Contents: Avifauna S7 - 1

7.1 Introduction S7 - 1

7.2 Objectives S7 - 4

7.3 Legislation, Standards & Guidelines S7 - 5

7.4 Assessment Approach S7 - 5

7.5 Baseline Conditions and Sensitive Receivers – Desktop Study S7 - 12

7.6 Baseline Conditions and Sensitive Receivers – Field Surveys S7 - 24

7.7 Construction Phase Impact Assessment S7 - 66

7.8 Operation Phase Impact Assessment S7 - 66

7.9 Mitigation & Best Practices Measures S7 - 77

7.10 Environmental Monitoring & Audit S7 - 78

7.11 Conclusions & Recommendations S7 - 78

7.12 References S7 - 79

7 Avifauna

7.1 Introduction

7.1.1.1 This section presents the approach to and the findings of the ecological impact assessment of avifauna. The aim of the ecological impact assessment is to examine the avifauna and other components of the ecological habitats within the assessment area in order to protect, maintain or rehabilitate the natural environment. Special attention shall be paid to avoid impacts on wildlife groups or habitats / species with conservation interests including but not limited to migratory birds, breeding visitors and uncommon resident species.

7.1.1.2 There is extensive literature on the potential and actual effects that wind farms have on birds (e.g. Percival, 2003; Drewitt & Langston, 2006; Langston & Pullan, 2006). The construction and operation of commercial scale wind farms onshore or offshore has been found to produce a variety of effects including:

· Loss of habitats or particular foraging areas;

· Presenting a barrier to bird movement;

· Displacing birds from the area;

· Adversely affecting birds’ feeding grounds or food sources;

· Presenting a collision risk to birds.

7.1.1.3 The assessment shall identify and quantify the potential ecological impacts to the natural environment and the associated wildlife groups and habitats / species arising from the proposed Project including its construction and operation phases as well as the subsequent management and maintenance of the proposed development. The assessment has been conducted for installation of 67 nos. of 3MW turbines (Scenario A) and also for 40 nos. of 5MW turbines (Scenario B).

7.1.1.4 For the purpose of the avifauna impact assessment, the Study Area includes the wind farm area and its surroundings to a varying extent, depending on the specific elements being considered. These areas are defined as follows:

· Desktop Study Area: the circled area as displayed in Figure 7.1 to cover the sea area within approximately 20km radius from the centre of the proposed wind farm.

· Field Survey Area: covering the wind farm area and an adjacent area of approximately 130km² for both Scenarios, as displayed in Figure 7.2 and Figure 7.3 respectively.

· Assessment Area(s): the wind farm area plus additional 0.5km, 1km or 2km buffers for both scenarios, as displayed in Figure 7.4 and Figure 7.5 respectively.

7.1 Objectives

7.1.1.1 The aim of the avifauna impact assessment is to consider all potential impacts upon resident and migratory bird species and their habitats from the construction and operation of the proposed Project. In this regard the duration and focus of the baseline surveys was expanded beyond the scope of the EIA Study Brief.

7.1.1.2 Specific objectives of the assessment include:

· Collect information from desktop study and field surveys to establish an ecological baseline for the assessment area. The field surveys include both resident and migratory birds and covered a 20-month period;

· Identify and predict potential ecological impacts during construction and operation of the proposed development;

· Evaluate the significance and acceptability of the identified impacts;

· Recommend effective and practicable alternatives and mitigation measures;

· Recommend the need for and the scope of an appropriate monitoring and audit programme.

7.2 Legislation, Standards & Guidelines

7.2.1.1 Reference shall be made to local legislation governing flora, fauna and habitat conservation. Directly relevant legislation includes:

7.2.1.2 Wild Animals Protection Ordinance (Cap. 170) provides for the protection of species listed in ' Schedule 2 ' of the Ordinance by prohibiting the disturbance, taking or removal of such animals, their nests and eggs. This Ordinance excludes fish and marine invertebrates, but does allow for the protection of all marine mammals found in Hong Kong waters.

7.2.1.3 Protection of Endangered Species of Animals and Plants Ordinance (Cap. 586) gives effect in Hong Kong to the Convention on International Trade in Endangered Species of Wild Fauna and Flora signed in Washington D.C. on 3 March 1973; to regulate the import, introduction from the sea, export, re-export, and possession or control of certain endangered species of animals and plants and parts and derivatives of those species; and to provide for incidental and connected matters.

7.2.1.4 Regionally / internationally protected species: such as those species listed in the following:

· International Union for Conservation of Nature and Natural Resources (IUCN) Red Data Book;

· Convention on International Trade in Endangered Species (CITES);

· List of National Key Protected Species in the Mainland PRC; and / or

· Species considered sensitive and / or of local / regional / international conservation concerns by published literature.

7.2.1.5 EIA -TM (Annexes 8 and 16); and EIAO Guidance Notes No. 6, No. 7 and No. 11.

7.3 Assessment Approach

7.3.1 Desk-top Study Information Sources

7.3.1.1 A desktop study has been conducted to review records of migratory and resident avifauna that currently or may potentially utilise the Study Area. The information and data sources under review include:

· Pilot Project to Increase Awareness of the Ecological Importance of the Breeding Colonies of Terns in Hong Kong (ECF Project 23/2002) (Unpublished report; HKBWS, 2003);

· Seabird Migration Survey in Southern and South-eastern Hong Kong Waters, spring (HKBWS, 2006) (ECF Project 2005-10);

· The Population and Breeding Ecology of White-bellied Sea-eagles in Hong Kong (Tsim et al, 2003);

· 2002 – 2007 Monitoring Data of White-bellied Sea Eagles in Hong Kong (Agriculture, Fisheries and Conservation Department (AFCD), unpublished data);

· The Avifauna of Hong Kong (Carey et. al., 2001);

· Tern Surveys conducted by HKBWS (unpublished data);

· The Birds of Hong Kong and South China (Viney et. al., 2005).

7.3.2 Approach for Conducting Field Surveys

7.3.2.1 Additional / novel field surveys are necessary to supplement or to fill the information gap of the baseline conditions generated from desktop study. Several types of field surveys have been widely documented (e.g. Department for Environment, Food and Rural Affairs, 2005; SNH, 2005) and adopted (e.g. NERI, 2000; Seascape Energy Ltd., 2002; Camphuysen, et al, 2004; RPS, 2006), including:

· Boat-based Survey

· Aerial Survey

· Radar Survey

7.3.2.2 Table 7.1 presents a summary of different types of bird surveys with regard to their documented advantages and limitations.

Table 7.1 Summary of Boat-based Survey, Aerial Surveys and Radar Surveys with Regard to Advantages and Limitations

Types of Surveys	Advantages	Limitations
Boat-based Surveys	· Most sensitive methods to detect obtrusive and low-flying birds; · Good in identifying birds to species level · Allow collection of behavioural information on birds such as (feeding, movements between roosts, flight heights), and more detailed information on bird characters, e.g. age and sex.	· Poor in estimating total numbers for large population of birds; · Poor in terms of obtaining a snapshot of distribution at any one time;
Aerial Surveys	· Good in terms of obtaining a snapshot of distribution at any one time; · Allow surveys of large area at any one time; · Allow good estimate of relative abundance and densities for large population of birds across a seascape;	· Poor in terms of identifying obtrusive or low-flying birds; · Poor in terms of identifying birds to species level; · Unable to provide detailed information such as behaviour, flight height or direction.
Radar Surveys	· Allow surveys during night time; · Allow quantification of marked passage movements by significantly large flocks of migrating or moving birds	· Poor to provide information for bird identification; · Sensitive to human disturbance; · Only allow collection of information at fixed points.

7.3.2.3 Based on the results of desk-top study as conducted under Sub-section 7.5, the most appropriate type of survey methodology has been selected to conduct the field survey as described under Sub-section 7.6.

7.3.3 Collision Risk Impact Assessment & Evaluation

Collision Risk Calculation

7.3.3.1 Several collision risk models for wind farm birds have been developed in recent years. Among these models the Scottish Natural Heritage (SNH) bird collision risk model (Band et al, 2007) is regarded as generally robust and has been most frequently used for several years in Scotland and more recently in the rest of the UK (Madders and Whitfield, 2006). This model has been adopted for this EIA Study as presented under Sub-section 7.8.

7.3.3.2 Collision calculation is based on the worst-case wind farm configuration, being that with the largest blade pitch angle (relative to the rotor plane of the turbine), the smallest rotation period of turbines (the fastest speed of the blade), the maximum bird sizes, and the slowest flight speed of the bird obtained from available documentation or literature. Table 7.2 displays the worst-case configuration for the two proposed turbine options.

Table 7.2 Configuration of the Proposed Turbine Options

Variables	Scenario A	Scenario B
Risk Area of the wind farm (WF + 1km)	34.2 km^2	36.1 km^2
Rotor diameter	90 m	120 m
Number of turbines	67	40
Rotation period	3.33 seconds	4.96 seconds

7.3.3.3 Collision risk calculations have been prepared for two behaviour scenarios: one extreme worst-case scenario that assumes birds take no action to avoid collision, and one recognizing that most birds do take avoiding action (Band et al, 2007). For the latter behavioural scenario a “conservative” 95% ‘avoidance factor’ has been applied as suggested by the SNH guidelines (http://www.snh.org.uk).

7.3.3.4 Stage 1 of the model predicts the number of bird flying through rotors based on field observations. In Stage 2, the model predicts the probability (collision probability) of a bird to be hit by a wind farm turbine when it makes a transit through a rotor. Unlike Stage 1, the collision probability is independent of the abundance of birds (i.e. independent of field data) but depends only on the size of a bird (wingspan / bird length) and its flight speed.

Number of Birds Flying Through Rotors (Stage 1)

7.3.3.5 Under Stage 1 of the model, the amount of flight activity within the proposed wind farm site was quantified and expressed by the number of “bird transits” per season (as defined in sub-sections 7.6.1.9 to 7.6.1.11 the term “season” or “bird season” refer to particular periods to indicate occurrence of key bird population of concern within the Study Area, including migratory birds and breeding birds) within the rotor swept volume (Vr) as follow:

1. Identify a ‘flight risk volume’ (V_w):

This value was taken as the risk area of the wind farm multiplied by the rotor diameter. The risk area was taken as 34.2 km² for Scenario A, which represents the wind farm area (approximately 15.7 km²) plus 1 km buffer. For Scenario B, the risk area taken as 36.1 km², which represents the wind farm area (approximately 16.56 km²) plus 1 km buffer.

2. Calculate the combined volume swept out by the wind farm rotors (V_r):

V_r = N x πR² x (d + l) where N is the number of wind turbines, d is the depth of the rotor back to front, and l is the length of the bird.

3. Estimate the proportion (P_w) of the time that the species spent flying within the flight risk volume (V_w) in each bird season (this will be defined in the “Field Survey Methodology” under sub-section 7.6.1 below):

To allow the use of the model, the point count data obtained from the field survey were converted into time budget data for bird activity in the wind farm area under observation. The time budget (in “bird seconds”) for flight activity was predicted from the survey data collected at the five fixed-point count locations (P3 to P7) at the proposed wind farm. At each location, a circular “visible envelope” of 1km radius was defined by the sighting limit of the observer (assumed as up to 1km to sufficiently cover small birds).

Except for raptors and birds in near-shore area, the majority of birds sighted in the offshore environment (the proposed wind farm site) were observed in straight flight. For a bird in straight flight within the visible envelope, the longest time for it to become lost from the observer would therefore be the time for it to travel 1km. To adopt the most conservative approach in data conversion for collision risk calculation, each single bird count from the five sampling locations was therefore expressed as the time (in terms of “bird seconds”) required for the bird species to travel 1km.

The flight activity (P_w : proportion of time that a species spent flying within the “flight risk volume (V_w)”) was then taken as the bird seconds spent by the species within the risk area (taken as the wind farm area plus 1km buffer), divided by the duration of the survey (assuming a total of 2.5 hours at five locations per each survey trip). The value was then adjusted by multiplying by the overall proportion of the species that were observed flying at risk heights (>30m) to give the proportion P_w.

4. Estimate the bird occupancy (n_w) within the flight risk volume in each season/ survey period:

n_w = P_w x daily usage x site usage

Daily usage by the bird was assumed as 7 hours per day (i.e. the average survey duration of the whole Study Area per trip, of which birds were assumed most active in the Study Area); and

Site usage by the bird species was estimated from the duration of species that persisted in each bird season plus a 3-day buffer period from the starting and ending dates of the survey period. This was taken as the number of days between the first and the last calendar dates for which the species persisted in the study area plus 6 days (as a buffer period).

5. Calculate the bird occupancy of the volume swept by the rotors (n_r):

n_r = n_w x (V_r / V_w) (in bird seconds)

6. Calculate the time (t) taken for a bird to make a transit through the rotor and completely clear the rotors:

t = (d + l) / v where v m/s is the speed of the bird through the rotor

7. Calculate the number of bird transits through the rotors:

Number of birds transits through rotors = n_r / t

Collision Likelihood (Stage 2)

7.3.3.6 The probability of a bird flying through a rotor being hit depends on the size of the bird (both length and wingspan), the breadth and pitch of the turbine blades, the rotation speed of the turbine, and the speed of the bird. Appendix 7B displays a sample of the spreadsheet containing calculations of the collision probabilities.

Impact Evaluation

7.3.3.7 A recognized assessment methodology developed by Scottish Cultural Heritage (SNH) and British Wind Energy Association (BWEA) (Percival et al, 1999; 2001) has been used for this Study. The assessment methodology considers bird issues not to be of significance if either of two positions are satisfied:

· Where no important bird populations are identified in the vicinity of the proposed development, or

· Where important bird populations have been identified but where there is substantive evidence that a significant impact will not occur.

7.3.3.8 Given the recent development of offshore wind farms and a limited research base on how bird populations are affected by these developments the second position is difficult to achieve at this time. Therefore, a precautionary approach should be considered to avoid important protected areas or populations of birds for any new offshore developments. The assessment methodology developed by SNH and BWEA provides a framework (in a form of cross-tabulation matrix as presented in Table 7.5) to indicate significance of impact of offshore wind farm development on birds by giving priority to species / populations of high sensitivity in rating impact significance.

7.3.3.9 The assessment approach consists of three stages:

1. Determination of the sensitivity of the feature potentially affected (Table 7.3)

2. Determination of magnitude of effects on birds (Table 7.4)

3. Assessing the significance of the potential impacts by using cross-tabulation of “Sensitivity” and “Magnitude” (Table 7.5)

Table 7.3 Determination of Ornithological Significance

Sensitivity	Determination Factor
Very High	Cited interest of Special Protection Areas (SPAs) (strictly protected sites classified under the Bird Directive in UK. Cited means mentioned in the citation text for the site as a species for which the site is designated or notified. In Hong Kong, since there is no designated areas particularly for protection of birds, areas designated as the Sites of Special Scientific Interests (SSSIs) in regard of their ornithological importance are considered of “Very High” sensitivity to potential wind farm impacts in this study.
High	Other species that contribute to the integrity of a designated area for conservation. Local population of more than 1% of the national population of a species. Ecologically sensitive species, e.g. Accipitridae raptors and Sternidae terns (Langston & Pullan, 2006)
Medium	Regionally important population of a species, either because of population size or distributional context.
Low	Any other species of conservation interest not covered above.

Table 7.4 Determination of Magnitude of Effects on Birds

Magnitude	Definition
Very High	Total loss or very major alternation to key elements/ features of baseline conditions such that post development character/ composition/ attributes will be fundamentally changed and may be lost from the site altogether. Guide: >80% of population / habitat loss
High	Major alternation to key elements/ features of the baseline (pre-development) conditions such that post development character/ composition/ attributes will be fundamentally changed. Guide: 20 – 80% of population/ habitat loss
Medium	Loss or alternation to one or more key elements/ features of the baseline conditions such that post development character/ composition/ attributes of baseline will be partially changed. Guide: 5 – 20% of population/ habitat loss
Low	Minor shift away from baseline conditions. Change arising from the loss/ alternation will be discernible but underlying character/ composition/ attributes of baseline condition will be similar to pre-development circumstances/ patterns. Guide: 1 – 5% of population/ habitat loss
Negligible	Very slight change from baseline condition. Change barely distinguishable, approximating to the “no change” situation. Guide: <1% of population/ habitat loss

Table 7.5 Matrix of Magnitude and Sensitivity for Determination of Impact Significance

Magnitude	Sensitivity
		Very High	High	Medium	Low
	Very High	Very High	Very High	High	Medium
	High	Very High	Very High	Medium	Low
	Medium	Very High	High	Low	Very Low
	Low	Medium	Low	Low	Very Low
	Negligible	Low	Very Low	Very Low	Very Low

7.4 Baseline Conditions and Sensitive Receivers – Desktop Study

7.4.1 Migratory and Visitor Seabird Populations

7.4.1.1 The most updated published checklist of Hong Kong avifauna, “The Avifauna of Hong Kong” (Carey et al, 2001) documents a total of 41 seabird species recorded in Hong Kong, of which 40 species belong to seasonal passage migrants or visitors. The most latest seabird survey (HKBWS, 2006) conducted by the Hong Kong Bird Watching Society (HKBWS) identifies two additional migratory seabird species, Parasitic Jaeger Stercorarius parasiticus and Short-tailed Shearwater Puffinus tenuirostris in the southern and south-eastern Hong Kong waters.

7.4.1.2 There are also three seabird species recently recorded in Hong Kong Waters: Vega Gull Larus vegae in Mai Po; Japanese Cormorant Phalacrocorax capillatus at Po Toi Island; White-tailed Tropicbird Phaethon lepturus at Southern waters of Waglan Island and also at SE waters during a seabird boat trip conducted by HKBWS in May 2008.

7.4.1.3 Table 7.6 summarises all these 45 migratory / visitor seabird species that have been recorded in Hong Kong waters so far.

Table 7.6 Migratory and Visitor Seabird Species in Hong Kong

Seabirds	Principal Status*	Recorded in E / SE Waters?
Family Alcidae (Auks)
Ancient Murrelet Synthliboramplus antiquus	W	Yes
Family Fregatidae (Frigatebirds)
Christmas Island Frigatebird Fregata andrewsi	-	No
Lesser Frigatebird Fregata ariel	OV	Yes
Great Frigatebird Fregata minor	-	Yes
Family Laridae (Gulls)
Brown-headed Gull Larus brunnicephalus	W, M	No
Mew Gull Larus canus	M	No
Yellow-legged Gull Larus cachinnans	W, M	Yes
Black-tailed Gull Larus crassirostris	W, M	Yes
Slender-billed Gull Larus genei	-	No
Glaucous-winged Gull Larus glaucescens	M	No
Heuglin’s Gull Larus heuglini	W, M	Yes
Glaucous Gull Larus hyperboreus	-	No
Pallas’s Gull Larus ichthyaetus	W, M	Yes
Little Gull Larus minutus	-	No
Relict Gull Larus relictus	-	No
Black-headed Gull Larus ridibundus	W, M	No
Saunders’s Gull Larus saundersi	W, M	No
Slaty-backed Gull Larus schistisagus	W, M	Yes
Vega Gull Larus vegae	-	No
Black-legged Kittiwake Rissa tridactyla	-	No
Family Pelecanidae (Pelicans)
Dalmatian Pelican Pelecanus crispus	W	No
Family Phaethontidae (Tropicbirds)
White-tailed Tropicbird Phaethon lepturus	-	Yes
Family Phalacrocoracidae (Cormorants)
Japanese Cormorant Phalacrocorax capillatus	-	Yes
Great Cormorant Phalacrocorax carbo	W	No
Family Procellariidae (Shearwaters)
Streaked Shearwater Calonectris leucomelas	-	Yes
Short-tailed Shearwater Puffinus tenuirostris	-	Yes
Family Scolopacidae (Sandpipers)
Red Phalarope Phalaropus fulicarius	-	No
Red-necked Phalarope Phalaropus lobatus	M	Yes
Family Sternidae (Terns)
Whiskered Tern Chlidonias hybridus	M	Yes
White-winged Tern Chlidonias leucopterus	M	Yes
Aleutian Tern Sterna aleutica	AM	Yes
Bridled Tern Sterna anaethetus	Su, M	Yes
Gull-billed Tern Sterna nilotica	M	Yes
Caspian Tern Sterna caspia	M	Yes
Common Tern Sterna hirundo	M	Yes
Roseate Tern Sterna dougallii	Su	Yes
Black-naped Tern Sterna sumatrana	Su	Yes
Sooty Tern Sterna fuscata	-	Yes
Little Tern Sterna albifrons	M	Yes
Greater Crested Tern Strena bergii	-	Yes
Family Stercorariidae (Jaegers and Jaegers)
Long-tailed Jaeger Stercorarius longicaudus	SpM	Yes
Parasitic Jaeger Stercorarius parasiticus	-	Yes
Pomarine Jaeger Stercorarius pomarinus	-	Yes
Family Sulidae (Boobies)
Brown Booby Sula leucogaster	-	No
Red-footed Booby Sula sula	-	Yes

*Notes: M: passage migrant; AM; passage migrant in autumn; SpM: passage migrant in spring; Su: summer visitor; W: winter visitor; OV: occasional visitor, “-”: no information available.

7.4.1.4 Of these 45 migratory and/or visitor seabird species, 29 species have been recorded from the eastern or the south-eastern waters of Hong Kong.

7.4.1.5 The seabird survey conducted by HKBWS in spring 2006 covered the main migration path of seabirds in Hong Kong during spring migration period (March –May), as displayed in Figure 7.6. The survey recorded a total of 8,750 seabird individuals from 23 species on 22 survey days, as summarised in Table 7.7. The daily counts of seabirds ranged from 61 to 969 (HKBWS, 2006).

Table 7.7 Summary of Numbers of Seabirds Recorded in HKBWS Seabird Survey in Spring 2006 (HKWBS, 2006)

Seabirds	Number	% of Total
Family Alcidae (Auks)
Ancient Murrelet Synthliboramplus antiquus	3	0.03
Sub-total	3	0.03
Family Laridae (Gulls)
Yellow-legged Gull Larus cachinnans	2	0.02
Black-tailed Gull Larus crassirostris	2	0.02
Heuglin’s Gull Larus heuglini	158	1.81
Slaty-backed Gull Larus schistisagus	1	0.01
Unidentified Gull Larus sp.	8	0.09
Sub-total	171	1.95
Family Procellariidae (Shearwaters)
Streaked Shearwater Calonectris leucomelas	52	0.59
Short-tailed Shearwater Puffinus tenuirostris	15	0.17
Unidentified Shearwater Puffinus sp.	3	0.03
Sub-total	70	0.80
Family Scolopacidae (Sandpipers)
Red-necked Phalarope Phalaropus lobatus	6618	75.63
Sub-total	6618	75.63
Family Sternidae (Terns)
Whiskered Tern Chlidonias hybridus	6	0.07
White-winged Tern Chlidonias leucopterus	754	8.61
Aleutian Tern Sterna aleutica	200	2.28
Bridled Tern Sterna anaethetus	55	0.63
Gull-billed Tern Sterna nilotica	5	0.06
Caspian Tern Sterna caspia	4	0.05
Common Tern Sterna hirundo	212	2.42
Roseate Tern Sterna dougallii	2	0.02
Black-naped Tern Sterna sumatrana	258	2.95
Sooty Tern Sterna fuscata	1	0.01
Little Tern Sterna albifrons	1	0.01
Greater Crested Tern Strena bergii	10	0.11
Unidentified Tern Chlidonias sp. / Sterna sp.	219	2.50
Sub-total	1727	19.73
Family Stercorariidae (Jaegers and Jaegers)
Long-tailed Jaeger Stercorarius longicaudus	113	1.29
Parasitic Jaeger Stercorarius parasiticus	13	0.15
Pomarine Jaeger Stercorarius pomarinus	17	0.19
Unidentified Jaeger Sterocrarius sp.	18	0.21
Sub-total	161	1.84

7.4.2 Breeding Seabird Populations

7.4.2.1 Of the documented 45 migratory and / or visiting seabird species in Hong Kong referred in Table 7.6, only the Black-naped Tern Sterna sumatrana, the Bridled Tern Sterna anaethetus and the Roseate Tern Sterna dougallii are reported as breeding visitors. These terns are pelagic, and only move to land during breeding summers. Black-naped Tern was first reported breeding in 1983, with both Roseate Tern and Bridled Tern reported breeding in 1985.

7.4.2.2 Since these earliest records summer breeding colonies of all the three tern species have been recorded at offshore islands Waglan Island, East Ninepin and Kong Tau Pai (Carey et al, 2001) as displayed by Figure 7.7, although terns have not been recorded at East Ninepin since 1997.

7.4.2.3 Table 7.8 summarises the data collected by HKBWS between 1993 – 1997 on the minimum number of adults and fledged juveniles of the three breeding tern species at the East Ninepin.

Table 7.8 Minimum Number of Adults and Fledged Juveniles of Black-naped Tern Sterna sumatrana, Bridled Tern Sterna anaethetus, and Roseate Tern Sterna dougallii recorded between 1993 – 1997 at the East Ninepin by HKBWS (Carey et al, 2001)

Year	Black-naped Tern		Bridled Tern		Roseate Tern
	Adult	Juvenile	Adult	Juvenile	Adult	Juvenile
1993	80	10	8	0	40	4
1994	85	1	1	0	46	0
1995	35	2	0	0	3	0
1996	10	0	0	0	0	0
1997	0	0	0	0	0	0

7.4.2.4 In 2003, a government-funded project was conducted by HKBWS, namely “Pilot Project to Increase Awareness of the Ecological Importance of the Breeding Colonies of Terns in Hong Kong (ECF Project 23/2002)”. It was estimated in the study that Bridled Tern was the most abundant breeding tern species, with over 500 birds recorded during summer 2003. The breeding population for Black-naped Tern was estimated to exceed 200 birds, while Roseate Tern was estimated to range from 32 to 60 birds in summer 2003. The total population of breeding terns in Hong Kong during summer 2003 was estimated to be between approximately 740 and 820 birds (HKBWS, 2003).

7.4.2.5 The study also first identified breeding tern colonies on Waglan Island and Kong Tau Pai. The breeding tern colony on Waglan Island consisted of more than 200 terns including all three breeding tern species, and was the second largest breeding tern colony in Hong Kong waters (second to the population at Shek Ngau Chau in NE waters). For Kong Tau Pai, more than 40 breeding Black-naped Terns were found (HKBWS, 2003).

7.4.2.6 Although there has been no formally published data on breeding terns in Hong Kong since 2003, summer breeding tern surveys conducted by HKBWS are still on-going. Available results of the tern surveys from the online discussion forum of HKBWS’s website (http://www.hkbws.org.hk, retrieved in 2007) are extracted and summarised in Table 7.9 below.

Table 7.9 Available Results of On-going Breeding Tern Surveys in SE Waters of Hong Kong Conducted by HKBWS (Extracted from HKBWS's website)

Date	Black-naped Tern		Bridled Tern		Roseate Tern
	Adult	Juvenile	Adult	Juvenile	Adult	Juvenile
22 Jul 05	130	uncertain	13	uncertain	8	uncertain
11 Jul 06	174	uncertain	67	uncertain	10	uncertain
21 Jul 06	174	10	39	1	53	uncertain

7.4.3 Resident Populations

7.4.3.1 White-bellied Sea Eagle (WBSE), Haliaeetus leucogaster, belongs to the Family Accipitridae, is one of the ten fish eagle species in the world (ECA, 2007). WBSEs are monotypic species that inhabit coastal areas and offshore islands. Juveniles of WBSEs are dispersive, while breeding pairs tend to be more sedentary within their own territories. Nests of WBSEs are usually found on tall trees or on remote coastal cliffs (Tsim et al., 2003).

7.4.3.2 WBSE has a world distribution from western India through SE Asia to Australia (Ferguson-Lees & Christie, 2005). In Mainland China, WBSE is resident in Guangdong, southern Fujian, and occasionally occurs in Jiangsu and Hainan (Cheng, 1987).

7.4.3.3 Although the conservation status of WBSE is determined as “Least Concern (LC)” in accordance with the “The IUCN Red List Categories and Criteria (2001 Categories and Criteria (Version 3.1))” (IUCN, 2008), it is listed under the Animals and Plants (Protection of Endangered Species) Ordinance (Cap. 586), and in Appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). It is also considered “uncommon resident” in the AFCD biodiversity database and highly sensitive to human disturbance. It is also considered to have regional conservation concern by Fellowes et al (2002).

7.4.3.4 An on-going monitoring programme of local WBSEs was first started in 2001 by AFCD to record local distribution and provides long-term monitoring of the species. A total 154 sighting records from 55 sighting locations were obtained from November 2001 to July 2007 (AFCD unpublished data). Figure 7.8 indicates the sighting locations of WBSE in Hong Kong.

7.4.3.5 The most recent estimates from confirmed sighting records up to 2003 suggest that there are possibly 39 WBSEs in Hong Kong, including 23 adults and 16 immature / juveniles (Tsim et al., 2003), and a total of 8 confirmed nests / breeding pairs. The most updated information (up to July 2007) from AFCD unpublished monitoring data show that the number of recorded breeding pairs increases from 8 pairs in 2002/03 to 12 pairs in 2006/07.

7.4.3.6 The locations of sighting records suggest that the eastern and southern waters of Hong Kong generally support more WBSEs than the western waters.

7.4.3.7 A total of 7 breeding nests have been reported in SE waters that fall within the Study Area of the proposed wind farm, as displayed in Figures 7.9 and 7.10. Of these locations, Ninepin Islands, Wang Chau and Steep Island are the closest breeding locations to the proposed wind farm (approximately 5.5 km, 5.1 km and 8.0 km from the proposed wind farm for Scenario A respectively, and approximately 6.3 km, 5.2 km and 7.7 km from the proposed wind farm for Scenario B, respectively). Surveys of these 7 breeding nest locations have been conducted by AFCD since 2002, and the results are summarised in Table 7.10.

7.4.3.8 A study on foraging behaviour of two breeding pairs of WBSE at Yeung Chau (Sai Kung) (during incubation period) and Tai Ngam Hau (during chick-rearing period) was conducted by AFCD between 2001 and 2003. Observations of the foraging pairs at Yeung Chau and Tai Ngam Hau suggest that the breeding WBSEs foraged most frequently between 7a.m. and 11a.m., and between 3p.m. and 7p.m., with the peak foraging time between 5p.m. and 7p.m. Foraging range of the breeding pairs ranged from 0.05km to 2km in radius, while their home range were estimated to be 3 – 4km during breeding period. More foraging attempts were recorded during the chick-rearing period (Tsim et al., 2003).

Table 7.10 Summary of the Results of WBSE Breeding Site Survey in SE Waters Between 2002/03 and 2006/07 (AFCD unpublished data)

Surveyed Site	Year
	2002/03	2003/04	2004/05	2005/06	2006/07
Ninepin Island	U	U	S(P)	U	X
Steep Island	U	U	U	X	X
Tai Ngam Hau	S(2)	S(2)	S(1)	S(2)	S(2)
Tsim Chau	S(1)	U	F(P)	F	S(1)
Tsang Pang Kok	U	U	U	S(1)	F
Yeung Chau (Sai Kung)	F	F	F	F	F
Wang Chau	U	U	U	U	S(1)

Notes:

S(1): Success – Bred successfully; (1) – Number of fledglings;

S(P): Probably success – No fledgling seen; judgment based on observation on adult’s behaviour, e.g. bringing food back to the nest;

F: Fail – Adults abandoned the nest during breeding period; or abnormal breeding behaviour, e.g. prolonged incubating period;

F(P): Probably fail – None of fledgling nor feeding behaviour were observed. But the breeding pair stayed at the nesting site for the whole breeding period;

X: No nest – either because the pair didn’t attempt breeding or no WBSE inhabit the site during breeding season;

U: Uncertain – Insufficient data.

7.4.3.9 A study on post-release monitoring of two immature White-bellied Sea Eagles after rehabilitation jointly conducted by Kadoorie Farm and Botanic Garden (KFBG) and Agriculture, Fisheries and Conservation Department (AFCD) in 2002 (Griffiths and Tsim, 2004) by use of radio transmitters showed that both immature birds were able to fly well and establish their territory ranges within very short period of time after release (< 1 month). The radio-tracking results suggested that all recorded positions of the two immature WBSEs were all confined to near-shore coastal areas, within approximately 2.5km from the nearest shore.

7.4.4 Conclusion of Desktop Study

7.4.4.1 A desktop study has been conducted to cover the sea area (as displayed in Figure 7.1) within 20km radius from the centre of the proposed wind farm site. Although there are no designated areas for conservation of avifauna within the desktop Study Area, general seabird populations within the area including passage migrants, visitor breeders and the resident coastal bird populations have been identified and reviewed. Ecological profiles of key species have been established based on available information on distribution, abundance, breeding and foraging behaviour.

7.4.4.2 Results of the desktop study suggest that approximately 2/3 (or 29 out of the total 45) of migratory seabird species in Hong Kong occur in SE waters. Among the migratory and visitor seabird species, only three tern species, including Black-naped Tern, Bridled Tern and Roseate Tern breed in Hong Kong during summers. Breeding colonies of the three tern species have been recorded at offshore islets (Kong Tau Pai, the East Ninepin, and Waglan Island) in the SE waters.

7.4.4.3 White-bellied Sea Eagle (WBSE), the resident coastal raptor in Hong Kong, also occurs within the desktop Study Area. Review of confirmed sighting records and nesting locations suggests that the SE waters generally support more WBSEs than other parts of Hong Kong. A total of 7 nesting locations have been recorded within the SE waters (the desktop Study Area) of Hong Kong.

7.5 Baseline Conditions and Sensitive Receivers – Field Surveys

7.5.1 Field Survey Methodologies

7.5.1.1 Based on the results of desktop study, boat-based field survey was proposed for use in this study based on the following considerations:

· Abundance of existing birds within the Study Area are predicted to be low: Daily bird counts covering the majority of seabird population in S and SE HK waters ranged from 61 – 969 (HKBWS, 2006). The numbers are far too low compared to those obtained in other wind farm studies (e.g. NERI, 2000; RPS, 2006). Boat-based surveys would be the most cost-effective approach for quantifying birds of low abundance.

· Most seabirds are of low flight altitude: the results of the HKBWS 2006 seabird study showed that more than 96% of the observed birds belong to terns (~20%), the sandpiper, Red-necked Phalarope (~76%), and shearwaters, (~1%), which usually fly at low altitudes or near water surface.

· Absence of marked passage movements of migrating / moving birds: No moving flocks of birds of significant size or marked passage of moving / migration birds have been identified in the Study Area from previous surveys / studies. Instead, all migratory / moving birds in Hong Kong offshore area were found to occur individually or in small groups of several birds from previous records.

7.5.1.2 A fixed transect route for daytime bird survey (approximately 7 hours per each survey day) has been designed to cover the proposed wind farm and the adjacent area. The transect route covers Tathong Channel and offshore islets that are confirmed or susceptible nesting ground for breeding terns and WBSE. The transect route overlaid with the proposed wind farm layout for both Scenario A and Scenario B turbine options are displayed in Figure 7.11 and Figure 7.12 respectively.

7.5.1.3 In order to avoid missing birds in key areas and to allow estimation of population sizes of breeding colonies, fixed-point counts were conducted at a total of 9 fixed survey points, including P1 to P8, and Kong Tau Pai. Point count locations have been selected to cover the project area of concern (i.e., all four corners and the centre of the wind farm site area) and four coastal locations from which bird flights would originate / breeding activity would be centred.

7.5.1.4 For P1 to P8, fixed-point counts were conducted for 30 minutes at each of the points. For Kong Tau Pai, the known current breeding location in the SE waters for terns, point-counts were performed until all terns (including adults and juveniles) on the islet were counted and recorded.

7.5.1.5 All boat-based surveys were conducted by a professional ornithologist, with the best experience in coastal and seabird survey in the HKSAR and with specific survey experience in the Study Area. Additional staff support, including support for breeding bird point counts, was provided as appropriate, depending on sighting frequency.

7.5.1.6 The boat-based survey involved scanning on both sides of the boat by the observer to ensure no under-counting. For each bird sighting, the position of the observer was recorded using a GPS navigator. An estimate of the distance of bird from observer was made and aided by the use of a range finder to allow the analysis of bird distribution across environmental gradients (e.g. distance gradients away from the proposed wind farm).

7.5.1.7 For all bird sightings along the transect route, information including species identity, number of individuals, behaviour (e.g. at flight or foraging), height and direction of flight, maturity of the birds (adult / juvenile), and whether or not the birds fly through the Wind farm area was recorded.

7.5.1.8 All the bird surveys were conducted at daytime with good weather conditions (e.g. at Beaufort scale <5), to enable the best visibility and observer efficiency, and hence data reliability.

7.5.1.9 Survey duration / periods were defined and represented by four “bird seasons” according to the general understanding of occurrence of various types of seabirds or bird population across the SE waters of Hong Kong: Spring Migratory Period (March – mid June), Summer Breeding Period for terns (mid June – August), Autumn Migratory Period (September – November), and the Winter Period (December – February) for some winter breeders such as the WBSE.

7.5.1.10 Boat-based surveys were conducted more frequently during migratory periods when observation opportunities of birds offshore were expected to be greatest. The following frequency was basically followed as far as practicable:

· Spring Migratory Period 2006 (May – mid June 2006) [Frequency: 2x /week]

· Summer Breeding Period 2006 (mid June – August 2006) [Frequency: 2x /month]

· Winter Period 2006 - 2007 (December 2006 – February 2007) [Frequency: 2x /month]

· Spring Migratory Period 2007 (March – May 2007) [Frequency: 2x /week]

· Summer Breeding Period 2007 (August 2007) [Frequency: 1x /week]

· Autumn Migratory Period 2007 (September – November 2007) [Frequency: 1x / week]

· Winter Period 2007 (December 2007) [Frequency: 1x /week]

7.5.1.11 The survey during the Spring Migratory Period in 2006 and 2007 was designed to cover spring migration of seabirds during the months March through May (HKBWS, 2006). The Summer Breeding Period covered the breeding season of the three breeding terns in Hong Kong, while the Autumn Migratory Period (September – November) covered the autumn migration of seabirds from September through November. The Winter Period (December – February) aimed to cover the key breeding season for WBSE. Less frequent surveys were conducted in 2007 than in 2006 for the summer breeding periods and the winter periods, as the 2007 surveys only aimed to expand the scope of work beyond the EIA Study Brief by collecting additional data to supplement the primary findings of 2006 surveys. Table 7.11 summarises the dates (a total of 59 days) of boat-based surveys conducted between May 2006 and December 2007.

Table 7.11 Dates of Boat-based Surveys Undertaken in the Study Area Between May 2006 and December 2007.

Survey Period	Dates
Spring Migratory Period 2006	2006 May: 23, 26, 30 2006 June: 2, 5, 9, 12, 15
Summer Breeding Period 2006	2006 July: 4, 18 2006 August: 5, 19, 30
Winter Period 2006 - 2007	2006 December: 23, 30 2007 January: 12, 24 2007 February: 8, 22
Spring Migratory Period 2007	2007 March: 7, 10, 15, 16, 20, 26, 29 2007 April: 2, 6, 10, 12, 16, 19, 23, 26, 30 2007 May: 11, 12, 17, 22
Summer Breeding Period 2007	2007 August: 16, 24, 30
Autumn Migratory Period 2007	2007 September: 6, 13, 19, 27 2007 October: 5, 11, 18, 25 2007 November: 2, 10, 17, 24
Winter Period 2007	2007 December: 1, 7, 15, 23, 29

7.5.1.12 Except for the Autumn Migratory Period 2007 which generally covers a full “Autumn” period from September through November, other survey periods in 2006 and 2007 cover different selective periods of a particular season. As such, for the collision risk presented in Sub-section 7.8 survey data from 2006 and 2007 has been pooled for the risk analysis on a ‘species per season’ basis.

7.5.2 Field Surveys Findings

Abundance and Distribution

7.5.2.1 Total counts for the whole survey period (between May 2006 and December 2007) are summarised in Table 7.12 and Table 7.13 for Scenario A and Scenario B, respectively.

7.5.2.2 Peak daily counts for the whole survey period are summarised in Table 7.14 and Table 7.15 for Scenario A and Scenario B, respectively. Appendix 7A presents details of each bird survey.

Table 7.12 Total Counts of Bird Species Recorded during Boat-based Surveys within the Study Area (the proposed wind farm area (WF) with 0.5km, 1km and 2km buffers) for Scenario A

Bird Type	Species	WF	WF + 0.5km	WF + 1km	WF + 2km	Whole Area
Passerines	Black Drongo Dicrurus macrocercus	7	7	7	7	7
	Blue Rock Thrush Monticola solitarius	0	0	0	0	1
	Chinese Bulbul Pycnontus sinensis	0	0	0	0	6
	Collared Crow Corvus torquatus	0	0	0	0	1
	Crested Myna Acridotheres cristatellus	0	0	0	0	20
	Dollarbird Eurystomus orientalis	0	0	0	0	1
	Large-billed Crow Corvus marcorhynchos	0	0	0	0	3
	Oriental Turtle Dove Streptopelia orientalis	0	0	0	0	3
	Yellow Wagtail Motacilla flava	0	5	5	8	8
	Yellow-bellied Prinia Prinia flaviventris	0	0	0	0	1
Raptors	Black Kite Milvus migrans	0	3	4	6	615
	Bonelli’s Eagle Hieraaetus fasciatus	0	0	0	0	1
	Chinese Goshawk Accipiter soloensis	1	1	1	1	5
Raptors	Common Buzzard Buteo buteo	0	0	0	0	1
	Common Kestrel Falco tinnunculus	0	0	0	0	2
	Eurasian Hobby Falco subbuteo	0	0	0	0	1
	Grey-faced Buzzard Butastur indicus	0	0	0	0	1
	Osprey Pandion haliaetus	0	1	1	1	5
	Peregrine Falcon Falco peregrinus	0	0	0	0	12
	Unidentified Raptor	0	0	0	1	1
	White-bellied Sea Eagle Haliaeetus leucogaster	0	0	0	0	138
Seabirds	Aleutian Tern Sterna aleutica	33	65	71	103	154
	Ancient Murrelet Synthliboramphus antiquus	0	0	0	0	2
	Black-naped Tern Sterna sumatrana	6	12	12	14	1048
	Black-tailed Gull Larus crassirostris	28	37	38	40	48
	Bridled Tern Sterna anaethetus	89	191	222	246	883
	Common Tern Sterna hirundo	46	85	89	99	167
	Greater Crested Tern Sterna bergii	0	2	2	2	3
	Heuglin’s Gull Larus heuglini	9	12	13	13	14
	Litter Tern Sterna albisfrons	1	2	2	2	2
	Long-tailed Jaeger Stercorarius longicaudus	1	2	6	6	6
	Pomarine Jaeger Stercorarius pomarinus	6	9	14	14	14
	Roseate Tern Sterna dougallii	0	6	7	7	181
	Short-tailed Shearwater Puffinus tenuirostris	0	0	0	0	2
	Streaked Shearwater Calonectris leucomelas	4	8	8	8	8
	Unidentified Gull Larus sp.	1	1	1	1	1
	Unidentified Jaeger Stercorarius sp.	17	17	17	17	17
	Unidentified Tern Sterna sp.	20	25	33	48	87
Seabirds	White-winged Black Tern Chlidonias leucopterus	43	72	74	80	126
Swallows / Swifts	Barn Swallow Hirundo rustica	44	53	53	55	88
	Little Swift Apus affinis	1	2	2	3	48
	Pacific Swift Apus pacificus	0	0	0	0	230
	Black-crowned Night Heron Nycticorax nycticorax	0	0	0	0	1
Waders & Waterbirds	Cattle Egret Bubulcus ibis (coromandus)	20	47	47	47	47
	Chinese Pond Heron Ardeola bacchus	0	0	0	0	4
	Common Kingfisher Alcedo atthis	1	1	1	1	1
	Curlew Numenius arquata (orientalis)	0	0	0	0	1
	Great Egret Egretta alba	0	0	0	0	1
	Greater Sand Plover Charadrius leschenaultii	0	0	1	1	1
	Green Sandpiper Tringa ochropus	0	1	1	1	1
	Grey Heron Ardea cinerea	11	12	12	12	12
	Little Egret Egretta garzetta	26	45	45	65	141
	Northern Shoveler Anas clypeata	10	10	10	10	10
	Pacific Golden Plover Pluvialis fulva	3	3	3	5	5
	Pacific Reef Egret Egretta sacra	0	0	0	0	80
	Red Knot Calidris canutus	0	0	0	15	15
	Red-necked Phalarope Phalaropus lobatus	128	243	247	283	722
	Ruddy Turnstone Arenaria interpres	7	7	7	7	7
	Unidentified egret Egretta sp.	0	0	0	0	30
	Unidentified shore bird	0	12	12	12	12
	Whimbrel Numenius phaeopus	0	0	0	0	33
	White-breasted Kingfisher Halcyon smyrnnsis	0	0	0	0	1
	Wood Sandpiper Tringa glareola	0	31	31	31	37
Total		563	1030	1099	1272	5124

Table 7.13 Total Counts of Bird Species Recorded during Boat-based Surveys within the Study Area (the proposed wind farm area (WF) with 0.5km, 1km and 2km buffers) for Scenario B

Bird Type	Species	WF	WF + 0.5km	WF + 1km	WF + 2km	Whole Area
Passerines	Black Drongo Dicrurus macrocercus	7	7	7	7	7
	Blue Rock Thrush Monticola solitarius	0	0	0	0	1
	Chinese Bulbul Pycnontus sinensis	0	0	0	0	6
	Collared Crow Corvus torquatus	0	0	0	0	1
	Crested Myna Acridotheres cristatellus	0	0	0	0	20
	Dollarbird Eurystomus orientalis	0	0	0	0	1
	Large-billed Crow Corvus marcorhynchos	0	0	0	0	3
	Oriental Turtle Dove Streptopelia orientalis	0	0	0	0	3
	Yellow Wagtail Motacilla flava	5	5	5	8	8
	Yellow-bellied Prinia Prinia flaviventris	0	0	0	0	1
Raptors	Black Kite Milvus migrans	0	3	4	6	615
	Bonelli’s Eagle Hieraaetus fasciatus	0	0	0	0	1
	Chinese Goshawk Accipiter soloensis	1	1	1	1	5
	Common Buzzard Buteo buteo	0	0	0	0	1
	Common Kestrel Falco tinnunculus	0	0	0	0	2
	Eurasian Hobby Falco subbuteo	0	0	0	0	1
	Grey-faced Buzzard Butastur indicus	0	0	0	0	1
	Osprey Pandion haliaetus	0	1	1	1	5
	Peregrine Falcon Falco peregrinus	0	0	0	0	12
	Unidentified Raptor	0	0	1	1	1
	White-bellied Sea Eagle Haliaeetus leucogaster	0	0	0	0	138
Seabirds	Aleutian Tern Sterna aleutica	36	69	71	103	154
	Ancient Murrelet Synthliboramphus antiquus	0	0	0	0	2
Seabirds	Black-naped Tern Sterna sumatrana	10	12	12	14	1048
	Black-tailed Gull Larus crassirostris	22	37	38	40	48
	Bridled Tern Sterna anaethetus	119	197	224	246	883
	Common Tern Sterna hirundo	44	86	90	99	167
	Greater Crested Tern Sterna bergii	0	2	2	2	3
	Heuglin’s Gull Larus heuglini	9	13	13	13	14
	Litter Tern Sterna albisfrons	1	2	2	2	2
	Long-tailed Jaeger Stercorarius longicaudus	2	2	6	6	6
	Pomarine Jaeger Stercorarius pomarinus	7	13	14	14	14
	Roseate Tern Sterna dougallii	0	6	7	7	181
	Short-tailed Shearwater Puffinus tenuirostris	0	0	0	0	2
	Streaked Shearwater Calonectris leucomelas	4	8	8	8	8
	Unidentified Gull Larus sp.	1	1	1	1	1
	Unidentified Jaeger Stercorarius sp.	17	17	17	17	17
	Unidentified Tern Sterna sp.	21	28	30	48	87
	White-winged Black Tern Chlidonias leucopterus	49	72	74	80	126
Swallows / Swifts	Barn Swallow Hirundo rustica	46	53	53	55	88
	Little Swift Apus affinis	1	2	2	3	48
	Pacific Swift Apus pacificus	0	0	0	0	230
	Black-crowned Night Heron Nycticorax nycticorax	0	0	0	0	1
Waders & Waterbirds	Cattle Egret Bubulcus ibis (coromandus)	29	47	47	47	47
	Chinese Pond Heron Ardeola bacchus	0	0	0	0	4
	Common Kingfisher Alcedo atthis	1	1	1	1	1
	Curlew Numenius arquata (orientalis)	0	0	0	0	1
	Great Egret Egretta alba	0	0	0	0	1
Waders & Waterbirds	Greater Sand Plover Charadrius leschenaultii	0	0	1	1	1
	Green Sandpiper Tringa ochropus	0	1	1	1	1
	Grey Heron Ardea cinerea	11	12	12	12	12
	Little Egret Egretta garzetta	30	45	45	65	141
	Northern Shoveler Anas clypeata	10	10	10	10	10
	Pacific Golden Plover Pluvialis fulva	3	3	3	5	5
	Pacific Reef Egret Egretta sacra	0	0	0	0	80
	Red Knot Calidris canutus	0	0	0	15	15
	Red-necked Phalarope Phalaropus lobatus	159	242	246	283	722
	Ruddy Turnstone Arenaria interpres	7	7	7