CONTENTS 7. IMPACTS ON BIRDS Terms of reference Assumptions Sources of information 7 3

Transcription

1

2 CONTENTS 7. IMPACTS ON BIRDS Terms of reference Assumptions Sources of information DESCRIPTION OF ASPECTS THAT WOULD CAUSE IMPACTS ON BIRDS Wind turbines Electrical connection Construction of access roads Temporary construction area DESCRIPTION OF THE AFFECTED ENVIRONMENT Vegetation types and bird habitats Avifauna in the study area IDENTIFICATION OF KEY IMPACTS PERTAINING TO AVIFAUNA PERMIT REQUIREMENTS IMPACT ASSESSMENTS Collisions Mortalities resulting from collisions with the wind turbines Collisions with the proposed power line Displacement Displacement due to disturbance Habitat change and loss Mitigation measures Wind turbines Power line Monitoring requirements 7 15 Appendix 7A: Micro Habitats 7-19 Appendix 7B: Sensitivity Map 7-24 Table 7.1: Red listed species (excluding marine species) recorded in 3422BB and 3322DD by SABAP1 (Harrison et al 1997) and SABAP2 ( 7-8 Table 7.2: Impact assessment table 7-17 pg 7-1

3 7. IMPACTS ON BIRDS Terms of reference The terms of reference for the bird impact assessment study are as follows: Describe the affected environment and the determine status quo: The existing environment must be described and the bird communities most likely to be impacted must be identified. Different bird micro-habitats must be described as well as the species associated with those habitats. Indicate how a resource or community will be affected. Typical impacts that could be expected from the development must be listed as well as the expected impact on the bird communities. Impacts must be quantified (if possible) and a full description of predicted impacts (direct and indirect) must be provided. Gaps in baseline data must be highlighted and discussed. An indication of the confidence levels must be given. The best available data sources must be used to predict the impacts, and extensive use must be made of local knowledge (if available). Assessment of impacts: The potential impact on the birds must be assessed and evaluated according to the magnitude, spatial scale, timing, duration, reversibility, probability and significance. Propose and explain mitigation measures. Practical mitigation measures must be recommended and discussed. Summarise residual impacts after mitigation. An impact summary table must be provided, discussing expected impacts before and after mitigation. Indicate a monitoring programme. If a need for a monitoring programme is evident, it must be highlighted and a programme proposed. Mapping of sensitive areas: Bird sensitive areas must be mapped in a sensitivity map for easy reference Assumptions This study made the basic assumption that the sources of information used are reliable. However, it must be noted that the following factors may potentially detract from the accuracy of the predicted results: The SABAP1 data covers the period Bird distribution patterns fluctuate continuously according to availability of food and nesting habitat. Sources of error in the SABAP databases, particularly inadequate coverage of some quarter degree squares. This means that the reporting rates of species may not be an accurate reflection of the true densities in quarter degree squares that were sparsely covered during the data collecting period, as was the case with several of the squares (for a full discussion of potential inaccuracies in SABAP1 data, see Harrison et al, 1997). It must be noted that in this instance both the 3324DD and 3424BB quarter degree squares have been well covered with data being recorded on a combined total of 545 SABAP1 checklists, and 150 SABAP2 checklists. pg 7-2

4 Little detailed, verified information on the micro-habitat level was available on bird occurrence, densities and movements, therefore all conclusions are based on secondary sources. The only primary observations were those conducted during the site visit. Wind facilities are a relatively new development in South Africa. An extensive body of knowledge of avian interactions with wind generation facilities in a southern African context has yet to emerge; therefore strong reliance had to be placed on studies from overseas. Some speculation with regard to how South African birds are likely to interact with the proposed wind facility was therefore unavoidable. With certain classes of birds, particularly cranes and bustards, very little research has been conducted on potential impacts with wind facilities worldwide. The precautionary principle was therefore applied in assessing the potential impacts on species belonging to these classes. The World Charter for Nature, which was adopted by the UN General Assembly in 1982, was the first international endorsement of the precautionary principle. The principle was implemented in an international treaty as early as the 1987 Montreal Protocol and among other international treaties and declarations is reflected in the 1992 Rio Declaration on Environment and Development. Principle 15 of the Rio Declaration 1992 states that: in order to protect the environment, the precautionary approach shall be widely applied by States according to their capabilities. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall be not used as a reason for postponing costeffective measures to prevent environmental degradation. There have been few comprehensive studies, and even fewer published, peer-reviewed scientific papers on the impacts of wind farms on birds. Many studies suffer from a lack of before and after, or wind farm area and reference area comparisons, or a total lack of assessment of relevant factors such as collision risk, differences in bird behaviour between night and day, or are of inadequate duration to provide conclusive results (Langston & Pullen 2003). It is therefore inevitable that an element of speculation will enter the conclusions in this report, due to inconclusive and sometimes contradictory scientific evidence on the nature and extent of the impacts caused by wind farms, and the lack of any research on this topic in South Africa Sources of information The following information sources were consulted in order to conduct this study: Bird distribution data from the Southern African Bird Atlas Project (SABAP Harrison et al, 1997) obtained from the Avian Demography Unit of the University of Cape Town, was used as a means to ascertain which species occur within the study area. A data set was obtained for the QDGCs (quarter degree grid cells) within which the development will take place, namely 3424BB and 3324DD. A QDGC corresponds to the area shown on a 1: map (15 minutes of latitude x 15 minutes of longitude = 15' x 15') and is approximately 27 km long (north-south) and 23 km wide (east-west). The SABAP data were supplemented with SABAP2 data for the relevant QDGCs. These data are much more recent, as SABAP2 was only launched in May 2007, and should therefore be more accurate. For SABAP, QDGCs were the geographical sampling units. For SABAP2 the sampling unit has been reduced to pentad grid cells (or pentads); these cover 5 minutes of latitude by 5 minutes of longitude (5. 5.). Each pentad is approximately km. This finer scale has been selected for SABAP2 to obtain more detailed information on the occurrence of species and to give a clearer and better understanding of bird distributions. There are nine pentads in a QDGC. pg 7-3

5 Additional information on large terrestrial avifauna and habitat use was obtained from the Coordinated Avifaunal Roadcounts (CAR) project of the Animal Demographic Unit (ADU) of the University of Cape Town. The conservation status of all bird species occurring in the aforementioned quarter degree squares was determined with the use of the Eskom Red Data Book of Birds of South Africa, Lesotho and Swaziland (Barnes 2000). A classification of the vegetation types from an avifaunal perspective in the quarter degree squares was obtained from SABAP1. Detailed satellite imagery from Google Earth (imagery date 2004) was used in order to view the study area on a landscape level and to help identify bird habitat on the ground. An extensive review of the relevant international literature on birds and wind farm impacts was conducted and is included in Chapter 15 of this report. Information on the micro-habitat level was obtained during a site visit in September An attempt was made to investigate the total study area as far as is practically possible, and to visit potential sensitive areas identified from Google Earth imagery. The information obtained when the farm Sunnyside was inspected in 2007 was also taken into account. Telephonic interviews with regard to Blue Crane flight and foraging patterns were conducted with Kevin Shaw (ornithologist, Cape Nature), Kevin McCann (ex-chairman of the South African Crane Working Group), and Bronwyn Botha (ex-field worker of the Overberg Crane Working Group). An interview was conducted with Mr. Rob Malan, one of the landowners, on 29 September 2009, with regards to the birds that are present on his farm. Information was also obtained in 2007 from Mr Mark Holiday of the farm Sunnyside, with regard to the birds that he has observed on his property. Technical details of the planned infrastructure were obtained from Mainstream Renewable Power Jeffrey s Bay (Pty) Ltd. 7.2 DESCRIPTION OF ASPECTS THAT WOULD CAUSE IMPACTS ON BIRDS The following aspects of the project could potentially adversely affect the avifauna: Installation of wind turbines Electrical connections Construction of access roads Temporary construction areas Wind monitoring mast Wind turbines Currently it is planned to install between 40 and 85 turbines depending upon the power output of the individual units (1.5 to 3 MW) Similarly the expected hub height ranges from m and blade diameter from 70 m to 120 m depending upon the output of the unit. The turbines are to be supported on reinforced concrete foundations of approximately 20m x 20m x 2.5 m depth. Electrical transformers will be placed beside each turbine. pg 7-4

6 Gravel surfaced hard standing areas (approximately 40 m x 20 m) will be constructed adjacent to each turbine for use by cranes during construction and retained for maintenance use throughout life span of the project Electrical connection The wind turbines will be connected to each other and to the substation using in most cases buried (approximately 1 m deep) medium voltage cables, except where a technical assessment of the proposed design suggests that overhead lines are appropriate. A new sub-station (approximate size 90 m x 120 m) and transformer will be constructed adjacent to the 132kV line to connect to the Eskom grid. The connection from the substation to the Eskom grid line will be a short section of overhead line Construction of access roads Gravel access roads will be constructed onto the site from the public road (three options proposed). An internal road network to the turbines and other infrastructure (substation and operation and maintenance building) will be constructed. The road network may include turning circles for large trucks, passing points and culverts over gullies and rivers. All roads will be approximately 10 m in width, including cabling and drainage. Upgrading of certain existing roads may take place Temporary construction area A lay down area, besides an access route, with a maximum approximate area of 10,000 m 2 will be constructed. This hard-standing area could be temporary or if the landowner prefers, left for his/her use. The overall site compound for all contractors would be a approximately of 5000 m 2. Approximately five borrow pits (subject to appropriate permits) will be distributed around the site. Existing borrow pits will be used as far as possible. The size of these pits will be dependent on the terrain and need for material. At the end of construction these pits will be backfilled as much as possible using excavated material from the foundations and rehabilitated. 7.3 DESCRIPTION OF THE AFFECTED ENVIRONMENT Vegetation types and bird habitats Vegetation structure is more critical in determining bird habitat than actual plant composition (Harrison et al. 1997). Therefore, the description of the vegetation presented in this study concentrates on factors relevant to birds, and does not give an exhaustive list of plant species which occur in the study area. pg 7-5

7 The description of the vegetation type occurring on site makes use of information presented in the Atlas of southern African birds (Harrison et.al. 1997). The criteria used to amalgamate botanically defined vegetation units, or to keep them separate were (1) the existence of clear differences in vegetation structure, likely to be relevant to birds, and (2) the results of published community studies on bird/vegetation associations. It is important to note that no new vegetation unit boundaries were created, with use being made only of previously published data. The proposed development sites are situated within the Fynbos biome (Mucina and Rutherford 2006). The Fynbos biome is characterized by a high diversity in plant species composition and endemism. This diversity is not paralleled in its avifaunal composition, and Fynbos is regarded as relatively poor in avifaunal diversity compared with other southern African biomes. The endemic Fynbos avifauna consists of the Cape Rockjumper Chaetops frenatus, Victorin s Warbler Cryptillas victorini, Cape Sugarbird Promerops cafer, Orange-breasted Sunbird Anthobaphes violacea, Protea Seedeater Crithagra leucopterus and Cape Siskin Crithagra totta. The Black Harrier Circus maurus, a southern African endemic, also uses the Fynbos biome extensively for breeding. In the study area, these endemics are either absent or very sparsely distributed. There are however populations of Red listed species which are not restricted to the Fynbos biome. Whilst some of the distribution and abundance of birds in the study area can be explained by the description of vegetation types above, it is even more important to examine the micro habitats available to birds. These are generally evident at a much smaller spatial scale than the vegetation types, and are determined by a host of factors such as vegetation type, topography, land use and man made infrastructure. The micro-habitats observed in the study area during the field visit are described below. Examples of each micro habitat can be seen in Appendix 7.A. Irrigated pastures. The study area, especially the sections south of the N2 highway, contains extensive cultivated pastures. The area s most important economic activity is dairy farming, and the pastures have replaced most of the indigenous Fynbos, especially along the coastal flats. The pastures are important for several species, including Red listed species such as Blue Crane, Black-winged Lapwing and Denham s Bustard (see Table 7.1). Non Red listed species include African Sacred Ibis Threskiornis aethiopicus, Spurwinged Goose Plectropterus gambensis and Black-headed Heron Ardea melanocephala. In the summer months, large flocks of White Storks Ciconia ciconia frequent the pastures. Within the borders of the study area, very few irrigated pastures are found. Fynbos. The remaining areas of Fynbos are of importance for Red listed species such as Secretarybird, Denham s Bustard and Black Harrier (see Table 7.1). Other, non-red data species that are likely to be encountered here are Rock Kestrel Falco rupicolus, Jackal Buzzard Buteo rufofuscus, Steppe Buzzard Buteo vulpinus, Helmeted Guineafowl Numida meleagris and Red-necked Spur-fowl Phalaropus lobatus and, in degraded areas, Crowned Lapwing Vanellus coronatus and Spotted Thick-knee Burhinus capensis. There are extensive remaining areas of Fynbos, especially against the hill slopes which have not been cleared for cultivation. Old lands and pastures. There are several areas in the study area where the original Fynbos vegetation was cleared when agriculture was practiced at some stage in the past (mostly wheat farming). These areas are now reverting to a form of grassy Fynbos, which constitutes ideal habitat for Red listed Blue Crane, Denham s Bustard and Secretarybird (see Table 7.1). Some of the old lands have been planted with indigenous grasses which are intermingling with indigenous Fynbos. These areas are also very suitable for the species mentioned above, as well as foraging Black Harrier and White-bellied Korhaan. pg 7-6

8 Raptors such as Peregrine Falcon and Lanner Falcon will also hunt for birds in the cleared areas. Dams. The area contains several dams and water bodies, mostly man made but some also natural and seasonal. These dams and pans, depending on the shape, can be important for some bird species. Dams with shallow sloping sides are suitable for a wider range of species. In the context of this study, shallow dams with sloping sides are important roost sites for Blue Cranes and White Storks. These dams will also be frequented by a variety of waders and ducks, as well as Red listed Black Stork, Yellow-billed Stork, Greater Flamingo and Lesser Flamingo (see Table 7.1). Drainage lines. The study area contains a drainage line, the Swart River, that bisects the study area. The banks of the river itself are heavily infested with alien Black Wattle Acacia mearnsii and other exotic species. The drainage line itself may be of importance to Halfcollared Kingfisher. Some of the larger trees in the drainage lines may be used by Secretarybird for breeding. Wetlands. Some of the dams in the study area have associated wetland areas, which may be of importance to Blue Cranes and Red listed African Marsh Harrier (see Table 7.1) Avifauna in the study area The proposed wind facility is located in the 3422BB and 3322DD QDGCs. Within these cells, a total of 20 Red listed bird species was recorded by SABAP1 during the bird atlas period (excluding marine species), and 15 Red listed species thus far by the SABAP2 project. The Red listed species that have been recorded are listed in Table 7.1. The conservation status, habitat preferences as well as the potential for a species to occur at the proposed wind farm site is also presented in the table. pg 7-7

9 Table 7.1: Red listed species (excluding marine species) recorded in 3422BB and 3322DD by SABAP1 (Harrison et al 1997) and SABAP2 ( Common Name Scientific Name Conservation Status (Barnes 2000) Likelihood of occurrence at study site Black Stork Ciconia nigra NT Medium Secretarybird Sagittarius serpentarius NT Medium Martial Eagle Polemaetus bellicosus VU Low Peregrine Falcon Falco peregrinus NT Medium Lanner Falcon Falco biarmicus NT Medium Blue Crane Anthropoides paradiseus VU High Denham s Bustard Neotis denhami VU High Aghulhas Long-billed Lark Certhilauda brevirostris NT Low Knysna Woodpecker Campethera notata NT Low Half-collared Kingfisher Alcedo semitorquata NT Low African Crowned Eagle Stephanoaetus coronatus NT Low Habitat requirements (Barnes 1998; Barnes 2000; Hockey et al 2005; Young et al 2003; Harrison et al 1997; personal observations) Cliffs for roosting and breeding, and rivers and dams for foraging. Could be encountered at some of the dams, and at pools in the Swart River. Grassland, old lands, open woodland. Most likely to be encountered in fynbos, pastures and fallow lands. Breeds in trees in gorges. Diverse habitats, from open grassland and scrub to woodland. Typically found in flat country. May be encountered in fynbos and old agricultural areas reverting to fynbos. A wide range of habitats, but cliffs (or tall buildings) are a prerequisite for breeding. May hunt over open areas and fynbos. Generally prefers open habitat, but exploits a wide range of habitats. May hunt over open areas and fynbos. Old lands, pastures, wetlands, dams and pans for roosting. The species is fairly common in the study area. Fynbos, old lands and pastures. The species is fairly common in the study area. Fallow and recently ploughed fields, sparse shrub land dominated by renosterveld. Marginally recorded in the study area. Occurs in fynbos forest patches, or on the edges of afromontane forest. Little suitable habitat in the study area. Prefers fast flowing, clear streams. May be present along the Swart River when flowing strongly after good rains. Forest, including gallery forest, dense woodland, and forested gorges in savanna and grassland. Also in eucalyptus and pine plantations. Little suitable habitat in study area. VU = Vulnerable NT = Near threatened pg 7-8

10 7.4 IDENTIFICATION OF KEY IMPACTS PERTAINING TO AVIFAUNA To be effective, wind farms must be sited in open, exposed areas where there are high average wind speeds. This means that they are often proposed in upland, coastal and offshore areas, thus potentially affecting important habitats for breeding, wintering and migrating birds. The effects of a wind farm on birds are highly variable and depend on a wide range of factors including the specification of the development, the topography of the surrounding land, the habitats affected and the number and species of birds present. With so many variables involved, the impacts of each wind farm must be assessed individually. The principal areas of concern with regard to effects on birds are listed below. Each of these potential effects can interact, either increasing the overall impact on birds or, in some cases, reducing a particular impact (for example where habitat loss or displacement causes a reduction in the number of birds using an area which might then reduce the risk of collision). Collision with the wind turbines Collision with the proposed 500 m new stretch of power line Displacement due to disturbance Habitat change and loss 7.5 PERMIT REQUIREMENTS No specific legal requirements are applicable pertaining to avifauna. The applicable environmental legal requirements are fully covered in the Draft Scoping Report dated February From an international perspective, the Convention on Biological Diversity (1992) is applicable. The overall objective of the Convention is the conservation of biological diversity and the sustainable use of its components and the fair and equitable sharing of the benefits. 7.6 IMPACT ASSESSMENTS Collisions Mortalities resulting from collisions with the wind turbines The majority of studies of collisions with wind turbines have reported relatively low levels of mortality. This is, perhaps, largely a reflection of the fact that many of the studied wind farms are located away from large concentrations of birds. It is also important to note that many records are based only on finding corpses, with no correction for corpses that are overlooked or removed by scavengers (Drewitt & Langston 2006). Relatively high collision mortality rates have been recorded at several large, poorly sited wind farms in areas where large concentrations of birds are present (including Important Bird Areas (IBAs), especially migrating birds, large raptors or other large soaring species, eg Altamont Pass in California, USA, Tarifa and Navarra in Spain. In these cases, actual deaths resulting from collision are high, notably of the Golden Eagle Aquila chrysaetos and Eurasian Griffon Gyps fulvus, respectively. With the exception of the White Stork Ciconia ciconia in summer, large flocks of soaring birds are not a characteristic of the present study area. In a recent study in Spain, it was found that the distribution of collisions with wind turbines was clearly associated with the frequencies at which soaring birds flew close to rotating blades (Barrios & Rodriguez 2004). Patterns of risky flights and mortality included a temporal component (deaths concentrated in some seasons), a spatial component (deaths aggregated in space), a taxonomic pg 7-9

11 component (a few species suffered most losses), and a migration component (resident populations were more vulnerable). Collision risk depends on a range of factors related to bird species, numbers and behaviour, weather conditions and topography and the nature of the wind farm itself, including the use of lighting. Clearly, the risk is likely to be greater on or near areas regularly used by large numbers of feeding or roosting birds, or on migratory flyways or local flight paths, especially where these are intercepted by the turbines. Risk also changes with weather conditions, with evidence from some studies showing that more birds collide with structures when visibility is poor due to fog or rain, although this effect may be to some extent offset by lower levels of flight activity in such conditions. Birds that are already on migration, however, cannot avoid poor weather conditions, and will be more vulnerable if forced by low cloud to descend to a lower altitude or land. Fortunately, the phenomenon of mass migrations is not a feature of the study area. Strong headwinds also affect collision rates and migrating birds in particular tend to fly lower when flying into the wind (Drewitt & Langston 2006). Accepting that many wind farms result in only low levels of mortality, even these levels of additional mortality may be significant for long-lived species with low productivity and slow maturation rates (e.g. Blue Crane, Lesser Flamingo, Greater Flamingo, Martial Eagle and Denham s Bustard), especially when rarer species of conservation concern are affected. In such cases there could be significant effects at the population level (locally, regionally or, in the case of rare and restricted species, nationally), particularly in situations where cumulative mortality takes place as a result of multiple installations (Carette et al 2009). Large birds with poor manoeuvrability (such as cranes and bustards) are generally at greater risk of collision with structures. Species that habitually fly at dawn and dusk or at night are perhaps less likely to detect and avoid turbines (e.g. cranes arriving at a roost site after sunset, or flamingos flying at night). Collision risk may also vary for a particular species, depending on age, behaviour and stage of annual cycle (Drewitt & Langston 2006). While the flight characteristics of cranes, flamingos and bustards make them obvious candidates for collisions with power lines, it is significant that these classes of birds (unlike raptors) do not feature prominently in literature as collision victims of wind turbines. It may be that they avoid wind farms entirely, resulting in lower risks of collision (see the discussion of Displacement below). The precise location of a wind farm site can be critical. Particular topographic features may be used for lift by soaring species (Barrios & Rodriguez 2004; De Lucas et al 2008) or can result in large numbers of birds being funnelled through an area of turbines (Drewitt & Langston 2006). For example, absence of thermals on cold, overcast days may force larger, soaring species (e.g. Martial Eagle and Secretarybird) to use slopes for lift, which may increase their exposure to turbines. Birds also lower their flight height in some locations, for example when following the coastline or crossing a ridge, which might place them at greater risk of collision with rotors. In the present case, the entire study area is located on a flat area, and based on the site visit observations and by studying the contour lines, no obvious funnels could be detected. There is the potential for ridge soaring by raptors including Red listed species such as Lanner Falcon, Peregrine Falcon, Martial Eagle and Secretarybird in certain parts of the study area (see Appendix 7.B, Sensitivity Map). Local, low altitude movement by species such as Blue Crane and Denham s Bustard also happens frequently (pers. obs), and may be influenced by the topography. The size and alignment of turbines and rotor speed are likely to influence collision risk, however, physical structure is probably only significant in combination with other factors, especially wind speed, with gentle winds resulting in the highest risk (Barrios & Rodriguez 2004; Stewart et al 2007). De Lucas et al (2008) found that turbine height and higher elevations may increase the risk pg 7-10

12 (taller = more victims), but that abundance was not directly related to collision risk, at least for the Eurasian Griffon Gyps fulvus. Aviation warning lights on turbines may increase the risk of collision by attracting and disorientating birds. The effects of lights in these circumstances are poorly known, though collisions of large numbers of migrants with illuminated structures, especially during overcast nights with drizzle or fog, are well documented. The current advice is to use the minimum number of intermittent flashing white lights of lowest effective intensity (Drewitt & Langston 2006). It is not known if the use of lights on the outer turbines alone, which would perhaps result in more diffuse lighting, would be less likely to disorientate birds than a single bright point source. It must be noted that the risk of nocturnal collisions with lighted turbines has been studied within the context of large numbers of nocturnal migrants in the northern hemisphere, which is not a feature of the current study area. A review of the available literature indicates that, where collisions have been recorded, the rates per turbine are very variable with averages ranging from 0.01 to 23 bird collisions annually (the highest figure is the value, following correction for scavenger removal, for a coastal site in Belgium and relates to gulls, terns and ducks amongst other species) (Drewitt & Langston 2006). Although providing a helpful and standardized indication of collision rates, average rates per turbine must be viewed with some caution as they are often cited without variance and can mask significantly higher rates for individual turbines or groups of turbines (Everaert et al 2001 as cited by Drewitt & Langston 2006). Some of the highest levels of mortality have been for raptors at Altamont Pass in California (Howell & DiDonato 1991, Orloff & Flannery 1992 as sited by Drewitt & Langston 2006) and at Tarifa and Navarre in Spain (Barrios & Rodriguez unpublished data as sited by Drewitt & Langston 2006). These cases are of particular concern because they affect relatively rare and long-lived species such as Griffon Vulture Gyps fulvus and Golden Eagle Aquila chrysaetos which have low reproductive rates and are vulnerable to additive mortality. At Altamont, Golden Eagles congregate to feed on super-abundant prey which supports very high densities of breeding birds. In the Spanish cases, extensive wind farms were built in topographical bottlenecks where large numbers of migrating and local birds fly through a relatively confined area due to the nature of the surrounding landscape, for example through mountain passes, or use rising winds to gain lift over ridges (Barrios & Rodriguez 2004). Although the average numbers of fatalities annually per turbine were generally low at Altamont Pass and Tarifa, ranging from 0.02 to 0.15 collisions/turbine, overall collision rates were high because of the large numbers of turbines involved (over 7000 at Altamont). At Navarre, corrected annual estimates ranging from 3.6 to 64.3 mortalities/turbine were obtained for birds and bats (unpublished data). Thus, a minimum of 75 Golden Eagles are killed annually at Altamont and over 400 Griffon Vultures are estimated (following the application of correction factors) to have collided with turbines at Navarre. Work on Golden Eagles at Altamont Pass indicated that the population was declining in this area, thought to be at least in part due to collision mortality (Hunt et al 1999, Hunt 2001 as sited by Drewitt & Langston 2006). It must be noted that the Jeffrey s Bay study site is not a regular migration funnelling point for large birds. No estimate can however be made of potential collision rates, as a result of the lack of data Collisions with the proposed power line Even though the impact is expected to be low as the line will only have a maximum length of 500 m, collisions could still take place and the possible impact of these need to be addressed. Negative interactions between wildlife and electricity structures take many forms, but two common problems in southern Africa are electrocution of birds (and other animals) and birds colliding with power lines (Ledger & Annegarn 1981; Ledger 1983; Ledger 1984; Hobbs & Ledger 1986a; pg 7-11

13 Hobbs & Ledger 1986b; Ledger et.al. 1992; Verdoorn 1996; Kruger & Van Rooyen 1998; Van Rooyen 1998; Kruger 1999; Van Rooyen 1999; Van Rooyen 2000). Electrocution is not envisaged to be a problem on the proposed 132kV line. Collisions, on the other hand, could be a potential problem. Collisions kill far more birds annually in southern Africa than electrocutions (Van Rooyen 2007). Most heavily affected are bustards, storks, cranes and various species of water birds. These species are mostly heavy-bodied birds with limited maneuverability, which makes it difficult for them to take the necessary evasive action to avoid colliding with power lines (van Rooyen 2004, Anderson 2001). Unfortunately, many of the collision-sensitive species are considered threatened in southern Africa - of the 2369 avian mortalities on distribution lines recorded by the Endangered Wildlife Trust (EWT) between August 1996 and October 2007, 1512 (63.8%) were Red listed species (Van Rooyen 2007). In the Overberg region of the Western Cape, which has a very similar Red listed species composition and habitat use as the Jeffrey s Bay study area, power line collisions have long been recorded as a major source of avian mortality (Van Rooyen 2007). Most numerous amongst power line collision victims are Blue Crane and Denham s Bustard (Shaw 2007). It has been estimated that as many as 10% of the Blue Crane population in the Overberg are killed annually on power lines, and figure for Denham s Bustard might be as high as 30% of the Overberg population (Shaw 2007). These figures are of great concern, as the powerline collisions possibly represent an unsustainable source of unnatural mortality. Unfortunately, the dynamics of the collision problem is poorly understood. In the most recent study on this problem in the Overberg, Shaw (2007) identified cultivated land and the region itself as the significant factors influencing power line collision risk. Lines that cross cultivated land pose a higher risk, as expected, as this is the preferred habitat of Blue Cranes in the Overberg. In the current study area, it can be postulated that the old lands and pastures will be higher risk from a power line collision perspective, as this constitutes primary habitat for Blue Crane and Denham s Bustard. Collision rates are higher for birds in flocks, as they may panic, or lack visibility and room for maneuver because of the close proximity of other birds (APLIC, 1994). Other factors, such as proximity to dams, wind direction and proximity to roads and dwellings did not emerge as significant factors, but she readily admits that her broad-scale analysis may have been too crude to demonstrate their effects. It is for example a well known fact that cranes are particularly vulnerable to power lines skirting water bodies used as roosts, as they often arrive there or leave again in low light conditions (pers. obs.) Displacement Displacement due to disturbance The displacement of birds from areas within and surrounding wind farms as a result of visual intrusion and disturbance effectively can amount to habitat loss. Displacement may occur during both the construction and operational phases of wind farms, and may be caused by the presence of the turbines themselves through visual, noise and vibration impacts, or as a result of vehicle and personnel movements related to site maintenance. The scale and degree of disturbance will vary according to site- and species-specific factors and must be assessed on a site-by-site basis (Drewitt & Langston 2006). Unfortunately, few studies of displacement as a result of disturbance are conclusive, often because of the lack of before-and-after and control-impact (BACI) assessments. Onshore, disturbance distances (in other words the distance from wind farms up to which birds are absent or less abundant than expected) up to 800 m (including zero) have been recorded for wintering waterfowl (Pedersen & Poulsen 1991 as cited by Drewitt & Langston 2006), though 600 m is pg 7-12

14 widely accepted as the maximum reliably recorded distance (Drewitt & Langston 2006). The variability of displacement distances is illustrated by one study which found lower postconstruction densities of feeding European White-fronted Geese Anser albifrons within 600 m of the turbines at a wind farm in Rheiderland, Germany (Kruckenberg & Jaene 1999 as cited by Drewitt & Langston 2006), while another showed displacement of Pink-footed Geese Anser brachyrhynchus up to only m from turbines at a wind farm in Denmark (Larsen & Madsen 2000 as cited by Drewitt & Langston 2006). Studies of breeding birds are also largely inconclusive or suggest lower disturbance distances, though this apparent lack of effect may be due to the high site fidelity and long life-span of the breeding species studied. This might mean that the true impacts of disturbance on breeding birds will only be evident in the longer term, when new recruits replace existing breeding birds. Few studies have considered the possibility of displacement for relatively short-lived passerines (such as larks), although they found increased densities of breeding grassland passerines with increased distance from wind turbines, and higher densities in the reference area than within 80 m of the turbines, indicating that displacement did occur at least in this case. The consequences of displacement for breeding productivity and survival are crucial to whether or not there is likely to be a significant impact on population size. In the absence of any reliable information on the effects of displacement on birds, it is precautionary to assume that significant displacement will lead to a population reduction (Drewitt & Langston 2006). Studies show that the scale of disturbance caused by wind farms varies greatly. This variation is likely to depend on a wide range of factors including seasonal and diurnal patterns of use by birds, location with respect to important habitats, availability of alternative habitats and perhaps also turbine and wind farm specifications. Behavioural responses vary not only between different species, but between individuals of the same species, depending on such factors as stage of life cycle (wintering, moulting, breeding), flock size and degree of habituation. The possibility that wintering birds in particular might habituate to the presence of turbines has been raised (Langston & Pullen 2003), though it is acknowledged that there is little evidence and few studies of long enough duration to show this, and at least one study has found that habituation may not happen (Altamont Pass Avian Monitoring Team 2008). A recent systematic review of the effects of wind turbines on bird abundance has shown that increasing time since operation resulted in greater declines in bird abundance (Stewart et al as sited by Drewitt & Langston 2006). This evidence that impacts are likely to persist or worsen with time suggests that habituation is unlikely, at least in some cases (Drewitt & Langston 2006, Altamont Pass Avian Monitoring Team 2008). In the present study area, it can be reasonably inferred that sensitive species such as Whitebellied Korhaan, Denham s Bustard and Blue Crane will be affected by the noise (and the movement) arising from the construction and operation of the turbines. It is known that the Whitebellied Korhaan requires areas of suitable habitat well away from anthropogenic activities (high human densities). The White-bellied Korhaan is extremely sensitive to human intrusion and will promptly vacate areas when humans are detected, and may often flush away from human intrusion at distances of up to one kilometre measured between the korhaan individuals and the observer (Niemand 2009). Likewise, Morrison (1998) found that the probability of finding Blue Crane nests decreases as the number of roads in an area increases. She further found that Blue Cranes actively avoided tar and gravel roads, houses and areas of agricultural activity when selecting a nest site. It can therefore be postulated that the noise and movement at the wind farm will most likely serve as a deterrent to the species. Indications are that the Great Bustard Otis tarda (a species related to the Denham s Bustard) is displaced by wind farms within one kilometre of the facility (Langgemach 2008). From personal observations it is clear that the Denham s Bustard is very sensitive to anthropogenic activity and is likely to react in the same manner. The effect of birds altering their migration flyways or local flight paths to avoid a wind farm is also a form of displacement. This effect is of concern because of the possibility of increased energy pg 7-13

15 expenditure when birds have to fly further, as a result of avoiding a large array of turbines, and the potential disruption of linkages between distant feeding, roosting, moulting and breeding areas otherwise unaffected by the wind farm. The effect depends on species, type of bird movement, flight height, distance to turbines, the layout and operational status of turbines, time of day and wind force and direction, and can be highly variable, ranging from a slight 'check' in flight direction, height or speed, through to significant diversions which may reduce the numbers of birds using areas beyond the wind farm (Drewitt & Langston 2006). A review of the literature suggests that none of the barrier effects identified so far has a significant impact on populations (Drewitt & Langston 2006). However, there are circumstances where the barrier effect might lead indirectly to population level impacts; for example where a wind farm effectively blocks a regularly used flight line between nesting and foraging areas, or where several wind farms interact cumulatively to create an extensive barrier which could lead to diversions of many tens of kilometres, thereby incurring increased energy costs. It is not possible to make any firm projections in the current study area as to the significance of this potential impact as a result of a lack of data. It has to be assumed that it could be a factor for several species, including sensitive Red listed species such as White-bellied Korhaan, Denham s Bustard, and Blue Crane, which make frequent low altitude flights Habitat change and loss The scale of direct habitat loss resulting from the construction of a wind farm and associated infrastructure depends on the size of the project but, in general, is likely to be small per turbine base. Typically, actual habitat loss amounts to 2 5% of the total development area (Fox et al as sited by Drewitt & Langston 2006), though effects could be more widespread where developments interfere with hydrological patterns or flows on wetland or peatland sites (unpublished data). Some changes could also be beneficial. For example, habitat changes following the development of the Altamont Pass wind farm in California led to increased mammal prey availability for some species of raptor (for example through greater availability of burrows for Pocket Gophers Thomomys bottae around turbine bases), though this may also have increased collision risk (Thelander et al as sited by Drewitt & Langston 2006). In the study area, direct habitat loss is not regarded as a major impact on avifauna, relative to other impact such as disturbance Mitigation measures Wind turbines Mitigation measures fall into two broad categories: best-practice measures which could be adopted by any wind farm development and should be adopted as an industry standard, and additional measures which are aimed at reducing an impact specific to a particular development (Drewitt & Langston 2006). Amongst others, examples of best practice measures are (Drewitt & Langston 2006): Ensuring that key areas of conservation importance and sensitivity are avoided; Implementing appropriate working practices to protect sensitive habitats; Providing adequate briefing for site personnel and, in particularly sensitive locations, employing an on-site ecologist during construction; Implementing an agreed post-development monitoring programme through planning or licence conditions; pg 7-14

16 Siting turbines close together to minimize the development footprint (subject to technical constraints such as the need for greater separation between larger turbines); Where possible, installing transmission cables underground (subject to habitat sensitivities and in accordance with existing best practice guidelines for underground cable installation); Marking overhead cables using deflectors and avoiding use over areas of high bird concentrations, especially for species vulnerable to collision; Turning to more site-specific mitigation, it is necessary to prepare a site management plan designed to reduce or prevent harmful habitat changes following construction, and to provide habitat enhancement as appropriate. The following site specific mitigation measures are proposed: Ensuring that key areas of conservation importance and sensitivity are avoided: See Appendix 7.B for a map of the area, indicating the most sensitive areas from a Red listed species perspective. It is appreciated that avoiding construction of turbines in all these sensitive areas will not be economically viable. Due to the expected lack of large concentrations of Red listed species (e.g. roost sites or funnelling points) close to any of the proposed turbine sites, no potential no-go areas have been identified from a bird perspective. Implementing appropriate working practices to protect sensitive habitats: Habitat destruction should be limited to what is absolutely necessary for the construction of the infrastructure, including the construction of new roads. Providing adequate briefing for site personnel and, in particularly sensitive locations: personnel should be adequately briefed on the need to restrict habitat destruction, and must be restricted to the actual building sites. Implementing an agreed post-development monitoring programme through planning or licence conditions: It is imperative that a BACI (Before After Control Impact) study is conducted to test the assumptions in his report. There is an urgent need for baseline information on wind farm impacts in South Africa (see 6.5 below). Where possible, installing transmission cables underground in accordance with existing best practice guidelines for underground cable installation Power line The alignment for the power line has not been finalized. It is therefore suggested that anti-collution devices are installed. The Double Loop Bird Flight Diverter is proposed and one on each line would be sufficient for a 500 m stretch of power line Monitoring requirements At this early stage of wind power development in South Africa, the magnitude of the impact of the proposed project will be local only i.e. the footprint itself. If development of additional wind facilities takes place in an uncontrolled manner, the impact will increase significantly, which could lead to fragmentation of (in particular) Blue Crane and Denham s Bustard habitat and increased levels of disturbance. The cumulative impact of additional wind facilities will therefore have to be closely monitored. However, the fact of the matter is that wind energy is likely to become an increasingly important source of energy in South Africa. In the interest of science and conservation, it would therefore be very valuable if a monitoring programme could be implemented at the Jeffrey s Bay wind facility in order to start building a knowledge base of actual impacts (or the lack of) at local wind facilities. The proposed facility is within an hour s drive from the Nelson Mandela Metropole University, which would be the ideal institution to implement such a programme. It is therefore pg 7-15

17 recommended that the CSIR approaches the University before the construction of the facility with a view to the design and implementation of the programme, should the proponent agree to provide the necessary funding. pg 7-16

18 Table 7.2: Impact assessment table Impact description Bird collisions, particularly Red listed species, with the wind turbines Power line: Bird collisions with the 500 m power line Displacement due to disturbance Habitat change and loss due to the footprint of the infrastructure Status Extent Duration Intensity Probability Significance (without mitigation) Negative Negative Negative Negative Local (within 5km of the development) Local (within 500 m of the development) Local (within 5km of the development) Local (within 5km of the development) Long term >15 years Long term >15 years Long term >15 years Long term >15 years Medium Medium Medium Probable for raptors, but not for cranes and bustards. Probable (raptors, cranes and bustards) Probable for cranes and bustards. Improbable for raptors. Low Medium Low Mitigation No mitigation is required due to the likelihood of key species (cranes and bustards) being displaced from the wind farm area. Occasional interaction with raptors (particularly Secretarybird and Lanner Falcon) cannot be excluded, but mitigation not necessary due to low expected frequency of interaction. Fit anti-collision markers to the conductors in high risk areas No practical mitigation is possible for this impact, as it would effectively nullify the wind development, as the majority of the study area is suitable for cranes and bustards. Significance (with mitigation) N/A Low N/A Confidence level Medium/low due to lack of South African precedents. High Medium/low due to lack of South African precedents. Low Probable Low No mitigation is possible for this impact N/A Medium pg 7-17

19 Chapter 7: Impact on Birds APPENDICES TO CHAPTER 7 Appendix 7A: Micro Habitats 7-19 Appendix 7B: Sensitivity Map 7-24 pg 7-18

20 Chapter 7: Impact on Birds Appendix 7A: Micro Habitats Figure 1: An example of indigenous Fynbos against a slope. These slopes may be used by raptors for slope soaring on cold, overcast days Figure 2: An example of an old agricultural field which is reverting back to Fynbos 7-20 Figure 3: Figure 4: Indigenous Fynbos in the study area. This habitat is used by Denham s Bustard, Black Harrier and Secretarybird Old wheat fields on the horizon. This habitat is favoured by Blue Crane, Denham s Bustard and Secretarybird Figure 5: Another example of cleared Fynbos in the form of old agricultural fields Figure 6: A dam in the study area. Dams are important roosting areas for Blue Cranes and Greater Flamingo Figure 7: The Swart River in the study area. Drainage lines are heavily infested with alien vegetation Figure 8: Pastures in the study area. This habitat is important for Blue Crane and Denham s Bustard pg 7-19

21 Chapter 7: Impact on Birds Figure 1: An example of indigenous Fynbos against a slope. These slopes may be used by raptors for slope soaring on cold, overcast days. Figure 2: An example of an old agricultural field which is reverting back to Fynbos pg 7-20

22 Chapter 7: Impact on Birds Figure 3: Indigenous Fynbos in the study area. This habitat is used by Denham s Bustard, Black Harrier and Secretarybird. Figure 4: Old wheat fields on the horizon. This habitat is favoured by Blue Crane, Denham s Bustard and Secretarybird. pg 7-21

23 Chapter 7: Impact on Birds Figure 5: Another example of cleared Fynbos in the form of old agricultural fields. Figure 6: A dam in the study area. Dams are important roosting areas for Blue Cranes and Greater Flamingo. pg 7-22

24 Chapter 7: Impact on Birds Figure 7: The Swart River in the study area. Drainage lines are heavily infested with alien vegetation. Figure 8: Pastures in the study area. This habitat is important for Blue Crane and Denham s Bustard. pg 7-23

25 Chapter 7: Impact on Birds Appendix 7B: Sensitivity Map pg 7-24

26 Chapter 7: Impact on Birds Sensitive habitat for Blue Crane, Denham s Bustard, Secretarybird, and various raptors Sensitive habitat for slope soaring species e.g. Martial Eagle, Secretarybird pg 7-25