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


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.


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 summaÂ 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


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.


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.


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).


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.


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


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


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.


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;


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).


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).


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.


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).


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.


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).


Figure 3 – Bat caves located in the vicinities of Richards Bay wind energy facility.


Figure 4 - Nature reserves surrounding Richards Bay wind energy facility site.


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.


Figure 5 - Bat Sensitivity areas identified in the Richards Bay wind energy facility EIA report (from CES,

2012).


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).


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:

http://www.darklife.co.za/Caves/


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


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


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


(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


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


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.


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.


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


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


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.


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.


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).


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).


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.


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.


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”.


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.


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


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


Very Severe / Very

Beneficial

Very severe change to the

affected system(s) or party(ies)

A very substantial benefit to the


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


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


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;


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.


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.


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.


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.


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


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.



IUC

N

So

uth

Afr

ica R

ed

Lis

t


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


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



IUC

N

So

uth

Afr

ica R

ed

Lis

t


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


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.


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 *


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


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


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.


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


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


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-


Species

2012 2013

Total

May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. April


Chaerephon pumilus/Mops

condylurus 25 2 0 4 12 0 10 20 0 0 0 0 73


aegyptiaca 0 0 0 0 0 0 0 1 0 0 0 0 1


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


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



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


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).


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


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


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.


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).


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.


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


Nu

mb

er

of p

ass

es/

ho

ur

Low Height High Height


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


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


Nu

mb

er

of p

ass

es/

ho

ur

7m 30m 10m 60m


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


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)


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


Tem

pera

ture

(ºC

)

Nu

mb

er

of

pass

es/

gri

d c

ell

Month/Season

Average number of passes Temperature


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


Hu

mid

ity (

%)

Nu

mb

er

of

pass

es/

gri

d c

ell

Month/Season

Average number of passes Humidity


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


Win

d s

peed

(km

/h)

Nu

mb

er

of

pass

es/

gri

d c

ell

Month/Season

Average number of passes Wind speed


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.


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


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


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).


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


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


Avera

ge w

ind

sp

eed

(m

/s)

Nu

mb

er

of p

ass

es/

ho

ur

Average number of passes/hour Wind speed


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


Avera

ge t

em

pera

ture

(ºC

)

Nu

mb

er

of p

ass

es/

ho

ur

Average number of passes/hour Temperature


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


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


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%


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)


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%


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.


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


Nu

mb

er

of

pass

es/

gri

d c

ell

Month/Season


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


Nu

mb

er

of

pass

es/

ho

ur


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


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


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


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


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


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-


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


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.


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 *



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


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 ***


Response


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.


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).


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


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.


Figure 34 – Location of the roosts identified during pre-construction monitoring field surveys at the

Richards Bay wind energy facility site.


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


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


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.


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:


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.


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).


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.


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.


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


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


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


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.


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.

http://www.bendigoadvertiser.com.au/

http://bats.org.au/uploads/about-bats/batcarebrisbane_electrocution1.pdf


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


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.


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


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.


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.


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


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


Impact

Operational phase

Without mitigation Long term Localised Slight Definite Moderate

With

mitigation Long term Localised Slight May Occur Low

No-Go


With


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.


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


Impact

Operational phase

Without mitigation Long term Study area Severe Probable Moderate

With

mitigation Medium term Localized Moderate May Occur Low

No-Go


With



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).


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.


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.


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


Roosting

and

Habitat

sensitive

Without Permanent Study area Severe (Probable) High


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.


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


Impact

Operational phase

Without mitigation Long term Study area Moderate Probable Moderate

With

mitigation Long term Study area Slight Unlikely Low

No-Go


With



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.


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


Impact

Operational phase

Without mitigation Long term Regional Severe Probable Moderate

With

mitigation Medium term Study area Slight May Occur Low

No-Go


With


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.


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;


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


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).


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.


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


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).


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


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


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


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


confirmed bat roosts and, more importantly, will allow as well the establishment of the baseline

situation for the control area to be implemented.


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.


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.

http://www.wbwg.org/colorado/colorado.htm


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.


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.

http://www.batsgauteng.org.za/Bat_Calls_SA.pdf

http://www.wwindea.org/


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.

http://www.iucnredlist.org/

http://www.iucnredlist.org/


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.


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.

http://www.r-project.org./

http://www.r-project.org./


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.


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.


Figure 40 – Sampling points and transects location.


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



1100m

Orientation: N

Altitude: 60m

Slope: 12.90%




RB3



200m

Orientation: NE

Altitude: 110m

Slope: 3.99%





Point Description Photo

RB4



70m

Orientation: SW

Altitude: 100m

Slope: 3,83%




RB5

Biotope: Houses and trees around


600m

Orientation: SE

Altitude: 120m

Slope: 7.35%




RB6



1100m

Orientation: S

Altitude: 40m

Slope: 3.57%





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


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



hottentotus 4 0 1 18 1 3 0 1 0 2 0 0 30


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


aegyptiaca 0 0 0 0 3 0 0 0 0 0 0 0 3


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/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


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 viridis 0 0 0 0 2 0 0 0 0 0 0 0 2


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


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


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


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).


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


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


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


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,


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


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).


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



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


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


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)




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)


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


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)


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


WTG 43

P


WTG 45

P


WTG 47

P


WTG 48

P


WTG 1

P

Habitat sensitivive turbine

WTG new 4

P P P

Habitat sensitive turbine

WTG 8

P P P


WTG 9

P P P


WTG 10

P P P


WTG 11

P

P


WTG new 5

P P P


WTG 12

P P


WTG 13

P

P


WTG 15

P P P


WTG 16

P P P


WTG 18

P P


WTG 34

P P

P Habitat sensitive turbine

WTG new 1

P P


WTG 21

P P


WTG 22

P P


WTG 25

P

P


WTG 32

P


WTG 33

P


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


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


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


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.


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.


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).


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.

Richards Bay Wind Energy Facility - CESNET Bay Wind Energy Project TK06.… · Bio3 owns a Merlin...

Documents

Transcript of Richards Bay Wind Energy Facility - CESNET Bay Wind Energy Project TK06.… · Bio3 owns a Merlin...