromicia capensis Cape serotine x ? * Medium-High
- 11 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Family Specie Common name
Migration or
long
movements
Clutter
forager
Clutter-
edge
forager
Open air
forager
High
flight Low flight
Attracted
by light
Collision
known
(EUROBATS,
2013)
Risk of
collision
(Sowler &
Stoffberg,
2012)
VESPERTILIONIDAE Neoromicia nana Banana bat x ? * Medium-High
VESPERTILIONIDAE Neoromicia zuluensis Zulu serotine ? ? * Medium-High
VESPERTILIONIDAE Pipistrellus hesperidus Dusky pipistrelle x ? * Medium
VESPERTILIONIDAE Scotophilus dinganii Yellow-bellied house bat x Medium-High
VESPERTILIONIDAE Scotophilus viridis Green house bat x Medium-High
RHINOLOPHIDAE Rhinolophus blasii Blasius's horseshoe bat x Low
RHINOLOPHIDAE Rhinolophus capensis Cape horseshoe bat x Low
RHINOLOPHIDAE Rhinolophus clivosus Geoffroy's horseshoe
bat x Low
RHINOLOPHIDAE Rhinolophus darlingi Darling's horseshoe bat x Low
RHINOLOPHIDAE Rhinolophus simulator Bushveld horseshoe bat x x Low
RHINOLOPHIDAE Rhinolophus swinnyi Swinny's horseshoe bat x Low
MOLOSSIDAE Chaerephon pumilus Little free-tailed bat x High
MOLOSSIDAE Mops condylurus Angolan free-tailed bat x High
MOLOSSIDAE Tadarida aegyptiaca Egyptian free-tailed bat x ? * High
EMBALLONURIDAE Taphozous mauritianus Mauritian tomb bat x High
EMBALLONURIDAE Taphozous perforatus Egyptian tomb bat x High
- 12 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.5. A PPEND IX V – BR IEF D ESCRIP TION O F BAT S PECIE S WITH OCC UR RENCE
IN R IC HARDS B AY SITE
Species with confirmed presence at the study area
1. Sundevall’s leaf-nosed bat (Hipposideros caffer) is widely distributed throughout southern
Africa, mostly in the Eastern Cape and KwaZulu-Natal. This species roosts in caves,
sinkholes and cavities, being able to share structures with other species. It is closely tied
to savannah woodland, where it is closely associated with riparian locations (Monadjem et
al., 2010).
2. Egyptian silt-faced bat (Nycteris thebaica) has been recorded in almost all southern African
countries, with the exception of Lesotho. This species roosts during the day in caves,
burrows, culverts and large trees. It can also be found in night roosts where the
individuals consume their prey and socialise with conspecifics (Monadjem et al., 2010).
This species habitat appears to be related with savannah and karoo biomes, avoiding open
grassland. Being a clutter forager, this specie forages at low altitudes.
3. Wahlberg’s epauletted fruit bat (Epomophorus wahlbergi) is one of the most abundant bat
species being widely recorded in the eastern parts of South Africa, including KwaZulu-
Natal. This species is associated with forest and forest-edge habitats and may travel
several kilometres to reach fruit trees (Monadjem et al., 2010).
4. Lesser long-fingered bat (Miniopterus fraterculus) is a widespread species in the eastern part
of South Africa, and has been recorded along the coast to southern and western
KwaZulu-Natal. This is a cave-dependent species using probably different locations for
winter hibernation and summer maternity roosts (Monadjem et al., 2010).
5. Natal long-fingered bat (Miniopterus natalensis) occurs widely in South Africa, however with
more records in the southern and eastern part, including KwaZulu-Natal. This species is
mostly associated with savannahs and bushlands, using these habitats as a clutter-edge
forager. As Miniopterus fraterculus, M. natalensis is a cave-dependent species, using different
locations for hibernation and reproduction (Monadjem et al., 2010).
- 13 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
6. Long-tailed serotine (Eptesicus hottentotus) occurs widely but sparsely, is a clutter-edge
forager that uses woodland and rocky regions. This species roosts in small groups, mainly
in caves and rock crevices (Monadjem et al., 2010).
7. Temminck’s myotis (Myotis tricolor) is a widespread species found in the eastern coast of
South Africa, and has been recorded from eastern to western KwaZulu-Natal. This is a
cave-dependent species, with separate winter and summer roosts, being therefore
associated with mountainous areas (Monadjem et al., 2010).
8. Anchieta’s pipistrelle (Hypsugo anchietae) occurs from coastal KwaZulu-Natal throughout
northern South Africa. This specie is associated with well-wooded locations, such as
riparian vegetation in savannah and woodland (Monadjem et al., 2010).Zulu serotine
(Neoromicia zuluensis) is possibly present in the study area, since records of this specie
have been found in northern KwaZulu-Natal. It appears to be associated with woodland
savannah and riparian habitats (Monadjem et al., 2010).
9. Cape serotine (Neoromicia capensis) has a widespread distribution in South Africa and
apparently tolerates a wide range of environmental conditions, being present in arid semi-
desert, grassland, forests and savannas, using clutter-edges for foraging. This species can
use diverse roosts, such as buildings, barks of trees and foliage, and usually a single or a
small number of individual occupy each roost (Monadjem et al., 2010).
10. Banana bat (Neoromicia nana) is a very abundant species that occurs in the eastern and
northern parts of South Africa, and has been recorded in the northern part of KwaZulu-
Natal. This species roosts in furled banana leaves, as well as roofs of houses. It is also
associated with riparian vegetation and forest patches (Monadjem et al., 2010).
11. Schlieffen’s twilight bat (Nycticeinops schlieffeni), occurs widely in the eastern and northern
parts of South Africa being already recorded in KwaZulu-Natal. It is usually associated
with riparian vegetation along rivers, and roosts in crevices in trees and in houses
(Monadjem et al., 2010).
12. Dusky pipistrelle (Pipistrellus hesperidus) has been recorded through KwaZulu-Natal, being
also widely recorded in the eastern parts of South Africa. It appears to be associated with
riparian vegetation and forest patches in the vicinities of water. Though its roosting habits
- 14 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
are not well known, this species has already been found in a crack in rocks (Monadjem et
al., 2010).
13. Yellow-bellied house bat (Scotophilus dinganii) is a widespread species, having already been
recorded in KwaZulu-Natal. It is associated with the savannah biome; however seems to
avoid open habitats such as bushlands. This species roosts in several structures, including
holes in trees and roofs of houses (Monadjem et al., 2010).
14. Green house bat (Scotophilus viridis) is mostly confined to Eastern Africa, though this species
can be under-sampled and have a larger distribution. It may occur in savannahs, avoiding
open habitats such as bushlands, perhaps due to the lack of roosting sites. The Green
house bat can use holes in trees and roofs of houses for roosting (Monadjem et al., 2010).
15. Little free-tailed bat (Chaerephon pumilus) is an abundant species in Eastern Africa, becoming
more infrequent in the Western Africa. This species roosts in groups of variable size,
naturally in narrow cracks in rocks or trees but can also be found in buildings, such as
house roofs and ceilings or other locations inside the building where cracks and narrow
crevices can be found (Monadjem et al., 2010).
16. Angolan free-tailed bat (Mops condylurus) is widespread and abundant. This specie chooses
narrow crevices in rock faces and caves as well as hollow trees. Is an open-air forager
often feeding throughout the night (Monadjem et al., 2010).
17. Egyptian free-tailed bat (Tadarida aegyptiaca) is abundant and widespread throughout
Southern Africa. This species roosts in small to medium-sized groups, from dozens to
hundreds of individuals. The preferred structures for roosting vary from caves to rock
crevices, hollow trees, and cracks in the bark of old trees. The Egyptian free-tailed bat can
also be found in buildings, mostly in house roofs (Monadjem et al., 2010).
18. Mauritian tomb bat (Taphozous mauritianus) is an open-air forager that prefers open
habitats and avoids closed forests. This species is present in a variety of savannah
woodlands, mostly in the eastern and western parts of South Africa. It is also known to
occur through KwaZulu-Natal (Monadjem et al., 2010).
19. Geoffroy's horseshoe bat (Rhinolophus clivosus) occurs widely in South Africa although being
absent from the arid interior. It roosts in caves and mine adits where it forms colonies of
- 15 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
several thousands of individuals. They use night roosts (usually a tree) where they
consume the captured prey. This species can be associated with arid savannah, woodland
and riparian forest. This bat is a clutter forager (Monadjem et al., 2010).
20. Bushveld horseshoe bat (Rhinolophus simulator) occurs from KwaZulu-Natal towards
northern Africa where it is found roosting in caves and mine adits, forming large colonies.
A colony of 150 individuals is known in Doornhoek Tunnel (near Pietermaritzburg),
where females stay during winter and males stay throughout the year. This specie is
associated with savannah woodland and riparian forest (Monadjem et al., 2010).
21. Welwitsch’s myotis (Myotis welwitschii) is sparsely distributed in the eastern part of South
Africa, existing two records of the species from KwaZulu-Natal and Free State. This
species has been found associated with mountain areas with woodland, roosting in foliage.
It’s a clutter-edge forager (Monadjem et al., 2010).
Species with potential occurrence at the study area
1. Percival's short-heared trident bat (Cloeotis percivali) is sparsely present from northern
KwaZulu-Natal, Swaziland and northern South Africa. Though this is not an abundant
species and few specimens have been collected it seems that individuals may roost in
narrow crevices and may be associated with woodland (Monadjem et al., 2010).
2. Hairy silt-faced bat (Nycteris hispida) is mostly found in the northern and eastern regions of
South Africa. This species can be found in northern KwaZulu-Natal, being therefore
possible to exist at Richards Bay wind energy facility site. For roosts this species may use
a great variety of habitats, not being restricted to dark cavities (Monadjem et al., 2010).
3. African straw-coloured fruit bat (Eidolon elvum) is a widely distributed species as a non-
breeding migrant. Though little information is known regarding this specie, individuals can
forage up to 59km from their roost. However no bat roosts of this species are known in
the study area vicinities (Monadjem et al., 2010).
4. Peter’s epauletted fruit bat (Epomophorus cryturus) is widespread and abundant in the eastern
parts of South Africa, having already been recorded in KwaZulu-Natal. Being a fruit bat,
- 16 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
this species does not echolocate. It is associated with forest-edge habitats, mostly riparian
forest (Monadjem et al., 2010).
5. Egyptian rousette (Rousettus aegyptiacus) is a large fruit eating bat, occurring from Cape
Town to the extreme southwest of South Africa and along the coast to KwaZulu-Natal.
It’s the only fruit eating bat that echolocate. Being totally dependent on caves to roost
this species occurrence is conditioned by its availability (Monadjem et al., 2010).
6. Variegated butterfly bat (Glauconycteris variegata) is a poorly known species, appearing to be
associated with forest and moist woodland (Monadjem et al., 2010).
7. Damara wooly bat (Kerivoula argentata) is mostly present in the eastern part of South
Africa, being recorded north of Durban throughout eastern KwaZulu-Natal. This specie
roosts in foliage, where its pelage aids as camouflage (Monadjem et al., 2010).
8. Lesser woolly bat (Kerivoula lanosa) can be found in the eastern parts of South Africa, from
the western tip towards KwaZulu-Natal along the coastal areas. Considering the little
information available on this species it is possible that its distribution is constrained by the
roosts availability (Monadjem et al., 2010).
9. Zulu serotine (Neoromicia zuluensis) occurs in the northern part of KwaZulu-Natal where it
has been found in woodland habitats, closely associated with riparian habitats. This is not
a very well known species, as their roosting preferences were not yet ascertain
(Monadjem et al., 2010).
10. Blasius's horseshoe bat (Rhinolophus blasii) is restricted to southern Africa, being found in
KwaZulu-Natal. It roosts in caves and mines where it forms small groups of individuals.
This specie appears to be associated with mountain and savannah habitats (Monadjem et
al., 2010).
11. Cape horseshoe bat (Rhinolophus capensis) is an endemic species of the extreme South
Africa southwest , with occurrence only between the Eastern Cape until the south border
of Namibia. This specie roosts in caves and mines forming colonies of a thousand of
individuals approximately. As a clutter forager this specie forages predominantly in the
canopy of trees, in fynbos and karoo biomes (Monadjem et al., 2010).
- 17 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
12. Darling's horseshoe bat (Rhinolophus darlingi) occurs mostly in the northern part of South
Africa, being also possible to be found in the Western Cape. The species is associated
with arid savannah in the west part of southern Africa. It roosts in caves and mine adits,
being able to form groups of several dozens of individuals. This bat specie is a clutter
forager (Monadjem et al., 2010).
13. Swinny’s horseshoe bat (Rhinolophus swinnyi) occurs from the Eastern Cape and KwaZulu-
Natal north through northeast South Africa. The distribution of this specie appears to be
limited by the availability of adequate roosts, such as caves and old mines (Monadjem et
al., 2010).
14. Large-eared giant mastiff bat (Otomops martiensseni) has a very localised distribution in
southern Africa, being regularly observed in the Durban area. This specie roosts in the
roofs of houses in the Durban area, though in Kenya it is known to use caves. It prefers
woodland types where hunts for Lepidoptera as an open-air forager (Monadjem et al.,
2010).
15. Egyptian tomb bat (Taphozous perforatus) southern African population is very poorly
known. This specie appears to be however present in savannah woodlands with open
area, and roosts in dark crevices of rocky outcrops, sandstone overhangs and buildings
(Monadjem et al., 2010).
- 18 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.6. A PPEND IX VI – SU MMARY OF THE I NFORM ATION REC EIVED FO R TH E
CO MPILA TION O F THI S REP ORT
Table 21 – General information received and used to compile the present report.
Type of information Contents Date of reception
Weather data
Excel files with weather data: wind speed,
temperature, air humidity and precipitation
22nd March 2013
8th May 2013 10th May 2013
Wind Energy Facility Layout Shapefile with WEF turbine locations 25th March 2013
Sampling locations Shapefile with sampling locations; Manual transects; Static detectors; live-trapping;
roosts locations
25th March 2013 26th March 2013
27th March 2013 04th April 2013
Bat Equipment Installation Report NSS report (pdf) 27th March 2013
3rd Bat Progress Report (NSS) NSS report (pdf) 05th April 2013
Wind turbine specifications E-mail communication on the wind turbine
to be implemented in the WEF 10th May 2013
Detectors settings Excel files with the SM2+ bat detector
configurations and record times 20th May 2013
Static detectors details
E-mail communication concerning the
status of the detectors and major problems encountered
31st May 2013
Roost and live-trapping information E-mail communication concerning the NSS observations on roosts and live-trapping
6th June 2013
Table 22 – Total recordings (*.wac files) from static detection. Number of files refers to the *.wac files
received, prior to the conversion to *.wav format and screening process. One *.wac file may contain several
bat passes and noise recordings.
Sampling detector
Folder Sub-folder Number of files
Size (Gb)
RB1
20130501_RB1_WAC (PT) - 561 18.7
RB1_20120427 to 20120430 - 43 1.63
April-May_2012 - 167 7.85
February_2013
Fev_2013 15 0.995 (995Mb)
RB1-1-19022013 117 28.7
RB1-1-20022013 16 0.995 (995Mb)
RB1-2-19022012 113 28.2
RB1-3-19022013 111 28.1
RB1-4-19022012 55 12.8
January_2013 20130123_RB1_WAC 679 115
July-August_2012 - 103 1.41
RB1_20120619 to 20120718 - 449 8.19
June-July_2012 - 52 3.3
- 19 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Sampling
detector Folder Sub-folder
Number
of files Size (Gb)
March_2013 20130327_RB1_WAC 509 19.7
November_2012
RB-1-1-05122012 126 9.74
RB-1-2-05122012 197 17.5
RB-1-3-05122012 60 6.46
October_2012 201210_RB1_WAC 261 20.2
September_2012 201209_RB1_WAC 136 5.29
RB2
August_2012 RB2 309 45.2
RB2_20120815 to 20120910 - 347 64.2
October_2012 - 6 1.87
Rb2_test_Data_July_2012 - 10 0.114 (114 Mb)
September_2012 RB2-3-15092012 42 9.53
RB3 RB3_July-August 2012_analise - 102 2.25
RB3_test_data_Jul_2012_analise - 6 0.0907 (90.7Mb)
RB4A
20130501_RB4Top_WAC - 946 21.9
December_2012 RB2A-1-06122012 37 0.666 (666Mb)
Jan 2013_RB4A 20130123_RB4A_WAC 1269 26.9
March_2013_RB4A 20130327_RB4A-Top_WAC 922 11.7
RB4A February RB4A-1-19022013 786 22.5
RB4B
December_2012 RB2B-1-06122012 37 0.666 (666Mb)
Jan 2013_RB4B 20130123_RB4B_WAC 660 30
March_2013_RB4B 20130327_RB4B-Bot_WAC 77 2.05
RB4B February RB4B-1-19012013 388 20.4
RB5
December_2012 RB3-1-06122012 25 1.93
RB3-1-20122012 124 7.71
January_2013 20130115_RB5_WAC 159 9.87
RB3-1-03012013 139 15.1
March_2013_RB5 20130325_RB5_WAC 50 1.92
RB5 February - 158 14.4
RB6
20130116_RB6_WAC - 157 7.67
December_2012 RB4-1-06122012 14 1.15
RB4-1-20122012 146 5.89
RB6 February RB6-1-2013 149 5.52
RB6_20130220 to 20130304 - 153 6.89
- 20 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Table 23 – Total recordings (*.wac files) from manual detection. Number of files refers to the *.wac files
received, prior to the conversion to *.wav format and screening process. One *.wac file may contain several
bat passes and noise recordings.
Month Folder Sub-folder Number of files
Size (Gb)
May_2012 RB Transect 1 - 158 0.0689 (68.9Mb)
RB Transect 3 - 433 0.142 (142 Mb)
Jul_2012 RB Transect 4 - 14 0.0326 (32.6 Mb)
RB Transect 5 reversed - 9 0.249 (249 Mb)
Sep_2012 13092012 - 21 0.485 (485 Mb)
Oct_2012 RB partial transect 24102012 - 8 0.0729 (72.9 Mb)
Nov_2012 29 November Richards Bay - 13 0.210 (210 Mb)
30 November Richards Bay - 16 0.346 (346 Mb)
Dec_2012
1 December Richards Bay - 13 0.487 (487 Mb)
2 December Richards Bay - 13 0.428 (428 Mb)
4 December RB Canefields - 4 0.0489 (48.9 Mb)
Jan_2013
RB19 1742_RB_Transect_20130119_WAC 41 0.330 (330 Mb)
RB20 1742_RB_Transect_20130120_WAC 13 0.364 (364 Mb)
RB21 1742_RB_Transect_20130121_WAC 23 0.165 (165 Mb)
RB22 1742_RB_Transect_20130122_WAC 15 0.485 (485 Mb)
Feb_2013 RB19
18022013 15 0.283 (283 Mb)
19022013 11 0.232 (232 Mb)
RB20 20022013 10 0.127 (127 Mb)
Mar_2013 20130325_RB_Transect_WAC - 12 0.538 (538 Mb)
20130326_RB_Transect_WAC - 10 0.0392 (39.2 Mb)
Apr_2013 20130105_RB_Transect_WAC - 12 0.182 (182 Mb)
20130205_RB_Transect_WAC - 17 0.469 (469 Mb)
- 21 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Figure 41 – Timeline of the information received from the detectors placed at Richards Bay Wind Energy Facility site.
Date
27-04-2012
30-4-2012
30-4-2012
30-4-2012
01-5-2012
02-5-2012
03-5-2012
04-5-2012
05-5-2012
06-5-2012
07-5-2012
08-5-2012
09-5-2012
10-5-2012
11-5-2012
12-5-2012
13-5-2012
14-5-2012
15-5-2012
16-5-2012
17-5-2012
18-5-2012
21-5-2012
22-5-2012
23-5-2012
24-5-2012
25-5-2012
26-5-2012
27-5-2012
29-5-2012
30-5-2012
31-5-2012
01-6-2012
02-6-2012
03-6-2012
04-6-2012
06-6-2012
07-6-2012
08-6-2012
09-6-2012
10-6-2012
11-6-2012
12-6-2012
13-6-2012
14-6-2012
15-6-2012
16-6-2012
17-6-2012
18-6-2012
19-6-2012
18-7-2012
19-7-2012
20-7-2012
21-7-2012
22-7-2012
23-7-2012
24-7-2012
25-7-2012
26-7-2012
27-7-2012
28-7-2012
29-7-2012
30-7-2012
31-7-2012
01-8-2012
02-8-2012
03-8-2012
04-8-2012
05-8-2012
06-8-2012
07-8-2012
08-8-2012
09-8-2012
10-8-2012
11-8-2012
12-8-2012
13-8-2012
14-8-2012
15-8-2012
16-8-2012
17-8-2012
18-8-2012
19-8-2012
20-8-2012
21-8-2012
22-8-2012
23-8-2012
24-8-2012
25-8-2012
26-8-2012
27-8-2012
28-8-2012
29-8-2012
30-8-2012
31-8-2012
01-9-2012
02-9-2012
03-9-2012
04-9-2012
08-9-2012
09-9-2012
10-9-2012
11-9-2012
12-9-2012
13-9-2012
14-9-2012
15-9-2012
16-9-2012
17-9-2012
18-9-2012
19-9-2012
20-9-2012
21-9-2012
22-9-2012
23-9-2012
24-9-2012
25-9-2012
26-9-2012
27-9-2012
28-9-2012
29-9-2012
30-9-2012
01-10-2012
02-10-2012
03-10-2012
04-10-2012
05-10-2012
06-10-2012
07-10-2012
08-10-2012
09-10-2012
10-10-2012
11-10-2012
12-10-2012
13-10-2012
14-10-2012
15-10-2012
16-10-2012
17-10-2012
18-10-2012
19-10-2012
20-10-2012
21-10-2012
22-10-2012
23-10-2012
24-10-2012
25-10-2012
26-10-2012
27-10-2012
28-10-2012
29-10-2012
30-10-2012
31-10-2012
01-11-2012
02-11-2012
03-11-2012
04-11-2012
05-11-2012
06-11-2012
07-11-2012
08-11-2012
09-11-2012
10-11-2012
11-11-2012
12-11-2012
13-11-2012
14-11-2012
15-11-2012
16-11-2012
17-11-2012
18-11-2012
19-11-2012
20-11-2012
21-11-2012
22-11-2012
23-11-2012
24-11-2012
25-11-2012
26-11-2012
27-11-2012
28-11-2012
29-11-2012
30-11-2012
01-12-2012
02-12-2012
03-12-2012
04-12-2012
05-12-2012
06-12-2012
07-12-2012
08-12-2012
09-12-2012
10-12-2012
11-12-2012
12-12-2012
13-12-2012
14-12-2012
15-12-2012
16-12-2012
17-12-2012
18-12-2012
19-12-2012
20-12-2012
21-12-2012
22-12-2012
23-12-2012
24-12-2012
25-12-2012
26-12-2012
27-12-2012
28-12-2012
29-12-2012
30-12-2012
31-12-2012
01-1-2013
02-1-2013
03-1-2013
04-1-2013
05-1-2013
06-1-2013
07-1-2013
08-1-2013
09-1-2013
10-1-2013
11-1-2013
12-1-2013
13-1-2013
14-1-2013
15-1-2013
16-1-2013
17-1-2013
18-1-2013
19-1-2013
20-1-2013
21-1-2013
22-1-2013
23-1-2013
24-1-2013
25-1-2013
26-1-2013
27-1-2013
28-1-2013
29-1-2013
30-1-2013
31-1-2013
01-2-2013
02-2-2013
03-2-2013
04-2-2013
05-2-2013
06-2-2013
07-2-2013
08-2-2013
09-2-2013
10-2-2013
11-2-2013
12-2-2013
13-2-2013
14-2-2013
15-2-2013
16-2-2013
17-2-2013
18-2-2013
19-2-2013
20-2-2013
21-2-2013
22-2-2013
23-2-2013
24-2-2013
25-2-2013
26-2-2013
27-2-2013
28-2-2013
01-3-2013
02-3-2013
03-3-2013
04-3-2013
05-3-2013
06-3-2013
07-3-2013
08-3-2013
09-3-2013
10-3-2013
11-3-2013
12-3-2013
13-3-2013
14-3-2013
15-3-2013
16-3-2013
17-3-2013
18-3-2013
19-3-2013
20-3-2013
21-3-2013
22-3-2013
23-3-2013
24-3-2013
25-3-2013
26-3-2013
27-3-2013
28-3-2013
29-3-2013
30-3-2013
31-3-2013
01-4-2013
02-4-2013
03-4-2013
04-4-2013
05-4-2013
06-4-2013
07-4-2013
08-4-2013
09-4-2013
10-4-2013
11-4-2013
12-4-2013
13-4-2013
14-4-2013
15-4-2013
16-4-2013
17-4-2013
18-4-2013
19-4-2013
20-4-2013
21-4-2013
22-4-2013
23-4-2013
24-4-2013
25-4-2013
26-4-2013
27-4-2013
28-4-2013
29-4-2013
30-4-2013
01-5-2013
RB1
RB2
RB3
RB4A
RB4B
RB5
RB6
- 22 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.7. A PPEND IX VII – POT ENTIAL BAT CO LLISI ON R ISK WI TH WIND TU RBI NES
FR OM AC CORDIN GLY T O SO WLER AND STOFF BERG (2012)
- 23 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.8. A PPEND IX VIII – TUR BINE IMPA CT A NALYSIS RES ULTS
Turbine
Criteria
Group
1 2 3 4 5 6 7
Roosts
Buffer
around
roosts
Buffer
around
reproduction
roosts
Buffer
around
bat
features
Riverine
vegetation
and dams
Native
vegetation
High
Activity
areas
WTG 40
P
Roosting sensitive turbine
WTG 41
P
Roosting sensitive turbine
WTG 43
P
Roosting sensitive turbine
WTG 45
P
Roosting sensitive turbine
WTG 47
P
Roosting sensitive turbine
WTG 48
P
Roosting sensitive turbine
WTG 1
P
Habitat sensitivive turbine
WTG new 4
P P P
Habitat sensitive turbine
WTG 8
P P P
Habitat sensitive turbine
WTG 9
P P P
Habitat sensitive turbine
WTG 10
P P P
Habitat sensitive turbine
WTG 11
P
P
Habitat sensitive turbine
WTG new 5
P P P
Habitat sensitive turbine
WTG 12
P P
Habitat sensitive turbine
WTG 13
P
P
Habitat sensitive turbine
WTG 15
P P P
Habitat sensitive turbine
WTG 16
P P P
Habitat sensitive turbine
WTG 18
P P
Habitat sensitive turbine
WTG 34
P P
P Habitat sensitive turbine
WTG new 1
P P
Habitat sensitive turbine
WTG 21
P P
Habitat sensitive turbine
WTG 22
P P
Habitat sensitive turbine
WTG 25
P
P
Habitat sensitive turbine
WTG 32
P
Habitat sensitive turbine
WTG 33
P
Habitat sensitive turbine
WTG 3
P
P P P P Roosting & Habitat
sensitives
WTG 4
P
P P
Roosting & Habitat
sensitives
WTG 14
P
P P P
Roosting & Habitat
sensitives
WTG 17
P
P P
Roosting & Habitat
sensitives
WTG 26
P
P
P P Roosting & Habitat
sensitives
- 24 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
WTG 31
P
P P
P Roosting & Habitat
sensitives
WTG new 3
Non sensitive
WTG 20
Non sensitive
WTG new 2
Non sensitive
WTG 36
Non sensitive
WTG 38
Non sensitive
WTG new 6
Non sensitive
WTG 42
Non sensitive
WTG 46
Non sensitive
- 25 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
9.9. A PPEND IX IX – PR OPOSED BAT MONIT ORI NG PR OGRAM ME
9.9.1. O BJECT IVES
The primary aims of this monitoring program are the assessment of the potential impacts resulting
from the construction and operation of the wind energy facility over the bat community in the
study area. Therefore the main objectives of this monitoring program are:
a) Identify the potential changes in the bat community present within the Richards Bay wind
energy facility site and the eventual exclusion effect (avoidance of the wind facility area
during the operational phase of the project);
b) Assess the use of roosts in the wind energy facility development footprint as well as the
surrounding area;
c) Quantify bat fatalities associated with the wind energy facility during the operation phase
of the project;
d) Propose measures to monitor mitigate or, if unavoidable, compensate identified potential
impacts.
In order to meet these objectives the same general methodological approach implemented during
the pre-construction phase of the project should be implemented.
Additionally, it is recommended that vehicle based transects are used mainly as exploratory
methodology. Manual surveys should be conducted through implementation of fixed sampling
points distributed within the wind energy facility area (may be defined along transects), covering
all the different habitats present in the area. This sampling points should be conducted at least
once each survey. A control area, with similar characteristics of the wind energy facility area
should be defined and sampling points conducted with the same frequency and in similar biotopes.
The methodologies to be implemented should follow the general guidelines presented in the South
African Good Practice Guidelines for Surveying Bats in Wind Farm Developments (Sowler and Stoffberg,
2012).
9.9.2. M ONITO RING PROTO COLS
The overall monitoring program should be implemented throughout every phase of the wind
energy facility project for at least three years after the facility becomes operational. The
monitoring programme should be revised after this period, considering the results obtained its
continuation should be evaluated.
The methodological approach to be implemented should be similar to the one implemented
during the pre-construction phase and to which this report refers to (see section 2). It is
proposed that the manual surveys should be implemented by means of sampling points and the
- 26 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
general guidelines for this methodology are presented bellow. This methodology shouldn’t
substitute the vehicle based transects methodology already being implemented.
Bellow the general guidelines for the additional methodologies, not included in the methodological
approach presented in this report (section 2) necessary to implement during the construction and
operational phase of the project is presented.
9.9.2.1. M anu al det ect ion
The bat monitoring should be implemented in order to evaluate the activity patterns at the wind
energy facility site and, at least, one control area. Collecting this information, should allow:
Determination of the bat species that use the site;
Determination of bat activity index;
9.9.2.1.1. Methodology
The methodology to be implemented should follow the general guidelines presented in the South
African Good Practice Guidelines for Surveying Bats in Wind Farm Developments (Sowler and Stoffberg,
2012). The manual detection of ultrasounds should be conducted with a manual ultrasound
detector that allows saving the bat recordings for further analysis and identification.
Manual surveys should comprise sample points of, at least, 5 minutes each, along vehicle transects.
Each point should be characterised according to: minimum distance to the future turbines, slope,
dominant orientation, biotope, minimum distance to a water source and minimum distance to
known roosts, lunar phase, cloudiness, temperature and wind (speed and direction). At each 5
minute sampling point, all bat passes17 heard and observed should be recorded, as well as all the
passes detected between points. The surveys should start 30 minutes before the sunset ensuring
that bat species that emerge early in the evening can be included in the surveys (Sowler and
Stoffberg, 2012).
9.9.2.1.2. Sampling locations and Sampling perio ds
Transects should be established in the wind energy facility and in a separate and similar control
area(s), crossing the main biotopes present in the area. In each transect the sampling
points,should be established with a minimum distance of 200m, in between each other to avoid
pseudo-replication.
Surveys should be conducted at least once a month (a minimum of one survey per month). Each
sampling point should be conducted at least once per month for at least a full calendar year during
17 Contacts with bats detected by visual observation or ultrasonic detection of bat calls.
- 27 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
the construction phase, and at least three years after the project becomes operational
(operational phase).
9.9.2.2. B at mor tal ity
9.9.2.2.1. Methodology
The methodology to be implemented should follow the general guidelines presented in the Best
South African Good Practice Guidelines for Surveying Bats in Wind Farm Developments (Sowler and
Stoffberg, 2012) and the international best practices.
At onshore facilities the fatality estimation is based on carcass searches around wind turbines.
However, the number of carcasses found during the searches do not correspond to the real
number of bats killed by the wind farm, since not all carcasses are detected by searchers or, some
carcasses are removed given the time elapsed between searches, (e.g. by scavengers or decay)
from the site. Thus, to estimate the real mortality it is necessary to determine the associated bias
correction factor and adjust the observed mortality through the use of appropriate fatality
estimators.
Whenever bat and bird monitoring plans are simultaneously being implemented at a wind energy
facility the bat collisions and bird collisions assessment could be combined, following the same
general methodological approach.
9.9.2.2.1.1 Carcass searches
Regarding bat mortality evaluation, searches of dead bats around all the wind energy facility wind
turbines during the operational phase are proposed. The search plot will depend on the wind
turbine characteristics (hub height and rotor diameter) and should be larger than the area
covered by the rotor diameter with an addition of at least 5 meters. This area should be regularly
inspected for bat casualties. The observer should adjust its dislocation speed to the terrain
characteristics, inspecting as much area as possible. According to the terrain characteristics the
observer may conduct the survey through parallel transects, or by dividing the area in four
different quadrants, and carefully searching for any signs of bat collision incidents (carcasses,
dismembered body parts, injured bats). All evidence should be documented, being the evidence
collected in adequate preserving conditions, for further laboratory analysis.
9.9.2.2.1.2 Searcher efficienc y and carcass remov al trials
Field trials should be conducted to determine the observed mortality correction parameters such
as carcass detection by observers and carcass removal (e.g. by scavengers).
In carcass removal trials, carcasses should be placed at a minimum distance of 500 m from each
other. Once placed, carcasses should be checked to determine the time of removal for each one.
- 28 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
For the searcher efficiency trials, carcasses should be randomly placed around the turbines and
then searched by the observers in order to assess their efficiency rate.
In both trials, the type of carcasses used should mimic the dimensions and body size of the
existing wild species in the study area, such as rats.
9.9.2.2.2. Sampling locations and Sam pling perio ds
Mortality inspection, carcass detection and carcass removal should be implemented in the
operational phase of the project for at least three years, except if stated otherwise. All the
turbines implemented should be surveyed.
9.9.2.2.2.1 Carcass searches
Preferably the mortality inspection surveys should be conducted weekly (if not possible, then the
surveys must be conducted at least every 15 days, or monthly in the worst case scenario)
(Strickland et al., 2011), covering the whole annual period (Bernardino, 2008).
9.9.2.2.2.2 Searcher efficienc y and carcass remov al trials
The carcass removal trials should be performed during four seasons: winter, spring, autumn and
summer. In each campaign, the rat carcasses placed on site should be checked daily. The number
of carcasses used should be limited, in order not to attract too many scavengers.
In searcher efficiency trials, carcasses should be placed within the search plot of each turbine, if
the habitats have no significant variation throughout the year, the trial could only be performed
during one season of the year.
In order to obtain an accurate measure of the observed mortality, search efficiency rates and
scavenging rates should be assessed during the first operational year of the wind energy facility.
9.9.2.2.3. Data analysis
The results from the trials conducted should provide the evaluation of the following parameters:
Correction factor for carcass detection by field observers;
Correction factor for carcass removal by scavengers and environmental factors;
Real mortality estimates in the wind energy facility, during its operational phase.
To properly calculate the real mortality associated to the wind energy facility it is essential to
adopt a fatality estimator that adjusts the observed casualties by the estimated bias correction
terms. In the last years research has been conducted on this matter and several estimators have
been proposed. However, so far there is still lacking a universal estimator that ensures good
quality estimates under all circumstances (Bernardino et al. 2013).
- 29 - Bat Communi ty Moni tor ing – f ina l Report (pre- construct ion phase)
Therefore, when estimating the bat fatality associated to the wind energy facility the best
estimator available at the time should be used, which performance must be demonstrated in peer-
reviewed studies.
9.9.3. R EPORT S PRE PARAT ION A N D CON TENTS
A technical report containing the parameters referred to in the previous chapters should be
delivered at the end of each year of monitoring. In this document an evaluation of the adequacy of
monitoring protocols applied should be conducted, as well as an evaluation of the existence of any
detectable potential impacts occurring over the bat community in the impacted area, due to wind
energy facility and associated infrastructures. In these reports, a data comparison from results of
previous years should be performed, in order to obtain more reliable conclusions. For this
reason, the final monitoring program reports should present review of results obtained over the
previous years when the monitoring activities were implemented.